Introduction to Cognitive Psychology - Department of Psychological ...

Introduction toCognitivePsychologyGreg FrancisSpring 2013MWF 1:30 - 2:20 pmSmith Hall, Room 108http://www.psych.purdue.edu/~gfrancis/Classes/PSY200/index.html

Prof. Greg FrancisIntroductionPSY 200Greg FrancisLecture 01Four great mysteriesHumans face four great mysteries about theuniverse1) Why is there something instead of nothing? This is the domain of physics Most of us are not going to understand the ideasFour great mysteries.Purdue UniversityPurdue UniversityFour great mysteriesHumans face four great mysteries about theuniverse2) How did life form? This question is addressed at the boundary betweenchemistry and biologyFour great mysteriesHumans face four greatmysteries about theuniverse3) Why is there so muchdiversity of life? This is the domain ofbiology Evolution and naturalselection answer thisquestionPurdue UniversityPurdue UniversityFour great mysteriesHumans face four great mysteries about the universe4) What is the basis of human intelligence andconsciousness? Cognitive psychology and neuroscience Far from a complete answer Lots of issues to discussCourse outline Neuroscience -- EXAM (16%) Perception, Attention & Memory -- EXAM (16%) Memory & Mental representation -- Exam (16%) Language -- Exam (16%)Reasoning Cumulative Final (16%)Purdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology1

Prof. Greg FrancisTopicsTopicsDiscuss a sample of issues in cognitivepsychology / cognitive neuroscienceFor example Whats the deal with left and right brains? Why does everyone love Prozac?Try to relate cognitive psychology tostories you may have heard in the popularpressIdentify how the topics can help you to bea better person Why telephone operators seem rude. Why there is a gate at the first floor stairwayin the Psychology building. What to do if you are drunk while studying foran exam. What is the plural of walkman?Purdue UniversityPurdue UniversityTextbookThere is no textbookLecture notes are used insteadIf you want a book, borrow from a pastclassI am adding links to optional readings tothe syllabus Not for every subjectLecture notesDownloadable from theclass web page Adobe Acrobat (pdf)format Reduced form (6 to apage)Purdue UniversityPurdue UniversityBoiler castVodcast of the lectures will be provided on iTunesDo not believe that these are a substitute for attending lectureLink and instructions are on the class web pageCourse web pageSyllabus on the web http://www.psych.purdue.edu/~gfrancis/Classes/PSY200/index.html! updates to the syllabus links to labs Study guides for the exams Links to optional readings Grades will be posted after the first examThis course does not use BlackboardPurdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology2

Prof. Greg FrancisExamsMix of multiple choice and short answer Students are often surprised at how muchdetail is expected in the short answerquestionsDetailed study guides will be provided atleast one week in advance of an examBeware the scheduling of the final exam! The exam is during the final exam week There are few excuses for changing the dateClass participationOccasionally I need assistance during a lectureI will randomly select names from the class listIf you are absent, you lose class participationpoints (few excuses)Sometimes everyone in class is checkedContributes to 10% of class grade.Usually 6-10 checks per semesterPurdue UniversityPurdue UniversityHomeworkCogLabJava-based programs that can be run through aweb browser (e.g., Firefox, IE Explorer, Opera,Safari, Chrome).You participate in classic experimentsTotal lab grade contributes to 10% of your classgrade.Grade is based solely on completing theexperiment, not on the quality of the dataPurdue UniversityCogLabLabs are listed on the syllabusThey must be completed by 10:00 am at thedate indicated in the syllabus else you get no creditSince I wrote CogLab, you get access to theexperiments for free (a $30 value!)See handout for instructions on getting startedRegistration code is on a label on the instructionsFirst lab is due at 10:00 am on Wednesday!Purdue UniversityStraight scale 98% -100% A+ 93% - 97% A 90% - 92% A- 88% - 89% B+ 83% - 87% B 80% - 82% B- 78% - 79% C+ 73% - 77% C 70% - 72% C- 68% - 69% D+ 63% - 67% D 60% - 62% D-0% - 59% FGrading No rounding up: 82.99 is a B-Last semester: Fall 12# %A 13 6B 61 30C 74 35D 40 18F 23 11Purdue UniversityGradingLast semesters gradesE1 E2 E3 E4 Final CogLab% CP%Average 71.3 69.9 73.2 80.7 76.0 83.5 75.3Max 98 99 99 98 98 100 100Min 32 27 36 22 31 17.6 0Purdue UniversityPSY 200: Intro. to Cognitive Psychology3

Prof. Greg FrancisImportance of attending classCounts for 10% of final class gradeCorrelates well with final class grade (r=0.63)Instructor office hours Monday, Wednesday, Friday, 10:00 - 11:00am Or by appointment Psych 3186 Email: gfrancis@purdue.eduPurdue UniversityPurdue UniversityTeaching assistantsGrade examsKeep track of gradesMay lecture when I am out of townHave office hoursProvide out-of-class study sessions forexamsTeaching assistant Jihyun Kim Office: PSYCH, Room 3194Office hours: Monday and Wednesday, 2:30 – 4:00 pmEmail: kim510@purdue.edu!Purdue UniversityPurdue UniversityMindi CogdillTeaching assistant Office: PSYCH 3143Office hours: Tuesday and Thursday,12:30 – 1:30 pm, Friday 2:30-3:30 pmEmail: mcogdill@purdue.edu!Purdue UniversityAttitudeDuring lectures: turn off cell phones, dont read newspapers, dontplay gamesQuestions are always welcome. I can adjust my lecturing paceaccordinglyPrint out the lectures and bring them to class. Take notes duringclass. Not everything is on the slides.Everything we talk about in class is importantThis class is an introductory class, but that does not mean it is easy Its like Introduction to Physics or Introduction to Chemistry Almost every other subtopic in psychology depends on the ideas incognitive psychology Everything is at least 10,000 times more complicated than what wediscussIf you dont find a topic interesting, just wait a weekPurdue UniversityPSY 200: Intro. to Cognitive Psychology4

Prof. Greg FrancisNext timeCognitive neuroscienceThe brainThe modularity hypothesisCogLab on Brain asymmetry due!Whats the deal with left and right brains?Purdue UniversityPSY 200: Intro. to Cognitive Psychology5

Prof. Greg Francis 1/2/12Brain partsPSY 200Greg FrancisLecture 02What’s the deal with left andright brains?Purdue University!! The source of cognition(consider transplant!)!! Weighs about 3 pounds!! Damage to some parts resultin immediate death ordisability!! Damage to other partsseems to have no effect!!! What brain parts areimportant to cognition?!! How do we discover the roleof each brain part?The brainPurdue University!! Fore-brain!!cortex!! Hind-brain!!brain stem)The brain!! Hind-brain!!(brainstem)The brainLimbic Lobe(sexual behavior,emotional behavior,memory)Hypothalamus(appetite, thirst,temperature,hormones)Thalamus(sensorygateway,exceptsmell)Purdue UniversityCerebellumPurdue University (muscle control, learning)!! Cortex!! Similar to a thick,crumplednewspaper page!! Grooves (fissuresor sulci) separateregionsFore-brain!! Cortex!! Similar to a thick,crumplednewspaper page!! Grooves (fissuresor sulci) separateregionsFore-brainPurdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology 1

Prof. Greg Francis 1/2/12Fore-brainFore-brain!! Cortex!! Similar to a thick,crumpled newspaperpage!! Grooves (fissures orsulci) separate regions!! Cortex!! Similar to a thick,crumplednewspaper page!! Grooves (fissuresor sulci) separateregionsPurdue UniversityPurdue UniversityContralateral processing!! Processing in the brain is done on theopposite side of your organs!! Control of your right arm is from the leftside of your brainContralateral processing!! Neural fibersfrom the eyecross on wayto cortex!! Information from your left field of viewgoes to the right side of your brainPurdue UniversityPurdue UniversityContralateral processingCut corpus callosum!! Brainhemispheresareconnected bya mass ofneural fiberscalled thecorpuscallosum!! Behaviorchanges verylittle!! Subtle effectsPurdue Universityhttp://www.youtube.com/watch?v=ZMLzP1VCANoPurdue UniversityPSY 200: Intro. to Cognitive Psychology 2

Prof. Greg Francis 1/2/12!! If “nut” flashes on leftscreen!! subject cannot name it!! subject can pick up nutwith left hand!! If “nut” flashes on rightside!! subject can name it!! subject cannot pick upnut with left hand untilhe says “nut” out loudBrain sides1981 Nobel Prize for Roger Sperry!Purdue UniversityBrain sides!! Results led to further study and common belief that!! Left side: language, analytical, classification, Western rationalization!! Right side: art, music, recognition of faces and shapes, Easternmysticism!! Vast oversimplification!! in a normal brain, bothsides are involved inmany tasks!! Results do support the ideathat different parts of thebrain are involved indifferent cognitive tasks(modularity hypothesis)Purdue University!! In the CogLabexperiment Brainasymmetry!! You saw a pair ofchimeric faces andwere asked tochoose which onelooked youngerCogLabChimeric faces!! The faces were made by taking a normal face and anartificially aged faceTimePurdue UniversityPurdue UniversityChimeric faces!! Take opposite halves of the facesChimeric faces!! Put them together, and make the other face by flipping it!! Thus, both faces are mirror images, they contain the sameinformationPurdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology 3

Prof. Greg Francis 1/2/12CogLab!! If you stare at the middle of face, theinformation on the left goes to theright hemisphere!! which is supposed to be better able todeal with face information than the lefthemisphere!! So the information on the left side of theface should dominate the face decision!! This face should look relatively youngCogLab!! If you stare at the middle of face, theinformation on the left goes to theright hemisphere!! which is supposed to be better able todeal with face information than the lefthemisphere!! So the information on the left side of theface should dominate the face decision!! This face should look relatively oldPurdue UniversityPurdue UniversityCogLab!! So, we would expect people with brain asymmetrieswould usually choose as younger, the face that has theyounger half on the left side!! Left handed people typically do not show as much brainasymmetry effects as right handed people!! Thus, we can look for differences between left and righthanded people!! We expect right handed people to select the face with theyounger image on the left more often than left handed peoplePurdue UniversityCogLab!! Here is the data average for the class!! 191 right handed people!! 15 left handed people!! In the right direction, but a small effectLeft handed Right handedPercentage of 50.95 56.13choices withyounger faceon leftPurdue UniversityCogLab!! Here is the data average for the global data set!! 16,056 right handed people!! 1676 left handed people!! Some evidence of a very weak effectLeft handed Right handedPercentage of 56.4 57.2choices withyounger faceon leftPurdue UniversityAnatomy!! The cortex contains large fissures thatseparate five major areas!!Limbic (already discussed)!!Occipital!!Parietal!!Temporal!!Frontal!! Each has distinct propertiesPurdue UniversityPSY 200: Intro. to Cognitive Psychology 4

Prof. Greg Francis 1/2/12Occipital lobeParietal lobe!! Receivesinformation fromthe eye!!Most investigatedarea of the brain!! Sensations ofpain,temperature,touch, pressurePrimarysensoryareaPurdue UniversityPurdue UniversityParietal lobe!! Primary sensoryarea!!sensitivityinvolvesdisproportionateareas of the brain,relative to size ofbody part!! Hearing!!speech (left)!!music (right)!! Memory andattention!!visual recognitionTemporal lobePurdue UniversityPurdue UniversityFrontal lobe!! Divide lobes into areasBroadman areas!! Largest part ofcortex!!planning!!prediction!!motor area!!speech area!! e.g. Broadman!! Area 1, Area 12, … Area 200!! Some special names: V1, V2, V3,…Purdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology 5

Prof. Greg Francis 1/2/12!! Can partly identifyfunction by looking atnerves coming in andout of area!! Pathways throughareas seem to beinvolved in differentkinds of cognitive tasksArea functionPurdue UniversityColor!! There is order andfunction even withinan area!! The cortex is asheet of neurons!! In its thickness are6 layers of neurons!! numbered 1-6!! sometimes includesubdivisions (4a, 4b,4c!, 4c", …)Brain layersPurdue UniversityConclusionsNext time!! Lots of research in this area!! New brain regions are being mapped outdaily with ever increasing resolution!! Cognitive neuroscience relies strongly onthe “modularity hypothesis”!! Putting everything together is very difficult!! Brain scans!! EEG recordings!! MRI scans!! PET scans!! Functional MRI!! How to study the brain without killing someone.Purdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology 6

Prof. Greg Francis 1/2/12Brain scansPSY 200Greg FrancisLecture 03How to study the brain withoutkilling someone.Purdue UniversityScanning!! Technology provides insight into brain processes!! EEG recordings!! MRI!! Functional MRI!! Non-invasive!! Maps of brain activity!! The goal is to relate brain events to cognitiveeventsPurdue UniversityResolution!! For almost every technique we have to worry about itsability to discriminate differences in!! Space: whichplace is active?!! Time: whendoes somethinghappen?!! Finer resolutionis usually better!! But can bedifficult to dealwith so muchdataElectroencephalogram!! EEG!! The brainproduceselectrical activity!! Put electrodeson the headPurdue UniversityPurdue UniversityEEG!! Watch the electrical current change through time whilereading sentences (averaged across many trials)!! Good temporal resolution!! Kutas & Hillyard (1980)SemanticanomalyBrain maps!! You can analyze the EEG signals in many different ways!! Compare the signal strength for different situations!! Ayahuasca is a Brazilian psychoactive teaPurdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology 1

Prof. Greg Francis 1/2/12!! Poor spatialresolution!! You never reallySpatial resolutionknow which part ofthe brain is makingthe current!! Lots of work toimproveMagnetic Resonance Imaging!! Magnetic fieldforces protons inyour body to lineup!! pulses of radiointo fieldbouncesprotons around!! as they returnto normalposition, theyemit a signalthat can bedecoded into amapPurdue UniversityPurdue UniversityMagnetic Resonance Imaging!! MRI Scans: Like an x-ray machine, butcan look at soft tissue (like lungs, heart,..)!!Very good spatial resolutionMagnetic Resonance Imaging!! MRI Scans: Like an x-ray machine, but can look at softtissue (like ankles, my brain,..)!! Very good spatial resolution!! millimetersPurdue UniversityPurdue UniversityMRI ScansMRI Scans!! MRI providesa “slice” at atime!! Take multipleslices to buildup full image!! Can identifyanatomicaldifferences betweenbrains!! Alcoholic has largerventricles and thinnercorpus callosumNobel prizewinning work!Purdue University!! Note comparingacross brains is a bittricky!!! Everyone’s brain is abit differentPurdue UniversityPSY 200: Intro. to Cognitive Psychology 2

Prof. Greg Francis 1/2/12!! 3-D maps!! NormalMRI ScansMRI Scans!! Non-invasive, no side effects!! Allows early detection of brain disease,tumors,…!! Fantastic spatial resolution!! But…!! Ataxia:!!loss of motorcontrol!!it only shows structure!!no way to know what a brain area doesPurdue UniversityPurdue UniversityFunctional MRI!! Just like MRI, but with a new analysis!! MRI differentiates between different types of tissue (celltypes)!! Functional MRI differentiates between active andinactive neurons: concentration of oxygen!! The measurement is called the “blood oxygen leveldependent” (BOLD)»! It roughly tracks the flow of blood in the brain»! More active neurons recruit more bloodPurdue University!! Color maps showstrongest“responses”!! E.g., fMRI scan of awoman after a stroke!!Blue/green:normal blood flow!!Red/black:abnormal bloodflowFunctional MRIPurdue UniversityFunctional MRIScanning!! Consider this fMRI scan!! Very good spatialresolution!! millimeters!! Pretty goodtemporal resolution!! Seconds!! (Silva, 2002)!! It shows regional cerebralblood volume (rCBV)!! You cannot tell how/ifdifferent regions are involvedin different activities!! Breathing!! Digestion!! Thinking about exams!! …..Purdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology 3

Prof. Greg Francis 1/2/12Differences!! We have to look for differences in activity!! Alzheimer’s patients have reduced brain activityA simple experiment!! Suppose you run an fMRI experiment where a person alternatesbetween seeing a blank screen and a face!! You take multiple fMRI scans with half recording brain activity duringthe blank and half recording brain activity during the face!! Add them up pixel by pixel for each conditionViewing blankViewing facePurdue UniversityPurdue UniversitySubtraction method!! Subtract the fMRI signals produced by one condition from the fMRIsignals produced by another condition!! The difference map indicates those brain regions that are involved inthe different cognitive tasks!! It requires a sophisticated statistical analysis to avoid false positives!Viewing blankViewing facePurdue UniversityDifference map!! What is usuallyreported is just thedifference map!! Colors mark placesin the brain that arestatistically differentbetween conditions!! Czisch et al. (2009)for rare tones vs.frequent tones!! The map would bedifferent if itcompared rare tonesversus speechReportingPurdue UniversityFunctional MRIFunctional MRI!! Color maps showstrongest“responses”!! e.g., during a taskthat requires covertspatial attentioncompared to one thatdoes not requireattention!! When moving apointer to a targetbox compared to nomovement!!“activity” in areasinvolved in vision,planning, and motorcontrolPurdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology 4

Prof. Greg Francis 1/2/12Limitations!! Brain scans do not really tell us how the brainworks!! the scans just tell us approximately where in the brainsomething occurs!! sometimes it can tell approximately when!! Even trying to find the place may be problematic!! Lots of cognitive abilities involve many different areasof the brain!! Most of the time theories of cognition are derivedfrom experimental psychology!! Brain studies explore how to implement the theoriesPurdue UniversityCommon misconception!! Brain scans demonstrate a physiological basis tothings that were thought to be emotionally orcognitively based!! e.g., MRI scansof stutters!! in fact, allbehavioral traitsare physiologicallybasedPurdue UniversityConclusions!! Lots of research in this area!! Technology is improving in many ways!! There are many other types of scanningtechnologies!! Computerized Axial Tomography (CAT)!! Diffusion tensor imaging (DTI)!! Single Photon Emission Computed Tomography(SPECT)!! Near Infrared Spectroscopic Imaging (NIRSI)!! Magnetoencephalography (MEG)!! Positron Emission Tomography (PET)Purdue UniversityNext time!! How do we use brain scans to studycognition?!! How good are the scans?!! What is really being measured?!! How to read someone’s mind.Purdue UniversityPSY 200: Intro. to Cognitive Psychology 5

Prof. Greg Francis 1/2/12Brain scansPSY 200Greg FrancisLecture 04Scanning!! Brain scanning techniques like fMRIprovide spatial and temporal patterns ofactivity across the brain!! We want to analyze those patterns todiscover how the brain worksHow to read someone’s mind.Purdue UniversityPurdue UniversityfMRI!! Passive listening vs. active listening!!Vannest et al. (2009)!! Twenty children (ages 11-13) complete threetasks!!Passive listening: hear a female speaker tell a 30-second story!!Active response: hear the same speaker tell a story in5 second segments of two sentences. Scanningoccurred after the sentences (silence). Answerquestions!!Random tones: no task, just listenPurdue UniversityfMRI!! The colors show thedifference mapsrelative to listening tothe tones!! Common activity(breathing, digestion,hearing machinenoise,…) issubtracted out!! The colors are notbrain activity!Purdue UniversityPassivelisteningActiveresponsefMRI!! More and differentactivity for activelistening compared topassive listeningPassivelisteningfMRI!! Does more activity for theactive response mean activeresponse listening is “better”than passive listening?Passivelistening!! (Could it be!! Tested children onotherwise?)comprehension of stories!! PL: 75.1% correct, SD=12.7Activeresponse!! AR: 79.1% correct, SD=9.1!! No real difference incomprehensionActiveresponsePurdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology 1

Prof. Greg Francis 1/2/12Sensory substitution!! Some scientists look for replacements to lost perception!! For example, there is a tongue display unit that attempts to present spatialinformation for blind peopleSensory substitution!! People can use a TDU to discriminate shapes!! Kaczmarek, Bach-y-Rita & Tyler (1998)!! Link to video on class web pagePurdue UniversityPurdue UniversitySensory substitution!! So what happens in the brain?!! Are there responses from areas typically involved in shapeperception? Or in areas related to touch on the tongue? (or both orneither?)!! fMRI differencessuggest that using theTDU involves areas ofmotor cortex!! Not areas that aretraditionally for visualperceptionSensory substitution!! This is the kind ofquestion that can bestbe answered with brainscan technologyPurdue UniversityPurdue UniversityReading minds!! Long-term, the goal of fMRI research is to be able to analyze a brain scanand identify what a person is thinking!! There are several attempts to do this (Haynes et al., 2006)!! Adding or subtracting numbersReading minds!! Based on the choice at the end, we can deduce whether the subject choseaddition or subtraction for that trial!! Make an fMRI scan during the selection process!! Whether to add or subtract numbersScan "now"Purdue UniversityPurdue UniversityOr scan"now"PSY 200: Intro. to Cognitive Psychology 2

Prof. Greg Francis 1/2/12Mind reading!! It is possible to build a recognition system that distinguishes (with71% accuracy) the brain patterns for addition and subtraction!! Depends on the place in the brain!! Different places for intention and execution!! You can read the mind of these subjects!Thought reconstruction!! Another research group analyzed fMRI responses toreproduce a shown imagePurdue UniversityPurdue UniversityThought reconstruction!! Performance depends on where the signals come from!! Fewer errors for “lower” brain areas!! Where do you stop?, the retina?, the lens of the eye?Purdue UniversityThought reconstruction!! These kinds of studies are mostly a demonstration oftechnology!! we already know the brain represents visual information!!! Before the study was run, we knew that there weredifferences in the brain when we see different images!! The percept is based on the brain, so there must bedifferences!!! These kinds of studies tell us that theneurophysiological differences between cognitiveevents can be measured by these brain scanningtechnologies!! A bit like statistical powerPurdue UniversityMind reading with EEG!! You can purchase an EEG device to read your mindand control a computer!! www.emotiv.comMind reading limits!! fMRI: If subjects decide to multiply numbers, a system trained todistinguish between subtraction and addition is clueless!! Thought reconstruction: As the number of possible images to beshown increases, it becomes harder to reconstruct the shownimage!! Emotiv: very limited set of mind reading possibilities!! This is a very limited form of mind reading!! People do better than this every day by watching people behave(posture, eyes, skin tone)!! The military and police are quite interested in these approaches!! Companies are interested in identifying what you think!! Proctor & GamblePurdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology 3

Prof. Greg Francis 1/2/12Problems / limitations withscanning!! So much data that is difficult to know what to do with everything!! Statistical analysis is complicated!! In a small brain scan, you may have 64 x 64 voxels x 10 slices»! =40,960 voxels overall!! Some of those voxels will give different responses just by chance!! Difficult to compare across subjects!! Slightly different anatomy!! Blurring of images is difficult to deal with!! Sometimes blur together brain areas, across a fissure, that areactually far apart on surface of cortex!! Some cognitive events are faster than the technology can track!! Can only measure the brain, cannot manipulate itPurdue UniversityStatistics!! It is easy to do the statistics incorrectly (it has taken a while for thefield to sort this out)!! Bennett et al. (2010) ran a study where the subject was shown aseries of photographs depicting people in social situations with aspecified emotional valence, either socially inclusive or sociallyexclusive. The subject was asked to determine which emotion theindividual in the photo must have been experiencing.!! fMRI contrasts were computed between the scans for the twotypes of emotional valencePurdue UniversityStatistics!! There is a significant difference in fMRI activity for some regions ofthe brain!! Medial brain cavity and upper spinal columnStatistics!! The “subject” in this study was a mature Atlantic salmon(sex unknown)!! The “active” regions identified by the fMRI are due tochance!! Even with purely random noise, there will be somestatistically significant findings!! The brain has lots of random noise!! These problems can be reduced but never entirelyeliminated!! They are common to many areas of psychology, not just brainscansPurdue UniversityPurdue UniversityNeurons!! The brain cells that are responsible for cognitionare neurons!! Dendrite!! input!! Soma!! integrateA neuron!! Axon!! output!! Myelin sheath!! insulatePurdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology 4

Prof. Greg Francis 1/2/12A neuron!! There are many different types of neurons!! We will describe only the most common characteristicsHow many neurons?!! Estimates of 10^11 neurons in the human brain!!100,000,000,000; one hundred billion!!estimates of 100,000 per cubic millimeter!!(about the resolution of functional MRI)!! Millions are active at any given timeNo. 21 mmPurdue UniversityPurdue UniversityConclusions!! Brain scans usually look at differences in brain“activity”!! Lots of technical (and ethical) issues!! The goal is to be able to look at a map (or movie)and be able to read someone’s mind!! That is many years off!! questionable if it can even be done with these methodsaloneNext time!! What is the neural activity that producesbrain scans?!! How do neurons transmit information toother neurons?!! Why does (nearly) everyone like Prozac?Purdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology 5

Prof. Greg Francis 8/12/11Neurons and neurotransmittersPSY 200Greg FrancisLecture 05Neurons!! The brain cells that are responsible for cognitionare neuronsWhy does (nearly) everyonelove Prozac?Purdue UniversityPurdue UniversityInput / outputInput at dendrites!! Electrical signal!! Established by therelative amount ofcharged ions insideversus outside the cellmembrane!! Inputs change theresting potential of thecell!! Output identifies whenthe cell potential hasincreased a lotelectrode-70 millivoltsresting levelPurdue UniversityNA+K+NA+K+!! Changes the cell membranepotential!! which causes furtherchanges in the cell’schemistry!! which causes furtherchanges in the membranepotential!! Strong enough input crossesa threshold and the cell fires!! action potentialhttp://www.mind.ilstu.edu/curriculum/neurons_intro/action_potential_1.0.swfPurdue UniversityA neuron!! An action potentialgenerated at thesoma travelsdown the axon tothe terminalsA neuron!! The action potential then affects the membranes ofother cells’ dendritesPurdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology 1

Prof. Greg Francis 8/12/11Output!! Myelin is like insulation for thecell’s axon!! it insures that the signalgenerated by the actionpotential is strong!! Jumps electrically rather thanthe normal chemical exchanges!! In multiple sclerosis the body’simmune system attacks myelin!! physical problems (paralysis)!! cognitive problems (memory,reasoning, judgement)!! cause unknown (300,000people)Output!! The output of a neuron is either excitatory or inhibitory on theother neuron it reaches!! Excitatory: when our neuron sends an output, the receivingneuron is more likely to produce an action potential!! Inhibitory: when our neuron sends an output, the receivingneuron is less likely to produce an action potentialPurdue UniversityPurdue UniversityNetworks!! Cognitive behavior is related togroups of neurons workingtogether!! Include excitation and inhibition!! more laterEpilepsy!! Disease of central nervous system!! causes mostly unknown!! Seizures!! bursts of electrical activity travelling through networks inthe brain!! brain activity is out of control!! epileptic fits!! Isolated seizures also occur due to high fever,lack of oxygen, or head injuryPurdue UniversityPurdue UniversityEpilepsyEpilepsy!! EEG recordingsare often used todiagnose epilepsy!! Many differenttypes of epilepsy,with different EEGpatterns!! One theory (but not yet proven) is that epilepsypatients’ inhibitory cells are not working properly!! Excitatory cells activate everything until they “exhaust”themselvesPurdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology 2

Prof. Greg Francis 8/12/11!! Treatment generally involves!! drugs, diet, avoiding stress,keeping regular schedule!! In extreme cases surgeryprevents seizures fromspreading throughout thebrainEpilepsyA balanced brain!! The brain is a dynamic system!! at multiple levels!! Neurons!! balance between “forces” inside and outside of cell membraneallows for action potentials!! Networks!! balance between excitation and inhibition!! Without these balances you do not thinkPurdue UniversityPurdue University!! Axon --> dendritesNeural connectionsMolecular structure!! Molecules have a particular threedimensionalshapewaterbenzenehttp://www.mind.ilstu.edu/curriculum/neurons_intro/neurotransmission_classic_3.0.swfPurdue UniversityPurdue UniversityMolecular structureMolecular structureTNT!! Different molecules have differentshapesaspirin!! Neurotransmitters are just molecules!! At least 50 different neurotransmitters!!dopamine, norepinephrine, serotonin,acetylcholine, glutamate, gammaaminobutyricacid (GABA)!! All with different shapes!Purdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology 3

Prof. Greg Francis 8/12/11ReceptorReceptor!! Very largemolecules calledproteins!! Similar to a filter!! accepts someneurotransmitters!! rejects others!! When it accepts a neurotransmitter, itstarts a chain reaction of events!!physical, chemical, electrical!!locally changes the cell membrane»!depolarization (excitation)»!hyperpolarization (inhibition)Purdue UniversityPurdue UniversityNeurotransmitters!! Different neurotransmitters are associatedwith different properties!!actually neurotransmitter and receptor pairs!! neural!! cognitive!! behavioralTourette’s syndrome!! Inherited (~200,000 in US)!! Behavior!! Tics»! Simple: eye blinking, facial grimacing, sniffing»! Complex: coordinated patterns, sniffing objects,jumping, twisting!! Too much dopamine!! Treated with Haldol (among others)!! blocks dopaminehttp://www.cbsnews.com/video/watch/?id=1033515nPurdue UniversityPurdue UniversityParkinson’s!! Lack of dopamine!!Many different causes!!In extreme cases, patients are “frozen”!! Give patients large doses of L-DOPA!!a precursor of dopamine!!sometimes solves the problem!!lots of side effects!! Awakenings, by Oliver SacksDrugs!! Interact with neurotransmitters in lots of ways, forexample!! Replace: accepted by receptor and with similar effect!! Production: increase or decrease!! Reuptake: knock out enzymes that removeneurotransmitter from receptor, neurotransmitter has abigger effect!! Blocking: enter receptor but does not trigger reaction,partly closes receptor protein so neurotransmittercannot enterhttp://you.video.sina.com.cn/b/10602624-1406529597.htmlPurdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology 4

Prof. Greg Francis 8/12/11!! Some forms of depressionseem to be related to limits inthe use of theneurotransmitter serotonin!! Prozac is a selectiveserotonin reuptake inhibitor(SSRI)!! it keeps serotonin bound to areceptor for longer thanusual, thereby increasing itseffect!! Prozac is one of the mostwidely prescribed drugs in theworld!ProzacOther drugs!! Amphetamines: release of norepinephrin or dopamine!! LSD: resembles serotonin!! Phenothiazine drugs: block dopamine!! Curare: blocks acetylcholine!! Cocaine: prolongs effects of dopamine!! Morphine: resembles small set of neurotransmitters called endorphinpeptides (modulate pain perception)!! Tetrahydrocannabinol (active ingredient in marijuana): binds tosome neuroreceptors, but it’s not clear what it doesPurdue UniversityPurdue UniversityConclusions!! Neural action potentials!! Shape of proteins!! Specific use of neurotransmitters for certain behaviors!! Current work on identification of role of neurotransmitters!! Lots of money to be made!! Lots more complicated than what we’ve seen hereNext time!! Neural sensitivity!! Neural codes!! Receptive fields!! CogLab on Blind spot due!!! How do you recognize your grandmother?Purdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology 5

Prof. Greg Francis 8/14/12Receptive fieldsAction potentialPSY 200Greg FrancisWith enough excitatory input, a cell produces anaction potential that sends a signal down its axonto other cells But a single action potential has little effectLecture 06If the input stays present, the cell producesanother, and another,…How do you recognize yourgrandmother? A rapid series of action potentials can influence othercellsThe number of action potentials in a certainlength of time determines the firing rate of thecellPurdue UniversityPurdue UniversityFiring rateSpecificity8 spikes (action potentials) during 300 ms Firing rate of 27 Hz (27 spikes per second)Two key questions in cognitive neuroscience are What stimulus (or stimuli) makes a given cell fire at a strongrate?» something red?» a pen?» your grandmother? What does it mean when a given cell fires strongly?» You are thinking of something?» Seeing something?» Remembering something?Purdue UniversityPurdue UniversityReceptive fieldThe set of stimuli that reliably changes acells firing rate.A stimulus could excite the cell above normal firing rateOr inhibit the cell below normal firing rateReceptive fieldReceptive fields are very useful for studies of spatial perceptionTouch involves sensitivity to pressure on skinThe loops indicate theregions where a singleneuron responds topressurePurdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology 1

Prof. Greg Francis 8/14/12Receptive fieldReceptorsReceptive fields are very useful for studies of visual perceptionVision involves spatial patterns of lightThe back of the eyecontains tightlyLightpacked sensorscalled rods andcones that detectlight at a particularlocationPurdue UniversityPurdue UniversityReceptorsReceptorsLight sensors (rods and cones) respond to lightat a particular location in the back of the eye produces a neural responseA receptor has a simple receptive field it responds to light of the right wavelength (color) and the rightpositionLightPurdue UniversityPurdue UniversityWhere nerves leavethe back of the eye,there are no lightreceptors light that hits this spot isnot visibleBlind spotBlind spotIn CogLab you mapped your blind spotmydatalookslikethisPurdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology 2

Prof. Greg Francis 8/14/12Blind spotHeres the average data for the whole class141 participantsNetworkLight receptors do not just pass informationto the brainNeurons are hooked together in an inhibitorywayLightPurdue UniversityPurdue UniversityEvidence of inhibitionEvidence of inhibitionStimulation of a center region alonegives a strong responseAdding light to a surrounding regioncan reduce the responseLight4035Light4035Firing rate3025201510Firing rate3025201510550Time0TimePurdue UniversityPurdue UniversityReceptive fieldThe receptive field of this cellincludes any place on the retinawhere light excites the cell and anyplace where light inhibits the cellOn-center, off-surround-+Purdue UniversitySpatial responsesThe spatialpattern ofexcitation(center) andinhibition(surround)means the cellis sensitive tothe location ofa small spot oflightPurdue UniversityPSY 200: Intro. to Cognitive Psychology 3

Prof. Greg Francis 8/14/12Spatial responsesSimple cellsYou have manydifferent suchcells withreceptive fieldsOn-center, off-surround cells send action potentials tosimple cells in parts of visual cortex, which have orientedreceptive fieldsthat arecentered atdifferentlocationsThey responddifferently to anedgePurdue UniversityPurdue UniversitySimple cellsSimple cellsOn-center, off-surround cells send action potentials tosimple cells in parts of visual cortex, which have orientedreceptive fieldsSimple cells come in a variety of types,but all are sensitive to bars or edges of apreferred orientation at a particularlocationPurdue UniversityPurdue UniversityInformation processingFor simple cells, an image like thisInformation processingFor simple cells, an image like this is coded something like thisStrongresponsesat edges!Purdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology 4

Prof. Greg Francis 8/14/12Complex cellsComplex cellsMany simple cells feedinto a complex cell,which is insensitive todirection of contrast andresponds to an orientedMany simple cells feed into acomplex cell, which isinsensitive to direction ofcontrast and responds to anoriented bar in many differentplacesbar in many differentplaces Often these cells are alsosensitive to directions ofmotionPurdue UniversityPurdue UniversityReceptive field hierarchyHigher order cellsReceptive fields inherit some properties fromComplex cells feed into hypercomplex cells, whichlower-level cellsare sensitive to some types of curves and visualBut they also gain new selectivity by interactingformswith each other (and across levels)Receptive fields seem to get ever more complexWhat does this mean?Is there a grandmother cell?Purdue UniversityPurdue UniversityReceptive fieldGrandmother cellsIn the inferior temporal cortex of monkeysIt could be that a single cell has a receptive fieldsome cellsappear to havereceptive fieldsselectively tuned to respond to the image of yourgrandmother but it is unlikelythat respond tomonkey faces, inprofile not enough cells cell deathReceptive fields become less useful as we searchfor neural representations of non-sensory concepts What is the receptive field of a neuron that codes loveor trust?Purdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology 5

Prof. Greg Francis 8/14/12ConclusionsReceptive fields any stimulus that affects a cells firing rate excitatory inhibitoryVery useful for studies of the visualnervous systemLots of issues left unresolvedNext timeNetworks of neuronsConnections between cellsFeedback -- resonanceCogLab on Receptive fields due!Seeing things that are not there.Purdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology 6

Prof. Greg Francis 1/5/11Neural networksPSY 200Greg FrancisLecture 07Seeing something that is notthere.Purdue UniversityReceptive field!! As we saw last time, acell’s receptive fielddepends to a large extenton the receptive fields ofother cells!! (e.g., complex cells dependon simple cells)!! Today we look at someissues involved innetworks of neuronsPurdue UniversityFeedbackResonance hypothesis!! Initially cell firing rates may vary a lot!! Cell 1 can affect cell 2, whichcan affect cell 3, which canaffect cell 1 again, which…!! What happens to cell firingrates?!! high firing rate==> active321!! In some networks cell firing rates stopchanging much (unless outside input changes)!! The remaining active cells are those thatsupport each others’ activities throughexcitation: resonance!! Inactive cells are inhibited by the active cells!! low firing rate ==> inactive!! Mental awareness ==> resonancePurdue UniversityPurdue UniversityA “simple” model!! A cell’s activation is on or off (one or zero)!! Cell connections (synapses or weights) arereciprocal!! Cells update activations one at a time!! Cell activations are calculated with the ruleai&" $ 1 if wijaj> 0= #%$ 0 if & wijaj! 0Purdue UniversityA simple model!! Neural connections(synapses) are described asweights on the links betweencells2 w 21!! Input to a cell is the summed 31multiplication of sendingw 41activation and weight w 34w 15!! Reciprocal weights have 45w ij = w ijPurdue UniversityPSY 200: Intro. to Cognitive Psychology 1

Prof. Greg Francis 1/5/11Demonstration!! Cell activities do settle down eventually!! Final pattern of activities satisfiesconstraints of the network connections!! Error correction capabilities!! Can tolerate the loss of some cells!! Emergent properties of the network!!no single cell has these propertiesFeedback!! Feedback in networks can act to “clean up”noisy sensory information to make itconsistent with what our systems expect!! In a very real way, what we see, hear, taste,smell, touch, and think, is biased by ournetwork’s expectation!! A network’s expectation is established by itsconnection weights!! excitation -- inhibitionPurdue UniversityPurdue UniversitySeeing things that are notthere!! Do you see a square in front of thepac men?Seeing things that are notthere!! Neurons in area V2 of your brain“create” the illusory contoursPurdue UniversityPurdue UniversityFeedback!! Similar situation with the illusorycircle seen hereConclusions!! Networks of neurons have propertiesdifferent from single cells!!emergent properties!!stable activities!!multiple constraints!!tolerance to errors and cell loss!! Structure of connections (synapses)determines the final pattern of responsesPurdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology 2

Prof. Greg Francis 1/5/11Next time!! How networks learn!! Changing connections!! Learning rules!! Self-organization!! A problem with virtual reality.Purdue UniversityPSY 200: Intro. to Cognitive Psychology 3

Prof. Greg Francis 1/5/11ProblemProblem!! The network coordinatingeye-hand systems, adjustsitself!! Extended use of thecomputer cameras makesthe user adapt so his eyesare where the cameras are!!! After taking the cameras off,it takes some time to adaptback!! Eye-hand coordination is off!! Could be a problem forsurgeons and pilots!Purdue UniversityPurdue UniversityOther adaptations!! Inverted prisms!! Fortunately, the adaptations return tonormal pretty quickly!! Kind of like the feeling you get afterroller-skatingConclusions!! Learning in neural networks!!changing connections!!relatively simple rules!! Much of our perceptual and motorbehavior is based upon this type ofcontinuous learning!! It’s not clear if more cognitive learning issimilarPurdue UniversityPurdue UniversityNext time!! Neural networks for visual perception!!brightness!!color!!form!! Why we see color afterimages.Purdue UniversityPSY 200: Intro. to Cognitive Psychology 3

Prof. Greg Francis 1/5/11Brightness contrast!! Two receptive fields inside the middle squarereceive the same excitatory and inhibitory signals!! Little responseBrightness contrast!! Receptive fields on the corner!!Receive the same excitation at the center!!differ in the amount of inhibition in the surroundlessinhibitionmoreinhibitionPurdue UniversityPurdue UniversityBrightness contrastBrightness contrast!! Thus, the visual system computes brightness assomething like local contrast!! It’s a property of the center-surround cells!! Our percept of brightness is determined by the responsesof cells at contrast edges!! As a result, things that have equal physical intensities canlook dramatically different (next slide)Purdue UniversityPurdue UniversityBrightness contrastBrightness contrastPurdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology 2

Prof. Greg Francis 1/5/11Brightness contrastHermann gridPurdue UniversityPurdue University!! Seems related to on-center,off-surround cells!! Cells at intersections receivemore inhibition than cells atsingle roadsHermann gridFilling-in!! We see color and brightness inside objects!! so edge information must fill-in to the interior!! It sometimes gets things messed up!! Water color effect!! How do we explain the otherversion?Purdue UniversityPurdue University!! Brightnessinformationspreads acrosssurfaces!!Craik-O’Brien-CornsweeteffectFilling-in!! Carefully fixatethe pink center!! If you keep youreyes very still, itwill disappear!! The yellow fills-in!Filling-inPurdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology 3

Prof. Greg Francis 1/2/12Visual dynamicsPSY 200Greg FrancisLecture 10Why (CRT) computer monitorswork.Purdue UniversityFlicker!! A flashing light looks constant if it ispresented rapidly enough!! The frequency of flashing at whichsubjects do not detect flicker is called theCritical Flicker Frequency (CFF)!!about 50 Hertz (50 on-off cycles in a second)!!20 millisecond durationsPurdue UniversityCFF!! Establishes minimum characteristics ofelectronic devices!! Lights flicker at 120 Hz!! we spend a lot of time in darkness!! Computer (Cathode Ray Tube, CRT)monitors and TV’s flicker at around 60 Hz!! better monitors go faster!! Liquid Crystal Display (LCD) monitors workdifferently»! Although some still flickerPhosphor!! The phosphor on a computer screentypically glows less than 10milliseconds!! ten thousandths of a second!! The gun reactivates the phosphorevery 17 milliseconds!! Thus, at any given time 1/3 of thescreen is dark!! the percept persists in your head!Purdue UniversityPurdue UniversityPersistenceExperiment!! What is the source of the persisting percept?!!Receptors in the eye?!!Receptive fields?!!Network interactions?!! Bowen, Pola & Matin (1973)!!subjects adjust duration of a blankstimulus so onset of probe matchedperceived offset of the target!!Cognitive (memory)?!! Studies support network interactions+Purdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology 1

Prof. Greg Francis 1/2/12Strange property!! As the target’sduration orluminanceincreases!!its persistencedecreasesNetworks!! Feedback isimportant!! Produces a persistingresponse!! DemonstrationInput from eyesPurdue UniversityPurdue UniversityExplanation!! Francis, Grossberg & Mingolla (1994)!! Something has to reset the network!!else it would keep “persisting” forever!! Two mechanisms!!(1) new inputs inhibit old responses!!(2) afterimages act as new inputs!! Note: afterimages get stronger as durationand luminance increase!Purdue University!! Offset of input fromExplanationthe eyes produces anafter response!!e.g., due tocompetition fromorthogonally tunedcells!! Offset responseinhibits persistingresponseInput from eyesPurdue University!! As the target’sduration orluminanceincreases!! the afterimageproduced at targetoffset increases instrength!! so there is strongerinhibition to breakthe feedback!! so the persistenceof the originalpercept decreasesExplanationPurdue UniversityWait a minute!! If visual percepts persist for over 100milliseconds, why doesn’t the world seemblurry?!!There should be smears of objects as theymove or as we move!! There must be something else preventingsuch blurring!!maskingPurdue UniversityPSY 200: Intro. to Cognitive Psychology 2

Prof. Greg Francis 1/2/12No maskNo mask!! Write down all the letters you see!! Write down all the letters you seePurdue UniversitySR KMW DPurdue UniversityNo maskMasking demonstration!! Write down all the letters you see!! Write down all the letters you seePurdue UniversityPurdue UniversityMasking demonstrationMasking demonstration!! Write down all the letters you see!! Write down all the letters you seeDQ FJR PPurdue UniversityX X X XX X X XX X X XPurdue UniversityPSY 200: Intro. to Cognitive Psychology 3

Prof. Greg Francis 1/2/12Significance!! The mask appeared after the target turned off!! The target was presented all by itself for a briefperiod of time!! However, our visual system is unable todevelop a complete percept of a scene in asuch a period of time!! Thus, the XXX mask interferes with processing ofthe letters by shortening their persisting responses!! And prevents perceived blurring of changing scenes!! Masking is often used to investigate otheraspects of cognition (we’ll see examples later)Purdue UniversityMotion!! In simple animals (like flies and frogs), we know howmotion is detected!! Demo on web page!! Reichardt detectorhttp://neurovision.berkeley.edu/Demonstrations/matthew/reichardt.htmlPurdue UniversityDetecting Motion!! Humans have something like Reichardt motion detectors!! at lots of different positions in the visual field!! sensitive to lots of different motion directions!! sensitive to lots of different motion speeds!! Think of them as receptive fields that vary in both space and time!! Many aspects of how we perceive motion follow from the propertiesof Reichardt motion detectorsApparent Motion!! When objects move, there is a continuous path ofmotion!! Reichardt motion detectors do not requirecontinuous motion!! and, continuous paths are not necessary for motion to beseenTime1Time2Purdue UniversityPurdue UniversityApparent Motion!! For just two stimuli, it does not depend on!! color!! shape!! attention!! cognitive primingApparent Motion!! The percept of motion does depends on!! stimulus duration!! interstimulus interval (50-200 msec)!! distancePurdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology 4

Prof. Greg Francis 1/2/12!! For a Reichardt detector toindicate motion, the signal fromthe second area must follow thesignal from the first by just theright length of time!! Vary the Interstimulus Interval(ISI) between the stimuli!! The time between offset of thefirst stimulus and onset of thesecond stimulusTimingISI too short"ISI about right"Korte’s laws!! Apparent motion was highly studied at the turn of the 20th century!! Korte (1915) noted that to get good motion, you needed to increasethe ISI between the stimuli as the distance between them increased!! CogLab data!! (169 participants- from class, 12,824 from global)ISI too long"Purdue UniversityPurdue UniversityMotion representation!! One conclusion of studies of apparent motionis that motion is a fundamental percept!! It has an explicit representation in the visual system!! You could imagine otherwise, we can be aware ofsomething moving withoutactually seeing the movement!! Apparent motion is the sourceof motion for all movies andanimationPurdue University!! Just like for color andorientation, we mightexpect an aftereffect ofmotion!! competition betweenopposite directions ofmotion!! Left-right!! Up-down!! habituating gate!! offset of one direction leads torebound in otherMotion aftereffectPurdue UniversityAfter effect!! Motion can be adopted by non-moving stimuli!! http://www.michaelbach.de/ot/mot_adapt/index.html!! Dynamic vision!! flicker!! persistence!! network dynamics!! Masking!! Reichardt detectors!! Apparent motion!! Motion aftereffectConclusions!! Also used to investigate other areas of cognitionand types of mental problemsPurdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology 5

Prof. Greg Francis 1/2/12Next time!! Face perception!! Visual agnosia!! Emotions!! Aftereffects!! Why do women wear cosmetics?Purdue UniversityPSY 200: Intro. to Cognitive Psychology 6

Prof. Greg Francis 1/2/13Face perceptionPSY 200Dr. FrancisLecture 11FacesPeople are very sensitive to faces Identity Health Interpretation of environmentWhy do women wearcosmetics?Purdue UniversityPurdue UniversityMotherTheresa?"FacesPeople see faces where they do not existFacial FeaturesPeople focus on some key features of faces Eyes, mouth And can fail to notice when things are askewThese images do not look so very different from each other Margaret Thatcher IllusionMartians?"Face-ofgodinWTC on9-11?"Purdue UniversityPurdue UniversityFacial FeaturesAnother way to describe what happens is that people are unableto piece together the features and so do not detect the true facialexpressionThese images do not look so very different from each otherFacial AdaptationFurther evidence for representations of facial featurescome from face adaptation studiesPurdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology 1

Prof. Greg Francis 1/2/13Facial beautyFacial beautyAverage faces are almost always beautifulAverage faces are almost always beautifulPurdue UniversityPurdue UniversityFacial beautyFacial beautyAverage faces are almost always beautifulAverage faces are almost always beautifulhttp://www.faceresearch.org/demos/averagePurdue UniversityPurdue UniversityFacial beautyAveraging is not the only thing that mattersThe average of attractive faces is more attractive than the average of unattractivefacesMore sophisticated averaging (morphing) to avoid blurringAverage of fourfaces rated as veryattractive"Average of four facesrated as veryunattractive"CosmeticsWhy do people (especially women) wear cosmetics? Averageness: smooths the face, which is a characteristic of average faces Symmetry: blemishes on one side are hidden from other Youth: coloring and blush Femininity: female attractiveness is related to femininity, which is related to some simplevariables (Russell, 2009)These two faces are identical except that the facial skin on the left isdarker than for the face on the right (eyes and mouth are identical)Purdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology 2

Prof. Greg Francis 1/2/13CosmeticsIt makes sense for people to interpret faces this way, because it reflects truedifferences between male and female facesMore so than men, the eyes and mouth of women tend to be darker than their skinCosmeticsContrast alone is enough to modify femininity (most neurons in the visual systemare sensitive to contrast)The two images have the same skin luminance, but the one on the right has darkermouth and eye regionsWomen wearing cosmetics have greater contrast than women not wearingcosmetics (Russell, 2009)Purdue UniversityPurdue UniversityFacial agnosiaSome people (maybe 3% of the population) cannot recognize faces or reademotions from facesFamous faces test:Prosopagnosiahttp://www.faceblind.org/facetests/index.phpMany people have it all their lives but do not realize it Recognize people by characteristics other than facesCannot recognize themselves in a photograph by face aloneThey know there are different faces, but they all look the sameWith fMRI, we can identify placesin the brain that responddifferently to faces compared toother stimuliThis image looks like a youngman with a cap in profile There is a set of images like this(called Mooney images)FMRI of facesPurdue UniversityPurdue UniversityInverting the image changesthe percept dramaticallyWhat happens in the brain?fMRI scan while viewing eachtypeSubtract responses to find thedifference map There should be activity thatis present when seeing a facebut not present when notseeing a face Activity at other visual areasshould be similarFMRI controlfMRIStronger responses to facesFusiform Face Area (FFA) Region in the fusiform gyrus Region of the temporal lobe Especially in right hemispherePurdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology 3

Prof. Greg Francis 1/2/13Face area?However, these results do not necessarily mean the FFA is responsible fordetecting facesHumans are very practiced at detecting faces, maybe the FFA is involved inpractice and expertise Gauthier et al. (2000) Participants see a sequence of objects (cars, faces, birds, other), judge similarity acrosstrials Participants include car and bird experts (everyone is a face expert)Face area?Bird and car experts show extra activity in right hemisphere FFA whenjudging faces or judging items in their area of expertise Note: the left side of the scan image is the right hemisphereSo faces may be special only because we practice a lotPurdue UniversityPurdue UniversityGrandmother cells?We earlier dismissed the idea that responses from a single cell mightcorrespond to seeing your grandmotherHowever, this does not mean that some cells may not be quite specific intheir responsesSome epileptic patients have electrodes implanted in their brains to helpcontrol their seizures These electrodes can also be used to record responses to different stimuli Quiroga et al. (2005) made such recordings in the left posterior hippocampusfrom eight patientsJennifer Aniston NeuronSome cells respond to very specific sets of stimuliPurdue UniversityPurdue UniversityPitt-Aniston NeuronSome cells respond to very specific sets of stimuliHalle Berry NeuronSome cells respond to very specific sets of stimuliPurdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology 4

Prof. Greg Francis 1/2/13Sydney Opera House NeuronSome cells respond to very specific sets of stimuliSydney Opera House & Pitt-Aniston NeuronBut lots of cells respond more generallyPurdue UniversityPurdue UniversityGrandmother cellsThese results indicate that a neuron may have great selectivity in itsresponsesHowever, the researchers could not test all possible images. Maybe theJennifer Aniston cell would also respond to other images that were nottestedThus, it may be a mistake to suppose that there is a single cell whosefiring corresponds to thinking about Jennifer AnistonPurdue UniversityConclusionsFace detectionFeatures and illusionsEmotionsBeautyAgnosiaExpertiseGrandmother cellsPurdue UniversityNext timeAttentionWhat is attention?What does it do?How could you not see it?Purdue UniversityPSY 200: Intro. to Cognitive Psychology 5

Prof. Greg Francis 8/12/11AttentionPSY 200Greg FrancisLecture 12How could you not see it?Attention!! The world contains more information thanwe can fully interpret or process all at once!! The ability to deal with some stimuli andnot others is attention!!not clear if there is an attentive system!!or if attention derives from other systemsPurdue UniversityPurdue UniversityInformation processing!! Modern theories see cognition as information processing!! much like a computer!! Different systems have different capabilities, capacities, and speeds!! Necessarily, some information is ignored because it is notprocessedAttention!! Part of attention seems to be due to mental effort onyour part!! attending a lecture!! ignoring whispering around you!! Part of attention seems a natural side effect of mentaleffort!! ignoring the “uhs” and “ums” from a speaker!! ignoring the feel of clothes on your body!! Part of attention seems effortless!! a loud noisePurdue UniversityPurdue University!! Pick any card!! Memorize itMagic trick!! really study the card carefully to be certain you have itmemorizedMagic trick!! Now the computer will shuffle the cards and presentthem againPurdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology 1

Prof. Greg Francis 8/12/11Magic trick!! Hey, the card you selected is missing!! How did I know which card you would select?Attention as processing!! Mental effort either is or requires attention!! Consider the following video (Simons & Chabris, 1999)!! Try it on the web at http://members.tripod.com/~andybauch/magic.htmlPurdue University!! you will see twogroups of peoplewearing either blackor white shirts!! each group istossing around aball!! count how manypasses are madeby the white groupPurdue UniversityAttention as processing!! Did you notice an unusual event in the video?!! Let’s look at it again!! Attention is a powerful force!! when you are busy doing something else, you can failto observe even very significant things!! that’s why it is not a good idea to talk on a cell phonewhile driving a carDrawing attention!! Attention is focused by meaningful stimuli!! Also focused by certain stimulus characteristics,especially changes!! Flashes of light!! Movement!! Color!! Think of advertising signs!! ExamplePurdue UniversityPurdue UniversityDrawing attention!! Suppose these cues were masked by otherchanging stimuli!! You might not notice the change at all!! DemonstrationAttention!! Another way to interpret these results is thatattention is necessary to detect stimuluschanges!! Explains how people can “look” but not “see”!!walking into doors!!driving into trains!!detecting changes on a radar screen!!why magicians use flashes of light!Purdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology 2

Prof. Greg Francis 8/12/11Attention effects!! An illusion because people do not see stimuli orsomething else?!! E.g., do not remember, do not recognize!! In general, attention seems to strengthenstimulus representations if they are weak!! Perceptual processing is better with attentionalfocus!! Detection!! Masking!! Discrimination!! Some neurons respond stronger to stimuli thatare attendedAttentional paradox!! If attention strengthens perceptual representations, we shouldlose perceptual veridicality!! We might expect what is schematized below!! But we normally do not experience this!! Attention generally seems to strengthen information about astimulus that is not perceptual!! Maybe edges rather than surfaces?StimulusAttend redAttend greenReynolds et al., 2000"Purdue UniversityPurdue University!! Sometimes attentioncan change perceptualproperties!! But then we have anincorrect perception ofthe properties of thevisual scene!! So it is difficult tounderstand howattention is helpinghereAttentional paradoxMore demos!! If time permits, here’s some more demos!! http://viscog.beckman.uiuc.edu/djs_lab/demos.html!! Field!! Living room!! Phone call!! Lunch conversation (9 changes)!! Paris sceneTse, 2005"Purdue UniversityPurdue UniversityConclusionsNext time!! Attention can have very powerful effects!!help processing of focused on things!!can cause unawareness of unattended things!! Not precisely defined!!characteristic of processing?!!An “extra” system?!! Methods of studying attention!! What things influence attention!!Timing, features!! CogLabs on Attentional blink and Visualsearch due!!! Should you pay $59.95 for Mega-speedreading?Purdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology 3

Prof. Greg Francis 1/2/13AttentionPSY 200Greg FrancisLecture 13Should you pay $59.95 forMega-speed reading?Purdue UniversityAttentionWe saw last time that attention can havevery powerful effects when it is focused on one thing, you ignoreother thingsToday we want to consider some morespecific properties of attention and look at experimental methods that areused to study attentionPurdue UniversityCharacteristics of attentionBy identifying the properties andcharacteristics of attention we can deduceproperties of the underlying systems thatare involved in cognition whether attention is thought of as a system or as a by-product of other systemsLook at temporal featuralPurdue UniversityAttentional blink Suppose you have to identify rapidly presented (100ms) letters e.g., detect J and/or K in a stream of lettersMPKRWSPurdue UniversityAttentional blinkTurns out that detection of first letter tends to makedetection of the second letter very difficult if it immediately follows the first Attentional blinkMPKRJSAttentional blinkMeasure frequency of detection class data (183 observers)Implies thatdetecting thefirst lettercausesyou to miss thesecond letter!Purdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology 1

Prof. Greg Francis 1/2/13Attentional blinkSuggests that processing the first targetletter, J uses up resources that would otherwise be usedto process second target letter, KSpeed readingEver seen the ads for speed reading?One is for a program called Mega Speed Reading claims to teach you to read 25,000 words per minuteJ attentional focus and refocus takes time and forthis task takes approximately 400 msSLNBJRKH...Once processing is done (afew hundred milliseconds),there is no difficultydetecting second letter.Purdue University ==> 2.4 milliseconds for each word impossible, if only by attentional blink!The seller is skimming at best,and lying at worstPurdue UniversityVisual searchVisual searchProofreadingMuch of our time is spent looking for various thingsTime and type of processing can also be measured byhaving observers respond as quickly as possible whenthey detect a target Staplers, Road signs hypothesize that tasks that involve attention will beslower than perceptually based tasks expect attention to depend on the number of things thatmust be searchedTypically, we distinguish between a target anddistracters by one or more features we vary the number of distracters and measure reaction timePurdue UniversityPurdue UniversitySet SizeSet SizeIn some situations, more distracters make search takelongerIn some situations, more distracters make search takelonger fast slowPurdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology 2

Prof. Greg Francis 1/2/13Feature SearchFeature SearchBut if the target differs from the distracters in the right way,search can be fast even with lots of distractersBut if the target differs from the distracters in the right way,search can be fast even with lots of distracters pop out pop outThis often happens when the target has a unique featureThis often happens when the target has a unique featurerelative to the distractersrelative to the distracters shape colorPurdue UniversityPurdue UniversityConjunctive SearchVisual search experimentBut if the target has shared features with differentdistracters, search is difficult No pop outThis often happens when the target is defined by aconjunction of features relative to the distracters Orange rectangle: color and shapeFour types of responses 1) Feature - present (can respond as soon as see target) 2) Feature - absent (must examine all stimuli before suretarget is not present) 3) Conjunctive - present (can respond as soon as seetarget) 4) Conjunctive - absent (must examine all stimuli beforesure target is not present)Purdue UniversityPurdue UniversityCogLab feature searchCogLab feature searchFew distracters -easyMany distracters - still easyPurdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology 3

Prof. Greg Francis 1/2/13Conjunctive searchConjunctive searchFew distracters - fairly easyMany distracters - difficultPurdue UniversityPurdue UniversityVisual searchCogLab data (182 observers)InterpretationFeature maps: color, shapeFeature search can identify target within either feature mapColorShapebluecornersgreenarcsNo searchingrequired!"Purdue UniversityPurdue UniversityFeature maps: color, shapeInterpretationFeature maps: color, shapeInterpretationFeature search can identify target within either feature mapConjunctive search cannot identify target within either feature map aloneColorShapeColorShapebluecornersbluecornersgreenarcsNo searchingrequired!"greenarcsRequires searchby comparisonacross featuremaps.Serial processthat takes timePurdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology 4

Prof. Greg Francis 1/2/13bluegreenFeature maps: color, shapeInterpretationConjunctive search cannot identify target within either feature mapColorShapecornersarcsPurdue UniversityRequires searchby comparisonacross featuremaps.Serial processthat takes timeConjunctive search fortarget absent has aslope twice as steep asfor target present Because when thetarget is present youfind it, on average,after searching half theitems and then canstop the search For target absentsearches, you mustsearch all items toverify each is not thetargetVisual searchPurdue UniversityAutomaticityWhen a task is unfamiliar itseems to require a lot ofattention to performLater it requires less attention riding a bike driving a car typing tying shoelaces» http://www.fieggen.com/shoelace/knots.htmPurdue UniversityAutomaticityThe process whereby a task goes fromrequiring a lot of attention to requiring little iscalled automatizationMany tasks are automatizable color naming word namingCan measure effects by pitting anautomatized task against a non-automatizedtaskPurdue UniversityStroop task Stroop (1935)Identify the color of ink for wordsStroop effectIt takes longer when the words are colornamesDemonstration measure reaction timePurdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology 5

Prof. Greg Francis 1/2/13Stroop effectStroop effectWord name interferes with ink colornaming ink color does not generally interfere withword naming lots of studies on Stroop effectMany effects that are similar to it Simon effect for pointingYou can try them both on CogLab Not required, no creditPurdue UniversityPurdue UniversityExplanationWord reading is well practiced especially among college undergraduatesConclusionsMethods of studying attention attentional blink so it occurs quickly and is automaticColor naming is unpracticed, so it occursslowly and requires attention visual search Stroop taskCharacteristics of attention timingWith two tasks, both trying to report on a color the automatic one tends to mess up theunpracticed one, it takes more mental effort (andtime) to do the unpracticed task role of perceptual featuresAutomaticityPurdue UniversityPurdue UniversityNext timeIntersection of attention, perception, and memory Iconic memory echoic memorySerial position curvesCogLab on Partial report due!Why telephone operators seem rude.Purdue UniversityPSY 200: Intro. to Cognitive Psychology 6

Prof. Greg Francis 8/12/11Sensory memoryPSY 200Greg FrancisLecture 14Why telephone operatorsseem rude.Purdue UniversityMemory!! Humans demonstrate memory when theybehave in a way that could only be basedupon previous experience!!does not necessarily imply that there arememory systems!! Memory could be a by-product of othersystems (vision, audition, language,…)Purdue UniversityPerception to memory!! Suppose you want to know how muchinformation is available in a single visualglanceWhole Report!! Write down as many letters as you see!! How would you measure it?!! It turns out it’s a complicated taskbecause it involves perception, attention,and memoryPurdue UniversityPurdue UniversityWhole ReportWhole Report!! Write down as many letters as you see!! Write down as many letters as you seeE G M BT W V XP F A SPurdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology 1

Prof. Greg Francis 8/12/11Whole report results!! Subjects report 4.5 letters onaverage (Sperling, 1960)!! Subjects claim they saw moreletters, but lost the percept whilethey reported!!they cannot report fast enough!! How can we tell if percept is lost?Purdue UniversityPartial report!! Same type of letter matrix!! Indicate which row to report after the matrixdisappears!! choice of row is random!! Suppose the subject reports 3 of 4 letters fromany row!! =>3/4ths of each row was available!! ==> entire field was available!! This is essentially how college tests aredesigned!Purdue UniversityPartial reportPartial report!! Write down letters from the indicated row!! Write down letters from the indicated rowPurdue UniversityZ K O LU Y V NC S D PPurdue UniversityPartial reportSensory memory!! Write down letters from the indicated row!! Temporal characteristics!! delayPurdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology 2

Prof. Greg Francis 8/12/11Sensory memorySensory memory!! Temporal characteristics!! delayC R H JO K Q EN S D MPurdue University!! Temporal characteristics!! delayPurdue University!! Temporal characteristics!! delaySensory memoryResults!! Vary delay to watch decay from memory!! compare to CogLab dataNumber of letters correct432100 0.2 0.4 0.6 1Delay of tone (seconds)Purdue UniversityPurdue UniversityResults!! CogLab data (187 participants)Results!! We can test on any row and get essentially the sameresult!! so, the number of letters that actually persist and is available isfound by multiplying by the number of rowsPurdue UniversityNumber of letters available10864200 0.2 0.4 0.6 1Delay of tone (seconds)Purdue UniversityPSY 200: Intro. to Cognitive Psychology 3

Prof. Greg Francis 8/12/11!! Performance is betterthan the whole-reportIconic/sensory memoryprocedure because youcan focus attention onjust one row before thepercepts fade away!! Fading percepts are dueto visual persistence,which we talked aboutearlier!! We call the use of thisinformation iconicmemory!! Large capacityNumber of letters available10864200 0.2 0.4 0.6 1Delay of tone (seconds)!! The partial-reportexperiment can bemodified in severalInfant iconic memory!! Short duration Purdue UniversityPurdue Universityways!! Blaser and Kaldy(2010) modified it totest iconic memoryof infants!! 60, 6 month oldinfants!! Infants show apreference to look atthe changed object forsmall enough set sizes!! Infant iconic memory isInfant iconic memoryactually quite similar to adults!! Ask adults to report thelocation of the changed coloritem!! They do better thaninfants (who did notunderstand the ‘task’)!! Look for sharp drop inperformance as set sizeincreases!! Estimate items in memoryMasking!! Masking effects can influence iconic memory!!persistence-based memory is very brief, and iseasily destroyed by a mask!! Iconic memory is!!brief!!easily disturbed!! Adults = 5.75!! Infants = 5.0Purdue UniversityPurdue UniversityPartial report with maskingPartial report with masking!! Write down letters from the indicated row!! Write down letters from the indicated rowPurdue UniversityI J T RH V P LC G K MPurdue UniversityPSY 200: Intro. to Cognitive Psychology 4

Prof. Greg Francis 8/12/11Partial report with masking!! Write down letters from the indicated rowX X X XX X X XX X X XPurdue UniversityMasking!! With the mask you do not have enough time to focusattention on the indicated row!! Iconic memory is so brief (less than a second) that itprobably has little to do with “normal” memory!! Too brief to be useful for many situations (except maybeknowing how to reach for something just after lights go out)!! Other similar systems are more notablePurdue UniversityEchoic memoryEchoic memory!! Other senses have a similar type ofpersistence or sensory memoryASWFHVDEX!! Properties!!Longer duration(seconds)!!Smallercapacity!! Significant forsome memorytasksNumber of lettersavailable5432100 1 2 3 4Delay (seconds)Purdue UniversityPurdue UniversityImmediate serial recall!! After given a list of items!!e.g., digits, letters, words,…!!subject must report them back!!1) no delay (immediate)!!2) in the correct order (serial)!!3) no cues (recall, not recognition)!! Plot percentage correctly recalled againstposition of item in listPurdue UniversitySerial position curve!! Often, subjects recall first and last itemsbestProbability of recall10.80.6 PrimacyRecency0.40.201 2 3 4 5 6 7Position in listPurdue UniversityPSY 200: Intro. to Cognitive Psychology 5

Prof. Greg Francis 8/12/11Modality effect!! Recency depends on the modality ofpresentationProbability of recall10.80.6Auditory0.4Visual0.201 2 3 4 5 6 7Position in listPurdue University!! Explanation:Modality effect!! In this task, recency depends on sensory memory!! It takes time to report all the items in the list, in order!! in the visual presentation, iconic memory of the last itemis gone before subject tries to report it (poor recall)!! in the auditory presentation, echoic memory of last itemis still present when subject tries to report it (good recall)!! Thus, auditory presentation shows recency, butvisual does not!! We will explain the primacy effect laterPurdue UniversitySuffix effectSuffix effect!! Auditory presentation only!! Cue to report is either a word or a tone24168792, 4, 1, 6, 8, 7, 9Purdue University!! Recency when cue to report is a tone!! Loss of recency when cue to report is a wordDemonstrationProbability of recall10.80.60.40.201 2 3 4 5 6 7Position in listPurdue UniversityToneWord!! Not affected by!!practice!!meaning of cue word!!common vs. rare wordSuffix effect!! Words are physically different from tones!!suffix word acts like a mask to wipe out last word inlist from echoic memory!!the situation is similar to being unable to report theletters in the partial report task with the X-masksPurdue UniversityPhone operators!! Call information for a number!! Operators are very short!!give the number!!do not say “goodbye” or “have a nice day”!! Avoiding the suffix effect!!!you would forget the last part of the phonenumber if they finished with pleasantriesPurdue UniversityPSY 200: Intro. to Cognitive Psychology 6

Prof. Greg Francis 8/12/11Conclusions!! Partial report experiment!! Sensory memory!! iconic memory (visual)!! echoic memory (auditory)!! Relation to immediate serial recall (recency)!! modality effect!! suffix effect!! significance for phone operatorsNext time!! Memory!! Modal model!! short term memory!! long term memory!! Experiments!! CogLabs on Brown-Peterson and Serial position due!!! Why it is difficult to win a pizza at Little Caesars.Purdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology 7

Prof. Greg Francis 8/16/11Two store modelPSY 200Greg FrancisLecture 15!! Simple viewMemory!!memory is a container of past impressions andknowledge!!memories can leak-out, decay away!! Not very realistic!!need to explain why memories disappearWhy it is difficult to win a pizza at LittleCaesar’s.Purdue UniversityPurdue UniversitySimple view!! The container theory of memory does not explain, forexample,!! why some memories are very long lasting (my childhood cartrips to Utah)!! why some memories are very brief (my wife asks me to takeout the trash)!! We are not going to get a full theory of memory, butwe can start to get an outline!! and identify some misconceptions about memoryEbbinghaus’ experiments!! First memory experiment (1885)!! Measure how long it takes to learn a list of nonsensesyllables perfectly!! NOF, QAP, HOS, LEQ, FIK, MEC, KIJ, HOM, NEM, MOJ!! How long does the memory last?!! In what form does the memory last?!! How does it affect future behavior?!! Does it help relearn the list at a later time?Purdue UniversityPurdue UniversityEbbinghaus!! Relearn the list at later points in time!!a different list each time!! Measure how long it takes to relearn the list!! Calculate savingsSavings = TimeoriginalTime! TimeoriginalPurdue UniversityrelearnForgetting curve!! Savings =1!! subjects do not need to relearn, perfect memory!! Savings=0!! subjects show no evidence of earlier learningSavings1.210.80.60.4Still not 0!0.200 5 10 15 20 25 30Time (days)Purdue UniversityPSY 200: Intro. to Cognitive Psychology 1

Prof. Greg Francis 8/16/11Significance!! Ebbinghaus’ results suggest that memories canlast a very long time, in some form!! Memories were believed to be “stored” in a memorysystem and did not just fade away (otherwise, thecurve should not asymptote above zero)!! Memory loss was believed to be due to interference ofother memories!! Other experiments challenge this viewMemory task!! See (or hear) a trigram of consonants!! Report it back in order!! Ebbinghaus’ results suggest goodmemory until other letters are alsomemorizedWRMPurdue UniversityPurdue UniversityRetention!! Peterson & Peterson (1959)!!Brown (1958)!! Give subjects trigram!!ask them to count backwards by 3’s andthen recall trigram779, 776, 773,...WRM 782Purdue UniversityRetention!! Vary duration of counting backward!! Numbers are different from letters, youmight not expect any interference!!but they can have very strong interferenceSuggestssome memorieslast only afew seconds!Proportion correct10.80.60.40.200 3 6 9 12 15 18Retention interval (seconds)Purdue UniversityRetention!! The results of the Brown-Peterson studysuggest that some aspects of forgettingare process driven!!keeping a memory “active” requires effort!!if you are distracted by another task, youcannot apply the effort to keep the memory!!similar to our observations about attentionand processingRetention!! The results of the Brown-Peterson study alsosuggest that some aspects of forgetting arepassive!! even if you are distracted, you can recall the trigram ifonly a short time has passed!! if many seconds have passed, while you aredistracted, you cannot recall the trigram!! memory has “decayed”, or something like decay, whileyou were doing the distracting taskPurdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology 2

Prof. Greg Francis 8/16/11!! CogLab dataRetentionAnother experiment!! 179 subjects!! Not everyone shows interference on this taskPurdue University!! Memory span!!how many items canyou correctly recallimmediately afterexposure?!! “The magic number7+/-2:…”!!Miller (1956)95 78 6 34 9 2 75 9 0 1 36 8 2 9 3 47 2 5 1 6 9 39 1 0 3 5 7 3 23 9 0 5 7 4 2 1 6Purdue UniversityInterpretation!! There exist two types of memory systems!! Long Term Memory (LTM)!! high capacity (no limit)!! long duration (forever)!! Ebbinghaus’ experiment!! Short Term Memory (STM)!! small capacity (~7 items)!! short duration (seconds)!! Memory span, Brown-PetersonPurdue UniversityPizza!! There used to be a LittleCaesar’s in W. Lafayette with agame where you could win apizza!! must repeat a sequence of flashinglights!! The sequence gets longer until youmake a mistake!! need a sequence length >7 to winmuch»! nearly impossible with STMproperties!! http://www.freegames.ws/games/kidsgames/simon/simon.htmPurdue UniversityModal Model of Memory!! Atkinson & Shiffrin (1968)!! Multiple stages of memory!! STM plays a dominant role in active memory!! Requires transfer between STM (STS) and LTM (LTS)Modal Model of Memory!! When something is memorized!! Items are first held in STM (temporary store)!! Items may transfer to LTM (permanent store)!! Takes time to transferPurdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology 3

Prof. Greg Francis 8/16/11Free Recall Serial Position Curve!! Given almost any list of items!! Subjects remember the first and last few itemsbest (free recall, not immediate serial recall)1Probability of recall0.8RECENCYPRIMACY0.60.40.201 2 3 4 5 6 7Position in listPurdue UniversitySerial position curve!! The effect of position is robust across many types of lists!! words!! letters!! numbers!! pictures…!! Here’s theCogLab data!! Demo!! (171 subjects)Percent correct"Position in list"Purdue UniversitySerial position curve!! In some situations the serial position curve canbe explained by different properties of STM andLTMProbability of recall10.80.60.40.20PRIMACY:Use LTMRECENCY:Use STMNEITHER LTM NOR STM1 2 3 4 5 6 7Position in listPurdue UniversityConclusions!! Short Term Memory (STM)!! Long Term Memory (LTM)!! STM / LTM distinction is one of thestrongest conclusions of cognitivepsychology!! Accounts for quite a bit of data!! Many details are unresolvedPurdue UniversityNext time!! Expansion of STM into!! Working memory!! central executive!! phonological store!! visuo -spatial sketchpad!! CogLab on Sternberg search due!!! Why there is a gate at the first floor stairway inthe Psych building.Purdue UniversityPSY 200: Intro. to Cognitive Psychology 4

Prof. Greg Francis 8/16/11Working memoryPSY 200Greg FrancisModal Model of Memory!! Atkinson & Shiffrin (1968)!! Today we focus on the Short-term store (Short termmemory)Lecture 16Why there is a gate at the first floorstairway in the Psych building.Purdue UniversityPurdue UniversitySearch of memory!! How is memory searched?!! Sternberg hypothesized three types of searches!! Explore by varying the number of items in memory set5 3 2 9(similar to visual search experiments)!! measure reaction time!!Sternberg (1969)5 3 2 95 3 2 9Purdue University8NOTypes of searches!! (1) parallel: target item is compared to all the items inmemory at the same timeNO!! the answer (yes or no) is returned after all items have beenchecked8 8 8 8885 3 2 98 8 8 88Purdue UniversityTypes of searches!! (1) parallel: target item is compared to all the items inmemory at the same timeYES!! the answer (yes or no) is returned after all items have beenchecked3 3 3 3335 3 2 93 3 3 33Purdue UniversityReaction time isthe same for ayes response!! If parallel searchMemory search!! number of items doesnot matter!! Yes and No responsesare both flatReaction timeSet sizePurdue UniversityPSY 200: Intro. to Cognitive Psychology 1

Prof. Greg Francis 8/16/11Types of searches!! (2) serial terminating: target item is compared to eachitem one after the other!! the answer (yes or no) is returned after the target is found orall items are searchedTypes of searches!! (2) serial terminating: target item is compared to eachitem one after the other!! the answer (yes or no) is returned after the target is found orall items are searchedNO5 3 2 9YES5 3 2 9Reaction time isfaster for a yesresponse8888 8 8883333 33Purdue UniversityPurdue UniversityIf self-terminating searchTypes of searches!! (3) serial exhaustive: target item is compared to each!! Go through itemsone-by-one until findtarget!! RT increases with setsize!! YES RT’s shorter thanNO RT’s!! Lines have differentslopesReaction time70060050040030020010001 2 3 4 5 6Set sizeYESNOitem one after the otherNO!! the answer (yes or no) is returned after all items are searched(regardless of whether target is found or not)8885 3 2 98 8 888Purdue UniversityPurdue UniversityTypes of searchesIf exhaustive search!! (3) serial exhaustive: target item is compared to eachitem one after the otherYES!! the answer (yes or no) is returned after all items are searched(regardless of whether target is found or not)3335 3 2 93 33 33Reaction time isthe same for ayes response asfor a noresponse!! Go through everyitem and then reportanswer!! RT’s increases withset size!! YES RT increases thesame as NO RT’s!! Lines are parallelReaction time70060050040030020010001 2 3 4 5 6Set sizeYESNOPurdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology 2

Prof. Greg Francis 8/16/11Hypothetical searches!! So, we have three hypothetical ways of searchingSTM!! They predict very different patterns of reaction time as afunction of memory set size!! Sternberg runs the experiment to see how the datacomes out!! You ran a version of the experiment in CogLabSearch of memory!! Sternberg’s data support exhaustivesearch!! Sternberg’sdata supportexhaustivesearch!! Here’s theCogLab data(182participants)=“Yes”=“No”Purdue UniversityPurdue UniversitySearch of memory!! Implications: Search of STM!! 1) is serial, one item at a time»! and checking each item takes approximatelythe same length of time»! Approximately 40 milliseconds (CogLab data isa bit slower, 50 milliseconds)!! 2) is exhaustive»! search always goes through all itemsSearch of memory!! These results were a bombshell in 1969!! finer analysis of cognition than anyone expectedwas possible!! used a thought experiment about different types ofsearches to generate precise testable predictionsabout cognition»! subsequent research found that there were othertypes of searches that complicate the conclusions!! counter-intuitive finding»! why should search be exhaustive?»! seems inefficient!Purdue UniversityPurdue UniversityInterpretation!! Exhaustive search makes sense if search of STMis done by some process that is!! very efficient (can search very quickly)!! dumb (doesn’t bother to stop itself)!! initiated by some other system (a controller)STMControllerSearchmemoryprocessPurdue UniversityController!! Controlling attentional system!! supervises!! coordinates!! starts and stops relatively independent processes!! e.g.!! Search short term memory!! Search long term memory!! walking down stairs!! gate in psychological sciences building!! DoorsPurdue UniversityPSY 200: Intro. to Cognitive Psychology 3

Prof. Greg Francis 8/16/11Other aspects of STM!! At about the same time, another studyindicated important characteristics ofphonological and visuo-spatial systems!! Brooks (1968)!! two types of tasks (visuo-spatial and phonological)!! two types of responses (visuo-spatial andphonological)!! Identifies two types of systems that arerelatively separate!! A complicated experiment!! Part 1: spatial mental task(diagrams)!! visual imagerySeparate systems!! classify corners (top or bottomcorner?)!! “yes” if top or bottom!! “no” if not top or bottomFPurdue UniversityPurdue University!! Part 2: verbal mentaltask!!read sentenceSeparate systems!!categorize words (nounor not?Purdue UniversityA bird inthe hand isnot in thebush.!! Either!!verbally!!spatiallyTwo response typesYES NOYES NONO YESYES NONO YESPurdue Universityno, yes, no,no, yes, no,...Results!! Measure time to finish mental task for eachresponse type!! diagrams -- pointing!! sentence -- pointing!! diagrams -- verbal!! sentence -- verbal!! ResultsResults!! when you have to respond by pointing, it is easier towork with sentence information than diagram information!! when you have to respond verbally, it is easier to workwith diagram information than sentence informationResponse taskPointing"Verbal"Mental taskDiagrams"28.2s"11.3s"Sentences"9.8s"13.8s"Purdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology 4

Prof. Greg Francis 8/16/11Significance!! The results suggest that there are tworelatively separate systems!!one deals with visuo-spatial information andmust do the pointing response and mentaldiagram task!!one deals with verbal information and must dothe spoken response and the sentence taskInterference!! These system have only limited resources and capabilities!! Asking a system to do two things at once (e.g., pointing andmental diagram) slows down the system!! Splitting responsibilities across the systems (e.g., spokenresponse and mental diagram) can be done quicklyDiagramsPointingVisuo-spatialinformationVerbalinformationSentencesVerbalDiagramsPointingVisuo-spatialinformationVerbalinformationSentencesVerbalPurdue UniversityPurdue UniversityAll together now!! Sternberg’s study suggests the existence of a “controller”that tells other systems what to do!! Brook’s study suggests separate systems that dealspecifically with viso-spatial and verbal information,respectively!! Baddley (1986) put these ideas together into a model ofworking memoryCentralexecutiveWorking memory!! Current thought, awarenessWorking memory"!! extension of short-term memory!! small capacity!! rapid forgetting!! Processor of information!! not a storage device!! hypothesizes mechanisms that lead to memory propertiesVisuospatialsketchpadPhonologicalloopPurdue UniversityPurdue UniversityConclusions!! Sternberg’s study!!controller system!! Brook’s study!!separate visual and verbal systems!! Baddley’s working memory model!!Central executive!!Visuo-spatial sketchpad!!Phonological loopPurdue UniversityNext time!! Properties of phonological loop!! Data!! phonological similarity effect!! articulatory suppression!! word length effect!! irrelevant speech effect!! CogLab on Memory span due!!! A problem with IQ tests.Purdue UniversityPSY 200: Intro. to Cognitive Psychology 5

Prof. Greg Francis 8/16/11Working memoryWorking memoryPSY 200Greg FrancisLecture 17A problem with IQ tests.Purdue University!! Current thought, awareness!!extension of short-term memory!!small capacity!!rapid forgetting!! Processor of information!!not a storage deviceVisuospatialsketchpad!!hypothesizes mechanisms that lead to memorypropertiesPurdue UniversityCentralexecutivePhonologicalloopPhonological loopLoop capacity!! Two components!! Articulatory control process(ACP)»! converts non-speechinformation into speechcode»! rehearsal / refresh!! Phonological store (PS)»! similar to how we firstdescribed STM (itemsdecay from memory)»! Refresh restarts the decayprocessPhonological loopACPPS!! How many items can bekept in the phonologicalloop?!! Depends on two factors!!Duration before decay fromPS!!Speed of rehearsal!! Spinning coins!Phonological loopACPPSPurdue UniversityPurdue UniversityMagic number?!! We earlier noted that memory span was about 7 items (+/- 2)!! The phonological loop suggests that it is not the number ofitems but their rehearsal duration!! To recall a list of items you must rehearse them all before anyof them fade!! The duration of decay in the PS!! Memory span should follow the equation!! Span = (Rehearsal Rate) X (PS decay time)!! Measure memory span (s): around 7 items!! Measure verbal rehearsal rate (r): around 4 items per secondfor English speakers!! Estimate duration of decay in PS (d)!! d=1.75 secondsPurdue University!! Capacity of theEffect of rehearsal ratephonological loop dependson the rate of rehearsal (r)!! A set of items that takeslonger to rehearse shouldbe harder to remember!! more likely that some itemswill drop out before you getback to the first itemPhonological loopACPPSPurdue UniversityPSY 200: Intro. to Cognitive Psychology 1

Prof. Greg Francis 8/16/11Effect of rehearsal rate!! Explains differences across groups of people!! Age effects in children!! Hitch, Halliday & Littler (1989)!! This implies that it is not theloop size that changes withage, but the rate ofrehearsalSpan=1.68(Oral Reading Rate)+0.71Word length effect!! Memory span isrelated to the lengthof words!! Number of syllables!! Nicely matched bychanges in readingspeed!! Rate of rehearsalPurdue UniversityPurdue University!! Somelanguages arespoken morequickly thanothers!! Should allowlarger memoryspan!! it doesLanguage effects!! Ellis & Henley (1980)Relation to IQ!! investigated complaints about WISCintelligence scores!! Welsch children tended to score lowerthan English children!! Part of the exam checks memory span!! and the slower rate of speech in Welsch partlyexplains the difference!! bilingual Welsch students tested in English got betterscores than when tested in WelschPurdue UniversityPurdue UniversityArticulatory suppression!! Subjects sees (hears) a list of phonemes!! Also repeats a phrase over and over!! e.g., “tippy-toe, tippy-toe, tippy-toe,...!! Recall is worse!! True for both auditory and visual presentation!! (Recall for visual may be better than auditory because there issome information in the visuospatial sketchpad as well)Proportion recalled10.90.80.70.60.50.40.30.20.10VisualAuditoryPurdue UniversitySilentSuppressionArticulatory suppression!! Repeating phrase ties up the ACP!!Without rehearsal more forgetting occursVisualPhonological loopAuditoryACPtippy-toePSPurdue UniversityPSY 200: Intro. to Cognitive Psychology 2

Prof. Greg Francis 8/16/11Phonological similarity!! Memory of a list of items is worsewhen the items sound the samePhonological similarity!! All items are stored in phonological loop!! similar sounding items interfere with each other in thephonological loop!! two possibilities:»! 1) harder to rehearse (effect in the ACP)»! 2) fade more quickly (effect in the PS)B F H NBG P TBetterrecallWorserecallPhonological loopACPNFPSH BPhonological loopACPG PBPSTPurdue UniversityPurdue UniversityPhonological similarity!! All items are stored in phonological loop!! similar sounding items interfere with each other in thephonological loop!! two possibilities:»! 1) harder to rehearse (effect in the ACP)»! 2) fade more quickly (effect in the PS)Phonological loopPhonological loopLocus of similarity effect!! Studies find a phonological similarity effect for auditory stimuliunder articulatory suppression!! We suggested two possibilities:»! 1) harder to rehearse (effect in the ACP)»! 2) fade more quickly (effect in the PS)!! Since the phonological similarity effect is there even when theACP is not involved, it must be possibility 2 (in the PS)ACPFPSH B NACPPS B GP TProportion recalledPurdue UniversityPurdue UniversityCogLab data!! The CogLab experiment on memory span shows datain agreement with our expectations (189 subjects)!! Item type Final list length!! Numbers ! ! !6.47!!! Letters that sound different !5.66!!! Letters that sound similar !5.01!!! Short words ! ! !5.20!!! Long words ! ! !4.17About 7Phonologicalsimilarity effectWord lengtheffectIrrelevant speech effect!! Does irrelevant “background” sound affectmemory?!!E.g., studying with the TV on!! Three groups of subjects recall consonants!!1) no backgroundbest!!2) background = nonsense words!!3) background = noise burstsworstPurdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology 3

Prof. Greg Francis 8/16/11Irrelevant speech effect!! The presence of phonemes in thebackground is critical to the effect!!strong effect when background is spoken inGerman, even for English speakers!! Suggests that background phonemesinterfere in the PS!! Study with classical music if you needsomething!Visuospatial sketchpad!! Briefly holds visual and spatialinformation!! Important for tasks that requireinteraction with theenvironment or searching!! Properties seem to vary acrossindividuals!! VSS of video gamers differs fromnon-video gamersVisuospatialsketchpadCentralexecutivePhonologicalloopPurdue UniversityPurdue UniversityVisuospatial sketchpad!! West et al. (2009)!! Participants identify whether or notthe 2 second display contains a nonmoving“swimmer”!! Video Game Players!! Minimum 3-4 days a week!! Action video game!! Minimum 2 hours each day!! Non video game players!! Little to no video game usage in thepast 6 monthsPurdue UniversityVisuospatial sketchpad!! Video game players do better on the detection task!! Both do similar on noting the absence of the targetPercent1009080706050403020100HitsCorrectrejectionsPurdue UniversityVGPsNVGPsConclusions!! Data accounted for by phonological loop!! word length effect!! phonological similarity!! articulatory suppression!! irrelevant speech effect!! Don’t listen to lyrical music while studying!! Classical music is fine!! Go ahead and play videogames!! There are limits, of courseNext time!! Review for Exam 2!! After exam 2!! Encoding specificity!! What to do if you are drunk while studying for anexam.Purdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology 4

Prof. Greg Francis 8/17/11Encoding specificityPSY 200Greg FrancisLecture 18What to do if you are drunk whilestudying for an exam.Purdue UniversityContext!! The context within which you learn and recallcan have a profound impact on your memory!! e.g., part-set cueing!! given part of a set that has been learned, subjects recallfewer of the remaining items than with normal free recall!! Interference of recall»! Have to keep checking if an item you recall isalready on the list!! DemonstrationPurdue UniversityContext!! The effect of part-set cueing suggests that to measure memoryyou must consider the conditions at test!! Memory is more often about discrimination of memory traces andnot about the strength of memory traces!! Similar to visual search experimentsPurdue UniversityFeaturesearchConjunctivesearchContext!! But memory is not exactly the same as visual search!! Information must be encoded in memory as well asrecalled!! Such encoding can alter what features are stored as part of the memory!! Which changes the discrimination of subsequent recall!! It turns out, that to maximize recallability!! the effort and conditions at the time of learning must beconsistent with the properties and conditions of the test!! Encoding specificity principlePurdue UniversityRepresentative study!! Subjects in two groups!! see the same words, but have different tasks!! This changes the encoding of information in memorySemantic judgementCHEESEThe manthrew the ballto the______.Rhyme judgementPurdue UniversityCHEESESNEEZETest!! Each group is then split into two subgroups that vary in the recall ofinformation!! 1) Normal recognition task!! 2) Shown a word and asked if any of the target words rhymed with thisword987Study65Semantic4Rhyme3210Number of wordsrecognizedStandardRhymeRecognition testPurdue UniversityPSY 200: Intro. to Cognitive Psychology 1

Prof. Greg Francis 8/17/11Significance!! It is not that one learning strategy is better thanthe other!! or that one testing strategy is better than the otherEncoding specificity!! Memory is better when the cues available duringrecall match the cues available during encoding!! Tulving & Osler (1968)!! Encoding (learning) of information and recall ofinformation need to matchCueNo cue!! This means it is very difficult to test forabsolute memory!! subject’s performance depends on many factorssoar-EAGLEHEALTH-nurseWHEEL-autopaper-BOOK...EAGLEHEALTHWHEELBOOK...Purdue UniversityPurdue UniversityEncoding specificity!! Each group split into two groups for testingCuerecall!! cue is always related to target wordsoarnurseautopaper...No cuePurdue University!! ResultsWords recalledEncoding specificity!!cue does not always help1614121086420No cueCueTest displayPurdue UniversityStudy no cueStudy cueSurrounding context!! Deep-sea divers learn words (Godden& Baddeley, 1975), either!!on land (dry)!!under water (wet)Surrounding context!! Test for recall on land (dry) or under water (wet)StudyEAGLEHEALTHWHEELBOOK...Purdue UniversityEAGLEHEALTHWHEELBOOK...Purdue UniversityPSY 200: Intro. to Cognitive Psychology 2

Prof. Greg Francis 8/17/11Significance!! Decompression tables for divers!! want to remember when under water!! generally study while on land!! Researchers working under water have difficultyrecalling their details on land!! E.g., counts of species!! How do you know if something is forgotten?!! changing context may allow subject to recall seeminglyforgotten information!! forgetting = retrieval problem?Purdue UniversityForgetting!! Forgetting is not always a characteristic of amemory system, or your brain!!although it could be in some cases, it is notalways!! Forgetting must be defined operationally!!specify the task and context of retrieval!!You can never be certain that if you are placedin a different context you will still show forgettingPurdue University!! Goodwin et al. (1969)Internal context!! Subjects drink 10 oz. 80 proof vodka, mixed insugar-free lemon-lime drink!! or a similar tasting drink (bit of vodka on top of drink)!! Test memory 24 hours laterInternal context!! Best recall if study and test states are the same!! Similar effects for marijuana cigarettes (Eich et al., 1975)Study state!! Subjects are either:!! Sober at study, sober at test!! Drunk at study, sober at test!! Sober at study, drunk at test!! Drunk at study, drunk at testPurdue UniversityPurdue UniversityTesting!! So, if you are intoxicated while studying for anexam!!and you didn’t study before!! You should be intoxicated while taking theexamMood!! Mood has a similar effect (Eich et al, 1994)!! Mood induced by music and directed thoughtsStudy moodBut do notexpect to dovery wellGET HELP!Purdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology 3

Prof. Greg Francis 8/17/11Classrooms!! Is memory better when you are tested in the same room aslectured?!! significant for final exams!!! Smith et al. (1978)!! Subject studied words in one of two contexts (on separate days)!! Varied classroom and dress of experimenterClassrooms!! Subjects recall all words either in the context of Day 1 or Day 2(different contexts for different subjects)!! Recall was best for words that were studied in the test contextStudy contextPurdue UniversityPurdue UniversityConclusions!! Context!! Encoding specificity!!memory best if study and test are similar!! Cues!! Environment!! StateNext time!! Discrimination in memory!! Proactive interference (PI)!! Release from PI!! CogLab on False memory due!!! How to take a test.!! Mood!! ClassroomsPurdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology 4

Prof. Greg Francis 1/2/13Memory discriminationPSY 200Greg FrancisLecture 19DiscriminationMany cognitive tasks require you todiscriminate between events/stimuli Is this a real smile? Is this fruit ripe? Is the stapler on the desk?The same kind of discrimination is required for memoryHow to take a test.Purdue UniversityPurdue UniversityDiscriminationFalse memoryDiscrimination is difficult because memoriescan come from lots of different sourcesConsider so-called False memory studies as in CogLab subject views a list of words the list of words have something in common» they are all related to a target wordAn example list is smooth, bumpy, road, tough, sandpaper, jagged, ready,coarse, uneven, riders, rugged, sand, boards, ground,gravel the special target is rough, which is not shown to thesubjectAfter viewing the list, the subject must go through aset of words and identify which ones were in thejust seen list some words were in the list some words were not seen» including the special targetPurdue UniversityPurdue UniversityFalse memoryFalse memoryThe main finding is that the special target is oftenidentified as part of the just seen list even though it was notSometimes people will even report that theyrecall seeing the special target but this is impossible because it was never shownCogLab data (193 participants) Type of selected items ! !Percentage of recalls! In original list ! ! !75.5! Normal distractor (not in list) ! 7.4! Special distractor (not in list) !70.3!Purdue UniversityThese types of findings suggest that ourmemories are not necessarily accurate, we can remember things thatnever occurred able to be manipulated, to a certain extent, I can makeyou have certain memoriesWhy does the false memory effecthappen?Purdue UniversityPSY 200: Intro. to Cognitive Psychology 1

Prof. Greg Francis 1/2/13False memoryWith every to-be-remembered item you store someinformation, but not only information about the item Other information is automatically generated as well smooth, bumpy, road, tough, sandpaper, jagged, ready,coarse, uneven, riders, rugged, sand, boards, ground, gravelMemory: Item 1 Memory: Item 2 Memory: Item 3smoothroughbabybumpysmoothroughbabyroadPurdue Universitybumpysmoothroughtoughbaby nailsroadFalse memoryAt the end of the trial, you have a lot of items in memory thatare related to the list Some of them are items that were actually on the list and some of them are items thatwere generated but not actually on the list Reporting all items from memory is not going to lead to good performance Both types of memory items are real, but only one type matches the physical stimuliGood performance on this task requires discrimination betweenmemories generated by physical stimuli and memories generated byinternal processes Source monitoringjagged stonesbumpysmooth peaks Memorytoughroughsandpaperbaby nailsgravel roadunevenPurdue UniversityDiscriminationInterferenceGood memory recall usually requires not onlyrecall of an item from memoryYou also must identify the correct item relativeto the appropriate context or time frame The current trial The context of the experiment Relative to an earlier eventRetroactive interference (RI) new information prevents recall ofprevious information e.g., Overwriting a computer file.Proactive interference (PI) prior learning prohibits new learning e.g., Learning new cultural customs. At a particular moment in timePurdue UniversityPurdue UniversityProactive interferenceMay be due to a variety of effects One is that memory involves discriminating new from oldVisual memory See a set of photos Then see a test photo and decide if new or oldMemory: Trial 1 Memory: Trial 2 Memory: Trial 3Proactive interferenceSuppose you see this building as the test on the third trial You have a match in memory, but is from trial 1, not trial 3 You may report it having been shown in trial 3In general, previous trials make memory discriminationmore challengingMemory: Trial 1 Memory: Trial 2 Memory: Trial 3Purdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology 2

Prof. Greg Francis 1/2/13Proactive interferenceOne finds proactive interference for lots of memory tasksCogLab serial position experiment I looked at recall of the first letter in each list, averaged acrossall studentsProportion of times firstletter is recalled10.950.90.850.80.750.70.650.61 2 3 4 5 6 7 8 9 10 11 12 13 14 15TrialPurdue UniversityRelease from PIProactive interference weakens for differentstimulus typesRun two Brown-Peterson type experimentsControlExperimentalXJFTrial 1 XJFWRMTrial 2 WRMDBLTrial 3 DBLNRXTrial 4 942Purdue UniversityRelease from PIRelease from PITrials 1-3 show build up of PIExperimental group shows release of PI onTrial 4Percentage correct1008060Control40Experiment2001 2 3 4Trial Purdue UniversityWorks for many kinds of memory tasksMany kinds of stimuli5, 7, 9, 1vs5, 7, 9, HANDNews storiesD, D, D, DvsD, D, D, FNICE, SUNNY, ENJOY, PUPPYvsNICE, SUNNY, ENJOY, KILLPurdue UniversityMemory systemWorking memoryEvery memory system must have at least twocomponents/processes Storage retrievalWe have described proactive interference as beingdue to difficulty discriminating new items fromprevious itemsBut there is an alternative explanation Proactive interference might prevent items from beingstored and thereby make them unrecallableFor example, working memory has a storageinterference hypothesis for the phonological loopWorking memory suggests that interference canoccur by blocking ACP rehearsal (articulatory suppression,Brown-Peterson task, word length effect) within the PS when items sound similar both of these block the storage of items (items fall out of theloop)Purdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology 3

Prof. Greg Francis 1/2/13Testing modelsTest location of PI by changinginstructions after the list is presentedExperiment stimuli are names of indoor and outdoorgames subjects usually do not notice that wordon the fourth trial is an indoor game andothers are outdoor gamesTake two groups of subjects one has traditional PI type experiment one is told of difference on fourth trial, atthe time of testTrial 1Trial 2Trial 3Trial 4Purdue UniversityFOOTBALLSOCCERBASEBALLWALLYBALLInterference at recallIf PI prevented the last item from being stored your telling asubject that the fourth item was an indoor sport, shouldmake no difference (other than guessing) but it makes a big difference, they show release from PIPercentage correct1008060Not told40Told2001 2 3 4TrialPurdue UniversityACTHow to take a testAvoid PIAnswering successive questions on thesame topic hurts recall after answering unrelated questions go back to questions you cannot answer less proactive interference should recall moreT1, T1, T1, T1vsT1, T1, T1, T2Purdue UniversityLong-long term memory Bahrick et al (1975)392 subjects brought in their high schoolyearbook asked to matchnames with faces or recall names forfacesPurdue UniversityLong-long term memoryWhy is memory generally poorer with time?DiscriminationOne can suggest at least two explanations for poorermemory as time elapses1) Retroactive interference from subsequent faces/names2) The mental representation of temporal positionbecomes blurred as a function of time Recent events are easy to discriminate in time~25 years~50 years Long ago events are difficult to discriminate in timePurdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology 4

Prof. Greg Francis 1/2/13ConclusionsDiscriminationRetroactive interferenceProactive interferenceRelease from PIStrong effectsKnowing about can help in everyday tasksNext timeConstructive memoryFlashbulb memoriesMemory misattributionMisleading questionsHow good is eye-witness testimony?Purdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology 5

Prof. Greg Francis 1/2/13Constructive memoryPSY 200Greg FrancisLecture 20Memory testLast lecture you watched me for 50 minutesWhat was I wearing?How good is eye-witness testimony?Purdue UniversityPurdue UniversityDiscriminationThe task is difficult because you have to doseveral things Recall information that might be related to the task from memory Determine if the memory is actually for the correct event Determine if the memory is actually for the correct moment in time Gauge your confidence in the memorys validityAll of this suggests that performance on a memory taskinvolves discriminating informationTo address the discrimination problem, people engagein a constructive process to report memoriesNo forgetting? Brain surgeon (Penfield, 1959) Epilepsy patients stimulate brain regions before operating want to know what is being removedConscious patients report vivid memories unable to recall normally» "she saw herself as she had been while giving birth to herbaby." stimulation of temporal lobesIn the image, numbers indicateplaces where stimulationevoked different reportedexperiencesPurdue UniversityPurdue UniversityNo forgetting?Suggests that memories are stored but normallyunreachable (context things again)Basis for ideas of memory repression (and a few self-help books) The results are usually misunderstood Actually only occurred for 5% of patientsEven worse…Penfield (1959) the memories are nearly impossible to verify the few attempts find that the memories are not true» people describe places they have never visited, impossibleevents, fantasy,… The patients have epilepsy» Stimulation may have triggered something like an epilepticseizure (which can have hallucinations)It is more likely that stimulation feels like a memory, even thoughit is not your awareness of remembrance is a product of your brain it can be stimulated, even without a real memoryWhat do we mean by a valid memory?Purdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology 1

Prof. Greg Francis 1/2/13Flashbulb memoriesHighly emotional events tend to produce strongmemoriese.g. JFK assassination Challenger explosion Oklahoma City bombing Earthquakes September 11, 2001 ...Flashbulb memoriesPeople vividly recall detailssurrounding event where they were when theyheard what people said clothing worn time of day,..People are confident abouttheir reports however...Purdue UniversityPurdue UniversityFlashbulb memories Talarico & Rubin (2003) On September 12, 2001Asked volunteers to answer questionsabout their memory of The WTC attack An ordinary event (volunteers choice)Three groups for follow up 7 days later 42 days later 224 days laterRecalling of details was the same forWTC attack and ordinary eventPurdue UniversityFlashbulb memoriesThe memories of the WTCattack were more vividSubjects believed thosememories were more likely to bereliableOrdinary memories Faded in vividness Belief decreased over timePurdue UniversityFlashbulb memoriesSubjects confidence in their memory can bemisleading retelling of the story (I remember vividly when Kennedywas shot. I was…) probably reinforced the story Maybe not the true memoryFlashbulb memories are a real phenomenonabout the experience of memory, but probably not super-memoryMemory misattributionDonald Thomson was accused of rape and picked out ofa lineup by the victim (Schacter, 1996)He was on live TV at the time of the rape Ironically, he was discussing memory of faces for eyewitnesstestimonyThe victim had the TV on at the time of rape misattributed the face on TV for the face of her attacker very accurate report of the crime, otherwisePurdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology 2

Prof. Greg Francis 1/2/13Eyewitness testimonyI will show you a series of slides and then ask yousome questionsPurdue UniversityDr. Kim VuPSY 332: Cognition Purdue University 14Purdue UniversityDr. Kim VuPSY 332: Cognition Purdue University 16Purdue UniversityDr. Kim VuPSY 332: Cognition Purdue University 18PSY 200: Intro. to Cognitive Psychology 3

Prof. Greg Francis 1/2/13Purdue UniversityPurdue UniversityPurdue UniversityPurdue UniversityQuestionsEyewitness testimonyDid the bus, which came by, come from the left or theright?Later, show slides and ask subjects if they were part of theoriginal setDid another car pass the Red Datsun while it was at real slide contains YIELD signthe intersection with the stop sign? fake slide contains STOP signDid you see a bicycle?Did you see the taxi cab?Did you see if the policeman wrote anything down?Purdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology 4

Prof. Greg Francis 1/2/13Eyewitness testimonyEyewitness testimonyMisinformation effect Loftus & Palmer (1974)Compare accuracy according to pre-test questions Subjects without a misleading question--90% accurate Subjects with a misleading question -- 20% accurate In a follow-up, the experimenters asked those with misleadingquestions if they thought they were misled» 90% say noPaying money for correctness also had no effectLoftus, Miller &Burns (1978)Themisinformationeffect getsstronger with aweek delaybefore thememory testPercentage correct80706050403020100Consistent None MisleadingTypes of questions"DelayedImmediatePurdue UniversityPurdue UniversityMemory implantsLoftus has a procedure that implants amemory of being lost in a mall Basically just have subject read a plausible story (withsome details that could be true) Get family members to pretend the story is true Later the subject remembers the story as somethingthat happened to him/herIt is very easy for a therapist to implant falsememories into patientsConclusionsMany techniques believed to provide accuratememories, do not flashbulb memories brain stimulationMemory is constructiveMemories can be easily influenced by questions,interpretation, and contextPurdue UniversityPurdue UniversityNext timeAmnesiaAnterograde amnesiaRetrograde amnesiaUnusual characteristicsRepressionWhats wrong with my wife?Purdue UniversityPSY 200: Intro. to Cognitive Psychology 5

Prof. Greg Francis 8/16/11AmnesiaPSY 200Greg FrancisLecture 21What is wrong with my wife?Purdue UniversityFundamental fact!! There is no method other than objectphysical evidence to verify the accuracy of amemory!! Memory is a cognitive experience!!Confidence in the memory is another cognitiveexperience!!You can be very confident and still be wrong!! Of course, we must be correct fairly often, orour lives would be a total mess!Purdue UniversityAmnesiaAmnesia!! Loss of memory or memory abilities!! retrograde: forgetting events prior to the injury!! anterograde: forgetting events after the injury!! In most cases amnesia is limited in scope andduration!! like when my brother Joe slipped while playing frisbeePurdue University!! Scope and duration!! Retrograde amnesia for one patient5 months 8 months16 monthscomacomacoma1 year2 weeks2 years4 yearsto infancy to infancyto infancyPurdue UniversityTimeofaccidentWhat’s wrong with my wife?!! Nothing!!! But she cannot remember anything beforeher senior year in high school!! motor cycle accident!! complete retrograde amnesiaPurdue UniversityAn unusual case!! Side issues!! Sense of smell!! Mild anomia!! Odd aphasia (language deficit)!! She is able to learn and remember new information!! Remarkably unaffected by the loss of memories!! Personality!! Parents!! college!! makes study of retrograde amnesia difficultPurdue UniversityPSY 200: Intro. to Cognitive Psychology 1

Prof. Greg Francis 8/16/11What is lost?!! How can someone who loses theirchildhood memories go to college the nextyear?!!memories cannot be “wiped clean”!!perhaps they are just not directly accessible!!forgetting = recall problem?!! More generally,What is lost?!!while patients with retrograde amnesia forget theirnames, parents, addresses,…!!they do not generally forget how to walk, talk,solve problems»!Although they may have problems…!!Different types of memory systems»!controversial!Purdue UniversityPurdue UniversityMemory systems!! There are many different types of memory!! Amnesia seems to affect declarative more than nondeclarative memoryAnterograde amnesia!! Some patients have amnesia that preserves pastmemories but prevents formation of new memories!! many are long-time alcoholics who did not eat properly»! which leads to a thiamine deficiency»! which leads to Korsakoff’s syndrome!! Leonard in MementoPurdue UniversityPurdue UniversityPatient HM!! Surgery on hippocampus (to control epilepsy)!! anterograde amnesia!! unable to learn anything new!! Thought it was 1953!! shocked by age of face in his mirror!! Could not stand to read newspapers!! reintroduced himself to doctors, nurses,…!! Could carry on a conversation!Purdue UniversityAnterograde amnesics!! Fairly normal STM digit span (~7 items)!! But very difficult to extend digit span!! how many trials to repeat back list correctly?Mean trials to criterion3025201510500 5 10 15 20 25Number of digits to rememberPurdue UniversityControlsPatientsPSY 200: Intro. to Cognitive Psychology 2

Prof. Greg Francis 8/16/11!! Normal recency!! Abnormal primacySerial position curve!! Consistent with STM-LTM dichotomyPercent correct10090807060504030201000 5 10 15Item number Purdue UniversityControlsAmnesics!! Could learnsome things!!! E.g.,location ofhospitalcafeteria!! E.g. mirrordrawingtaskPatient HMPurdue UniversityPatient HM!! Mirror drawing task (Milner, 1968)!! HM had no knowledge of doing the task before!Purdue UniversityAmnesia-like memory!! Some aspects of memory seem very much likeamnesia!!infantile “amnesia”!!repressed memories!! Careful studies are difficult to come by becausethe memories (and absence thereof) must beverified!!remember the “fundamental fact” at the start oftoday’s lecturePurdue UniversityInfantile amnesia!! Most people report that they cannot rememberanything that happened to them before age 4yearsNumber of memoriesreported201510500 2 4 6 8Age at time of eventPurdue UniversityInfantile amnesia!! Reason is unknown, but the best theory goeslike this…!!children younger than 4-years-old view the worlddifferently from adults!!by encoding specificity, one needs to be in asimilar state as study to best recall something!!adults are very different from children, and thisprevents recall of early memoriesPurdue UniversityPSY 200: Intro. to Cognitive Psychology 3

Prof. Greg Francis 8/16/11RepressionRepression!! Psychotherapists (e.g. Freud) suggested thatinfantile amnesia occurred because much ofchildhood is filled with painful events and memoryof the pain is prevented by psychological defensemechanisms (repression)!! This is very unlikely!! In a laboratory, showing evidence of repressionrequires!!being unable to remember something!!being able to recover the memory through therapy!!proving that the recovered memory is accurate!! people do remember painful events well!! laboratory studies find no evidence of repressed memoriesPurdue UniversityPurdue UniversityRepressionConclusions!! In therapy, clinicians often claim evidence of repressionwith!! dream interpretation!! patterns in symptoms!! recovering a memory through hypnosis!! None of these techniques demonstrate a verifiedmemory!! Retrograde amnesia!! Anterograde amnesia!! Learning in anterograde amnesics!! Infantile amnesia!! Repression!! Among carefully controlled memory research, there isno evidence of repression!Purdue UniversityPurdue UniversityNext time!! Study schedules!! Levels of processing!! Mnemonics!! Sleep!! How to improve your memory without spending$20.Purdue UniversityPSY 200: Intro. to Cognitive Psychology 4

Prof. Greg Francis 8/14/12Improving memoryPSY 200Greg FrancisLecture 22How to improve your memory withoutspending $20.MemoryWe seem to be unable to control ourmemories learn things we dont want to remember unable to learn things we want to rememberIs there any reliable cue that something will beremembered? no but there are several tricks you can useto improve memory in certain situationsPurdue UniversityPurdue UniversityEncoding specificityWe know that memory is best when study and test contexts aresimilarEncoding specificityTest subjects in a neutral context(after another 3 hour interval) For example, testing in the study classroomBut variability in study promotes more general recall Smith et al. (1978) Subjects studied words twice: either in same context or different contexts (3 hourinterval between contexts)Look at proportion correct recall Highest with variable study contextsAdvice: if you want to remembersomething in lots of contexts, study inlots of contextsSession 1 Learning contextSession 2ContextClassroom 1 Classroom 2Classroom 1 0.41 0.69Classroom 2 0.53 0.39Purdue UniversityPurdue UniversityLevel of processingMemory can be influenced by depth of processing at the timeof study Craik & Tulving (1975)Subjects observe words with associated tasksquestion YES NOIn capitalletters?Rhyme withthing?Synonymfor heavy?BOOKspringbulkybooksprintbrownPurdue UniversityNumber of words recalledLevel of processingRecall is better as depth of processing increases More distinctive memories are created, which helps subsequent recall By varying depth of processing, you can construct memories that aremore likely to be recalled2520151050Capital Rhyme SynonymStudy taskPurdue UniversityPSY 200: Intro. to Cognitive Psychology 1

Prof. Greg Francis 8/14/12Levels of ProcessingLevel of processing is more important than intent tolearn (Hyde & Jenkins, 1973)11 groups of subjects 1 control group: told they will be tested to recall the words» not given any study task 10 experimental groups split to perform a study task» Pleasant-unpleasant rating» Estimate frequency of word usage» E-G checking: does word contain an E or a G?» Identify part of speech: noun, verb,…» Sentence framing: which sentence does word best fit in? For all experimental groups, either» (a) Intentional learning : told they will be tested to recall thewords» (b) Incidental learning: not told they will be testedPurdue UniversityIntentionRecall (out of 24 words) varies a lot with taskNot much variation with intention to learnPurdue UniversityImplicationsJudgments of LearningPeople do not know the best way to remember! There was nothing stopping controls from making pleasantunpleasantratings they apparently did not, or they would have done better!Advice: study interactively read notes rewrite notes rephrase notes teach someone elseMore generally, people are not usually good atestimating whether something will be rememberedPurdue University Nelson & Dunlosky (1991)Subjects study a pair of words (e.g., OCEAN – TREE)Estimate how likely they are to be able to rememberone word if shown the other (JOL). Given OCEAN, how likely to remember the associated itemlater? This is the subjects estimate of their ability to use LTMMake judgment either Immediately after studying the pair Delayed to later in the experimental trialsNote: students studying for an exam often use theimmediate approach for a JOL to decide if they needto continue studyingPurdue UniversityJudgments of LearningImmediate JOLs donot match memoryperformance (at theend of the experiment) Especially for highJOLsDelaying the JOLleads to fairly accurateJOLsAdvice: study, wait,estimate learningMethod of lociUsed by ancient Greeks to remember complicated speechesTo remember a list of words or key ideas visualize walking around an area with distinctive landmarks link the items to be remembered with landmarks by using bizarremental imagery to recall items in order, mentally walk through area (any ordered sequence will work -- e.g., a childrens rhyme)Memory piggybacks on the easy recallability of the bizarreimageryPurdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology 2

Prof. Greg Francis 8/14/12Method of locie.g., grocery listITEMS LOCI Add vivid,bizarre imageryhot dogscat foodtomatoesdrivewaygarage interiorfront doorDigit spanSF learned to increase hisdigit span to 81 digits (anyrandom sequence) 200 hours of practice Ericsson, Chase & Faloon (1980)Created a hierarchy of memorytricksTechnique did not transfer to othermemory tasks (e.g., letters)Essentially learned to use long-termmemory for the digit span taskPurdue UniversityPurdue UniversitySleepMany types of memory improve with sleepSome type of consolidation of memoriesThe effect is not just time Although time also has an effectWell look at one representative study Ellenbogen et al. (2007)Subjects learn to identifyorder relationshipsbetween random shapesOnly shown one pair at atimeSubjects have to learn/memorize the appropriateanswer to each pairSleepPurdue UniversityPurdue UniversitySleepThere is a ordered arrangement to the stimuliIf you know this arrangement, deciding for any pair is easy But subjects are never explicitly told about this arrangementSubjects are split in to threegroups, according to whenthey are tested 20 minutes later 12 hours later 24 hours laterNo differences when testedon the originally studieditemsSleepPurdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology 3

Prof. Greg Francis 8/14/12SleepSleepSubjects are split in to threegroups, according to whenthey are tested 20 minutes later 12 hours later 24 hours laterBig differences when testedon new pairs that fit theordered structure E.g., A>C, C>E, B>DHalf of the 12 hour grouphad sleep and half did notIt makes a difference forpairs of items that are farapart in the orderedstructure1-degree: A>C, B>D,…2-degree: A>D, B>E,…Advice: study early! Get some sleep!Purdue UniversityPurdue UniversityA common approach to studying isto use flash cards (or somethingsimilar)Two steps to studying 1) Read material on both sides(study) 2) Practice test the material (givenone side, try to recall the informationon the other side)What should you do when yousuccessfully recall the informationduring the practice test? Continue to study? Continue to test? Set aside and focus on other cards?Practicing recall Karpicke & Roediger (2008)Practicing recallSubjects study 40 Swahili - English word pairs mashua – boat kaka -- brother Test for English given Swahili:» mashua --???Four groups of subjects, that differ after an item is correctly recalled ST (study-test): subject studies and continually tested over every pair SnT (study on non-recalled - test on all): when a subject recalls a pair, it is nolonger studied, but it continues to be tested STn (study all, test only on non-recalled): when a subject recalls a pair, itcontinues to be studied, but it is not tested SnTn (study on non-recalled, test on non-recalled): when a subject recalls apair, it is not studied or tested againA week later, everyone is testedPurdue UniversityPurdue UniversityPracticing recallLearning stylesStandard advice is that once you learn something, study something elseThis is not good advicePerformance is best whenevery pair is tested, even ifyou have alreadydemonstrated it ismemorizedThe amount of time spentstudying the words doesnot matter so muchSuggests that you learn howto recall the informationAdvice: Test yourself!Keep practicetestingDont practicetestA common approach in education is to identify a students learning styleand then teach for that styleLots of tests to identify a students learning styleThere do seem to be real differences in what style people indicate theypreferPurdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology 4

Prof. Greg Francis 8/14/12Learning stylesUnfortunately, there is absolutely no evidence that reported learning stylepreference has anything to do with learningPashler et al. (2009) observed that to demonstrate evidence that learningstyle influenced learning, you have to show a particular kind of interactionof effectLearning stylesPashler et al. (2009) then reviewed hundreds of studies purporting to showevidence for learning styles, but only ever found effects like theseBut these only indicate an advantage for a type of learning or a methodPurdue UniversityPurdue UniversityWhy is the idea popular?Learning stylesIt fits with the American ideal of everyone beingcapable of learning if given the chance (no childleft behind)It allows parents (and students) to blame theeducational system for failure rather than lack ofmotivation or abilityIt lends itself well to statistical quirks of findingjust the right method for a given studentIts a generalization of the experience that agiven student benefits from a new explanation ofmaterialPurdue UniversityConclusionsLots of ways to improve memory Encoding specificity level of processing Judgments of Learning method of loci Sleep Practice testing Learning stylesPurdue UniversityNext timeMental representationPrototypesExemplarsPropositionsCogLab on Prototypes due!What is a shoe?Purdue UniversityPSY 200: Intro. to Cognitive Psychology 5

Professor Greg Francis 1/6/11Representation of knowledgePSY 200Greg FrancisLecture 23What is a shoe?Concepts!! What is the information in Long Term Memory?!! May be several different types!! We have knowledge about the world!! Due to personal experience!! Or due to language!! Such information must be in some kind of format,which we call concepts!! But what are the concepts?!! what is the concept of “dog,” “walking,” or “free-marketcapitalism”?Purdue UniversityPurdue UniversityConcepts!! We will look at three topics in concepts!!Definitions (don’t really work)!!Prototypes (closer to how humans think)!!Exemplars (more likely than prototypes)!! And then combinations of concepts!!propositionsDefinitions!! Plato (and Socrates) spent a lot of efforttrying to define terms like virtue andknowledge!!they were largely unsuccessful!! the 20th century philosopher Wittgensteinwondered if definitions of even simpleconcepts were possiblePurdue UniversityPurdue UniversityDefinitions!! Consider the concept shoe, you might define it asWebster’s does!! A covering for the human foot, usually made of leather,having a thick and somewhat stiff sole and a lighter top.!! Anything resembling a shoe in form, position, or use.!! Lots of shoes fit this definitionDefinitions!! Consider the concept shoe, you might define it as Webster’sdoes!! A covering for the human foot, usually made of leather, having athick and somewhat stiff sole and a lighter top.!! Anything resembling a shoe in form, position, or use.!! But now consider some situations and decide if they are reallyshoes!! A shoe that is intended for display onlyPurdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology 1

Professor Greg Francis 1/6/11Definitions!! Consider the concept shoe, you might define it as Webster’sdoes!! A covering for the human foot, usually made of leather, having athick and somewhat stiff sole and a lighter top.!! Anything resembling a shoe in form, position, or use.!! But now consider some situations and decide if they are reallyshoes!! a shoe filled with cement, which cannot be worn!! a covering worn on the handsof a person without legs whowalks on his hands!! And this? "Definitions!! The difficulty is the same one that Plato and Socrateshad trying to define virtue!! for any definition you come up with, I can find examples that donot seem to fit the definition!! But we all know what a shoe is!! so our knowledge of this concept must not be based on someprecise definition!! Note, scientists can (sometimes) create precisedefinitions (e.g., a dog is defined by a DNA pattern orby mating abilities)!! but the definition is somewhat arbitraryPurdue UniversityPurdue UniversityPrototypes!! Perhaps what defines a concept is similarityamong its members!!there may be no absolutely necessary characteristics!!there may be no absolutely sufficient characteristics!! Prototype theory supposes that similarity isjudged relative to a prototype example of theconcept!!e.g., an ideal, average, or most frequent version of theconceptPrototypes!! In prototype theory it is possible for an object tobe “more” or “less” a certain concept!! Consider the concept “coffee cup”Purdue UniversityPurdue UniversityPrototypes!! In prototype theory it is possible for an object tobe “more” or “less” a certain concept!! Consider the concept “coffee cup”!!and variations (some are “cup-ier” than others)Prototypes!! In prototype theory it is possible for an object tobe “more” or “less” a certain concept!! Consider the concept “coffee cup”!!and variations (some are “cup-ier” than others)Purdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology 2

Professor Greg Francis 1/6/11Prototypes!! In prototype theory it is possible for an object tobe “more” or “less” a certain concept!! Consider the concept “coffee cup”!!and variations (some are “cup-ier” than others)Prototypes!! In prototype theory it is possible for an object tobe “more” or “less” a certain concept!! Consider the concept “coffee cup”!!and variations (some are “cup-ier” than others)Purdue UniversityPurdue UniversityPrototypes!! Lots of experiments suggest the role ofprototypes!!Posner & Keele (1968): learning category names forrandom dot patterns!!Discriminate two sets of random dot patterns!!Each pattern is a variation of one of two prototypepatterns!! PrototypesAPrototypesBPurdue UniversityPurdue UniversityPrototypes!! variations are made by moving some of the dotsvariant of Avariant of BPrototypes!! The key test is done after subjects learn to classify thevariants!! reaction time for judgment is recorded for stimuli they have neverseen before»! new variants»! the prototypes!! reaction time is faster for the prototypes!! which suggests that the mental representation of the categories(concepts) are built to favor the prototype of the category!! Look at CogLab data!! subjects learn to classify many different variants!!they never see the prototypes themselvesPurdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology 3

Professor Greg Francis 1/6/11Prototypes!! Results are based on data from 169 participants.!! Pattern type Reaction time (ms)!! Prototypes ! !686!!! Variants ! !719!!! Unanswered by this (and many other) experiments is what a prototype is:!! a “thing” that resides in memory and contains information about the categoryfeatures!! the result of processing information!! A bit of thought suggests it is the result of processing informationPrototypes!! Consider the types of concepts you can have!! and how specific they can be!! things: bird, dog, chair, shoe,…!! actions: walking, running, sleeping,…!! goal-derived: “things to eat on a diet”, “things to carry out of a housein case of a fire”,...!! ad hoc: “things that could fall on your head”, “things you might seewhile in Paris”, “gifts to give one’s former high school friend who hasjust had her second baby”,...!! When studied, these concepts all seem to have prototypecharacteristicsPurdue UniversityPurdue UniversityPrototypes!! We can generate new concepts from oldconcepts!!it’s inconceivable that every possible prototype existsready to be used!!some must just be built as they are needed!!perhaps even the prototypes for simple concepts like“bird” or “shoe” are also just built when they areneeded!! A theory that can account for this processingapproach is exemplar theoryPurdue UniversityExemplars!! A concept consists of lots of examples of theconcept!!e.g., a “coffee cup” concept might contain lots ofexamples of coffee cupsPurdue UniversityExemplars!! Comparing an object to see if it is a coffee cupinvolves comparing it to each example in memoryand seeing if it matches anything well enoughExemplars!! Even if it is a new object, it may match severalexemplars well enough to generate an overallresponse to indicate it is a coffee cupPurdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology 4

Professor Greg Francis 1/6/11Exemplars!! Some coffee cups seem prototypical becausethey match lots of exemplars!!that’s what defines a prototypeExemplars!! Unlike prototype theory, exemplar theory also containsinformation about the variability of examples within aconcept!! Thus, we know that pizzas have an average size of 16inches but can come in lots of different sizes!! And we know that foot-long rulers have an average sizeof 12 inches, but essentially no variability in sizePurdue UniversityPurdue UniversityComplex associations!! How do we represent a concept thatinvolves combinations of concepts?!!e.g., “Dogs chase cats.”!!e.g., “Last Spring, Jacob fed the pigeons inTrafalgar Square.”!! Need to identify the role of each conceptPropositions!! Higher order ideas!! things doing something!! Statement that is either true or false!! things cannot be judged true or false!! e.g., Book, Albert, Threw, Professor, Test, Gave!!consists of an ordered list of concepts»!e.g., (relation:X, Agent:Y, Object:Z)Albert threw the book.(relation:Threw, Agent:Albert, Object:Book)Purdue University(Threw, Albert, Book)Purdue UniversityProposition!! Network Representation!!The proposition connects the appropriate conceptnodesagentAlbertProposition!! Network Representation!! The proposition connects the appropriate concept nodesProfessoragentthrewrelationAlbert threw the bookgaverelationThe professor gave a testobjectobjectbookPurdue UniversitytestPurdue UniversityPSY 200: Intro. to Cognitive Psychology 5

Professor Greg Francis 1/6/11!! Network RepresentationProposition!! The proposition connects the appropriate concept nodesProposition!! Network Representationlast SpringDogtimechaserelationagentDogs chase catsJacobagentLast Spring, Jacob fed pigeonsobjectpigeonspigeons inTrafalgar SquareinrelationobjectcatPurdue UniversityrelationfeedslocationTrafalgar SquarePurdue UniversityProposition!! One way of combining concepts!!there are also other theories of how to do this!! Used a lot in Artificial Intelligence!! Do humans represent interactions ofconcepts with propositions?!! Some experimental evidence!! Ratcliff & McKoon (1978)Proposition!! study phase»! subjects are asked to memorize a set of 504 sentences»! 18 - 1 hour sessions!!! test phase»! show words and have subjects decide if they were in thestudy sentences or not»! measure reaction time for words from the sentencesThe bandit who stole the passport faked the signaturePurdue UniversityPurdue UniversityProposition!! Network RepresentationProposition!! In the test phase, a word is given and thesubject responds as quickly as possibleobjectpassportobjectpassportpassportbandit stole passportagentbanditagentfakedrelationbandit stole passportagentbanditfakedrelationbandit faked signaturebandit faked signaturerelationobjectrelationobjectstolePurdue UniversitysignaturestolePurdue UniversitysignaturePSY 200: Intro. to Cognitive Psychology 6

Professor Greg Francis 1/6/11Proposition!! In the test phase, a word is given and thesubject responds as quickly as possibleProposition!! The expectation is that activation will flow through theentire proposition that includes this wordobjectpassportpassportobjectpassportpassportagentbanditfakedagentbanditfakedbandit stole passportrelationbandit stole passportrelationbandit faked signaturebandit faked signaturerelationobjectrelationobjectstolePurdue UniversitysignaturestolePurdue UniversitysignatureProposition!! So, if the next word is part of the same proposition, asubject will respond even fasterProposition!! If words are from different propositions, noprimingobjectpassportbanditobjectpassportpassportagentbanditfakedagentbanditfakedbandit stole passportrelationbandit stole passportrelationbandit faked signaturebandit faked signaturerelationobjectrelationobjectstolePurdue UniversitysignaturestolePurdue UniversitysignatureProposition!! In the test phase, a word is given and thesubject responds as quickly as possibleProposition!! Activation will flow through the entire proposition thatincludes this wordobjectpassportpassportobjectpassportpassportagentbanditfakedagentbanditfakedbandit stole passportrelationbandit stole passportrelationbandit faked signaturebandit faked signaturerelationobjectrelationobjectstolePurdue UniversitysignaturestolePurdue UniversitysignaturePSY 200: Intro. to Cognitive Psychology 7

Professor Greg Francis 1/6/11Proposition!! When the next word is shown, its node has not beenprimed, so it responds more slowlyobjectpassportbandit stole passportrelationstoleagentbanditbandit faked signaturePurdue UniversitysignatureobjectsignaturefakedrelationProposition!! Test Phase : Priming Task!! compare RTs for second in a pair of words!! within a common proposition (bandit -- passport)!! between propositions (passport -- signature)!! not related in sentence (horizon -- signature)!! interested in RT to second word in each pair!! Ratcliff & McKoon (1978)!! results»! within same proposition words: 561 msec»! between proposition words: 581 msec»! unrelated: 671!! evidence of priming by propositional activation!! We think in propositions!Purdue UniversityConclusions!! Concepts!!definitions!!prototypes!!exemplars!! Propositions!!Evidence we think in terms of propositionsNext time!! Other types of knowledge!! Mental images!! mental rotation!! mental scaling!! limitations of!! CogLab on Mental rotation due!!! Is a picture in your head like a picture in the world?Purdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology 8

Professor Greg Francis 8/14/12Mental imageryMental representationPSY 200Greg FrancisHow do you mentally representknowledge? concepts (prototypes, exemplars)Lecture 24 propositions mental images, mapsIs a picture in your head like a picture inthe world?Purdue UniversityPurdue UniversityPerceptionWe have knowledge about, and memoriesof, perceived stimuli sights smells touches soundsAre these converted into propositions, orconcepts or is there something else?Purdue UniversityWhen we see thisimage how do werepresent theinformation in theimage? analog: copy ofimage in head andwe can retrieve itImages symbology: convert topropositions/conceptsPurdue UniversityImagesImagesI can remember theimage on theprevious slide and itfeels like I picturethe image in myhead but is it really? could it be aproposition?shortswearsgirlRuns toballIf you ask me questions about the previous slide,my answers would not necessarily identify therepresentationshortswearsgirlObservableRuns toballThere was a girl running toa soccer ball. She wore red shorts.MentalPurdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology 1

Professor Greg Francis 8/14/12Pure propositionsLets look at the arguments for a purelypropositional representationLook at this picture, Ill ask you questions about itWorking with imagesIf this image was printed on a piece of paper, you wouldhave no problem answering the questions about itIf you had an exact copy of the image in your head, youwould expect you could look at the copy and make allkinds of judgments but you cannot how you interpret the image to a large extent determines whatyou know about itMental images are not exactly like real images this tends to be particularly true for memory of images verbal descriptions dominate memory for imagesPurdue UniversityPurdue UniversityAnswer these questions:Another exampleHow did you do?Another exampleWhich is further North, Seattle, Washington orMontreal, Canada?Which is further East, Detroit, Michigan orIndianapolis, Indiana?Which is further South, Boston, Massachusetts orNew York City, New York?Which is further West, Reno, Nevada or SanDiego, California?Purdue UniversityPurdue UniversityPropositionsSo this suggests that mental images are not exactly like realimages and something like propositional information likely influencesreports that are ostensibly based on mental images or mentalSan DiegomapsinagentrelationNevadaSan Diego is inCaliforniaobjectobjectCaliforniaCalifornia is west of NevadaagentrelationWestPurdue UniversityReno is in NevadaagentRenoinrelationPropositionsIt is clear that propositional information influencesmental imagery but is it all propositions? are there mental images, as we tend to experience them?Is there any reason to believe that mental imagesare at all analogous to real images? yesPurdue UniversityPSY 200: Intro. to Cognitive Psychology 2

Professor Greg Francis 8/14/12Representation of mental imagesImagine you have a mental image of a lion If the mental image is a description (set ofpropositions), then it should include descriptiveinformation (head, mane, ears, whiskers,, tail,…)» size of body parts shouldnt matter much(except as information in the proposition)Representation of mental imagesImagine you have a mental image of a lion If the mental image is a depiction, then it shouldinclude factors such as head, mane, ears, and theirsizes and relative positions» self reports of mental images suggest thedepicted version, but how to measureexperimentally?Purdue UniversityPurdue UniversityRepresentation of mental images Kosslyn (1976) Ask subjects to quickly answer questions like:» Does a lion have a head? (big body part)» Does a lion have claws? (small body part) subjects in two groups» 1) form a mental image of a lion» 2) think about a lion, but without a mental image Subjects forming a mental image respond more quickly to the headquestion than the claw question» presumably because the head is bigger in the mental imageRepresentation of mental imagesSubjects that did not form a mental image respond more quicklyto the claws question than the head question presumably because the propositions about a lion having claws aremore accessible in some network of propositionsSo, different ways of thinking about a lion can lead to differentpatterns of access to information which implies mental images are different than propositions16001600Reaction time (ms)140012001000800600400200ImageryNo imageryReaction time (ms)140012001000800600400200ImageryNo imagery0SmallpartBig partPurdue University0SmallpartBig partPurdue UniversityMental rotationShepard & Metzler(1971) 3-D shapes rotated in plane or in depthCogLab experiment 163 participantsThe more rotated thestimuli are, the longerit takes to make aresponseMental rotation decide if shapesare same aredifferent close to linear increasein RT measure time todecidePurdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology 3

Professor Greg Francis 8/14/12Mental rotationVividnessThis type of experiment has been taken as strongevidence that mental images are not just propositionsThere are individual differences in reports of mental imagevividness imagined movement of the mental image (rotation) resemblesactual movement It takes time to mentally move through a mental space» The CogLab data suggests it is about 100 degrees/second» 10 milliseconds for each degree some people report their mental images are just like real images others report they are fuzzy and vague some people report no mental images at all (10%) Nevertheless, people all do basically the same on many tasks thatseem to require operating on mental images no reason why propositions would give data that incorporatespatial and temporal relations between aspects of the mentalimagesPurdue UniversityPurdue UniversityfMRIfMRIHugdahlet al. (2006) had subjects perform a mentalrotation task or a comparison task (did not require rotation)in an fMRI scannerCompared to the 2-D condition, the 3-D condition hasmore activity in the superior parietal lobesMalesFemalesPurdue UniversityPurdue UniversityConclusionsNext timeMental representations of pictorialConsciousnessinformationDualismThere are some things you cannot do withmental imagesPropositions are importantArtificial intelligenceQualiaMental images are not just propositionsMental rotation taskDo you see red like I see green?Purdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology 4

Greg Francis 1/2/13ConsciousnessPSY 200Greg FrancisLecture 25Do you see red like I see green?What is consciousness?Awareness of events, stimuli, thoughts,selfA sequence of meaningful itemsStream of thoughtsDistinct from unconscious processing(e.g., hearing a sentence, retrievinginformation from memory,…)Purdue UniversityPurdue UniversityHistoryDescartes dualism (Cartesian dualism) pineal gland link between body and spirit how they could connect was a real problemMaterialismNearly all scientists are materialists, butold ideas die hardA lot of work (PET scans, fMRI,…) looksfor the site of consciousness a special physical transformation thalamusMind-body problemMaterialism (the brain is the mind) or the mind derives from the brain reticular formation quantum mechanics distributed awarenessPurdue UniversityPurdue UniversityBeforeconsciousnessA turning pointA common view is that there is a moment/place which/where before something wasnot conscious and which after it isconsciousBut this is not true in the brainConsciousnessAfter consciousnessDistributed processingInformation processing is spatially andtemporally distributed in the brainProcessing changes with new stimuliThere really is no moment ofconsciousness different brain areas know different things atdifferent timesPurdue UniversityPurdue UniversityPSY200 Cognitive Psychology 1

Greg Francis 1/2/13An analogyWhen did the British empire learn of the end of the War of1812? treaty signed in London months before the Battle of NewOrleans word was not received by British troops in America until twoweeks after the Battle of New Orleans (January 8, 1815)An analogyFor complicated systems like the British empire (andhuman brains) different parts know different things at different times there is no official moment of knowledge no official moment of consciousness!Demonstration when does the class know/understand?Purdue UniversityPurdue UniversityHow / Why?There is no moment becauseinformation is distributed in the brain Both in space and timeCan distributed processing really produceconsciousness, or must there besomething else to put it all together? can consciousness arise from non-consciousprocessors? (artificial intelligence?)Purdue UniversityHow do you know aperson is conscious/intelligent? They behave in a waythat we interpret asconsistent with aconscious beingTuring test: apply thesame logic to acomputer if a conversation with acomputer isindistinguishable from aconversation with ahuman Then conclude thecomputer is intelligentTuring testPurdue UniversityArtificial intelligenceNo computer has passed anything but a weakform of the Turing test lack sufficient schemas, creativity, general knowledgeIt is worth noting that other things would also notpass a Turing test children mentally impaired people mute people people who speak a language we do not understandPassing a Turing test is not necessary forconsciousnessPurdue UniversityCaptchaThe basic ideas are implemented in severalmethods for computer securityCompletely Automated Public Turing test to tellComputers and Humans ApartPurdue UniversityPSY200 Cognitive Psychology 2

Greg Francis 1/2/13Turing testThe Turing test is only one way to demonstrateintelligence and a rather strict one at that not passing the Turing test does not mean that acomputer is not intelligent of course, it doesnt mean the computer is intelligenteitherVariations on Turing testDoubtersMany people have suggested that computerscannot, in principle, become intelligent they argue that purely symbolic computations cannotlead to consciousness and humans use emotion, insight, intuition,intentionality instead of simple computationLets look at two arguments against strong AI discriminate conversation between a child and acomputer look at a conversation and decide which was thecomputer ...Purdue UniversityPurdue University1. The Chinese room (Searle)Imagine you are in a room with two slots and a book Slot 1: someone sends you notes with Chinese characters on them Book (written in English): in the book you can look up the Chinesecharacters and write down corresponding Chinese characters onanother piece of paper Slot 2: you can send your piece of paper out this slotSlot 1 Slot 21. The Chinese room (Searle)If the book provides rules on how to answer questionsin Chinese then you can answer written questions in Chinese even though you do not know Chinese!Consciousness (in general, understanding) is not afunction of the thing (or person) who implements therulesBut consider it from the point of view of a personoutside the room Who is sending messages inPurdue UniversityPurdue University1. The Chinese room (Searle)You are having a conversation with someone You have to decide if the person understands what you are saying(its the Turing test) You ask them to describe the wall of their room They report it is green …and so on…Slot 2 Slot 11. The Chinese room (Searle)Searles point is that We know the person in the room does not understandChinese We might be fooled into thinking they do based on theirresponses to the questions Thus, the Turing test is a bad testBecause the Turing test is essentially the samestructure The computer plays the role of the person in the roomPurdue UniversityPurdue UniversityPSY200 Cognitive Psychology 3

Greg Francis 1/2/13However,...Searle has set up a deceptively simple scenario the Chinese room may be an impossibilityYou can imagine a situation where one has a bookwith rules to answer questions in Chinese but only if you do not think too hard in reality, there may be no such book! if the questions can be on almost any topic, thenunderstanding is required for that type of complexprocessingAnd understanding is generally restricted toconsciousness Or maybe one needs to conclude that such an advancedbook has potential consciousnessPurdue UniversityAnd moreover...At a smaller level of computation, it is hard to seehow consciousness could not be (theoretically)possible in computersEach cell in your head is data in - data out suppose cells were gradually replaced by tinycomputers that kept all processing the same» Neuromorphic chips would you claim that at some point you are no longerconscious?There is nothing fundamental about organicconsciousnessPurdue University2. QualiaSome researchers object to the very idea thatcomputers could become conscious They argue that some things in consciousness are notjust computation e.g., consider the color red There seems to be a particularly subjective experienceof seeing something red2. QualiaConsider two people who see the world incolor oppositesQualia for person 1A red apple with agreen leafQualia for person 2A red apple with agreen leafPurdue UniversityPurdue University2. QualiaClearly, theres a big difference in theperceptual experience of these people,but their behavior is essentially the same And there seems no way to distinguish oneexperience from the other Its the unmeasureable experience that is aqualia2. QualiaQualia proponents argue, for example, you can learn all there is to know about light waves,photoreceptors, neural transduction and coding ofcolor,… But suppose you never see any red objects Your knowledge will not tell you what you willexperience when you first see the red of an apple Indeed, you could be tricked into believing a greenapple was red (if you had never seen green either)Purdue UniversityPurdue UniversityPSY200 Cognitive Psychology 4

Greg Francis 1/2/132. QualiaConclusionsBut this is a defeatist argument, or a pointless oneConsciousness if I knew everything about light, photoreceptors, andneural representation of colors, then I would be able toknow what I will experience when I see red it is difficult (maybe impossible for any single human) toknow (or even imagine knowing) all that information in anacademic sense but that doesnt mean that such information does notexistIts partly an empirical question But no one can do the experimentdistributed processing in the brain no site of consciousness no time of consciousnessChinese roomQualiaArtificial IntelligenceDaniel Dennet Consciousness ExplainedPurdue University(1991)Purdue UniversityNext time Review for exam 3 After Exam 3 Start a discussion of language Language as an instinct pidgins and creolesWhy we do not have to worry about teachinglanguage in school.Purdue UniversityPSY200 Cognitive Psychology 5

Prof. Greg Francis 1/6/11The language instinctPSY 200Greg FrancisLecture 26Why we do not have to worry aboutteaching language in school.Purdue UniversityLinguistics!! Study of language (Noam Chomsky)!!sentences!!words!!sounds!!structure!!interpretation!! The language instinct!!Pinker (1994)Purdue UniversityPreconceptions!! We tend to think of language as!!a great invention of human cognition!!taught to children!!taught in schools!!a cultural invention!! This is wrong!!!instead, language is an instinctPurdue UniversityCultural influences!! Culture does influence language!! Consider words in English»! Some derived from the invading Normans (1066)(considered sophisticated and polite)»! Some derived from the Anglo-Saxon language of theBritish Isles (considered crude by the invaders)!! Norman: perspiration, dine, deceased, desire, urine,excrement!! Anglo-Saxon: sweat, eat, dead, want, piss, shit!! But this is not what determines our capability tohave language!Purdue UniversityBiology!! Language is a specialized skill ofhuman animals!!Darwin (1871)!! Humans instinctively learn language!!effortless!!unconscious!!procedural knowledgePurdue UniversityLearning!! Like all skills, language needs the properenvironment to be developed!!blinded birds cannot navigate by the stars!!Atlantic Ocean turtles that navigate by magneticfields need to be in the correct ocean!! Language development needs exposure toother people for communication!!but it needs surprisingly less exposure than youmight suspectPurdue UniversityPSY 200: Intro. to Cognitive Psychology 1

Prof. Greg Francis 1/6/11Child learningLearning!! Children do not learn language by simplyimitating others!!otherwise they would never come up withstatements likeDon’t giggle me!We holded the baby rabbits.I’m felling!Purdue University!! Instead, each child reinvents language!!difficult to test because we rarely get to see alanguage created from a non-language!!however, there are cases!!! Slave plantations in the South Pacific mixedtogether people of many differentlanguages!!create a jargon called a pidginPurdue UniversityPidgin!! For example, in New Guinea!!pidgin is similar to English (rulers of the plantation)woman: ‘meri’ (Mary, generic word for woman)another man’s wife: ‘meri bilong enaderfelo man’hair: ‘grass bilong hed’helicopter: ‘mixmasta bilong Jesus Christcoffin: ‘die bokus’piano: ‘bokus bilong teeth yu hitim teeth bokus is cryPurdue UniversityPidgin!! The Ten commandments in pidgen!!as translated by the Alexishafen Catholic Mission in1937!! 1. Mi Master, God bilong yu, yu no ken mekim masalai end oltambaran.!! 2. Yu no ken kolim nating nem bilong God.!! 3. Yu must santuium sande.!! 4. Yu mast mekin gud long papamama bilong yu.!! 5. Yu no ken kilim man.!! 6. Yu no ken brukim fashin bilong marit.!! 7. Yu no ken stilim samting.!! 8. Yu no ken lai.!! 9. Yu no ken duim meri bilong enaderfelo man.!! 10. Yu no ken laik stilim samting.Purdue UniversityLearning!! In Hawaii at the turn of the century!!workers from China, Japan, Korea, Portugal,The Philippines, and Puerto Rico werebrought in to harvest sugar!!they developed a pidgin!!some were still alive in 1970 and interviewedto see how the pidgin workedLearning!! Pidgin is not a true language!!word order is arbitrary!!no rules!!no tenses!!no prefixes or suffixesMe cape buy, me check make.!!can only be understood in context of theconversationHe bought my coffee; he made me out a check.I bought coffee, I made him out a check.Purdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology 2

Prof. Greg Francis 1/6/11Creoles!! The children of these workers speak verydifferently!!if removed from parents (and so unable to learnnative tongue)!!they transform the pidgin into a full-fledgedlanguage»!tenses, rules, prefixes, suffixes,...!! Find the same type of transformations amongchildren learning sign-languagePurdue UniversitySign language!! Nicaraguan schools for the deaf (1979)!!tried to teach children to lip-read (poorresults)!!but children started making a pidgin on theplayground»!Lenguaje de Signos Nicaraguense(LSN)!! New students took the pidgin andcreated a language (creole)!!Idioma de Signos Nicarguense (ISN)Purdue UniversitySign language!! You can even see the invention of language ina single child!! “Simon,” a deaf boy who also had deaf parents!!parents learned American Sign Language (ASL) latein life and so are not very good at it!! Simon had little contact with other deaf people!!but his signing was much better than his parents!!! Language learning is not imitation!Education!! There is always a group of people who say that weneed to get back to the “basics” of education!! including studies of grammar!! usually, these are veiled versions of racism!! In fact, children do not learn language in school!! No one learns to speak by properly identifying nouns,pronouns, prepositional phrases, verbs, adverbs,…!! Education is good for reading and writing!! but writing is dramatically different from speaking!! and reading is dramatically different from listeningPurdue UniversityPurdue UniversityEducation!! But then how do we explain thatuneducated people speak improperly?!! e.g. gang member in HarlemYou know, like some people say if yourgood an’ shit, your spirit going’t’heaven…’n’ if you bad, your spirit goin’to hell. Well bullshit! Your spirit goin’ tohell anyway, good or bad.Education!! This person is not speaking with bad grammar, buthe is also not speaking in Standard AmericanEnglish (SAE)!! He’s speaking in a dialect called Black EnglishVernacular (BEV)!! Both languages have certain rules!! His statements obey the rules of BEV precisely!!! Consider contractions of wordsPurdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology 3

Prof. Greg Francis 1/6/11Rules!! In SAE you can replace some word pairswith contractions!!“They are” --> “They’re”!!“He is” --> “He’s”!! But you cannot always do this!!“Yes he is!” -->? “Yes he’s!”!!“Who is it?” -->? “Who’s it?”!! BEV has similar types of rulesPurdue UniversityRules!! BEV allows speakers to drop some words!!“…if you are bad…” --> “…if you bad…” isgrammatically correct!! BEV does not allow word dropping arbitrarily!!“Yes he is!” -->? “Yes he!”!!“Who is it?” -->? “Who it?”!! It is difficult for a non-speaker of BEV to noticethe application of the rulesPurdue UniversityLanguage!! So if everyone is speaking a language, which iscorrect?!!none, they are just different!!they are different dialects of English!! Linguist Max Weinreich!!“A language is a dialect with an army and a navy.”!! The dialect you speak may give away yourpersonal history, but it is not fundamentallyworse than any other dialect.Purdue UniversityConclusions!! Language is an instinct!!specialized skill among humans!!children need little tutoring to learn language!!children invent language if one is not readilyavailable!! Language follows rules!!even when it doesn’t seem toPurdue UniversityNext time!! Grammar!! Long term dependencies!! Phrases!! Language universals!! Dr. Francis says something new!Purdue UniversityPSY 200: Intro. to Cognitive Psychology 4

Greg Francis 1/6/11Phrase treesPSY 200Greg FrancisLecture 27Dr. Francis says something new!Language!! Conveys information!! Allows us to know about things we havenever experienced!!moon flights!!mating habits of tigers,…!! How do we do it?!! Two key aspectsPurdue UniversityPurdue University!! SymbolsSymbols and grammar!!words are arbitrary!!the sound “dog” has nothing to do with dogs!!compare driving on parkway to parking ondriveway, blueberries and cranberries,hamburger…!! Grammar!!the order of words matters!!Dog bites man. vs. Man bites dog.Purdue UniversityGrammar!! Discrete combinatorial system!!combinations of words!! How many combinations?!!if interrupted anywhere in the middle of asentence, there are about 10 words one couldselect before finishing it up!!if sentences average around 20 words, thatmeans there are 10^20 sentencesPurdue UniversityGrammar!! But in fact, there are infinitely manydifferent sentences!!there is no limit to how long a sentence canbe!! For any sentence I give you, you canalways make it longer by addingsomething like!!Professor Francis said that, “….”Vastness!! It is amazing how powerful language is!! You have probably never heard thefollowing sentence!!moreover, it is probably its first utterance inhuman history, but you understand it anyhowWe still have a month of classes to go, but it is so warm that itfeels like it is already dead week.Purdue UniversityPurdue UniversityPSY200 Cognitive Psychology 1

Greg Francis 1/6/11Grammar!! You not only understand language, yousense when a sentence is ungrammatical!!Is raining.!!The child seems sleeping.!!Sally poured the glass with water.!!It’s a flying finches, they are.Sometimes youstill understandwhat was meant!!!Rarely is the question asked: Is our childrenlearning? (a joking George W. Bush)Grammar!! You can also have sentences withoutmeaning that are perceived as grammatical!!Colorless green ideas sleep furiously.!!If we don’t succeed, we run the risk of failure.(a not joking Dan Quayle)!!‘Twas brillig, and the slithy tovesDid gyre and gimble in the wabe:All mimsy were the borogoves,And the mome raths outgrabe.Purdue UniversityPurdue UniversityGrammar!! These properties of language suggest that yourknowledge about language grammar is a basiccomponent of language systems!! It is distinct from both meaning and understanding!! Much of linguistics explores the rules of language!! we are interested in how people perceive grammar!! this is different from the grammar rules you may havelearned in school!»! Which often focus on forming sentences that are easyto understandModern linguistics!! Noam Chomsky used the properties ofgrammar to demonstrate that language isquite different from other types of learningthat might occur!!it’s not like learning to play a piano!!or learning about statistical regularities in theenvironment (stimulus-response)Purdue UniversityPurdue UniversityNonsense sentences!! Think about the sentence!!Colorless green ideas sleep furiously.!! What is the probability that in normal life youwould hear the word “green” follow the word“colorless”?!!it must be close to zero!! But we recognize it as a grammaticallycorrect sentence!Purdue UniversityStatistics!! If you just learned the statisticalcombinations of words, you might thinksomething like this was a grammaticalsentenceHouse to ask for is to earn our living byworking towards a goal for his team in oldNew York was a wonderful place wasn’t iteven pleasant to talk about and laugh hardwhen he tells lies he should not tell me thereason why you are is evidentPurdue UniversityPSY200 Cognitive Psychology 2

Greg Francis 1/6/11Statistics!! The previous paragraph creates coherentgroups of 4 words at a time (generator madesure 4 words were with high probability)!! Maybe by including a larger number of wordsgrouped together you can insure that everysentence is appropriate!! Actually you cannot!!Because sentences have no maximum lengthPurdue UniversityLong-term dependencies!! Language has rules that determine what types ofwords can be used and when!! A word choice early in a sentence can have an effectat the end of a sentenceHow Ann Salisbury can claim that Pam Dawber’s angerat not receiving her fair share of acclaim for Mork and Mindy’ssuccess derives from a fragile ego escapes me.1) “at not receiving ” --> noun “acclaim”2) “anger” --> “derives” (singular)3) “How” --> “escapes” (number)Purdue UniversityLong-term dependencies!! Chomsky demonstrated that long termdependencies can be very long!!Consider “If…then…” and “Either…or…” sentencesIf the girl eats ice cream, then the boy eats hot dogs.Either the girl eats ice cream, or the boy eats hot dogs.Recursion!! In fact, any sentence can go inside the “if…then”part of a sentence!! embed a sentence in a sentence!! Thus the following is a (ugly) valid sentenceEither if the girl eats ice cream, then the boy eats icecream, or if the girl eats ice cream then the boy eats candy.!! recursion cannot be learned by statistics, it has to bebased on rulesPurdue UniversityPurdue UniversityPhrases!! Every sentence is built out of phrasesThe happy boy eats candy.The first three words form a unit calleda noun phrase (NP)The happy boyWhat identifies anoun phrase?This is not the same analysis you didin grammar school!Purdue UniversityPhrases!! All noun phrases obey certain rules!!rewrite rulesNP-->(det)A*Nthe happy boy!!NP -- noun phrasethe boy!!det -- determinant: “the”, “a”, “an” John!!A -- adjectivethe tall slender woman!!N -- noun!!( ) -- optional!!* -- as many as you wantPurdue UniversityPSY200 Cognitive Psychology 3

Greg Francis 1/6/11Phrase tree!! It helps to describe rules as phrase trees!! Specifies both what can be used in thephrase and where it must be usedNPdet A NPhrases!! Similarly, there are rules for all sorts ofphrases in a language!! There may be many ways to rewrite aphrase!S-->NP VPVP-->V NPS -- sentenceNP -- noun phraseVP -- verb phrasethe happy boyPurdue UniversityPurdue UniversityLexicon!! We also need a mental dictionary(lexicon) that specifies parts of speech!!N --> boy, girl, candy, hot dogs, ice cream,…!!V --> eats, likes, bites,…!!det --> a, the, one,…!!A --> lucky, tall,...Purdue UniversityPhrase tree!! With rewrite rules and a mental dictionary,you can create a sentence by linking therules togetherSNPVPVNPdet A NNthe happy boyeatsice creamPurdue UniversitySockets!! In a phrase tree, a phrase is like a componentthat snaps into the right placedetthe!!any appropriate phrase works! (even nonsensephrases)SNPVPVNPA A NNcolorless green idea eatsice creamPurdue UniversityUsefulness!! It is important to appreciate how the phrase treeapproach simplifies the description of language!! Consider how we learn a new word and knowhow to use it!! If you learn that a word is a noun, you canimmediately use that noun in many differentwaysPurdue UniversityPSY200 Cognitive Psychology 4

Greg Francis 1/6/11Learning phrases!! You do not have to relearn the role of the word“boy” for each useThe boy eats candy.Long term dependencies!! Phrase trees have no problems with long-termdependencies and recursion!! The rewrite rules provide the structure neededto insure the right if-then combinationI like the happy boy.I gave the new boy a cookie.The happy boy’s cat eats candy.S--> either S or SS--> if S then SS -- sentenceeither -- the word “either”or -- the word “or”if -- the word “if”then -- the word “then”Purdue UniversityPurdue UniversityPhrase tree!! A phrase tree can handle this type of sentenceSifSSthenS or Sthe boy eats hot dogseitherthe girl eats candythe girl eats ice creamPurdue UniversitySignificance!! Rules and phrase trees allow us to identifyfundamental characteristics about howhumans communicate!! Consider all the ways you mightcommunicate!!Morse code, 0-1’s, English, Spanish, tappingtoes, beeps,..!!an infinite number of ways to create a languagePurdue UniversityLanguage similarity!! All human languages are very similar,compared to the possibilities!! In some sort of language space all our 6000languages are clustered togetherHumanMartianDolphinAlphacentaurianPurdue UniversityLanguage universals!! There are several types of universals!! For example, in English the normal pattern ofsentences is!! Subject-Verb-Object!! (There are exceptions: “A bear he shot.”)!! This pattern is true for most of the world’slanguages!! 98% of languages have the Subject before the Object (theVerb location varies across languages)!! 80% of languages have the Subject before the Verb (theObject location varies across languages)Purdue UniversityPSY200 Cognitive Psychology 5

Greg Francis 1/6/11Language universals!! Most language universals involve co-appearance oflinguistic features!! For example, if a language’s preferred word order isSubject-Object-Verb!! the language is likely to form questions by adding somewords at the end of the question!! If a language’s preferred word order is Subject-Verb-Object (like English)!! the language is likely to form questions by adding somewords at the beginning of the question!! “Where did he…?”, “When did they…?”Conclusions!! Language consists of!!symbols (words)!!grammar (rules)!! Language is best described as phrasetrees!!explains long term dependencies!! Language universalsPurdue UniversityPurdue UniversityNext time!! Words!! Mental lexicon!! Morphology!! Structure!! CogLab on Word superiority due!!! What is the plural of “walkman”?Purdue UniversityPSY200 Cognitive Psychology 6

Professor Greg Francis 8/14/12WordsGrammarPSY 200Greg FrancisThe rules of phrases rules for combining phrases universals for all languagesLecture 28What is the plural of walkman?So why do we have so difficult a timecommunicating with people that speak otherlanguages?Purdue UniversityPurdue UniversityWordsWords are specialEven if all languages have similarrules for combining phrases, they usedifferent wordsWords are symbols that are arbitraryin many respectsWe are better at identifying a word thanan individual letter Word superiority effect--?-E or L--?-E or L dog is nothing like a dog is it rote memorization?» partly, but it is also more than thatLXXXXFELTXXXXPurdue UniversityPurdue UniversityWords are specialNote: context does not give the answer Just knowing there was a word should not help FEET FELTLXXXX--?-E or LFELTXXXXPurdue University--?-E or LWords are specialNevertheless, a letter in a word is better identifiedCogLab class data Results are based on data from 155 participants" Condition ! !Percent correct detections" Target letter in a word "67.4" Target letter in isolation "66.7"CogLab global data" Results are based on data from 39,489 participants" Condition ! !Percent correct detections" Target letter in a word "77.2" Target letter in isolation "76.0"No convincing explanationPurdue UniversityPSY 200: Intro. to Cognitive Psychology 1

Professor Greg Francis 8/14/12Word rulesMorphologyPart of linguistics identifiesthe rules for working withwords (morphology) Show this page to apreschoolerThe child will say wugseven though he has neverseen a wug before This implies that there mustbe a rule for pluralizing nounsThe rules of word formationIn many respects English has a verylimited morphology nouns have two forms verbs have four formsduckducksquackquacksquackedquackingPurdue UniversityPurdue UniversityMorphologyOther languages have many morevariations Italian and Spanish have 50 forms of each verb classical Greek has 350 forms of each verb Turkish has 2 million forms of each verb some languages build entire sentences aroundone complex verbThere are rules for these formsPurdue UniversityMorphologyEnglish can convey this information in as many waysas other languages, but we use grammatical phrasesto do soSimple present tense General truths: Ducks quack. Habitual action: I quack like a duck when I wake up.Present Perfect Progressive To express duration of an action that began in the past, hascontinued into the present, and may continue into the future:The duck has been quacking for two hours, and he hasntfinished yet.Other languages have different verb forms to indicatethese conditionsPurdue UniversityMorphologySuffixesOn the other hand, English morphologyallows one to easily create new wordsfrom old words add suffixes and prefixesteachteachableteacher unteachable teachablenessEnglish has lots of these derivational suffixes-able-age-al-ate-ed-en-ify-ion-ish-ize-an-ant-ance-ary-er-ful-hood-ic-ism-ist-ity-ive-ness-ory-ous-yExamplesof morphemesYou probably donot consciously knowwhat some of thesemean, but yourlanguage system does.Purdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology 2

Professor Greg Francis 8/14/12CompoundingEnglish also allows new words to be createdout of other words and combinations can be combinedtoothtoothbrushbrushunmicrowaveabilityRulesSo what are the rules?One looks to be easy to pluralize a noun, add -sNNstem Ninflection NstemNNinflectionbootyliciousPurdue Universitydog -swug -sPurdue UniversityRulesMore detail on rulesThe creation of compound nouns also followsa simple ruleNstemNstemNstem NstemConsider a rule that creates an adjective out ofa verbOr a noun out of a verbAstemNstemNstemtoothNstembrushdogbiteBy the way, ignore what grammarschool might have taught you, theseare not adjective-noun phrasesVstemcrunchAstemaffix-ableVstemcrunchNstemaffix-erPurdue UniversityPurdue UniversityRootSome morphemes canonly be attached to certaintypes of words a root is a word that cannotbe split into smaller parts some morphemes attachonly to rootsNstemNstemNstemaffix thus, Darwinianisms is aword, but Darwinismians is-ismnotNroot Nrootaffix» -ian must attach to a rootDarwin -ianPurdue UniversityNNinflection-sLexiconTo keep track of what can attach to what,there must be a mental dictionary ofmorphemes -able -er ...» adjective stem affiix; means capable ofbeing Xd;attach me to a verb stem» noun stem affix; means one who Xs;attach me to a verb stemPurdue UniversityPSY 200: Intro. to Cognitive Psychology 3

Professor Greg Francis 8/14/12ExceptionsExceptionsThese exceptions generally come from otherYou can probably think of lots of exceptions tolanguages (with appropriate rules)these types of rules English adopts the words but not the rules many words seem to follow arbitrary rulesThese exceptions tend to be very common wordsPluralization, past tense» mouse, mice teach, taught» leaf, leaves buy, bought» man, men fly, flew drink-drank sink-sank throw-threw ring-rang sit-satblow-blewAll derive from a proto-Indo-European languagethat formed past tense by replacing one vowelPurdue Universitywith anotherPurdue UniversityLanguage familiesMany languages are related to each other and havebroad familiesVery special casesWhat is the plural of walkman? Which is why you see so many similar words across languages walkmans? walkmen? Neither feels quite rightTo answer this question we have to understandhow the word walkman is formed and what it isabout this tells us how to pluralize the wordPurdue UniversityPurdue UniversityHeadsHeads and compoundsMost words have a head that indicates whatthe word is aboutIn English it is always the rightmostmorpheme crunchable a thing that can be x-ed cruncher a thing that does x workman a type of person sawtooth a type of toothThe plural form of a compound word is basedon the plural form of the head of the compoundword toothbrush --> toothbrushesNstem sawtooth --> sawteeth snowman --> snowmen Nstem NstemIs walkman a compound noun? yes, but it is not normal walkmanPurdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology 4

Professor Greg Francis 8/14/12Headless compoundsHeadless compoundsSome compound words are headlessHow can you tell?Sony corporation suggests that the plural ofwalkman is a walkman is not about a type of man thus, the head is not what the word is about this tell us that walkman is more like a new wordthan a compound word (e.g., electricity)A similar analysis explains the plural form ofnamesWalkman Personal StereosToronto Maple Leafs (not Leaves)For headless compounds the irregular pluralform is not appropriateFlorida Marlins (not Marlin) no plural form feels correct because the word hasno headI am sick of all the Mickey Mouses in thisadministration. (not Mickey Mice)Purdue UniversityPurdue UniversityConclusionsNext timeWords follow rulesInterpreting languageParsersMorphology / morphemesMore ambiguityLexiconComputers vs humansExplains characteristics of languageCogLab on Lexical decision due!This is a valid sentence: Buffalo buffalo buffalo buffalobuffalo buffalo buffalo buffalo.Purdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology 5

Professor Greg Francis 1/2/13ParsingPSY 200Greg FrancisLecture 29This is a valid sentence:Buffalo buffalo buffalo buffalo buffalobuffalo buffalo buffalo.Purdue UniversityLinguisticsSo far we have looked at grammar wordsFrom the point of view of generating asentenceToday we look more closely at interpreting asentencePurdue UniversityEliza / EmacsEliza / Emacs Weizenbaum (1966)I am the psychotherapist. Please, describe your problems. Each timeyou are finished talking, type RET twice.I am sad.Why do you say you are sad?My parents wont give me a new car.Tell me something about your family.I have five brothers and two sisters.You seem to dwell on your family.They are important.Is it because of your plans that you say they are important?Purdue UniversityI am the psychotherapist. Please, describe your problems. Each timeyou are finished talking, type RET twice.Hi, my name is Greg.How do you do? What brings you to see me?I am showing a demonstration for class.Why do you say you are showing a demonstration for class?Because I am.Can you elaborate on that?Well, I am a professor.Tell me about your professor.I am the professor!Eliza just picks upon key phrases andgenerates somestock responses.Purdue UniversityParsingWho does what?Eliza does not really understand anything Try it yourself at http://www.manifestation.com/neurotoys/eliza.php3The difference between Dog bites man. Man bites dog.Requires identifying the subject, object, and verbThe system that does this is called a parserYou can learn quite a bit about a sentences meaningby knowing the phrase tree structure of the sentence indicates some aspects of meaningThe green idea eats the girls candy.We know the sentence is about an idea rather than agirl we also know the idea is doing the eatingPurdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology 1

Professor Greg Francis 1/2/13Building phrase treesYou create a sentence with ideas in your headThose ideas are converted into appropriate words andphrase trees to convey those ideasSometimes two different ideas can give rise to thesame sentence leads to ambiguous sentences the parser does not work in the same way as the creatorAmbiguous sentencesConsider the followingYoko Ono will talk about her husband John Lennonwho was killed in an interview with Barbara Walters.Two cars were reported stolen by the Purdue policeyesterday.Tonights program discusses stress, exercise, nutrition,and sex with former Celtic forward Scott Wedman, Dr.Ruth Westheimer, and Dick Cavett.Purdue UniversityPurdue UniversityAmbiguous sentencesAmbiguous sentencesThe writer had in mind a phrase tree likeBut a reader/listener could interpret it likeVdiscussVPNPNsexwithPPNPDick CavettVdiscussVPNPNwithsexPPNPDick CavettNo photoPurdue UniversityPurdue UniversityMentaleseThat two different internal thoughts cangive rise to the same language statementis interesting it suggests that we think in some way that isdifferent from languageParsingParsing is something like building aphrase tree in reverseLets parse through a simple sentenceword by word The dog likes ice cream.S a mentalese, if you willNPVPPurdue Universitydet N Vthe dog likesNPNPurdue Universityice creamPSY 200: Intro. to Cognitive Psychology 2

Professor Greg Francis 1/2/13ParsingTwo problemsOnce every slot is filled, the sentence isparsed a mental click of understandingEach word has its role definedParsing is complicated in two ways (1) Phrases are not always consistent with word order (2) The same spoken sounds are sometimes used forwords with different meanings (noun vs verb vsadjective) and the order of the phrases identifies themeaning (usually)Purdue UniversityPurdue UniversityWord orderWord orderThis sentence is relatively easy to parse,even though it is a complicated sentenceThis sentence is not as easyOne type of phrase is embedded in anotherSPP PP PPRemarkable is the rapidity of the motion of the wing of the hummingbird.SPPThe rapidity that the motion has is remarkable.Purdue UniversityPurdue UniversityWord orderThis sentence is nearly impossibleS PP PP PPThe rapidity that the motion that the wing that the hummingbird has has has is remarkable.Purdue UniversityDifficult sentencesThese sentences are difficult for humans because oflimited memory when a phrase tree includes many unfilled branches of thesame type (PP) the parser becomes confused as to which phrase isassociated with a new word ends up backtracking to sort out the phrases sometimes falls apart (has has has)The grammar generator and the parser are differentthings in your language system these are grammatically correct sentences they are not good sentencesDont you make sentences like thesemake meshow youyour exams!Purdue UniversityPSY 200: Intro. to Cognitive Psychology 3

Professor Greg Francis 1/2/13Word ambiguityWord ambiguityA word by itself is often ambiguousConsider a parser trying to follow the phrase The plastic pencil marks...A differently designed phrase tree handles thenew word The plastic pencil marks...SSdetNPNVPWord pencil isinconsistent withstructure created!detNPANVPWord pencil isconsistent withstructure created!the plastic pencilthe plasticpencilPurdue UniversityPurdue UniversityWord ambiguityBut you run into the same problem with theword marks (noun or verb?) The plastic pencil marks were ugly. (noun) The plastic pencil marks easily (verb)Parsers build phrase trees on the fly, sobacktracking is often required many times it is so fast that we do not notice seems effortlessLexical decisionIt is not effortless and it can be shown with anexperimentThe experiment is a variation of the lexical decisiontask, which you did in CogLabIn the lexical decision experiment, you see a pair ofwords/non-words, and we measure the reaction timefor you to decide if the second word is a word RT is faster if the second word is semantically related to thefirst wordPurdue UniversityPurdue UniversityLexical decisionThe data find that RT is faster if the second word is meaningfullyrelated to the first wordCogLab DataResults are based on data from 180 participants [Global 53,570 participants] Condition ! ! !Reaction time (ms)! Associated words ! !686 [Global: 619]! Unassociated words !685 [Global: 629]! Word then non-word !870! Non-word then word !766! Two non-words ! !871!Evidence of ambiguityWe can apply the lexical decision task to the ambiguityof parsing (Swinney, 1979)Consider the following paragraph, which subjectslistened to Rumor had it that, for years, the government had beenplagued with problems. The man was not surprised when hefound several spiders, roaches, and other bugs in the cornerof his room.The word bugs is ambiguous insects vs surveillance devicesPurdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology 4

Professor Greg Francis 1/2/13Evidence of ambiguityNo one notices the ambiguityBut, give a lexical decision test for words verses nonwords Flashed visually on a screen just after the word was spoken Subjects respond faster for words related to either definition ofbugSentence ambiguitySome sentences never resolve their ambiguities Ingres enjoyed painting his models nude. Visiting relatives can be boring. I saw the man with the binoculars.Context often helps more laterant sew spyfasterslowerfasterPurdue UniversityPurdue UniversitySentence ambiguityInterestingly, people often miss ambiguities insentences Time flies like an arrow.Humans recognize only one interpretationComputer algorithms can find 5 interpretations all grammatically correct!Sentence ambiguityTime flies like an arrow.(1) Time proceeds as quickly as an arrow proceeds.(2) Measure the speed of flies in the same way thatyou measure the speed of an arrow.(3) Measure the speed of flies in the same way that anarrow measures the speed of flies.(4) Measure the speed of flies that resemble an arrow.(5) Flies of a particular kind, time-flies, arefond of an arrow. (Fruit flies like a banana.)Purdue UniversityPurdue UniversityAmbiguity and computersOr consider the following (valid) sentence thatcomputer algorithms can correctly interpret Buffalo buffalo buffalo buffalo buffalo buffalo buffalobuffalo.Heres a hint to make it understandable inprincipleChicago horses (that) Milwaukee cowsintimidate (also) intimidate Cincinnatipigs.SignificanceThese types of results suggest that wordsand grammar are not enough to insurecommunicationIn a certain sense a speaker and listenermust already be agreeing about the topicbefore anything can be communicatedThus, we can understand the followingdiscourse Woman: Im leaving you.Purdue University Man: Who is he?Purdue UniversityPSY 200: Intro. to Cognitive Psychology 5

Professor Greg Francis 1/2/13Schemas / scriptsSchemas / scriptsCognitive devices describe stereotypical properties of a situation e.g., restaurant scene involves table, waiter, drinks, tips,…Giving computers the general knowledge oflife needed to create something like schemasis very difficultFill-in the missing information that is critical forunderstanding language (and events in general) explains why it is difficult to communicate across cultures,even with a common languageThis is why computers do not carry onconversations with youLots of work going on in artificial intelligence toSchemas provide the context to remove the almostaddress this problemconstant ambiguities of languagePurdue UniversityPurdue UniversityConclusionsNext timeUnderstanding languageParsingPhrase trees (in reverse)SpeechPhonemesArticulation / coarticulationAmbiguitiesComputer generated interpretationsMissing information / schemasWhy do we say razzle-dazzle instead ofdazzle-razzel?Purdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology 6

Professor Greg Francis 1/6/11SpeechPSY 200Greg FrancisLecture 30Why do we say “razzle-dazzle” instead of“dazzle-razzle”?Purdue UniversityLanguage!! Many levels!!grammar!!phrases!!words!! All humans, who can, communicate throughspoken language!!how does language depend on speech?!!what are the units of speech?Purdue UniversityIllusions!! When you hear what I say, you think youhear at least!!separate words!!separate syllables!! But you do not!!words actually overlap in the speech signal!!it is nearly impossible to take a speech signaland cut it up into separate wordsPurdue UniversityIllusions!! The “blurriness” of speech explains some long-heldconfusions!! Oronyms (Mondegreens)The good candy came anyways. The good can decay many ways.It’s a doggy-dog world.Purdue UniversityWhy the blur?!! The ear is a bottleneck!!analogous to the critical flicker frequency in theeye!!the ear can distinguish as separate onlyif they are given at less than 20 hertz»!20 clicks per second!!above that, a series of clicks sounds like acontinuous buzzBut...!! Speech is seemingly perceived muchbetter!! Normal speech provides 10 to 15 distinctphonemes each second!! Fast speech is 20 to 30 phonemes persecond!! Artificially fast speech is 40 to 50phonemes per secondPurdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology 1

Professor Greg Francis 1/6/11Phonemes!! pho!neme \'fo-,nem\ n[F phoneme, fr. Gk phonemat-, phonema speechsound, utterance, fr. phonein to sound](ca. 1916):a member of the set of the smallest units of speechthat serve to distinguish one utterance fromanother in a language or dialect, the \p\ of pat andthe \f\ of fat are two different phonemes in English>Phonemes!! Speech is made of phonemes!! Different combinations of phonemescorrespond to different syllables andwords!! We seemingly hear more phonemes thanthe ear can actually handle!!how?Purdue UniversityPurdue UniversityPacking!! If the ear can only distinguish up to 20 sounds persecond!! and we can interpret speech that seems to contain 50phonemes per second!! then the speaker must be combining many phonemestogether to overcome the limits of the ear!! The listener hears the 20 (or so) sounds in a second,but interprets them as more than 20 differentphonemesPacking!! If phonemes are being smashed togetherthere must be some blurriness!!and this can lead to misinterpretations!! This is also why there are few computers toread to the blind!!they do not know how to combine phonemes inthe right wayPurdue UniversityPurdue University!! So what are phonemes?Speech!! All speech is made of sounds!! sound is a pattern of pressure on the ear!! a tuning fork vibrates back and forth to make the sound of apure tone!! Frequency of vibration corresponds to pitch of the sound!! Speech consists of lots of patterns of this sort!! With many different overlapping frequencies!! Lungs pushair out tomake asound!!otherorgansshapesoundPhysiologyPurdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology 2

Professor Greg Francis 1/6/11Example!! Note where your tongue is as you say!!bet butt!!beet bat!! The position of the tongue shapes the vocaltract and makes different sounds!!!this is true for all vowelsExample!! Note what your lips do as you say!! boot book!! The lips add additional frequencies to make differentsounds!! Thus, you can hear someone smile across atelephone!!! Vowels are all distinguished by the shape of the vocaltractPurdue UniversityPurdue UniversityConsonants!! Consonants are more complicated!!different type of control of air flow!! (1) Voicing: vibration of vocal cords!!/b/, /d/, /m/, /w/, /v/ (voiced)!! /p/, /t/, /f/ (not voiced, or unvoiced)!! (2) Place of articulation:!!/d/, /t/ (upper gum)!!/m/, /b/, /p/ (lips)!!/f/, /v/ (lip and teeth)Purdue UniversityConsonants!! (3) Manner of articulation!!/d/, /t/ (stop)!! /m/ (nasal)!!/f/, /v/ (fricative)!! Each consonant is uniquely identified by itsvoice (or not) and its place and manner ofarticulation!!Some languages have a few other characteristicsas well (e.g., tone)Purdue UniversityFun!! Why do we say razzle-dazzle instead ofdazzle-razzle?!!for phrases like this, people always first say theword with a leading consonant that impedes air flowthe leastsuper-duperhelter-skelterharum-scarumhocus-pocuswilly-nillyroly-polyholy molywalkie-talkienamby-pambywing-dingherky-jerky mumbo-jumboPurdue UniversityIt’s arule!Phonemes!! English uses 22-26 (it depends on how you count)combinations of voicing, place, and manner ofarticulation (and 20 vowels)!! Rotokas (Papua New Guinea) uses 6 (and 5 vowels)!! Khoisian (Bushman) uses 141»! Uses clicks as consonants!! No language uses some possible sounds!! raspberries, scraping teeth, squawking,…!! Note, these sounds are used for communication, but not aspart of language!!! Japanese does not distinguish /r/ from /l/Purdue UniversityPSY 200: Intro. to Cognitive Psychology 3

Professor Greg Francis 1/6/11Rules!! To say a word, we must combine phonemes!! In every language there are rules (trees) thatdescribe what phonemes can follow otherphonemes!! Thus, we can identify possible words fromimpossible words!!plast ptak!!vlas rtut!!thole hladCompression!! Moving the tongue (and otherarticulators) around is difficult and takestime!!to say sounds faster, people usecoarticulation!!shape tongue in advanced preparation forthe next phoneme!!this influences the sound of phonemes!!nypip dnom Purdue UniversityPurdue UniversityCoarticulation!! We generally do not notice theseadjustments!!we are tuned to recognize the new soundsas coarticulation!! This is the main reason computers havea hard time recognizing human speech!Purdue UniversityCoarticulation!! Notice that your tongue body is in differentpositions for the two /k/ sounds in!!Cape Cod!! Note too, that the /s/ becomes /sh/ in!!horseshoe!! And /n/ becomes /m/ in!!NPR!! You can enunciate these “correctly”, but incasual speech you do not!Purdue UniversityCoarticulation!! There are rules for how to coarticulate!! When a stop-consonant appears betweentwo vowels, you do not actually stop!!flapping!! slapped --> slapt!! patting --> padding!! writing --> wridingPurdue UniversitySpelling!! We have often observed that written languageis different from spoken language!! George Bernard Shaw (among others)complained about spelling in English!!he noted you could spell “fish” as “g-h-o-t-i”gh -- tough o -- women ti -- nation!! He offered a prize in his will for someone tocreate a good alternative to English spellingPurdue UniversityPSY 200: Intro. to Cognitive Psychology 4

Professor Greg Francis 1/6/11Spelling!! It is true that English spelling does not seem to agreewith pronunciation!! a problem for learning how to read!!! Nor should it!! if words were spelled the way they were pronounced, wewould lose the visual connection between words!! slap --> slapped would become slapt!! write --> writing would become wridding!! National Public Radio --> NPR would become MPROther approaches!! There are other written forms of language that avoidsome of these problems!! The most sensible writtenlanguage is probably theKorean hangul!! Drawn characters indicatehow consonants arepronouncedPurdue UniversityPurdue University!! Speech!! Blurring!! Phonemes!! Articulation!! Coarticulation!! SpellingConclusionsNext time!! Learning language!! Babies!! Children!! Learning a second language!! When should you learn a foreign language?Purdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology 5

Professor Greg Francis 8/15/12Language developmentLanguagePSY 200InstinctiveGreg Francis your brain is wired to work with grammars, words,phrasesLecture 31Learning you do have to learn some specifics for yourWhen should you learn a foreignlanguage?native tongue rules wordsPurdue UniversityPurdue UniversityLearningBabies and phonemesWhat is learned?How does a child learn?How much about language does a childknow?Infants have linguistic skills as soon as theyare born babies are interested in new things attach a tape player to a pacifierWhen have you mastered language? each suck causes the player to play a soundHow do you learn a second language?Repetition of the same sound leads toboredom and fewer sucksWhat do babies do? ba, ba, ba, ba, ba, ba, ba, ba,Purdue UniversityPurdue UniversityBabies and phonemesBabies and phonemesWhen the syllable changes babies suck more often ba, ba, …, pa, pa, pa, pa, pa,…Moreover, they hear things the way adults do you can change the pronunciation (timing) of ba and still hear it asba (CogLab data) babies hear it the same wayBA"PA"Identification taskSame/different taskPurdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology 1

Professor Greg Francis 8/15/12BabiesBabies hear all phonemes, even ones theirparents cannot distinguishBabies, even newborns, do show a preferencefor what will become their native tongue occurs because they hear mothers voice while in thewombMostly prefer the melody, stress, timing French infants like French and Italian equally well playing language backwards keeps many consonantsbut distorts melody (babies are not interested)Language developmentBetween 5-7 months, babies start making sounds clicks, hums, hisses, smacks,…Between 7-8 months babies start babbling insyllables ba-ba-ba-ba-ba neh-neh-neh da-da-da-da-daPurdue UniversityPurdue UniversityBabblingBabbling sounds are the same in all languages patterns are common across languagesBy the end of the first year babies combine syllablesto sound like words neh-nee da-dee meh-nehBabbling is important children who do not babble often show slower speechdevelopment deaf children babble with hands, if parents use signlanguagePurdue UniversityBabblingBabbling teaches child how sequences of musclecombinations lead to different sounds necessary to produce speechBy about 10 months babies learn the sounds of theirnative tongue they can no longer distinguish phonemes that are not part ofthe language Part of learning is forgetting!Purdue UniversityLanguage stagesNearly all children learn language in stages 1) Cooing (first several months) 2) Babbling (~6 months) 3) One word utterances (~1 year) 4) Two-word utterances and telegraphic speech(1-3 years) 5) Basic adult sequences with grammar(~4 years)The rate of learning varies substantiallyPurdue UniversityLearning wordsChildren learn words with ridiculous easeAn average 6 year old knows 13,000 words learned one new word every two waking hours this is without knowing how to read!The average high school graduate knowsabout 60,000 different words (not countingcompound words and such) means that in 17 years of life (not counting the firstone), they learned an average of 10 new wordseach day (one word every 90 waking minutes)Purdue UniversityPSY 200: Intro. to Cognitive Psychology 2

Professor Greg Francis 8/15/12Around 18 monthsChildren learn simple rules of syntax All dry. All messy. All wet. I sit. I shut. No bed.All hell breaks looseAfter mastering 2-word strings, toddlers go crazy onlanguageConsider changes in language (year;month) (2;3) Play checkers. Big drum. I got horn. No pee. See baby. See pretty.Content is similar for all languages objects appear, disappear, move,… people do things, see things,… ask questions, who, what, where,... (2;5) Now put boots on. Where wrench go? What that paperclip doing? (2;7) Ursula has a boot on. Shadow has hat like that. (2;9) Where Mommy keep her pocket book? Show yousomething funny. (2;11) Why you mixing baby chocolate? I finishing drinking allup down my throat.Purdue University (3;1) You went to Boston University? Doggies like to climb up.Purdue UniversityErrorsThree year olds make lots of grammaticalerrors that is because there are lots of opportunities forerrors but pick any particular grammatical rule and youfind most three year olds obey it most of the time this is amazing because there lots of cases that youwould expect would be difficult to learnExpected errorsConsider a child hearing adults talk and howthey might incorrectly apply what they learnOut of 66,000 sentences, children never madethese errorsHe seems happy. -->Does he seem happy?He is smiling. -->Does he be smiling?He did eat. -->He didnt eat.He did a few things. -->He didnt a few things.Purdue UniversityPurdue UniversityErrorsChildren do make errors, but the errors are consistentwith rules of languageChildren often over generalize a rule -s to pluralize a noun» Mouses, leafs -ed to make the past tense of a verb» My teacher holded the baby rabbits and we patted them.» Hey, Horton heared a Who.» I finded Renee.» Once upon a time a alligator was eating a dinosaur and thedinosaur was eating the alligator and the dinosaur was eatenby the alligator and the alligator goed kerplunk.Purdue UniversityOvergeneralizationThese past tense forms sound wrong because Englishhas around 180 irregular verbs inherited from other languages These past-tense forms are not derived from rulesIrregular forms have to be memorized, word by wordIf a child cannot remember (in its lexicon) s/he defaults to the ruleThese errors are for the most difficult parts of alanguage to learn Because they dont follow the normal rulesPurdue UniversityPSY 200: Intro. to Cognitive Psychology 3

Professor Greg Francis 8/15/12OvergeneralizationsWe know this is the most difficult part of language because adultsmake the same kind of mistakes tread - trod strive -strove dwell - dwelt slay-slew rend - rentsmite - smoteSound weird because we do not often hear them many adults regularizethe words treaded, strived, dwelled,slayed, rended, smited thus language changes…!Purdue UniversitySyntax errorsChildren make similar mistakes in applyinggrammatical rulesIn English there is a causative rule that applies tosome verbs and not others takes a verb meaning to do something and converts it to averb meaning to cause to do somethingThus you can say The butter melted. --> Sally melted the butter. The ball bounced. --> Hiram bounced the ball.But you cant say the second of each pair I like sausage. --> I am liked of sausage. I giggled. --> Sally giggled me.Purdue UniversitySyntax errorsChildren over generalize application of the causativerule to inappropriate verbs I go to the bathroom. --> Go me to the bathroom. Aunt Jane died. --> The tiger will come and eat David andthen he will be died and I wont have a little brother anymore. I drink with a cup. --> Yawny Baby - you can push her mouthopen to drink her.Many of these errors would be fine in other languages!The situation is similar to the special past tense verbs This rule must be memorized as applying to some verbs andnot othersSyntax errorsAdults also misapply the causative rule Sparkle your table with Cape Cod classic glass-ware. Well, that decided me. This new golf ball could obsolete many golf courses. If she subscribes us up, shell get a bonus. Boiler up!Childrens errors tend to track the more difficultaspects of a language, relative to other languages Adults make the same kinds of mistakes for still more difficultto remember casesPurdue UniversityPurdue UniversitySecond languageIt is difficult (and rare) for an adult to becomefluent in a second language children do it easilyWhat accounts for the difference? most likely it is age there seems to be a critical period during whichlanguage can be learned beyond age six (or so) it becomes more difficult to learna language (first or second)Second languageHigh school and college (or later) is just about theworst time to try to learn a second languageIt should be in kindergarten or preschoolImmigrants who arrive after age 6 may never fullylearn a second language Children who fail to learn any language by age 6never do they might create a pidgin of some sortPurdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology 4

Professor Greg Francis 8/15/12ConclusionsNext timeLanguage developmentStages of learningErrorsSecond languageLanguage & brainBrocas aphasiaWernickes aphasiaAnomiaLanguage ability of chimpsWhats the big deal about Nim Chimpsky?Purdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology 5

Professor Greg Francis 8/17/11Language and the brainPSY 200Greg FrancisLecture 32What’s the big deal about NimChimpsky?Language!! Properties!!grammar!!phrases!!words!! Instinct!!different from other types of learning!!special areas in the brain related to language!!evolution: can similar brains learn language?Purdue UniversityPurdue University!! Some stroke patientsshow agrammaticalspeech!! Seem to know whatthey want to sayBroca’s aphasia!!But are unable to say itRighthemisphereLefthemisphereBroca’s aphasia!! Some stroke patients showagrammatical speech!! repetition!! short sentences!! true for both written and spoken!! no problem controlling mouth»! e.g. blowing out candlesDo you drivehome on weekends?Why, yes…Thursday,er, er, er, no, erFriday…Barba-ra…wife…and, oh, car…drive…purnpike…you know…reset and…teevee.Purdue UniversityPurdue UniversityBrain damageBroca’s aphasia!! Broca’s area ==> Broca’s aphasia!! Wernicke’s area ==> Wernicke’s aphasia!! Mr. Ford!! omitted endings (-ed, -s)!! omitted function words (or, be, the)!! skipped function words when reading (or, be, the) but readsimilar sounding words (oar, bee)!! named objects and recognized names!! high (nonverbal) IQPurdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology 1

Professor Greg Francis 8/17/11Broca’s aphasia!! Difficulty getting ideas across!! Patient BL was asked to describe this pictureB.L.: Wife is dry dishes. Waterdown! Oh boy! Okay Awright.Okay ...Cookie is down...fall, andgirl, okay, girl...boy...um...Examiner: What is the boydoing?B.L.:Cookie is...um...catchExaminer: Who is getting thecookies?B.L.: Girl, girlExaminer: Who is about to falldown?B.L.: Boy...fall down!Purdue UniversityBroca’s aphasia!! Could understand questions if gist couldbe deduced from content words!!Do you use a hammer for cutting?!!Does a stone float on water?!! Failed to understand anything requiringgrammatical analysis!!The lion was killed by the tiger, which one isdead?Purdue UniversityBroca’s area!! Plays a role in learningthe rules of a language!! Musso et al. (2003)!! fMRI while subjects judgewhether Italian sentencesare grammatically corrector not!! At start, subjects did notknow rules of Italian!! Broca’s area is activated!! Signal correlates withcorrect identificationRighthemispherePurdue UniversityLefthemisphereWernicke’s aphasia!! Other stroke patientsalso showagrammatical speech!! Seem to be able to saythings!!But what they say isalmost meaninglessRighthemispherePurdue UniversityLefthemisphere!! Patients showWernicke’s aphasia!!poor comprehension!!poor vocabulary!!“empty” speechWhatbringsyou to thehospital?Boy, I’m sweating, I’m awful nervous, you know, once ina while I get caught up, I can’t mention the tarripoi, amonth ago, quite a little, I’ve done a lot well. I impose alot, while on the other hand, you know what I mean, Ihave to run around, look it over, trebbin and allthat sort of stuff...Purdue University!! Difficulty getting ideasacrossWernicke’s aphasiaH.W.:First of all this is falling down, just about, and is gonna fall downand they're both getting something to eat...but the trouble is this is gonnalet go and they're both gonna fall down...but already then...I can't seewell enough but I believe that either she or will have some food that'snot good for you and she's to get some for her too...and that you get itand you shouldn't get it there because they shouldn't go up there and getit unless you tell them that they could have it. and so this is falling downand for sure there's one they're going to have for food and, and didn'tcome out right, the uh, the stuff that's uh, good for, it's not good for youbut it, but you love it, um mum mum (smacks lips)...and that sothey've...see that, I can't see whether it's in there or not.Examiner:Yes, that's not real clear. What do you think she's doing?H.W.:But, oh, I know. She's waiting for this!Examiner:No, I meant right here with her hand, right where you can'tfigure out what she's doing with that hand.H.W.:Oh, I think she's saying I want two or three, I want one, I think, Ithink so, and so, so she's gonna get this one for sure it's gonna fall downthere or whatever, she's gonna get that one and, and there, he's gonna getone himself or more, it all depends with this when they fall down...andwhen it falls down there's no problem, all they got to do is fix it and goright back up and get some more.Purdue UniversityPSY 200: Intro. to Cognitive Psychology 2

Professor Greg Francis 8/17/11Wernicke’s aphasia!! Most aphasias involve damage to morethan just one specific areaAnomia!! Damage around Wernicke’s area produces a deficit in theability to name things!! e.g., after a stroke in this area CB cannot retrieve nouns he wantsto useC.B. Uh, well this is the...the...of this. This and this and thisand this. These things going in there like that. This is...thingshere. This one here, these two things here. And the other onehere, back in this one, this one...look at this one.Examiner Yeah, what's happening there?Purdue UniversityC.B. I can't tell you what that is, but I know what it is, but Idon't know where it is. But I don't know what's under. I knowit's you couldn't say it's ... I couldn't say what it is. I couldn'tsay what that is. This shu-- that should be right in here. That'svery bad in there. Anyway, this one here, and that, and that's it.This is the getting in here and that's the getting around here,and that, and that's it. This is getting in here and that's thegetting around here, this one and one with this one. And thisone, and that's it, isn't it? I don't know what else you'd want.Purdue UniversityAnomia!! Sometimes anomia can be remarkablespecific!! Some patients have difficulty with only certaintypes of nouns!!concrete vs abstract (chair vs trust)!!nonliving vs living (table vs dog)!!animals and vegetables vs food and body parts!!colors!!proper namesPurdue UniversityBrain and language!! Recall that the left side of the brain is more involvedin language than the right side!! Broca’s and Wernicke’s areas are on the left hemisphere!! However, the right hemisphere can also work withlanguage!! left handed people!! hemispherectomies (age matters!)Purdue UniversityBrain and evolution!! We’ve argued that language is an evolved instinct!! differences in brains account for differences in abilities!! One might hope to find proto-language abilities in“close” animals to humans!! Chimpanzees, apes!! Anatomically, there are many similarities betweenhuman brains and apes and chimpanzeesPurdue UniversityBrain and evolution!! Cantalupo & Hopkins (2001)!! Broadmann’s area 44 (part of Broca’s area in humans)!! Compare area on left and right hemispheres»! Larger on left for humans and apeschimpanzeesgorillasbonobosPurdue UniversityImplieshemisphereasymmetriesthat underlaylanguagebeganat least 5million yearsagoPSY 200: Intro. to Cognitive Psychology 3

Professor Greg Francis 8/17/11Chimpanzee language!! In the 1960s several research groups reportedteaching chimpanzees American Sign Language(ASL)!! after failure to teach spoken language!! other groups taught chimps to press symbols on a computerkeyboard or string magnetized plastic shapes on a board!! Claimed to teach chimps hundreds of words!! and chimps created new compound words»! swan -> water bird»! stale Danish -> cookie rock»! See videoProblems!! Just like with Eliza (the computer therapist) it iseasy to attribute language ability where it doesnot really exist (9 month old children)!! You can teach an animal a lot using simpleconditioning tricks!! Researchers were quick to excuse mistakesas “play”, “jokes”, “puns”, “metaphors”,...Purdue UniversityPurdue UniversityWord counts!! For example, a deaf student on one research teamlater commented that she saw much fewer signs thanthe non-deaf students!! Like!! seems the researchers counted almost any hand movementas a sign!! scratch --> “scratch”!! pointing --> “you”!! finger to mouth --> “drink”!! hugging --> “hug”!! reaching --> “give”!! kissing --> “kiss”Purdue UniversityNim Chimpsky!! A relative of other “signing” chimps!!with more careful judging probably learnedapproximately 25 words!!moreover, the “signs” were variations of the naturalmovements of chimps in the wild!! The chimps did not learn ASLPurdue UniversityGrammar!! Chimps failed to learn the rules of ASLgrammar!!unable to understand complex signs!! Seemingly able to understand complexsentences!!Would you please carry the cooler to Penny?!! But really, the chimp need only understand twowords: cooler and Penney!!the rest can be guessed!Purdue UniversityGrammar!! Likewise, the chimps never produced complex sentences!! They tended to “say” things like the following!! Nim eat Nim eat.!! Drink eat me Nim.!! Tickle me Nim play.!! Me eat me eat.!! Me banana you banana me you give.!! Banana me me me eat.they communicatebut not with reallanguage!! Give orange me give eat orange me eat orange give me eatorange give me you.Purdue UniversityPSY 200: Intro. to Cognitive Psychology 4

Professor Greg Francis 8/17/11Evolution!! Note, it would have been interesting if chimpscould learn language!!and not inconsistent with the idea that we have alanguage instinct!! But the failure of chimps to learn languagedoes not go against the idea that languageevolved in humans!!as some people have proposedPurdue UniversityEvolution!! Chimps are the closest evolutionary relatives ofhumans!!so if any non-human animal could learn language itwould probably be chimps!! But in evolutionary history, chimps and humanssplit from a common ancestor millions of yearsago!! Humans evolved a language skill and chimpsdid notPurdue UniversityConclusions!! Language and the brain!! Broca’s aphasia!! Wernicke’s aphasia!! Anomia!! Chimps!! Review for exam 4!! After exam 4Next time!! Decision making!! Framing effects!! Risks!! Alternatives!! What every consumer should know before they buy.Purdue UniversityPurdue UniversityPSY 200: Intro. to Cognitive Psychology 5

Greg Francis 1/7/11Decision makingPSY 200Greg FrancisLecture 33What every consumer should knowbefore buying.Purdue UniversityDecision making!! We have to make lots of choices!!course selections!!elections!!housing!!job!!cancer treatment!! What affects our choices?!! How do we make choices?Purdue UniversityTopics!! Influences on decision making!!framing effects!!risks!!alternatives!!loss aversion!! Effects are often relatedFraming effects!! Your decisions are influenced by the way aset of choices is presented!! The child custody problem!!two versions, essentially the same!!lead to different choicesPurdue UniversityPurdue UniversityVersion 1: award frame!! Imagine that you serve on the jury of an only-child solecustodycase following a relatively messy divorce. The factsare complicated by ambiguous economic, social, andemotional considerations, and you decide to base yourdecision entirely on the following few observations. To whichparent would you award sole custody of the child?!! Parent A: average income, average health, average workinghours, reasonable rapport with child, relatively stable sociallife!! Parent B: above-average income, very close relationship withchild, extremely active social life, lots of work-related travel,minor health problems64%Version 2: deny frame!! Imagine that you serve on the jury of an only-child solecustodycase following a relatively messy divorce. The factsare complicated by ambiguous economic, social, andemotional considerations, and you decide to base yourdecision entirely on the following few observations. To whichparent would you deny sole custody of the child?!! Parent A: average income, average health, average workinghours, reasonable rapport with child, relatively stable sociallife!! Parent B: above-average income, very close relationship withchild, extremely active social life, lots of work-related travel,minor health problems55%Purdue UniversityPurdue UniversityPSY200 Cognitive Psychology 1

Greg Francis 1/7/11Framing effects!! So Parent B is the choice to award custodyand to deny custody!!but one necessarily precludes the other!!! Subjects are biased by the task at hand!!focus on different characteristics depending onwhether they are considering awarding ordenyingFraming effects!! Your decisions are influenced by the way aset of choices is presented!! The Asian disease problem!!two versions, essentially the same!!lead to different choicesPurdue UniversityPurdue UniversityVersion 1: Saving frame!! Imagine that the US is preparing for the outbreak ofan unusual Asian disease, which is expected to kill600 people. Two alternative programs to combatthe disease have been proposed. Assume that theexact scientific estimates of the consequences ofthe programs are as follows:!! If program A is adopted, 200 people will be saved.72%!! If program B is adopted, there is a 1/3 probabilitythat 600 people will be saved and a 2/3 probabilitythat no people will be saved.Version 2: Dying frame!! Imagine that the US is preparing for the outbreak ofan unusual Asian disease, which is expected to kill600 people. Two alternative programs to combatthe disease have been proposed. Assume that theexact scientific estimates of the consequences ofthe programs are as follows:!! If program A is adopted, 400 people will die.!! If program B is adopted, there is a 1/3 probabilitythat nobody will die and a 2/3 probability that 600people will die. 78%Purdue UniversityPurdue UniversityChoices and framing!! The two problems are essentiallyidentical, except that the choices arephrased differently!!200 people saved = 400 people dead!!2/3 probability that no one is saved = 2/3probability that 600 will die!! But the phrasing makes a difference inthe choices of subjects!!why?Purdue UniversityRisks!! Risk corresponds to those events thatoccur with probability!!will I like the next movie starring Matt Damon?!!will I live to be 50?!!will the dice show double sixes?!! Events that occur with certainty are withoutrisk!!the sun will rise tomorrow!!I will be older tomorrow!!I will give you an A if your grade is 90 or abovePurdue UniversityPSY200 Cognitive Psychology 2

Greg Francis 1/7/11Risk!! Humans sometimes prefer risky optionsover non-risky options!!and vice-versa!! When the choices are perceived as losses!!subjects tend to be risk-seeking!! When the choices are perceived as gains!!subjects tend to be risk-averse!! Decision making is open to manipulation!!subjects can contradict themselvesPurdue UniversityRisk: monetary choices!! Assume yourself richer by $300 than you aretoday. You have to choose between72%!! A) a sure gain of $100.!! B) 50% chance to gain $200 and 50% chanceto gain nothing.!! Subjects tend to prefer the sure gain!!risk averse with perceived gainsPurdue UniversityRisk: monetary choices!! Assume yourself richer by $500 than you aretoday. You have to choose between!! A) a sure loss of $100.!! B) 50% chance to lose nothing and 50%chance to lose $200.64%!! Subjects tend to prefer the risky option!!risk seeking with perceived lossesPurdue UniversityNotice!! Selecting A) in either situation means you endup with $400!!$300 + $100!!$500 - $100!! Selecting B) in either situation means you endup with either $500 or $300!!$300 + $200 or $300 + $0!!$500 - $0 or $500 - $200!! People do not just look at the “bottom line”!!which is why businesses emphasize that approachPurdue UniversityAlternatives: version 1!! Imagine you are shopping for a new car and havenarrowed down your choices to three models.According to a consumer magazine, the cars’ ridequality (RQ) and gas mileage (GM) are rated asModel RQ GMAsteroid 100 27Bravo 80 33Comet 100 21!! Which car do you select?69%29%2%Alternatives: version 2!! Imagine you are shopping for a new car and havenarrowed down your choices to three models.According to a consumer magazine, the cars’ ridequality (RQ) and gas mileage (GM) are rated asModel RQ GMAsteroid 100 27Bravo 80 33Clarion 60 33!! Which car do you select?19%79%2%Purdue UniversityPurdue UniversityPSY200 Cognitive Psychology 3

Greg Francis 1/7/11Notice!! Subjects hardly ever select the Comet or theClarion!!you might think they do not enter the decisionmaking process at all!!!but they do!! The comparison of Asteroid and Comet clearlyfavors the Asteroid!!it is less clear for the Bravo and Comet!!it is the reverse for ClarionPurdue UniversityConsumer beware!! Stores are very aware of this type ofbehavior!! Thus, they often stock merchandise for thesole purpose of influencing your purchasingbehavior!!usually towards a more expensive model!! Likewise companies make low-end modelssimply to bias you toward higher endmodels and against the competitionPurdue UniversityLoss aversion!! Another general property of decision making is thatpeople tend to be more sensitive to losses than togains!! thus people rarely take an “even-bet”!! the loss of $10 is more significant than the gain of $10!! This is also why fans of sports teams think thereferees treat their team unfairly!! In a game of basketball, each team will!! have fouls called on them when they shouldn’t have (a loss)!! Commit fouls that are not called (a gain)!! But the gains don’t count as much as the losses!! So in a truly fair game both teams (and their fans) feel as ifthey were treated unfairly!! of course one team wins, so it feels that it overcame theinjusticePurdue UniversityLoss aversion!! The same phenomenon ruins many marriages/relationships!! When your partner does something for you (a gain) itdoesn’t count as much as when your partner doessomething against you (a loss)!! Thus, you perceive your relationship as overall not beingworth the trouble (even if your partner is good as often asbad)!! That’s why therapists suggest that in successfulrelationships people must learn to forgivePurdue UniversityLoss aversion!! By definition a choice is a loss or a gaindepending on where you start!! As a result, loss aversion dramatically affectsmany types of choices by magnifying thosecharacteristics of a choice that leads to aperceived loss!! Consider choosing a jobPurdue UniversityChoosing a job: 1!! You have decided to leave your current job. It islocated so far away from your apartment that itrequires an 80-minute commute each way. But you dolike the fact that your job involves much pleasantsocial interaction with your coworkers. Your search fora new job has given you two options and now youmust choose between them. Which job would youprefer?!! Job A: Limited contact with others, commuting time 20minutes.!! Job B: Moderately sociable, commuting time 60 67%minutes.Purdue UniversityPSY200 Cognitive Psychology 4

Greg Francis 1/7/11Choosing a job: 1!! You have decided to leave your current job. The jobinvolves only a ten-minute commute, which you ratherlike. But your job leaves you isolated from coworkersfor long periods of time. Your search for a new job hasgiven you two options and now you must choosebetween them. Which job would you prefer?!! Job A: Limited contact with others, commuting time 20minutes.70%!! Job B: Moderately sociable, commuting time 60minutes.Loss aversion!! In each case the subjects tend to choose the optionthat produces the least loss! keep sociable coworkers in version 1! minimizing commuting time in version 2!! Note, this means subjects are not just choosing whatthey perceive to be the best job overall (again, notlooking at the bottom line)! but are instead choosing the best job relative to the currentsituation!! a very strange phenomenon!Purdue UniversityPurdue UniversityConclusions!! Influences on decision making!! Framing effects!! Risk aversion (perceived gains)!! Risk seeking (perceived losses)!! Loss aversion!! Problem solving!! Expertise!! Analogy!! Set effects!functional fixedness!! InsightNext time!! What does that “aha!” feeling mean?Purdue UniversityPurdue UniversityPSY200 Cognitive Psychology 5

Greg Francis 1/7/11Problem solvingPSY 200Greg FrancisLecture 34What does that “aha” feeling mean?Purdue UniversityProblem solving!! A hallmark of intelligence!!often used as a definition of intelligence!! Seem to get something from nothing!! We will not explain exactly how it happens!!but we can look at some characteristics ofproblems and problem solving»!what makes for an easy (or hard) problem?»!what makes for a good (or bad) problemsolver?Purdue UniversityTopics!! We will look at a number of factorsthat influence our ability to solveproblems!!expertise!!analogy!!set effects»!priming»!incubation»!functional fixedness!!insightPurdue UniversityExperts!! Some people learn how to solve particulartypes of problems!! What makes an expert different from anovice?!! Experts know how to describe problems!!other than that, there seems to be nofundamental difference (even for geniuses!)Purdue UniversityAn example of experts!! Chi, Feltovich & Glaser (1981)!! Take second year physics students (novices) and ask them toclassify a bunch of physics problems!! they tend to group them by surface similaritiesAn example of experts!! Have experts (PhDs) classify the sameproblems!!grouped according to how to solveExpert 2: “Conservation of energy”Novice 2: “Angular velocity; momentum, circular things”Novice 3: “Rotational kinematics, angular speeds, angularvelocities”Novice 6: “Problems that have something rotating, angularspeed”Novice 1: “These deal with blocks on an inclineplane”Novice 5: “Inclined plane problems, coefficient offriction”Novice 6: “Blocks on inclined planes with angles”Expert 3: “Work-energy theorem. They are allstraight-forward problems.”Expert 4: “These can be done from energyconsiderations. Either you should know the principleof conservation of energy, or work is lostsomewhere.”Expert 2: “These can be solved by Newton’s second law”Expert 3: “F=ma; Newton’s second law.”Expert 4: “Largely use F=ma; Newton’s second law.”Purdue UniversityPurdue UniversityPSY200 Cognitive Psychology 1

Greg Francis 1/7/11Chess expertsChess experts!! Chase & Simon (1973)game!! Chase & Simon (1973)random!! Show subjects a chess board andthen clear it!! have subjects recall positions ofthe chess pieces on the board!! Master players are better thanbeginners when the pieces arepositioned as in the middle of areal game!! Experts have schemas that allowthem to organize the piecepositions!! They only need to remember theschemaCorrect pieces3025201510501 2 3 4 5 6MasterBeginner!! Show subjects a chessboard and then clear it!! have subjects recallpositions of the chesspieces on the board!! Master players are worsethan beginners when thepieces are positionedrandomly!! Experts try to use theschemas, but they end upmisremembering the actualpiece positionsCorrect pieces25201510501 2 3 4 5 6MasterBeginnerTrialPurdue UniversityTrialPurdue UniversityExpert schemas!! In general, experts have lots of problemsolving schemas specific to their domain ofexpertise!!given such and such; do such and such!!allows them to organize information in a way thatallows for easy recall and easy use!! Expertise in one domain does not transfer toanother!!except for especially useful skillsPurdue UniversityAnalogy: Attack-Dispersion Story!! The many roads to a dictator’s fortress aremined so that small groups of men maypass, but a large group will be destroyed. Ageneral knows that his army can defeat thefortress if he can get his entire army toattack at once, but he cannot take his armydown a single road all at once without losingtoo many men.!! What should he do?Purdue UniversitySolution!! The general should split up his army intomany smaller units!!each unit takes a separate road to converge onthe fortress simultaneously and defeat thedictatorfortressPurdue UniversityAnalogy: Parade-dispersion!! A dictator wants to show off his army so hetells a general to plan a parade of the army. Hedemands that the general insure that the armyis seen and heard across the entire kingdomsimultaneously. The dictator also demands thatthe parade be the most impressive ever at thefortress. Splitting up the army would allow it tobe seen everywhere, but would make thedisplay at the fortress unimpressive.!! What should the general do?Purdue UniversityPSY200 Cognitive Psychology 2

Greg Francis 1/7/11Analogy!! The solution is the same!!split up the army and have them converge tothe fortress from different parts of the kingdom!! Glick & Holyoak (1980)!!subjects read stories like these and wereasked to solve the problems!!even when shown one solution and told that itcould be applied by analogy to another!!subjects used analogies only 20% of the timeAnalogy!! Analogies are actually very difficult to apply!!need to identify what is common between twoproblems!! Analogies are often applied after twoproblems are solved and well understood!!it is then easier to see what is common!! One of the problems handed out can besolved by analogy to these two problemsPurdue UniversityPurdue UniversitySet effects!! “Set” refers to “mind set” (or something likethat)!!negative set: bias toward solving a problem makesit more difficult!!positive set: bias toward solving a problem makesit easier!! You can be biased by lots of things!!problem statement!!previous methods of reaching solution!!general knowledgePurdue University!! Buddhist monk problemPast experience!! One morning, exactly at sunrise, a Buddhist monk began toclimb a tall mountain. The narrow path, no more than a footor two wide, spiraled around the mountain to a glitteringtemple at the summit. The monk ascended the path with anaverage speed of 3 mph. He reached the temple shortlybefore sunset. After several days of fasting and meditationhe began his journey back along the same path, starting atsunrise and walking an average speed of 5 mph. Is there aspot along the path that the monk occupied on both trips atprecisely the same time of day?Purdue UniversitySelf-imposed limits!! Nine dots problem!!Draw four straight lines, passing through all nineof these dots, without lifting your pencil from thepage.Set effects: Representation!! Mutilated chessboard!!62 squares remain!!can 31 dominoes cover them?!! Easy solution!Purdue UniversityPurdue UniversityPSY200 Cognitive Psychology 3

Greg Francis 1/7/11Set effects: Priming!! Safren (1962)!! Unscramble the following anagrams!! 12.2 seconds per wordKOBO CTURK STTESet effects: Incubation!! Cheap necklace problem!! Opening a link costs $2!! Closing a link costs $3!! Go from given to goal state for no more than $15CHUOC ANCYD DRINEF7.4 seconds per wordKMLI GRAUS RECMAFOEFCE TEESW IKRDNPurdue UniversityPurdue UniversitySet effects: Incubation!! Silveira (1971)!! Control: Work on problem for half an hour!!55% solve problem!! Exp A: Work on problem for half an hour in15 minute sections, half-hour distracter task!!64% solve problem!! Exp B: Work on problem for half an hour in15 minute sections, 4 hour distracter task!!85% solve problemPurdue UniversitySet effects: Incubation!! Subjects often get stuck using an approachthat goes nowhere (set)!! After a break they are more likely to try adifferent approach!! Brainstorming tries to avoid set effects byallowing free “dreaming” of solutions!!most of the ideas are worthless, but the approachis still beneficialPurdue UniversitySet effects: Functional fixedness!! How can you tie two strings together?!! Dunker (1945)!!“MacGyver”Purdue UniversityInsight!! Intuitively, we sometimes feel as if we have astrong insight into a problem and its solutionbecomes obvious!!the “aha” feeling!!is it real?!!what does it correspond to?!! Using problems like the “Bronze coin” andthe “Tree planting” problems!!subjects judge their progress with a “warmth”ratingPurdue UniversityPSY200 Cognitive Psychology 4

Greg Francis 1/7/11Insight!! Warmth stays mostly steady, right up toproposing a solution!!the “aha” feelingUnfortunately,the feeling doesnot necessarilyindicate acorrect solutionConclusions!! Effects on problem solving!! Expertise!! Analogy!! Set effects!! Insight!! Solve remaining problems!! Tumor problemPurdue UniversityPurdue UniversityNext time!! Wrapping up the course!! Other courses to take/avoid!! Paths to pursue!! Graduate school!! Advice for further explorationPurdue UniversityPSY200 Cognitive Psychology 5

Greg Francis 8/15/12Now what?CoursePSY 200Greg FrancisNeurophysiology brain characteristics neural networksLecture 35 perceptionMemoryAdvice for further exploration characteristics theoriesPurdue UniversityPurdue UniversityCourseLanguage/LinguisticsWhat else?Most of psychology requires experimentation,you need characteristicsCognition decision making problem solving PSY 201: Introduction to quantitative topics inpsychology (statistics) PSY 203: Introduction to research methods inpsychology (experimental methods)More statistics PSY 202 Introduction to Quantitative Psychology STAT 225 Introduction to Probability Models STAT 311 Introduction to Probability STAT 350 Introduction to StatisticsPurdue University STAT 511 Statistical MethodsPurdue UniversityOther coursesResearch PSY 355: Stereotyping & prejudice PSY 577 (IE 577): Human factors inengineering PSY 337: Social cognitionIndustrial engineering may have otherrelevant courses Look for human factorsPSY 390 Research in… Actively participate in a research laboratory Details vary dramatically across labs Advisors can identify some positions Talk to faculty about possibilitiesResearch Focused Honors program 3 semester sequence (starts Spring of penultimateyear) Design and carry out your own research study (withguidance from a faculty member)Purdue UniversityPurdue UniversityPSY200 Cognitive Psychology 1

Greg Francis 8/15/12Useful backgroundUseful backgroundComputersMathematics Most experiments are run on computers Models are simulated on computers Learn to program in a computer language» MatLab, C / C++, Java, Basic Possible courses CS 15800 C Programming CS 17700 Programming With Multimedia Objects CS 18000 Problem Solving/Object-Oriented Programming Most psychologists have little mathematicalbackground But it is especially useful for cognitive psychology Take as much mathematics as you can, especially» Calculus (MA 161, 165 or 223)» Finite (discrete) mathematics (not easy to get at Purdue WestLafayette)» Linear (matrix) algebra (MA 262, 265)» Differential equations (MA 266) CS 24000 Programming In C CNIT 15500 Introduction to Object-Oriented Programming CNIT 17500 Visual ProgrammingPurdue UniversityPurdue UniversityBrain characteristicsFurther study PSY 222: Introduction to behavioral neuroscience SLHS 304: Anatomy and physiology of the speech and hearingmechanism PSY 320: Behavioral neuroscience of sensation & arousal PSY 322: Neuroscience of motivated behavior PSY 324: Introduction to cognitive neuroscience PSY 512: Neural systems PSY 390s with Davidson, Kemmerer, Kinzig, Chester, Powley,Swithers, KaganovichThere is a behavioral neuroscience concentration inpsychology More natural sciences than typical psych degreePurdue UniversityNeural networksFurther study PSY 514: Intro. to mathematical psychology (sometimes) Artificial intelligence in computer science or electricalengineering PSY 390 with FrancisPerception PSY 310: Sensory & perceptual processes Artificial intelligence in computer science or electricalengineering PSY 511: Psychophysics PSY 520: Attention & performance PSY 390 with Dzhafarov, Francis, Pizlo, ProctorPurdue UniversityFurther studyFurther studyMemory:Language (many courses in Speech, Language, andHearing Sciences - SLHS) PSY 311: Human learning & memory SLHS 227: Elements of linguistics PSY 314: Introduction to learning SLHS 309: Language development PSY 410: Animal memory & cognition PSY 518: Memory & cognition PSY 390 with Karpicke, Nairne, Schweickert SLHS 401: Language & the brain PSY 426: Language development PSY 526: Psycholinguistics PSY 390 with Hollich, Kemmerer, Kaganovich, Alex Francis(SLHS)Purdue UniversityPurdue UniversityPSY200 Cognitive Psychology 2

Greg Francis 8/15/12Further studyConclusionsProblem solving & decision making PSY 285: Consumer behaviorCourses to consider if you liked thiscourse PSY 318: Problem solving & decision making PSY 514: Introduction to mathematicalpsychologyHow to get involved in research PSY 390 with Pizlo, Proctor, SchweickertPurdue UniversityPurdue UniversityNext timeReview for final examPurdue UniversityPSY200 Cognitive Psychology 3