Contents

32 PART I: General IssuesBOX 2.1CAN DOGS DETECT CANCER? ONLY THE NOSE KNOWSResearch on methods to detect cancer took aninteresting turn in 2004 when investigators reportedthe results of a study in the British MedicalJournal demonstrating that dogs trained to smellurine samples successfully detected patients’bladder cancer at rates greater than chance (Williset al., 2004). This research followed up manyanecdotal reports in which dog owners describedtheir pets as suddenly overprotective or obsessedwith skin lesions prior to the owners’ being diagnosedwith cancer. Interest in the story was sogreat that similar demonstrations were conductedon television programs such as 60 Minutes.Skeptics, however, cited the example ofClever Hans to challenge the findings, arguingthat the dogs relied on researchers’ subtle cuesin order to discriminate samples taken fromcancer vs. control patients. Proponents of thestudy insisted that the researchers and observerswere blind to the true status of the samples socould not be cuing the dogs. More recent studiessuggest mixed results (e.g., Gordon et al., 2008;McCulloch et al., 2006). Researchers in this newarea of cancer detection have applied for researchfunding to conduct more experiments. We nowawait the results of these rigorous studies to tellus whether dogs can, in fact, detect cancer.Key Conceptmanipulates in order to determine their effect on behavior are called theindependent variables. 1 In the simplest of studies, the independent variablehas two levels. These two levels often represent the presence and the absence ofsome treatment, respectively. The condi tion in which the treatment is present iscommonly called the experimental condition; the condition in which the treatmentis absent is called the control condition. For example, if we wanted tostudy the effect of drinking alcohol on the ability to process complex informationquickly and accurately, the independent variable would be the presence orabsence of alcohol in a drink. Participants in the experimental condition would1Sometimes the levels of the independent variable are selected by a researcher rather than manipulated.An individual differences variable is a characteristic or trait that varies across individuals; forexample, sex of the participants (male, female) is an individual differences variable. When researchersinvestigate whether behavior differs according to participants’ sex, they select men and womenand examine this factor as an individual differences variable. As we will see in Chapter 6, there areimportant differences between manipulated and selected independent variables.