Givaudan-Roure Lecture - Association for Chemoreception Sciences

101 Poster [ ] Olfactory CNS Physiology and Coding
PET ACTIVATIONS DURING SMELLING OF
ANDROSTENONE: OSMICS VERSUS ANOSMICS
Boyle J.A. 1, Zatorre R. 1, Pause B. 2, Jones Gotman M. 1 1McGill
University, Montreal, Quebec, Canada; 2Christian-Albrechts-
University, Kiel, Germany
We used positron emission tomography to evaluate the neural
correlates involved in perception of the endogenous steroid
androstenone. Our aim was to investigate whether differences exist in
brain activations between subjects with high sensitivity (low thresholds)
for androstenone and those with a specific anosmia. We screened 87
subjects using a staircase procedure to establish individual thresholds.
Twelve subjects unable to detect the steroid in its crystal form were
selected for an anosmic group, and 12 subjects capable of consistently
detecting a 1.4 x 10-7 M solution of androstenone comprised an osmic
group. Subjects were scanned during passive smelling of phenyl ethyl
alcohol (PEA), pyridine, androstenone (AND) or the diluant (for
baseline). Our first analysis compared brain responses of the two groups
to two of these stimuli: PEA and AND versus baseline. Activations for
PEA were observed in olfactory regions, including orbitofrontal and
piriform cortex, in both subject groups. However, osmics and anosmics
differed in their response to AND. Osmics showed more activity in the
dorsomedial area of the left parietal cortex and had no significant
olfactory activations, whereas anosmics showed more frontal
activations in the right dorsomedial and orbitofrontal regions. The
results for PEA are consistent with other studies of olfactory
processing, showing activations in olfactory cortices, whereas the AND
results suggest differential function in the human brain related to
individual sensitivity to androstenone. Implications of the parietal lobe
activations will be discussed. Supported by Canadian Institutes of
Health Research
102 Poster [ ] Olfactory CNS Physiology and Coding
FUNCTIONAL NEUROIMAGING OF ODOR IMAGERY
Djordjevic J. 1, Zatorre R. 1, Petrides M. 1, Boyle J. 1, Jones Gotman M. 1
1Montreal Neurological Institute, Montreal, Quebec, Canada
We used positron emission tomography to investigate brain regions
associated with odor imagery. We compared changes in brain activation
during active imagination of odors versus those during nonspecific
expectation of olfactory stimuli, and during experience of physically
presented versus imagined odors. Sixty-seven healthy volunteers were
screened for their odor imagery (with a paradigm developed in a
previous study), and 12 of them, selected as “good odor imagers”,
participated in the neuroimaging study. Comparison of odor imagery
with general expectation of odors revealed activation in the left primary
olfactory cortex (POC) including piriform cortex, as well as in the
insula bilaterally, and the left posterior orbitofrontal cortex (OFC). On
the other hand, comparison of odor perception with an odorless baseline
resulted in increased activation bilaterally in the piriform and insular
cortex and in the right posterior OFC. Results of a conjunction analysis
revealing regions of common activation during odor perception and
imagery were convergent with the results obtained by the subtraction
method: left piriform cortex and bilateral insula were activated both
when smells were perceived and imagined. Findings also suggested a
differential role of the right versus left posterior OFC in perceptual
versus imaginal olfactory processing: whereas the right OFC was
activated during odor perception, the same region of the left hemisphere
was activated during odor imagery. Overall, the findings indicated that
neural networks engaged during odor perception and imagery partially
overlap.
Supported by Canadian Institutes of Health Research
26
103 Slide [ ] Olfactory CNS Physiology and Coding
NEURAL SUBSTRATES UNDERLYING MENTAL IMAGERY
OF PLEASANT AND UNPLEASANT SMELLS
Bensafi M. 1, Porter J.A. 2, Khan R.M. 1, Mainland J. 1, Johnson B. 3,
Zelano C. 4, Sobel N. 1 1Neuroscience, University of California,
Berkeley, Berkeley, CA; 2Psychology, University of California,
Berkeley, Berkeley, CA; 3Bioengineering, University of California,
Berkeley, Berkeley, CA; 4Biophysics, UC Berkeley, Berkeley, CA
Patterns of neural activity in orbito-frontal cortex (OFC) reflect odor
valence: the medial OFC responds preferentially to pleasant odors,
whereas the lateral OFC responds preferentially to unpleasant odors.
Whether this dissociation exists during odor imagery in the absence of
any odorants is unknown. Here, we approached this question using
fMRI (4T). Sixteen subjects (8f) were tested in an event-related design
with 5 trial types, each repeated 22 times (TR=500ms, ISI=30sec). In 3
sensory trial types, a pleasant odor (strawberry), an unpleasant odor
(rotten eggs) and a non-odorized control were presented by air dilution
olfactometer. In 2 imagery trial types, no odorants were presented, but
subjects were asked to imagine the same smells (strawberry, rotten
eggs). Replicating previous reports, initial analysis revealed a
dissociation during perception whereby strawberry induced a significant
activation in medial OFC, whereas rotten eggs induced an increase in
lateral OFC (p