A. Psychophysical ratings of odorant nameability, pleasantness and intensity. Odorants classified as unnameable were rated as significantly more difficult to name than odors classified as nameable. However, subjects rated both classifications of odorants to be equal in pleasantness and intensity. B. Subjects performance on the delayed match to sample task. Subjects performed better in the conditions with a 5 second delay between odor presentations than the conditions with a 10 second delay between odor presentations. There was no difference in performance across odor subcategories.

Each row of the figure shows the deconvolved time course of activity in each region. In each row, the left panel shows the nameable and unnameable 5-second-delay conditions while the right panel shows the nameable and unnameable 10-second-delay-conditions. To the right of each panel is the binned response of each condition. Bar graphs depict the integral under the curve in response to sniff-one of the delayed match to sample task. Error bars represent the standard error. In all subregions shown in the figure, smelling unnameable odorants elicited more activity than nameable odorants. A. Activity in right frontal piriform cortex. B. Activity in right frontal olfactory area. C. Activity in left frontal piriform cortex. D. Activity in left frontal olfactory area.

A. The deconvolved time course of activity in left inferior frontal gyrus area. Although we analyzed data from opercular, orbital and triangular inferior frontal gyri, all data shown in the figure are from the opercular region. The smaller panels show the binned response of the 10-second delay nameable and unnameable odorant conditions. Bar graphs depict the integral under the curve in response to sniff-one of the delayed match to sample task. Error bars represent the standard error. Smelling nameable odorants elicited more activity than unnameable odorants. B. The deconvolved time course of activity in right inferior frontal gyrus area is shown as in A. On the right side, there was no difference between conditions. C. Memory-related activity in left inferior frontal gyrus. Shown is the time course of activity that has been analyzed identically to the data shown in figure 3. The level of activity during the delay period was greater when subjects were remembering nameable odorants as compared to unnameable odorants.

In the scanner, subjects performed an olfactory delayed-match-to-sample memory task where they were presented with two sequential odorants and then asked whether the two odorants were the same or different. The odorants were presented either 5 or 10 seconds apart, and were either nameable or unnameable smells (more specifically, the task instructions were presented 5 or 10 seconds apart, so that sniffs were technically 9 and 14 seconds apart). All subjects received the same odorants, but those odorants differed across subjects in terms of their nameability (see figure S2). The figure shows a schematic of a trial that a participant would experience in the scanner. The blue lines show the olfactometer output, the black line shows time elapsed, and the red lines show the auditory stimulus that the subjects heard through headphones throughout the experiment.

Participants sniffed equally during nameable and unnameable conditions. The left panel shows the mean sniff, in liters per minute, for nameable and unnameable conditions. The center panel shows sniff duration, in liters per minute, for nameable and unnameable conditions. The right panel shows sniff maximum, in liters per minute, for nameable and unnameable conditions.

Each panel shows the time course of activity that has been normalized such that the maximum response to sniff-one is equal to 1. This enabled comparisons of activity levels during the delay period. Only the nameable (in black) and unnameable (in red) 10-second-delay conditions are shown along with an additional control condition in which subjects remembered auditory pitches (in green) for 10 seconds. Bar graphs depict the minimum activity level from 6-seconds after the first sniff to the time that the subject was instructed to prepare for the second sniff. This provides a measure of how well activity levels were sustained during the delay (note that because the bar graphs in this figure depict the minimal value, the bar may reveal a negative value even when the mean was positive, e.g., black line in panel A). A. Memory-related activity in right frontal piriform cortex. B. Memory-related activity in right frontal olfactory area. C. Memory-related activity in left frontal piriform cortex. D. Memory-related activity in left frontal olfactory area. In all subregions, the level of activity during the delay period was greater when subjects were remembering unnameable odorants as compared to nameable odorants. In the olfactory area, activity levels were hierarchically organized in the following manner: unnameable>nameable>auditory pitch. In frontal piriform cortex, remembering unnameable odors elicited more activity than nameable odors, but remembering nameable odors and auditory pitches both elicited a similar level of activity during the delay (unnameable>nameable = auditory pitch).