A predictive nose: Olfactory identification is more reliant than visual identification on predictive object cues

Abstract

Identifying objects in the environment is one of the most important features of our senses. Toward this goal, predictions are made from known environmental regularities. It is unknown whether different sensory systems may rely differently on cues for prediction and evaluation. Olfaction, the sense of smell, provides a means to test this hypothesis. People are often unable to identify odor stimuli based on scent alone, and cues from vision and audition are often needed. We tested whether olfactory (relative to visual) identification is especially dependent on source-based predictive cues. Participants listened to word cues (object-based, e.g. “lilac”, and category-based, e.g. “flower”), and then quickly determined whether a visual or olfactory target stimulus matched or did not match the word cue. As predicted, we found that the greatest response time differences between matching and non-matching trials occurred when the word cue was object-based and the target was olfactory. This interaction effect disappeared when participants were cued with category-based words (“fruit” or “flower”). A follow-up fMRI experiment revealed an increased involvement of the auditory cortex in anticipation of olfactory relative to visual objects. Furthermore, olfactory error signals activated a transmodal cortical network involving primary olfactory and visual cortices, as well as the anterior cingulate cortex. Our results suggest that olfactory identification may be especially reliant on predictive object cues and involve a vast, sensory-specific cortical error detection system. These results provide new insights into the organisation of sensory-cognitive hierarchies in the human brain.