Presentation Details
| Spatiotemporal dynamics of odor feature processing in the human brain Sarah Cormiea, Naz Dikecligil, Joel Stein, Isaac Chen, Kathryn Davis, Jay Gottfried. University of Pennsylvania, Philadelphia, PA, USA |
Abstract
Every odor carries with it a wealth of information about its source as well as its relevance to the smeller. With each sniff, the human olfactory system works to answer questions such as: “What is this smell?”, “Do I like it?”, “Can I eat it?”. It remains unknown whether our olfactory system resolves all possible dimensions of odors and whether it does so in parallel or in a hierarchical manner. To investigate the spatiotemporal dynamics of distinct odor features in the human brain, we paired an odor feature rating task with high-temporal-resolution intracranial EEG recordings in patients undergoing invasive monitoring for treatment of intractable epilepsy. Local field potentials were continuously recorded throughout the behavioral task via surgically implanted depth electrodes. On each trial, participants rated a real-world odor (e.g., lemon, pine, cheese, shampoo) on one of three dimensions: (i) pleasantness, (ii) edibility, or (iii) identity. Participants’ ratings reliably differentiated pleasant and unpleasant odors as well as edible and inedible odors. Participants also overwhelmingly chose true labels over incorrect foil labels. Consistent with prior work, we observed robust odor-evoked changes in oscillatory activity in piriform cortex. Early analyses find above-chance decoding of neural responses based on all three odor feature dimensions. Further, time-resolved decoding analyses reveal that the three features emerge and evolve along distinct timelines. These results suggest that oscillations in primary olfactory cortex maintain overlapping yet separable codes for odor identity, pleasantness, and edibility. Ultimately, this work seeks to uncover the contents, sequence, and dimensionality of odor stimulus information embedded in olfactory neural signals as they unfold over time.
No part of this publication may be reproduced, distributed, or transmitted in any form or by any means, including photocopying, recording, or other electronic or mechanical methods, without the prior written permission of the author.
No part of this publication may be reproduced, distributed, or transmitted in any form or by any means, including photocopying, recording, or other electronic or mechanical methods, without the prior written permission of the author.
Content Locked. Log into a registered attendee account to access this presentation.