Olfactory-visual causal inference in perception and the brain (OLVICI)

• Applicant: Professor Tim Rohe, Ph.D.
• Subject Area: Biological Psychology and Cognitive Neuroscience ; General, Cognitive and Mathematical Psychology
• Term: since 2022
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 515110556

Humans integrate sensory signals from their physical (e.g., vision, audition, touch) and chemical senses (i.e., olfaction, gustation) into a coherent multisensory perception of their environment, for example when they eat food. To achieve veridical multisensory perception, the brain should only integrate signals that arose from a common cause, but should segregate those from independent causes. A priori, the brain does not know which causes generated the sensory signals: To infer the signals’ uncertain causal structure, human observers rely on causal sensory evidence such as the signals’ spatiotemporal disparity for the physical senses. Yet, whether chemosensation, especially olfaction, invokes principles of causal inference is unexplored. A priori, odorants could arise from internal or external sources: When we eat food, our brain must infer whether olfactory, gustatory and visual food signals (among others) arose from a common object in our mouth, and not an external source. Only then the brain should combine the signals into a multisensory perception of food flavour. Further, it is unknown whether causal inference affects humans’ evaluation of a flavour’s pleasantness. This Project Proposal will investigate how the brain performs olfactory-visual causal inference to form our perception of food objects. The research will significantly extend our knowledge of multisensory causal inference and chemosensation: First, we can learn whether the principle of causal inference generalizes from the physical to the chemical senses. In contrast to the physical senses, the chemical senses provide more qualitative than quantitative perceptual information and accumulate sensory evidence more slowly. Second, applying the causal inference model will allow to investigate olfactory-visual perception using a principled account which might be especially helpful to more deeply understand human food perception and the evaluation of food pleasantness. To characterize how humans infer the causal structure of olfactory-visual signals and evaluate their pleasantness, we suggest two work packages: In psychophysical and EEG studies on healthy human participants, we first investigate how the brain accumulates qualitative olfactory-visual causal evidence of food stimuli to integrate or segregate their olfactory-visual signals depending on their causal structure. Second, we characterize how the brain infers the causal structure of olfactory-visual food stimuli, evaluates their pleasantness and finally expresses the pleasantness in emotional facial motion. Therefore, we will combine psychophysics, computational modelling, and uni- as well as multivariate pattern-analysis of EEG data. The results will significantly advance our understanding of how the brain forms, evaluates and expresses the multisensory perception of food flavour. This knowledge is highly relevant for the design of food products and nutrition-related psychopathological processes in eating disorders.

• DFG Programme: Research Grants