Influence of value on unconscious sensory information sampling

• Applicant: Dr. Marcus Rothkirch, since 5/2022
• Subject Area: Cognitive, Systems and Behavioural Neurobiology ; Human Cognitive and Systems Neuroscience
• Term: from 2019 to 2022
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 416529504

Sensory information is often ambiguous and compatible with conflicting interpretations. Therefore, our brains are regularly faced with the task of deciding which of several conflicting interpretations to select for conscious perception. Previous evidence indicates that the value of sensory information plays an important role in perceptual decisions under conflict, suggesting that top-down preferences selectively affect sensory information sampling. Such value-based selective sampling might critically depend on dopamine and glutamate as indicated by pharmacological manipulation studies of value processing and top-down expectations on sensory processing. However, the computational bases and neural implementation of value-based selective sampling mechanisms in perceptual decisions under conflict remain poorly understood. The central question of the project proposed here is how value affects unconscious sensory information sampling via dopaminergic and glutamatergic neurotransmission. We plan to answer this question by investigating the effect of reward on perceptual decisions under interocular conflict using the continuous flash suppression (CFS) paradigm. In CFS, a high-contrast dynamic stimulus shown to one eye renders a low-contrast stimulus shown to the other eye invisible for several seconds, which elegantly enables the investigation of the sensory processes that precede a conscious perceptual decision. The porposed research program combines state-of-the-art computational modeling with behavioral, functional magnetic resonance imaging (fMRI) and pharmacological modulation, and is structured in three work packages (WPs). In WP1, we will devise a behavioral experiment to quantify the effects of reward on perceptual decision making under conflict. Computational modeling in a Bayesian predictive coding framework will be applied to formalize the influence of value on unconscious information sampling under CFS. In WP2, we will use fMRI to examine the neural implementation of the effects of reward on perceptual decision making under conflict. Model-based fMRI will be used to identify the neural representations during unconscious sensory information sampling under CFS at different levels of the visual processing hierarchy. In WP3, we will use the experimental paradigms established in WP1 and WP2 in conjunction with pharmacological manipulation of prediction and prediction error signaling using the NMDA receptor antagonist ketamine and the D2 receptor antagonist amisulpride. By specifically addressing the role of stimulus value on the unconscious determinants of perceptual decisions, we expect this project to make a substantial contribution to our understanding of the neural mechanisms underlying decision-making in humans.

• DFG Programme: Research Grants
• Ehemaliger Antragsteller: Professor Philipp Sterzer, Ph.D., until 4/2022