Understanding adaptive criterion setting in perceptual decision-making at the implementational level: focus on dorsomedial striatum and medial prefrontal cortex

• Applicant: Professor Dr. Maik C. Stüttgen
• Subject Area: Cognitive, Systems and Behavioural Neurobiology
• Term: since 2024
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 543128862

Perceptual decision-making (PDM) is the process of using sensory information to select an appropriate course of action. For example, an animal may decide to eat or not to eat a certain fruit based on its visual appearance. Importantly, such sensory-based decisions are influenced by non-sensory factors – in the present scenario, a hungry animal will be more inclined to eat a certain fruit than a satiated one, even if eating that fruit is potentially harmful.This situation can be mimicked in a laboratory setting in which rat subjects are required to make choices depending on sensory evidence (i.e., discriminating between several different stimuli) and non-sensory factors (e.g., the probability of receiving a reward for a certain choice). In a natural setting, response-outcome contingencies change frequently – for example, a visit to a food patch may be successful on one day but unsuccessful on the next, and vice versa. Similarly, we confront our subjects with unsignaled changes in reward probability which require them to react flexibly to harvest the maximum number of rewards – a process called ‘adaptive criterion setting’ (ACS).In our previous work, we have developed quantitative models which describe how subjects behave adaptively in this scenario. Furthermore, we have begun to observe and manipulate brain areas which support PDM-ACS, among them medial prefrontal cortex (mPFC). In the present proposal, we argue that ACS is crucially dependent on the dorsomedial striatum (DMS), one of the main projection targets of mPFC. To that end, three experiments are proposed. In Experiment 1, single-unit recordings in DMS will be conducted during PDM-ACS to examine the neuronal representation of different decision variables on a trial-by-trial basis. These decision variables will be extracted from behavioral data through quantitative fits of both detection-theory-based and reinforcement learning models. In Experiment 2, we will optogenetically inhibit DMS neurons at specific time points during task performance to dissect its causal contribution to ACS. In Experiment 3, we will perform pathway-specific inhibition of mPFC projections to DMS, asking which decision variables are communicated and whether (and when) they are used for adaptive behavior in our task.The results can then be interpreted in the context of the quantitative models in an effort to connect the behavioral and neurophysiological levels of investigation. Moreover, these results will help to clarify the overall function of the dorsomedial part of the striatum.

• DFG Programme: Research Grants
• Co-Investigator: Professor Dr. Frank Jäkel