Linking sensory uncertainty with motivation and memory

点击翻译按钮获取中文摘要

Abstract: Processing information that reduces uncertainty is a key role of the brain. Emerging evidence shows that, in addition to using information that is given to them, animals actively seek to obtain information about future rewards or semantic (trivia) questions. Information demand (ID) is robust even when it is non-instrumental (cannot be exploited to increase reward gains). Moreover, non-instrumental ID activates the reward-sensitive dopaminergic midbrain and enhances memory, suggesting that uncertainty, independently of reward gains, is a source of motivation for decision making and cognitive function. However, little is known about the neural mechanisms mediating these effects of uncertainty. I will address this question by designing a new task that measures perceptual ID and allows me to examine how signals of sensory uncertainty relate to those of motivation and memory. I identified visual images – “texforms” – that reliably activate human ventral temporal cortex (VTC) and can be distorted in a controlled fashion to parametrically manipulate the uncertainty about animate/ inanimate stimulus categories. Second, I will use a generative model to decode trial-by-trial sensory uncertainty from stimulus-evoked multi-voxel activity in the VTC. Finally, I show preliminary evidence that perceptual ID (the desire to reveal the identity of an ambiguous image) is related to behavioral reports of confidence (a proxy of uncertainty). In Aim 1, I will examine the hypothesis that sensory uncertainty in VTC relates to midbrain activity either directly or mediated by univariate responses to confidence (such as those in the dorsal anterior cingulate cortex). In Aim 2, I will examine the hypothesis that VTC uncertainty modulates the hippocampus and generates ID-related memory enhancements. Furthermore, I will explore the possibility that distinct network interactions mediate the effect of ID on immediate and delayed memory. Together, the findings will bring significant new insights into the pathways through which multivariate and univariate signals of uncertainty motivate actions that reduce uncertainty and affect memory. In addition, the results will expand our understanding of pathways that produce maladaptive effects of uncertainty in disorders of mood and anxiety.