LC-ACC interactions supporting adaptive, feedback-driven decisions

点击翻译按钮获取中文摘要

PROJECT SUMMARY/ABSTRACT Our long-term goal is to understand how the brain processes information in a flexible, context-dependent manner to support effective decision-making. Many previous studies of decision-making focused on how the brain accumulates evidence used to select a particular action, which was shown to involve modulations of persistent activity of individual sensory-motor neurons that prepare the appropriate action. However, the brain must also often accumulate evidence in a flexible manner across actions, about which little is known. We propose that decisions requiring the flexible accumulation of feedback-related evidence across actions depends on interactions between the Anterior Cingulate Cortex (ACC), a cortical structure on the medial surface of each cerebral hemisphere that has widespread connectivity with other parts of the brain, and the brainstem nucleus locus coeruleus (LC), which is the primary source of the neuromodulator norepinephrine (NE) to the rest of the brain. The ACC and LC have strong, reciprocal connections and are thought to interact in ways that support key features of cognition, including adaptive information processing, but the details of these interactions are not well understood. Our primary hypothesis is that these interactions modulate activity patterns of populations of ACC neurons that implement a process of across-trial evidence accumulation that uses reward and error feedback to govern decisions to switch behavioral choices. We are particularly interested in understanding how these modulations relate to changes in coordinated variability in ACC that can have major effects on how neural populations process information. To test this hypothesis, we use simultaneous, complementary measurements of neuronal activity from single and populations of neurons from the two brain areasin the context of two tasks that require different forms of across-trial accumulation of feedback information to guide saccadic decisions. We have three Specific Aims. Aim 1 is to understand how activity patterns of individual neurons in the LC relate to performance on these tasks. Aim 2 is to understand how relationships between neuronal activity patterns in the LC and ACC relate to performance on these tasks. Aim 3 is to use a combination of manipulations to identify causal contributions of temporally precise, pathway specific activity patterns from LC to ACC on task performance. Together these Aims will provide new mechanistic and computational insights into how LC-related modulations of ACC population activity support ACC's role in flexibly linking performance monitoring and control across multiple trials. Our findings will have direct relevance to numerous constructs in the Research Domain Criteria (RDoC) framework and thus have broad significance to fields that aim to understand the neural substrates of complex behaviors and their dysfunction in certain mental disorders.