Mesoscale dynamics underlying expectation bias in the orbitofrontal cortex

点击翻译按钮获取中文摘要

Project Summary Adaptive behavior critically depends on the ability to make predictions from mental models, or cognitive maps. These allow us to act preemptively when events are expected, for instance by shifting our attention, thoughts, and behavior. However, when expectations are incorrect, such biases may be inappropriate and should be suppressed. An inability to override the effects of expectations is a common feature of psychiatric disorders such as depression and anxiety, for instance when fixed negative predictions excessively bias thoughts and behavior. To help us understand the neural basis for these symptoms, this project examines how expectations influence motivated behavior and are suppressed when they are not appropriate. The orbitofrontal cortex (OFC) is critical for using mental models to create expectations. Although neurons in OFC are well-known to signal the value of expected outcomes, how these expectations are balanced with new information to adaptively guide behavior is not well understood. One hurdle is that OFC is large and functionally heterogeneous, particularly in primates, making it difficult to glean a comprehensive understanding of how the region contributes to expectations and resulting behavior. Here, we propose a novel approach that will provide an unprecedented view of OFC activity and functional organization. We will use micro-electrocorticography (µECoG) arrays custom-designed to cover nearly the entire orbital surface in the macaque brain. This mesoscale approach records epicortical signals, which are high signal-to-noise, cover large anatomical areas, and have excellent spatiotemporal resolution. We will combine µECoG technology with a novel behavioral task for monkeys, in which they expect and choose between different rewards. Behavioral effects of expectations are quantified by the intensity of instrumental responses to obtain an expected outcome. We have shown that this behavior is biased by prior expectations and slowly adjusts when contradictory information accumulates. To understand how expectations are dynamically signaled by OFC, we will enlist cutting-edge computational methods that leverage the large-scale nature of the µECoG recordings. We will investigate both dominant signals in OFC and subregional variation that produce biasing expectations and update them with new information. Together, this project will provide a uniquely holistic view of OFC function. In doing so, it will advance our understanding of expectation and bias in motivated behavior, and lay the groundwork for future studies investigating OFC’s role in broader circuits that integrate sensory, memory, and emotional processing to create the mental models that guide adaptive behavior.