Sarah R. Heilbronner - US grants

DESCRIPTION (provided by applicant): Addiction may be caused by disordered decision making and reward processing. Thus, understanding the neural mechanisms of reward-guided decision making has the potential to help those addicted to drugs of abuse. One particular brain region, posterior cingulate cortex (CGp), has been implicated in a variety of decision making processes relevant to addiction, including those choices involving risk, time, and social information. Recent studies have suggested that neuronal responses in CGp may encode the subjective value of a chosen option during decisions about risk and time. Thus, a central hypothesis of CGp function is that it is part of a specialized system that integrates information from multiple sources into a single common metric of value. However, one major limitation of these earlier studies is that they have focused on a single decision context (e.g.., risk or delay), but true subjective value representations should be independent of context. Another mystery about CGp function concerns its relationship with the anterior cingulate cortex (ACC), particularly in those decisions involving delays, risk, and social information. A comparison could potentially shed light on the roles of both regions, as well as the mechanisms of reward-based decision making in general, as there are strong reciprocal connections between the two. I am proposing three separate experiments to address these gaps in the literature. I will record responses of single neurons in CGp and ACC during a task involving three different decision contexts relevant to addiction: risk, delay discounting, and social reinforcement. I hypothesize that firing rates of CGp neurons will signal subjective value independent of decision context, and that response modulations of ACC neurons will be greater than those of CGp neurons during decisions involving social reinforcement, whereas CGp neurons will have relatively larger modulations to decisions involving risk. Finally, I will test whether firing patterns observed in CGp play a causal role in decision making by reversibly inactivating this region. Addiction can be framed as a problem of decision making, of failing to make good decisions about risks, rewards, and delaying gratification. Knowing how the brain makes decisions can be of enormous use in trying to understand and solve the complex health problem of addiction to drugs of abuse. I propose to study how a particularly mysterious brain region known as cingulate cortex guides how we process rewards and make decisions, which may advance our progress toward solving the problem of addiction.

DESCRIPTION (provided by applicant): The default mode network (DMN) is a set of brain regions consistently activated at rest and deactivated during task performance. DMN activity is abnormal in many neurological and psychiatric disorders, including major depressive disorder (MDD) and drug addiction. The two main parts of the DMN are the medial prefrontal cortex (mPFC) and posteromedial cortex (PMC), two areas that are spatially distant and have distinct canonical functions. While there is some evidence of direct connections between these general cortical areas, little is known about their specifics and how they link the network together as a whole. I hypothesize that the DMN is linked together by specific hub subareas that contain converging connections from multiple DMN regions, and the cingulum bundle and the internal capsule are the major connecting white matter bundles for this network. Thus, the goal of this proposal is to delineate how anatomical connections and pathways between mPFC and PMC allow them to operate as a network. Combining traditional anatomical techniques with diffusion magnetic resonance imaging (dMRI), I will determine the connections that underlie the DMN. Defining these specific DMN connections will establish the circuitry subserving neuroimaging results. This basic knowledge is fundamental and is the first step in understanding the changes in the DMN in disease. Functional connectivity within the DMN may result from direct anatomical links, indirect ones, or some combination of both. The first overall aim of this grant i to delineate the direct and indirect connections between subregions of the mPFC and PMC to find potential zones of converging connections. Furthermore, dMRI has identified psychiatric disorders, including MDD and addiction, with specific abnormalities within white matter pathways that likely link DMN structures. These abnormalities likely reflect disruption of specific connections. However, the fibers traveling through any given location within a white matter bundle remain unknown. The second and third aims are to establish white matter pathways that connect DMN structures using tracing and dMRI methods. This will link dMRI and anatomical tract-tracing, testing the validity of dMRI and providing a guide for how to interpret its results.

Project Summary The posteromedial cortex is a brain region that is especially abnormal in many psychiatric diseases, including depression, schizophrenia, and anxiety. However, it has traditionally received little attention from neuroscientists, in part because we do not understand its underlying biology. We propose to remedy this problem by uncovering posteromedial cortical circuits using tract-tracing in nonhuman primates and diffusion- weighted magnetic resonance imaging in nonhuman primates and humans. This pipeline will allow us to create a segmentation of the posteromedial cortex according to its anatomical connectivity. First, we propose to tile the nonhuman primate posteromedial cortex with tract-tracer injections, which will allow us to analyze its connectivity with other, better-understood brain regions. We will use these data to segment the posteromedial cortex according to its anatomical connectivity. We expect that different segments will connect with distinct, well-defined networks within the brain. Next, we will collect diffusion-weighted neuroimaging data in nonhuman primates and humans. Although this method allows noninvasive detection of some connections, it is highly susceptible to errors. We will use the principles of white matter organization identified in the tract-tracing data to guide and debug these neuroimaging data. Because tract-tracing cannot be performed in humans, this pipeline offers a rare possibility to infer anatomical connectivity in the human brain. Finally, based on prior neuroimaging experiments, scientists have hypothesized that the posteromedial cortex may connect with a particularly large number of brain regions and networks. Thus, we would like to determine the extent of axonal collateralization in the neurons of this brain region. In other words, does a single posteromedial cortical neuron project to many other brain regions, or are the circuits largely separate? Together, we expect these projects will elucidate PMC anatomical connectivity in such a way that large-scale neuroimaging networks can be linked with specific neuronal circuits.

PROJECT SUMMARY: TRANSLATIONAL NEUROPHYSIOLOGY CORE The purpose of the Translational Neurophysiology Core is to link the individual Projects by facilitating cross- species, cross-method behavioral and neural data collection and analysis. Our Center relies on standardized EEG/fMRI acquisition and processing, as well as consistent DPX and Bandit task data collection with human control subjects and participants with early psychosis (PROJECTS 3 and 4). In addition, ensemble recordings and local field potential data, along with DPX and Bandit task behavioral data, will be collected in nonhuman primates (PROJECT 1) and mice (PROJECT 2). The cross-modal aspect of neural recordings and behavioral data acquisition presents unique challenges, which will be addressed through three Service Aims. Service Aim 1: Support for behavioral data acquisition. Our Center will collect trial-by-trial data in humans from identical experimental behavioral paradigms (the DPX and Bandit tasks) acquired across both clinical and normative human samples, studied in multiple settings (e.g. PROJECTS 3 & 4). It is important that these paradigms be developed to be administered in a highly reliable manner using portable digital methods, and that the data collected from these paradigms be stored, processed, integrated, and visualized in a secure, reliable, and accessible manner so that findings can be compared across Projects. Service Aim 2: Standardized collection, preprocessing, and analysis of simultaneously acquired EEG and fMRI data in humans. Our goal is to ensure that these data are collected, preprocessed, and analyzed in an identical manner across PROJECTS 3 and 4, with a focus on acquiring measures that can be used by the COMPUTATIONAL CORE to assess activity timing, excitatory-inhibitory balance, and system noise. Service Aim 3: Cross-species neuroanatomical translation. We will facilitate translation of neural circuits across species by using connectivity data (combination of diffusion MRI and tract- tracing) to identify homologies across nonhuman primates (PROJECT 1), mice (PROJECT 2), and humans (PROJECTS 3 & 4).