Charan Ranganath - US grants

The goal of the proposed project is to more precisely characterize, cognitively and neurally, processes involved in short-term, or working memory (WM; Baddeley, 1986), and memory for events, or episodic memory (Tulving, 1972). Although WM and episodic memory are typically discussed separately, there is a growing consensus among researchers that common cognitive processes, referred to as executive functions, contribute to both WM and episodic memory performance. Unfortunately, although researchers have proposed several models to specify what these executive functions might be and how they are involved in WM and episodic memory, identification of the cortical networks subserving these functions has lagged behind. Available evidence indicates that regions of prefrontal cortex (PFC) are particularly involved in these functions. For example, neuroimaging studies of WM suggest that whereas ventral regions of PFC are involved in keeping representations "on-line" in consciousness (active maintenance), dorsal regions are more involved in performing operations upon these active representations (manipulation). Although these finding have emerged from studies of WM, it is possible that the same mechanisms also contribute to the encoding and retrieval of specific perceptual details of events that are useful for determining the source of an episodic memory (e.g., determining when, where, and how an event took place, differentiating perceived events from thoughts and mental images). Thus, encoding and retrieval of this source-specifying information may require executive functions that contribute to both WM and episodic memory.

DESCRIPTION (provided by applicant): The ability to remember a past event, or episodic memory is fundamental to almost every act of daily living. Tragically, episodic memory is severely disrupted in psychiatric (e.g., Schizophrenia) and neurological (e.g., traumatic brain injury) conditions, and patients with memory disorders are frequently unable to work or live independently. The goal of the present project is to investigate the neural mechanisms of temporal memory (memory for when) in the human brain. The available evidence indicates that the prefrontal cortex (PFC) contributes to temporal memory, but it is unclear when or how the PFC contributes and it is also unclear whether different PFC subregions play different roles in temporal memory. We therefore propose to address 4 important and inter-related questions: (1) does the PFC support memory for temporal order even when it is incidental or task-irrelevant? We will use functional magnetic resonance imaging (FMRI) and time frequency analyses of electroencephalography (EEG) data to test the hypothesis that the PFC contributes to memory for temporal context information even when it is incidental or irrelevant to encoding or retrieval intentions. (2) Does the PFC contribute to both working memory (WM) and long-term memory (LTM) for temporal information? We will use FMRI and EEG to test whether the PFC supports both retention of temporal order information in WM and the encoding and retrieval of temporal information in LTM. (3) What is the relationship between the role of the PFC in memory for associations based on temporal order and its role in memory for other kinds of associations? We will use FMRI to test the extent to which similar or different PFC subregions may be involved in forming associations based on temporal order, semantic relatedness, or spatial contiguity. (4) Is there a hierarchical representation of temporal order information in PFC? We will use FMRI to test whether progressively rostral areas in PFC may be involved in encoding and maintenance of temporal context information at increasingly broad timescales. Questions 1 & 2 focus on identifying the conditions under which the PFC contributes to memory for temporal order, whereas questions 3 & 4 test predictions about the relative involvement of different PFC subregions in temporal memory. Collectively, the proposed studies comprehensively investigate the mechanisms for temporal memory in the human brain. Results from these studies will allow us, for the first time, to develop a detailed model of how the PFC and other brain areas support memory for temporal context. Basic research clarifying how the PFC supports memory processes can provide a foundation for new diagnostic and therapeutic approaches to memory disorders.

DESCRIPTION (provided by applicant): The ability to successfully form memories for arbitrary associations is essential for most acts of daily living. Numerous studies have shown that the hippocampus plays a critical role in associative memory. Recent work indicates that the perirhinal cortex (PRc) is also involved, but its functional role in associative memory is poorly understood. This research program will use functional magnetic resonance imaging (FMRI) and novel behavioral paradigms to test three theories of how the PRc contributes to associative recognition: (1) Unitization theory asserts that the PRc can support familiarity-based recognition of novel associations if the paired items are encoded as a single unit, but that the hippocampus is required for recollection of relations between items that are encoded as separate units. Novel FMRI experiments are proposed to test whether the PRc specifically contributes to recognition of associations between pairs of items that were encoded as a single unit. Parallel behavioral studies will test whether unitization influences indices of familiarity. (2) Domain theory asserts that the PRc supports familiarity-based recognition for associations between items from the same processing domain, but that the hippocampus is required to support recollection for associations between items from different processing domains. Novel FMRI studies are proposed to test whether the PRc supports associations between items from the same processing domains, but not associations between items from different domains. In addition, behavioral studies will test whether within-domain, but not across-domain associations influence the behavioral indices of familiarity. (3) Binding of items and contexts (BIC) theory asserts that the PRc represents item information, the PHc represents context information, and that this information is bound by the hippocampus. FMRI experiments are proposed to test whether recall of item information will be associated with PRc activity, whereas the recall of contextual information will lead to PHc activity. By testing the three theories, this proposal can provide the foundation for the development of a comprehensive theory of MTL function that can incorporate the influences of encoding processing, stimulus domain, and retrieval processing. Several psychiatric (e.g., schizophrenia) and neurological (e.g., Alzheimer's disease and traumatic brain injury) disorders are associated with memory impairments and with medial temporal lobe dysfunction. The proposed research may lead to improved diagnostic and therapeutic approaches to these disorders. PUBLIC HEALTH RELEVANCE Basic research on the mnemonic functions of the medial temporal lobe region is critically important because several psychiatric (e.g., schizophrenia) and neurological (e.g., Alzheimer's disease and traumatic brain injury) disorders are associated with memory impairments and MTL dysfunction. Such memory disorders can have a devastating effect on patients'quality of life. Research clarifying the basic neural mechanisms of memory can lead to improved diagnosis and treatment of these disorders.

DESCRIPTION (provided by applicant): Working memory (WM) and episodic long term memory (LTM) dysfunction is a core feature of schizophrenia (SZ) that limits psychosocial function, is unresponsive to current medications, and shows only modest improvements with cognitive training, demonstrating the need to understand basic neural mechanisms to guide development of new treatments. Our previous research, and that of others, has identified a specific pattern of memory strengths and weaknesses in SZ, suggesting a dual-network model in which dorsolateral prefrontal cortex (DLPFC) and hippocampal (HI) memory circuits important for relating items to a specific spatial or temporal context to support recollection is disproportionately impaired, whereas ventrolateral prefrontal cortex (VLPFC) and perirhinal cortex (PRC) memory circuits important for encoding item information in support of familiarity based retrieval are relatively intact. The goal of the current project is to link this dual- networ model to fundamental molecular and neural candidate mechanisms. We will use an innovative multimodal neuroimaging approach integrating magnetic resonance spectroscopy (MRS), electroencephalography (EEG) and functional magnetic resonance imaging (fMRI) to test the central idea that there is a disruption in GABAergic inhibition in the DLPFC in SZ, which contributes to the reduced theta oscillations, DLPFC and HI fMRI impairments, and disproportionate relational memory deficits observed in prior research and preliminary data. This multimodal imaging approach will employ temporal context versus item memory paradigms recently validated in high profile studies of healthy individuals, will assess medication effects (y having equal numbers of medicated and unmedicated participants), and will investigate previously observed relationships with clinical symptoms (disorganization and negative symptoms) and psychosocial function (UPSA-B). The following Specific Aims will be addressed: Aim 1 - Assess WM and LTM for item and temporal context information in healthy controls (HC) and people with SZ; Aim 2 - Use EEG to identify neural oscillations associated with WM maintenance and LTM encoding of item and temporal context information in HC and SZ; Aim 3 - Use MRI to assess GABA concentrations and patterns of fMRI activity during LTM retrieval of item and temporal context information in HC and SZ. By linking hypothesized neurophysiological mechanisms to specific memory strengths and weaknesses the proposed research avoids a generalized deficit explanation of memory dysfunction in SZ, identifies relative strengths that can be enhanced through cognitive remediation, and identifies biomarkers of memory impairment at the molecular, physiological and behavioral level that can guide new treatment development.

By 2060, the number of Americans 65 and older is projected to more than double from 46 million to 98 million, 24% of the total population. With this comes an increased prevalence of Alzheimer?s disease (AD), which will create significant burden on our society and government. At present, screening tools capable of differentiating healthy aging from AD are most effective a decade or more after the preclinical stage, when potential treatments would be most effective. Thus, discovery of novel and specific tools for assessing the aging brain are of utmost importance. Typical studies of cognitive ability involve recognition of learned objects or simple word associations. However, in real-world situations, the content of an event is segmented from a flow of multimodal information. Segmentation and representation of events is supported by a posterior-medial network (PMN) of brain areas. Critically, this very same brain network features the first regions affected by pathological accumulation of amyloid beta (A?), a key characteristic of AD. A recent report from an NIA working group defined asymptomatic A? accumulation as the earliest indicator of preclinical AD. Given the functional role of the PMN, we propose that this stage of disease may not be truly asymptomatic: subtle functional deficits may be evident if properly probed. To address this, we have developed a naturalistic paradigm to characterize event representation in the brain and subsequent memory. We aim to test the novel hypotheses that the brain?s ability to segment and represent complex events is compromised in preclinical AD, and that the extent of this disruption is predictive of deficient memory for the experienced events. We will acquire functional MRI (fMRI) scans while participants view a video narrative depicting naturalistic scenarios. We will additionally test memory performance related to details of the events depicted in the video both in and out of the scanner. Using representational similarity analysis (RSA) and machine learning analyses of functional MRI (fMRI) data, we will examine differences in the way the brain represents events into advanced aging in participants with and without ?asymptomatic? amyloid deposition (status obtained via existing PET scan data). This combination of approaches is highly innovative because current translational measures do not assess memory for rich, dynamic events that make up the majority of real-world experience. This project is expected to significantly improve our understanding of neural and cognitive disruptions that differentiate healthy aging from preclinical AD. By studying how the brain chunks and represents events, and how this relates to memory for those events, we can reveal significant insights into the way AD-related pathology affects day-to-day living. This can provide a mechanistic framework for understanding subtle, subjective memory complaints. The results of this work are anticipated to significantly advance our understanding of memory decline in the earliest possible stages of AD, providing a mechanistic basis for subtle issues that have to date been difficult to assess in the clinic.

A key feature of human learning is use of the past to predict what is likely in the future. For example, people rely on past experience to predict that thunder is often followed by a downpour. But there are different ways to learn how events are related. For example, some events co-occur (thunder and rain), while others are causally related (cloud cover and rain). People can also learn long-range relationships?for example, that rain will lead to more flowers, which leads to more insects, even though rain may not directly affect insects population. This project will investigate how different brain areas support these different kinds of learning abilities, and will fill important gaps in understanding exactly how the brain changes in response to experience. As such, the proposed research will significantly advance our understanding of how the brain supports learning and memory. Because predictive and causal relationships are at the heart of many high-level cognitive activities including reasoning, language, decision making, the proposed research will have a broad scientific impact. More broadly, the findings may have important implications for education by elucidating how we learn, and for artificial intelligence, in which causal reasoning is a major frontier. Additional impacts on education and society will also be enabled through our outreach activities.

A set of neural areas, important for memory, show learning-related changes that represent predictive relations. However, it is unknown how predictive memories form, or what they reflect about observed experience. This project will investigate which ?core predictive memory? areas support learning of causal or non-causal predictive relations. Experimental aims will focus on three major principles of causal learning: sensitivity to confounds, temporal specificity, and representation of structure. Computational models will make specific predictions about how these areas should respond to evidence if they learn according to classic theories from theoretical neuroscience. A well-established fMRI measure of relational strength will then test these model predictions in core memory areas. The first aim will test whether core memory areas are sensitive to confounds, or reflect simple co-occurrence. The second aim will test whether core memory areas are temporally specific. This will resolve whether and which areas learn temporally precise, causal relations as opposed to cumulative predictive relations better suited for planning. The third aim will test whether core memory areas represent explicit structure. Overall, these findings will sharpen and deepen understanding of how predictive memory areas work together to support higher order cognition.

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.