Thad Polk - US grants

One of the central tenets of modern cognitive neuroscience is that the neural architecture of cognition is organized into anatomically segregated subsystems (modules) that perform different functions. It is usually plausible to assume that the underlying neural segregation was genetically predetermined, but there are cases that this hypothesis cannot easily accommodate. Furthermore, a growing body of evidence in neuroscience has demonstrated that experience can and does lead to what might be called quantitative changes in brain organization (e.g., changes in the size of an existing functional region). It is therefore natural to ask: Can experience lead to qualitative changes in the brain's modular organization (such as the neural segregation of a function that was not previously localized) or are neural modules always hard-wired by genetics? Previous patient work demonstrating selective impairments in experience-dependent functions (e.g., reading and writing) suggests that experience may indeed be able to produce neural segregation, although the interpretation of those findings can be controversial. We propose to develop converging evidence relevant to the issue by using an independent methodology (neuroimaging) that complements lesion studies. Making progress on this issue could have important implications for our understanding of neural organization, for the interpretation of neurological patients, and ultimately for the rehabilitation of patients with brain damage. The method we propose to use is functional magnetic resonance imaging (fMRI). fMRI has a number of features that make it ideal for addressing the issue of experience-dependent neural segregation: (1) its spatial resolution (1-5mm3) is sufficiently fine to detect anatomic segregation of functional modules, (2) significant results can typically be obtained in individuals making it unnecessary to average across subjects (a critical point because the underlying neural organization under study may differ across individuals), and (3) it can be used with human subjects who are neurologically intact. Our plan is to develop converging evidence regarding experience- dependent neural segregation by investigating three sub- questions: (1) Is letter recognition segregated from digit recognition? (2) Is semantic knowledge about numbers segregated from semantic knowledge about other categories? (3) Are different languages segregated in bilinguals? We chose these domains for four reasons. First, in each case the putatively segregated function can be selectively impaired by brain damage (suggesting that it may be segregated). Second, in each case it is problematic to assume that any underlying neural segregation could be hard-wired. Third, the functions can be relatively well matched with control tasks (making it more difficult to attribute positive results to artifacts). And fourth, the functions can be readily studied using fMRI (e.g., they do not require extensive movement). By studying these three domains in parallel we hope to develop converging evidence that will allow us to draw more general conclusions relevant to the issue of experience-dependent neural segregation.

Research on tailoring for health-related behaviors and decision-making has increased dramatically since the early 1990's. Results from numerous randomized trials suggest that tailored communications for cancer prevention and control have a greater influence on behavioral and decision making outcomes than one-size-fits-all approaches. Until recently, however, we had little understanding of why tailored communications influenced behavior, or which components of these communications were effective. CECCR1 funding helped us to organize a concerted interdisciplinary effort to open the black box of health communications interventions, which led to identifying a broad array of psychosocial and communications components relevant to health behavior change and decision making. We also identified important individual characteristics that moderate the impact of health communications messages. Results from several of these studies have already been published, and are leading to significant new research funding and a new generation of health communications investigators. Moreover, CECCR1 results are also being adopted i health care, employer, pharmaceutical, and government settings of the U.S. and other parts of the world The overarching aims of CECCR2 are to: (1) Extend our tailoring research beyond the prevention area to the broader cancer care continuum, including early detection, treatment, and long-term survival; (2) Extend our tailoring research to new clinical and post-treatment settings; (3) Deepen our understanding of the key psychosocial and communications components identified in CECCR1, including motivation, ethnic identity, risk perception, and cognitive processing; (4) Explore methods of tailoring to patient preferences for shared decision making; (5) Develop new social and cognitive neuroscience strategies for identifying immediat impact and mechanisms of health communications messages; (6) Develop new interdisciplinary collaborations with scientists and research institutions; (7) Train a new generation of health communications scientists and practitioners; and (8) Disseminate both the scientific and practical results of our research efforts. Woven through CECCR2 research are crosscutting interests related to: tailoring and relevant communications channels; reaching underserved populations through more relevant and easier-to-proces content; physiological mechanisms of communication effect; and methodological issues of design, data collection, and measurement CECCR2 involves four primary research sites and networks, and collaborations with 34 research investigators from 13 institutions. Because of the ambitious scope of our proposed studies, we have obtained over $10 million in matched funding from other sources to help us carry out the proposed research.

DESCRIPTION (provided by applicant): While tailored smoking cessation interventions have generally demonstrated positive outcomes, the mechanisms by which people respond to these interventions are less well understood. We are proposing a series of studies that examines the neural substrates associated with smoking cessation messages individually tailored to smokers'needs and interests. The proposed project uniquely combines cognitive neuroscience and public health communications methods to investigate how smokers respond to tailored smoking cessation messages. We have demonstrated in a preliminary functional magnetic resonance imaging (fMRI) study that smokers receiving high-tailored smoking cessation messages (compared to low-tailored cessation messages) showed greater medial prefrontal cortex (MPFC) activation, consistent with the role of MPFC in processing self-relevant material. To further investigate the neural substrates that respond to specific tailored smoking cessation messages and importantly the relationship of these findings to subsequent actual smoking cessation, we propose two studies, each with two phases. Study 1a (Phase I) investigates the role of self-relevance in targeted messaging for smoking cessation. We will examine the relative activation of the MPFC in each subject during exposure to highly self-relevant, high-tailored smoking cessation messages in contrast to low-tailored self-relevant messages and a control self-relevance task. 88 cigarette smokers will receive high-tailored and low-tailored self-relevant smoking cessation messages during fMRI, and will also complete a control task testing self-relevance (Phan et al., 2004). The findings will inform whether greater MPFC activation to high- tailored self-relevant smoking cessation messages is associated with greater self-referential processing. In Study 1b (Phase 2), we will administer to each Study 1a participant a previously tested web-based tailored smoking cessation program immediately following the fMRI scanning (Study 1a) session. Subjects will be followed up in four months and assessed on actual smoking abstinence. Study 1b will explore associations between specific brain region activations from self-relevant smoking messages and subsequent cessation. Study 2 investigates the role of two different types of smoking cessation messages: motivational messages (why to quit) and instructional messages (how to quit) in targeted messaging for smoking cessation. In Study 2a (Phase1), another 88 cigarette smokers will receive high-tailored motivational messages, high-tailored instructional messages, and neutral messages during fMRI. In Study 2b (Phase 2), Study 2a participants will receive the web-based tailored smoking cessation program immediately after the fMRI scanning session and then be followed up in four months to determine actual smoking abstinence. Study 2b will explore the relationship between specific brain region activations in response to tailored motivational and instructional cessation messages and subsequent cessation. The results of these studies will begin to identify specific pathways for optimal message tailoring to make the largest impact on smoking cessation. Nicotine addiction is an important public health concern. The project combines public health, psychology, and neuroscience research methods to understand how smokers respond to individualized smoking cessation messages in an effort to improve smoking cessation programs.

This project is developing computational agents that operate for extended periods of time in rich and dynamic environments, and achieve mastery of many aspects of their environments without task-specific programming. To accomplish these goals, research is exploring a space of cognitive architectures that incorporate four fundamental features of real neural circuitry: (1) reinforcing behaviors that lead to intrinsic rewards, (2) executing and learning over mental, as well as, motor actions, (3) extracting regularities in mental representations, whether derived from perception or cognitive operations, and (4) continuously encoding and retrieving episodic memories of past events. A software framework called Storm facilitates this exploration by enabling the integration of independent functional subsystems, allowing researchers to easily plug in and remove different subsystems in order to assess their impact on the overall behavior of the system.


Cognitive architectures are being tested by exposing them to a wide variety of novel environments with unpredictable (and non-repeatable) extrinsic rewards, but in which many actions could lead to intrinsic rewards (e.g., surprise). To assess flexibility, an automated environment generator exposes agents to environments that are unknown in advance to the artificial agent or human researcher. To assess robustness, cognitive systems are being exposed to many variants of the same environment to ensure that the systems can learn from past experience and generalize when appropriate. And to assess autonomy, systems' must operate effectively for extended periods of time in a dynamic environment.

In the longer term, flexible and robust cognitive architectures being devloped under this research will have application as the 'brains' of robotic and software systems in emergency, miltary, and a wide variety of other societal and service realms.

This award is funded under the American Recovery and Reinvestment Act of 2009 (Public Law 111-5).

DESCRIPTION (provided by applicant): Developmental dyslexia is highly heritable, but the search for susceptibility genes is challenging because the pathway from genotype to dyslexic phenotypes is very complex. We hypothesize that neural markers of dyslexia that have been discovered using neuroimaging could be endophenotypes: intermediate traits unseen by the unaided eye that are associated with dyslexia but that are more directly linked to the underlying genotype. We propose to (1) explore the factors underlying neural markers of dyslexia (Aim 1), (2) explore the relationship between the neural factors and behavior (Aim 2) and (3) explore the relationship between the neural factors and a specific genotype that is associated with dyslexia risk (Aim 3). We hope to find a neural factor that is associated with both a behavioral phenotype and an underlying genotype. Such a marker would be a promising candidate endophenotype that could make the search for susceptibility genes more straightforward, and in turn could lead to novel biological treatments for dyslexia in the future. PUBLIC HEALTH RELEVANCE: Identifying the genes that contribute to dyslexia could revolutionize treatment, but is very challenging. We plan to search for a neural signature of dyslexia that is on the pathway from genes to the behavioral impairments in the hopes of making the search for dyslexia genes more straightforward and successful.

? DESCRIPTION (provided by applicant): Cognitive, sensory, and motor function all usually decline with age, but some people age much more gracefully than others. Why? We hypothesize that one important factor is individual differences in neural distinctiveness. In previous work we've repeatedly found that neural activation patterns in response to different stimuli are significantly less distinctive in older compared with younger adults. Furthermore, we've found that older adults with more distinctive neural representations perform significantly better than others on a range of fluid processing tasks. Animal work confirms these age-related reductions in neural distinctiveness and suggests that declines in the neurotransmitter gamma-aminobutyric acid (GABA) may be an important cause. Building on this work, we propose to investigate the scope, cause, and consequences of age-related changes in neural distinctiveness. First, we will use functional MRI to test whether neural distinctiveness declines in auditory cortex, somatosensory cortex, and motor cortex, like it does in visual cortex (Aim 1: Scope). Second, we will use magnetic resonance spectroscopy to measure GABA levels and use the GABA agonist Lorazepam to manipulate GABA levels in order to investigate whether age-related reductions in GABA cause reductions in neural distinctiveness (Aim 2: Cause). Third, we will collect an extensive battery of measures assessing behavioral performance in the same participants in order to assess the relationship between neural distinctiveness and behavior (Aim 3: Consequences). The proposed studies will provide novel insights into how the brain changes with age, whether those changes can predict successful aging, and whether changes in GABA levels are a contributing cause. Such insights could provide the basis for novel interventions to slow, or conceivably even reverse, the behavioral declines seen during normal aging.

Project Summary/Abstract A growing body of research has found that neural representations are less distinctive in older relative to younger adults, a phenomenon known as age-related neural dedifferentiation. Neural dedifferentiation has been associated with many of the behavioral impairments typically observed in healthy aging, and evidence collected by our group during the original funding period has found a strong relationship between neural dedifferentiation and gamma-aminobutyric acid (GABA), the brain?s major inhibitory neurotransmitter. Specifically, we found that GABA levels are lower in older vs. younger adults and that older adults with less GABA exhibit greater dedifferentiation. We propose to take this line of research in two new directions. (1) We will conduct a longitudinal study and explore age-related trajectories of change in GABA, neural distinctiveness, and behavior, as well as directional relationships among these measures (e.g., Do GABA levels at age 65 predict subsequent declines in neural distinctiveness? Does neural distinctiveness at age 65 predict behavioral declines over the next decade). (2) We will also examine how dedifferentiation is related to Alzheimer?s pathology by collecting the same measures in patients with Mild Cognitive Impairment in whom amyloid beta and tau burden has already been measured. The proposed studies will provide novel insights into how the brain changes with age and with neuropathology, whether those changes can explain the observed behavioral deficits, and whether changes in GABA levels are a contributing cause. Such insights could lead to novel interventions to alleviate the behavioral impairments associated with healthy and pathological aging.