John D.E. Gabrieli - US grants

[unreadable] DESCRIPTION (provided by applicant): This proposal aims to examine how healthy aging influences the psychological and neural bases of economic choice. Older individuals face important and often complex decisions about retirement, income distribution, insurance, and other financial and health-related matters. These decisions invariably have consequences that extend into the future, and usually involve some risk. There is clear evidence that healthy aging alters many of the psychological capacities and neural systems that are likely to be engaged by such economic decisions. This proposal develops an integrated approach to age-related changes in the neural mechanisms involved in both temporal and probabilistic discounting, providing an in-depth and integrated approach to this important domain. [unreadable] [unreadable] Temporal discounting refers to the observation that individuals experience diminishing value from anticipated money or goods as they become more remote in time. Experiment 1 proposes methodological improvements over prior studies of temporal discounting that should allow unambiguous fMRI measurement of neural responses to immediate and delayed alternatives, which may invoke different - and possibly competing - neural circuits. Because aging is known to impact functional activation within the frontal-striatal system and associated control behaviors, this proposal hypothesizes that aging will affect the functional connectivity of these networks such that age-related differences in discounting behavior will be correlated with differential activity in the reward and control networks. [unreadable] [unreadable] Decisions involving risk require discounting of uncertain consequences relative to certain ones. Experiment 2 proposes to measure probabilistic discounting using similar procedures employed for time discounting, and to image the same participants. By examining both temporal and probabilistic discounting in the same individuals, we will be able to establish whether they involve distinct neural systems, and whether age affects temporal and probabilistic discounting in the same way. [unreadable] [unreadable] The relationship of individual differences among older adults to choices in economic domains is unknown. Neuropsychological and behavioral measures of executive control will be collected for each participant so that the status of control processes can be related to discounting behavior and brain function in Experiments 1 and 2. In addition, trait measures of individual differences in time preferences, impulsivity, and risk-seeking will be collected in order to determine whether such traits are correlated with functional activation in executive control or reward processing circuits. In addition, the relationship of age differences in discounting to socio-economic and socio-emotional/personality dimensions will be assessed in order to distinguish the effects of age-associated neuro-cognitive changes from environmental and personality effects. [unreadable] [unreadable] Older individuals face important and often complex decisions about retirement, income distribution, insurance, and other financial matters, and there is clear evidence that healthy aging alters many of the psychological capacities and neural systems that are likely to be engaged by such economic decisions. By examining behavioral and neural indices of two types of economic discounting (temporal and probabilistic), this proposal aims to establish whether they involve distinct neural systems, and whether age affects temporal and probabilistic discounting in the same way. In addition, this proposal examines how variability among older adults in executive control function, socio-economic status, and socio-emotional and personality dimensions affect discounting behavior and its neural correlates. [unreadable] [unreadable] [unreadable]

[unreadable] DESCRIPTION (provided by applicant): The goal of this proposal is to discover the neural underpinnings of the normal development of declarative memory (conscious memory for events and facts). Declarative memory is critical for everyday life and essential for education, but virtually nothing is known about the normal functional development of the neural circuits underlying either the encoding or the retrieval of declarative memories. We propose to use functional magnetic resonance imaging (fMRI) to examine the development of declarative memory systems in the maturing human brain. Preliminary Studies motivate the initial hypothesis that medial temporal lobe mnemonic functions mature prior to prefrontal cortical mnemonic functions. We propose 9 cross sectional and 2 longitudinal experiments to elucidate the normal development of declarative memory. In each experiment, we compare 20 children ages 8-12, 20 adolescents ages 13-17 and 26 adults ages 18-30 (stratified by age and sex). All participants will be characterized with behavioral measures in terms of cognitive and memory abilities. The first stage of memory involves the encoding of current experiences into long-term memory. In Experiments 1-6, we examine the neural substrates of (1) verbal memory encoding for item and source; (2) visuospatial memory encoding (3) self-instructed intentional vs. experimenter-instructed incidental memory encoding; (3) controlled versus automatic encoding; (4) the intention to encode vs. the success of encoding; (5) the relation between maintaining information in working memory and memory encoding; and (6) control of attention and memory encoding. The final stage of memory performance involves the retrieval of information from long-term memory in service of a current goal. Experiments 7-9 examine the retrieval of declarative memory. We examine the neural substrates of (1) retrieval of visuospatial memories; (2) retrieval of source information when source is well remembered; and (3) retrieval for source information when source is less well remembered. In each experiment, we relate behavioral measures to fMRI measures, and compare between children adolescents and adults. In addition, we propose to re-test two group of subjects after three years to examine the longitudinal development of declarative memory encoding and retrieval. According to the American Academy of Child and Adolescent Psychiatry, one in ten children suffer from a learning disorder. It is impossible to begin to understand the brain bases of learning disorders without a reasonable understanding of how learning develops in the typical, healthy brain. We hope these studies will provide a useful step forward in developing a cognitive neuroscience framework of normal memory development that can be used to begin to understand the brain basis of learning disorders. [unreadable] [unreadable] [unreadable]

This award provides funds which permit Drs John Gabrieli and Robert Desimone to purchase an Elekta Neuromag system for human brain imaging. The instrument will be integrated into the Martinos Imaging Center at MIT and will complement the Center's existing neuroimaging capabilities currently based on magnetic resonance imaging. The Neuromag has 306 MEG channels plus up to 124 channels for simultaneous EEG. Magnetoencephalography (MEG) is a powerful technology for noninvasive imaging of human brain activity. It measures magnetic signals given off by active brain cells and it combines good spatial with very high temporal resolution. In particular, MEG makes it possible to study the flow of information through the brain during cognitive tasks and can reveal the high-frequency oscillations which are fundamental to the physiological mechanism by which different brain areas communicate with each other. The technique therefore is extremely well suited to answering a wide range of basic questions about human brain function. For example it has made important contributions in areas such as somatosensory and auditory processing, and particularly in the study of language, a distinctively human capability for which the rapid neural processes involved are not well resolved by fMRI. It is also possible to use this technique on children, including newborns, and this approach has the potential to transform understanding of early brain development including development of perceptual abilities, language and many other aspects of brain function.

Although the justification for purchase of this instrument rests on the basic research which will be performed and the insights into normal brain functioning which will result, MEG also has clinical applications. It is a potentially powerful tool for translational research, for example for identifying patterns of brain activity that might serve as biomarkers for brain disorders. The instrumentation can also be used to study the neural effects of educational interventions in such areas as reading or musical training and as a basis for developing human brain-machine interfaces.

DESCRIPTION (provided by applicant): The goal of this training program is to train the next generation of developmental scientists with the skills and knowledge necessary to make real headway in understanding the development of the human mind and brain and how atypical development leads to pediatric disorders of cognition and emotion. To this end, our students will be trained in multiple approaches to developmental cognitive neuroscience, so that they can forge links between behavioral studies of cognitive development in humans, computational theories of learning and development, and biological mechanisms of brain development. Specifically, beyond receiving a firm foundation in traditional approaches to developmental cognitive science, our students will be trained in six new frontier areas of developmental research that are being actively pursued in our Department. First, advances in computer science and machine learning have led to new computational models (e.g., of inductive inference) that are for the first time providing real traction on many longstanding problems in cognitive development;this work involves collaborations with faculty in the Computer Science and AI Laboratory. Second, there has been a new level of integration among psycholinguistic, linguistic, and neuroscience approaches towards understanding language. Third, to enhance the connection between traditional developmental work and neuroscience, our faculty are now advancing the technical frontiers of developmental cognitive neuroscience, for example by developing methods that enable us to conduct fMRI scans on ever younger children. Fourth, many of us are focusing on social cognition, which not only constitutes a core domain of cognition, but which further includes many component mechanisms that are likely to play a necessary role in the development of other domains of cognition. Fifth, virtually everyone in our group is integrating their work on normal development with parallel studies of developmental disorders, including autism, dyslexia, and specific language impairment. In addition, many members of our Department are working on the biological mechanisms underlying developmental disorders of the brain. Finally, many faculty in our Department are investigating the basic neural mechanisms of brain development, an area that will be increasingly important in understanding cognitive development as the links between cognitive science and neuroscience strengthen over the next decade. Through intensive exposure to all of these areas, our students will be ideally positioned to shape the future of research in cognitive development.

DESCRIPTION (provided by applicant): The long-term objective of this research is to understand the brain basis of developmental dyslexia, one of the most common specific learning disabilities, and to advance early identification of dyslexia so that early intervention can minimize the documented negative influence of dyslexia on student achievement, self-perception, and long-term life outcomes. Dyslexia typically results from a deficit in phonological awareness (the ability to manipulate speech sounds of language) that precedes and impairs learning to read, but the underlying cause of this deficit has not yet been determined. Neuroimaging methods, including event-related potentials (ERPs), functional magnetic resonance imaging (fMRI), and diffusion tensor imaging (DTI), have identified brain differences in children with dyslexia. Nearly all of these studies, however, involve older children with demonstrated reading failure, so two essential questions remain unanswered. First, what brain differences lead to dyslexia (i.e., are present in 5-year-old kindergartners prior to reading instruction in the 1st grade)? Second, can brain measures significantly enhance our ability to predict which pre- reading children at risk for dyslexia in kindergarten actually go on to become dyslexic by second grade? To answer these questions, we propose a longitudinal study that involves (1) screening 1000 pre-reading kindergartners to identify 120 children at risk for dyslexia and 60 children not at risk; (2) perform MRI, fMRI, DTI, and ERP experiments in these 180 kindergartners to identify brain differences in children with versus without risk for dyslexia; (3) longitudinally follow the language and reading development of these children to discover which at-risk children actually progress to dyslexia at the end of 2nd grade; and (4) use various statistical methods, including multivariate statistics, to improve the accuracy with pre-reading kindergartners can be identified as being at true risk for dyslexia. This study is novel in its multimodal imaging with young children, its longitudinal follow-up, and its translational health aim of developing methods to accurately identify young children at true risk for dyslexia so that such children can be offered early intervention to minimize their learning difficulties.

DESCRIPTION (provided by applicant): Specific language impairment (SLI) and autism spectrum disorders (ASD) are complex highly heritable disorders that involve primary impairments in language and communication. A subtype of ASD includes impairments in language that parallel those found in SLI, including two core clinical markers for SLI: deficits in phonological working memory, as measured on nonword repetition tasks, and deficits in grammatical morphology, particularly in marking tense on verbs. The goal of this project is to investigate the brain bases of these two core deficits in SLI, ASD, and typical development of these core language abilities. Aim 1 will investigate the neural basis of phonological working memory (nonword discrimination) and grammaticality judgments in 96 typically developing children aged 5-12. For the nonword discrimination task, we predict that increases in nonword length will result in greater recruitment of bilateral superior temporal gyrus (STG) and left inferior frontal gyrus (LIFG), with developmental changes primarily in LIFG. In the grammaticality judgment task we predict that errors involving omission of obligatory tense (e.g., past tense -ed) will lead to greater activation in LIFG and left STG, compared to correct sentences. Aim 2 will implement the same experimental paradigms in 48 children with SLI and 96 children with ASD (48 with and 48 without co-morbid language impairment) aged 8-12 years old, who will be compared to age- and language-matched typical controls. We predict that in children with SLI the degree of functional connectivity between IFG and STG will be reduced and that functional activation in STG and/or LIFG and functional connectivity between these regions will correlate with behavioral performance in both experiments. We also predict that the ASD children with language impairment will show similar atypical activation patterns as in the SLI group and that over time, both groups will show activation patterns that more closely resemble the typical children. This is the first developmental study on the neurobiological bases of core impairments in language in SLI and ASD. The findings will pave the way to identifying potential biomarkers for these disorders and will open up the possibility of providing intensive interventions for school-aged children that are individually tailored to biomarker profiles. PUBLIC HEALTH RELEVANCE: This project is designed to identify the developmental brain bases of core areas of language impairment in children with the two most prevalent developmental disorders of language using a combination of neuroimaging and behavioral measures. The findings from this research have the potential of providing new biomarkers for developmental language disorders that can be used in assessment and identification of children. By demonstrating the capacity for change in the brain regions that support language processing in school aged children with language disorders, new approaches to interventions for older children will become more feasible and influence clinical practice.

DESCRIPTION (provided by applicant): This project aims to develop a novel potential treatment for common neuropsychiatric disorders, specifically chronic pain syndrome and affective disorders including depression, anxiety, and post-traumatic stress disorder (PTSD). We have previously shown that real-time functional magnetic resonance imaging (rt-fMRI) can provide a feedback signal (neurofeedback) that allows people to gain learned voluntary control of activation in the rostral anterior cingulate (rACC), which in turn alters pain perception. Here, we aim to develop a brain-validated electroencephalography (EEG) neurofeedback method that involves simultaneous EEG and fMRI recordings. This system will then be used to determine EEG signatures reflective of regulation of rACC and of the amygdala. Multiple psychiatric states (e.g., depressive, anxious) are associated with an inability to regulate emotion and appear to be characterized by dysfunction of the limbic circuit including the ACC and amygdala. However, current psychiatric treatment approaches typically lack specificity regarding brain networks and mechanisms. Based on research studies indicating that individuals can be trained to regulate their own mental and physiological responses through online feedback, the proposed study aims to develop a novel tool for enhancing physiological resilience through psychological training using brain signals as the basis for such feedback. Thus, neurofeedback can provide a potential, complementary/alternate approach for treatment of neuropsychiatric disturbances. The combination of simultaneously acquired high spatial-resolution fMRI data and high temporal-resolution EEG data together with advanced statistical learning approaches provide a plausible way to infer and generate a robust and effective EEG neurofeedback signals that can target even deep brain structures such as the rACC or amygdala as the source of feedback. Given the high prevalence of psychopathology and the non-specific targeting of current treatments, the development of a robust method for learned regulation of specific limbic regions is important. Specifically, the new EEG-neurofeedback tool could be a novel, rational, neuroscience-based therapeutic intervention that could become a portable, easy-to-use, and low-cost clinical tool for improving self-control over brain working. In the long-term such a technique could be useful in alleviating chronic pain and regulating emotional instability and agony in cases of depression and anxiety syndromes. PUBLIC HEALTH RELEVANCE: The inability to self-regulate our affective/mental states is an underlying cause of multiple disorders, including chronic pain syndrome, depression, and anxiety that have serious social and psychological consequences. This project aims to develop a novel system that enables people to learn to regulate pain and emotion, which could lead to improved outcomes for major neuropsychiatric diseases that are often unresponsive to current treatments.

? DESCRIPTION (provided by applicant): This proposal is submitted in response to NIH Funding Opportunity Connectome Related to Human Disease (U01) and in response to NIMH's priority/disease area of interest mood and anxiety disorders. This is a collaborative effort among researchers at the Massachusetts General Hospital (MGH), Massachusetts Institute of Technology (MIT), McLean Hospital, Boston University, and the Human Connectome Project (HCP) at Washington University in St. Louis. We believe that the combination of (1) state-of-the art MRI technology and methods at the MGH Martinos Center for Biomedical Imaging, (2) an active collaboration with the HCP to validate neuroimaging harmonization, (3) a Boston-wide consortium of experienced and expert clinical researchers, and (4) a transdiagnostic focus across the anxiety and depression spectrum can deliver a high- quality dataset that meets the specification of the Funding Opportunity. We propose to focus on an area of great clinical need and public health implication: better understanding of psychiatric disorders in adolescence. We target anxiety and depression as diseases that affect many adolescents across multiple traditional psychiatric diagnoses, that are strongly associated with two leading causes of death in adolescents and young adults (suicide and substance-abuse related accidents), and that are understood to frequently have developmental roots leading to lifelong psychiatric disorders. Our research approach is guided by two principles (1) careful adherence to HCP protocols so as to develop large-scale, integrated, and growing data sets available to the scientific community, and (2) a research approach aligned with two constructs from the NIMH Research Domain Criteria Project (RDoC), specifically: (a) the Acute Threat/Fear construct, which is associated with atypical structure and function in specific neural networks, especially amygdala, orbitofrontal cortex (OFC), and ventral medial prefrontal cortex (vmPFC); and (b) the Reward Prediction Error construct, which is associated with OFC, ventral striatum, and the midbrain ventral tegmental area. Across four years we aim to (1) operationalize MRI data collection and behavioral characterization that is harmonized and validated with the Human Connectome Project (HCP); (2) recruit and characterize clinically and behaviorally, 225 adolescents ages 14-15 with and without anxiety and/or depression (180 patients, 45 controls); and (3) perform and analyze HCP imaging with participants. We hypothesize that greater activation in the amygdala-OFC circuit will correlate with more severe scores on measures of fear, and that lesser activation of the striatal-OFC circuit will correlate with more severe scores on measures of reward-error expectancy. We will also (a) examine whether neuroimaging analyses are enhanced with artifact-detection tools and physiological aliasing correction that are publicly available and could be integrated with the HCP, and (b) create an age-specific human tract atlas and tools for automated reconstruction of white-matter tracts involved in the above circuits, which will also be made publicly available.

Difficulties in math and reading prevent many students from succeeding in STEM education which demands both math and reading skills. Dyslexia is the most common type of learning disability affecting reading proficiency but many students with dyslexia also exhibit dyscalculia, or math disability. To date, little research has sought to understand the extent of the neurobiological connections between dyslexia and dyscalculia.

This proposal will explore the cognitive underpinnings of dyscalculia observed in children with dyslexia. Dyslexia and dyscalculia have traditionally been studied as separate development difficulties but this project will seek to determine whether math difficulties that exist in children with dyslexia are similar or different from the math difficulties that exist in children with dyscalculia only. The investigators will recruit 150 children aged 10-12 and screen them for dyslexia, dyscalculia, and both dyslexia and dyscalculia. A battery of behavioral measures in both reading and math are planned along with neuroimaging (fMRI) data collection methods.

This project is supported by NSF's EHR Core Research (ECR) program. The ECR program emphasizes fundamental STEM education research that generates foundational knowledge in the field. Investments are made in critical areas that are essential, broad and enduring: STEM learning and STEM learning environments, broadening participation in STEM, and STEM workforce development. The program supports the accumulation of robust evidence to inform efforts to understand, build theory to explain, and suggest intervention and innovations to address persistent challenges in STEM interest, education, learning and participation.

? DESCRIPTION (provided by applicant): The Martinos Imaging Center at MIT is a core facility that provides researchers with access to cutting-edge brain imaging technologies, including magnetic resonance imaging (MRI), magnetoencephalography (MEG), electroencephalography (EEG) and functional near-infrared spectroscopy (fNIRS). The Center has two Siemens Magnetom Trio 3T MRI scanners, one of which is almost ten years old and approaching the end of its expected lifespan. We propose to upgrade this instrument with Siemens' Prisma technology, which will greatly enhance its capabilities and life expectancy. In particular, this upgrade, branded by Siemens as Prisma-fit, will greatly improve the quality of diffusion imaging and will also provide better signal/noise in other imaging modalities, along with better instrument stability, lower acoustic noise, and shorter scan times. Moreover, because the upgrade involves complete replacement of the gradient coils, the life expectancy of the instrument will also be prolonged, essentially to that of a brand new scanner. The upgraded scanner will support research by multiple groups of users, including the 15 principle investigators (PIs) listed in this proposal. These researchers are conducting studies in many areas of basic and clinical neuroscience, including studies of brain function in healthy subjects, its development in children, and its disruption in many different brain disorders. In particular, NIH-funded projects by major users on this proposal include studies of autism, dyslexia, attention, social cognition, color vision, scene perception, and the anatomical structure and development of the human brain. Additional projects focus on schizophrenia, traumatic brain injury, auditory perception and hearing loss, and the development of new MRI-based technologies. In addition to human neuroimaging, the upgraded scanner will also be advantageous for many studies on animal models, allowing us to draw new parallels between mechanistic animal studies and human brain function. In sum, the upgraded MRI scanner will support a large and highly productive group of users, enabling them to remain at the forefront of neuroimaging research for many years into the future.