Department of Brain and Cognitive Sciences
The human brain is the most complex, sophisticated, and powerful information-processing device known. To study mechanisms of the brain and mind, the Department of Brain and Cognitive Sciences combines the experimental technologies of neuroscience and psychology with the theoretical power of computational methods and cognitive science. The department's research focuses on several themes, with each studied at multiple levels of analysis using cellular, systems, computational, and cognitive approaches. The department's faculty also constitute the majority of investigators in the Picower Center for Learning and Memory and the McGovern Institute for Brain Research.
Brain Mechanisms of Vision and Movement
Peter Schiller's lab has studied the role of cortical inhibitory circuits in target selection with visually guided saccadic eye movements, by applying minute quantities of GABA agonists and antagonists to selected brain areas. They found that inhibitory circuits play a central role in visual analysis in the striate cortex and in controlling visually guided eye-movements in the frontal eye fields. These results suggest that the generation of each saccadic eye movement requires not only that excitatory signals be sent to the brainstem oculomotor centers, but also that inhibitory circuits be inactivated.
Mriganka Sur's lab discovered a new property of neurons in early visual cortex. We move our eyes several times a second as we scan a scene, and we tend to successively view locations that differ in structure and composition from the previously viewed location. Experiments demonstrated that such scanning improves visual discrimination, and also the response selectivity of visual cortex neurons. Such rapid plasticity of cortical responses constitutes a simple form of learning that continually influences vision.
Research in Emilio Bizzi's lab has shown that in the motor areas of the monkey's frontal lobe, there is a population of neurons which is selectively recruited during motor learning. They are also studying patients with impaired mobility caused by a stroke affecting the motor areas of the central nervous system to assess the feasibility of a remotely supervised, computer-enabled physical therapy device, administered over high-speed telecommunications.
Learning and Memory
Ann Graybiel's lab has provided an understanding of how activity states of the forebrain are controlled and modulated during motor activity, procedural learning and cognition. This has major clinical relevance for disorders such as Parkinson's and Huntington's diseases, and for neuro-psychiatric disorders such as Tourette syndrome, obsessive-compulsive disorder, attention deficit disorder, and major depression.
Earl Miller's lab has made key discoveries of the neural basis of the high-level concepts, abstractions and functions that guide intelligent behavior. In trained monkeys, they found neural representations of perceptual categories ("cat" vs. "dog"), abstract rules ("same" vs. "different"), and the numbers 1–5, together with an understanding of the neural dynamics underlying short-term memory.
Anthony Wagner's lab carries out fMRI research examining how human memory is organized and supported by the mind and brain. Advances over the past year include: demonstrating that keeping information in active memory contributes to the formation of durable long-term memories and new word learning; characterization of frontal lobe processes that guide semantic memory retrieval; and delineation of the role of specific subregions within the medial temporal lobes during memory formation.
Mathew Wilson, in collaboration with Susumu Tonegawa, demonstrated for the first time the role of the circuits within hippocampal area CA3 in the retrieval of partially cued memories. They identified an explicit physiological correlate of memory retrieval through a process known as "pattern completion." The lab also advanced understanding of the mechanisms of sequence memory formation and retrieval with the demonstration of sequence memory reactivation during slow-wave sleep.
Joshua Tenenbaum, who joined BCS in January 2002, studies learning and inference in humans and machines. Developments in his lab include: the discovery of an apparently universal scaling law of similarity comparison, determining how features of objects or concepts are weighted by people in judging their similarity; improvements to a powerful approach to nonlinear dimensionality reduction that is becoming widely used for scientific visualization and pattern analysis tasks across a number of disciplines; and a novel framework for understanding how people use intuitive theories to learn about the meanings of words and the causal structure of their environment, drawing rich generalizations from very limited perceptual evidence.
Brain Development and Plasticity
Morgan Sheng's lab uncovered three specific proteins that control the growth of dendritic spines (specialized postsynaptic compartments in brain synapses), and revealed the subunit-specific rules that govern the surface delivery of AMPA receptors (an important postsynaptic neurotransmitter receptor) in neurons.
The focus of Guosong Liu's lab is to identify the principles that guide the formation of functional neural circuits. Recently, his lab discovered that the excitatory/inhibitory synapse ratio in a single dendritic tree is always conserved and the total amount of excitatory synaptic inputs per dendritic branch is scaled according to the surface area of the tree. This led them to propose a new rule that governs the organization of synaptic inputs on a dendritic tree.
Elly Nedivi's lab has been studying the transcriptional regulation of cpg15, a gene they isolated in a forward genetic screen for activity-regulated genes that may play a role in synaptic plasticity. They recently found that cpg15 expression is mediated by molecules and signaling pathways previously associated with synaptic plasticity during learning and memory, and development. Cpg15 may be one of a set of CREB effector genes whose activity-dependent regulation confers neurons with the capacity for circuit restructuring.
William Quinn's lab studies the fruit fly, Drosophila, which has two kinds of long-term memory. One requires protein synthesis and is mediated by gene transcriptional events. Another (consolidated memory) does not require protein synthesis and depends on the product of the radish gene. The cell-signaling pathway leading from and through the ras gene product has been implicated in several developmental and cell-interactive events, notably in human cancers. An aberrant ras gene, when expressed in fly brain structures called mushroom bodies, blocks immediate learning, but the same aberrant gene, when expressed in a pair of large cells that project to the mushroom bodies, blocks memory while leaving immediate learning intact.
Lera Boroditsky, who joined the faculty in fall 2001, has uncovered several ways in which languages shape the way their speakers think about objects, time, and events.
Work in Ted Gibson's lab has focused on determining what constraints make sentence comprehension easy or difficult. It appears that integration cost is sensitive to the kinds of referential expressions (e.g., a pronoun, a name, a full noun phrase) that intervene between two syntactic dependents, and that storage cost affects ambiguity resolution. Another project has demonstrated that in sentence production, the likelihood of producing an intonational boundary is dependent on: the size of the syntactic constituent that was just finished; and the size of the syntactic constituent that is about to be produced. Applications of this work include computer speech generation.
Steven Pinker is conducting a large study of language development in twins, which aims to assess genetic effects on the timing of language development. He is also engaged in research on the neural bases of lexical memory and
grammatical computation using fMRI . His sixth book, The Blank Slate, on the denial of human nature in modern intellectual life, will be published in September 2002.
The Test of Grammatical Impairment (TEGI) is a result of Ken Wexler's research on linguistic development. This is the first standardized test to select children with Specific Language Impairment that has a scientific basis, and it is expected that it will be the most viable and accurate way to determine which children have this syndrome.
The Friday Departmental Colloquia followed by tea has become an important department-wide event. The weekly lunch presentations organized and run by the graduate students—Brain Lunch and Cog Lunch—continue to flourish, as do the Brain and Machines seminar series sponsored by CBCL and the Plastic Lunch series organized by PCLM. The special events—the semi-annual Teuber Lectures, the annual Bidwell Lecture, and the BCS Reception at the Society for Neuroscience Annual Meeting—have also grown in popularity. Our semi-annual department newsletter is enhancing in scope, and is serving both as a link within our expanding department and a means of keeping our alumni informed of our activities.
Twelve graduate students entered in fall 2001, while twelve students graduated with the PhD. Nine of the latter have assumed postdoctoral positions; one is a continuing MD student at Harvard Medical School; one is now a MS student in MIT's Health Sciences and Technology Program; and one is working on neural network models for advanced process control in industry. Five students were honored for excellence in undergraduate teaching, one student was awarded for excellence in graduate teaching, and three were commended for continuing dedication to teaching. In addition, one student won the Institute-wide Goodwin Medal, which recognizes teaching "conspicuously effective over and above ordinary excellence."
The department had a total of 123 undergraduates this year, and the 37 graduating seniors have been replaced by 34 freshmen. One graduating senior won a Rhodes scholarship, one of two awarded at the Institute and the only one awarded to an undergraduate. In addition, another graduating senior won the Karl Taylor Compton Prize, the highest award presented by the Institute to students and student organizations in recognition of excellent achievements in citizenship and devotion to the welfare of MIT. In MIT's 2002 graduating class, 11 of 58 Phi Beta Kappas were BCS majors.
The Brain and Cognitive Sciences Society (BCSS) was officially launched in the fall of 2001 as an undergraduate-run organization whose aim is to provide information about graduate school, career opportunities, and classes and requirements, in addition to promoting speakers and conferences that may be of interest to BCS undergraduates. The organization also aims to unite the BCS community at MIT by encouraging more interactions between undergraduates and faculty, as well as among undergraduates themselves.
Emilio Bizzi was named Institute Professor. Lera Boroditsky received the Searle scholars award and the Surdna Foundation research award . James DiCarlo received the 2002 Pew Scholar award in the biomedical sciences and became an Alfred P. Sloan research fellow for 2002. Ann Graybiel received the James Rhyne Killian Jr. faculty achievement award from MIT and the National Medal of Science for 2002.
Nancy Kanwisher was appointed a MacVicar faculty fellow. Earl Miller was named associate director of the Picower Center for Learning and Memory. Steven Pinker was designated a humanist laureate by the International Academy of Humanism. Tomaso Poggio was named chairman of the Scientific Advisory Board of IRST (Trento, Italy) in 2002. Morgan Sheng was elected president of the Society of Chinese Neuroscientists of America. Mriganka Sur received a distinguished alumnus award from the Indian Institute of Technology, Kanpur Fellow, Neuroscience Research Program. Matthew Wilson received a 2002 Picower Scholars award. Richard Wurtman was appointed Smithies lecturer, Oxford University in May 2002.
More information about the Department of Brain and Cognitive Sciences can be found on the web at http://web.mit.edu/bcs.