Institute of Cognitive and Brain Sciences
University of California at Berkeley
3210 Tolman Hall MC 1650
Berkeley, CA 94720-1650
Administration support for the Institute is provided by the staff of the Helen Wills Neuroscience Institute. See the administration page for help and information.
All talks are in 5101 Tolman Hall, 11am-12:30pm.
Presenter: Gwenaelle Douad, Oxford University
Title: Predicting progression to Alzheimer's and capturing treatment effects in mild cognitive impairment: what can be gained from using MRI?
Abstract: I'll discuss two recent papers emphasising the benefit of imaging people at risk for Alzheimer's disease (AD). In the first one (Douaud et al., J Neurosci 2013), we've examined both volumetric and brain microstructure abnormalities in 13 amnestic patients with mild cognitive impairment (MCI), who progressed to AD no earlier than 2 years after baseline scanning, in order to focus on early, and hence more sensitive, imaging markers. We compared them to 22 stable amnestic MCI patients who did not show progression of symptoms for at least 3 years. In the second one (Douaud et al., PNAS 2013), an imaging study based on a randomized controlled trial in elderly subjects with MCI, we've explored whether it was possible to prevent atrophy of key brain regions related to cognitive decline and Alzheimer's disease by modifying nongenetic risk factors: in this instance by lowering elevated plasma homocysteine using B vitamins.
Presenter: Gary Marcus, New York University
Title: Computational diversity and the mesoscale organization of the cortex
Abstract: The human neocortex participates in a wide range of tasks, yet superficially appears to adhere to a relatively uniform six-layered architecture throughout its extent. For that reason, much research has been devoted to characterizing a single "canonical" cortical computation, repeated massively throughout the cortex, with differences between areas presumed to arise from their inputs and outputs rather than from "intrinsic" properties. There is as yet no consensus, however, about what such a canonical computation might be, little evidence that uniform systems can capture abstract and symbolic computation (e.g., language) and little contact between proposals for a single canonical circuit and complexities such as differential gene expression across cortex, or the diversity of neurons and synapse types. Here, we evaluate and synthesize diverse evidence for a different way of thinking about neocortical architecture, which we believe to be more compatible with evolutionary and developmental biology, as well as with the inherent diversity of cortical functions. In this conception, the cortex is composed of an array of reconfigurable computational blocks, each capable of performing a variety of distinct operations, and possibly evolved through duplication and divergence. The computation performed by each block depends on its internal configuration. Area-specific specialization arises as a function of differing configurations of the local logic blocks, area-specific long-range axonal projection patterns and area-specific properties of the input. This view provides a possible framework for integrating detailed knowledge of cortical microcircuitry with computational characterizations. With Adam Marblestone, MIT and Tom Dean, Google
Presenter: Lara Buchak, UC Berkeley
Title: Risk and rationality
Abstract: This talk is about the principles that govern rational decision-making in the face of risk. A distinctive feature of these decisions is that individuals are forced to consider how their choices will turn out under various circumstances, and to decide how to trade off the possibility that a choice will turn out well against the possibility that it will turn out poorly. The orthodox view is that there is only one acceptable way to do this: rational individuals must maximize expected utility. The contention of this talk, however, is that the orthodox theory (expected utility theory) dictates an overly narrow way in which considerations about risk can play a role in an individual's choices. In particular, the orthodox theory holds that preferences combine two internal attitudes, represented as utilities and credences. I argue for an alternative, more permissive, theory of decision-making: one that allows individuals to pay special attention to the worst-case or best-case scenario. This theory, risk-weighted expected utility theory, holds that preferences combine three internal attitudes, represented as utilities, credences, and risk attitudes. This theory provides a more plausible analysis of each of these attitudes. Furthermore, I argue that decision-makers whose preferences can be captured by risk-weighted expected utility theory are rational.
Presenter: Paul Smolensky, Johns Hopkins University
Title: Gradient symbolic computation
Abstract: I will introduce a notion of Gradient Symbolic Computation (GSC) which is formally specified at two inter-related levels of description. At the micro-level, GSC is a type of neural-network computation, combining two dynamics in parallel. One is a stochastic optimization process, which favors well-formed distributed representations (optimally satisfying weighted grammatical and statistical distributional constraints). The other dynamical process creates attractors at all the distributed representations that realize discrete symbol structures. At the macro-level, GSC representations are gradient symbol structures, describable in two equivalent ways. In the first, a representation is a discrete set of structural relations (e.g., dominance and precedence in a tree) in which the elements being related are blends of symbols with differing activation levels (thus the root of a tree might be labeled 0.3*X + 0.1*Y). In the alternative description, a representation is a set of discrete symbols each occupying a weighted blend of structural positions (so an argument of a verb might fill the role 0.2*Subject + 0.5*Object). The talk will illustrate: conceptual differences between GSC and probabilistic models over discrete structures; formal results on the computational power of GSC; and applications to linguistic theory and language processing. A case study to be sketched in the talk is a highly intuitive analysis of French liaison, in which orthographically-final consonants are sometimes pronounced and sometimes silent. The GSC analysis is simply that in the lexicon, such consonants are of reduced activation and occupy a blend of word-final and word-initial positions. This case illustrates the potential of GSC to resolve the long-standing deadlocks pervasive in grammatical theory in which one theory is favored by one subset of the data, while an inconsistent theory is favored by a different subset. GSC offers the possibility of resolving the deadlock through a weighted blend of the two theories that accounts for both subsets of data.
Presenter: Reza Shadmehr, Johns Hopkins University
Title: A memory of errors in sensorimotor learning
Current view of motor learning suggests that when we revisit a task, the brain recalls the motor commands it previously learned. In this view, motor memory is a memory of motor commands, acquired through trial-and-error and reinforcement. Here we show that the brain controls how much it is willing to learn from the current error through a principled mechanism that depends on the history of past errors. This suggests that the brain stores a previously unknown form of memory, a memory of errors. A mathematical formulation of this idea provides insights into a host of puzzling experimental data, including savings and meta-learning, demonstrating that when we are better at a motor task, it is partly because the brain recognizes the errors it experienced before.
Center for "big data" in psychology to be established at Berkeley
ICBS members recently received a grant from the National Science Foundation to establish a new research center
Alison Gopnik on play and learning
A story in the New York Times highlights Gopnik's recent work on pretend play
How do you refer to your mother's older sister?
Terry Regier and Charles Kemp find human kinship systems are simple and informative
Len Talmy receives Gutenberg Research Award
The first time this prestigious award has been given to a cognitive scientist.
Tania Lombrozo receives the Janet Taylor Spence Award
A significant early career award from the Association for Psychological Science.
What do babies think?
Alison Gopnik gives a presentation at TED.
Neurotechnology prize for Jose Carmena
Carmena received the New York Academy of Scienes Aspen Brain Forum Prize in Neurotechnology
Forget Mozart, Try Kant.
Alison Gopnik on the inquiring baby.
Jerry Feldman awarded the Berkeley Citation
Tribute for leadership in promoting interdisciplinary research and opportunities for study in computer science.
The Interpretation of Dreams, 2010
Jack Gallant uses fMRI to take a peek at what's on your mind
New Advances in Brain-Machine Interface Technology
Jose Carmena and Karunesh Ganguly demonstrate that the brain can consolidate a neural representation for long-term prosthetic control
Conference on Neurocognitive Development, July 12-14, 2009
Conference proceedings available on-line.
George Lakoff named Richard and Rhoda Goldman Distinguished Professor
Professorship in Linguistics and Undergraduate and Interdisciplinary Studies for 2009-13
Fillmore FEST: Conference in honor of Charles Fillmore, July 31 - Aug 2
Celebrate Chuck's 80th birthday at Berkeley during this special meeting.
The Mental Navigator
Alison Gopnik's new book, "The Philisophical Baby" is discussed in the Boston Globe and SF Chronicle
Archived Files from past ICBS Seminars and Workshops
Click for links to materials
Berkeley Cognitive Science Undergraduates to Host Conference
First annual California Cognitive Science Conference on April 25
Darwin Day Celebrated at ICBS
Webcasts from the workshop, "Evolution in the Cognitive Sciences"
Alison Gopnik offers her ideas for revolutionizing undergraduate education.
In a Slate commentary, Gopnik suggests how to create "scientists in the dorm".
A Personal Experience with Synesthesia
Sherri Roush discusses how sounds make her skin tingle-- literally.
Alison Gopnik sounds off in Slate on mirror neurons.
Rea2007 essay on, "What the myth of mirror neurons gets wrong about the human brain."
Paul Kay joins discussion in NY Times on how language influences thought
Science Times article discusses new views on an old debate.
Art Shimamura awarded a Guggenheim Fellowship
Fellowship to support his efforts on a new book taking a neurocognitive approach to the psychology of art and aesthetics.
Conference on Religion and Cognitive Science, Jan 16-18
Hosted by The Graduate Theological Union and the Cognitive Science Program at UC Berkeley
George Lakoff to receive inaugural Giulio Preti Prize
The Regional Council of Tuscany honors Lakoff's contributions in the study of Science and Democracy.
Steve Palmer discusses the science of aesthetics in a presentation at Google.
Updated web materials for Jerry Feldman's book, 'From Molecule to Metaphor: A Neural Theory of Language'.
Check out recent Updates, Teaching Tips, and a Readers' Roadmap- a remarkable graphical guide to each chapter produced by John Torous, a recent student in CogSci 110 at Berkeley.
Looking back into the future.
Eve Sweetser's cross-cultural research on how we speak about time.
How do People Make Predictions?
New ICBS Member Tom Griffiths puts Bayes rule to the test
Linguistic Modulation of Color Perception Differs for the Left and Right Cerebral Hemispheres.
ICBS members Kay and Ivry use neuropsychological methods to revisit the Whorf hypothesis
Alva Noe brings the brain and body together in his new book.
Conversation with Alva Noe on the just published, "Action in Perception"
LEONARDO review of recent Synesthesia Conference co-hosted by ICBS
Review of the The Fourth Annual National Conference of the American Synesthesia Association, Inc.
"What Do You Believe Is True Even Though You Cannot Prove It?"
ICBS member Alison Gopnik and other scientists consider unsolvable problems
George Lakoff of Linguistics and ICBS on the Language of Politics
UC Berkeley professor George Lakoff tells how conservatives use language to dominate politics.