Paul R. Adams
Professor
PhD, University of London
Life Sciences Building - Room 518
Office: (631) 632-6938
Fax: (631) 632-6661
Training
Paul Adams attended Cambridge University (UK) and received a B.A. in Physiology and Pharmacology in 1968. He continued his education at London University (UK), receiving a Ph.D. in Pharmacology in 1974. Professor Adams was on the faculty of the University of Texas from 1977 to 1981. He came to the State University of New York at Stony Brook as Associate Professor of Neurobiology and Behavior in 1981 and was promoted to the rank of Professor in 1984. He held concurrent appointments of Professor of Neurology from 1987 to 1989 and Professor of Pharmacological Sciences from 1987 to 2002. From 1987 to 1995 he was a Howard Hughes Medical Institute Investigator. Professor Adams was awarded a MacArthur Foundation Prize in 1986 and was elected a Fellow of the Royal Society in 1991. He has served on the Editorial Boards of numerous journals (including, currently, Frontiers in Neural Circuits)
Research
A brain is a network of billions of interconnected neurons, and synaptic connections
have modifiable strengths. The modifications occur as a result of correlated activity
of the connecting neurons, and underlie learning, which can progressively improve
the utility of the behavior the brain generates in response to sensory input. I’m
interested in 2 interrelated aspects of this process: (1) because synapses are extremely
densely packed, modifications at one synapse can spuriously affect the strengths of
neighboring synapses (“crosstalk”); (2) because of this problem, and other learning
difficulties, individual brains may fail to learn useful solutions, but inter-brain
communication (“networks of networks”) might allow sharing of rare but useful individual
learning. We study these issues using simplified, mathematical models of network plasticity.
We have shown that even extremely low levels of crosstalk can prevent useful learning.
We propose that this problem could be overcome, in the neocortex, by a process we
call “Hebbian proofreading”, using some neurons (e.g. in layer 6) as detectors and
editors of correlated activity between other neurons (e.g. in upper layers). We have
also shown that inter-brain communication fosters both useful and spurious learning,
and are exploring social mechanisms that might overcome this difficulty.