|
Randy Bruno
Personal Webpage
After I had received all my decision letters, choosing a graduate
program seemed like an impossible task. My choice was influenced
heavily by one aspect in particular: The community of scientists
interested in the brain and its behavior was seemingly not larger or
more diverse anywhere else in the world than in Pittsburgh. This was
so important because I had been looking for a place that would help
unify my own diverse interests. My bachelors had been in cognitive
science, I had later worked on modeling issues as a research assistant
in statistics, and I had finally recognized what my career should be
in 'neuroscience'. While desiring intense training in one discipline, I
didn't want to surrender a multi-disciplinary perspective. This
thinking led me to choose the Neuroscience Graduate Training Program
at the University of Pittsburgh and to join the Center for the Neural
Basis of Cognition.
The general question I wanted to ask was: How does a biological neural
network perform any perceptual or cognitive function? To answer this
computational question, one needs to know the biophysical properties
of the neurons involved, the pattern of connections between them, and
the nature of the network's inputs. (Tough stuff! Can you think of a
complete description of a real neural network in any system?) After
several lab rotations, I posed this question to Dr. Daniel Simons
(Pitt Neurobiology). His reply was 'I think I have a problem for you
to work on' and, in some sense, defined the start of my doctoral
thesis.
The CNBC is a diverse place. Dan, who I asked to be my advisor, has
had a long-standing collaboration with Dr. Bard Ermentrout (Pitt
Mathematics), and together they had produced a computational model of
the thalamocortical circuit in the somatosensory system. The model
makes many assumptions regarding the patterns and strengths of
connections. My research has primarily focused on determining
experimentally the exact nature of those connections. I recorded from
thalamic and cortical neurons in vivo, mapped out their receptive
field properties, and used cross-correlation techniques to infer the
presence and strength of monosynaptic connections, if any, between the
cells. Inferring anatomical connections from the spiking activity of
neurons requires sophisticated statistics. I knew of a powerful, but
previously unused, method for doing this and was able to turn to
Dr. Valerie Ventura (CMU Statistics) for advice. In the end, my
experiments not only confirmed many of the major predictions of our
network model but also indicated features needing revision.
The CNBC community is something more than simply a place to find
advice. While a graduate student, I've had the privilege of so many
others' perspectives and insights, shared in formal and informal
ways. Throughout the year, distinguished speakers from other
institutions, invited for the annual colloquium series, impart their
ideas and opinions in their lectures and meetings with students. The
student-organized 'brain bags', talks by fellow graduate students
spread throughout two universities and numerous departments, have been
enormously helpful in learning about questions (and even fields) other
than my own and in getting to know the people who work so hard on
them.
These days, I am currently finishing manuscripts and
preparing to complete and defend my thesis (tentatively in fall
2002). In January 2003, I will begin a postdoc position at the Max
Planck Institute in Heidelberg, Germany, with Bert Sakmann (winner of
the 1991 Nobel Prize in Medicine). I expect and hope that, while I am
there and everywhere subsequent, the CNBC's unique view will remain
with me. I hope that view becomes yours, too.
Home Program/Department: CNUP/Neurobiology
Advisor: Daniel J. Simons
Year: Fifth
Research Interest: Computation in small local circuits; thalamocortical connectivity of the somatosensory system
|
