The hierarchical hypothesis (Re-send of lost final message; thanks, Dave.)

[Richard Granger]

We were long puzzled by our biological models' persistent tendency to
produce different cortical outputs over successive theta cycles, and
eventually recognized that what these outputs might be encoding was
sequential hierarchical information (the hypothesis whose formal statement
ultimately appears in the 1990 Science paper).  The hypothesis was novel
and exciting to us -- but perhaps the finding was already obvious to
everyone except us (and the reviewers and editors at Science).  We've
received many private messages indicating that most people in the field do
recognize the history as we've described it, and that it is easy to
misconstrue in hindsight (the continuing special investigation into our
lack of foresight notwithstanding).  Our thanks to the many writers of
those messages!   (One friend reminded us that Postmodernism has clearly
shown that authors know far less than their Text knows, and accordingly
admonished us to "shut up, sit down, and listen to the story of your life
as it Really Happened."  :-)

On the other hand, the discussion is shot through with useful threads
twining around understanding of the distinctions and relationships among
experimental findings, construction of simulations, observation of
simulations, and formal characterization and simplification of results,
much as in physics.  As we come to better understand the differential
nature of experiments, versus simulations, versus formal characterization,
our ability to talk constructively across the boundaries of computation and
biology correspondingly improves.

Finally, it's worth noting that this hypothesis (that cortical neurons
differentially respond over sequential cycles, yielding successive
hierarchical information) is readily differentiated from other hypotheses.
Perhaps, then, comfort can be taken from the realization that physiological
and behavioral evidence may one day demonstrate that some competing
hypothesis is correct after all.


-Rick Granger
 granger at uci.edu


[A number of writers have requested references to subsequent publications
of ours, so a partial list is appended.   (Those of you who requested
references to our research on ampakines, I'll send that list in a separate
message.)  ]

Selected topical bibliography since '90:

Ambros-Ingerson, J., Granger, R., and Lynch, G. (1990).  Simulation of
paleocortex performs hierarchical clustering.   Science, 247: 1344-1348.

Granger, R., Staubli, U., Powers, H., Otto, T., Ambros-Ingerson, J., and
Lynch, G. (1991).  Behavioral tests of a prediction from a cortical network
simulation.  Psychol. Sci., 2: 116-118.

McCollum, J., Larson, J., Otto, T., Schottler, F., Granger, R., and Lynch,
G. (1991).  Short-latency single-unit processing in olfactory cortex.  J.
Cog. Neurosci., 3: 293-299.

Anton, P., Lynch, G., and Granger, R. (1991).  Computation of
frequency-to-spatial transform by olfactory bulb glomeruli.  Biol. Cybern.,
65: 407-414.

Granger, R., and Lynch, G. (1991).  Higher olfactory processes: Perceptual
learning and memory.  Current Opin. Neurosci., 1: 209-214.

Coultrip, R., Granger, R., and Lynch, G. (1992).  A cortical model of
winner-take-all competition via lateral inhibition.  Neural Networks, 5:
47-54.

Lynch, G. and Granger, R. (1992).  Variations in synaptic plasticity and
types of memory in  cortico-hippocampal networks.  J. Cog. Neurosci., 4:
189-199.

Granger, R. and Lynch, G. (1993).  Cognitive modularity: Computational
division of labor in the brain.  In: The Handbook of Neuropsychology, New
York: Academic Press.

Gluck, M. and Granger, R. (1993).  Computational models of the neural bases
of learning and memory. Annual Review of Neurosci. 16: 667-706.

Anton, P., Granger, R., and Lynch, G. (1993).  Simulated dendritic spines
influence reciprocal synaptic strengths and lateral inhibition in the
olfactory bulb.  Brain Res., 628: 157-165.

Coultrip, R. and Granger, R. (1994).  LTP learning rules in sparse networks
approximate Bayes classifiers via Parzen's method.  Neural Networks, 7:
463-476.

Kowtha, V., Satyanarayana, P., Granger, R., and Stenger, D. (1994).
Learning and classification in a noisy environment by a simulated cortical
network.  Proceedings of the Third Annual Computation and Neural Systems
Conference,  Boston: Kluwer, pp. 245-250.

Granger, R., Whitson, J., Larson, J. and Lynch, G. (1994). Non-Hebbian
properties of LTP enable high-capacity encoding of temporal sequences.
Proc. Nat'l. Acad. Sci., 91: 10104-10108.

Myers, C., Gluck, M., and Granger, R. (1995).  Dissociation of hippocampal
and entorhinal function in associative learning: A computational approach.
Psychobiology, 23: 116-138.

Ozeki, T., Shouval, H., Intrator, N. and Granger, R. (1995).  Analysis of a
temporal sequence learning network based on the property of LTP induction.
In: Int'l Symposium on Nonlinear Theory, Las Vegas, 1995.

Kilborn, K., Granger, R., and Lynch, G. (1996).  Effects of LTP on response
selectivity of simulated cortical neurons.  J. Cog. Neurosci., 8: 338-353.

Granger, R., Wiebe, S., Taketani, M., Ambros-Ingerson, J., Lynch, G.
(1997).  Distinct memory circuits comprising the hippocampal region.
Hippocampus, 6: 567-578.

Hess, U.S., Granger, R., Lynch, G., Gall, C.M.  (1997).  Differential
patterns of c-fos mRNA expression in amygdala during sequential stages of
odor discrimination learning.  Learning and Memory, 4: 262-283.