[CTN] informal talk: Dr. Adam Johnson (Bethel University), Thu Aug 23, 3.30pm in PAS 2464

[Matthijs van der Meer]

Dear all,

The official CTN seminar series kicks off again on Sept. 11
(http://ctn.uwaterloo.ca/research/upcomingevents.html).

However, I thought there may be interest in an informal talk by my
friend and colleague Adam Johnson. Adam uses tools and perspectives from
statistics and machine learning to create models learning and
exploration in the brain, and the hippocampus in particular. (Recent
paper: "The hippocampus and exploration: dynamically evolving behavior
and neural representations",
http://www.frontiersin.org/Human_Neuroscience/10.3389/fnhum.2012.00216/abstract).

Adam's talk whill be this Thursday (Aug 23) at 3.30pm in PAS 2464.

If you'd like to meet with Adam, please let me know! Approximate title
and abstract for his talk follow below.

Hope to see you there,

-- Matt


Dr. Adam Johnson
Assistant Professor, Department of Psychology
Bethel University, St. Paul, MN

http://people.bethel.edu/~acjohnso/

Title: Learning and exploring in the hippocampus

Abstract: Recent experimental studies suggest that the hippocampus plays
a central role in the development of schematic representations that
contribute to single trial learning and memory consolidation (Tse et
al., 2007, 2011). We use a hierarchical Bayesian approach to model
developing schematic representations, single trial learning, and memory
consolidation in the paired associate task. The model captures the
fundamental behavioral observations from the paired associate task and
provides an intuitive statistical understanding of task schemas and
time-variable memory consolidation. Using this probabilistic approach,
we show how multiple hippocampal maps develop as the product of task
schemas that represent the intrinsic statistical characteristics of a
behavioral task and how reactivation of previous experience contributes
to consolidation. Finally, we integrate our probabilistic approach with
an information theoretic approach to show how animals can most
efficiently explore their environment and memory using previous
experience.