Norman Hippocampus and Cortex Complementary Learning Memory Model 2010

Jun 06, 2012 07:56 PM

Note: paper is both online and in my pdf colection.

appeared in Hippocampus in 2010;  single author = Kenneth Norman  (Princeton);

gave this talk in Stfd Psych (Jordan Hall)  on June 6.

ABSTRACT: We describe how the Complementary Learning Systems

neural network model of recognition memory (Norman and O’Reilly

(2003) Psychol Rev 104:611–646) can shed light on current debates

regarding hippocampal and cortical contributions to recognition memory.

We review simulation results illustrating three critical differences

in how (according to the model) hippocampus and cortex contribute to

recognition memory, all of which derive from the hippocampus’ use of

pattern separated representations.

Pattern separation makes the hippocampus

especially well-suited for discriminating between studied items

and related lures; it makes the hippocampus especially poorly suited for

computing global match; and it imbues the hippocampal ROC curve

with a Y-intercept > 0. We also describe a key boundary condition on

these differences: When the average level of similarity between items in

an experiment is very high, hippocampal pattern separation can fail, at

which point the hippocampal model will start to behave like the cortical

model. We describe the implications of these simulation results for

extant debates over how to describe hippocampal versus

 cortical contributions and how to measure these contributions. VC

2010 Wiley-Liss, Inc.

page 1   KEY IDEA:  The most important of these ideas is that

HC implements  PATTERN SEPARATION  (HC is biased to

assign DISTINCT REPS to input patterns, regardless of their similarity,

but cortex does not ......

COMPLEMENTARY LEARNING SYSTEMS FRAMEWORK... =CLS

division of labor btwn HC and Cortex... McClelland 95.

According to the CLS framework,

neocortex forms the substrate of our internal model of the

structure of the environment.

In contrast, hippocampus is specialized for rapidly and

automatically memorizing patterns of

cortical activity, and so they can be recalled later

based on partial cues (pattern completion).

The model posits that neocortex

learns incrementally; each training trial results in relatively

small adaptive changes in synaptic weights. These small changes

allow cortex to gradually adjust its internal model of the environment in respons e to new informatio n.

The other key prop-erty of neocorte x (according to the model) is that it assigns

simila r (overlapping) rep resentation s to similar stimuli.

Use of overlapping repre sentation s allows corte x to repre sent the

shared structure of events and therefore makes it possi ble for

cortex to gen eralize to novel stim uli based on their simila rity to

pre viously expe rienced stimuli.

 In contr ast, the mod el posits

that hippo campus assigns distinct pattern separated responses to stimuli,

 regardless of their similarity.  This property

allows hippocam pus to rapidly memorize arbitary patterns of

cortical activity with out suff ering unacceptably  high (catastrophic)  

levels of interference.

Norman Hippocampus and Cortex Complementary Learning Memory Model 2010

Attachments