Parallel and Interactive Memory

1
Parallel and Interactive Memory Systems in
the Mammalian Brain
2
Human Brain
3
(No Transcript)
4
Sagittal view showing coronal levels
Adapted from Swanson (1992)
Swanson, L.W. (1992). Brain Maps Structure of
the Rat Brain. Amsterdam Elsevier.
5
Level 14 of 73
Coronal section
Adapted from Swanson (1992)
Cresyl violet stain
Swanson, L.W. (1992). Brain Maps Structure of
the Rat Brain. Amsterdam Elsevier.
6
Level 28 of 73
Coronal section
Dorsal Hippocampus
Amygdala
Adapted from Swanson (1992)
Formol-thionin stain
Swanson, L.W. (1992). Brain Maps Structure of
the Rat Brain. Amsterdam Elsevier.
7
Associative memories
1) Stimulus-response (S-R) 2) Stimulus-affect
(S-A) 3) Stimulus-stimulus (S-S)
8
(No Transcript)
9
Radial Arm Maze Tasks
1) Spatial Win-Shift (SWS) 2) Conditioned Cue
Preference (CCP) 3) Cued Win-Stay (CWS)
10
Spatial Win-Shift (SWS) Task
11
Cued Win-Stay Task

12
Conditioned Cue Preference Task


Food
Paired
Unpaired
No Food
Training
Testing
13
Summary of Radial Arm Maze Results
McDonald White (1993). A triple dissociation of
memory systems Hippocampus amygdala, and
dorsal striatum. Behavioral Neuroscience, 107,
3-22.
14
Parallel Independent Memory Systems
S-R
S-A
S-S
Dorsal Striatum
Lateral Amygdala
Hippocampus
Motor systems and Behavior
15
Interactions Among Memory Systems
Three major types of interactions could occur
and depend on the demands of particular tasks
for descriptive purposes, we can define these in
terms of three major types of tasks Category A
tasks fit precisely the mnemonic function of a
single system, and the other systems may
interfere with or have no influence on the
acquisition of category A tasks. Tasks in
category B can be acquired by more than one of
the memory systems, with each system acting
alone. Tasks in category C may require the
function of two or more of the systems for
accurate performance. (McDonald White, 1993,
p. 17).
McDonald White (1993). A triple dissociation of
memory systems Hippocampus amygdala, and
dorsal striatum. Behavioral Neuroscience, 107,
3-22.
16
Concurrent Cue-Place Task in the Water Maze
Involves acquisition of both S-R and S-S
information. However, its a mix of category A
and B tasks. Some days allow simultaneous
acquisition of S-R and S-S associations (category
B), while others require exclusive use of S-S
information (category A). At the end of the
task, the two types of associations are placed in
competition to determine which has greater
control of performance (category A).
McDonald, R.J., White, N.M. (1994). Parallel
information processing in the water maze
Evidence for independent memory systems involving
dorsal striatum and hippocampus.
Behavioral and Neural Biology, 61, 260-270.
17
Water maze Cue-place training
Two days with the visible platform at a fixed
spatial location
Third day with the submerged platform at the
same location
18
Cue-Place Acquisition
21 cue-place series repeated 3 times (9 days)
Competition Test
Visible platform moved to opposite quadrant
adapted from McDonald, R.J., White, N.M.
(1994). Parallel information processing in the
water maze Evidence for independent
memory systems involving dorsal striatum and
hippocampus. Behavioral and Neural
Biology, 61, 260-270.
19
Water maze Competition Test
Cue Response
Place Response
20
McDonald White (1994)
McDonald, R.J., White, N.M. (1994). Parallel
information processing in the water maze
Evidence for independent memory systems involving
dorsal striatum and hippocampus.
Behavioral and Neural Biology, 61, 260-270.
21
(No Transcript)
22
Swanson, L.W. (1992). Brain Maps Structure of
the Rat Brain. Amsterdam Elsevier.
23
Experiment 1 Bilateral Lesions
Devan, B.D., White, N.M. (1999). Parallel
information processing in the dorsal striatum
Comparison to hippocampal function. Journal
of Neuroscience, 19, 2789-2798.
24
(No Transcript)
25
(No Transcript)
26
Devan, B.D., White, N.M. (1999). Parallel
information processing in the dorsal striatum
Comparison to hippocampal function. Journal
of Neuroscience, 19, 2789-2798.
27
Experiment 1 Competition Test Results
Table. Number of Rats That Swam to the Former
Platform Location (Place Responders) Versus the
New Visible Platform Position (Cue Responders) on
the Competition Test (Day 10)
Devan, B.D., White, N.M. (1999). Parallel
information processing in the dorsal striatum
Comparison to hippocampal function. Journal
of Neuroscience, 19, 2789-2798.
28
Interactive Memory Systems
S-S
S-R
Hippocampus
Lateral Striatum
Medial Striatum
competition
Motor systems and Behavior
29
Interactive Memory Systems
S-S
S-R
Hippocampus
Lateral Striatum
Medial Striatum
Impaired spatial learning
No competition
S-R enhancement/S-S impairment
30
Interactive Memory Systems
S-S
S-R
Hippocampus
Lateral Striatum
Medial Striatum
Normal spatial learning during acquisition
No competition
S-R impairment/S-S enhancement
31
Interactive Memory Systems
S-S
S-R
Hippocampus
Lateral Striatum
Medial Striatum
Normal spatial learning during acquisition
No competition
(S-S)-R impairment/S-R enhancement
32
Interactive Memory Systems
S-S
Hippocampus
?
Motor Systems and Behavior
33
Interactive Memory Systems
S-S
Hippocampus
Disconnection
?
Motor Systems and Behavior
34
Anatomical Connectivity
Medial prefrontal cortex
Caudate- Putamen
Hippocampus
35
Complete Disconnection
Medial prefrontal cortex
Caudate- Putamen
Hippocampus
Commissurotomy
36
Cross-unilateral lesion (partial disconnection)
Medial prefrontal cortex
Caudate- Putamen
Hippocampus
37
Experiment 2 Unilateral Lesions
Devan, B.D., White, N.M. (1999). Parallel
information processing in the dorsal striatum
Comparison to hippocampal function. Journal
of Neuroscience, 19, 2789-2798.
38
(No Transcript)
39
Devan, B.D., White, N.M. (1999). Parallel
information processing in the dorsal striatum
Comparison to hippocampal function. Journal
of Neuroscience, 19, 2789-2798.
40
Experiment 2 Competition Test Results
Table. Number of Rats That Swam to the Former
Platform Location (Place Responders) Versus the
New Visible Platform Position (Cue Responders) on
the Competition Test (Day 10)
Devan, B.D., White, N.M. (1999). Parallel
information processing in the dorsal striatum
Comparison to hippocampal function. Journal
of Neuroscience, 19, 2789-2798.
41
Experiment 2 Competition Test Results
Because the crossed-unilateral lesion resulted in
a significant cue bias on the competition test we
can conclude that connections (direct and/or
indirect) between the hippocampus and medial
striatum are important for allowing S-S
information to compete with S-R behavior.
Because a complete disconnection was not
necessary, the result has the added advantage
that alternative interpretations based on partial
disconnection of other structures besides
hippocampus and medial striatum, through
commissurotomy, cannot account for the finding.
Devan, B.D., White, N.M. (1999). Parallel
information processing in the dorsal striatum
Comparison to hippocampal function. Journal
of Neuroscience, 19, 2789-2798.
42
Parallel and Interactive Memory Systems in
the Human Brain
43
Probabilistic Classification Task
Knowlton, B.J., Mangels, J.A., Squire, L.R.
(1996). A neostriatal habit learning system in
humans. Science, 273, 1399-1402.
44
(A) Performance on the probabilistic
classification tasks by controls (CON, n 15),
amnesic patients (AMN, n 12), patients with
Parkinsons disease (PD, n 20), and a subgroup
of PD patients with the most severe symptoms
(PD, n 10). (B) Performance on the
declarative memory task. Both PD and PD groups
exhibited entirely normal declarative memory for
facts about the testing episode, despite their
poor performance on the task itself. In
contrast, amnesic patients exhibited a severe
impairment in declarative memory for the testing
episode but normal performance on the
classification test.
Knowlton, B.J., Mangels, J.A., Squire, L.R.
(1996). A neostriatal habit learning system in
humans. Science, 273, 1399-1402.
45
Probabilistic Classification Tasks for FMRI
Studies
Poldrack, R.A., Clark, J., Paré-Blagoev, E.J.,
Shohamy, D., Moyano, J.C., Myers, C., Gluck,
M.A. (2001). Interactive memory systems in
the human brain. Nature, 414, 546-550.
46
a. Activation for FB compared to baseline (yellow
increase, blue decrease) b. Activation for
PA compared to baseline c. Regions exhibiting
significant differences between FB and PA tasks
d. Plot of task related signal change from the
MTL region exhibiting maximal task-dependent
differences against a region in the right caudate
that exhibited significant negative correlation
with the MTL in functional connectivity analysis.
Each data point represents a single subject.
Poldrack, R.A., Clark, J., Paré-Blagoev, E.J.,
Shohamy, D., Moyano, J.C., Myers, C., Gluck,
M.A. (2001). Interactive memory systems in
the human brain. Nature, 414, 546-550.
47
Results from event-related FMRI study of FB
category learning (experiment 2). a, Regions
exhibiting significant evoked activation (yellow)
or deactivation (blue) for classification trials.
Yellow arrow highlights region of caudate
activation, white arrow highlights region of MTL
deactivation. b, c, Depiction of parametric
change in modelled evoked haemodynamic response
across the initial 450-s scanning run (averaged
across subjects) in b, left body of caudate
nucleus (-12, 3, 21), and c, left MTL (-24, -3,
-24). Red indicates positive, event-related
response, blue indicates negative event-related
response.
Left body of caudate nucleus
Left medial temporal lobe
Poldrack, et al. (2001). Interactive memory
systems in the human brain. Nature, 414, 546-550.
48
What can/should we conclude from this study? The
authors state that the results provide the
first substantive evidence, to our knowledge, for
competition between memory systems in the human
brainthe present study provides direct evidence
for competition at the neural level by
demonstrating three essential features of the
MTL-striatum interaction. First, it shows that
engagement of MTL and striatum is modulated by
whether the task encourages the use of
declarative versus nondeclarative memory
processes or strategies. Second, it demonstrates
that engagement of these regions is negatively
correlated across subjects. Third, it
demonstrates rapid reciprocal changes in the
engagement of these regions. These data are
concordant with animal lesion studies
demonstrating that the memory systems based on
the MTL and striatum can compete with one another
during learning.
Poldrack, R.A., Clark, J., Paré-Blagoev, E.J.,
Shohamy, D., Moyano, J.C., Myers, C., Gluck,
M.A. (2001). Interactive memory systems in
the human brain. Nature, 414, 546-550.
49
However, the authors also state that their
computational theory interprets both the
earlier animal data and the present human imaging
data as implying an interaction between the
hippocampus and other brain structures, in which
the hippocampus has a modulatory role in learning
by developing new stimulus representations during
early phases of training which are used by the
striatum to develop complex stimulus-response
associations.
This quote does not imply or suggest, but
directly states, that the hippocampus and
striatum interact cooperatively! So how can the
results be the first substantive evidence of a
competitive interaction?
Poldrack, R.A., Clark, J., Paré-Blagoev, E.J.,
Shohamy, D., Moyano, J.C., Myers, C., Gluck,
M.A. (2001). Interactive memory systems in
the human brain. Nature, 414, 546-550.