Dudai, Macphail
, Schacter ,italices signify other sources
What structures are involved in learning/memory?
I. How is this question addressed.
A. Establish a very specific paradigm. eg.
Water maze, Nictitating membrane response, NMR
B. Establish what the CR is going to be
1. non-specific response
Readings: Dudai Ch 11 pgs 166-176
other resources: Martinez & Kesner (1991) Learning and Memory a Biological
View. Academic Press. Ch 9.
Bear et al. (1996) Neuroscience: Exploring the Brain. Williams & Wilkins.
pg 415
2. specific
Readings: Dudai various places
other sources:
-
leg flexion in response to shock to the leg
-
eyeblink in response to airpuff
-
eyblink in response to glabalar tap
-
Nicitating membrane response to airpuff
-
potentiated startle
C. Determine the circuitry involved.
-
electrophysiology - cell firing - are changes in cell firing correlated
to the occurance of the CR?
-
lesioning - do lesions prevent the acquisition and retention of CRs.
-
direct stimulation - can stimulation of a region elicit a UCR or CR.
-
direct stimulation - can stimulation of a region substitute for the CS.
II.Actual structures involved in storage of memories.
A. Sensory-motor pathway
1. Aplysia -
-habituation, sensitizatoin, dishabituation
-classical conditioning
-instrumental conditioning
2. Spinal cats - multisynaptic circuit
- habituation, sensitization, dishabituation
- classical conditioning (Patterson, Cegavske,
Thompson 1973)
- instrumental conditioning (pg154) leg-water-shock
.
B. Cerebellum
1. NMR - (pgs 176-183)
Paradigm-
UCS 100msec pulse
CS -350msec 1-kHz 85dB tone
CR long latency retraction of eye (80msec), coverage by eylid
and nictating membrane (>100msec)
UCR - short latency retraction of eye (7msec), coverage by eylid and
nictiating membrane (22msec).
Standard pairing - 1 session/day
120 trials/session
250 msec btw tirals
Multiple sessions - usually took 1 - 4 days
criterion 70 to 80% CRs
Circuitry -
-
UCS /UR
-
light air puff-UCR is eliminated by lesions of trigeminal-accessory abducens,
Cranial nerve VI or VII (depending on UCR). Stimulation of theses same
nuclei can produce a UCR. Therefore, circuit involves trigeminal
(spinal or descending), Cranial Nerves VI and VII.
-
lesions of DAO (inferior olivary) eliminate the ability of the airpuff
to act as a UCS.
-
CS/CR
-
As stated above for the UCR, stimulation of Cranial nerves VI and VII can
cause extension of the NMR.
-
stimulation of accessory abducens and facial nuclei as well as red nucleus
can produce an extension of the nictitating membrane.
-
Stimulation of areas in the deep cerebellar nuclei that have shown a pairing
associated increase in response also elicit NMR.
-
lesions of all the above areas plus the coclear nuclei eliminates
the CR.
-
Extraneous circuitry: models of CS aquisition are found however lesions
of these regions do not disrupt learning.
Evidence for storage in Cerebellum:
Cerbellar Cortex
-
CS and UCS information seem to converge onto cells (purkinje cells) in
the cerebellar cortex.
-
Some cells exhibit a "model" of the CR.
-
If stimulation of the lateral reticular nucleus ( a source of mossy fibers)
was used as a CS unilateral lesion of lobule HV1 abolish and prevent reacquisition
of the CR. This result was unfortunately not replicable with tone or light
CSs but was repeated .
Deep Cerebellar nuclei
-
Collaterals of inputs from the pontine nuclei and the inferior olivary
complex enter the deep cerebellar nuclei and presumable converge
there.
-
Some cells exhibit a "model" of the CR.
-
Lesions abolihs the CR. However, when activity in these nuclei is
suppressed with lidocaine during training CS-UCS pairing still produces
an association because a CR will be elicited after training eventhough
none was present during training.
Not in the DAO/inferior olive: lesions of DAO (inferior olivary)
after training decrease the CR in a similar fashion to when extinction
procedures are invoked but does not eliminate the CR immediately.
(but see Yeo et al.)
More complex paradigms or monitoring of different CRs had slightly different
results:
2. VOR - vestibular ocular reflex. (pg 183) stabilizes the eye against
changes in head position.
Paradigm - change the relationship between where an object is in space
and where we are in space. The process is sometimes referred to as Error
detection.
Circuitry -
CS - labyrinth of inner ear - mossy fibers -
UCS - ocular signal - climbing fibers -
Evidence for storage in cerebellum
Ito : Demonstrated that models of CS aquisition were evident in cerebellar
nuclei and cortex. Lesions of said structures disrupted both learning and
retention of learned information.
Cerebellar plasticity: LTD (pg 185) conjoint activation of climbing
fibers allows for parallel fibers to produce a large enough calcium signal
that then activates processes that decrease the efficacy of future parallel
fiber activation. This actually serves to increase the output because
the Purkinje cells act as suppressors of activity in the deep cerebellar
nuclei. Thus a decrease in Purkinje activity would serve to increase
activity in deep cerebellar nuclei.
1-4 Hz stimuli of both the climbing and parallel fibers for a period
of 4-6 minutes will result in selective attenuation of the parallel-fiber
Purkinje neuron synapse (typically 20-50 %). No changes occur in the amplitude
of the climbing fiber response. This can be seen in vivo, in slices
and in primary cultures. Furthermore, this process is independent
of the activation of NMDA receptors. The induction of changes in synaptic
efficacy seem to require increases in post-synaptic calcium and the expression
seems to due to changes in the sensitivity of the postsynaptic receptors.
Activation of AMPA and metabotropic glutamate receptors by parallel fibers
during increases in intracelllular calcium (naturally caused by stimulation
of climbing fibers) has been shown to be necessary.
Interestingly Ekerot & Kato found that the optimal interval
for CS-US presentation was to have the climbing fiber (US) precede the
parallel-fiber stimulation by 125- 250 msec. Schreurs and Allcon
(1993) found that the best interval was 50msec with US preceding CS.
Chen and Thompson (1992) however, show the optimal interval to be 250msec
with the parallel fiber stimulation preceding the climbing fiber stimulation.
(Linden & Connor 1995 Long-term Synaptic Depression Annual review of
Neuroscience 18:319-357)
(but see Llinas - inferior olivary complex)
Formal models of how the cerebellar may be accomplishing the feat of compensating
for or changing reactions to various stimuli were first developed by Marr
and Albus (dudai 185). Before it was found that parallel fibers decreased
their efficacy when paired with climbing fiber stimulation Albus hypothesised
that, based on cerebellar microcirtuitry that the site of storage for memory
would be at the parallel fiber Purkinje cell synapse but he said that the
strength of the connections would increase. Marr on the other hand
modified Albus' formal proposal by pointing out that if the cerebellum
were to be responsible for the behavioral changes the synaptic strength
at the parallel fiber - Climbing fiber synapses would need to decrease.
C. Inferior olivary complex -
Paradigm: motor compensation in response to vestibular lesion ( pg
183) Llinas
Circuitry:
CS vestibular information comes through mossy fibers
UCS - visual inputs come through inferior olive climbing fiber pathway
Evidence for storage in inferior olivary complex:
Compensation is blocked by inactivation of the inferior olive and lesions
of the cerebellum.
Retention is blocked by post-training lesions of the inferior olivary
nucleus suggesting that the change necessary for compensation occurs in
the inferior olivary nucleus.
D. Red nucleus - (pg 184-185) Tsukahara -
Paradigm - tone is paired with an electric shock to the forelimb in
cats
Circuitry -
CS - Same as for NMR
UCS - slightly different more akin to eyeblink in cat.
Evidence for storage in red nucleus:
Stimulation of cortico rubral tract can substitute for CS.
Stimulation of cerebellar output to red nucleus can substitute for
UCS
Electrically induced plasticity caused both electrophysiological changes
and anatomical ones.
E. Neocortex
1. Lashley's experiments (Ch 10 )
Paradigm - complex maze
Circuitry - unknown
Evidence for storage in Neocortex: -
Lesion of Neocortex, if extensive enough, causes lose of memory for
running maze-
concluded that memory is stored as a distributed representation
2. Penfield (Ch 10)
Paradigm - stimulation of various areas of cerebral cortex in humas
Circuitry - cortex - motor output
Evidence -
When stimulating in temperal lobes patients described various presumed
experiences from their past.
It was later found that much confabulation was evident.
3. Woody - eyeblink (pgs 171- 175) -already discussed with respect to systems
that are associated with motivation or reinforcement i.e. stimulation of
lateral hypothalamic nucleus facilitates retention.
Paradigm-
UCS glabella tap
UCR - short latency EMG from appropriate facial muscle (obicularis
oculi)
CS -100 msec 60-70dB click
CR short latency >50msec EMG from appropriate facial muscle (obicularis
oculi)
Trace conditioning - US -CS separated by 20-400msec 4- 6 sessions/day
150 trials/session
10 sec btw tirals
Multiple sessions
criterion 80% CRs
Circuitry -
UCS- ??
CS - ventral auditory nucleus (VII), Suprasylvian cortex (sensory ctx),
pericruciate cortex 'motor ctx' - facial nucleus
Evidence for storage in Neocortex:
Changes in excitability of cortical neurons paralleled the aquisition
of the CS.
Lesions of the rostral cortex (motor) suppressed the CR but not the
UR.
F. Hippocampus: after break
G. Prefrontal cortex: after break