Learning

1
Learning Memory
Kevin Silber k.silber_at_derby.ac.uk
2
Types of Memory Research

• Early research into learning and memory

• Learning in Aplysia

• Long-Term Potentiation

• The Search for the Engram

3
Early Research Into Memory

• Hailed by many as the founder of modern day
  physiological research into memory, Lashley
  carried out the first systematic studies.
• He believed memory to be a function of the
  cerebral cortex rather than of subcortical areas.
• He reasoned that if he lesioned various parts of
  the cerebral cortex he ought to find memory
  deficits.

Karl Spencer Lashley
4
(No Transcript)
5
Early Research Into Memory

• His experiments of the 1930s suggested two laws.
• The law of mass action stated that the degree of
  memory deficit was directly proportional to the
  amount of cortex removed.
• The law of equipotentiality stated that an equal
  amount of damage produced the same degree of
  memory loss irrespective of where in the cortex
  that damage occurred.

6
Early Research Into Memory

• After several years of researching he concluded
  that

it is not possible to demonstrate the isolated
location of a memory trace anywhere within the
nervous system. The complexity of the functions
involved in reproductive memory implies that
every instance of recall requires the activity of
literally millions of neurons.
7
Early Research Into Memory

• One of the first studies was conducted by Ungar
  and McConnell in the 1960s.

• They believed that memory was the result of the
  synthesis of new proteins.
• They trained flatworms called Planaria to avoid a
  light stimulus.

8
Early Research Into Memory

• They then chopped up the trained worms and fed
  them to untrained worms.
• They found that the naive, untrained worms took
  fewer trials to learn to avoid the light and
  concluded that the memory-containing proteins had
  been transferred to the naive worms

9
(No Transcript)
10
Learning Aplysia

• Kandel and co-workers have used a marine snail
  called Aplysia to investigate simple learning
  mechanisms.

11
Learning Aplysia

• Short-term and long-term habituation

• Short-term and long-term sensitisation

• Also, classical conditioning

12
(No Transcript)
13
(No Transcript)
14
Learning Aplysia

• Kandel et al. have investigated what the
  mechanism of this habituation might be.
• They discovered that the habituation takes place
  at the synapse between the sensory neuron coming
  from the siphon and the motor neuron feeding the
  gill.

15
Learning Aplysia
presynaptic mechanism
MOTOR NEURON CELL BODY
SENSORY NEURON
postsynaptic mechanism
16
Learning Aplysia

• In order to discover whether the mechanism was
  pre or post-synaptic they undertook a series of
  neurochemical experiments.
• They discovered that the mechanism was
  presynaptic and that it involved a decrease in
  the amount of neurotransmitter being released.
• They later found that this decreased release was
  due to a decrease in the amount of Ca2 ions
  entering the terminal bouton.
• You should be able to assess for yourselves why
  this decreases the amount of transmitter released.

17
(No Transcript)
18
Long-Term Potentiation

• The hippocampus is often implicated in memory
  research so it is no surprise that researchers
  have looked here for memory mechanisms.

19
Long-Term Potentiation

• Long term potentiation (LTP) was a phenomenon
  discovered in the 1970s.
• If you stimulate the pathway going into the
  hippocampus (the perforant path) with a tetanic
  stimulus then the EPSP recorded from hippocampal
  cells is potentiated when a further single pulse
  is presented.

20
Cross section through the hippocampus
Hippocampal neurons stained with cox golgi
method. (Schneider Laboratory, MIT Department of
Brain and Cognitive Sciences.)
21
Long-Term Potentiation
Initial response to stimulation of the perforant
path.
22
Long-Term Potentiation
Stimulation of the perforant path with a tetanus
stimulus.
23
Long-Term Potentiation
Potentiated response to stimulation of the
perforant path of the same intensity and duration
as the original stimulus.
24
4 Properties of Long-Term Potentiation
1) LTP may be rapidly induced The application of
1 or more high frequency stimulation (tetanic)
sequences may induce LTP.
2) Co-operativity LTP can be induced by intense
tetanic stimulation of a single pathway, or by
weaker simulation of multiple pathways.
25
4 Properties of Long-Term Potentiation
3) AssociativityWhen sub-thershold stimulation
of a single pathway is insufficient to induce
LTP, simultaneous strong stimulation of another
pathway will induce LTP at both pathways.
Cooperation and association in LTP likely share
the same underlying neurobiological mechanism
4) Localisation Once LTP occurs at a synapse, it
will remain localised to that synapse and will
not propagate to adjacent synapses.
26
Early Late Long-Term Potentiation
LTP is often divided into two phases
1) E-LTP is an early, protein synthesis-independen
t phase lasting between 1 and 5 hours.
2) L-LTP is a late, protein synthesis-dependent
phase lasting from days to months.
27
Long-Term Potentiation

• The EPSP is not only potentiated but this
  potentiation can last for a few weeks - hence
  long term potentiation.
• Lynch and Baudry examined a possible mechanism
  for this LTP and first suggested that previously
  unavailable postsynaptic receptors became
  available after this tetanic stimulation.
• However, further research has revealed that the
  mechanism is rather more complicated and involves
  the activity of different types of receptor on
  the postsynaptic membrane.

28
Long-Term Potentiation
29
(No Transcript)
30
(No Transcript)
31
Long-Term Potentiation

• A further discovery was the fact that whilst the
  induction of LTP appears to rely on postsynaptic
  mechanisms, the maintenance of LTP appears to be
  presynaptic.
• This means that there must be a mechanism for
  retrograde communication between the neurons such
  that the postsynaptic change gets translated into
  a presynaptic change.
• The precise mechanism of this communication is
  still not properly known.

32
The Search for the Engram

• What we have looked at so far are approaches
  which have tried to investigate the neuronal
  mechanisms involved in memory.
• These methods typically involve using tissue
  slices rather than looking at behaviour in whole
  animals.
• Thompsons group have investigated a classical
  conditioning paradigm in a live animal.
• They have looked at classical conditioning of the
  nictitating membrane response in the rabbit.

33
(No Transcript)
34
The Search for the Engram

• First, a word about the paradigm.
• The UCS is a puff of air to the cornea which
  elicits closure of the nictitating membrane
  (third eyelid).
• The CS is a tone
• Forward conditioning is used with a 350 ms CS and
  a 100 ms UCS which co-terminate.

CS
UCS
35
The Search for the Engram

• Thompson recorded multiple unit activity from the
  hippocampus and found that as the behavioural
  response started to appear during the CS-UCS
  interval (i.e. before the UCS was presented) so
  too did the hippocampal activity follow suit.
• Indeed, the hippocampal activity slightly
  preceded the behaviour suggesting that it might
  be driving the nictitating membrane response
  rather than responding to it.

36
The Search for the Engram

• Thompson decided to investigate whether or not
  the hippocampus was essential for acquisition of
  the response or for its retention.
• He removed the hippocampus of untrained rabbits
  and found that they still were able to learn the
  classically conditioned response.
• Aha, he thought, the hippocampus doesnt seem to
  be essential for acquisition so it must be
  necessary for retention.
• So he trained rabbits to produce conditioned
  responses and then he removed the hippocampus.

37
The Search for the Engram

• Much to his surprise, he found that the rabbits
  were still able to produce the conditioned
  response.
• It seemed that the hippocampus was going to the
  trouble of marking the learning that was taking
  place but did not seem to be necessary for it.
• Thompson reasoned that maybe classical
  conditioning, being a phenomenon found in lower
  animals than the rabbit, was controlled by a
  phylogenetically older structure than the
  hippocampus.

38
(No Transcript)
39
The Search for the Engram

• He turned his attentions to the cerebellum which
  is present in animals which have not evolved
  sufficiently to possess a hippocampus.
• The cerebellum responded in the same way as the
  hippocampus. However, this time he found that the
  cerebellum was, indeed, necessary for the
  acquisition of the conditioned response.
• Furthermore, the unconditioned response remained
  present, suggesting strongly the cerebellum was
  involved in memory and not merely the motor
  control of the nictitating membrane response.

40
The Search for the Engram

• Thompson concluded that the cerebellum was the
  seat of the learning of the nictitating membrane
  response in the rabbit.
• Why, then, was the hippocampus performing in the
  way that it was?
• Thompson reasoned that at the beginning of
  training an organism does not know what level of
  learning is going to be needed. Thus all
  structures involved in memory must actively
  respond to the situation.

41
The Search for the Engram

• If the learning remains simple then older
  structures are able to cope with the entire needs
  of the learning task.
• However, if the learning is more complex then
  more highly evolved mechanisms will be needed.
• Had the task Thompson used been more complicated
  than simple classical conditioning then one would
  presume the hippocampus to play a more essential
  role.

42
(No Transcript)