Memory and learning

1
Memory and learning

• LeDoux, chapters 5 and 6

2
Types of memory

• There is general agreement that there are several
  different types of memory, each of which is
  predominantly in a different part of the brain.

3
Declarative vs. procedural memory

• Declarative memory (explicit memory)
• facts
• dates
• events
• Hippocampus is critical
• Procedural memory (non-declarative/implicit)
• how to perform an act (ride a bicycle)
• basal ganglia (dorsal striatum / caudate-putamen)
  is critical

4
(No Transcript)
5

• Patients with Alzheimer's disease are unable to
  learn or remember ordinary facts (declarative
  memory) but are normal or nearly normal at
  learning and remembering how to do things
  (procedural memory).

6
Memory experiment

• Alzheimer's patients learned and remembered how
  to read complex words in a mirror as well as
  normal control subjects
• Were unable to recall the training session or the
  fact that they had acquired this skill.

7
Classical (Pavlovian) conditioning

• Another kind of memory is distinct, both
  behaviorally and anatomically, from declarative
  or procedural memory.
• Pavlovian conditioning is a form of learning
  based on the tendency of certain natural events
  (food presentation) to elicit involuntary
  responses (salivation) with little or no training.

8

• Initiating event Unconditioned stimulus (US)
• Response pattern Unconditioned response (UR)
• Another "neutral" stimulus (ringing a bell),
  besides the US, does not usually elicit the UR.

9

• If another "neutral" stimulus (ringing a bell) is
  presented simultaneously several times with the
  Unconditioned stimulus (food), the "neutral" will
  be able to elicit the Unconditioned response
  (salivation).
• The "neutral" stimulus (ringing a bell) is then
  called the Conditioned stimulus (CS).
• The response (salivation) is then called the
  Conditioned response (CR)

10

• This process of learning an association between a
  CS and a CR is called Pavlovian or classical
  conditioning, or sometimes "associative
  learning".
• Pavlovian conditioning occurs automatically, with
  no control, voluntary participation, or (usually)
  even awareness on the part of the individual to
  whom it occurs.

11

• Evidence in animals and humans indicate that the
  amygdala is critical for classical conditioning.
• The studies indicate that the amygdala mediates
  expression of conditioned rewarding and approach
  behaviors as well as conditioned aversive and
  escape responses (e.g., "freezing" in mice).

12
Fear conditioning

• A simple form of associative learning (Pavlovian
  conditioning)
• Animals learn to "fear" a previously neutral
  stimulus (conditioned stimulus, CS), because the
  US has been presented at the same time as an
  aversive stimulus (unconditioned stimulus, US)
  such as a foot shock.
• Conditioned animals, when exposed to the CS, tend
  to refrain from all movement except breathing
  ("freezing").

13

• Freezing responses can be triggered with two
  different types of CS, each working via different
  parts of the brain
• - In "cued conditioning", the CS is simply a tone
  (e.g., 85 dB, 2800 Hz), and lesions in the
  amygdala, but not the hippocampus, appear to
  disrupt this type of conditioning.
• - In "contextual conditioning", rodents become
  conditioned to the "context" in which they were
  exposed, such as a particular location.
  Contextual conditioning is thought to depend on
  both the amygdala and the hippocampus.

14
Memory and the hippocampus

• In 1950, a young man, known now by his initials,
  H.M. underwent brain surgery in Hartford,
  Connecticut.
• H.M. was one of several patients in whom parts of
  the temporal lobe were removed in an effort to
  control epilepsy.

15

• The temporal lobe is one of the four major
  divisions (lobes) of the brain, and is often the
  place in the brain attacked by epilepsy.
• In H.Ms case, temporal lobe areas were removed
  on both sides of his brain.
• After the surgery, his epilepsy was better, but
  he no longer had the ability to acquire new
  memories.

16

• H.M became probably the most famous case in
  neurological history, and has been the subject of
  many studies.
• Much of the initial work was carried out by
  Brenda Milner and her colleagues in Montreal.

17

• Milner found that, although H.M could recall many
  of the events of his earlier life, he was unable
  to form new memories for experiences that
  occurred after the surgery.
• He could remember things for a few seconds
  (short-term memories) but he couldnt convert
  this information into long-term memories.

18

• Analysis of H.M.s lesion, based on the surgical
  report, indicated that the main temporal lobe
  areas affected were the hippocampus, amygdala,
  and parts of the surrounding cortex.
• By comparing H.M.s lesion with those in other
  patients, it seemed that the hippocampus was the
  area damaged most consistently in memory deficits.

19

• At first, it was thought that H.M. had lost all
  ability to acquire new memories.
• However, it was found that he could learn certain
  tasks.

20

• Brenda Milner asked H.M. to copy a picture of a
  star viewed through a mirror.
• To do this, the movements had to be done in the
  direction opposite from the way the seemed they
  should be made.
• This task is hard at first, but eventually most
  people can do it, and H.M. was no exception.

21

• H.M. learning the mirror drawing task, and he
  retained the learning.
• But if asked about drawing using the mirror, he
  had no conscious memory of having done it.

22

• Suzanne Corkin of MIT found that H.M. also
  improved with practice in another manual skill
  learning task one in which he was required to
  keep a stick held in his hand on a dot spinning
  on a turntable.
• As with the mirror drawing task, the more times
  he did it, the better he got.
• His ability to form memories about how to make
  precise movements (motor skills) seemed intact.
• Subsequent work has shown that other regions of
  the brain (the basal ganglia) are primarily
  responsible for remembering motor skills.

23

• Much is now known about the hippocampus, but we
  will mention just a few points.
• Information about the external world comes into
  the brain through sensory systems that relay
  signals to the cortex, where sensory
  representations of objects and events are created

24

• Outputs of each of the cortical sensory systems
  converge in parahippocampal region (also known as
  the rhinal cortical areas) which surrounds the
  hippocampus.
• The parahippocampal region integrates information
  from the different sensory modalities before
  sending it to the hippocampus proper.

25

• The hippocampus and parahippocampal region make
  up what is now called the medial temporal lobe
  memory system, which is involved in explicit or
  declarative memory.

26
(No Transcript)
27

• The connections between the hippocampus and the
  neocortex are all more or less reciprocal
• The pathways that take information from the
  neocortex to the rhinal areas and then into the
  hippocampus are mirrored by pathways going in the
  opposite direction.
• Cortical areas involved in processing a stimulus
  can thereby also participate in the long-term
  storage of memories of that stimulus.

28

• The rhinal areas serves as convergence zones,
  brain regions that integrate information across
  sensory modalities and create representations
  that are independent of the original modailty.
• As a result, sights, sounds, and smells can be
  put together in the form of a global memory of a
  situation.

29
(No Transcript)
30

• Convergence zones allow mental representations to
  go beyond perceptions to become conceptions.
• They make possible abstract representations that
  are independent of concrete stimulus.
• The primate neocortex has several cortical areas,
  more than in other mammals.

31

• Because the hippocampus receives inputs from
  several convergence zones in the rhinal region,
  it can be thought of as a superconvergence zone.
• It can form explicit memories about many
  domain-specific systems, such as face- and
  language-processing systems, allowing us, for
  example, to form a memory that includes both what
  someone says and what he looks like.

32

• Many researchers believe that explicit memories
  are stored in the corical systems that were
  involved in the initial processing of the
  stimulus, and that the hippocampus is needed to
  direct the storage process.

33
Early experiments on drug effects on memory

• Certain post-training treatments can modulate
  memory storage in ways that enhance or prevent
  retention.
• First observed with stimulant drugs
• strychnine (very low doses)
• amphetamine
• caffeine

34

• Early studies showed that drugs that inhibit
  protein synthesis also inhibit long-term memory
  formation.
• Several inhibitors of RNA synthesis or protein
  synthesis block long-term memory, but do not
  affect short-term memory.

35
Gene transcription, translation, and memory

• DNA is transcribed to produce RNA
• RNA is translated to produce protein
• DNA -gt RNA -gt protein
• Transcription factors are proteins that regulate
  what genes are transcribed (expressed).
• Transcription factors typically bind near the
  promoter region of a gene (the on/off switch).

36
Amphetamine improves learning

• Rats were put into a cage where they could drink
  water.
• After being put in the cage, the rats heard a
  series of 10 second tones, each terminated with a
  brief foot shock.
• The shock caused the animals to stop moving
  (freeze).
• After several such tone-shock pairings, the rats
  acquired a conditioned freezing response, which
  lasted for several minutes each time the tone was
  presented.

37

• The next day, the rats were placed in the
  drinking cage.
• Tone came on when they began to drink.
• Animals froze
• Duration of freezing was used as a measure of the
  rats' memory for the tone-freezing association.
• Rats that experiences more pairings (12) froze
  significantly longer than rats that had fewer
  pairings (2).

38

• Some rats got drug injections immediately after
  their experience of the tone-shock pairings.
• Rats that got two pairings followed immediately
  by a saline injection froze for slightly longer
  than rats that got two pairings but no injection.
• However, rats that got two pairings followed
  immediately by an amphetamine injection froze for
  about the same length of time as rats that got 12
  pairings (but no drug).
• gt Amphetamine improves learning if it is given
  immediately after training

39

• Another group of rats got 2 pairings followed by
  amphetamine injection 2 hours later.
• These rats froze for the same length of time as
  rates that received saline or no injection,
• gt amphetamine had no effect if it was given 2
  hours after the training

40

• The results indicate that the immediate
  amphetamine injections improved the rats' memory
  for the tone-freezing association.
• The fact that the delayed drug injection had no
  effect is consistent with the idea that the
  memory was susceptible to modulation only during
  a consolidation period that lasted less than two
  hours.

41
How drugs act on synapses

• Neurons communicate with each other at synapses
  using chemical neurotransmitters.
• This provides the bases for drugs (and poisons)
  to affect synaptic transmission.
• Drugs with chemical properties similar in some
  way to those of neurotransmitters can act on
  synapses to alter behavior and thoughts
  (psychotropic or psychoactive drugs)

42

• Drugs that increase synaptic transmission are
  "agonists".
• Drugs that block or reduce synaptic transmission
  are "antagonists".

43
(No Transcript)
44

• About 25 neurotransmitters are known in the
  mammalian brain.
• Most psychoactive drugs act on the synapses of a
  single neurotransmitter.
• These synapses often occur in different,
  functionally unrelated parts of the brain,
  controlling many different behaviors
• The psychological actions of drugs can be quite
  complex and difficult to predict

45
To affect the brain, drugs must cross the
blood-brain barrier

• Access to the brain from the circulatory system
  is controlled by the blood-brain barrier (BBB).
• This barrier is made up of a layer of cell
  surrounding the blood vessels that supply the
  brain.
• These cells determine the degree to which
  substances in the blood can enter the brain.

46

• Fat-soluble substances (e.g., alcohol) cross the
  BBB more easily than water soluble substances.
• Drugs and hormones with large molecular weights
  do not easily pass the BBB.
• Some substances, including glucose and insulin,
  are actively transported into the brain.
• The degree to which drugs cross the BBB is
  critical to their effects on memory.

47
Learning and memory

• What is the nature of the neural changes that
  constitute learning and memory?
• Most neuroscientists today believe that
  alterations in synaptic connectivity underlie
  learning, and that memory is the stabilization
  and maintenance of these changes over time.
• How does experience change synapses, and what
  makes changes last?

48

• Changes in synapses resulting from the
  simultaneous (or near simultaneous) activation of
  neurons is generally thought to be the basis of
  all learning, including procedural, declarative,
  and conditioned learning.
• We will see that the central role of synaptic
  changes in learning and memory provides the bases
  for the action of neurologic drugs.

49

• By the early 1950s, several studies had shown
  that repeated delivery of a brief electrical
  stimulus to a nerve pathway could alter synaptic
  transmission in that pathway could, in other
  words, produce synaptic plasticity.

50
Hebbian learning

• The most-widely accepted theory of how
  information is stored in the nervous system is
  based on a concept first described by D.O. Hebb,
  now called Hebbian learning.
• Start with the idea that each perception evokes a
  unique set pattern of neural activity.
• The set of activated neurons are connected to
  each other, and reactivate each other for a short
  period of time.

51

• Hebb suggested that this period of recurrent
  activation repeatedly activates the synapses
  connecting the neurons, causing the synapses to
  undergo permanent changes. These changes
  facilitate future activation of the synapses.
• The pattern of permanently facilitated synapses
  increases the probability that on future
  occasions activation of one part of some of the
  neurons will activate the rest of the neurons,
  leading to recall of the information it
  represents.

52
Hebbs proposal

• when an axon of cell A is near enough to excite
  cell B or repeatedly and consistently takes part
  in firing it, some growth process or metabolic
  changes take place in one or both cells such that
  As efficiency, as one of the cells firing B, is
  increased.

53
(No Transcript)
54
Memory consolidation

• When first acquired, memories are stored in a
  labile state (represented by Hebb's recurrent
  activation phase) and are subject to disruption
  by external events.
• With the passage of time their storage may become
  more permanent (Hebb's synaptic changes) and are
  less susceptible to disruption.
• This process by which memories become permanent
  is called "consolidation". The interval during
  which the hypothesized process of synaptic change
  occurs is called the consolidation period.

55
Long Term Potentiation A mechanism for memory

• In the mid-1960s Terje Lomo in Oslo noticed that
  a brief burst of electrical stimuli delivered to
  nerves going to the hippocampus in a rabbit led
  to a dramatic and long-lasting increase in
  transmission (a bigger electrical response to a
  test stimulus after, as compared to before, the
  burst) at synapses in the hippocampus.
• This is now called Long Term Potentiation (LTP)

56

• Later, Lomo and Bliss put a stimulating electrode
  in the nerve pathway going into the hippocampus
  and a recording electrode in the hippocampus
  itself.
• They delivered a single electrical stimulus to
  the pathway, and recorded the electrical response
  of the postsynaptic neurons.
• This served as the baseline, the standard against
  which the rest of the experiment was gauged.

57

• Next, they gave the potentiating stimulus a
  brief burst of many repeated pulses. Then they
  started testing again with a single pulse, and
  continued testing periodically for several hours.
• They found that, after the potentiating pulses,
  the synaptic response got bigger, relative to the
  baseline response, and remained bigger for hours.

58
(No Transcript)
59

• Although Bliss and Lomo had produced long-lasting
  change in the postsynaptic response by
  electrically stimulating neural pathways, rather
  than by having the animals actually learn
  something, they realized that they had identified
  a mechanism that might be able to translate
  neural activity generated by environmental
  stimuli into changes in synaptic efficiency.

60

• This mechanism might be used for learning and
  memory.
• The fact that the discovery was made in
  hippocampal tissue supported their speculation
  that changes in the efficiency of synaptic
  transmission might account for memory.
• LTP has since been induced in many areas of the
  nervous system.

61

• Computer simulations have demonstrated that
  information (memories) can be stored and recalled
  in a "neural network" in which the weights
  between "neurons" are altered as a result of
  learning.