Consolidation & Forgetting (Week 4)

Question 1

Forgetting

Answer 1

Forget lots of material.
- not a failure to encode- this means we never internalised info in the first place.
Forgetting = inability to recall something that could’ve been recalled at another time.
- must have formed memory to forget it.

Question 2

Inaccessibility

Answer 2

Sometimes we need a cue in order to retrieve memory.
- in absence of the cue, we have issue of inaccessibility.
Impossible to disprove the inaccessibility account.

Question 3

Is forgetting adaptive?

Answer 3

Believed that we need to forget something in order to generalise across various contexts.
Idea is opposed by individuals with autobiographical memories.

Question 4

Rate of forgetting

Answer 4

Ebbinhaus’s (1885) nonsense syllable experiment:

• over course of days, kept relearning lists.
• forgetting was measured as percent savings (comparison of immediate testing vs. testing after delay).
• info is rapidly forgotten at first- then slows down.(non-linear function).

Question 5

Explanations of forgetting

Answer 5

Trace decay: memories just fade over time.
Interference: memories (particularly if similar) get confused.
Cue-dependent forgetting: issue of accessibility and not availability.
Consolidation: new memories are fragile.

Question 6

Trace decay

Answer 6

eg. image of old photo fades.
Model of working memory (B & H, 74): within phonological loop.
Without rehearsal the phonological info fades from the store.
- word length effect (B etal, 75): more time elapses for longer words before rehearsal; therefore longer words should be forgotten to a greater extent.

Brown-Peterson (1958): trigrams.

• required to count backwards by threes.
• 0 sec delay = 90% correct recall.
• 18 sec delay= 7%.
• retention duration= approx. 20 secs.

Question 7

Problems with decay

Answer 7

Often times you have competing interpretations with interference.
P & P (1959)- shows more forgetting with delay of 18 seconds.
- but the first trial comparison the forgetting is small; and over many trials it increases.
- all of the trigrams on different trials begin to interfere with the new ones you’re trying to learn/
- explaining the difference between first trial comparison and many.

Jenkins & Dallenbach (1924): suggests time is not the only important variable.
- information isn’t lost at the same rate- if sleeping you have better recall later on than if awake.

Question 8

Interference

Answer 8

Ability to remember current info is disrupted by previous learning and future learning.
- the more similar the memories he greater the confusion.
Proactive interference: old info interferes with learning new info (eg. P & P, 59).
Retroactive: new info interferes with retention of old info

Question 9

Evidence for RI

Answer 9

Waugh & Norman (1965) probe digit task:
- ppts listening to 16 digits and last digit tells you which to report- (eg. 5 is at the end, so go to last occurrence of 5 and report the number after it).
Performance was the same with 1 number per second and 4 numbers per second.
Performance was different depending on number of intervening digits- worse for more.

Question 10

Evidence for PI

Answer 10

Wickens et al (1976):

• ppts listened to three words, counted back for 15 secs, recalled words.
• four trials, different words on each trial.
• words from a new category caused a release from PI.

Question 11

Interference problems

Answer 11

Does not explain why rate of forgetting (Ebbinghaus) slows over time.
- as even more interference should result in accelerated forgetting.
Overlaps with other accounts:
- time-based decay.
- cue-dependency.

Question 12

Cue-dependent forgetting

Answer 12

Search engine analogy: web page might exist somewhere but in order to find it you need the write search terms (cue).

Failure to retrieve as appropriate cue has not been presented- memory exists but issues in accessing it.
Meeter et al (2005): examined forgetting curve for newsworthy events:
- performance better for multiple choice questions (52%) compared to free recall (31%).

Tulvig & Pearlstone (1966): found ppts recalled twice as many words when given a cue compared to free recall.

• but performance not 100% so other explanations needed.
• perhaps appropriate cue not yet been presented; hence why cue-dependent forgetting is non-falsifiable hypothesis (not a good thing).

Question 13

Cue-dependent forgetting: context (or state) dependent memory studies

Answer 13

Memory is better when context at learning and retrieval is the same- context=cue.

• Diving (Godden & Baddeley, 75).
• Alcohol (Goodwin et al, 69).

Question 14

Cues as hooks

Answer 14

More cues means more hooks= easier to snag.
Cue overload and interference- a single cue that applies to many memories is less effective than one that applies to a unique memory (Watkins, 78).

Question 15

Consolidation- theory to propose memories become less fragile over time (could explain shape of Ebbinghaus curve)

Answer 15

Two types:

• synaptic: changes in connectivity between neurons (occurs rapidly).
• systemic: memories become independent from the hippocampus and move to the surrounding cortices (more long-term).

Question 16

Information storage at the synapse

Answer 16

Hebb (1948): learning and memory represented in the brain by physiological changes at the synapse.
- neural record of experience- cells that fire together wire together.
Long term potentiation (LTP): enhanced firing of neurons after repeated stimulation.
- structural changes and enhanced responding- neurons in hippocampus, amygdala and cortex show long term enhances response (potentiation) after a learning episode.

Question 17

Researching the mind-memory consolidation

Answer 17

Memory for recent events is fragile.
If disrupted, recent memories can fail to be consolidated.
New info can interfere with memory consolidation.
Protein synthesis inhibitors can disrupt memory consolidation.
Sleep can aid the consolidation process.

Question 18

Memory consolidation: behaviour approach

Answer 18

Muller & Pilzecker (1900) had ppts learn two lists of words:

• IV= one group learned second list immediately after first; other group had six minute delay.
• DV= recall for first list of words.
• found with no delay recall was 28% and with 6 min delay recall was 48%.

Gais et al (2007) the effect of sleep.

• one group learned a list of words shortly before going to sleep,; other group many hours before going to sleep.
• DV= memory for list of words measured two days later.
• found group that slept straight after learning forget less (recalled more)- no interference.

Question 19

Memory consolidation: physiological approach

Answer 19

Gais et al (2007) effect of sleep on memory consolidation.

• measured brain activity using fMRI.
• found differential brain activity between two groups during retrieval.

Question 20

Synaptic consolidation

Answer 20

Periods of acute anterograde amnesia assist memories formed just prior to amnesia
Suggest new memories interfere with the synaptic consolidation process.

Question 21

Anterograde amnesia = inability to form new memories after incident

Answer 21

eg. HM- surgery to relieve severe epilepsy.
- lesion of medial temporal lobe.
- severe anterograde amnesia but some learning remained: improvement on mirror tracing task despite no episode recall of learning task or memory for experimenters (Corkin, 68).

Traditional understanding of AA using multistore model (A & S, 68).
- cannot encode from STM to LTM.

Question 22

AA re-explained

Answer 22

But some amnesic patients capable of long term retention when interference is minimised.
Cowan et al (2004): 6 densely amnesic patients given a list of 15 words to remember.
Recall test was (1) immediately after, (2) following unfilled interval, (3) following filled interval.
- 4/6 patients had better memory following unfilled 10 min interval.
Seem to benefit from minimal interference- so are they just highly susceptible to interference, or are they rehearsing in STM?

Rehearsal interpretation examined by Dewar et al, 2009:

• if ppts rehearsing in STM there should be no difference between early and late interference (as both would abolish rehearsal).
• performance was better with no interference but also better when interference in last 3 minutes.
• inconsistent with absence of LTM.

Question 23

New cognitive model of forgetting (Dewar et al, 2010)

Answer 23

Essentially a set of synaptic and systemic consolidation that bridges the gap between STM and LTM.
This particular area is susceptible to interference.
In terms of anterograde amnesia- this is what is depicted to be the cause; explained by a depletion of consolidation resources.

This model can explain why some get anterograde amnesia following TBI- can remember what happened straight away after injury but not after a while.
- failure to consolidate memory.

Question 24

Systemic consolidation

Answer 24

Memories are purportedly moved out of hippocampus into surrounding cortices:
- longer term process- months to years.

Question 25

Retrograde amnesia = inability to remember memories prior to incident

Answer 25

Methodological issue: how do you know what individuals knew prior to injury?
- solution use well-known events or ask family members.

Patient PZ had just completed memoirs prior to amnesia: good source (Zola-Morgan et al, 83):
- older memories more resilient than recent memories- for both semantic and episodic memories (Mann et al, 03).

Question 26

RA: systemic consolidation

Answer 26

Temporally graded amnesia (greater loss for more recent memories) is consistent with systemic consolidation.
Assumes that hippocampus and associated areas operate as intermediate storage structure.
Older memories get moved into the neocortex (outer later of brain):
- therefore, damage to the hippocampus (and/or associated areas or connections) affects new memories.
- older memories survive as not hippocampal- dependent.

Question 27

Summary

Answer 27

AA: can be explained by synaptic consolidation being disrupted- but perhaps only applicable to declarative memory.
RA: effects (older memories surviving) can be explained by systemic consolidation.