Forgetting in LTM

1
Forgetting in LTM

• Availability vs accessibility
• Interference
• Suggests that information forgotten from LTM has
  disappeared completely
• Cue dependent forgetting
• Suggests that forgotten information is still
  stored, but is (temporarily) inaccessible

www.psychlotron.org.uk
2
Interference

• Forgetting occurs when information to be stored
  is similar to information already in LTM
• Retroactive - new info overwrites previously
  stored info
• Proactive - previously stored info prevents new
  info from being stored properly
• Predicts that forgetting will increase with
  similarity of information

www.psychlotron.org.uk
3
Interference

• McGeoch MacDonald (1931)
• PPs had to learn lists of adjectives, recall
  after a delay. Three conditions
• Did nothing between learning recall
• Learned additional unrelated material
• Learned additional adjectives
• Most forgetting in group 3
• Supports prediction that forgetting is a function
  of similarity

www.psychlotron.org.uk
4
Interference

• Tulving (1966)
• PPs asked to free recall word lists they had
  previously learned
• Recall tested on several different occasions
• Generally, PPs recalled about 50 of the words,
  but not always the same 50
• Suggests that words had not disappeared but had
  actually been inaccessible
• This is contrary to what interference theory
  suggests

www.psychlotron.org.uk
5
Interference

• Clearly it is possible to confuse similar
  information
• Some experiments support interference theory, but
  they are very artificial
• Information that has been forgotten often becomes
  recoverable later
• Unlikely that interference accounts for most of
  the forgetting we do

www.psychlotron.org.uk
6
Cue Dependent Forgetting

• Forgetting occurs when information becomes
  inaccessible
• We lack the appropriate retrieval cues that will
  allow us to locate it in LTM
• Retrieval cues can be external (context) or
  internal (state)
• Predicts that remembering will be better when
  state context are the same as at the time of
  learning

www.psychlotron.org.uk
7
Cue Dependent Forgetting

• Smith (1970) tested recall of a word list in the
  original learning context or a different room
• Same room 18/80 words
• Different room 12/80 words
• PPs who imagined themselves back in original room
  recalled avg. 17/80
• Strong evidence for role of context cues in
  retrieval

www.psychlotron.org.uk
8
Cue Dependent Forgetting

• Fair amount of support for role of state cues in
  forgetting/remembering e.g
• Goodwin et al (1969) heavy drinkers often
  forgot where they had put things when sober, but
  remembered once they had drunk sufficient alcohol
• Eich (1980) similar findings with heavy marijuana
  users

www.psychlotron.org.uk
9
Cue Dependent Forgetting

• Much research support for basic propositions.
• Retrieval seems to be most likely when conditions
  match those of initial learning
• Does not apply equally to all types of info
• E.g. procedural memories (skills) seem stable,
  resistant to forgetting and not reliant on
  retrival cues

www.psychlotron.org.uk