False Memory (Week 5) Flashcards
What is it?
Roediger & McDermott, 1995: “either remembering events that never happened, or remembering them quite differently from the way they happened”
Roediger & McDermott (1995): experiment 1 (DRM paradigm)
Examined false memory for related prototype words:
- 6 lists of 12 words were presented auditorially (1 word per 1.5 sec)- followed by a free recall of each list.
- then presented list of wordsl ppts rated confidence that each word was in the list (1-4).
- lists were composed of 12 related words linked by omitted prototype word (examining recall of unstudied ‘critical lure’).
Free recall produced typical serial position curve
- recall of studied words= 0.6.
- recall of unrelated (non-studies) words= 0.14.
- recall of critical (non-studied word)= 0.4 (critical lure).
- false memory effect not due to guess- 58% of ppts used highest confidence rating for non studied critical lure; 80% of non studied unrelated words received lowest confidence rating.
R & MD 95: experiment 2
Explored meta-cognition with the ‘remember/know’ paradigm.
- after recognition judgement, ppts state if they can mentally relive the experience (R=remember) or are confident that the item occurred in the list but are unable to relieve the experience (K=know/familiarity).
Studied 16 lists each containing 15 words (each list linked to prototype): recalled half the lists post presentation and completed a mental arithmetic task after the other half.
K and R for all lists and unstudied items followed.
In recall phase, ppts recalled 55% of lures (unstudied prototype words).
For K and R task, responses for lure closely matched the studied items- lures that were initially recalled were recognised at a higher rate (0.65) than studied items that were not initially recalled (0.5)- and ppts mostly said they R rather than K for the false memory lures.
- when incorrectly remember something, it strengthens that memory for incorrect false memory.
R & MD: explanations
As DRM study list is presented a ‘gist’ (general meaning) is developed.
- subsequent accurate recall depends upon item-specific memory and ‘gist’ (using ‘gist’ as a bias to say if word was old or knew).
- whereas false memory relies upon ‘gist’ but not item-specific meaning.
Accurate DRM performance necessitates distinguishing between ‘gist’ and item specific recollection.
Associative networks: seeing a particular word activates related words.
Importance of FMs
We often assume our memories are accurate- but our memory is not a retelling of what happened; we often build our memories when we recall them- often filling in blanks with what seems likely to go there.
- we change our memories using this process.
Memory malleability
Loftus (1979): subjects viewed slides of accident with either stop or yield sign.
- asked “did another car pass the car while it stopped at the stop/yield sign?”
- consistent condition = 75% accurate.
- inconsistent condition = 41% accurate.
Loftus & Palmer (1974): false memory from witness can result in a conviction. Ppts saw video of accident then estimated speed of impact:
- contacted = 31.8mph.
- hit = 34mpg.
- bumped = 39.3mph.
- smashed = 40.8mph.
- no broken glass in footage but those who heard ‘smashed’, 32% said they saw glass.
Loftus & Pickrell (1995)- can you get people to have memories of events that never happened?
Interviewed relatives to generate 3 true stories and on false about being lost while shopping as a child.
Given stories and asked to recall what they could of events.
- 68% recall of true events.
- 29% of false.
- true memories were rated as more clear and more words were used to describe them in recall.
False memory: alien abduction?
Reports of alien abduction “seem unlikely to have occurred?” (Clancy et al, 02).
DRM false memory task applied to different alien abduction groups:
- memory of abduction (recovered).
- believed they were abducted but no autobiographical memory (repressed).
- deny being abducted (control).
Those with recovered and repressed memories reported more false recall and false recognition that the control group- despite no difference in overall memory performance.
- may be a problem with source monitoring- those highly susceptible to false memories have a hard time distinguishing where the memories came from.
Applied importance of FM: forensic settings
80s and 90s spate of criminal trials premised on recovery of lost memories following therapy.
- not much research on FM at this point.
View that following trauma, unpleasant memories are repressed but can later be retrieved intact (eg. Tarr, 94).
- but might these memories arise from suggestive techniques during therapy and may in fact be false.
Recovered/FM debate
Memories of childhood abuse or trauma that are repressed and hen later recovered.
Self-help book, “The Courage to Heal (Bass & Davis, 88)” - states ‘if you have a pattern of symptoms such as low self-esteem, depression etc, the you were probably abused’
- but a problem- these symptoms also occur without abuse.
Some “recovered” memories for 1 or 2 years of age- hippocampus and frontal lobes are underdeveloped so no long term memories should exist of this age.
Power of suggestion
- interviewing witnesses.
- interviewing suspected criminals- admit to crimes they couldn’t possibly have committed.
- therapists- imagine childhood events to recover them; self help books: exercises (eg. who is the likely perpetrator? where would the abuse have been likely to occur?).
Can you differentiate between true and false memories by physiological responses?
Baioui et al (2012)- HR, skin conductance, respiration etc.
“a stimulus that someone has already encountered bears higher significance than a completely unknown stimulus…”
Methodology: Pictorial DRM task.
- semantically related component removed from image.
- 20+ min retention interval (completed personality questionnaire).
- critical lure included.
Found: higher recognition for related lures compared to unrelated (non studied) items.
- stronger physiological effects for the true false memories (lures); some awareness that the ‘lure’ is false.
Cabeza et al (2001) used fMRI to test neural mechanisms of FM
Medial temporal lobe provides different types of messages when stimuli are presented in a false memory paradigm.
Anterior hippocampus suggests that false items are similar to true (old) items- correct in respect to semantic info.
Posterior parahippocampal activity suggests that false items are similar to new items- correct in respect to sensory info.
Summary
In recognition and recall, FMs can be induced using semantically related ‘foils’.
Not just increased familiarity: ppts believe that the item was presented (K & R responses).
False memories of real life events can be induced- potentially devastating impact (legal context highlights importance).
Some indication that brain activity differs for true and false memories- mixed messages in MTL depending on type pf info about an item.
