Reconstructive Memory

• What is the memory like?

  Imaginative reconstruction of past events, influenced by how we encode, store and retrieve info
• Schema

  Parcels of information based on our own personal experiences
• How do schemas affect memory?

  The schema alters the content of our memories in order to make it consistent with our beliefs/previous experiences
• What is retrieval of memory like?

  Involves an active process of reconstruction using a range of information
• Confabulation

  Info is added to fill in the gaps in a memory in order to make sense of a story
• Bransford and Johnson procedure

  Showed ppts a passage about doing laundry, but it did not explicitly show it was about laundry.
• Bransford and Johnson results

  1/3 knew it was about laundry before reading and made sense of it. 1/3 read it and were told afterwards it was about laundry, then they could make sense of it, 1/3 read without knowing it was about laundry and did not make sense of it. This shows the use of schemas
• The War of the Ghosts (Bartlett) procedure

  Showed 20 students the Native American ghost story with unusual features, he asked them to recall it on certain occasions after hours, days, weeks or months or years. This is called serial reproduction as is a repeated measures design
• The War of the Ghosts (Bartlett) results

  Ppts would shorten the story when it was reproduced, with a range of 330-180 words. Ppts confabulated details, changing the strange features to be normal, eg hunting seals turned to going fishing.

  Ppts rationalised the story, coming up with explanations for baffling parts of the story, eg ppts missed out the 'ghosts' and just described a battle between native American tribes
• Loftus and Palmer (1974) procedure

  • Students were shown film clips of real car crashes
  • they were given a questionnaire, asking about the speed of the cars. some were exposed to the word "smashed" others to "hit" eg. "how fast where the cars going when they smashed into each other"
  • A week later they were retested on whether or not they saw broken glass, there was no broken glass
• Loftus and Palmer (1974) results

  • Ppts exposed to 'smashed' verb recalled a higher speed (average of 40.8mph) whereas the average for ppts exposed to hit was 34mph
  • A week later 12% of the control group (not asked the speed) recalled glass, 32% of the smashed group recalled broken glass
• Why does memory not change when we experience something new?

  There are no existing schemas
• Weakness of Bartlett WOTG

  Not scientific, no standardised procedure, no scoring system for measuring the change in recall other than no words recalled, so it is subjective
• Barlett WOTG issue and solution

  Getting uni students to recall a ghost story lacks mundane realism, but Bartlett claimed the task had to be strange to get ppts to remember and sharpen details
• Allport and PostmanFindings

  Schemas distort memory, especially prejudice as we remember how things should be
• Schema application to dementia

  When memory is lost, the world may be scary and confusing, schemas can help make sense. By using old music, activities and reminiscing people can be made to feel comfortable
• Reconstructive memory was proposed by Bartlett in 1932
• Allport & Postman Study

  • showed ppts a drawing of an argument on a subway train
  • they were asked to describe it to another ppt through serial reproduction
  • in the image, the black character was better dressed and more respectable than the white character, but after serial reproduction this was reversed
  • some ppts even described the black character as holding a knife
• Loftus concludes that eyewitnesses are unreliable because they are influenced by leading questions. When we reconstruct memories, we change them by incorporating new information we learned after the incident. We also incorporate our schemas (expecting broken glass after a “smash” - this is sharpening the memory). We cannot tell which parts of a memory are original and which parts are later changes and there’s no way of going back to the original.