Memory

• Interference


  When 2 pieces of information conflict with each other so then one or both are forgotten.
• Interference is more likely to occur when

  two pieces of information are similar/response competition- you will forget revising French then Spanish than you would English then Maths.
• Retro(old)active Interference


  New information interferes with previously learnt information

  Old information forgotten
• Proactive interference


  Past memories prevent new ones being stored in the LTM.

  New information forgotten (memory of an old phone number disrupting a new one)
• 😊Thousands of studies on interference


  Highly controlled expts giving a consistent evidence base that interference plays a role in forgetting. e.g Underwood had ps learn word lists & be tested 24 hours later. On early lists recall was 80% however later lists fell to 20%. This supports proactive interference as old information they had learnt was interfering with the new information being stored.
• ☹️Issues w/lab based research interference

  Tasks are artificial so no real motivation for remembering, conditions for interfering are unique. They were also often independent groups w/ small time gaps between conditions- real life memory tasks??

  Therefore interference may not occur to the same extent in more real-life settings and scenarios, so challenging interference as an explanation for forgetting.
• 😊Real life evidence for interference

  There is evidence of this type of forgetting taking place with a real life stimulus. This potentially overcomes the issue of lacking motivation to remember due to lack of relevance to real life because it uses a real life situation. Learning patterns of street names was easier with meaning/emotion attached, new info confused with old info after moving house. Suggests there is retrospective interference in real life situations.
• The loss of info may only be temporary (interference)


  Retrieval cues can be used to overcome interference therefore forgetting is due to a lack of cues. e.g Tulvig and Psoltka found when p's were given cued recall the effects of interference disappeared. P's remembered 70% of words regardless of how many lists they were given, info is there they just can't retrieve. Suggests retrieval failure is a more important explanation of forgetting.
• A leading question is


  a question that will imply something
• Post-event discussion is


  A discussion that happens after the event and can alter the memory of it.
• Conformity effect


  This term involves co-witnesses reaching a consensus view- this view may not be based on factual information.
• Loftus & Palmer


  Investigate how info in the form of a question can influence a witness' memory of an event. An opportunity sample of 45 American students took part in a lab study exp which there were 5 conditions. They were shown a video involving cars & were then asked to describe the event they had witnessed. "How fast was the car travelling when it... with the other car?"
• Loftus & Palmer findings

  How the question was phrased influenced the p's speed estimates.

  When the verb "smashed" was used, p's estimated the cars were travelling much faster than the verb in the "contacted" condition.
• Loftus & Palmer conclusions

  Response bias: different speed estimates occured because of the critical word.

  OR

  Memory is altered by critical word so they actually "see" the accident differently i.e more or less severe.
• Post event discussion research

  150 student p's shown a short film involving a car crash. They were split into 3 groups asked about the speed of the vehicles w/different verbs (except control group). One week later the p's returned & were asked about broken glass. Those in the "smashed" condition were more likely to think they saw broken glass and said a faster mph.
• Post event discussion research conclusion

  We are affected by leading q's, the way a q is worded can alter a event.
• Ecological validity of Loftus&Palmer's exps

  2 studies lack ecological validity- findings in the lab about the q's do not reflect real life situations surrounding a crash & speeds.

  In the lab they saw a video whereas irl there would be distractions.

  Felt normal & calm in the lab yet irl would be anxious.

  If they are wrong in lab no consequences, irl police questioning consequence.

  Research only tells us responses to a crash under unique circumstances.
• Ecological Validity of Loftus & Palmer


  Lacks population validity meaning it might not accurately be applied to a larger population.

  Used 150 students, some students might not be able to drive and understand speeds, a group of adults may be better.

  Therefore findings only tell us how students are influenced by leading q's around speed. Others might not be impacted by these if they are experienced drivers.
• 😊Researchin PED and false memories

  Gabbert et al. Procedure- p's watched a video of a girl stealing money from a wallet. Control group tested alone & experimental group in pairs. Co-witness group were told they had watched the same video however saw different perspectives of the same crime, only one witnessed the girl steal. P's in co-witness group discussed crime together.
• Gabbert et al findings


  71% of the witnesses in the co-witness group recalled info they had not actually seen.

  60% said that the girl was guilty despite the fact they hadn't seen her commit the crime. Even though they are asked alone they still didn't tell the truth, no peer pressure suggests they believe it.

  Conclusion- highlights the issue of PED & the powerful effect this can have on the accuracy of EWT.
• Evaluation of Gabbert et al.


  😊Different perspectives of same event- would happen in a real crime

  ☹️P's knew it was about memory so paid more attention than someone watching a real crime would.
• Multistore model of memory (MSM) A01

  1. info enters via senses
  2. SM/register - decay/displaced
  3. attention
  4. STM - maintenence rehearsal, decay/displaced
  5. elaborative rehearsel - LTM
  6. retrieval LTM-STM
  7. Absence of cues/interference
• MSM A01


  Coding- the form in which information is stored

  Capacity- how much information can be stored

  Duration- how long the info can be stored for
• Sensory store


  CODING- recieves information from our senses.

  DURATION- Information doesn't last long before passing into STM

  CAPACITY- lots of information enters this store, we only remember if we pay attention.
• Coding- how we process the info before changing it into a suitable form


  STM encoding is acoustic. In STM tasks people performed worse on acoustically similar words as STM is confused. The 2 stores code differently suggesting they are different stores.
• Capacity- amount of info which can be held


  STM capacity is 5-9 items, if you chunk together items you can remember more. P's had to repeat a string of numbers in the same order which increased until they could no longer recall them. If the capacity of the ST and LTM is different it suggests they are different stores.
• Duration- how long we can remember a piece of info

  STM duration is about 18 secs but rehearsing will make it last longer. P's had to remember trigrams (strings of letters) then count back in 3's to prevent rehearsal. Memory lasted no longer than 18secs. If LTM lasts 18 secs and LTM lasts a lifetime suggests they are different stores.
• LTM Coding


  Information in LTM coded semantically. Baddely's research supports as the list with semantically similar words were least effectively recalled after a 20 minute wait.
• LTM Capacity


  Unlimited, we can often recall more than we think we are able to.
• LTM Duration


  Potentially a lifetime. Bahrick asked graduates 50 years later to recall their classmates. Recognition group-for each photo p's were given a group of names and asked to match it w/the photo.

  Recall group- p's asked to name the person in the photo.

  Recognition condition was most accurate as 60% accuracy after 47years. Info in LTM can be returned to STM through retrieval.
• WMM


  Proposed by Baddely & Hitch to account for some of the limitations of MSM. Felt STM consists of multiple stores not just one unitary store and STM is an active process.
• Central executive


  Allocates tasks to the memory systems.

  Can process any type of info, limited capacity.

  Allows us to swap between tasks and focus on the most demanding.
• Phonological loop


  Capacity believed to be 2 seconds of what you would want to say.

  Deals with auditory sound & info and helps us preserve order of info.

  The phonological store codes our perception of sounds.

  The articulatory loop codes words we have heard or seen, called sub-vocal repetition.
• Visuo-spatial sketchpad


  Visual Cache- codes information about form and colour

  Inner scribe-arrangement of spatial information/memory of visual fields.
• Episodic buffer


  The "convenience store", holds and merges the info from- LTM,CE Visual and Audio.

  Allows us to work out how things relate- recording events.

  Due to temporary storage the CE would not be able to do this well.
• 😊WMM- Left prefrontal cortex active during CE tasks


  😊Real life evidence from brain damaged patients. KF w/ damaged STM was poorest at auditory tasks shows brain damage restricted his phonological loop not visual spatial sketchpad. Brain scans show different areas of the brain are involved in processing different types of STM information.
• ☹️WMM individuals such as KF cannot suppport a whole model

  We don't know if KF had issues w/attention before the injury or if something is causing underperformance in certain tasks. Weakness as we cannot replicate such studies on others therefore difficult to generalise findings.
• ☹️Lack of clarity of Central Executive

  Role is too vague, unsure of capacity, duration etc. More about attention than the actual store like the other three.

  Eslinger & Damasio- patient performed well on some tasks requiring reasoning, had poor decision making skills and spent hours on simple decisions. Idea of a single CE, must have several components.

  Difficult to design research/isolate the variables to test it alone.
• 😊WMM Baddely & Hitch dual task study


  P's asked to perform 2 tasks at a time involving repeating numbers & answering true or false questions. As digits increased p's slowed down but didn't make errors in verbal reasoning. Verbal task used CE and digits used phonological loop. Shows a store is not being overloaded.
• 😊WMM has real world applications

  Phonological loops role in reading allows for measures and understanding for dyslexia as Pl becomes overloaded. Model has allowed developement of effective teaching practises/children access education.