Memory

• Studies on capacity and duration of STM (A01)

  JACOBS (1887): RESEARCH INTO CAPACITY! digit span technique - ppts were given 4 digits and had to recall them in order. This was repeated, adding 1 digit every time until they couldn't recall. Mean capacity was 7+/- 2. (9.3 for digits, 7.3 for letters)

  
  

  MILLER (1956): FURTHER RESEARCH INTO CAPACITY! Proposed capacity of STM was 7+/-2 but these could be chunks to increase capacity e.g phone number.
• Evaluation of studies into duration and capacity of short term memory

  Evaluation of JACOBS: Supported by Miller. Criticism - Digital span increased with age, decreasing validity as it may be partially due to participant variables.

  
  

  Evaluation of MILLER: A03: Supported by Jacobs. Criticism - COWAN's study suggests STM is limited to 4 chunks

  
  

  Evaluation of PETERSON AND PETERSON: Criticism - MARSH's study suggests duration was shorter when ppts weren't expecting a test of recall (suggesting demand characteristics - the please you effect)

  Criticism - unable to know if it was displacement or decay.
• Research into capacity and duration of long term memory (AO1)

  SHEPARD: Ppts were shown 612 memorable pictures. An hour later, almost 100% were recognised. 4 months later, 50% were recognised.

  
  

  BAHRICK: ppts aged 17-74, their real yearbooks used. Ppts who's graduation was <15 years ago had 90% photo recognition. 70% photo recognition 48 years after graduation.

  
  

  Capacity and duration of LTM is potentially unlimited
• Evaluation of research into capacity and duration of LTM (2) (A03)

  SHEPARD: High internal validity as all ppts were shown the same pictures. Low external validity (low mundane realism) as it's unrealistic.

  
  

  BAHRICK: Low internal validity (lots of extraneous variables) but high external validity, high mundane realism (real yearbooks, realistic task)
• STM and LTM coding

  BADDELEY: Tested the effects of acoustic and semantic similarity on STM and LTM recall.

  Gave ppts lists of words which were acoustically similar (cat, mat) or dissimilar (pit, few) and words that were semantically similar (great, large) or dissimilar (good, hot).

  Found ppts had difficulty remembering acoustically similar words when asked to recall them straight after (STM recall).

  Found ppts did worse recalling the semantically similar words after a time interval of 20 minutes (LTM recall).
• Multi Store Model

  
• Evaluation of the MSM (A03)

  SUPPORT: Case of HM (Scoville and Milner): HM had brain surgery and had parts of his hippocampus removed (crucial to memory). He could remember STM things but not new LTM. Supports the linear direction shown in the MSM.

  
  

  SUPPORT: Glanzer and Cunitz: PPpts were asked to recall 20 monosyllabic words after 30secs of counting back in threes. Results showed better recall of the first and last words in the list, and less good recall of the middle words. Suggests seperate stores, displacement and decay. (primary and recency)
• Working Memory Model

  
• Evaluation of the WMM (A03)

  STRENGTH: Baddeley et al 1975: when ppts carried out 2 tasks (one visual and one verbal) performance was good. When both tasks were visual (both using the VSS) tasks couldn't be completed well.

  
  

  STRENGTH: Case of KF (Shallice and Warrington): KF had amnesia. He could remember digits when he read them but not when he heard them. Suggests multiple STM stores.

  
  

  LIMITATION: The episodic buffer waws only added in 2000 so may be oversimplified. Baddeley himself said the central executive needs to be more specified, suggesting it is also oversimplified.

  
  

  LIMITATION: Eslinger and Damasio: Decision making and reasoning may be done in the central executive, which isn't mentioned in the WMM. Shows that the WMM may be too simplistic.
• Tulving 1985 (types of LTM) (A01)

  EPISODIC: For personal events. Includes memories of when an event occurred and of the other people, objects, places and behaviours.

  SEMANTIC: For our knowledge of the world. This includes facts and our knowledge of what words and concepts mean.

  EPISODIC and SEMANTIC are both DECLARATIVE, meaning they need to be consciously recalled

  PROCEDURAL: For our knowledge of how to do things. This includes our memories of learning new skills.

  PROCEDURAL is NON-DECLARATIVE, meaning these memories can be recalled without deliberate effort.
• Evaluation of Tulving's model (Types of LTM)

  Case studies: Clive Wearing and HM

  + Both could remember some long term memories but not all, and the things they could and couldn't remember fits into Tulving's different stores.

  However they are ungeneralisable as they are very individual, and are natural experiments so there are no records of the 'before'. They could have had PTSD - an EV.

  
  

  Opposing study: Cohen and Squire

  Argue that episodic and semantic memories are stored together in one LTM store called declarative memory. Decreases Tulving's validity

  
  

  Research: (positives)

  Scientific evidence - used a brainscan on ppts, increases validity and reliability

  Shows different bits of the brain doing different types of LTM.
• Explanations of forgetting: Interference (A01)


  Proactive interference: When older memories preventing the recall of new memories.

  
  

  Retroactive interference: New memories preventing the recall of old memories.

  
  

  (The one in the title is the one TRYING to be remembered!)

  
  

  MCGEOCH AND MCDONALD 1931: Investigating interference: Asked 6 groups to remember a list of 10 words. They then asked them to remember an additional list (the IV - different levels of similarity between lists). The group with the most similar list (synonyms) had the worst recall of the original words. Shows that interference is strongest when memories are similar. Could be retroactive interference or proactive interference.

  
  

  Muller and Pilzecker (1900): The first people to investigate Retroactive Interference. List of nonsense syllables to learn for 6 mins. Half ppts asked to describe 3 paintings before recall, other half just waited for same interval and then asked to recall. Less good recall if asked to describe.

  
  

  Underwood (1975): Looked at Proactive Interference. Analysed a number of studies, and found that ppts learn beginning of word lists better than ends! If ppts memorised 10+ word lists, recall after 24 hours was 20%, compared to 70% recall if only asked to learn 1 list.
• Evaluation of Interference as an explanation of forgetting (A03)

  Strength: McGeoch and McDonald's methodology - lab experiment so it's highly controlled and has high internal validity. The study displays evidence for both types of interference (HOWEVER NOT WHICH IS WHICH!) and evidence for the effects of similarity. We can establish cause and effect.

  
  

  LIMITATION: McGeoch and McDonald's methodology: lab experiment so artificial! Unrealistic - low mundane realism and low external validity - the conditions in the experiment are manipulated so that interference is more likely to happen, so it may not be generalisable to the real world.

  
  

  Support: BADDELEY AND HITCH - real world evidence! asked rugby players to remember the names of the teams they had played in a season - better recall if they had played less matches (regardless of how long ago!) which shows interference in the real world!

  
  

  LIMITATION: Tulving and Psotka - found that when words were organised into categories recall was ~70%. This fell when ppts were given an extra list to learn, but when they were reminded of the categories (cues) it rose back to 70%. Suggests that interference may only occur due to lack of availability rather than accessibility and can be overcome with cues - may not be the best explanation since it can be overcome!

  
  

  (retrieval failure as a better explanation! alt limitation)
• Leading question definition


  A question which, because of the way it's phrased, suggests a certain answer. e.g. 'was the knife in his left had' leads you to think it was.
• Post-event discussion definition


  Occurs when there is more than one witness to an event. Witnesses discuss what they have seen with co-witnesses or with other people. May influence the accuracy of each witness' recall of the event.
• Explanation of forgetting: Retrieval failure (A01)

  Encoding specificity principle: TULVING (1983) states that a functioning cue has to be present at encoding (learning) and retrieval.

  
  

  Non-meaningful cues are unconscious - there are 2 types!

  Context-dependent forgetting: recall depends on an external cue (e.g. weather or place)

  State-dependent forgetting: recall depends on internal cue (being drunk or upset)

  
  

  GODDEN AND BADDELEY (1975) researched context-dependent forgetting! Got deep-sea divers to learn a list of words in 4 conditions: learning on land or underwater, and then recalling on land or underwater. Accurate recall was 40% lower in non-matching conditions (learn on land, recall underwater).

  
  

  CARTER AND CASSADAY (1998) researched state-dependent forgetting! Ppts learned a list of words in 4 conditions: Learning on antihistamine (making them slightly drowsy) or not, and recalling on antihistamine or not. Recall was significantly worse when conditions of learning and recall didn't match.
• Evaluation of Retrieval Failure as an explanation of forgetting (A03)

  Strength: There is a range of supporting evidence, such as Godden and Baddeley, Carter and Cassaday. Shows retrieval failure occurs in real life situations as well as labs, increasing the validity (E and I).

  
  

  Limitation: It's hard to find contexts as different as land and sea/ on and off antihistamines. Real life contexts aren't THAT different so may not be so applicable to real life.

  
  

  Support: Real world application - context-related cues have been used in the Cognitive Interview for cognitive reinstatement. Increases the ecological validity and general external validity as it's useful in the real world.

  
  

  Limitation: Interference is another explanation of forgetting as a result of one memory blocking another (supported by evidence from McGeoch and McDonald) however interference only explains limited examples of forgetting.
• Factors effecting Eye Witness Testimony (EWT): Anxiety (A01)

  JOHNSON AND SCOTT (1976):

  Condition 1 LOW ANXIETY: Ppt was in a waiting room next to a room where two people were talking calmly, and one walked out holding a pen.

  Ppts had to find a picture of the person who walked through with a pen out of 50 photos. 49% accuracy.

  Condition 2 HIGH ANXIETY: Two people shouting at each other. Broken glass heard, then one walked out holding a knife with blood on them. Recalling person from 50 photos, 33% accuracy.

  Higher accuracy with lower anxiety.

  
  

  YUILLE AND CUTSHALL (1986): 13 witnesses of armed robbery in Canada gave accurate reports 4 months after the crime, even though they were given 2 leading questions.

  High anxiety 88% accuracy

  Low anxiety 75% accuracy

  
• Evaluation of factors effecting Eye Witness Testimony (EWT): Anxiety (A03) >>>

  Limitation: PICKEL (1998): walked past ppts in hair salon with scissors, hand gun, wallet or raw chicken. EWT significantly poorer in highly unusual conditions (handgun and chicken). Suggests that surprise might negatively effect memory rather than anxiety!

  
  

  Limitation: Y&C is real life compared to J&S' field experiment. Can't create same levels of anxiety in controlled conditions and demand characteristics. Memory may work differently in the real world which may explain the different findings.

  
  

  LIMITATION: Ethical issues - In Johnson and Scott's experiment they purposefully induce stress to ppts, which isn't protecting them from psychological harm!

  
  

  LIMITATION: FAZEY AND HARDY: suggested catastrophe theory which predicts that when anxiety (physiological arousal) increases past optimum levels there is a sharp decline, because of the increased mental anxiety (worry) whereas the inverted-U only looks at physiological harm. Therefore Catastrophe Theory is a better explanation.

  
• Factors effecting Eye Witness Testimony (EWT): Misleading information (A01)

  LOFTUS AND PALMER (1974): To investigate whether leading questions (misleading info) distorted the accuracy of an eyewitness' immediate recall. Ppts were shown film clips of car accidents and filled in a questionnaire straight after. Questionnaire contained critical question (the question that the researcher is actually interested in) that was leading, about how fast the cars were going when they hit each other. There were 5 different conditions, and each condition had a different verb in the critical question. Different verbs got a different mean speed, even though all ppts watched the same clip.

  CONTACTED: 31.8 mph

  HIT: 34.0 mph

  BUMPED: 38.1 mph

  COLLIDED: 39.3 mph

  SMASHED: 40.5 mph
• Evaluation of factors effecting Eye Witness Testimony (EWT): Misleading information (A03)

  SUPPORT: LOFTUS AND PICKRELL (1995): Ppts were presented with 3 true stories of them aged 4-6 and one made up story (getting lost in a shopping mall). Ppts were asked to write down all the details of the events they recalled straight after reading them, and in interview and a week later. After a week, they were then told one of them was fake and 5/24 ppts thought the 'lost in the mall' memory was true.

  
  

  SUPPORT: BODNER ET AL (2009): Effects of post-event discussion are reduced if ppts are warned of the effects. Police officers are also trained and warned about how to ask questions and interview witnesses.

  
  

  Methodology:

  High internal validity (high control).

  Low external validity, low ecological validity - anxiousness from an actual memory wouldn't be there

  Possibly unethical.

  
  

  LIMITATION: YUILLE AND CUTSHALL - 13 witnesses to an armed robbery in Canada gave accurate reports 4 months after the crime, even though they were given 2 leading questions.

  High external validity (ecological and mundane realism)

  Low internal validity as it's a natural experiment (lots of EVs and not much control.)

  Leading questions had no effect!
• Outline the Cognitive Interview(A01)

  FISHER AND GEISELMAN (1992): found people remember things better when given retrieval cues (Encoding specificity principle)

  Therefore they came up with the Cognitive Interview.
• Cognitive Interview Evaluation

  STRENGTH: Cognitive interview lowers context-dependent forgetting as it facilitates context-related cues. Therefore it has many useful applications in the real world, increasing the ecological validity.

  
  

  LIMITATION: Mello and Fisher compared age with accuracy of recall using the CI. When ppts were under 6 years old, CI resulted in less accurate recall than a structured interview.

  
  

  STRENGTH: Kohnken et al (1999) consistently showed ECI to have more accurate recall.

  
  

  LIMITATION: it is harder! It has no time pressure, a rapport needs to be created between police and witness, police need to be more self aware, and a peaceful environment has to be made.
• Studies on capacity and duration of STM (AO1)

  PETERSON AND PETERSON (1959): RESEARCH INTO DURATION! Ppts had to recall trigrams (3 letters with no meaning). To prevent rehearsal, they had to count back in 3s/4s from a 3-digit number. Trial was repeated with a longer retention interval each time. With a 3 second retention period, 80% trigrams were recalled. With an 18 second retention period, <10% trigrams were recalled correctly.

  STM capacity is limited to 7+/-2. As a result information can get displaced if new info enters STM.

  STM duration is limited up to 30s. Information can decay if not rehearsed.
• Explanations of Forgetting - Interference (A01)

  Muller and Pilzecker (1900): The first people to investigate Retroactive Interference. List of nonsense syllables to learn for 6 mins. Half ppts asked to describe 3 paintings before recall, other half just waited for same interval and then asked to recall. Less good recall if asked to describe.

  Underwood (1975): Looked at Proactive Interference. Analysed a number of studies, and found that ppts learn beginning of word lists better than ends!

  If ppts memorised 10+ word lists, recall after 24 hours was 20%, compared to 70% recall if only asked to learn 1 list.
• STM and LTM Coding

  A03: Criticism

  Low external validity (mundane realism) as word lists are unrealistic. Low internal validity as Baddeley didn't consider visual coding.

  FROST found that LTM recall was related to visual as well.

  BRANDIMOTE found that STM coded visually when verbal was prevented.
• Evaluation of the MSM (A03)

  LIMITATION: Case of KF (Shallice and Warrington): KF had amnesia. He could remember digits when he read them but not when he heard them. Suggests multiple STM stores.

  LIMITATION: Lots of studies supporting the MSM used digits or words for ppts to remember, rather than real life things like faces and places. Means many of them may have low mundane realism and lower external validity.
• Outline the Cognitive Interview (A01)

  FISHER ET AL (1987): Examined real interviews of experienced detectives in Florida over 4 months.

  Number of issues found: Witnesses bombarded with a series of brief, direct and close-ended questions aimed to elicit facts; the sequencing of these questions often seemed out of sync with the witnesses own mental representation of the event; witnesses were often interrupted and not allowed to talk about their experiences.
• Outline the Cognitive Interview (A01)

  1. CONTEXT REINSTATEMENT

  Witness returns to original crime scene in their mind. Imagine environment and their emotions at that time.

  This is related to context-dependent forgetting and cues.

  2. REPORT EVERYTHING

  Witness is encouraged to include every single detail of the event, even though it may seem irrelevant to the witness or the witness is unsure about the details. Trivial details may trigger other important memories.

  3. RECALL IN REVERSE ORDER

  Events should be recalled in a different order from the original sequence. To prevent people reporting their expectations of what must have happened rather than repeating the actual events. Prevents dishonesty.

  4. RECALL FROM A DIFFERENT PERSPECTIVE

  Witness should recall the incident from other people's perspectives to disrupt expectations of what happened and schema.
• Factors Affecting EWT - Misleading Information (A01)

  LOFTUS AND PALMER did a second experiment a week later, the police interviewed ppts again and asked 'did you see any broken glass'. Those who had heard the verb 'smashed' a week before, were over twice as likely to report seeing broken glass than those who heard 'hit'. Suggests that a week later the original memory had been modified.

  GABBERT ET AL (PED): Ppts were in pairs, each person watched a video of the same crime but one person in each pair saw from a view where they could see extra details. They then discussed what they had both seen, before recalling them.

  Findings: 71% if the ppts mistakenly recalled aspects of the event that they didn't see in the video but had picked up in the discussion. The corresponding figure in a control group was 0%.

  Evidence of memory conformity and memory contamination.
• Factors effecting EWT: Anxiety (A01)

  Deffenbacher: Reviewed 38 studies.

  Emotional and physiological arousal enhances performance to a point. Past this point, arousal has a negative effect.

  Fazey and Hardy: Show that the graph might have a steeper drop as anxiety increases, than shown in Yerkes-Dodson's graph.

  Weapon focus - focusing on the weapon rather than the situation around the weapon - decreasing EWT.