Memory Case Studies

Miller (1956)

Procedure: Pps read a list of words and were then asked to repeat them in the same order.

Findings: Most people have a digit span of 7 + or - 2 items.

Not representative of real life (lacks ecological validity), pps might not perform well under pressure.

Peterson and Peterson (1959)

Procedure: Participants were presented with trigrams that had sets of nonsense syllables (e.g. BFM, CTG) and were asked to recall after 3, 6, 9, 12, 15 and 18 seconds. Pps were given an inteference task where they had to count backwards in 3's from a large number.

Findings: 3 seconds - 80% recalled, 6 seconds - 50% recalled, 18 seconds - 10% recalled.

Lacks mundane realism, biased results, trigrams are not meaningful information, pps under pressure.

Conrad (1964)

Procedure: Showed pps a random sequence of 6 constonants. One group got acoustically similar letters and another group got acoustically dissimilar letters. They had to write these down in the correct order straight after.

Findings: There were frequent errors in recall, and pps found it more difficult to recall similar sounding words; there is acoustic confusion.

Used artificial stimuli, students unrepresentative of the whole population.

Well controlled, no ethical issues.

Baddeley et al (1966)

Procedure: Gave pps either acoustically similar words, acoustically dissimilar words, semantically similar words or semantically dissimilar words. Pps had to write down the words 20 minutes later.

Findings: Similar sounding words were harder to remember. LTM encodes semantically.

Lab experiment, artificial, consent needed.

Used familiar words rather than constanents, no ethical issues.

Bahrick et al (1975)

Procedure: Pps (392 American ex-high school students ages 17-74) were asked to remember their classmates through either 1. Free recall, 2. A photo recognition test, 3. A name recognition test and 4. A name and photo matching test.

Findings: 90% accuracy in 4. After leaving high school 48 years ago this declined to 80% name recognition and 40% for face recognition. Free recall was far less accurate; 60% free recall after 15 years and only 30% accurate after 48 years.

Opportunity for rehearsal, results cannot be generalised into other information.

High mundane realism, high ecological validity.

Glanzer and Cunitz (1966)

Procedure: Presented a list of words to pps. One group had to recall immediately. A seperate group had to take part in a distractor task (counting backwards in 3's from a large number for 30 seconds).

Findings: The first groups resutls show an expected serial position curve. The second group didn't recall the words very well at all.

Artificial, lacks validity.

Highly controlled, easy to replicate.

Baddeley et al (1975)

Procedure: Gave pps visual presentations of words very briefly, then they were asked to write them down in serial order. Two conditions. 1. One syllable English words (harm, wit, twice) 2. Polysyllabic words (organisation, association, university).

Findings: First condition was easiest to recall.

Longer words are less familiar.

No ethical issues, well controlled.

Baddeley et al (1973)

Procedure: Gave pps a simple tracking task (following a spot of light with a pointer as it moved around a circular path) whilst carrying out a simultaneous visual imagery task (imagine looking at an angular block capital such as H, T, F and E). Pps were then asked to hold the image in their heads and then, starting at the bottom left hand corner, to respond to each angle as a yes if it included the bottom or top line of the letter and no if it didn't.

Findings: Pps found it very difficult to track the spot of light and accurately classify the corners.

Artificial, debriefing needed as some pps may have felt stressed.

Highly controlled, used repeated measures design to cancel out extraneous variables.

Poole and Lindsay (2001)

Procedure: Parents read their children (aged between 3 and 8) a story which contained elements of a science experiment they'd just done. They were then questionned about the experiment, and later asked to think very carefully about where they got that information from.

Findings: They had incorporated much of the new information into their original memory. Some older children revised their account of the science experiment  and extracted the post-event information. The younger children weren't able to do this.

Difficult to eliminate extraneous variables, artificial, consent needed.

Highly controlled variables.

• Gordon et al (2001) - (Reviewed child witness research) Young children can provide detailed and accurate statements but they are susceptible to suggestion and their accounts should be viewed with caution. 
• Yarmey (1984) - (Pps shown a staged event and then asked q's about it) 80% of the elderly compared to 20% of the younger adults failed to mention that the attacker had a knife in his hand.
• Cohen and Faulkner (1989) - (Shown a film of a kidnapping and were then read the narrative that was either consistent or inconsistent with the film) In a subsequent recall test, the elderly were found to be more susceptible to the effects of misleading information.

Loftus (1975)

Procedure: Pps were asked to sit outside a lab. In one condition, they pps could hear an equipment failure from inside the room followed by a man leaving the room with a pen. In the second condition, pps heard a hostile discussion, glass breaking, furniture overturning and then a man coming out of the room with a blood-stained knife. Later they were asked to identify the man that left the room from a set of photos.

Findings: Condition 1 - 49% accurate. Condition 2 - 33% accurate.

Could have caused psychological harm for participants in condition 2; debriefing is necessary.

High ecological validity.

Loftus and Zanni (1975)

Procedure: Showed pps a brief clip of a film of a car accident. Condition 1: Asked pps if they saw "the" broken headlight. Condition 2: Asked pps if they saw "a" broken headlight.

Results: Condition one - 17% said they did. Condition 2 - 7% said they did.

Artificial setting.

Real material used.

• Loftus (1975) - 17% reported seeing a barn when misleading information was used and only 3% of the control group made this error.
• Loftus and Palmer (1974) - Used different words and asked the speed of the car in the clip. Highest speeds - "smashed" and lowest speeds - "contacted"
• Schooler et al (1986) - Showed pps slides and asked pps not just to identify the slide they had seen but to give descriptive details. The more detail they give the more likely it is to be true.

• Fisher et al (1990) - Significantly increases the amount of information recalled.
• Milne and Bull (2002) - The procedures used (all four) singly produced more recall than the standard interview method. Most successful combination is Context Reinstatement and Report Everything.
• Holliday (2003) - More correct details being recalled.
• Kebbel et al (1999) - Used widely across the UK.

Time consuming, artificial studies (especially Fisher et al. 1990).

• Koehnken (1999) - Those questioneed using the cognitive interview recalled more incorrect information than those questioned using a standard technique.
• Geiselman (1999) - Children under the age of 6 years gave less a less accurate repot of events when using the cognitive interview.