Cognitive Psychology Studies

Peterson and Peterson:

Tested the duration of Short Term Memory.

Gave undergraduate students a CONSONANT SYLLABLE IE VDW, then a 3 digit number, from which they had to count back from in 3's or 4's for a set amount of time ie 3, 6, 12 seconds.

This prevented rehearsal.

The students were then told to recount the syllable.

They found that 90% remembered it after 3s, but only 2% after 18s. 

They concluded that STM will last about 20s.

Bahrick et al 

Got participants between ages 18-66

Gave them their high school yearbook, told them to identify the people.

The 66 year olds were still 70% accurate in the recall of their peers' names.

Suggests LTM lasts many years.

Shepard 

Gave participants 612 memorable photographs.

After 2 hours there was almost perfect recognition.

After 5 months recognition was still 50% accurate. 

Baddely

Gave 4 conditions of participants 4 different lists of words. Accoustically similar and dissimilar, and Sematically similar and dissimilar.

He asked them to recall the list they were given.

They were fine remembering acoustically similar in short term memory but got them muddled in long term memory.

This means long term memory stores words accoustically because the accoustically similar words got mixed up and were undistiguishable.

Semantically similar words got jumbled in STM which means that encoding in STM is semantic 

Jacobs and Miller 

Jacobs invented the serial digit span, with which he tested participants' short term capacity

It would look like:

9 0 2 6 7 4 9 1 2

He asked participants to recall it immediately in the in the right order.

It would start with 3 digits and with every correct recollection got 1 digit longer.

Found that, on average 7 letters were remembered and 9 numbers.

Miller expanded on this and said that STM can hold 7+-2 letters.

Sperling:

Asked participants to look at a screen, a 4 x 4 table of number flashed up for 50 milliseconds:

4 5 6 0 recall was poor when asked to recall the whole thing, but much better

1 5 3 8 when asked to recall 1 line, this shows that information decays 

2 9 5 7 quickly in sensory memory.

9 2 7 1 <- Like so

Recall was poor when asked to recall the whole thing, but much better when asked to recall 1 line, that shows that. 

Glanzer and Cunitz

Showed participants a list of words for 20 seconds, then told them to write the list down immediately after.

They found that the participants always remembered the first and last few words in the list. They said this was because the first words had transferred into long term memory, and the last few words still remained in STM

This is called Primary and Recency Effect.

KF: Accident caused brain damage.

This resulted in him having poor function storing verbal information in STM but not visual, which implies STM is more than one store.

Schacter said there was more than one type of LTM: Semantic, Episodic, Procedural and Perceptual.

Craik and Lockhart got people in 3 different conditions, gave them a written word to look at, then asked a question to look at processing.

Condition 1: Was the word in CAPITAL LETTERS? (Shallow processing)

Condition 2: Did the word rhyme with train? (Phonetic, medium processing)

Condition 3: Was the word a type of fruit? (Deep processing)

P's answered condition 3 most correctly, this implies levels of processing are more important than rehearsal.

Baddely and Hitch

Gave participants two tasks to perform simultaneously, one occupied the central executive, the second occupied either the central executive or the articulatory loop, or both.

Found that participants were slower performing when both tasks involved the central executive, and were quickest when they used the two different stores.

This suggests that a single store finds it hard to perform 2 simultaneous tasks, but stores work interdependantly very well.

A different study found that when given short words vs long words, they remembered the short words more as the long words did not fit on the articulatory loop.

Scans found that when performing a dual task there was increased activity in the brain.

Christianson

Interviewed 58 witnesses to bank robberies and asked them to recall the robbery in as much detail as they could.

Found that those who were threatened had more accurate and detailed recall than those who had just been a bystander.

Concluded that a heightened anxiety can improve accuracy of EWT.

Deffenbacher

Meta-analysis of many anxiety related EWT studies. Found that those who were extremely aroused had impaired recall and it was far less detailed. 

Yerkes- Dodson came up with curve of arousal.

Loftus #1

Showed participants in 7 conditions 7 different videos of car collisions, then gave them a question to answer, the question varied between groups:

'How fast were the cars going when they smashed into each other?'

'How fast were the cars going when they bumped each other?'

They changed the underlined word per condition. Faster speeds were given when more violent words were used, slower speeds were given when gentler words were used.

Loftus #2

Showed participants the same video of a car crash in every condition, but they were still given the different questions. They were then asked a second question:

'Did you see any broken glass'

They found that those who had been given the more violently worded questions said that they'd seen glass, even though there was none. 

Loftus (Again)

Participants placed 2 separate conditions, both heard an argument from an adjoining room, then a man emerged, and that's when the conditions differed:

Condition 1: A man emerged holding a pen and his hands were covered in butter.

Condition 2: A man emerged holding a penknife and his hands were covered in blood.

The participants were then given photos and asked to identify the emerging man.

Participants in condition 2 were much less accurate at identifying the man, Loftus suggested that this was down to the participants automatically focussing on the weapon, rather than focussing on the face of the man.

Another study, which used trackers to follow the eye movements of participants showed that, in a simulated weapon crime, participants were looking at the weapon most of the time. 

Parker and Carranza

Used primary school students and college students.

Participants were shown slide photos of a crime, they were then given multiple photos then asked to pick the criminal from the slides from the deck.

They found that Primary School students were better at selecting a person than the college students, but the college students were better at correctly identifying the criminal than the primary school kids.

Yamey

Stopped 651 random people, to ask them to identify a man that they had spoken to for 15 seconds 2 minutes before.

He found no significant difference between young and old participants and correct selection of the man.

Anastasi and Rhodes

Had three age groups of participants aged 19-25, 35-45, 55-65.

They were shown 24 photographs of people, all of which they had to rate for attractiveness. The photos were then taken away and they were given a distraction task. 48 photos were then given back to the participants and they had to pick out the original 24.

It was found that the younger two groups were more accurate in their identification.

They also found that all 3 of the groups were far more accurate at identifying photographs of people of a similar age to them.