Psychology - Memory

Explanations for forgetting

STM - duration, capacity, encoding

Duration - 18-30 seconds

Capacity - 5-9 bits

Encoding - Acoustically

LTM - duration, capacity, encoding

Duration - Unlimited

Capacity - Unlimited

Encoding - Semantically

Sensory Registers - duration, capacity, encoding

Duration - 1/4 second

Capacity - V. large

Encoding - Five senses

Jacobs (1887)

Aim – To investigate the capacity of STM

Procedure - Participants are made to recall a series of letters or numbers in serial order and to recall immediately after. They increase the digits each time.

Findings - It was found on average participants remembered 9 numbers and 7 letters.

Conclusion - STM has a capacity of 5-9 bits.

Jacobs (1887) A03

• High control, more valid

• Low mundane realism, less generalisable

• Low ecological validity, less generalisable

• Pioneering, useful

• Reliable, increases validity

Peterson & Peterson (1959)

Aim – To investigate the duration of STM

Procedure – Participants are given a trigram to remember then had to count back in 3s from a random number to prevent rehearsal. They has to recall the trigram after a certain amount of time (3s,9s etc).

Findings – After 3 seconds, 80% of trigrams were recalled. There was 50% recall after 6 seconds and less than 10% after 18 seconds.

Conclusion - STM has a duration of 18-30 secs.

Peterson & Peterson A03

• High control, high internal validity

• Low mundane realism, low external validity

• Low ecological validity, low external validity

• Reliable, increases validity

Bahrick (1975)

Aim – To investigate duration of LTM

Procedure – He asked participants, aged 17-74, to recall classmates using free recall, photo recognition, name recognition and name and photo matching.

Findings - After 34 years recall was 90% and after 48 it was 80%.

Conclusion - Duration of LTM is unlimited.

Bahrick A03

• High mundane realism, high external validity

• Low control, decreases internal validity

• Pioneering, useful

• Type of information, questionable

__Baddeley (1966)__

Aim – To investigate how memory is encoded

Procedure – Participants are given four sets of words (acoustically similar, acoustically dissimilar, semantically similar, semantically dissimilar) and asked to immediately recall them. Then they had to recall after 20 seconds.

Findings - More mistakes were made on acoustic words when using STM. Whilst using LTM, participants confused more semantic words.

Conclusion - STM encodes acoustically and LTM encodes semantically.

Baddeley A03

• High control, high internal validity

• Low mundane realism, low external validity

• Low ecological validity, low external validity

• Volunteer bias, not representative

Multi store Model

The multi store model begins with processing information from our surroundings. When information is detected by the five senses it is sent to the sensory registers which has a duration of ¼ of a second, but it has a large capacity. There are two main stores in the sensory register: iconic and echoic memory. If attention is paid, the information enters the short-term memory which has a capacity of 5-9 bits and is encoded acoustically. Information is lost after 18-30 seconds unless maintenance rehearsal is carried out and the information is able to be retrieved. If prolonged rehearsal is carried out, the information is passed to LTM which has unlimited capacity and duration and encoded semantically. LTM is able to retrieve information from the STM when it is needed.

MSM A03

• Pioneering, useful

• Supporting research, valid model

• Doesn’t account for flashbulb memories, limited in explaining memory store features

• Contradictory evidence, too simplistic

• Prolonged rehearsal isn’t needed, too simplistic

Case Studies - KF, HM, CW

KF had brain damage from motorcycle accident. He had no issue with LTM, but his STM was only 2 bits and he forgot auditory stimuli more than visual stimuli.

HM had brain surgery to cure epilepsy where his hippocampus was removed. He lost the ability to create new memories (anterograde amnesia). This showed he damaged his LTM, but his STM was still intact.

Clive Wearing had an infection that spread to his brain. He damaged his STM and parts of LTM. CW was able to recall how to play a piano but not his wedding day.

Working Memory Model

The Working Model Memory (WMM) was created by Baddeley and Hitch in response to criticism of the Multi Store Model of single unitary stores. People believed the STM had different parts, so the WMM showcases the STM. The model suggests that the working memory is an active store as it contains information that is currently thought about. When attention is paid to information, the information passes to the Central Executive, which holds overall control of the STM. The Central Executive is multi-modal as it deals with both stimuli but has a very limited capacity, so it passes information to one of two slave stores: Phonological Loop and Visuo-Spatial Sketchpad. Visual and Spatial information passes to the Visuo-Spatial Sketchpad which has two sub-stores. The visual cache stores visual information e.g., shape and colour. The second sub store, Inner Scribe, deals with the spatial relationship between objects e.g. distance. The Phonological Loop holds auditory information and passes it to its sub stores. The articulatory control system carries out maintenance rehearsal rehearsing the words in a loop for two seconds. The phonological store contains words you hear. In the 2000s, the Episodic Buffer was added to the WMM. The Central Executive and the Episodic Buffer pass each other information that the Episodic Buffer combines to make episodes/memories. This links to the Long-Term Memory.

WMM A03

• Supporting evidence from KF, valid model

• Support evidence from Dual Task Method, valid model

• Supporting evidence from PET scans, valid model

• Pioneering model, useful model

• Very little is known about the Central Executive, model is too simplistic

• Doesn’t account for musical memory, reduces validity

• Suggests spatial and visual information are linked to the VSS, too simplistic

Types of LTM - Episodic Memory

Ability to recall personal events (episodes) from our lives. These memories are complex and time-stamped, involving a conscious effort to recall the memories. Lots of information interwoven to produce a single memory.

Types of LTM - Procedural Memory

How to do things. These memories aren’t usually time stamped and we don’t have to consciously recall it.

Types of LTM - Semantic Memory

Shared knowledge of the world. These memories aren’t time stamped but we must consciously recall it. It is constantly being added to.

LTM Stores A03

• Clive Wearing support, valid theory

• Lack of variable control, findings are less valid

• Brain scans aren’t reliable, less valid

• Real life application, more useful

Interference

When one memory blocks another, causing one or both memories to be disrupted or forgotten.

Proactive Interference

When old memories disrupt the recall of new memories, causing new memories to be forgotten.

Retroactive Interference

When new memories disrupt the recall of old memories, causing old memories to be forgotten.

Interference Support - McGeoch and McDonald

Participants are made to learn a new list of words after being able to 100% accurately recall the first list. The new list ranged in similarity to the first list and then they had to recall the original list.

The more similar the words were to the original list, the worst the recall of the original list was. This shows us that interference causes more forgetting when words are similar.

McGeoch & McDonald A03

• Lab support, more valid

• high control, more valid

• Lacks ecological validity, less generalisable

Retrieval Failure

When the lack of cues means memory is available but not accessible.

Context-dependent forgetting

When memory is forgotten due to the lack of external cues e.g. environment

State-dependent forgetting

When memory is forgotten due to the lack of internal cues e.g. emotions

Encoding Specificity Principle

A helpful cue must be present at encoding and present at retrieval. If cues at encoding and retrieval are different, there will be some forgetting.

EWT - Misleading Information - Gabbert et al. (2003)

Aim: To investigate the effect of post event discussion on the accuracy of eyewitness testimony.

Procedure: 60 university students and 60 older adults watched a video of a girl stealing money from a wallet. They were either tested individually (control group) or in pairs (co-witness group). The pairs watched different videos but were told they’d watched the same one – one had actually watched the girl stealing. The pairs discussed the crime together. All participants completed a questionnaire on the event.

Findings: They found 71% of the witnesses in the co-witness group recalled information they hadn’t seen and 60% said the girl was guilty when they hadn’t seen her commit a crime.

Conclusion: Post-event discussions can change eye-witness testimonies.

EWT - Misleading Information - Loftus and Palmer

Aim – To investigate the effect of leading questions as misleading information on EWT.

Procedure – 45 students were shown 7 videos of a traffic accident, followed by a series of questions about “how fast were the cars going when they {smashed, collided, bumped, hit or contacted} each other?” The IV was the verb used and DV was the speed reported.

Findings – The more severe verbs resulted in higher speed estimations. Smashed had the highest speed of 40.8mph and contacted had the lowest speed of 31.8mph.

Conclusion – The harsher the verb used the higher the speed estimated, showing the leading question reduced the accuracy of EWT.

EWT - Anxiety - Johnson and Scott (1976)

Aim – To investigate if anxiety affects the accuracy of eyewitness testimony and facial recognition

Procedure – Participants sat in a reception area of a laboratory. The receptionist left the participants alone. They experienced one of two conditions:

1\.       ‘No weapon’ condition – They overheard a conversation about equipment failure then an individual (target) left the lab and walked past the participant holding a pen, hands covered in grease.

2\.       ‘Weapon’ condition – They overheard a heated exchange and the sound of breaking glass and crashing chairs, followed by the target running into the reception area, holding a bloodied letter opener.

They were then shown 50 photographs and asked to identify the target.

Findings – Those who saw the man holding a pen correctly identified the man 49% of the time and those who saw the man holding a knife correctly identified the man 33% of the time.

Conclusion – Those in the ‘weapon’ condition had higher anxiety so had lower accuracy.

EWT - Anxiety - Yuille and Cutshall (1986)

Aim – To investigate the effect of anxiety in a real life shooting.

Procedure – They interviewed 13 out of 21 witnesses, aged between 15 and 32, that the police originally interviewed, 4-5 months later.

Findings - They found those who took part in the follow up interview were accurate in their eye witness testimonies five months later. All of the major details stayed the same but minor details changed e.g. estimates of age, height and weight. The witnesses avoided leading questions and anxiety experienced at the time of the event had little or no effect on their memory of the event.

Conclusion – In real life cases of extreme anxiety, the accuracy of eyewitness testimony is not affected.

Cognitive Interview Techniques

Recall Everything – Witness recalls every single thing, no matter how minor they think the details are.

Context Reinstatement – Witnesses are asked to put themselves back into the environment in hopes of triggering a cue.

Recall in Reverse Order -  Witnesses start from a point of the event and go backwards in time.

Change in Perspective – Witnesses recall from another witnesses point of view.