Psychology - Memory

Coding - the format in which information is stored in the various memory stores 

Capacity - the amount of information that can be held in a memory store

Duration - the length of time information can be held in memory

SHORT TERM MEMORY                                               LONG TERM MEMORY 

Coding - mainly acoustic                                                Coding - mainly semantic

Capacity - 5 to 9 items                                                   Capacity - unlimited 

Duration - 18 to 30 seconds                                           Duration - up to a lifetime 

Baddelely (1966A and 1966B)

Split participants into four groups:

Group 1 - acoustically similar (words sounded similar) e.g. cat, cab, can

Group 2 - accoustically dissimilar (words sounded different) e.g. pit, cow, few

Group 3 - semantically similar (words with similar meanings) e.g. great, large, big

Group 4 - semantically dissmiliar (words with different meanings) e.g. good, hot, small

• Participants were shown original words and asked to recall them in order. 
• When asked to immediately recall (STM) participants did worse with acoustically similar words. 
• When asked to recall after 20 minutes (LTM) particiapnts did worse with semantically similar words

Artificial stimuli- Baddeley used artificial stimuli, rather than meaningful material. The words had no personal meaning to participants, which means that we should be cautious about generalising the findings to different memory tasks. For example, when processing more meaningful material, semantic coding may even be used in STM tasks. This is a limitation because the findings have limited application. 

Jacobs (1887)

Researcher gives participant 4 digits. 

Participant is asked to recall in the correct order out loud. If they get it correct, the researcher will read out 5 digits and so on until they can't recall correctly. 

Found that the mean span for digits was 9.3 and for letters was 7.3

Miller (1956)

Noticed things come in sevens (7 notes on a musical scale, 7 days of the week). 

Suggests that the span/capacity of STM is about 7 items (+/-2).

Noted that people can recall 5 words as well as 5 letters. This is done by 'chunking' and grouping sets of words together. 

Lacking validity - Jacobs' study was conducted a long time ago abd early studies lacked adequate control. For example, some participants may have been distracted while they were being tested, which could've affected their performance. There also may have been confounding variables that were not controlled. However, the results that were found have been confirmed by other research, which supports its validity

Too many chunks - Miller may have overestimated the capacity of STM. For example, Cowan (2001) reviewed other research and concluded that the capacity of STM was about 4 chunks. This suggests that the lower end of Miller's estimate (5 items) is more appropriate than 7. 

Peterson and Peterson (1959) - STM 

• Tested 24 undergraduates. Each student took part in 8 trials. 
• On each trial the student was given a trigram to remember. They were then asked to count back from any three digit number to prevent rehearsal. 
• On each trial they were told to stop after a different amount of time - 3, 6, 9, 12, 15, 18 seconds (retention interval). The longer the retention interval, the less correct responses.

Bahrick (1975) - LTM

• Studied 392 participants ages 17 and 74.
• High school yearbooks were obtained from particioants. 
• Recall was tested by:
• Photo-recognition test consisting of 50 photos, some from participant's yearbook.
• Free recall test where participants recalled names from their graduating class.
• Photo recognition = 90% accurate after 15 years and 70% after 48 years. 
• Free recall = 60% accurate after 15 years and 30% after 48 years 

Meaningless stimuli - Peterson and Peterson's study used meaningless stimuli. Trying to memorise trigrams doesn't reflect most real life memory activities. This study lacks external validity, as it doesn't reflect everyday and real life examples of memory. However, we do try to remember fairly meaningless things, such as phone numbers, so it isnt completely irrelevant. 

Higher external validity - Bahrick's study has higher external validity, as real life memories were studied. Other studies of LTM were conducted using meaningless pictures and recall rates were lower (Shepard 1967). The downside of such real life research is that confounding variables are not controlled. For example, participants could've looked at their yearbooks over the years and unconsciously rehearsed the information. 

Atkinson and Shiffrin (1968)

Sensory register - information enters through senses. Has several stores, one for each of the 5 senses. Coding = iconic (visual) and echoic (auditory). Capacity = over 1/2 million cells in each eye all storing information. Duration = less than half a second

STM - Coding = acoustic. Capacity = 5 to 9 items. Duration = 18 to 30 seconds. Maintenance rehearsal is when we repeat material over and over to ourselves. If we rehearse enough, information will pass to LTM. 

LTM - Coding = semantic. Capacity = unlimited. Duration = up to a lifetime. When we want to recall information it has to be transferred back to STM by retrieval. 

Supporting research evidence - There is research studies that support the theory that STM and LTM are qualitatively different. Baddeley found that we tend to mix up words that sound similar when using STM. We tend to mix up words that have similar meanings when using LTM. This is a strength because it clearly shows that coding in STM is coustic and in LTM is semantic. This supprts the MSM's view that the two stores are separate and independent. 

There is more than one type of STM - The MSM states that STM is a unitary store (only one type), however case studied from amnesia patients shows this can't be true. KF's STM for digits was poor when read out loud to him, but recall was much better when he could read the words himself. The unitary STM is a limitation because research shows that at least there must be one store to process auditory and another for visual information. This was later proposed by the working memory model. 

More than one type of rehearsal - According to MSM what matters is the amount of rehearsal. However, Craik and Watkins (1973) found that what matters is the type of rehearsal. Maintenance rehearsal doesn't transfer to LTM. Elaborate rehearsal is needed for long term storage (linking information to existing knowledge). 

Tulving (1985) argued that the MSM view of LTM was too simplistic and inflexible. 

Episodic memory - Ability to recall events (episodes) from our life. These memories are time stamped. We have to make a conscious effort to recall them. E.g. recent visit to the dentist, what breakfast you ate today.

Semantic memory - Stores knowledge of the world. Is not time stamped. Semantic knowledge is less personal. Is constantly being added to. E.g. applying to university, meanings of words, capital cities. 

Procedural memory - Memory for skills, actions or how to do things. Can recall without any conscious effort or awareness. We may find it hard to explain these skills to soemone else. E.g. driving a car, riding a bike, swimming. 

Clinical evidence to support - Cases of HM and Clive Wearing demonstrated that episodic memory can be damaged bur procedural and semantic memory can remain unaffected. For example, HM would not be able to recall stroking a dog an hour earlier, but would know what a dog was (semantic). Both patients knew how to tie their shoelaces and Clive Wearing could even remember how to play the piano (procedural).

Problems with clinical evidence - HM, Clive Wearing and KF were only used as examples and evidence after an injury when they experienced brain damage. You cannot be certain which parts of the brain are damaged until death, but most studies are on living patients. These individuals are case studies, which means that the findings cna't be generalised to explain everyone with brain damage, as not all people react and behave in the same way. 

Real life applications - Belleville et al (2006) demonstrated that episodic memories could be improved in older people who had a mild cognitive impairment. The trained participants performed better on a test of episodic memory after training, in comparison to a control group. Epsiodic memory is the type most often affected by cognitive impairment. Distinguishing between types allows specific treatments to be developed. 

Baddeley and Hitch (1974) states that the MSM was too simplistic. The MSM says that STM is a unitary system, however the WMM says that there are different systems for different types of information. 

Central executive - directs attention to particular tasks. Controls other systems by determining how resources will be allocated to slave stores. Can briefly store information, has a very limited capacity, is modality free.

Phonological loop - deals with auditory information. Includes written and spoken material. Temporarily retains language based information. Is subdivided into the phonological store: 'inner ear', stores words we hear, holds memory of sound for 2 seconds in the order it was heard. Articulatory rehearsal process: 'inner voice', allows maintenace rehearsal, capacity is thought to be about 2 seconds worth of what you can say.

Visuo-spatial sketchpad - processes visual and spatial information in a mental space. Has limited capacity of 3/4 objects. Logie (1995) subdivided it into the visual cache (stores visual data) and inner scribe (records the arrangement of objects in the visual field)

Episodic buffer - was later added by Baddeley in 2000. Facilitates communication between central executive and LTM. Brings information from all systems together into a single memory. Can be seen as storage for CE. Limited capacity of about 4 chunks. 

Clinical evidence - suppprts comes from KF case study. He could recall letters and digits but had difficulty with sounds, which suggests that his phonological loop was damaged but other areas were not. This supports the existence of separate visual and acoustic stores. 

Dual task performance - Studies of dual task performance supports the separate existence of the VSS. Baddeley showed that participants had difficulty doing two visual tasks than doing one visual and one verbal task at the same time. Tasks from the same store compete for the same slave system, but doing different tasks means there is no competition. 

Lack of clarity over the CE - some cognitive psycholgists suggests that the CE is unsatisfactory and doesn't really explain anything. The CE needs to be more clearly specified than just being 'attention'. This means that the WMM hasn't been fully explained. 

Brain scanning - Braver et al (1997) gave participants tasks that involved the CE while having a brain scan. They found greater activity in the left prefrontal cortex, and activity increased as the task got harder. As demands on the CE increase, it has to work harder to fulfil its function

Says we forget because one memory blocks another, causing one or both to be distorted or forgotten. 

Proactive interference - when an old memory interferes with a new memory. Works forward in time. Old prevents new e.g. the memory of and old phone number disrupts attemts to recall a new one. 

Retroactive interference - when a new memory interferes with an old memory. Works backwards in time. New prevents old e.g. the memory of a new car registration prevents the recall of an old one. 

Effects of similarity - interference is more likely when information are similar, due to response competition. McGeoch and McDonald (1931) studied retroactive interference by changing levels of similarity between materials. Participants learned a list of words until 100% accurate recall. Then learned a new list. 

Findings - when participants recalled the original list recall depended on the nature of the second list. Most similar materials (synonyms) produced worst recall. Shows that interference is strongest when memories are similar 

Evidence from lab studies - interference in memory is arguably the most consistently demonstrated findings. Thousanda of lab studies, such as McGeoch and McDonald, have provided the same conclusions. Most studoes how that both types of interference are very likely to be common ways we forget in LTM. This is a strenght becasue lab experiments control the effects of irrelevant influences. 

Artificial material - the stimulus used in most studies is a list of words. Learning a list of words is much more realistic than consonant syllables but is still far from the things we remember in real life. This is a limitation because the use of artificial tasks means interference may be more likely in a lab than in everyday instances. 

Time between learning - lab experiments are designed to maximise the possibility of interference. An example is the time between learning lists and then recalling them. These periods are relatively short for practical reasons; a participant could learn one list and then learn another 20 minutes later, but recall only 5 minutes after that. The wholoe process could be over in an hour, which is a limitation becase the time between learning and recall in real life could be days/weeks/months.

Forgetting becasue we do not have the necessary cues to access a memory. The memory is available but not accessible unless a suitable cue is provides. 

Encoding specificity principle - Tulving (1983) reveiwed research on retrieval failure and found a consistent pattern that a cue has to be present at encoding as well as retrieval in order to enable recall. If the cues at encoding and retrieval are different there will be some forgetting. 

Context dependent forgetting - external cues. Depends on the environment in which encoding takes place and this acts as a cue to remember in the same environment again. Godden and Baddeley (1975) - deep sea divers learned a list of words either on land or underater, and then recalled the list either on land or underwater. (learn land/recall underwater, learn land/recall land, learn underwater/recall land, learn underwtaer/recall underwater). Found recall was better when the environments were the same at encoding and retrieval. Recall was 40% lower when environment was different. 

State dependent forgetting - internal cues. Depends on the internal state of the participant when encoding takes place. Recall is easier in the same state. Carter and Cassaday (1998) gave anti-histhamine drugs to participants to make them drowsy. Participants had to learn a list of words/passages and recall them. (Learn on drug/recall on drug, learn on drug/recall off drug, learn off drug/recall off drug, learn off drug/recall on drug). Found recall was significantly worse when states were different. 

Supporting evidence - Godden and Baddely/Carter and Cassaday provide support for the theory. The amount of supporting evidence increases the validity of retrieval failure as an explanation for forgetting. This is true when evidence shows retrieval failure occurs in real life situations as well as lab experiments.

Questioning context effects - Baddelely (1997) argues context effects are actually not very strong, especially in real life. Contexts have to be very different before an effect is seen, for example it would be hard to find an exnvrinment as different from land as underwater. In contrast, learning soemthing in one room and recalling in another is unlikely to result in much forgetting.

Recall vs recognition - context effect may be related to the kind of memory being tested. Godden and Baddeley replicated the deep sea diver study using a recognition test instead of recall. Participants had to say whether they recognised a word from the list rather than retrieve them themselves. When testing recognition there were no context effects. 

Problems with ESP - it isnt testable and leads to a form of circular reasoning. When a cue produces a successful recall, we assume it was present at encoding and vice versa. This is only an assumprion; there is no way to establish is the cue was encoded or not. 

When incorrect information is given to the eyewitness after the evebt. Can be leading questions and post-event discussion between cowitnesses and/or other people. 

Leading questions - Loftus and Palmer (1974). Participants watched cips of car accidents and then answered questions. The leading question was to descibe how fast the car was going. "How fast was the car going when they hit each other?". The verb changed for each group of participants (hit, contacted, collided, bumped, smashed) and when 'smashed' was used the average speed estimate was 40mph but when 'contacted' was used the mean speed was 31mph. 

Another experiment was carried out and participants watched a clip of an accident. One group heard the verb 'smashed' and another heard 'hit'. They were later asked if they saw any broken glass on the floor. Those who heard 'smashed were twice as likely to recall seeing glass. 

The response bias explanation - wording of the question has no real effect on memory but does influence how they decide to answer. When asked about the 'smashed' car, participants are encouraged to choose a higher speed. 

Substitution explanation - Supported by second experiment. The wording of the question actually changes the memory of the clip. Participants who head 'smashed' weremore likely to report seeing broken glass than tho0se who heard 'hit'. 

Gabbert (2003) said that this occurs when there is more than one witness. They may discuss wheat they have seen and this may influence the accuracy of their recall of the event. 

She studied participants in pairs and each of them watched a video of the same crime filmed from different viewpoints. This meant that each participant saw elements that others did not. Both participants discussed what they had seen before completing an individual recall test. 

71% mistakenly recalled aspects of the event that they didn't see in the video. This was 0% for a control group who had no time to discuss. 

Has useful real life application - Is hugely important in preventing false witness statements. The consequences of inaccurate EWT can be very serious. Loftus (1975) believes that leading questions can have such a distorting effect on memory so police officers need to be careful when interviewing witnesses. Research into EWT is one ares which psychologists believe they can make a positive impact on people's lives. 

Artificial tasks - Loftus and Palmer's participants watched video clips which is different from witnessing it themselves. There is some vidence that emotions play a part in memory so this is a limtation becasue the same thing may not have happened if the participants were real witnesses to a car accident. 

Individual differences - There is evidence that older people are less accurate than younger people when giving an EWT. Anastas and Rhodes (2006) found that people from 18-25 and 35-45 were more accurate than those from 55-78.

Stressful situations create anxiety, which is a sate of emotional and physical arousal. It is unclear whether anxiety improves or worsens recall. 

Negative effect - Johnson and Scott (1976) said anxiety creates a physicological arousal in the body which prevents us paying attention to cues, so recall is worse. Looked at the effect of weapons on recall accuracy. While in a waiting room participants overheard a staged argument in the next room. Low anxiety condition - man walked through wating room carrying a pen with grease on his hands. High anxiety condition - argument was heated, was sound of broken glass, man walked out carrying a knife and covered in blood. 44% accurate recall in low anxiety, only 33% accurate in high anxiety condition. 

Positive effcet - Yuille and Cutshall (1986) said anxiety triggers the fight or flight response, which increases alertness and improves memory for the event as we become more aware of cues. A shop owner shot a thief while participants were present. Months later witnesses were asked to come back and rate how stressed they were at the time. Those who reported being most stressed had most accurate recall (88% compared to 75% for less stressed). 

Inverted U - Yerkes and Dodson (1908) concluded that as anxiety increases recall also increases because the increased level of alertness helps us remember. However, there comes a point where it begins to have the oposite effect. Deffenbacher (1983) said that is a witness experiences more than the optimal level of anxiety recall suffers. 

Irrelevance of the weapon focus effect - participants may focus on the weapon because they are suprised at what they see ratehr than out of fear. Pickel (1998) conducted an experiment using scissors, a handgun, a wallet or a raw chicken. Accuracy was significantly poorer in the unusual conditions (chicken and handgun). Suggests that weapon focus effect is due to unusualness rather than anxiety. 

Field studies - there is a lack of control over extraneous variables and things may happen that the researcher has no control over. An example is post-event discussion. This is a limitation because these variables may be responsible for the accuracy of the recall. 

Ethical issues - may subject participants to psychological harm. This is why real life studies are beneficial, as psychologists interview people who have witnessed a real event so there is no need to create it. 

The inverted-U is too simplistic - anxiety is hard to define and measure accurately. One reason is because it has many elements (cognitive, behavioural, emotional, physical). The inverted-U explanation assumes only one of these is linked to poor performance > physciological arousal. 

CI is a method of interviewing witnesses to help them retrieve more accurate memories. 

Fisher and Gieselman (1992) argues EWT could be improved if polive used better interviewing techniques. Report everything - wintesses are encouraged to report every detail, even the most trivial as they could trigger other memories. Reinstate the context - witnesses should mentally return to the original scene, they are told to imagine the environment which relates to context dependent forgetting. Reverse the order - events should be recalled in a different order which prevents people from reporting their expectations of the event and prevents dishonesty. Change perspective - wtinesses are told to recall the event from other people's perspectives. 

Fisher et al (1987) developed the enhanced cognitive interview which had additional elements to focus on the social dynamics of intercation. The interviewer needs to know when to establish eye contact, shouldn.t distract with unnecessary interruptions, uses open ended questions, controls flow of information, allows witness to say they 'don't know', reduce their anxiety. 

The CI is time consuming - police may be reluctant to use the CI because it takes longer than a standard police interview. More time is needed to establish a report with the witness. Also requires special training that many forces can't provide, so it is unlikely that the proper CI is even used. 

Some elements may be more valuable than others - Milne and Bulle (2012) found that eavh technique used on it's own produced more information than a standard interview. A combination of 'report everthing' and 'reinstate context' produced better recall thanany other conditions. This is a strength because is suggests that at least 2 elements should be used to improve police interviews, even if the full CI isn't used. 

Support for the effectiveness of the ECI - Kohnken et al (1999) conducted a meta-analysis from 50 studies. The ECI consistently produced more correct information than a standrad police interview. This is a strength because it indicates that there are real practical benefits. Research also shows that it gives police a better chance of catching the criminal, which benefits society as a whole.