Memory Revision Notes PDF

Preview

Summary

full notes...

Description

Memory – Paper 1 STM and LTM Capacity Capacity concerns how much data can be held in a memory store: STM limited, LTM infinite Capacity of STM assessed through digit span (Jacobs, 1887) Found average span for digits 9.3 items, 7.3 for letters, suggested easier to remember digits as only 9 v 26 letters Miller (1956) concluded span of immediate memory is around 7 items Noted people can count 7 dots, musical notes, letters and words Can recall 5 words as well as 5 letters because we chunk things together

Evaluation One criticism of Miller’s research is that the original findings have not been replicated Cowan (2001): review of studies, found that STM likely to be limited to 4 chunks Vogel et al (2001): capacity for visual information also 4 = lower end of range more appropriate, and STM is not as extensive as thought However, evidence has been found that supports the idea that the size of the chunk affects how many you can remember Simon (1974) found people had a shorter span for larger chunks i.e 8-word phrases, than smaller chunks = continues to support the view that STM has a limited capacity and refines understanding Another criticism of capacity is that there are individual differences Jacobs found digit span increased steadily with age (8yr 6.6 digits v 19yr 8.6 digits), age increase may be because of changes in brain capacity/development of strategies = capacity of STM is not fixed for everyone, and individual differences contribute to the capacity of STM

Duration LTM potentially lasts forever, STM not very long Peterson and Peterson (1959) used 24 students to study the duration of STM, tested over 8 trials PROCEDURE: On each trial pt were given a consonant syllable and 3-digit number, asked to recall the syllable after retention interval of 3,6,9,12,15 or 18 seconds, during which they had to count backwards from their 3-digit number FINDINGS: 90% over 3s, 20% correct after 9s and 2% after 18s = STM has a very short duration as long as rehearsal is prevented Bahrick et al (1975) tested 400 people between 17-74 on memory of classmates PROCEDURE: Photo-recognition test of 50 photos, some from yearbook; asked to free-recall names they could remember of classmates FINDINGS: Pt tested within 15yrs of graduating were 90% accurate in identifying faces, after 48 years declined to 70% Free recall accurate 60% after 15yrs, 30% after 48yrs

Evaluation A criticism of research investigating STM is that it is artificial Trying to remember consonant syllables does not truly reflect most everyday memory activities that involve remembering meaningful things, yet we do remember fairly meaningless things i.e groups of numbers (phone) or letters (postcode) = although the task was artificial, there is still some relevance to everyday life A criticism of the Petersons’ study is that it did not actually measure what it set out to

Pt were counting the numbers in their STM, which may displace the syllables to be remembered Reitman (1974): used auditory tones instead of numbers so displacement does not occur, found duration of STM was longer = forgetting due to displacement than decay in Peterson x2 study

Coding The way information is changed so it can be stored in memory – can be held in form of sounds (acoustic), images (visual) or meaning (semantic) Cat, cab, can, cap are acoustically similar but semantically different Great, large, big, huge are semantically similar but acoustically different Baddeley (1966a and 1966b) used word lists to test the effects of acoustic and semantic similarity on STM and LTM Found pt had difficulty remembering acoustically similar words in STM only, and semantically similar words posed problems in LTM = suggests that STM is largely encoded acoustically, LTM semantically

Evaluation Some experiments have shown that visual codes are also used in STM Brandimote et al (1992): found pt used visual coding in STM when given a visual task and prevented from verbal rehearsal in retention interval before performing visual recall task; pt used visual codes instead of normal translation; Wickens et al (1976): STM sometimes uses semantic code = STM not exclusively acoustic In general, LTM appears to be semantic, but this is not always the case Frost (1972): LT recall related to visual and semantic categories Nelson & Rothbart (1972): found acoustic coding in LTM = seems that coding in LTM is not simply semantic but can vary according to circumstances Baddeley’s method criticized – STM tested immediately after, LTM 20 mins = casts doubt over validity, not really testing LTM _______________________________________________________________________________________________

Multi-Store Model of Memory (Atkinson & Shiffrin, 1968) Multi-store as consists of 3 memory stores, linked to each other by processes that enable transfer of information from one to the next Sensory register is where information is held at each sense & corresponding brain areas Capacity very large and constantly receive information, but most receives no attention and remains in SR for very brief duration If a person’s attention is focused on one of the sensory stores, the data is transferred to STM – first step in remembering something Information held in STM to be used for immediate tasks Limited duration as in ‘fragile’ state and will decay relatively quickly if not rehearsed Information will disappear also if new information enters STM, displacing original as limited capacity Maintenance rehearsal, such as repetition, keeps information in STM but eventually creates LTM Atkinson & Shiffrin proposed direct relationship between rehearsal in STM and strength of LTM LTM potentially unlimited in duration and capacity Retrieval is process of getting information back from LTM, involving the information passing back through STM, making it available for use

Evaluation A strength of the MSM is that it acknowledges the qualitative differences between STM and LTM by representing them as separate stores. This has been shown by lab studies on capacity, duration and coding, and studies using brain scanning techniques have also shown a difference between STM and LTM. For example, STM is encoded acoustically, whilst LTM is encoded semantically and has a much longer duration; Beardsley found that the prefrontal cortex was active during STM and Squire et al found the hippocampus active during LTM. Therefore, the MSM portrays an accurate view of the differences between the two types of memory, as supported by Baddeley and Miller, and the model also has supporting studies supporting it. Case studies of individuals have shown areas of brain to be involved in STM and LTM e.g HM: damage caused by op to remove hippocampus to reduce epilepsy, personality and intellect intact but could not form new LTM, but could remember from before surgery = support for MSM’s notion of separate stores, as HM unable to transfer information from STM->LTM, but could retrieve information from LTM MSM suggests STM & LTM are unitary stores, but this is not supported by research WMM shows STM & LTM are both divided into qualitatively different stores; research shows different kinds of LTM, with each behaving differently = suggests MSM may be overly simplistic MSM criticised for emphasis on maintenance rehearsal Craik & Lockhart (1972): enduring memories created by processing, things processed more deeply are more memorable Craik & Tulvig (1975): list of nouns, asked pt questions involving shallow/deep processing (CAPS or sentence), remembered more words in deep = suggests that process of rehearsal does not fully explain the process of creating LTM – deep processing is part of the process MSM suggests that STM is involved before LTM BUT this claim has been questioned by other researchers Logie (1999): STM relies on LTM and therefore cannot come first; in order to chunk you need to recall meaningful groups of letters and such meanings are stored in LTM Ruchkin et al (2003): asked pt to recall a set of words and pseudo-words, found much more brain activity when real words processed, indicating the involvement of others areas of the brain than just STM = suggests that STM is part of LTM and not a separate store _______________________________________________________________________________________________

Working Memory Model (Baddeley and Hitch, 1974) Baddleley & Hitch (1974) felt STM was a number of different stores than just one If you do 2 things at the same time of the same nature, you perform less well If you do 2 things at the same time of different natures, there is no interference = suggests one store for visual processing and another for processing sounds, formed basis of WMM where ‘slave systems’ are organised by central executive Function of Central Executive is to direct attention to tasks, determining how resources are allocated Data arrives from senses or LTM, Limited capacity Phonological Loop has limited capacity, deals with auditory information & preserves order

Baddeley (1986) divided into phonological store (holds words you hear), articulatory process (words heard/seen, silently looped (form of rehearsal)) Visuo-Spatial Sketchpad used to plan spatial tasks, visual/spatial information temporarily stored here Logie (1995): VSS can be divided into visual cache and inner scribe (arrangement) Episodic Buffer added in 2000, extra storage, integrates info from stores, orders it into episode sent to LTM

Evaluation Shallice and Warrington’s study of KF provides support for the WMM because their findings show that KF had very poor STM recall for auditory stimuli, but increased STM recall for visual stimuli. This suggests that the components of memory which process auditory and visual stimuli are separate, as described in the WMM through the PL and VSS. Further support for the WMM comes from studies of dual-task performance, where each pt undertakes a visual and verbal task simultaneously, and shows decreased performance for such tasks. Consequently, this supports the idea that the CE has a very limited processing capacity and that the slave systems are in competition with each other for these tasks and resources. Neuroscanning evidence, such as that provided by Braver et al, has demonstrated a positive correlation between an increasing cognitive load processed by the CE, and increasing levels of activation in the prefrontal cortex. This supports the idea that the CE has the role of allocating tasks to slave systems and has a limited capacity, as reflected by the increased brain activation levels. As a result, this suggests the WMM is accurate in its mechanism of the CE. However, the CE has not been precisely defined. For example, the term ‘process’ is vague, and the CE may be made up of sub-components or be a part of a larger component itself in WM. This lack of a comprehensive explanation for each component draws doubts about the accuracy of its depiction of WM. Finally, there are problems with using case studies of individuals with brain damage. Some supporting evidence for the WMM comes from case studies of individuals such as KF, but such individuals may have difficulty in paying attention and so simply underperform on certain tasks. This is an important issue because the results of case studies are difficult, if not impossible, to generalise to the general population. _______________________________________________________________________________________________

Types of LTM EPISODIC MEMORY – concerned with personal experience and is an individual’s unique memory of a specific event or events where they were involved. Have 3 elements: details, context and emotions at time. SEMANTIC MEMORY – memories related to knowledge about the world, shared by everyone. Related to things, such as the function of objects, or what behaviour is appropriate in a situation. May be related to abstract concepts such as maths. PROCEDURAL MEMORY – concerned with skills, such as knowing how to tie a shoelace; it is remembering how to do something rather than knowing what to do. Typically acquired through repetition and practice, less aware of memories as they have become automatic.

Evaluation

Research support comes from Petersen et al, who demonstrated that semantic memories were recalled from the left prefrontal cortex, whilst episodic memories were recalled from the right PFC. This supports not only the idea that there are different types of LTM, but shows that they each have a different neurological basis because they are recalled from different parts of the brain. Further support comes from, Cohen and Squire, who drew a distinction between declarative and non-declarative memories. Declarative memories must be recalled consciously, whilst non-declarative memories may be recalled unconsciously. However, this is a different classification and organisation system as the one used by Tulving, suggesting that his depiction of LTM is not entirely accurate. Moreover, the cases of HM and Clive Wearing show how while one type of LTM may be impaired, the others will be unaffected. Wearing was able to still skilfully play the piano and understand the concept of music, but was unable to remember his wife visiting him 5 mins previously. This gives strong support to the idea that different areas of the brain are involved in the types of LTM, and confirms that the different types of LTM are separate. However, a problem with supporting research for types of LTM is that it relies on research involving patients with brain damage. It is difficult to conclude from patients such as HM the exact parts of the brain that are affected until after the patient has died. Also, damage to a particular area of the brain does not mean this area is responsible for a particular behaviour. This means that we cannot establish a causal relationship between a particular brain region and type of LTM. _______________________________________________________________________________________________

Explanations for Forgetting: Interference RETROACTIVE INTERFERENCE – observation that learning something new interferes with previously learned material, leading to that being forgotten. Muller and Pilzecker (1900): first identified RI where pt given an intervening task inbetween learning nonsense syllables and recalling them; performed less well than those without task McGeoch & McDonald (1931): forgetting of original material is greater if the intervening items presented prior to recall are more similar to original, something only RI rather than decay PROACTIVE INTERFERENCE – previously learned material interferes with current attempts to learn something leading to forgetting of current material Underwood (1957): analysed findings from various studies, found pt were less able to learn word lists presented later in a sequence, in comparison to those presented earlier on PI results from response competition between current and previously learned information at recall. The finding that the PI effect was greater in later lists is attributed to the increasing competition of multiple associations at recall.

Evaluation A weakness with this explanation is that the artificial stimuli used in these tasks, such as learning lists of random words with no personal meaning to the participants, means that the findings of interference studies are likely to have low mundane realism. This is because in real life, we are likely to learn lists of meaningful information, such as revision topics for psychology, which we draw links upon and also which have personal meaning to us. These factors may also influence the extent of forgetting, rather than influence. A second methodological criticism of interference studies, further suggesting that they lack mundane realism and reliability, is that they are often conducted in very short spaces of time, with participants recalling their words 1 or 2 hours after they have learnt them. This does not reflect the normal passage of time in everyday life, where we often find that several days pass until we need to recall such information e.g. in the case of an

exam. Therefore, this suggests that interference is unlikely to be a valid explanation for forgetting from the LTM. However, interference has been consistently demonstrated in several studies, but particularly in lab experiments. This increases the validity of the theory, due to the use of highly-controlled conditions in lab experiments, standardised instructions alongside the removal of the biasing effects of extraneous and confounding variables. Baddeley and Hitch found that, in a group of rugby players who had to recall their last game and the number of games they’d played that season (which would be different for each player), the number of games they’d played since was more important than the total time they’d been playing for. This can be explained in terms of interference, where the more games each player had played, the more likely the memories of these newer games would interfere or block the recall of older games i.e. retroactive interference. _______________________________________________________________________________________________

Explanations for Forgetting: Retrieval Failure Memory most effective if information that was present at encoding is also available at retrieval. The closer the cue to original, the more useful Forgetting in LTM mainly due to retrieval failure – the failure to find an item of information because of insufficient cues during retrieval. CONTEXT-DEPENDENT FORGETTING – recall of knowledge is greater when context present during learning and retrieval are the same Godden and Baddeley (1975): divers learned word list on land or underwater, then tested on land or underwater. Highest recall was when initial context matched recall environment STATE-DEPENDENT FORGETTING – recall is greater when an individual’s physical or psychological state is similar at encoding and retrieval. Goodwin et al (1969): asked pt to remember a list of words when drunk or sober, and then 24 hours later when drunk or sober. Recall best when in the same state both times

Evaluation Eysenck has suggested that retrieval failure may be one of the main reasons that we forget information from the LTM. This, alongside the strictly-controlled conditions of a lab experiment (reducing the biasing effects of extraneous and confounding variables), increases the validity of retrieval failure as an explanation for forgetting, due to more confidence being placed in these conclusions on the basis of such experimental designs. The findings from studies of retrieval failure may lack ecological validity. This is because Baddeley argued that it is difficult to find conditions in real-life which are as polar as water and land, for example, and thus questioned the existence of context effects in normal life. This suggests that retrieval failure may be best suited to explaining cases of forgetting where the cues associated with encoding and retrieval are uncommonly distinct, thus not providing an accurate depiction of forgetting in day to day life. Godden and Baddeley repeated their underwater, deep-sea diver experiment (1975) but tested for the recognition of learnt words, as opposed to recall, and found no significant difference in accuracy of recognition between the matched and non-matched conditions. This suggests that retrieval failure may only explain forgetting for some types of memory, tested in specific ways and under certain conditions, hence not being a universal explanation. This further suggests that the findings from studies of retrieval failure suffer from poor generalisability. The encoding specificity principle suffers from cyclical reasoning due to its overreliance on assumptions. For example, it may not always be the case that differences between cues at the time of encoding and recall causes retrieval failure, but the cyclical nature of the ESP suggests that it is so. _______________________________________________________________________________________________

Accuracy of EWT: Misleading Information

Loftus and Palmer (1974): Leading Questions EXPERIMENT 1 PROCEDURE: researchers showed 45 students 7 films of road accidents. After each film pt given a questionnaire with critical q...