Memory Flashcards
A01 and A03 in the format of well worked exam answers
Describe and outline the Multi- store model
The multi- store model of memory (MSM) developed by Atkinson and Shiffrin explains how information flows from one storage system into another in a linear way. The model is made up of three unitary permanent stores in memory which are: sensory register (SR), short term memory (STM) and long term memory (LTM). Each memory structure differs in terms of their capacity; which means how much information can be stored, their duration; which means how long information can be stored and coding; which is the form in which information can be stored. Information through the senses enters the sensory register and is coded in its original form for only a short duration; information is lost rapidly from sensory memory after only half a second. The STM is also limited in duration, around 18 seconds. It has a capacity of 5-9 items of information and encodes these items acoustically, by sound. LTM is for permanent storage as its capacity is unlimited, it has lifetime duration and it encodes semantically, by meaning. There are also different processes involved in transferring information through the different stores. Attention takes information from sensory memory to short term memory. Any information which is not attended to will decay and be lost. Maintenance rehearsal in the form of acoustic repetition holds information in STM and moves it to LTM, any new information or distractions preventing rehearsal in the STM will push the information out and it will be lost.
Draw the diagram for the multi store model
- refer to booklet
What are the key features of each store? (CCD)
Coding refers to the form in which a
memory is stored in a particular store
Capacity refers to the amount of information that can be held at any one time in memory.
Duration refers to the amount of time that a memory can be stored
Capacity, Duration and Coding of the sensory register
Capacity = Unlimited
Duration = A split second
Coding = In original form via one of the senses e.g. sight, touch, smell etc.
Sperling Sensory register research
To find the existence of sensory memory Sperling asked participants to look at a chart containing a grid of letters for less than a second and were asked to recall how many of the letters they could remember. The participants could remember only 4/5 letters, although they said they had been aware of more. Then Sperling got the participants to recall single rows of letters when particular tones were heard. High tone for top row, medium tone for middle row and low tone for the bottom row. On average 3 items were recalled from any indicated row. This shows that all the letters entered awareness, but only those attended to could be recalled. Research supports the idea that the SR has a large capacity but a very short duration.
Evaluation of the methodology used in Sperlings research into the sensory register
This research was carried out in a lab to ensure control over the setting and task, although this is good for internal validity, the findings don’t really apply in a real life setting; remembering number strings is not what you do in real-life and so it is not a valid test of everyday memory and so the results lack ecological validity. However, the findings have been replicated by many other psychologists showing the result to be reliable.
Capacity, Duration and coding of the short term memory
Capacity = 5-9 items
Duration = 18-30 seconds
Coding = Acoustic
Jacobs short term memory, capacity research
Jacobs conducted a lab experiment using a digit span test. 443 female students had to repeat back a string of numbers or letters in the same order and the number of digits/letters was gradually increased, until the participants could no longer recall the sequence. On average Jacobs found that the students could recall 7 letters and 9 numbers, which has been developed in the conclusion that STM capacity is 7+/-2 items. Others, like Miller, have also discovered that chunking can increase capacity (BBC or 01858 becomes one chunk of information rather than 3 or 5 distinct chunks).
Evaluation of the methodology used in Jacobs STM, capacity research
This research was carried out in a lab to ensure control over the setting and task, although this is good for internal validity, the findings don’t really apply in a real life setting; remembering number strings is not what you do in real-life and so it is not a valid test of everyday memory and so the results lack ecological validity. However, the findings have been replicated by many other psychologists showing the result to be reliable.
Peterson and Petersons Duration research in the short term memory
In a lab experiment Peterson and Peterson required 24 psychology students to briefly look at nonsense trigrams (strings of letters e.g. TZH) then count backwards in threes. The counting task preventing rehearsal so that only STM was being tested. They were then asked to recall the trigrams they had seen seconds before, in each trial the counting interval increased by 3 seconds. Most people remembered 80% of the trigrams after 3 seconds of counting but fewer than 10% of the trigrams were recalled correctly after 18 seconds. This shows that the duration of STM is around 18 seconds maximum.
Evaluation of the methodology used within Peterson and Petersons research into Duration in the STM
This memory experiment has been carried on a limited sample; 24 students who are gathered by opportunity sampling. This means the sample is bias as it is unlikely that all people will use memory in a similar way to students. This means the results are not representative of the target population (all humans) and so should only be generalised with caution.
Baddeley’s research into coding in the short term memory
In a lab experiment Baddeley divided 75 participants into 4 groups. Each group heard a different list of five words. Immediately after hearing the 5 words they were asked to recall them in the correct order. This was repeated 4 times. Participants given acoustically similar words performed the worst, with a 55% accuracy. They confused similar sounding words, such as recalling ‘cap’ instead of ‘cat’. Since acoustically similar words were recalled the least efficiently, it seems there is acoustic confusion in STM, suggesting STM is encoded on an acoustic basis.
Evaluation of the methodology used within Baddeley’s research into coding in the STM
This research was carried out in a lab to ensure control over the setting and task, although this is good for internal validity as possible extraneous variables like noise distractions have been controlled for, the findings don’t really apply in a real life setting; remembering random words is not what you do in real-life and so it is not a valid test of everyday memory and so the results lack ecological validity.
Capacity, Duration and coding of the Long term memory
Capacity = Unlimited
Duration = Lifetime
Coding = Semantic
Linton’s long term memory capacity research
Linton kept a detailed diary recording daily events and facts on cue cards over many years. Each day was represented by only a single word. After 7 years she tested her recall of the 11,000 events she had recorded. When given the cue word she was able to recall with 70% accuracy all the events and information recorded for that day showing the potentially huge capacity of LTM.
Evaluation of the methodology used within Lintons long term memory research into capacity
Linton’s diary study is a type of case study as she only studied one person – herself! This does not have good population validity because she may have a unique memory compared to other people and so her results cannot be generalised. Although she was testing her memory of events which is a fair test of everyday memory and is not as artificial as some lab experiments which means they have good ecological
Bahricks research into duration within Long term memory
Bahrick et al used high-school yearbooks to test very long-term memory for names and faces, using 400 participants aged 17 to 74 in their study. The participants were given various tests such as a free – recall test where they were asked to name all the people in their graduating class (free recall). For the participants who had left school nearly 50 years before the experiment, identification of names and faces stood at 70-80%. Considering the duration as well as the fact that the latter group were on average much older, this is a small difference and shows the potentially life time duration of LTM.
Evaluation of the methodology used within Bahricks study into duration within Long term memory
Testing memory of old school mates is a fair test of everyday memory and is not as artificial as some lab experiments which means the results have good ecological validity.
Baddeley’s research into how information is codded within long term memory
In a lab experiment Baddeley divided 75 participants into 4 groups. Each group heard a different list of five words. 20 mins after hearing the 5 words they were asked to recall them in the correct order. This was repeated 4 times. Participants given semantically similar words performed the worst. Since semantically similar words were recalled the least efficiently, it seems there is semantic confusion in LTM, suggesting LTM is encoded on a semantic basis.
Evaluation of the methodology within Baddeley’s research into coding within the LTM
This research was carried out in a lab to ensure control over the setting and task, although this is good for internal validity as possible extraneous variables like noise distractions have been controlled for, the findings don’t really apply in a real life setting; remembering random words is not what you do in real-life and so it is not a valid test of everyday memory and so the results lack ecological validity.
What does VLTM refer to?
this just means ‘very long term memory’ and refers to research such as Bahrick’s and Linton’s where memory over the course of years is tested, rather than just after 20mins like in Baddeley’s study.
What processes are involved in transferring information between different stores (MSM)
attention and rehersal
How is information transfered between stores using attention? (MSM)
If attention is focused on sensory memory, it is transferred to Short Term Memory. Anything else is ignored and lost to save space and prepare for more new information.
How is information transfered between stores using rehersal? (MSM)
Information is maintained in STM through maintenance rehearsal, which takes the form of acoustic (sound based) repetition. If information is not rehearsed it will decay as the duration is thought to be around 18 seconds. Adding meaning, known as elaborative rehearsal, leads to transfer from STM to LTM. The idea is that the more something is rehearsed the better it will be remembered.
Research about how information is transfered between stores using rehersal (MSM)
Murdock carried out a lab experiment which showed that when participants were presented with a list of 20 words and asked to immediately recall them, they tend to remember the first few and last few words and are more likely to forget those in the middle of the list. This is called the serial position effect as memory was affected by the order position of the word.
This supports the existence of separate LTM and STM stores because they observed this primacy and recency effect. Words early on in the list were put into long term memory (primacy effect) because the person has time to rehearse the word, and words from the end went into short term memory (recency effect).
Research against information being transfered between stores using rehersal (MSM)
Rehearsal is considered a too simple explanation to account for the transfer of information from STM to LTM. For instance, the model ignores factors such as motivation and strategy (e.g. mnemonics) which underpin learning.
Also, rehearsal is not essential to transfer information into LTM. For example, we are often able to recall information which we did not rehearse (e.g. celebrity gossip) yet unable to recall information which we have rehearsed (e.g. revision notes). Therefore, the role of rehearsal as a means of transferring from STM to LTM is much less important than Atkinson and Shiffrin claimed in their model.
Clive wearing case study as evaluation for the Multi Store Model
Clive wearing has amnesia (memory loss) because a herpes virus attacked his brain and damaged key areas associated with memory, such as the hippocampus which transfers STM to LTM. Clive is now unable to transfer new information into his long term memory but his short term memory is intact. For example he will forget a question he has been asked or a person he has just seen as little as 7-30 seconds later. This supports the MSM as it show that STM and LTM are separate stores. However Clive has lost his long term memory for events (fact and event memory is known as declarative memory) but his long memory for how to do things such as play the piano (memory for skills and habits is called non-declarative memory) remains unaffected. This actually criticises the MSM as it suggests there is not just one type of LTM but several.
One general weakness of the MSM
The MSM claims rehearsal is required to make long term memories, however in reality this is not the case and the type of information is also important. For example students can rehearse their revision notes and not remember them but can read a gossip magazine once and remember the information easily as it is more meaningful.
Other psychologists have proposed instead the idea of ‘deep processing’ as being the way in which LTM is formed. This idea states that originally well learning information with more meaning will be more likely to become lasting long term memory. This criticises rehearsal as being too simple to explain the process involved in making long term memories.
Comparing the WMM and the MSM
There are alternative models of memory such as the Working Memory model (WMM). This model claims STM is much more complex than in the MSM with different components working on visual or auditory based information. The WMM suggests that the single STM store in the MSM is too simple. However the WMM does not include LTM and so in this respect the MSM is a more full and complete model of memory even if it is perhaps too simple.
Overall the WMM is a much more complex account of STM than the multi-store model (MSM) which claims STM is just one unitary store. The evidence supports the WMM’s more complex ideas about different types of STM. However the WMM is an incomplete model of memory when compared to the MSM as it does not theorise about the workings of long term memory or sensory memory.
Describe and outline the Working Memory model
The working memory model was proposed by Baddeley and Hitch (1974) who viewed the short term memory (STM) as an active store based on processes, holding several pieces of information while they are being worked on. According to them, it has four separate components.
The central executive (CE) system selectively attends to particular types of information and delegates tasks accordingly to the two slave systems. It is ‘the manager’ of the working memory model. The phonological loop (PL), one of the slave systems is the inner ear and deals only with auditory information. It has two sub components; the articulatory loop where information is rehearsed sub vocally, and the phonological store where speech input is held for a very brief duration.
The second slave system, the visuo-spatial sketchpad (VSS) is the inner eye and deals with what things look like and where things are located in the environment and the relationship between things. It helps people to navigate and interact with their physical environment. A key assumption of the model is that we can do two tasks together if they use different slave systems but not if they use the same system as this overloads it and is too difficult.
The fourth component, added to the model in 2000, is the episodic buffer (EB) whose role is to integrate information from the CE, the PL and the VSS as well as from the long term memory. The component can deal with any type of information and acts as a general store in STM.
Draw the diagram of the WMM
- see work book
Coding and capacity of Phonological loop
It works on/codes spoken and written material; things that enter our memory in the form of auditory information limited capacity – set by how long it takes to sub vocally repeat the item
coding and capacity of visuo-spatial sketchpad
Non-auditory information. Visual and spatial items and the relationships between them. Codes information in the form of its visual features e.g. size & shape. limited capacity
coding and capacity of central executive
process/code any type of incoming information limited capacity
coding and capacity of the episodic buffer
process/code any type of incoming information limited capacity
Description of the central executive
This is the command centre for working memory (the manager) and is used with problem solving tasks and decision making. It has limited capacity and duration as it is not a store but more of an attention mechanism. It can process/code any type of incoming information. Its function is to delegate tasks accordingly to the relevant slave system and focus our attention, attaining a balance between tasks when attention needs to be divided e.g. talking whilst driving.
Description of the phonological loop
This is one of the ‘slave’ systems. This is a temporary store with a short duration. It also has a limited capacity. It works on/codes spoken and written material, or in other words, things that enter our memory in the form of auditory information (when you read written info you speak aloud the words in your mind and thus convert it to auditory info). As it is primarily an acoustic store, so confusions can occur with similar-sounding words. Baddeley further subdivided this loop into two parts, the phonological store (PS) which holds auditory memory traces for a few seconds before they decay and an articulatory process (AP) which allows words/sound that are heard or seen to be stored through rehearsal by silently repeating them (sub vocal repetition).
Description of the episodic buffer
This third ‘slave’ system or fourth component was added in 2000 by Baddeley. The WMM encountered problems in explaining how information from different subsystems in different formats/codes could be combined, and how it was capable of communicating with long-term memory. The EB is claimed to be a store of limited capacity capable of holding any format of information. It is a ‘buffer’ in the sense that it is a limited capacity temporary store that forms an interface or link between a range of systems all having different memory codes. It is ‘episodic’ in the sense that it is capable of holding episodes; integrated chunks of information that then became accessible to working memory.
description of the visuo-spatial sketchpad
The VSS is the ‘slave’ system that handles non-phonological information. It is a temporary store for visual and spatial items and the relationships between them; essentially, it is a store for what items are and where they are located. Codes information in the form of its visual features e.g. size & shape. The VSS therefore helps individuals to navigate and interact with their physical environment, with information being rehearsed and encoded through the use of ‘mental pictures’. Within the VSS is the inner scribe that deals with spatial and movement information such as planning out routes and the visual cache, which is a limited capacity store for visual info about form and colour.
A03 evaluation and research into the central executive
Bunge et al (2000) used brain scans to see which parts of the brain were most active when participants were completing two tasks (reading a sentence and recalling the final word in a sentence). Participants either did this simultaneously (at the same time), or one after the other. They found that the same brain areas were active in both the dual and single task conditions, but there was significantly more activation in the dual task indicating increased attentional demands shown by brain activity. This highlights the important role of the central executive that involves directing attention.
A03 evaluation and research into the phonological loop
Baddeley based the phonological loop on research he had conducted into ‘the word length effect’; that lists of short words are remembered better than longer words. Short words can be (silently) articulated faster than longer words and therefore more of them can be articulated in the time (approx 2 secs) before the trace decays. If the participants in a study are then asked to say something aloud at the same time as they are rehearsing material, then memory for the rehearsed material is impaired. This provides evidence for the existence of an articulatory rehearsal process with a duration of 2 seconds.
A03 evaluation and research into the episodic buffer
One finding that may support the EB is the concept of chunking, initially proposed by Miller. This assumes that STM memory is capable of holding only a limited number of chunks, regardless of the size of each chunk. To do this one must be able to use the LTM to recall what things can be grouped by meaning according to when we have seen them combined together in the past. The episodic buffer may undertake this task to help WMM function better. Baddeley showed the power of chunking by performing a simple lab experiment asking people to immediately recall wither unrelated words or words chunked together in a short meaningful sentence. Participants recalled around 7 sentence words and only around 5 unrelated words.
A03 evaluation and research into the visuo-spatial sketchpad
One finding that may support the EB is the concept of chunking, initially proposed by Miller. This assumes that STM memory is capable of holding only a limited number of chunks, regardless of the size of each chunk. To do this one must be able to use the LTM to recall what things can be grouped by meaning according to when we have seen them combined together in the past. The episodic buffer may undertake this task to help WMM function better. Baddeley showed the power of chunking by performing a simple lab experiment asking people to immediately recall wither unrelated words or words chunked together in a short meaningful sentence. Participants recalled around 7 sentence words and
only around 5 unrelated words.
General evaluation of the ecological validity when studying the stores within the WMM
As a general evaluation of any store you can comment that the research evidence tends to be unrealistic as the stimulus materials are artificial and simplistic compared what we need to recall in everyday life. All the experiments are also controlled lab experiments which means participants know they are being tested and so may concentrate harder on memorisation and attention than they would in everyday life. This means results lack ecological validity.
evaluating the central executive in the WMM
To evaluate the WMM there is little direct evidence for how the central executive works and what it does. It is very hard to measure as it does not work on a specific type of information and so the capacity of the central executive has never been measured. Bunge et al (2000) used brain scans to show that the same brain areas were active when participants were doing tasks at the same time or separately, but that there was significantly more activation in the dual task indicating increased attention demands shown by brain activity. This highlights the important role of the central executive that involves directing attention but it does not provide valid evidence as tasks were very simple and highly controlled meaning there was poor ecological validity.
evaluating the idea of having 2 separate slave systems in the WMM
There is convincing evidence to support the claim that there are separate slave systems for visual and auditory information. For example Gathercole & Baddeley found participants’ had more difficulty doing two visual tasks compared to doing one visual and one auditory task. This study supports the existence of separate stores for different types of info and the concept that they can work independently. However the tasks were things like describing the angles of a letter ‘F’ which do not represent fully real life memory and so have poor ecological validity.
A real life case study with better ecological validity is that of ‘KF’. Shallice & Warrington found that following brain damage ‘KF’ showed very poor digit span (usually less than 2 items) when presented with information verbally, but much a better recall when he was able to see the numbers visually written down. This supports the idea that there are different areas for auditory and visual tasks as the WMM suggests. However this case study is of a unique individual and so may not be fully generalizable to all human memory due to a lack of population validity.
Evaluating the capacity of the phonological loop within the WMM
There is research to support the claim that the phonological loop has a limited capacity and its articulatory process allows words that are heard or seen to be stored through rehearsal by silently repeating them (sub vocal repetition). Baddeley conducted a lab experiment into ‘the word length effect’; that lists of short words are remembered better than longer words. Short words can be (silently) articulated faster than longer words and therefore more of them can be articulated in the time (approx 2 secs) before the trace decays. If the participants in a study are then asked to say something aloud at the same time as they are rehearsing material, then memory for the rehearsed material is impaired. This provides evidence for the existence of an articulatory rehearsal process with a duration of only 2 seconds. However because this is a lab based study it lacks ecological validity; the words used were meaningless and so did not have any meaning for participants, it could be that in everyday memory we can recall more as it is relevant and important.
Evaluating the real life application of the WMM
Researchers today generally agree that short-term memory is made up of a number of components or subsystems. This means that the working memory model has real life applications as it makes sense of a range of real life tasks: reading (phonological loop); problem solving (central executive); navigation (visual and spatial processing). It follows, therefore, that understanding the components of this memory system and the interaction of these components might allow us to improve our working memory and reap the intellectual benefits this improvement brings. It has been suggested that some cognitive disorders, such as ADHD, are due to deficits in WM and could be improved by using Working Memory training programmes, for example teaching people to break down tasks into component parts to ease demands on the central executive.
What are the three types of long term memory?
procedural, episodic and semantic
Outline procedural LTM
Memories of motor skills/actions and muscle memories, located in the basal ganglia and cerebellum, they are unavailable for the unconscious inspection to explain verbally, and they do not enter conscious awareness, they are in general more resistant to forgetting or amnesia however we may not recall when we learned these memories
Outline episodic LTM
These are memories of life events and are stored in the hippocampus, they can be expressed verbally as they enter conscious awareness. It is common to lose as a result of forgetting and they are generally stored with a reference to time and place
Outline semantic LTM
These are knowledge or fact based and are stored in the hippocampus, they can be expressed verbally as they enter the conscious awareness however it is common to lose as a result of forgetting or amnesia, we may not recall when we learned them/encoded these memories
Evaluating the types of Long term memory through HM - who had his hippocampus removed
HM had his hippocampus removed to try to control his seizures. Following the operation he could no longer create episodic memories. For example he would watch the same film repeatedly and eat meals twice. However a psychologist named Milner discovered he still had his procedural LTM intact. She found that he improved each time he was asked to draw a star within an outline via a mirror reflection. He got more accurate each time despite having no memory of having done the test before. This showed that his skills based or motor memory did not disappear like his episodic memories and therefore supports the concept of different types of LTM.
Evaluating the types of Long term memory through clive wearing
Clive wearing has amnesia (memory loss) because a herpes virus attacked his brain and damaged key areas including the hippocampus. Clive has lost his long term episodic memory for events as he cannot remember having seen his wife only moments earlier. However, his long procedural memory for how to do things such as play the piano (memory for skills and habits is called non-declarative memory) remains unaffected. This suggests there is not just one type of LTM but several.
Evaluating the types of Long term memory through methodology
A lot of the evidence supporting different types of LTM is based on case studies of unique individuals with serious cases of memory loss. This means the results are not fully valid for two reasons. Firstly the individuals are not representative of healthy adults with functioning memory and so we cannot be sure that their memories are generalizable to normal memory. Furthermore case studies do not have internal validity as many variables are acting on the participant and so cause and effect cannot be claimed. For example HM also had a poor standard of education due to his childhood seizures and this may account for some of his memory problems.
Evaluating the types of Long term memory through Methodology
A lot of the evidence supporting different types of LTM is based on case studies of unique individuals with serious cases of memory loss. This means the results are not fully valid for two reasons. Firstly the individuals are not representative of healthy adults with functioning memory and so we cannot be sure that their memories are generalizable to normal memory. Furthermore case studies do not have internal validity as many variables are acting on the participant and so cause and effect cannot be claimed. For example HM also had a poor standard of education due to his childhood seizures and this may account for some of his memory problems.
Evaluating the types of Long term memory through neuroimaging - involving HM and Clive Wearing
As both HM and Clive wearing had damaged hippocampi it seems likely that episodic memories are formed there as these are the memories they can no longer form. There is also evidence from brain scan studies that different types of memory are stored in different parts of the brain. For example, Tulving et al. (1994) got their participants to perform various memory tasks while their brains were scanned using a PET scanner. They found that the episodic and semantic memories were both recalled from an area of the brain known as the prefrontal cortex. This area is divided in two, one on each side (or hemisphere) of the brain. The left prefrontal cortex was involved in recalling semantic memories. Episodic memories were recalled from the right prefrontal cortex. Neuroimaging evidence adds scientific credibility to the theory as it is objective due to the use of computers that can show memories in a physical way.
Describe and outline proactive and retroactive interference as explanations for forgetting due to an absence of cues
Interference explains forgetting as two lots of information becoming confused in memory Proactive interference is where old learning affects recall of new information, for example not remembering your new girlfriends name as you keep thinking of the old girlfriends name. Retroactive interference is where new learning affects recall of/overwrites old information e.g. learning a new postcode and forgetting your old one. Interference is more likely to occur when the two pieces of information are similar Interference between memories makes it harder for us to locate memories, and this is experienced as ‘forgetting’. The impact of passage of time/intervening events are the cause forgetting.
How is proactive and retroactive interferences methodology and explanation tested?
Most research involves paired associate learning. Participants are presented with word pairs (like Cat-pear, Floor-badger, Window-glass…). They are then presented with a second list of paired words in which the first words are repeated but the second word is changed (like cat-radio, floor-jelly, window-spanner). When testing proactive interference the experimental group are given both lists to remember and the control group just remember the second list. Both groups are then asked to recall the second list and the experimental group do less well because the first list has interfered with their recall. When testing retroactive interference the experimental group are given both lists to remember and the control group just remember the first list. Both groups are then asked to recall the first list and the experimental group do less well because the second list is interfering with their recall.
Evaluation of proactive and retroactive interference through lab experiments (McGeoch and McDonald)
McGeoch & McDonald tested the hypothesis that words with the same meaning would be more prone to interference than information that was unrelated. Participants were given a list of words that they practised until they could recall with 100% accuracy. They were then given another list to remember, for the control group these words were very different but for the experimental group the words had similar meanings to those in the original list. When asked to recall the first list it was found that the experimental group were significantly impaired, down to only 12% (previously they had achieved 100% accurate recall). In contrast the control group learning unrelated information in the second list produced less impairment. They therefore concluded interference was most likely to occur when material being remembered was similar. However, this study involved learning word lists which is not an ecologically valid representation of everyday memory as information to recall in real life is often more complex. This means results cannot be fully generalised to real life.
Evaluation of proactive and retroactive interference through field experiments (Baddeley and Hitch)
Most evidence that interference can explain forgetting frequently comes from artificial laboratory experiments using artificial tasks, so interference may not occur to the same extent in more real-life settings and scenarios, so challenging interference as an explanation of forgetting. However everyday/real life situations have shown interference can explain forgetting. For example Baddeley & Hitch tested rugby players’ memory of the names of players they had played against in that season. Crucially, they had played in different numbers of games due to injury or being left out of the team. They found that players who had played in fewer games recalled proportionately more names than those who had played in the most games. The researchers attributed this to interference experienced by players who had had more team mates over the season. This offers externally valid evidence of the explanation to compliment the internally valid results of lab experiments.
Evaluation of proactive and retroactive interference through practical applications (revision strategies)
This explanation has practical applications. For example students revising for exams can develop more effective revision strategies as a result of this theory by making similar information more distinct by making notes on different colour paper in order to prevent interference.
Describe and outline retrieval failure due to an absence of cues
Tulving proposed the concept of ‘cues’, which are signposts or links to memories that make recall possible. If these cues are missing then information will be forgotten.
This states that forgetting is problem with accessibility and not availability. In other words the memory is in there somewhere but there are problems in finding it.
This has good ‘face validity’ (it makes sense to us based on our personal experiences), for example we have all experienced ‘tip of the tongue phenomenon’, we know the name of that band but can we can’t seem to get to the answer. If one aspect of the target word is remembered e.g. first letter, then eventually we will be able to access the whole memory.
This theory argues that information in LTM is only accessible if the appropriate cues are present, like the first letter or number of syllables. This means that information that has been originally well learnt, as we have made lots of connections to it, will be less likely to be forgotten.
This is known as the encoding specificity principle and is the idea that a cue or clue to a memory needs to be present when the material is learned (encoded) and again present when it is recalled or retrieved. Tulving (1979), found that the closer the cue to the target word the better our recall. This is how mnemonics and memory tricks often work.
Some psychologists believe that all forgetting is cue dependent. That is, we never really forget anything form our LTM, provided some form of cue is given we could remember anything.
Organisation cues: Missing things like categories which help to group items in memory will cause forgetting
Context cues: Having to try to remember in a different external environment to the one you experienced when you were learning the information would cause forgetting. State cues: Being in a different emotional or physical state to when you learned the info would cause forgetting.
Evaluating retrieval failure due to an absence of cues as an explanation of forgetting through research evidence (Tulving and Pearlstone)
Tulving and Pearlstone got participants to recall 48 words that belonged to one of 12 categories. As each word was presented it was preceded by its category to act as a cue (e.g. Gem: sapphire, Gem: diamond, Gem: ruby, etc…). If the cue was then present at recall (gem) then overall recall for the 48 words was 60%. If the cue was not present then recall fell to 40%. However it is not always possible in real life to categorise information we learn and so the practical applications may be limited.
Evaluating retrieval failure due to an absence of cues as an explanation of forgetting through research evidence (Godden and Baddeley)
To test the idea of context dependent cues Godden & Baddeley got divers to learn and recall word lists either on dry land or underwater. There were 4 groups: 1. Learn Underwater &recall underwater, 2. underwater/dry land, 3. dry land/underwater, 4. dry land/dry land. Results showed that words learnt and recalled in different contexts (physical environments) were more likley to be forgotten. However, Baddeley himself realised that this experiment is extreme; the difference between land and sea is big. He realised that in real life it is rare for contexts to be so different, as a result context probably plays a minimal role in aiding recall.
Evaluating retrieval failure due to an absence of cues as an explanation of forgetting through research evidence (Goodwin et al)
To test the idea of state dependent cues Goodwin et al, found that participants who hid money while drunk were more likely to recall the hiding place when in a similar drunk state, and would forget it when sober, supporting the idea of state-dependent cues being an aid to recall.
Evaluating retrieval failure due to an absence of cues as an explanation of forgetting through practical application (revision strategies)
This explanation has practical applications. For example students revising for exams can develop more effective revision strategies as a result of this theory by making cues for information when they are learning it and then using the cues to recall the information. They can also use context dependent recall by revising in an environment that is similar to the exam environment, such as at a desk in silence.
Describe and outline how misleading information affects eye witness testimony
A key factor in affecting the accuracy of recall is misleading information.
It has been stated that memories are not an accurate snapshot of an event, but are reconstructed over time, influenced by schemas. Schemas are ready-made expectations based on previous experiences, moods, existing knowledge, contexts, attitudes and stereotypes. We use schemas to interpret the world. They help fill in the gaps in our knowledge and simplify the processing of information. This has implications for the reliability of EWT, because witnesses are not merely recalling facts as they happened, instead they are reconstructing memories and these reconstructions are biased by schemas active at the time of recall.
Leading questions are questions that make it likely that a participant’s answer will be affected or influenced by the words in the question, or even by the way it is asked (tone of voice and body language). This is because words in the question can activate pre-existing knowledge and expectations (known as a ‘schema’) will influences people to give answers based on these expectations. For example, if we witness at night and from a distance a crime being committed by a figure dressed in black baggy clothes and wearing a balaclava we have no way of accurately knowing the gender of the criminal. If the police officer interviewing us the next day using questions such as ‘what did the man steal?’, in our mind we now remember the criminal as a man because this leading question has influenced our memory of the event by activating our pre-existing expectations that criminals are men.
Post-event discussion can also add misleading information after the incident has occurred and it changes our memory in the same way as leading questions. Post-event discussion refers to a conversation between co-witnesses or an interviewer and an eyewitness after a crime has taken place which may contaminate a witness’ memory for the event. When co-witnesses to a crime discuss it with each other (post-event discussion), their eyewitness testimonies may become contaminated. This is because they combine (mis)information from other witnesses with their own memories. Furthermore due to the desire to conform co-witnesses may reach a consensus view of what actually happened. This is very important in terms of EWT as often EWT can be primary evidence in law courts as Juries can find EWT of overriding importance in their deliberations, yet 75% of cases where people have been found by DNA evidence to have been wrongly convicted, the original guilty verdict was formed on the basis of inaccurate EWT.
research evidence into misleading information (Loftus and Palmer)
Loftus & Palmer investigated the effect of leading questions by asking American students to estimate the speed of cars in a film concerning traffic accidents. They found the estimated speed was affected by the verb in the question, smashed yielded the highest estimate at 41mph and contacted the lowest at 31mph. This suggests leading questions affected participants schemas, influencing them to give an answer linked to their previous experiences of the verb. They then conducted a further experiment to test the idea that a completely false memory could be implanted by using misleading information, they found that more participants mistakenly remembered a week later seeing non-existent broken glass if given the verb ‘smashed’ rather than ‘hit’ was used, suggesting that at recall, misleading information can therefore create a new memory, this is known as the ‘reconstructive hypothesis’.
Evaluation of research evidence into misleading information (Loftus and Palmer)
However the lab experiments by Loftus and Palmer can be criticised as they lack validity. They do not have good population validity as they were conducted using American students who are not representative of the general population as they may have a different standard of memory and so results cannot be generalised. They also lack ecological validity as the participants were in a lab environment watching the crash on a video – this would differ to real life, for example levels of emotion, and so cannot be applied to real life situations. It is also unclear as to if the results could be down to demand characteristics – where participants in lab experiments alter their behaviour as they have guessed the aim – in which case the results would lack validity.
Research into post event discussion in misleading information
Gabbert et al. investigated the effect of post-event discussion on the accuracy of eyewitness testimony. First with 60 students then with 60 older adults, finding similar results each time. Participants all individually watched a video of a girl stealing money from a wallet. The participants were then put into pairs to discuss the crime and were told that they had watched the same video, however they had in fact seen different perspectives of the same crime and only one of the pair had actually witnessed the girl stealing. All of the participants then completed a questionnaire, testing their memory of the event. 71% of the witnesses recalled information they had not actually seen and 60% said that the girl was guilty, despite the fact they had not seen her commit a crime.
Evaluation of research into post event discussion in misleading information
However the results have low ecological validity as the witnesses knew they were taking part in an experiment and were more likely to have paid close attention to the details of the video clip. Therefore, these results do not reflect
everyday examples of crime, where witnesses may be exposed to less information. A positive of the experiment is that Gabbert et al. tested two different populations, university students and older adults and found little difference between these two conditions. Therefore, the results provide good population validity and allow us to conclude that post-even discussion affects younger and older adults in a similar way.
Describe and outline how anxiety effects the accuracy of recall
Many researchers have looked at the relationship between anxiety and accuracy of recall because most crimes involve some level of anxiety as they are distressing.
· Some suggest anxiety can divert attention from important features of a situation whereas others suggest it improves recall as focus is applied in a serious situation.
· The Yerkes Dodson inverted U hypothesis can be used to explain this phenomenon. This hypothesis suggests that improved accuracy during recall is associated with moderate levels of arousal. If the eye witness experiences high levels or low levels of arousal at the scene of the incident, then the accuracy of their recall decreases, therefore an optimum arousal level is preferred.
Research into how anxiety effects the accuracy of recall (Johnson and Scott)
Johnson & Scott carried out a Lab experiment to investigate if anxiety during an incident affects the accuracy of later recall. Participants were split into two conditions and then were asked to wait outside the lab. In condition 1 Participants heard a low key discussion taking place in the lab about equipment failure and then a man emerged holding a pen and with grease on his hands. In condition 2 participants heard a heated and hostile exchange between people in the lab and also heard breaking glass. A man then emerged holding a knife covered in blood. When asked to identify the man from 50 photos, p’s in condition 1 were 49% accurate, compared with 33% accuracy in condition 2. These findings suggest that the weapon may have distracted attention from the person holding it caused by the anxiety induced be seeing it. This ‘weapon focus effect’ means that participants have accurate recall of the central detail of the crime such as the weapon, but a poor memory for the peripheral details.
Evaluation of research into how anxiety effects the accuracy of recall (Johnson and Scott)
This lab experiment can be credited for being a controlled environment where all other potential extraneous variables, such as distractions from the criminal or variations in what each witness can see, are controlled for which leads to high internal validity. However the experiment does not fully represent a real life EWT situation, as there was only the suggestion of a crime in the experimental condition, anxiety levels may not have been as high as in real life where witnesses may see someone get seriously hurt. This means the results lack ecological validity. Furthermore even though Johnson & Scott tried to prevent demand characteristics by making them think they were waiting for the experiment to start they may still have realised their behaviour was being measured and acted differently to the way they would in real life, for example they may have paid more attention to the man. Overall, this means the results lack both external and internal validity to some extent and cannot fully support the role of anxiety in affecting EWT accuracy.
Research into how anxiety effects the accuracy of recall (Yuille and Cutshall)
In contrast to Johnson and Scott, Yuille & Cutshall investigated the anxiety levels and accuracy of recall of witnesses to a real life fatal shooting. 13 agreed to be re-interviewed by the psychologists. The other 8, which included the victim, did not want to take part as it would involve reliving the trauma. The interviews took place 4-5 months after the crime. They found that even after the time delay of errors were rare and the accuracy remained high, because they asked lots of questions and allowed free recall the researchers actually gained more details overall than the police did. Furthermore misleading information which was deliberately given by the psychologists had little effect on the witnesses.
Evaluation of the research into how anxiety effects the accuracy of recall (Yuille and Cutshall)
This research is very high in ecological validity as it was based on interviews of witnesses to a real life crime and so is a good test of real levels of anxiety and its effect. However natural experiments have a lack of control and so extraneous variables may have affected the results. For example those who suffered highest anxiety were also closer to the crimes and this may explain their increased recall as they had a better view. Also it was a volunteer sample which is bias and lacks generalizability, perhaps the 8 who did not agree to be re-interviewed refused because they were made the most anxious and also had the poorest memories of the crime. The inconsistency with results from lab based and field based research mean that psychologists are unsure as to whether anxiety increases or decreases accuracy of recall.
Describe and outline the cognitive interview as a way to enhance the accuracy of eye witness testimony
One way of improving the accuracy of EWT’s is the cognitive interview (CI) developed by Fisher and Geiselman to improve recall in police interviews.
· The CI was based on Tulving’s idea that there are several retrieval paths to each memory and information not available through one technique may be accessible through another.
· The second principle behind the CI involves Tulving and Thomson’s Encoding Specificity Theory. This theory suggests that memory traces are made up of several features and that as many retrieval cues as possible should be used to enhance recall.
· The CI is different to a standard police interview which includes short, closed questions aimed at prompting specific facts, it may include interrupting the witness and it is very restricted.
· In contrast the CI takes a lot more time, involves open ended questions and aims to reduce eye witness anxiety.
· There are 4 main techniques/components/strategies used as part of the CI.
o Mental reinstatement of context involves encouraging the witness to return to both the environmental context (place, weather) and emotional context (feeling) of the crime scene. Recalling how you felt and the context enhances recall as these details act as cues to recall.
o Change of narrative order involves recounting the crime in a variety of different chronological orders, e.g. from the end to the beginning. When events are recalled in forward order, witnesses reconstruct based on their schemas. If the order is changed they are more accurate.
o Change of perspective involves recounting the scene from different perspectives, e.g. from the offenders point of view, this encourages many retrieval paths.
o And finally, report everything involves recalling all information, even if it appears to have little relevance or is accompanied with a lower level of confidence. Witnesses might not realise that some details are important and details might help them recall significant information.
Research into the cognitive interview (Kohnken et al conducted a meta-analysis)
There is some evidence which suggests that the CI might not be completely effective in improving the accuracy of eyewitness testimony. Kohnken et al conducted a meta-analysis (review/combining of results) involving nearly 2500 cognitive interviews and found an increase in correctly recalled items using the cognitive interview compared with standard interview techniques. However, they also found that the CI was more likely to increase the number of incorrect items too, perhaps because more detailed recall increases the chances of making mistakes. This means that although the CI is effective at producing more correct information it is not effective in reducing inaccurate information.
Research into the cognitive interview (Fisher et al field study)
Fisher et al conducted a field study to support the use of the CI in real police work. 16 experienced detectives were divided into two matched groups based on groups based on their supervisors’ evaluations and previous interviewing success. One of the two groups received training on the CI, whereas the other did not. Although the two groups were matched on skill before the CI training, the trained group of detectives went on to elicit 63% more information than the untrained group in subsequent EWT interviews. The use of a matched pairs design increases the internal validity as participant variables were controlled for – the differences in information gained is more likely to be due to the cognitive interview that the pre-existing skills of the detectives. This means the study has both ecological and internal validity and so offers strong support for the effectiveness of the CI
Research into the cognitive interview (Milne and Bull)
There is research evidence to support the idea that the CI provides more accurate recall of crime. Milne and Bull found that each individual technique used on its own produced more information than the standard police interview. However, Milne and Bull found that using a combination of report everything and context reinstatement produced better recall than any of the other techniques. However it is the case that in real life different police forces use different versions and different combinations of the main features depending on time, budget and training. This means that gathering data on how effective the CI is can be difficult as it is hard to make comparisons as different features are used.
What is a main flaw of the cognitive interview?
It is time consuming
A practical problem with the CI is that it is time consuming, often requiring more time and training than officers have operational time and budgets for. It has been found that many police officers do not use the CI technique in less serious crimes as they do not have the time to carry out this type of interview. In fact police often used strategies to deliberately limit an eyewitness report to the minimum amount of information deemed necessary by the officer. It is also the case that different police forces use different versions and different combinations of the main features. This means that gathering data on how effective the CI is can be difficult as it is hard to make comparisons.
Does the cognitive interview technique work on children?
No.
Young children seem to find the instructions confusing and find it hard to put themselves into another person’s perspective, as a result it produces less reliable recall than with standard police interviews. Geiselman recommends that the CI is only used on children aged eight and over.
