Psychology: Memory and Attachments

                          Short-Term Memory          Long-Term Memory

Duration             Measured in seconds                Measured in hours, days

                                        and minutes                                     and years

Capacity              Less than 7 chunks                    Potentially Unlimited

Encoding               Acoustic or visual                              Semantic

Short-term memories don't last very long. Rehearsal helps us remember information in the short term memory, and if it's rehearsed it can transfer to long-term memory. 

Long-term memories last anywhere from 2 hours to 100 years plus, so anything that isn't short-term.

Case Study: Peterson & Peterson's Trigram Study

Peterson and Peterson carried out a study to test the duration of short-term memory. The experimenter reads out a trigram made up of three consonants, which were made to have no meaning. It is then followed by a three-digit number, and participants were asked to count down in 3s or 4s for 3, 6, 9, 12, 15 or 18 seconds. This was to stop them rehearsing the trigram. When asked to recall the trigram, participants remembered 90% after a 3-second interval, and only 2% after an 18-second interval. This suggested that short-term memory only lasts for a meximum of 20 seconds. However, critics of this study say that counting down displaces the information. They also believe participants are more likely to remember because they know they are going to be tested on the trigram so are relying on more than short-term memory. Finally, another study found the short term memory can last for up to 96 seconds.

Capacity of short-term memory - Miller suggested 'the magic number seven, plus or minus two'. This means the span of immediate memory is seven chunks of information, which makes it easier to remember. However, the size of the chunk does matter. Jacobs carried out a digit span study, where participants were asked to recall a number of digits as it steadily increases. He carried out the same study with letters, and found the average correctly recalled for digits was 9.3 items, and 7.3 letters.

Encoding in STM and LTM - Baddeley tested the effects of acoustic and semantic similarity on short- and long-term recall. He gave participants a list of words which were acoustically similar and dissimilar and semantically similar and dissimilar. They were asked to recall the list after a short period of time, and encountered acoustic confusion, meaning they were having trouble recalling the acoustically similar list. After a long period of time, participants encountered semantic confusion, meaning they had trouble recalling the semantically similar list. This means short-term memory encodes acoustically and long-term memory encodes semantically.

However, other studies have found that although STM encodes mainly acoustically and LTM encodes mainly semantically, they both also encode visually.

Componants of the multi-store model

• Sensory Memory - this is composed of several stores - the eyes, ears, nose, etc, and the corresponding areas of the brain. The sensory stores constantly receive information, but most does not receive attention. If it does receive attention, the data is transferred to short-term memory.
• Short-term Memory - information in the STM is in a fragile state and it will decay quickly unless it is rehearsed. It will also disappear if new information enters STM, displacing the original information. This happens because of the limited capacity.
• Long-term Memory - transferring information from STM to LTM is due to rehearsal. First, rehearsal stores information in STM, but the more it is rehearsed, the more lasting the memory will be. This maintenence rehearsal is mainly verbal.

Glanzer and Cunitz's Primacy-Recency Effect

They gave participants a list of 20 words. When asked to recall the list, particiapnts were better at recalling the words at the beginning and end of the list, and were worse at remembering those in the middle. This is because those at the start were rehearsed and therefor transferred to LTM, whereas those at the end were still being stored in STM.

Strengths

• There is strong evidence of three different stores, backing the basis of the MSM.
• It has clear predictions about memory which makes it easy for psychologists to test. 
• It provides an account of memory in terms of both sturcture and process.; the structures are the three stores, and the processes are attention and verbal rehearsal.

Limitations

• It oversimplifies the memories processes and structures.
• It suggests  that STM and LTM are initery stores, whereas evidence suggests that this is not true. Case Study: KF suffered brain damage which resulted in difficulty processing verbal information in STM, but a normal ability to process visual information. This suggests STM is not a single store. 
• Schachter et al suggested four long-term stores: semantic (knowledge about words etc), episodic (yesterday or last week), procedural (how to read etc) and perceptual-representation system (enhanced recognition of specific stimuli).
• Research has shown that maintenence rehearsal is the way memories are created and that enduring memories are created by the processig that you do.

Baddely and Hitch suggested that short-term memory is a number of different stores rather than just one. The components of the Working Memory Model:

• Central Executive - it has limited capacity. It directs attention to particular tasks, determining how 'resources' are allocated to tasks. The 'resources' are the three slave systems listed below.
• Phonological Loop - this too has a limited capacity. This deals with auditory information. Baddeley subdivided this loop into the phonological store and an articulatory process. The phonological store words you hear. The articulatory process is used for words that are heard or seen. These words are then looped, as a form of maintenance rehearsal.
• Visuo-Spatial Sketchpad - this is used for spatial tasks and visual information is temporarily stored here. Visual information is what things look likes and spatial information is the relationship between things.
• Episodic Buffer - this was added by Baddeley in 2000 because the model needed a general store. The visuo-spatial sketchpad and phonological loop deal with specific type of information, and the central executive has no storage capacity, meaning there was nowhere to store information that relates to both visual and acoustic information. It also has limited capacity, but integrates information from the central executive, phonological loop, visuo-spatial sketchpad and long-term memory.

Evidence

• The phonological loop explains the word-length effect. People cope better with shorted words - this can be put down to the fact the phonological loop can hold the amount of information you can say in 2 seconds, making it hard to remember and rehearse long words.
• Baddeley et al found that, when participants were shown words and then asked to immediately recall, their performance was much better when asked to recall sentences rather than unrelated words. This supports the idea of an immediate memory store for items that are neither visual nor phonological.
• KF shows that STM works indepentently of LTM, as he has no problem with long-term learning but some aspects of his short-term memory are impaired. His brain damage seemed to be restricted to the phonological loop because he had problems remembering auditory information whereas his ability of remembering visual information wasn't as limited. 

Strengths

• The model explains the word-length effect and the partial short-term memory difficulties experienced by KF and others.
• It offers a better account than the STM store in the multi-store model because it isn't being described as a unitary store but as one with many components. 
• In comparison with the MSM, the working memory model includes verbal rehearsal as an optional process rather than the only way information is kept in the immediate memory. It also amphasises process more than the MSM, which emphasised the structure. The working memory model also suggests that immediate memory holds the most recently activated portion of long-term memory.

Limitations

• What exactly is the central executive? It appears it is the same as 'attention'. Some psychologists feel it is too vague.
• Some of th ekey evidence for the working memory moodel comes from cases of brain damage, However, you cannot make 'before and after' comparisons, so it is not clear whether changes in behavious are caused by the damage.  The process of the injury is traumatic, which may in itself change behaviour.

Loftus and Palmer studied whether misleading information distored recall.

Experiment 1

• 45 students were shown 7 films of a car accident. They were split into groups and were asked a set of questions. There were asked: 'how fast were the cars going when they hit each other?'. Each group was given the question with a different verb, including smashed, collided, bumped or contacted in place of the word hit.
• The speed estimate was calculated for each group. 'Smashed' estimated the highest speed at around 41mph, with contacted at the lowest speed at around 30mph.

Experiment 2

• 3 groups were shown a one-minute film of a car accident. Group 1 was asked about the speed when the cars 'smashed', group 2 were given 'hit' and group 3 were not asked a question about speed. They were then asked to return one week later and were asked a series of questions, including whether they saw broken glass.
• Findings showed that participants that gave a higher estimate (with 'smashed') were more likely to think that they saw broken glass, suggesting that misleading post-event information does change the way information is stored.

Anxiety

• Negative effect - Deffenbacher et al carried out a meta-analysis of 18 studies, looking at the effects of heightened anxiety on accuracy of eyewitness recall. It was clear that there is support for the hypothesis that high levels of stress negatively impacted on the accuracy of eyewitness memory.
• Positive effect - some studies found the emotional arousal actually enhance accuracy of memory, as found when Christianson and Hubinette found when interviewing witnesses of a bank robbery that those witnesses who had been threatened were more accurate in their recall and remembered more details than onlookers.
• The weapon-focus effect - there is evidence that in violent crimes, arousal may focus the witness on more central details of the attack, than the more peripheral details. Loftus et al used two conditions, once involving a weapon and one not. Participants heard a conversation in a room next door. In one condition, a man emerged holing a pen, and in the other a man emerged holding a paper knife covered in blood.  When asked to identify the man from 50 photos, participants in condition 1 were 49% accurate, compared with the other where they were 33% accurate, suggesting the weapon may have distracted attention from the person.

Explaining the contradiction

• Deffenbacher suggests the contradiction in research can be explained by the Yerkes-Dodson Law, which states that performance improves up to an optimum point where it then starts to decrease.
• The studies which found an increase of accuracy (or performance) with an increase of anxiety or stress, were dealing with the left half of the graph, and the studies that found a decrease of performance with an increase of stress were dealing with the right side of the graph.

Research Supports the Weapon-Focus Effect

A meta-analysis of studies on the weapon-focus effect shows that presence of a weapon does decrease the accuracy of recall. Loftus et al monitored eyewitness' eye movement and found that presence of a weapon causes eyes to physically be drawn towards the weapon itself, and away from other details, such as the person's face.