Please enable JavaScript.

Coggle requires JavaScript to display documents.

Are there separate short-and long-term memory systems? (Neuropsychological…

- - - - Recently encountered items are better remembered than earlier ones, apparently provides good evidence for a distinction between STM and LTM: if there were a short-term store, it should boost the recall of recently presented items more than earlier ones. Pre-recency items are supposedly only recallable due to LTM
        
        Glanzer and Cunitz (1966) conducted an experiment to explore this, whereby participants were presented with lists of 15 words, one every 3 seconds, to freely recall. They observed a strong recency effect. Glanzer and Cunitz also observed that introducing a delay after the last item would reduce the recency (but not primacy) effect, whereas introducing an increase in presentation rate would increase the primacy (but not recency) effect. This double dissociation supports the notion of a distinct short-term store. Furthermore, their results support the theory that recall of pre-recency items relies on long-term memory and thus is a function of the extent to which they were rehearsed.
        
        Murdock (1962) has reported that in free recall tasks you often see primacy and recency effects, with middle items being recalled least well. This is what you would expect if pre-recency items were recalled from long-term memory as it reflects the ability to rehearse the items.
        
        Rundus (1971) tested this effect of rehearsal, finding a correlation between the number of times a word was rehearsed and the probability of it being recalled, for all items except the last. Better recall for the first rather than the middle items on the list (‘primacy effect’) can therefore be explained by the fact that the first items get rehearsed more times on average than do middle items on the list. At the beginning the participant only has one item to rehearse and more time to rehearse it, however after the presentation of the 10th word, the participant has much less time to rehearse it if the previous words are not to be forgotten. These extra rehearsals earlier on increase the probability that a word is stored in LTM. That rehearsal does not affect recall of the last items supports the idea that these are being recalled directly from STM rather than from LTM.
      - The explanation for recency effects on free recall tasks has however been questioned due to the discovery of long-term recency effects. Tzeng (1973) conducted a free recall task with a 20 second distraction between the presentation of each item and recall. He found that a recency effect persisted, which is in contrast with explanations for the role of a short-term memory store in the recency effect (e.g. Glanzer & Cunitz, 1966; Rundus, 1971). #
        
        If the recency effects observed were a result of later items’ accessibility in the short-term memory store, then a delay should decrease the ability to recall these items since they would have been displaced from the short-term memory store by the time of recall. The determinant of recall should therefore reflect rehearsal, which Rundus (1971) demonstrated to decrease as more items are presented. Consequently, recency in this case cannot be attributed to a distinct STM store.
      - Koppenaal and Glanzer (1990) have however attempted to explain the reason for observed long-term recency effects. They argue that receiving the same distractor task after every item means that participants learned to alternate between practicing the distractor task and rehearsing items, allowing the short-term memory store to carry out its normal functions. Additionally, if the distractor task is changed, the ability to alternate between the task and rehearsing items should be disrupted, and the long-term recency effects should disappear. This was supported as they found that if the final distractor task was changed, recency effects did not occur.
        
        Alternatively, Neath (1993) proposed that processing of items doesn’t end immediately after the distractor task begins. Instead, the two co-occur and the distractor task becomes a contextual retrieval cue. He claims that the type of distractor task serves as a contextual retrieval cue for the item, and so changing the distractor task for the last item means the item will be associated with a different context than the other items. Consequently, this will be less likely recalled if the main distractor task is used as a contextual retrieval cue. Following this, if a different distractor task is used after every item, then participants will use more than one distractor task as a retrieval cue. The last item will hence not be disrupted and long-term recency effects should reappear. Their study following this procedure supported this.
- - - - One initial problem with the SIMPLE model was the lack of rehearsal mechanism, causing difficulty explaining recall data in which rehearsal plays a significant role.
        
        However, Tan and Ward (2000) proposed that rehearsal changes the effective time at which list events happen – the time at which items are last brought to mind before recall. In their study, participants rehearsed aloud, permitting Tan and Ward to re-plot their recall data in terms of the point at which each item was last rehearsed during inter-stimulus intervals in the list. When fitted to these data, SIMPLE very accurately reproduced the observed recall curve. Their results demonstrated that only how recently the item had been rehearsed directly affected recall; presentation rate did not. Slower presentation rate gave participants more times to rehearse early-list items later in the list. Therefore, presentation rate affects pre-recency but not recency because a slower rate permits participants to rehearse early list items later in the list, resulting in different final rehearsal times for early list items. This explains Rundus’ (1971) findings that rehearsal benefits pre-recency but not recency items.
- - - - However, Keppel & Underwood (1962) employed this task, finding no differential forgetting (recall was the same for all items) on the first recall trial regardless of whether the item had been retained for three or 18 seconds. Differential forgetting only occurred on subsequent trials. This provides evidence against decay of information retained in STM, as recall would have been worse for items retained for longer. Differential forgetting only occurred on later trials in the study, due to a build-up of interference from information from previous trials; this type of interference is called proactive interference (PI).
        
        Wickens (1970; 1973) tested the PI hypothesis, predicting that if there is similar information to recall on each trial, PI should increase over trials. Hence, a change in the type of information used each trial should prevent PI occurrence. In their study they used the same type of information for the first three trials, and subsequently switched to a different type of information for the fourth. They found that PI occurred on the second and third trials, but not on the fourth, demonstrating a ‘release of PI’. This supported their hypothesis and hence the notion that STM and LTM comprise a unitary system.
        
        Instead of isolating STM and its processes, the Brown-Peterson paradigm can be seen to temporally magnify the retention of a specific item as if it were under a microscope (Crowder & Neath, 1991). This microscope analogy can be employed to propose the idea that the same principles apply to STM as to LTM. Crowder and Neath (1991) utilised the continuous-distractor paradigm to study temporal distinctiveness, conducting two parallel studies that explored both STM and LTM. In the both studies they found that the recency effect was largest in the condition in which the distinctiveness of the recency items was greatest and smallest in the condition in which the distinctiveness of the primacy items was greatest. Thus, the SIMPLE model was able to account for the results from the Brown-Peterson paradigm.
- - - - However, upon observing her rehearsal strategy, they realized it was abnormal: there was little evidence of rehearsing earlier items from the list at later points in time. This contrasts greatly with controls, who rehearsed early items throughout item presentation and consequently changed the effective time at which items occurred. This suggests that differences in rehearsal pattern could potentially explain amnesiacs’ selectively impaired recall of pre-recency items, since they seem to rehearse items in a way that harms pre-recency items more than recency items.
        
        Dewar et al. further investigated this idea, by introducing experimenter-imposed rehearsal: in between presentations of list words, participants read aloud words from earlier in the list, mimicking a typical rehearsal pattern. This design ensured that D.A. would rehearse earlier list items later in the list. They found that D.A. no longer showed highly suppressed recall of pre-recency items, and that her performance was slightly suppressed across all items, illustrating a universal deficit in her ability to make temporal discriminations. This supports the hypothesis that amnesiacs’ unusual rehearsal technique, rather than selectively impaired LTM, is responsible for the results.
        
        Race et al. (2013) corroborated these data demonstrating the impairment of STM and LTM in amnesiacs, finding that their ability their ability to remember a single face over intervals of either 8 or 15 seconds was impaired. Hence, the results are consistent with the SIMPLE model and the idea of a unitary memory system.
  - - - This suggests that brain areas that are implicated in LTM are also implicated in STM, supporting the idea that the two memory systems are not distinct. This activity may represent encoding and retrieval??? LOOK AT NEURO