View sample Cognitive Psychology of Short-Term Memory Research Paper. Browse other research paper examples and check the list of research paper topics for more inspiration. If you need a religion research paper written according to all the academic standards, you can always turn to our experienced writers for help. This is how your paper can get an A! Feel free to contact our custom writing services for professional assistance. We offer high-quality assignments for reasonable rates.
The concept of short-term memory has been of theoretical signiﬁcance to cognitive psychology since the late 1950s. Some investigators have even argued that ‘all the work of memory is in the short-term system’ (Shiﬀrin 1999, p. 21). However, others have claimed that short-term memory is an archaic concept and there is no need to distinguish the processes involved in short-term memory from other memory processes (Crowder 1993).
1. Deﬁnitions And Terminology
‘Short-term memory’ refers to memory over a short time interval, usually 30 s or less. Another term for the same concept is ‘immediate memory.’ Both these terms have been distinguished from the related terms ‘short-term store’ and ‘primary memory,’ each of which refers to a hypothetical temporary memory system. However, the term ‘short-term memory’ has also been used by many authors to refer to the temporary memory system. Thus, here the term ‘short-term memory’ is used in both senses.
It is important to distinguish ‘short-term memory’ from the related concepts ‘working memory’ and ‘sensory memory’. Some authors have used the terms ‘short-term memory’ and ‘working memory’ as synonymous, and indeed the term ‘working memory’ has been gradually replacing the term ‘short-term memory’ in the literature (and some authors now refer to ‘short-term working memory’; see Estes 1999). However, ‘working memory’ was originally adopted to convey the idea that active processing as well as passive storage is involved in temporary memory. ‘Sensory memory’ refers to memory that is even shorter in duration than short-term memory. Further, sensory memory reﬂects the original sensation or perception of a stimulus and is speciﬁc to the modality in which the stimulus was presented, whereas information in short-term memory has been coded so that it is in a format diﬀerent from that originally perceived.
2. Historical Development And Empirical Observations
An initial description of short-term memory was given by James (1890), who used the term ‘primary memory’ and described it as that which is held momentarily in consciousness. The intense study of short-term memory began almost 70 years later with the development of the distractor paradigm (Brown 1958, Peterson and Peterson 1959).
2.1 The Distractor Paradigm
In this paradigm, a short list of items (usually ﬁve or fewer) is presented to subjects for study. If the subjects recall the list immediately, perfect performance results because the list falls within their memory span. However, before the subjects recall the items, they engage in an interpolated activity, the distractor task, which usually involves counting or responding to irrelevant material. The purpose of the distractor task is to prevent the subjects from rehearsing the list items. The length of the distractor task varies, and its duration is called the ‘retention interval.’ By comparing performance after various retention intervals, it was found that the rate of forgetting information from short-term memory is very rapid, so that after less than 30 s little information remains about the list of items.
Another central ﬁnding from the distractor paradigm involves the serial position curve, which reveals accuracy for each item in the list as a function of its position. The curve is typically bowed and symmetrical with higher accuracy for the initial and ﬁnal positions than for the middle positions. The distractor task typically requires serial recall (i.e., recall in the order in which the items were presented; see Representation of Serial Order, Cognitive Psychology of). Two types of errors occur with this procedure. Transposition errors are order errors in which subjects recall an item in the wrong position. Nontransposition errors are substitution errors in which subjects replace an item with one not included in the list. For example, if subjects receive the list BKFH and recall VKBH, they make a nontransposition error in the ﬁrst position and a transposition error in the third. The bowed serial position curve reﬂects the pattern of transposition errors. The pattern of nontransposition errors shows instead a relatively ﬂat function with errors increasing slightly across the list positions.
Errors in the distractor paradigm can also be classiﬁed by examining the relation between the correct item and the item that replaces it. In the example, the subjects replace B with V. Because those two letters have similar-sounding names, the error is called an ‘acoustic confusion error’, and such errors occur often in the distractor paradigm. Even visually presented items are typically coded in short-term memory in an acoustic or speech representation. This fact was ﬁrst illustrated in an experiment in which subjects were shown a list of six consonants for immediate serial recall. The errors were classiﬁed by a confusion matrix, in which the columns indicate the letter presented and the rows indicate the letter actually recalled. A high correlation was found between the confusion matrix resulting from this memory task when subjects recalled a list of visually presented letters, and the confusion matrix resulting from a listening task when subjects heard one letter at a time and simply named it with no memory requirement (Conrad 1964).
2.2 The Free Recall Task
Another paradigm commonly used in the early investigation of short-term memory was the free recall task, in which subjects are given a relatively long list of items and then recall them in any order they choose. When the list is recalled immediately, the subjects show greater memory for the most recent items. This ‘recency eﬀect’ was attributed to the fact that the ﬁnal items in the list, but not the earlier ones, are still in short-term memory at the termination of the list presentation. Support for this conclusion came from the observation that if presentation of the list is followed by a distractor task, then there is no recency eﬀect, although as in the case of immediate recall there is a ‘primacy eﬀect,’ or advantage for the initial items in the list. The explanation oﬀered for the elimination of the recency eﬀect with the distractor task is that the ﬁnal list items are no longer in short-term memory after the distractor activity. Further support for this explanation came from ﬁnding that other variables like list length and presentation rate had diﬀerential eﬀects on the recency and earlier sections of the serial position curve. Speciﬁcally, subjects are less likely to recall an item when it occurs in a longer list for all except the most recent list positions. Likewise, subjects are less likely to recall an item in a list presented at a faster rate for all but the recency part of the serial position curve.
3. Theoretical Accounts
The most widely accepted account of short-term memory was presented in the 1960s in what was subsequently termed the ‘modal model’ because of its popularity. The core assumption of that model is the distinction between short-term memory, which is transient, and long-term memory, which is relatively permanent. The fullest description of the modal model was provided by Atkinson and Shiﬀrin (1968), who also distinguished between short-term memory and sensory memory.
3.1 The Buﬀer Model
Atkinson and Shiﬀrin’s ‘buﬀer model,’ as it has been called, is characterized along two dimensions (Atkinson and Shiﬀrin 1968). Following a computer analogy, the ﬁrst dimension involves the structural features of the system, analogous to computer hardware, and the second dimension involves the ‘control processes,’ or operations under the control of the subjects, analogous to computer software. The structural features of the buﬀer model include the sensory registers (with diﬀerent registers for each sense), short-term store, and long-term store. The control process emphasized is rote rehearsal, which takes place in part of short-term store called the ‘buﬀer.’ The rehearsal buﬀer has a small capacity with a ﬁxed number of slots (about four).
To account for performance in the free recall task, it is assumed that each item enters the buﬀer and when the buﬀer is full the newest item displaces a randomly selected older item. While an item is in the buﬀer, information about it is transferred to long-term store, with the amount of information transferred a linear function of the time spent in the buﬀer. Although information about an item thereby gets transferred to long-term store, the item remains in the buﬀer until it is displaced by an incoming item. At test, subjects ﬁrst respond with any items in the buﬀer and then try to retrieve other items from long-term store, with the number of retrieval attempts ﬁxed. These assumptions allow the model to account for the various empirical observations found with the free recall task. The model accounts for the recency eﬀect and its elimination with a distractor task because the ﬁnal items are still in the buﬀer immediately after list presentation but get displaced from the buﬀer by the interpolated material of the distractor task. The model accounts for the primacy eﬀect by assuming that the buﬀer starts out empty so the initial items reside in the buﬀer longer than subsequent items because they are not subject to displacement until the buﬀer is full. The eﬀect of list length is due to the ﬁxed number of attempts to retrieve information from long-term store. The longer the list, the smaller is the likelihood of ﬁnding a particular item. The eﬀect of presentation rate is due to the linear function for transferring information from short-term to long-term store. More information is transferred when the rate is slower so that retrieving an item from long-term store is more likely at a slower rate.
There have been numerous reﬁnements and expansions of the buﬀer model since it was ﬁrst proposed. These updated versions have been termed ‘SAM’ (search of associative memory) and ‘REM’ (retrieving eﬀectively from memory). However, these reﬁnements have been focused on the search processes in long-term memory, and the short-term memory component of the model remains largely intact (Shiﬀrin 1999).
3.2 The Perturbation Model
Whereas the buﬀer model explains the results of the free recall task, a popular model proposed by Estes (1972) explains the results of the distractor paradigm. According to this ‘perturbation model’, the representation in memory of each item in a list is associated with a representation of the experimental context. This contextual representation is known as a ‘control element.’ It is assumed that if the control element is activated, then the representations of the items associated with it are activated in turn, allowing for their recall. Forgetting is attributed in part to the fact that the context shifts with time so that subjects may be unable to activate the appropriate control element after a delay. Reactivation of the item representations by the control element does not only occur at the time of test. In addition, there is a reverberatory loop providing a periodic recurrent reactivation of each item’s representation, with the diﬀerence in reactivation times for the various items reﬂecting the diﬀerence in their initial presentation times. This reverberatory activity provides the basis for the short-term memory of the order of the items in a list. Because of random error in the reactivation process, timing perturbations result, and these perturbations may be large enough to span the interval separating item representations, thereby leading to interchanges in the order of item reactivations and, hence, transposition errors during recall. Because such interchanges can occur in either the forward or backward direction for middle items in the list but in only one direction for the ﬁrst and last items, the bowed serial position function for transposition errors is predicted. The symmetry in the serial position function is predicted with the assumption that timing perturbations do not start until all the list items have been presented. Because interchanges can also occur between the list items and the interpolated distractor items, the gradually increasing proportion of nontransposition errors is also predicted.
The original version of the perturbation model included only the perturbation process responsible for short-term memory, but subsequent research documented the need to include a long-term memory process in addition to the short-term perturbation process (Healy and Cunningham 1999). Other extensions allowed the perturbation model to account for a wide range of ﬁndings in the distractor paradigm and also to provide insights into other memory processes such as those responsible for memory distortions in eyewitness situations (Estes 1999; see also Eyewitness Memory: Psychological Aspects).
3.3 Theoretical Controversies And Complications
Although models of memory typically include the distinction between short and long-term memory, some investigators have pointed to problems with some of the evidence establishing the need to postulate a distinct short-term memory. For example, Craik and Lockhart (1972) argued that short-term memory cannot be distinguished by the exclusive use of speech coding, as suggested by Conrad (1964). Craik and Lockhart proposed an alternative framework called ‘levels of processing,’ according to which information is encoded at diﬀerent levels and the level of processing determines the subsequent rate of forgetting.
More recent arguments against the need for a separate short-term memory were made by Crowder (1993), who pointed out that the rapid forgetting across retention intervals in the distractor paradigm is not found on the ﬁrst trial of an experiment. He also pointed out that a recency eﬀect like that in immediate free recall is found in a number of tasks relying exclusively on long-term memory. Healy and McNamara (1996) dismissed some of these arguments with two general considerations. First, information can be rapidly encoded in long-term memory, so that, for example, recall can derive from long-term memory rather than short-term memory on the ﬁrst trial of an experiment before interference from previous trials has degraded the long-term memory representation. Second, although recency eﬀects occur in many memory paradigms, the speciﬁc properties and causes of the serial position functions diﬀer across paradigms.
Memory models also typically make the distinction between sensory and short-term memory. However, an important exception is Nairne’s ‘feature model’ (Nairne 1990). Instead of distinguishing between these two types of memory stores or processes, the feature model distinguishes between two types of memory trace features—modality independent (which involve speech coding) and modality dependent (which are perceptual but not sensory). According to this model, during recall subjects compare the features of the memory trace to the features of various item candidates. Forgetting occurs as the result of feature overwriting, by which a new feature can overwrite an old feature but only if the two features are of the same type (modality dependent or modality independent).
Although short-term memory is distinguished from both sensory and long-term memory by most contemporary models of memory, including a recent ‘composite model’ proposed by Estes (1999) to describe the current state of theorizing, many current models have broken down short-term memory into diﬀerent subcomponent processes. The most popular of these models is Baddeley’s ‘working memory model’ (Baddeley 1992), which includes an attentional control system called the ‘central executive’ and two ‘slave systems,’ the phonological loop (responsible for speech coding) and the visuospatial sketchpad (responsible for the coding of both spatial and visual information). Recent neuropsychological evidence (Martin and Romani 1994) has led to a further breakdown into semantic and syntactic components in addition to the components of Baddeley’s system.
Despite these controversies and complications, it seems clear that the concept of ‘short-term memory’ will continue in the future to play an important role in our theoretical understanding of cognitive processes.
- Atkinson R C, Shiﬀrin R M 1968 Human memory: A proposed system and its control processes. In: Spence K W, Spence J T (eds.) The Psychology of Learning and Motivation: Advances in Research and Theory 2. Academic Press, San Diego, pp. 89–195
- Baddeley A-A. 1992 Is working memory working? The ﬁfteenth Bartlett lecture. Quarterly Journal of Experimental Psychology 44: 1–31
- Brown J 1958 Some tests of the decay theory of immediate memory. Quarterly Journal of Experimental Psychology 10: 12–21
- Conrad R 1964 Acoustic confusions in immediate memory. British Journal of Psychology 55: 75–84
- Craik F I M, Lockhart R S 1972 Levels of processing: A framework for memory research. Journal of Verb Learning and Verb Behavior 11: 671–84
- Crowder R G 1993 Short-term memory: Where do we stand?. Memory and Cognition 21: 142–45
- Estes W K 1972 An associative basis for coding and organization in memory. In: Melton A W, Martin E (eds.) Coding Processes in Human Memory. Winston, Washington, DC, pp. 161–90
- Estes W K 1999 Models of human memory: A 30-year retrospective. In: Izawa C (ed.) On Human Memory: Evolution, Progress, and Reﬂections on the 30th Anniversary of the Atkinson–Shiﬀrin Model. Erlbaum, Mahwah, NJ, pp. 59–87
- Healy A F, Cunningham T C 1999 Recall of order information: Evidence requiring a dual-storage memory model. In: Izawa C (ed.) On Human Memory: Evolution, Progress, and Reﬂections on the 30th Anniversary of the Atkinson–Shiﬀrin Model. Erlbaum, Mahwah, NJ, pp. 151–64
- Healy A F, McNamara D S 1996 Verbal learning and memory: Does the modal model still work? Annual Review of Psychology 47: 143–72
- James W 1890 The Principles of Psychology. Holt, New York
- Martin R C, Romani C 1994 Verbal working memory and sentence comprehension: A multiple-components view. Neuropsychology 8: 506–23
- Nairne J S 1990 A feature model of immediate memory. Memory and Cognition 18: 251–69
- Peterson L R, Peterson M J 1959 Short-term retention of individual verbal items. Journal of Experimental Psychology 58: 193–8
- Shiﬀrin R M 1999/30 years of memory. In: Izawa C (ed.) On Human Memory: Evolution, Progress, and Reﬂections on the 30th Anniversary of the Atkinson–Shiﬀrin Model. Erlbaum, Mahwah, NJ, pp. 17–33