remembering
Oxford Companion to the Mind:
Top
remembering
If one asks teachers, students, or the proverbial man or woman in the street for the best available techniques for remembering something, the answers will be quite varied. However, one recurrent theme is sure to be: 'Repeat it!' A psychologist is likely to comment: 'Yes, but ... repetition by itself, mere repetition, does not help.' Yet the history of experimental investigations of memory is to a large extent concerned with mere repetition. In fact, the father of the experimental psychology of memory, Hermann Ebbinghaus, started his investigations in Germany in the last quarter of the 19th century by focusing almost exclusively on the effect of repetition. How many repetitions did it take to learn a list of words (or nonsense syllables)? How many trials were saved in relearning some list as a function of its prior repetition?
Both the common lore about repetition and the influence of Ebbinghaus dictated a preoccupation with the effect of repeated rehearsals. After all, it is well known that repeating a telephone number between looking it up and dialling it protects it from disappearance. And handsome, negatively accelerated learning curves resulted from numerous experiments that studied the effect of repetition on retention. But repetitive activities do not lead to effortless retrieval. Actually, when we try to remember an address, a name, the title of a book, or the plot of a play, we seem to engage in rather complicated search operations. The success of these operations depends not so much on how often we have repeated the required information in the past as on the proper embedding (the organization) of the target information within the larger flux of our knowledges and memories. When shopping for the weekend meals, we might retrieve the meats to be bought as a single memorial 'chunk', and liquid refreshments in another. Or another shopper might organize a mental shopping list by remembering what to buy in terms of what is where in the local supermarket. Both of these schema are kinds of organizations of the to-be-remembered things, and both require effort. Trying to recall the plot of a play, the rememberer might first recall vaguely the gist ('It was about a family who were always arguing') and then more and more details within coherent subdivisions ('Yes, there was the unhappy daughter and her pitiful suitor').
The notion that organization and structure are essential for memory retrieval is not novel. Extensive mnemonic techniques date at least to ancient Greece, where orators constructed complex spatial and temporal schema as an aid in rehearsing and properly presenting their speeches (see Yates 1966). In modern times the associationism of British empiricism and German experimentalism was seriously questioned during the first half of the 20th century by the Gestalt psychologists in general and by the British psychologist F. C. Bartlett in particular. Today we know in some detail what it is that repetition makes possible, what it is that is needed in addition to mere repetition.
A set of objects, events, or mental representations is said to be organized when consistent relations among the members of the set can be identified and specified. The result of such organization is called a structure. Structures may exist among events in the world as well as among mental events. A special kind of structure is the schema, which is a mental structure, specifically an organized representation of a body of knowledge. Thus, schema determine the expectations people have about events to be encountered, and about the spatial and temporal structure of those events.
The organization of to-be-remembered material takes time and conscious capacity. If we are told to remember a luncheon appointment while reading a book or watching our favourite television programme, conscious capacity is taken up by these primary activities and little organizational action will result. In order to remember the luncheon appointment we need to retrieve other plans (schema) about the specific day and 'fit in' the appointment. For example, we need to store such things as 'After the dentist, go to work, but go to the luncheon an hour later'. In the temporal organization of that day's plan, dentist, work, and luncheon will form an appropriate mental schema. And thinking about these plans (repeating them) will make their proper retrieval on the appointed day more likely. But again it is not mere repetition that provides a better schema, rather it is the anchoring of the relevant events within better, richer, and more accessible events that provides the more effective schema. Thus, each repetition provides an opportunity to relate the target event (the luncheon) to other events and thoughts. We may store the fact that our best friend will be at the luncheon, that it is held at a favourite restaurant, etc., etc. Each of these additions produces a more elaborate structure, and the more elaborate the retrieval opportunities for a target event the more likely it is that it will be recalled. Repetition provides opportunities for the organization of the to-be-remembered events.
While it is the case that most events are stored in long-term memory in complex, multistructured forms, certain frequently used structures can be identified. First there is the categorical or subordinate kind of structure in which a list of instances is stored under a general concept or label. To recall all the animals we know requires the use of such subordinate structures within a hierarchy of categories. Typically we gain access to some general animal category and then generate its subcategories such as domestic animals and catlike animals. Second there are coordinate structures of a few, usually less than five, events or things that are related to one another. Spatial structures, such as the directions of the compass, are one good example; another is the set of things called a table setting. Whereas in the categorical structure the higher-order label or node retrieves the lower instances, in the coordinate structure the members of the set act as retrieval points for one another. The third kind of structure is a serial or pro-ordinate structure in which a string of events is organized, usually in a temporal or spatial form. An excellent example is the way we retrieve the alphabet; another is the structure that represents the route we take to work from home. Parts of the serial string act as retrieval cues for subsequent things or events.
These idealized structures usually interact within any complex memorial event. More important, they are incorporated within the more general spatio-temporal schema mentioned earlier. Thus the understanding of a conversation involves the kinds of expectation inherent in our schema for social conversations, story schema tell us to look for crucial aspects and themes of a story, restaurants require that we have the proper schema for ordering from menus, talking to waiters, and so forth. The episodes of our daily lives are organized within such schema, which in turn incorporate the three kinds of structures described above.
Up to now the description of memory systems has focused on the recall of information. Another important kind of memory feat involves the recognition of previously encountered events. We are able to determine that people, rooms, foods, tunes are events that we have previously met, seen, tasted, heard. Not only do we know that we have encountered them before but we usually also know who or what they are. Conversely we sometimes know only that the event is familiar without knowing exactly who that person is, where we have seen that room before, what kind of food it is, what the name of a tune is. It is the latter phenomenon that has generally been studied by psychologists under the rubric of recognition.
The recognition of prior occurrence is a two-stage process involving two distinct mechanisms. One of them is a judgement of familiarity, the other a retrieval process essentially identical to that discussed for the recall of information. The judgement of familiarity is an automatic process, requiring no conscious effort and occurring as an immediate response to the event. However, the familiarity of information available of the event may be inadequate to make a confident judgement of prior occurrence. In that case a search process queries the long-term memory system whether the event in question is in fact retrievable. If such an attempt is successful then the event is considered to be 'old', i.e. having been previously encountered. Thus, recognition involves a judgement of familiarity which is supplemented by a retrieval attempt. For example, we meet someone who looks vaguely familiar, but the definite judgement that we 'know' that person is not made until we can recapture the place or context where we have previously encountered him or her.
The process of judging familiarity brings us back to the problem of repetition, because mere repetition does affect familiarity. The more frequently an event has been observed the more likely it is to be recognized on the basis of familiarity alone. Thus, repetition does have a function, but not for the retrieval of information. Repetition affects the process of integrating the representation of an event; it establishes its familiarity independent of its context or its relations to other mental contents.
Finally, errors of memory can obviously be of two kinds: retrieval errors and, less frequently, errors of familiarity judgements. Given the structural, schematic organization of memory storage, it is obvious that some events that 'fit' into the appropriate schema are likely to be retrieved even though they were not originally encountered. One might remember having witnessed an argument in a play because the structure of the play is stored under some general schema of 'family conflict', or one might 'recall' having seen a particular red armchair before, because it was stored as 'striking-looking furniture'. Thus, errors of memory are often even more instructive about the nature of mental structure than the normal recovery of information. See also déjà vu.
(Published 1987)
Both the common lore about repetition and the influence of Ebbinghaus dictated a preoccupation with the effect of repeated rehearsals. After all, it is well known that repeating a telephone number between looking it up and dialling it protects it from disappearance. And handsome, negatively accelerated learning curves resulted from numerous experiments that studied the effect of repetition on retention. But repetitive activities do not lead to effortless retrieval. Actually, when we try to remember an address, a name, the title of a book, or the plot of a play, we seem to engage in rather complicated search operations. The success of these operations depends not so much on how often we have repeated the required information in the past as on the proper embedding (the organization) of the target information within the larger flux of our knowledges and memories. When shopping for the weekend meals, we might retrieve the meats to be bought as a single memorial 'chunk', and liquid refreshments in another. Or another shopper might organize a mental shopping list by remembering what to buy in terms of what is where in the local supermarket. Both of these schema are kinds of organizations of the to-be-remembered things, and both require effort. Trying to recall the plot of a play, the rememberer might first recall vaguely the gist ('It was about a family who were always arguing') and then more and more details within coherent subdivisions ('Yes, there was the unhappy daughter and her pitiful suitor').
The notion that organization and structure are essential for memory retrieval is not novel. Extensive mnemonic techniques date at least to ancient Greece, where orators constructed complex spatial and temporal schema as an aid in rehearsing and properly presenting their speeches (see Yates 1966). In modern times the associationism of British empiricism and German experimentalism was seriously questioned during the first half of the 20th century by the Gestalt psychologists in general and by the British psychologist F. C. Bartlett in particular. Today we know in some detail what it is that repetition makes possible, what it is that is needed in addition to mere repetition.
A set of objects, events, or mental representations is said to be organized when consistent relations among the members of the set can be identified and specified. The result of such organization is called a structure. Structures may exist among events in the world as well as among mental events. A special kind of structure is the schema, which is a mental structure, specifically an organized representation of a body of knowledge. Thus, schema determine the expectations people have about events to be encountered, and about the spatial and temporal structure of those events.
The organization of to-be-remembered material takes time and conscious capacity. If we are told to remember a luncheon appointment while reading a book or watching our favourite television programme, conscious capacity is taken up by these primary activities and little organizational action will result. In order to remember the luncheon appointment we need to retrieve other plans (schema) about the specific day and 'fit in' the appointment. For example, we need to store such things as 'After the dentist, go to work, but go to the luncheon an hour later'. In the temporal organization of that day's plan, dentist, work, and luncheon will form an appropriate mental schema. And thinking about these plans (repeating them) will make their proper retrieval on the appointed day more likely. But again it is not mere repetition that provides a better schema, rather it is the anchoring of the relevant events within better, richer, and more accessible events that provides the more effective schema. Thus, each repetition provides an opportunity to relate the target event (the luncheon) to other events and thoughts. We may store the fact that our best friend will be at the luncheon, that it is held at a favourite restaurant, etc., etc. Each of these additions produces a more elaborate structure, and the more elaborate the retrieval opportunities for a target event the more likely it is that it will be recalled. Repetition provides opportunities for the organization of the to-be-remembered events.
While it is the case that most events are stored in long-term memory in complex, multistructured forms, certain frequently used structures can be identified. First there is the categorical or subordinate kind of structure in which a list of instances is stored under a general concept or label. To recall all the animals we know requires the use of such subordinate structures within a hierarchy of categories. Typically we gain access to some general animal category and then generate its subcategories such as domestic animals and catlike animals. Second there are coordinate structures of a few, usually less than five, events or things that are related to one another. Spatial structures, such as the directions of the compass, are one good example; another is the set of things called a table setting. Whereas in the categorical structure the higher-order label or node retrieves the lower instances, in the coordinate structure the members of the set act as retrieval points for one another. The third kind of structure is a serial or pro-ordinate structure in which a string of events is organized, usually in a temporal or spatial form. An excellent example is the way we retrieve the alphabet; another is the structure that represents the route we take to work from home. Parts of the serial string act as retrieval cues for subsequent things or events.
These idealized structures usually interact within any complex memorial event. More important, they are incorporated within the more general spatio-temporal schema mentioned earlier. Thus the understanding of a conversation involves the kinds of expectation inherent in our schema for social conversations, story schema tell us to look for crucial aspects and themes of a story, restaurants require that we have the proper schema for ordering from menus, talking to waiters, and so forth. The episodes of our daily lives are organized within such schema, which in turn incorporate the three kinds of structures described above.
Up to now the description of memory systems has focused on the recall of information. Another important kind of memory feat involves the recognition of previously encountered events. We are able to determine that people, rooms, foods, tunes are events that we have previously met, seen, tasted, heard. Not only do we know that we have encountered them before but we usually also know who or what they are. Conversely we sometimes know only that the event is familiar without knowing exactly who that person is, where we have seen that room before, what kind of food it is, what the name of a tune is. It is the latter phenomenon that has generally been studied by psychologists under the rubric of recognition.
The recognition of prior occurrence is a two-stage process involving two distinct mechanisms. One of them is a judgement of familiarity, the other a retrieval process essentially identical to that discussed for the recall of information. The judgement of familiarity is an automatic process, requiring no conscious effort and occurring as an immediate response to the event. However, the familiarity of information available of the event may be inadequate to make a confident judgement of prior occurrence. In that case a search process queries the long-term memory system whether the event in question is in fact retrievable. If such an attempt is successful then the event is considered to be 'old', i.e. having been previously encountered. Thus, recognition involves a judgement of familiarity which is supplemented by a retrieval attempt. For example, we meet someone who looks vaguely familiar, but the definite judgement that we 'know' that person is not made until we can recapture the place or context where we have previously encountered him or her.
The process of judging familiarity brings us back to the problem of repetition, because mere repetition does affect familiarity. The more frequently an event has been observed the more likely it is to be recognized on the basis of familiarity alone. Thus, repetition does have a function, but not for the retrieval of information. Repetition affects the process of integrating the representation of an event; it establishes its familiarity independent of its context or its relations to other mental contents.
Finally, errors of memory can obviously be of two kinds: retrieval errors and, less frequently, errors of familiarity judgements. Given the structural, schematic organization of memory storage, it is obvious that some events that 'fit' into the appropriate schema are likely to be retrieved even though they were not originally encountered. One might remember having witnessed an argument in a play because the structure of the play is stored under some general schema of 'family conflict', or one might 'recall' having seen a particular red armchair before, because it was stored as 'striking-looking furniture'. Thus, errors of memory are often even more instructive about the nature of mental structure than the normal recovery of information. See also déjà vu.
(Published 1987)
— George Mandler
- Bibliography
- Baddeley, A. D. (1976). The Psychology of Memory.
- Bartlett, F. C. (1932). Remembering.
- Mandler, G. (1985). Cognitive Psychology.
- Norman, D. A. (1976). Memory and Attention (2nd edn.).
- Yates, F. A. (1966). The Art of Memory.
Post a question - any question - to the WikiAnswers community:
Copyrights:
Oxford Companion to the Mind
The Oxford Companion to the Mind. Second Edition. Copyright © Oxford University Press, 2004. All rights reserved.
Read more