Cognitive relevance of schemas
The first research of this kind was started by Frederick Bartlett in the thirties of the last century (Bartlett, 1932). In his tests, Bartlett used the following procedure. Subjects were given material to read (usually a short story) and then asked to reproduce it after various time intervals. Bartlett noticed that the respondents were not successful in repeating the content verbatim, but rather in reproducing its meaning, relying on the so-called reconstructive memory. The main feature of this form of memory is combining information from the original material with the knowledge we already possess. At the same time, the material reconstructed in this way often contains elements that were arrived at by reasoning, or parts that relate to the content but are not part of the original material. Bartlett noticed two interesting moments in his findings. During the reproduction, the respondents omitted unimportant details, reducing the reproduction to the basic content, i.e. the meaning of the text they read. However, a far more interesting finding is that the respondents tended to “normalize” the story by introducing details that were not in the original story, as well as omitting details that did not fit their expectations or prior knowledge.
Distortion in the reproduction of the original material is a source of valuable information that can indicate the influence of existing knowledge and expectations on the memorization of new material. This influence is the basis on which it is possible to determine the cognitive status of schemes. If the distortion were unmotivated, we would not be able to say anything about the cognitive status of the schemes. However, if it were shown that the distortion is conditioned by knowledge and expectations based on already existing schemas, it would be an empirical argument that would confirm the status of cognitive relevance of schemas.
In order to examine the effect of prior knowledge on reproduction accuracy, Sullin and Dooling presented subjects with a short text (one paragraph) about a fictional character named Gerald Martin (Sulin & Dooling, 1974). The text begins with the title “The Rise of Gerald Martin” and continues with the following sentences: “Gerald Martin sought to overthrow the existing government in order to satisfy his political ambitions. Many in the country supported his efforts…”. Another group of subjects was given the same text, but Gerald Martin was replaced by Adolf Hitler. After five minutes, the respondents were given a list of sentences, and their task was to evaluate which of them were the same, similar or different from the sentences in the previously shown text. The obtained results show that the knowledge about Hitler greatly influenced the answers of the respondents who read the version of the story in which Hitler was the main character. A large number of subjects judged that the sentence “Hitler was obsessed with the desire to conquer the world” was present in the original text, when this was not the case. The distortion effect was even more pronounced when the subjects were asked to recognize the sentences a week after reading the original text. The existing schema about Hitler not only led to distortions, the respondents already added new content. Although it was not explicitly stated respondents understood that it was a question of the German government, in this case was manifested through the extension of the original text in which central figure is Hitler.
We mentioned that any new information must be integrated into a specific scheme (frame of reference) in order to gain its meaning. Without fitting into existing schemes, it is largely incomprehensible, it remains out of context, and its interpretation becomes-difficult. Schemes are, therefore, a necessary basis for designing new material.
Although in the majority of research in which the cognitive status of schemas was examined, verbal material was used, similar findings were obtained on non-verbal material as well. In the experiment by Brewer and Treyens, subjects were asked to wait in their office until the experimenter checked that a bottle of wine, a basket of food, a screwdriver and items not usually found in professors’ offices. When listing the objects, the respondents adapted to the scheme of a “professor’s cabinet”. All mentioned a desk with a typewriter, a table and a shelf, but very few respondents mentioned a bottle of wine and a basket of food. Moreover, some respondents indicated that there was a bookshelf in the cabinet, although there were no books (Brewer & Treyens, 1981) and here we have an example of reconstruction memory, i.e. adaptation to a certain scheme, whereby many elements are omitted and elements that were not present in the original situation are added.
The findings of the presented research show that schemas are independent of modalities, moreover, they are, as a rule, a combination of information obtained on the basis of several sensory modalities. In this sense, schemes are not only more abstract but also more flexible and comprehensive than individual concepts.
Schemas are not passive units of our knowledge, but active, dynamic factors that participate in designing new information and greatly facilitate their understanding. At the same time, they are constantly supplemented and changed, and some elements of the schemes disappear to be replaced by new ones. What is this dynamic based on and what makes a scheme a scheme? A clear demarcation between individual schemes is often impossible, and their already fluid boundaries become additionally relativized if one takes into account the fact that individual schemes can be parts of wider schemes, as well as the fact that individual elements can belong to different schemes. Despite this, schemata still emerge as more or less coherent memory entities.
The status of the cognitive relevance of schemes is confirmed by the results of research in which their impact on the design and reproduction of various contents was examined. These findings show that the interpretation of information is largely conditioned by schemas, that is, expectations based on previous experience and knowledge. With these findings in mind, the question arises as to what our expectations are based on and what underlies the schemes. Every expectation is based on the probability derived from previous events, that is, our experience related to certain events and situations.
It is quite certain that we will find a waiter in the restaurant, that there will be music in the restaurant is less likely but not unexpected, while the probability that we will meet a penguin in the restaurant (as in picture 13) is negligible. A scheme is, therefore, a set of individual elements of unequal probabilities that represent a more or less coherent whole. It is a feature of the scheme, but also of the event, concept, action and everything that can be covered by the scheme. Because concepts, events, and actions within a given schema occur with unequal probability, the boundaries between schemas become fluid. When a “restaurant” ceases to be a “restaurant” will depend on the degree of presence of elements that do not belong to the expected inventory of this term (scheme). But when this critical limit is reached, when we can state with certaint that we are no longer within the framework of the given scheme, it is a ques that remains unanswered for now.
It is certain that the cognitive system is extremely flexible and adaptable and t is able to place a given situation in a new scheme or modify an existing scheme way that corresponds to the greatest extent to the probabilities derived from the previous