What are cognitive maps?

Introduction

Edward C. Tolman
first identified what he later named cognitive maps as a result of a series of experiments conducted in the 1920s and 1930s. In these experiments, Tolman sought to discover whether learning occurred that might not be immediately reflected in performance—what he came to term “latent learning.”

In a typical experiment, Tolman constructed an intricate maze. Three groups of hungry rats ran the maze once a day for twelve days. The first group, the rewarded control group, received food for successfully completing the maze. The second group, the nonrewarded control group, received no food; its members investigated the maze. The third group, the experimental group, also received no food for the first ten days of the experiment, and its members simply surveyed the maze. On the eleventh day, the experimental group was provided with food; on the twelfth day, this group performed as well as the rewarded control group.

Tolman concluded that latent learning had taken place and that under the appropriate conditions, the experimental group reflected this learning through its performance—the successful completion of the maze. He asserted that the rats had formed “cognitive maps” that enabled them to solve the maze. Therefore, learning—defined here as the construction of cognitive maps—is not the same as performance. Although learning is reflected in performance, Tolman’s work strongly suggests that the appropriate context is necessary to elicit that performance. In the case of the experimental group, the context was the food the experimenters provided on the eleventh day of a twelve-day experiment. The rats, Tolman inferred, anticipated that successful completion of the maze on the next attempt would result in their receiving the desired food.

Research has supported Tolman’s pioneering work. In 1978, Emil W. Menzel used chimpanzees to illustrate the spatial dimensions of cognitive maps. He hid food in a field and then carried a chimpanzee around the field with him. He did not allow the ape either to eat or approach the food, preventing both instrumental conditioning and primary reinforcement from taking place. Later, the chimpanzee that had been shown the food’s hiding places and five experimental chimpanzees that had not seen them were released in the field. Invariably, the first chimpanzee went directly to the food. The experimental animals found their food through scrutinizing the area near the chimpanzee that had been shown the hidden supply of food or begging food from this chimpanzee.

In 1976, David S. Olton and Robert J. Samuelson demonstrated spatial memory (cognitive maps) in rats through employment of a radial maze. Each arm of the maze had food at the end of it. Through a series of manipulations, the experimenters demonstrated that the rats remembered which arms they had explored and at which arms they had been fed. They eliminated the possibility that the animals used smell to locate the food through altering the animals’ sense of smell. Furthermore, other researchers moved the maze to note whether other factors, such as tactile clues, influenced the rats. In each variation of the experiment, the rats behaved as if they were responding to spatial location and not tactile clues. In 1985, William Roberts and Nelly Van Veldhuizen demonstrated that pigeons, as well as rats, can work the radial maze.

Learning Theories

Cognitive theories in learning have gained in popularity. Tolman’s groundbreaking work established the concepts of cognitive maps, internal spatial memories of the animal’s relevant environment, and expectancy, an animal’s anticipation of a sequence of events in time. Tolman’s work, supported and developed through additional research, further established the distinction between learning and performance. Linguists such as Noam Chomsky express this distinction as one between competence and performance.

Cognitive anthropologists, influenced by Tolman’s work, have applied the concept of the cognitive map to the learning of individuals within sociocultural systems. Cognitive maps, in the anthropologists' view, provide guides to cultural behavior by organizing psychological, social, and cultural landscapes in terms of their relevant characteristics for members of any given society. Two famous covers of the New Yorker magazine illustrate this point. One cover is a New Yorker’s view of the West. In that view, the entire center of the United States comprises an area smaller than Midtown Manhattan. The view to the east is little better: the Atlantic Ocean becomes a puddle, and the geography of Europe is greatly distorted. The point is not that New Yorkers are more ethnocentric than other people; it is that all people exaggerate those aspects of their landscapes or environments that are most important to them, and they neglect those features that they consider unimportant.

A number of factors enter into perceptions of what is most relevant and what is not. Some of these are personal, such as age, gender, likes, and dislikes; others are social, such as class, ethnic group, and occupation. All factors, however, fit into a cultural context and take on meanings within that context. Anyone attempting to understand the manner in which people learn and demonstrate that learning through adequate and appropriate performance must take into account these factors and how they help shape an individual’s cognitive map.

In education, the prior-learning approach has sought to come to grips with these issues and to apply them to instructional ends. Essentially, this perspective maintains that it cannot be assumed that learning has not taken place merely because a student does not demonstrate the desired performance. William Labov demonstrated, for example, that the presumed inarticulateness of African American street youths was a function of the setting in which people had tested them. In more natural settings, Labov determined that they were, in fact, highly articulate.

Tolman’s insight that rats in a maze will demonstrate latent learning through performance when an appropriate stimulus is present has influenced cognitive therapy. This therapy is based on the hypothesis that people base their behavior on cognitive maps and expectancies. These internal representations of spatial relationships and anticipated sequences of events, based on past experiences—psychological, social, and cultural—form individuals’ perceptions of reality, even when these stimuli are not materially present.

Cognitive motivation theory is a “pull” theory of motivation, based on the hypothesis that people’s expectancies provide incentives for behavior. There are positive-incentive and negative-incentive motivations. Working for a promotion along paths anticipated to achieve that desired goal is an example of positive-incentive motivation. In contrast, a youngster who is developing his or her martial arts skills to deal with bullies who frequently beat him or her provides an example of negative-incentive motivation.

Values enter intimately into cognitive motivation theory. To motivate people, incentives must be valued. If people do not value an incentive, such as a promotion, they are less likely to perform the actions they associate with receiving that incentive. If they receive the goal without performing the behavior—for example, if someone receives a promotion undeservedly—they are less likely to value the goal. In sum, there is an intrinsic relationship between expectancies and value.

Moreover, relief and frustration enter the picture. Relief refers to not receiving an expected negative result (a person does not fail a test for which he or she did not study). Frustration involves failure to attain a goal for which a person has prepared. Failure to receive a promotion to which a person is entitled is an example of frustration. Relief is an example of positive-incentive motivation, and frustration is an example of negative-incentive motivation.

Albert Bandura and others have advocated cognitive behavior therapy based on the application of positive and negative incentives. Such therapy seeks to alter the expectancies and relational maps of clients. Thus, clients can relearn their environment through redrawing cognitive maps and altering their expectancies. There are many techniques employed to bring about these changes in spatial and event expectations. Therapists who employ Albert Ellis’s
rational emotive therapy
believe that the therapist should take a strong interventionist role in the therapy, aggressively confronting the client whenever he or she exhibits examples of irrational thought. These confrontations seek to force clients to learn new, more rational ways of thinking and, therefore, behaving.

Cognitive behavioral therapists seek to change a person’s inappropriate thoughts to more effective ones. They first learn what their clients are thinking and then relate these thoughts to inappropriate behavior. They seek to help their clients learn new thoughts that will result in more appropriate behavior. Patients are taught to “talk to themselves,” substituting good thoughts for bad. Rather than dwelling on failure, patients concentrate on success or positive aspects of their lives. A student taking an examination, for example, would stop thoughts of failure and remind him- or herself about how well test preparation had gone. Self-encouragement would replace self-disparagement.

Each of these applications is based on the theory that people’s behavior is based on internal representations of the world. Each person’s representations differ in some way from those of others. These representations influence both the way in which one learns about the world and the manner in which one represents that world. The application of Tolman’s work on cognitive maps and their related internal representations, or expectancies, has led to a deeper understanding of learning.

Relevance of Cognitive Maps to Psychology

One of the important issues for the field of cognitive psychology is that of representation. In general terms, how do humans store information in the brain? In 1969, Allen Paivio presented the dual code theory, which suggests that both analogical and verbal codes are used for representing information. Some information maybe stored in an imagelike form (analogue), while other information is stored in a verbal format. In 1973, Zenon Phylyshyn advocated the propositional hypothesis, which suggests that concepts are stored in an abstract form that captures the underlying relationship between ideas. People may experience images, but this experience is simply a by-product, an epiphenomenon, of the retrieval of information. Research into cognitive maps may provide some insight into the issue of representation.

Sources of Information for a Cognitive Map

The development of a cognitive map may include in the representation information from a variety of sources, including landmarks, route information, and survey information. This information may be incorporated into a cognitive map over time and is not necessarily mutually exclusive. Salient features such as landmarks, distinctive objects that stand out from the rest of the environment, provide a point of reference for orientation and navigation within an environment. Route knowledge is specific information regarding how to navigate from one location to another. The directions one gives to allow another to navigate from one location to another would be similar to route information. Survey knowledge provides an overview of the relationships between locations; this perspective has a better grasp of the global relationship between various locations. This type of knowledge is acquired from maps by traversing the environment from a number of different perspectives.

In addition to the type of information available, a number of other factors can influence the development of a cognitive map, such as angle, shape, and orientation. These factors may lead to distortions in generating a map. For example, streets that cross each other at an odd angle tend to be drawn closer to a ninety-degree angle than they really are. Still other factors that are not spatial in nature may affect the retrieval of information. In 1991, Keith Clayton and Ali Habibi discussed the confounding of time and space. Locations experienced close together in space are often experienced close together in time. Retrieval of information, in some cases, may be due to temporal proximity versus spatial proximity. Other such factors in this category may include semantic clustering (grouping together of similar items) and route knowledge (grouping of items spatially and temporally).

Methods for Assessing Cognitive Maps

A variety of methods have been used to test spatial knowledge, including location judgment (whether a location is closer to one reference point or another), recognition (whether the item is part of the map or not), distance estimations (how far it is between two locations), map drawing, providing directions, and pointing (what direction would one travel from a given location to another specific location). In a 1986 study, Timothy McNamara discussed the merits of using tasks such as location judgment and recognition that allow one to look at priming, which taps into automatic processes. These tasks may be informative in terms of how spatial information is organized in long-term memory. Other tasks, such as pointing, may be more informative in terms of strategies that are used to answer spatial questions.

In 1978, Albert Stevens and Patty Coupe asked subjects which is farther west, Reno, Nevada, or San Diego, California. Many incorrectly inferred that San Diego is farther west than Reno, since California is farther west than Nevada. In actuality, the state of California curves under the state of Nevada, making Reno farther west. This implies the use of information such as relative position within categories (states) rather than actual spatial information between the two locations.

Other Areas of Research in Cognitive Maps

Research suggests that the ability to form and use a cognitive map begins to develop around the age of three. Judy DeLoache, in her 1987 experiments, used a scaled model of a larger room to test children’s spatial knowledge. Children of various ages were asked to find a small hidden toy in the model. After finding the small toy, children were then asked to find the larger toy, which was hidden in the same location in the larger room. This task required the child to use the information from the scaled model to find the object in the larger environment. DeLoache suggested that the younger children had a difficult time using the scaled model of a room as a basis for a representation of the larger room. Research in this area continues to look at the application and development of these skills over time.

Studies on gender differences in spatial knowledge have focused on strategies such as wayfinding and direction pointing. The use of such spatial strategies tends to be gender specific. It appears that women prefer to use a strategy of wayfinding based on the use of landmarks, while men tend to prefer using more of a global or survey strategy. Studies suggest, however, that these differences may be due to preferred strategies versus actual differences in acquisition of spatial knowledge.

Reality Representation and Behavior

Tolman’s concept of the cognitive map grew out of a recognition that internal representations of reality influence behavior. Moreover, learning is not indistinguishable from performance—the two processes are analytically distinct. An organism may have learned behavior without demonstrating that behavior through performance. This latent learning can be elicited through the presentation of adequate incentives.

Tolman’s work in the 1920s and 1930s did much to advance the field of cognitive psychology at a time when behavioral psychology dominated the schools. It provided an additional dimension to learning, advancing the position of internal representations of reality. The empirical evidence offered to support cognitive theory has been impressive, and its status in psychology has advanced accordingly; it is often combined with behavioral concepts, as in the work of Bandura.

That combination has enabled educators and therapists to bring about behavioral changes based on changes in the manner in which students and patients perceive their worlds. New internal representations of the external environment can be brought about through changes in cognitive maps and expectancies. In turn, these changes alter the bases of decisions that influence an individual’s future behavior.

Tolman’s work has influenced linguistics, anthropology, sociology, and other social and behavioral sciences. Scholars in these areas have applied the concept of the cognitive map cross-culturally and within cultures to members of subgroups. Future work will apply it to the manner in which each individual negotiates his or her way within cultural and social systems. Continuing work in anthropology and sociology in the negotiated nature of sociocultural systems draws heavily on cognitive maps. Prior-learning theory is based on the idea of latent learning, and future work will continue to extract applications of significant value to education.

Future advances will likely occur in studies that investigate field dependence and independence in cognition as related to other aspects of culture, such as child-rearing patterns and subsistence practices. The continuing interest in the relationship between language and the cultural organization of reality holds promise for further advances in understanding and applying cognitive maps. The role of choice in the individual’s construction of these maps is also an area of intensive investigation.

Bibliography

Bandura, Albert. Social Foundations of Thought and Action: A Social Cognitive Theory. Englewood Cliffs: Prentice, 1986. Print.

Bukatko, Danuta, and Marvin W. Daehler. Child Development: A Thematic Approach. 5th ed. Boston: Houghton, 2004. Print.

Carbon, Claus-Christian, and Vera M. Hesslinger. "Attitudes and Cognitive Distances: On the Non-Unitary and Flexible Nature of Cognitive Maps." Advances in Cognitive Psychology 9.3 (2013): 121–29. Print.

Chomsky, Noam. Language and Mind. 3rd ed. New York: Cambridge UP, 2006. Print.

Fromkin, Victoria, Robert Rodman, and Nina Hyams. An Introduction to Language. 10th ed. Boston: Wadsworth, 2014. Print.

Gallistel, C. R. The Organization of Learning. Cambridge: MIT P, 1990. Print.

Labov, William. The Social Stratification of English in New York City. 2nd ed. New York: Cambridge UP, 2006. Print.

Matlin, Margaret W. Cognition. 8th ed. Hoboken: Wiley, 2013. Print.

Menzel, Emil W. “Cognitive Mapping in Chimpanzees.” Cognitive Processes in Animal Behavior. Ed. Stewart H. Hulse, Harry Fowler, and Werner K. Honig. Hillsdale: Erlbaum, 1978. 375–422. Print.

Olton, David S., and Robert J. Samuelson. “Remembrance of Places Passed: Spatial Memory in Rats.” Journal of Experimental Psychology: Animal Behavior Processes 2.2 (1976): 97–116. Print.

Pazzaglia, Francesca, and Angelica Moe. "Cognitive Styles and Mental Rotation Ability in Map Learning." Cognitive Processing 14.4 (2013): 391–99. Print.

Roberts, William A., and Nelly Van Veldhuizen. “Spatial Memory in Rats on the Radial Maze.” Journal of Experimental Psychology: Animal Behavior Processes 11.2 (1985): 241–60. Print.

Suzuki, Ikuo. "Effects of Sense of Direction on Internet Skill and Cognitive Maps of the Web." Computers in Human Behavior 28.1 (2012): 120–28. Print.

Tolman, Edward C. “Cognitive Maps in Rats and Men.” Psychological Review 55.4 (1948): 189–208. Print.

Tolman, Edward C. Purposive Behavior in Animals and Men. New York: Century, 1932. Print.

Zhu, Qing, Rubin Wang, and Ziyin Wang. "A Cognitive Map Model Based on Spatial and Goal-Oriented Mental Exploration in Rodents." Behavioural Brain Research 256 (2013): 128–39. Print.