Man and lower animals repeatedly do things to provide themselves with a change in stimuli. These activities can range anywhere from a shift in gaze about the room to a complicated sequence of responses that produces a new or different auditory, visual, or tactile experience. Moreover, animals will selectively attend to stimuli which are novel or particularly striking. Much of the time these behaviors do not subserve any of the primary biological drives such as hunger, thirst, sex, or escape from pain. Nor can they be satisfactorily explained in terms of an acquired drive. In keeping with the traditional position that an organism's actions are impelled by drives, stimulation drives have been postulated to account for these otherwise inexplicable behaviors. A curiosity or exploratory drive is one popular interpretation of an animal's apparent desire for commerce with new or different stimuli. Another explanation is that there is a general drive for sensory stimuli that increases in strength with an increase in the duration of stimulus deprivation. A drive for stimulus changes is thought to be operating concomitantly. Its fluctuations in strength are governed by the organism's second-to-second transactions with the environment. This is a relatively new formulation that may or may not prove satisfactory as research progresses.
The adequacy of the curiosity drive as an explanatory concept will be considered first.
The curiosity drive. Several investigators with quite divergent interests have contended that animals possess a fundamental tendency to explore and learn about the world. For example, W. S. Small (1901, p. 214), remembered for the original work on maze learning
in rats, was most impressed by the thorough manner in which his animals explored the mazes. He believed that rats have a basic desire to know about their new environment and that this desire for knowledge, which appears so fundamental to the rat's behavior, finds its most highly differentiated expression in man's search for scientific truth. A similar point of view was expressed by the French psychologist Ribot ( 1911, p. 368). He contended that a basic craving for knowledge as manifested by the “all-examining, all-embracing scrutiny of a Goethe” was not different in kind from the activity of an animal “that touches and smells an unknown object.” I. P. Pavlov, the Russian physiologist, spoke of an investigatory or “What-is-it?” reflex. Pavlov claimed that this reflex, in animals, results in immediate responsiveness to the slightest changes in the environment and enables animals to engage their receptor organs appropriately ( 1960, p. 12). Pavlov, too, traced the scientific endeavors of man back to this basic propensity of animals to learn about their surroundings.
Research on curiosity behavior . The notion that there exists a basic tendency to explore has been responsible for a large number of animal studies. These can be divided into two groups: (1) experiments designed to provide information on the motivational aspects of curiosity behavior; (2) experiments concerned with the responsiveness of animals to objects or places differing from one another in degree of novelty. Several reviews of these studies have been written, the most complete being Welker's (1961).
Motivational aspects of curiosity behavior . Several types of investigations have been used to study the motivational aspects of curiosity behavior. The animals employed most frequently in this kind of research are the white rat and the rhesus monkey. The rewards provided have been exploratory, visual, manipulative, and auditory experiences.
Exploratory rewards. The white rat is an avid explorer of unfamiliar settings. It seems to rely largely on olfaction, but visual and tactile stimuli are also known to influence curiosity behavior. Permitting the rat to investigate a new region as a reward for executing a specific response provides very satisfactory results. Rats will, for instance, cross an electrically charged grid in order to gain access to an area divided into many compartments. Rats will learn to turn a wheel that removes a barrier separating two compartments and thereby permits entrance into the adjacent one. No food or sexual partner awaits them in this compartment; it is simply a place to be explored, a place somewhat different from the one they had previously occupied. When rewarded, for each correct response, by a 60-second sojourn in a multiunit maze, rats will even learn a brightness-discrimination problem and perform proficiently throughout repeated test sessions.
Visual rewards. The rhesus monkey proves an excellent subject for the motivational aspects of curiosity behavior since it is a highly visual animal as well as an unflagging manipulator of almost any movable object within reach. In view of these basic propensities, the opportunity to watch environmental events and to manipulate small devices has been employed as reward for responding correctly in various discrimination learning problems.
In one study, results showed unquestionably that monkeys will learn to discriminate yellow from blue if rewarded by the opportunity to view the activities taking place in their environment.
Manipulative rewards. A color discrimination problem was also used to determine the value of manipulative rewards. Monkeys were presented with a board containing several pairs of screw eyes arranged vertically. One member of each pair was painted red, the other green. Although all of the screw eyes were fitted firmly to the board, the red ones could be removed easily. If the subject grasped one of these, he was permitted to take it from the board and handle it for several seconds. This was, in fact, defined as the correct response to the problem. An error consisted of either touching a red screw eye but not removing it,.or touching a green one. Within a few test sessions the monkeys responded differentially to the screw eyes, reaching for and manipulating the red ones, but infrequently touching those painted green. Obviously, discrimination learning occurred, and the only apparent reward for the performance was the opportunity to manipulate the devices.
This work has been extended to the study of manipulative tendencies in mice, kittens, and raccoons. Mice were confined in cages equipped with two levers. One lever was movable, the other was not. The recording apparatus registered automatically the number of times each lever was touched; the results showed an overwhelming preference for the one that could be moved. In other experiments, kittens and raccoons have been found to learn simple maze problems in return for the opportunity to manipulate such items as rubber balls, crumpled paper, or small devices mounted on springs.
Auditory rewards. The opportunity to hear what is happening in the environment can also serve effectively as a reward for discrimination learning. Rhesus monkeys were tested on a spatial discrimination problem. During testing, a subject was isolated from the rest of the colony and placed in a box located in a sound-treated room. A loudspeaker was fixed to the top of the box, and a microphone with its associated amplifier was placed in the room housing the colony. The box contained two levers. Pressing one of them resulted in 12 seconds of sounds emanating from the colony room. The other lever was “silent.” Each monkey showed a distinct preference for pressing the lever that provided the auditory stimuli. When sound reward was made contingent on pressing the previously silent lever, the monkeys changed their response pattern accordingly.
Persistence of curiosity behavior. Behavior maintained by exploratory rewards is found to be extremely persistent. This fact is of especial theoretical significance. For if one postulates that curiosity behavior is an expression of a fundamental drive, then it is imperative that the behavior be strong and persistent. With regard to visual exploration, it has been shown that monkeys will hold open a spring-loaded door several hours a day, day after day, if their efforts are rewarded by a view of a dynamic scene. One situation that invariably elicits considerable interest on the part of the test animal is a view of other monkeys. The persistence of manipulative behavior is also great. Monkeys will, for example, repeatedly manipulate devices throughout a 10-hour test session. The amount of manipulative behavior that takes place over a period of days is highly dependent on the animal's previous experience with the manipulanda. Those data available on responsiveness to auditory rewards are not adequate for evaluating the strength of any underlying motivational mechanism. All that can be said is that no decline in response frequency for sound reward was evident during a long series of weekly test sessions.
Factors that influence drive strength. It is clear that for monkeys the strength of an inferred curiosity drive can be increased by confining the animal to a monotonous environment. In the first experiment to demonstrate this, monkeys were kept in an empty illuminated box for two-hour, four-hour, or eight-hour periods without an opportunity to look outside. On completion of the visual restriction period, a small door for viewing purposes was unlocked and the animal was allowed to push it open. A colony of monkeys was outside. The records showed that the frequency of door openings was positively correlated with the duration of visual restriction.
Fear, on the other hand, serves to suppress curiosity behavior. The explanation is somewhat circular but appears, nonetheless, to be relatively reasonable. The experiments have been designed as follows: Animals are subjected to situations presumed, on the basis of other criteria, to be fear provoking. Any marked reduction in curiosity behavior is attributed to the suppressive effect of fear. For example, rats are shocked in a maze; monkeys are presented a close-up view of a large dog or the sound of its bark; or infant chimpanzees are exposed to a strange object. In each case the animal fails to investigate. The rat freezes; the monkey refuses to open the door to look or press a lever to hear; and the infant chimpanzee retreats to the rear of its cage. In most situations not involving painful stimuli, exploration gradually gains ascendancy over fear. This has been elegantly shown for the infant chimpanzee's behavior toward strange objects. A most orderly increase in curiosity behavior occurs with repetitive exposure to new and different objects.
The influence of hunger and thirst on curiosity behavior has elicited a great deal of interest. The general question posed in this line of experimentation is whether curiosity or exploration functions in the service of the nutritional needs. Clearly, if the amount of curiosity behavior is greater when an animal is hungry or thirsty the probability of finding food or water is increased and curiosity has definite survival value. The results of approximately a score of studies on the rat are markedly inconsistent. Much of the confusion can be traced to a confounding of exploration with the amount of locomotor behavior. In some situations animals locomote more when deprived; in others they locomote less. It is probably correct to say, however, that curiosity behavior increases under conditions of food or water deprivation. The most compelling data that support this statement are those from experiments in which (1) a variety of indices for curiosity, including such behaviors as manipulation, sniffing, and window peeking are used; and (2) the locomotive behavior is controlled by the apparatus design, leaving responsiveness to novelty as the critical variable. The thesis that exploratory tendencies in animals are fundamental is not necessarily contradicted by the findings that exploration increases with food and water deprivation. These experiments do imply, however, that a curiosity drive can interact with hunger and thirst drives.
Research carried out with infant rhesus monkeys has direct relevance to the question of the primacy of the curiosity drive. The infants begin manipulating objects within the first few days of life and learn simple problems for manipulative rewards by the time they are two to three months of age. At an equally early age they operate a lever in order to view objects outside their quarters.
Novelty and stimulus change . At this point the argument for the existence of a curiosity drive is half complete. The opportunity to look, listen, and manipulate appears to be rewarding for a variety of animals; and performances can be maintained by these rewards throughout prolonged and repeated test sessions. Curiosity, however, implies selective attention to the new and the different. Animals do, in fact, display such selective attention, but the motivational basis for such behavior may be other than curiosity. This particular problem has generated a long series of experiments; a few representative ones will be discussed briefly.
Spontaneous alternation. In the early studies, white rats were placed in a complex maze containing neither food nor water and the sequence of their movements was carefully noted. It soon became apparent that the rats were not moving about the areas in a haphazard fashion. Rather, when confronted with a choice, they usually entered the maze unit that they had least recently occupied. Later, rats were observed in a simpler situation— a maze constructed in the form of a T. In this apparatus, the animals are started at the stem of the T. On reaching the intersection of the T they enter one of the arms. When, after a few seconds, they are placed again in the maze, they are more likely to enter the opposite arm. This behavior, called spontaneous alternation, was first explained in terms of a response inhibition. It was suggested that the act of turning left or right at the choice-point initiated an inhibitory process, which decreased the probability that the same turn would be repeated on the next trial. Some investigators suggested that an exploratory tendency was responsible for spontaneous alternation, but under ordinary circumstances there was no way to decide whether animals were alternating turns or alternating places, that is, maze arms. Alternating turns by necessity brought the animal to a different place, and alternating places required the animal to make a different turn. By a rather ingenious arrangement of T mazes, it has been shown that rats will repeat the same turn if by doing so they will end up at a different place. Furthermore, they fail to alternate turns if the consequences lead them to the same place. These data were interpreted to mean that spontaneous alternation is an expression of the rat's tendency to respond positively to the more novel aspects of the environment.
Rejection of the familiar. The above opinion was not without its dissenters. Why, it was asked, could one not say that spontaneous alternation in the T maze reflects an active avoidance of the familiar arm rather than a positive approach to the more novel arm? This is no trivial question. Phrased in more general terms, it relates to the basic orientation of animals toward their environment. Are they attracted to the new or repulsed by the old? The data available support the idea that animals are attracted to new or relatively novel places and things. The experiment offering the most convincing evidence for this idea is one in which rats were forced to remain in the stem of a T maze for several seconds. A glass door permitted them to view either arm but prevented their entry. At the time of their confinement, one arm was painted white, the other black. Then the rats were removed and one of the maze arms was replaced so that both were either black or white. The animals were immediately placed back into the maze and this time they were free to enter either arm. Since the subjects had just been exposed to both black and white, presumably they were equally familiar with each color. And with both arms now being the same color the subjects should show no preference if avoidance of the familiar is the basis for exploratory behavior. Yet, almost every animal entered the newly inserted arm.
Reactions to novelty. Other experimenters have studied the rat's responsiveness to environmental change simply by introducing a novel object into the subject's cage and recording the amount of time the animal spends touching and sniffing the object. All experiments consistently demonstrate that rats are quick to notice the presence of anything new. The frequency of investigative responses, however, declines sharply within the first few minutes of stimulus exposure. Investigative behavior can be revived immediately by placing something else in the cage. Novelty effects even extend to eating behavior; novel foods are preferred initially to familiar foods.
The influence of novelty on investigative behavior has also been studied in primates, most of the work being carried out on chimpanzees. When presented with a new object, chimpanzees usually start handling it immediately. Like the rat—and, for that matter, like the monkey and the young child—chimpanzees soon become satiated. Introducing another object serves to reinstate manipulative activity. Generally speaking, more heterogeneous objects elicit more manipulative behavior. It has been argued that heterogeneity makes for greater novelty, which in turn increases the subject's responsiveness. This hypothesis is at least partially supported by other data. For instance, the amount of manipulation in young chimpanzees increases with increase in the number of novel stimuli provided by the object.
The curiosity drive as an explanatory concep t. Much of these data can be accounted for by postulating the existence of a curiosity drive. The strong tendency to manipulate, look, and listen certainly appears to be basic to the motivational system of the higher organisms. It is extremely difficult to explain these response tendencies in terms of other drives. The evidence on the importance of novelty in maintaining investigative activity lends further support to the curiosity drive concept. Selective attention to the new is precisely what one would expect if behavior is motivated by curiosity.
Curiosity implies, however, that an animal is motivated to learn about its world. It has been formally demonstrated that some animals can learn spatial relationships as well as relationships between events merely through the process of exploring the situation. There is little question that this is so for the young child. Nonetheless, differential attentiveness to environmental stimuli is not necessarily an information-seeking process. There are drives for sensory stimulation and stimulus change, as proposed at the beginning of this paper. It is suggested that animals' responses to certain stimuli do not reflect a desire to learn about them; learning may occur as a consequence of stimulus-seeking behavior.
Sensory drives. Drive for stimulus change . There are data collected whose explanation does not require a concept as elaborate as curiosity; simply, a change in stimulation can function as a reward. It has been shown, for example, that mice, rats, and monkeys confined in a darkened cage will repeatedly press a lever that momentarily turns on a light. Some data suggest that the onset of a neutral auditory stimulus in response to the lever can also reinforce lever-pressing behavior.
The recording of eye movements is another technique that may be employed to study what appears to be a fundamental requirement for stimulus change. Chimpanzees will immediately fixate on a panel when it is first illuminated; but the duration of this fixation drops rapidly, and the animal attends to other things. As discussed earlier, the relative novelty of a stimulus influences attentiveness. It could be argued then that this is the basis for the rewarding effect of stimulus change. But other stimulus characteristics, which before now have commanded little research interest, also affect an animal's responsiveness. Stimulus complexity is one. Lever-pressing frequency for light reward is increased over its normal level when the pattern of light describes a complex geometrical figure. Chimpanzees and man have been shown to fixate longer on complex, than on simple, stimuli.
Stimulus characteristics other than novelty influence the duration of visual attention in the monkey. When provided with a chance to view either of two motion pictures of identical content, they will look at each about the same number of times. But they will watch for longer periods that film which is (1) better focused, (2) more brightly illuminated, (3) oriented right side up, (4) in color, and (5) presented at normal speed. In this connection, chimpanzees looking out of a box will select that one of two windows which provides an undistorted view.
In summary, animals will act to effect a simple stimulus change. The level of responsiveness can be influenced by various stimulus characteristics, stimulus novelty being only one of many. This concept of stimulus change can account for much of the data previously described quite as adequately as can a curiosity drive. Moreover, the notion of stimulus change is trimmed of surplus meaning.
The drive for stimulation . The primary biological drives, such as hunger and thirst, increase in strength with deprivation and are reduced by intake of food and drink. The postulated drive for stimulation appears to behave comparably to these in that drive strength is influenced by the degree of deprivation—in this case, stimulus deprivation. Already mentioned is the fact that the monkey's responsiveness to those visual incentives provided by a view of a monkey colony increases with longer intervals of visual restriction. But there are more compelling data that argue for the existence of a stimulus drive. Monkeys confined in the dark for periods up to eight hours will press a lever for a flash of light more frequently as duration of deprivation increases. Comparable data have been reported for man.
Orderly and predictable satiation effects can be demonstrated also when light is used as a reward. Response frequency for light reinforcement declines during a test session; it varies inversely with the level of ambient illumination in the test box and is reduced by pre-exposure to an irregularly flashing light. There are additional data that indicate that rats, when given control over cage illumination, provide themselves with a nearly constant duration of light daily. This observation suggests the operation of a homeostatic mechanism and places the need for illumination in the same category as the basic physiological needs.
Complete visual deprivation or severe visual restriction continuing for the first several months of life have a profound effect on lever pressing for
light reinforcement. Response frequency is inordinately high. It should be noted that in these studies the reinforcement (light) is not confounded with any event associated with seeing, for the subjects had never experienced the world visually.
Animals will learn new problems and maintain performances on previously learned tasks when their efforts are rewarded by visual, auditory, or tactile stimuli. Closely related to this phenomenon are the findings indicating a fundamental tendency to seek stimulus change in the second-to-second transactions with the environment. These data cannot be explained adequately in terms of either primary biological drives or acquired drives. The operation of a curiosity drive has been suggested; and, indeed, most of the findings can be accounted for by this type of motivational agent. The writer, however, prefers to interpret these results in terms of a drive for stimulation and stimulus change. Additional data have been presented to support such a notion.
Irrespective of their eventual interpretation, the experiments discussed here have influenced significantly the theoretical treatment of motivation in man and animals. In fact, it is being argued now that the drive for stimulation is a general and omnipresent motivational mechanism and that hunger, thirst, and sex are mere specialized drives peculiar to the particular state of the organism. This represents an extreme position, but it serves to emphasize the increasing role that is being attributed to the stimulation drives as determinants of behavior.
Robert A. Butler
[Directly related are the entriesCreativity; Nervous System, article on Brain stimulation; Problem solving. Other relevant material may be found inDrives; Learning, article onreinforcement; Motivation.]
Fowler, Harry 1965 Curiosity and Exploratory Behavior. New York: Macmillan.
Pavlov, Ivan P. (1927) 1960 Conditioned Reflexes: An Investigation of the Physiological Activity of the Cerebral Cortex. New York: Dover. → First published as Lektsii o rabote bol'shikh polusharii golovnogo mozga.
Ribot, ThÉodule A. (1896) 1911 The Psychology of the Emotions. 2d ed. New York: Scribner. → First published in French.
Small, Willard S. 1901 Experimental Study of the Mental Processes of the Rat. Part 2. American Journal of Psychology 12:206-239.
Welker, W. I. 1961 An Analysis of Exploratory and Play Behavior in Animals. Pages 175-226 in Donald W. Fiske and Salvatore R. Maddi (editors), Functions of Varied Experience. Homewood, 111.: Dorsey Press.