2. BEHAVIORISTIC APPROACHES TO MOTIVATION. In the late nineteenth century psychology merged both philosophy and physiology together in an effort to explain motivational processes in human beings. Three factors caused a revolution in motivational thought. First, there was the explosive growth in theory and methodology of the natural sciences during the previous two centuries. Second, There was the shattering influence of Darwin's theory of evolution. These two influences affected the course of Hull's 1943 formulation (Korman, 1974). One psychologist profoundly influenced by these three was Watson (1913) who called for psychology to be objective in nature. He called for researchers to study the hard, physical matter of behavior that one could touch, feel, and experiment with in the same way that one could in physiology and physics. Watson believed that in this way psychology could stand beside the other sciences as a legitimate pursuit in the study of man. He declared psychology could operate from the same philosophical framework as these others sciences. The general approach would be to search for the "basic elements" of behavior, physically defined and understood. This basic unit of behavior was called, conditioned-response. Watson was able to recondition the child not to fear the rabbit by presenting it in a context where there was no strong fear-inducing stimulus. The child's conditioned fear response was eliminated the same way that it had originally been developed. Watson believed that he had developed a meaningful way of understanding the antecedents to behavioral choice and direction and that he could do it without resorting to what was, for him, the mysticism of conscious experience. Watson's theories were oversimplified explanations, as he and his followers (Kuo, 1921) thought they would be. The conditioned response (CR) to the conditioned stimulus (CS) is rarely, if ever, the same as the unconditioned response (UR) to the unconditioned stimulus (US). Hence, stimulus substitution cannot account for the change in the direction of behavior, one of the primary concerns of the psychology of motivation. They concluded that something else must be involved. Responses are not infinitely transferable. Brown (1961) made this same conclusion. He contended that this type of approach would hold that variables such as hunger and thirst, variables which are often called motivational because they arouse and direct behavior, are really just sets of stimuli and act like any other stimuli that become linked to responses through classical conditioning. He described Freud's theory of motivation as follows: 1. Basic instinctual drives serve as determinants of behavioral arousal; 2. Increase in internal stimulation results from cyclic patterns in instinctual drives (id); 3. Nature of instinctual drive that is salient at the time like hunger, thirst, sex, etc., results in motivation to satisfy needs; 4. Societal constraints and values as to appropriate means of instinctual drive reduction (super ego); 5. Strength of ego in being able to balance demand of id and superego results in need fulfillment; 6. Contemporary environmental expectancies as to manner in which currently salient instinctual drives may be reduced are developed; 6. Learned habits as to appropriate means of instinctual drive reduction result in the satisfaction of needs. 7. Current developmental stages that the person is in (oral, anal, phallic, latency, or genital), reveal preferred habits of drive reduction remaining from earlier stages, frustrations from earlier stages that the individual wishes to overcome; 8. All this results in the direction of behavior or choice (pp. 112-113). Freud's theory of motivation was largely untested due to its hypothetical rather than scientific foundation. In avoiding unconscious motivational dynamics, Brown and others moved toward a more scientific and observable treatment of motivation in human beings. Hull (1943) is remembered today as the author of a sophisticated attempt at constructing a rigorous, locally tight, mathematically oriented theory of motivation. He rejected mentalistic, subjective notions, such as the will, in favor of physically defined variables in dealing with the phenomena he called motivational processes. His work is described as having involved two major stages, with the first evolving into his 1943 system (Hull, 1943) and the second into his 1952 theory (Hull, 1952). Hull's theory included the physically oriented behaviorism of Watson and the notions of instinctual behavior stemming from Darwin's theory of evolution. Following Darwin, Hull believed that the problem of arousal of behavior, or why man originally behaved at all, stemmed clearly from evolutionary considerations. Whenever an organism was in a threatening situation, behavior was aroused and engaged in to eliminate the treat. An organism behaved when its survival was threatened, and the direction it took stemmed either from innate behavioral characteristics that had survival value or from learned behaviors that had been associated with survival in the past. Hull believed most forms of behavior could eventually be achieved. Hull (1943) attempted to provide some clear answers to the "why" of behavior or the reasons certain behaviors were engaged in more than others. He attempted to spell out in precise detail how these mechanisms worked, how they combined, and so on. He proposed that the newborn organism possessed a set of receptors capable of being stimulated by such sources as external stimuli and internal stimuli of the type associated with biological states of a threatening nature. Examples are stomach contractions (hunger), dryness of the mouth (thirst), and tissue injury (pain). These external and internal stimuli give rise to an internal state marked by two major characteristics. The first of these is a general drive state (called D by Hull), that acts as a general stimulant to the arousal of behavior in that it stimulates activation of whatever behavior tendencies exist in the organism at the time. The second characteristic is that these biological states have associated sets of physical stimulation unique to each state. Thus, hunger involves stomach contractions, thirst, a dry mouth, and so on, Hull thought, and it is these physical stimuli that determine the direction of behavior. The first type of behavior sequence that might be activated is the unlearned sequence, one innate to the organism possibly bred into him by evolutionary adaptation because of its survival value. The second type of behavior sequence that might be activated is that in which the organism responds to the stimuli using a learned behavior rather than an innately bred mechanism. This learned stimulus-response link Hull called H, for habit, and it is on the basis of biological survival value that these stimulus-response links develop, or are learned, or so Hull proposed. Thus, for Hull, D, drive, and H, habit, are conceptual, theoretical variables, rather than observable, physical stimuli. They occur and vary as a function of physically defined responses as a function of their values. At least four of the basic instinctual forces of behavior are the desires to reduce (1) hunger, (2) thirst, (3) pain from tissue injury, and (4) sexual stimulation. The 1943 Hullian system postulated a number of determinants and procedures for measuring these determinants. First, H, a habit, develops because the responses involved have biological utility in that they have led to the reduction of drive stimuli. The speed and development of H is a function of the following variables: 1. The closeness in time of the stimulus-response coupling to the actual stimulus reduction; 2. The number of reinforced trials; 3. The magnitude of reinforcement during training (p. 32) These procedures determine how D and H are measured. How do they combine in predicting behavior? Hull, basing his statements on the results of experiments of Perrin (1942) and Williams (1938), suggested that the strength of the impetus to respond is a multiplicative function of habit and drive, assuming the form: Behavior = f(D) X f(H) Most forms of complex human behavior cannot be viewed simply as the attempt to satisfy survival needs. These are the forms of motivated behavior that the Hullian tradition tended to report. Human behavior is symbolic, conceptual, and not easily controllable by what seems to be the physical type of external stimulation found in animal experiments. Hull's proposal was simple but illusory. He proposed that stimuli associated with the primary drive states elicit behaviors similar to those that the primary drive states do. Thus, it was by using the conditioned stimulus-conditioned reinforcers framework developed on the basis of contiguity that Hull proposed to account for complex behavior in his 1943 theory. Mowrer (1960) proposed a theory of behavior explicitly designed to use much of the Hullian approach and terminology in accounting for complex behavior. His approach dealt with conditioned reinforcers and their implications for controlling behavior. According to Mowrer, various stimuli become associated with various kinds of outcomes and thus assume symbolic significance to the organism. These stimuli assume reinforcing properties and are influences on later behavior. Mowrer distinguished at least four separate cases: 1. Stimuli that become associated with outcomes of behavior that are negative in nature; these become stimuli that are called "fear" stimuli. 2. Stimuli that become associated with outcomes of behavior that are positive. 3. Stimuli that become associated with outcomes of behavior that involve the disappearance of signals that meant the reduction of drives; these become "disappointment" stimuli. 4. Stimuli that become associated with outcomes of behavior that involve the termination of danger stimuli; these become "relief" stimuli. It was Mowrer's basic prediction that "hope" and "relief" stimuli become positive reinforcers of behavior and can be utilized to build habits since organisms are motivated to achieve them. "Fear" and "Disappointment" are negative outcomes that, Mowrer argued, organisms try to avoid and that will not build up into habits. Miller (1948) studied how a conditioned response could be predicted. In an experiment with rats, he found that when they were placed in a situation where they would receive shocks while attempting to escape from a maze that they would (a) attempt to escape even before the shock was administered, and (b) learn new habits that would enable escape from the box when earlier learned escapes were unavailable. The implications of this research finding were that fear could be considered a secondary or acquired drive because it acted like a drive even though it was not innate, but was learned, and conditions for its acquisition could be demonstrated experimentally. The fact that the behavior is learned and does not need to have any biological significance attached to it is very crucial. This research was one of the reasons why the adequacy of Hull's theoretical system was eventually challenged by many psychologists. The significance of anxiety for understanding neurotic behavior had been pointed out by Freud (1933), among others, and psychologists such as Mowrer (1960) spent significant portions of their careers studying the anxiety phenomenon by using experimental techniques inspired by Hullian theory. Miller suggested this in a 1948 article. The mechanism of acquired drives allows behavior to be more adaptable in complex variable situations. It also allows behavior to appear more baffling and apparently lawless to any investigator who has not had the opportunity to observe the conditions under which the acquired drive was established. One hypothesis is that neurotic systems, such as compulsions, are habits which are motivated by fear and are reinforced by a reduction in fear (pp. 100-101). One implication from this successful experimental development of an acquired drive was that it led to a proposed mechanism for integrating within a relatively theoretical framework the diverse multiplicity of human goals. Instead of going through the process of postulating separate motives for such things as money, status, prestige, a new car, a big house, and the like. Brown (1952) suggested: The important motivating component of many of the supposed acquired drives for specific goal objects is actually a learned tendency to be discontented or anxious in the absence of these goal objects. On this view, stimulus cues signifying a lack of affection, a lack of prestige, insufficient money, etc. would be said to acquire, through learning, the capacity to arouse an anxiety reaction having drive properties. This learned anxiety would then function to energize whatever behavior is directed toward goal objects by stimuli, and its reduction, following the achievement of these goals, would be powerfully reinforcing (pp. 1-21). Motivation can be a function of frustration. It is an emotion that causes many people to make life adjustments to meet desired needs. Frustration, according to Brown (1961) and Lawson (1965), is a conflict between two opposing tendencies. One response tendency is the one originally evoked by the situation (presumably some kind of goal response), and the other being some alternative response aroused by the frustrating interfering conditions themselves. This conflict between opposing tendencies leads to whatever could be said to be the unique behavioral consequences of frustration. Because frustration is defined in terms of the relationship between two hypothetical constructs - the opposing response "tendencies" - frustration is a higher order construct. It is defined in terms of first order constructs. One effect is an increase in drive. Frustration adds to the total motivation of the organism, and thus strengthens if the goal directed response is far stronger than any other behavior in the situation. The second effect is to produce unique internal stimuli. These stimuli may be related to other responses not previously present in the situation. Although there were studies to support Hull's prediction (Wolfe, 1936; Cowles, 1937), the hypothesis always had problems. For example, there was evidence that secondary reinforcers may quickly lose their effect over time if they are not associated often enough with the primary reinforcer (Isaacson, Hutt, and Blum, 1965; Wike, 1969). Other experiments showed secondary reinforcing effects over long periods of time. Zimmerman (1957) predicted that a secondary reinforcer can have long-lasting effects if it has been periodically associated with the primary reinforcer to begin with, rather than a constant association with it. Siegal and Milty (1969) found little experimental support for the proposal that stimuli signal the termination of electric shock take on secondary properties in the manner that Hull and Mowrer predicted. In his emphasis on Darwinian evolutionary theory as a logic for behavior and on a stimulus-reduction model as a basic explanatory system, Hull (1943) was quite similar to Freud. In other respects he was different. Hull was a physically oriented behaviorist who attempted to deal with motivational processes by postulating physiologically based stimuli of evolutionary significance as leading to the arousal and motivation of behaviors. Once behavior was aroused and drive (D) was operative, Hull suggested that direction was due to the linkage of physiologically based stimuli and responses. Despite his objective goals, logical and empirical problems appeared when Hull applied his theories to complex human processes. Hull's work became what he hoped it would, a stimulation for continued research by opponents and proponents. Fifty years later, his research can still be cited as the acme of what a theory should be: a thought-generating system that proved to be an immeasurable aid in the understanding of what motivates behavior. There were a number of problems with Hull's 1943 system. While stimuli associated with shock do assume fear or anxiety- arousing properties and do function as drives in the Hullian sense, stimuli associated with hunger and thirst deprivation states do not become conditioned stimuli and do not assume drive characteristics (Cravens and Renner, 1970). Another set of findings that did not support Hull's work was the evidence that obese individuals do not use biological deprivation cues for how and when they should eat. Their eating behavior seems to be a function of variables of having little or nothing to do with the degree of their physiological deprivation at the time (Nisbett, 1968; Schacter, Goldman, and Gordon, 1968). As a continuing demonstration of the difficulty of coordinating D, as conceptualized by Hull, to biological-need states and the various stimuli conditioned to them, Zajonc's program supported the theoretical assumption that the presence of other people may as a drive state in the Hullian sense. This added to the confusion as to what D is (Zajonc, 1965). There was considerable additional evidence that biologically linked need states as hunger, thirst, pain, and sex area function of many other variables besides physiological determinants. Vernon (1969) pointed out that (a) food preferences are very much culturally determined and some individuals will and do starve rather than eat forbidden foods; (b) sexual behavior is under hormonal influence for organisms low in the evolutionary hierarchy, but such behaviors are very much a function of learning experience for humans and higher animals; (c) there is some basis for biologically linked instinctual maternal behaviors in lower animals but such behavior is culturally determined in humans; and (d) fear and anxiety are very much culturally learned, with great differences between individuals and groups in the degree to which they will admit such reactions and base their behavior on them. The question is, does the learning of stimulus-response connections (habit) depend on primary and/or conditioned-stimulus (need) reduction? In a real sense, the logic of Hull's entire approach rested on the answer to this question. He based his system on the principle of the biological utility of behavior and its derivatives. While habits can be learned on the basis of primary or conditioned need reduction, it is not necessary that such reduction occur for learning to take place. The evidence for this statement comes from a variety of research sources, some of which follow: 1. Non-nutritive saccharin (a substance that passes through the body without providing any food value) can reinforce the learning of an instrumental response (Sheffield and Roby, 1950). 2. Infant monkeys can obtain satisfaction from ersatz mothers who have not provided milk (Sheffield, Wulf, and Backe, 1951). 3. Rats can be reinforced by copulating responses without being given the opportunity to ejaculate (Harlow and Zimmerman, 1959). 4. Habits can be learned when the outcome of the behavior involved is not the reduction of some primary or conditioned need, but rather the opportunity to explore some primary or conditioned need and to explore some "new" stimulus field (Harlow, 1953; Montgomery, 1953; Glanzer, 1953). By a "new" stimulus field is meant one that is different from the one in which the organism is currently behaving and one not previously associated with primary-need reduction. Blodgett's (1929) famous latent-learning experiments showed quite conclusively that drive reduction, which defines reinforcement in the Hullian sense, seemed to influence engaging in behavior patterns already known, but may not be necessary for learning them. He showed that in three groups of rats learning to run a maze. One of the groups was provided food at the end of the maze whereas the other two were not. At first glance, this seemed to support Hullian theory. However, after closer scrutiny of the data, a different conclusion presented itself. He inferred that both Groups II and III almost immediately attained the level of proficiency of Group I, suggesting that they knew how to run the maze all the time, but that there was no reason to do so. Once they learned that there was food available, then there was some reason to show the learned behavior patterns. The latent learning experiments were crucial for at least two reasons. First, by showing that need reduction is not necessary for learning, but that the opportunity for need gratification could generate quick, significant changes in behavior, considerable doubt was cast on the 1943 Hull theory. Second, these results suggested that a complete theory of motivation would have to include some consideration of how changes in the environment affect behavior, and among these effects might be the availability of the types of incentives used by Blodgett. Thus, it was this research that led Hull and his co-workers to develop K, a type of environmental incentive, as part of the later 1952 postulates and Spence's 1956 theory. The theoretically crucial question for Freud, Hull, and other researchers to ask is behavior just a desire to reduce stimulation? Is this where behavior arousal comes from, or is there some other motivating factor present? Miller and Dollard (1941) suggested a drive is a strong stimulus which impels action. Any stimulus can become a drive if it is made strong enough. The stronger stimulus, the more drive function it possesses. Their logic eventually led Hull to adopt their argument in 1952. There was considerable evidence to support Hull's, Miller's, Dollard's, and Freud's work. Behavior is sometimes more than just a reduction of stimulation and may lead to less behavior rather than more. The following studies support these notions. Olds and Milner (1954) found that direct electrical stimulation of certain parts of the brain can reinforce behavior in the same manner as food, thirst, and pain reduction. They found that here was a case where reinforcement can come from more stimulation, rather than less. Harlow (1953), Montgomery (1953), and Glanzer (1953) all showed that the opportunity to explore and see "newness" and "differences" can be a great incentive to behavior; "newness" and "difference" can be assumed to be increases in stimulation. Brown and Jacobs (1949) found that increased amounts of drive do not always result in increased physical activity and might actually result in less. As an example, they cited their own research finding that some anxious rats "freeze," rather than increase their motor behavior. This is one of those situations where it is possible to argue that "freezing" is a form of behavior that reduces the drive for the subjects. Research with different forms of drug addiction and drug- taking behavior indicated both support and difficulties for the Hullian drive concept. On the one hand, the taking of opiates and barbiturates provides support since in these cases the result is some kind of stimulation-reducing process. On the other hand, the taking of hallucinogenic drugs like hashish, marijuana, and LSD, and those termed, analeptic (cocaine and amphetamines), would be considered negative evidence since the outcome is generally increased excitement and stimulation (Cohen, 1965). The sum total of these findings suggest that behavior can clearly be shown to be more than a reduction of stimulation. Under some conditions, organisms may act in order to increase stimulation. Berlyne's (1960) theoretical hypothesis suggested that the stimulation organisms try to reduce is not environmental but rather in the reticular activating system (RAS) of the brain. Berlyne argued that it is a curvilinear function of physical environmental stimulus complexity, so that RAS arousal is greatest when environmental stimulation is either very great or very small. According to this system, the following behavioral predictions are made as a function of the given physical environmental stimulus condition. The crucial question here is the degree of support for the major innovation of the theory, namely, that low levels environmental stimulation lead to high RAS arousal. While there is some indication that boredom leads to high RAS arousal, the evidence to support this is still scant (deCharms, 1968, p. 101). The implication of this paradigm is that the previous history of the organism determines its degree of responsiveness to later positive and negative outcomes and how significant these possible or expected outcomes may be in affecting its behavior. The poorer the previous history, the less likely it is that "good" outcomes will be necessary in order to solidify or reinforce approach behavior. The following research studies indicate that the more an organism is punished for its behavior, the less likely it will engage in behavior to achieve rewards and avoid punishments in the future. Conversely, the less it is punished, the more it will be oriented to achieving "good outcomes," such as engaging in behavior to achieve positive outcomes and avoid those that are negative. There is a vast body of research that points to the fact that organisms have not been continuously positively reinforced for engaging in certain behaviors when no reinforcement is available than those who have a history of continuous positive reinforcement. Baron (1966) found that individuals who have high social reinforcement standards are more likely to perform in a manner designed to achieve high rewards than those who have low social reinforcement standards. Kaufman (1963) and Feather (1965) projected that self-perceived ability on a task based on previous task performance is positively related to later task performance. Korman (1967a, 1967b) hypothesized that individuals of high esteem are more likely to choose occupations where they perceive themselves to have a degree of mobility than those of low self- esteem. In a later study, Korman (1968) found that individuals who are told they are incompetent and cannot achieve specific goals on task, even though they have had no previous experience with the task, will perform worse than those who are told they are competent to achieve the task goals. Zander et al (1969), in studying group behavior, found that if they failed previously set goals, they increase the probability of their failing again. Shaw (1968) maintained that academic underachievers have a more negative self-concept than achievers. Brookover and Thomas (1963-64) found that there is a significant positive relationship between self-concept of ability and grade-point average. Maier's (1949) focused on individuals who had a long series of frustrations. They engaged in behavior that was (a) persistent beyond degree of reward, (b) not alterable by punishments, (c) non-goal oriented, and (d) not affected by consequences or by the anticipation of the same. Karsh (1962) and Miller (1960) found that human beings often over learned in response to becoming very competent at any task, and this increases as the effectiveness of punishment is meted out resulting in eventually suppressing responses. Miller (1960) supported the notion that human beings adapt to punishment and in so doing decrease its effectiveness to cause change in the individual. Holz and Azrin (1961) found that if a subject is habituated to receiving shock together with positive reinforcement during reward training, punishment during extinction can actually increase resistance to extinction. Masserman's (1943) research indicated that punishment can lead to self-defeating behavior oriented toward no goal. Finally, Miller, Butler, and Martin (1969) concluded that rewarding others has a greater influence on their behavior than punishing them. Spence (1960) and Amsel (1958) theorized that organisms put on a partial-reinforcement schedule learn to link two types of stimuli with approach responses in a given goal-box situation. They learn two habits, not one, under partial reinforcement. The Spence-Amsel attempt to account for what might be termed self- defeating behavior within a physical system received considerable research attention, with some results being supportive (Haggard, 1959; Goodrich, 1959) and others not so (Hill, 1968). These researchers supported the conclusion that individuals will perform in a manner consistent with other reinforcement patterns. From the previous discussion, it is clear that Hull's 1943 theory had an imposing list of problems that had to be overcome if the approach was to remain a viable one in the study of motivational processes. The decade of the 1950's saw revisions of the general framework by Hull, and his promising student, Kenneth Spence. The dilemma for Hull and Spence was to account for the findings of the latent-learning experiments and others similar to it within the physical system. There was evidence that organisms could and did change their behavior almost immediately when the levels of reward and/or incentive available in the environment changed. This accounted for the fact that behavior was not only determined by previous learning (habit) and current condition (drive), but also by the contemporary characteristics of the environment. Consider a rat running a maze to get to a "goal box" that has food in it. Once he gets to the food, he begins eating, and the more food, the longer and more vigorously he eats, generally up to the saturation point. Since this eating behavior becomes conditioned by stimulus-generalization processes to the various stimulus characteristics of the "goal box," the rat, when later reintroduced into the "goal box," immediately starts the eating behavior again, even before the food is present. This anticipatory behavior is engaged in prior to actual eating behavior and might be engaged in other parts of the maze prior to actually entering the "goal box." When the rat runs a maze, the physical stimuli affecting him are the stimuli of the maze and his anticipatory goal (or eating) responses at each of the maze prior to the goal (or food) box (Korman, 1974, p. 60). Spence (1960) described this condition mechanism as K, and he felt that K, as a general arousing agent acting in the same manner as D, should be placed in the same conceptual category. Behavior = D (drive) + K (conditioned mechanism) x Habit Hull (1952) proposed that D and K combine multiplicatively, whereas, Spence (1960) proposed that they combine additively, a difference illustrated by the above formula. Hull gave up the simplistic notion that all behavior stemmed from biological-physical sources. He changed his assumption that behavior or drive was aroused by biological-need deprivation by some stimulus conditioned to it. Instead, he eventually supported Miller and Dollard's (1943) assumptions that any strong stimulus can become a drive, the reduction of which is reinforcing. This change in focus presented several advantages. It freed Hull from having to tie all secondary drives back to a biological base and it enabled him to show that, experimentally, learning and performance, and competence, could take place without biological-need reduction. Five decades after the 1943 model was first presented, there are a number of conclusions that can be waged. First, the tradition is a strong one and remains highly viable as a source for continued resource. Second, environmental as well as biological and sociological factors can and do effect motivational change and Hull and Spence included these factors in their later theories. As an example of the pervasiveness of the Hullian scheme, Weiss (1968) adopted an approach to delimit the conditions under which attitude change was a function of persuasive communication. He suggested that the likelihood of change as a result of persuasive communication is greater the more we have changed our attitudes in the past. The more anxiety we feel about our current attitudes, and hence the greater the impetus toward change, the more reason we see for attitudinal change. Zajonc (1965) hypothesized why the presence of others will sometimes facilitate performance and sometimes debilitate it. He postulated that the presence of others is a source of drive, D. While other people can be directional influences on behavior in the sense of producing specific cues and reinforcement, Zajonc argued that they are also a source of general arousal, energizing all responses likely to be emitted in the given situation. Zajonc reported a number of studies that supported these predictions (Zajonc and Sales, 1966; Wheeler and Davis, 1967; Cottrell, Rittle, and Wack, 1967; Zajonc, Heingartner, and Herman, 1969). In addition, this particular application was further solidified and related to the general Hullian tradition by the theorizing of Cottrell et al. (1968), Weiss and Miller (1971), and Paulus and Murdock (1971). Evidence suggested by Bolles (1967) that these different sources of drive were not interchangeable in contributing to general arousal and were contrary to Hull's predictions. Cofer and Appley (1964) created two incentive-like constructs, known as the sensitization-integration mechanism (SIM) and the anticipation-invigoration mechanism (AIM), as conceptual substitutes for the D construct. Berger and Lambert (1968) described these mechanisms as follows: The first mechanism (SIM) is posited to explain the invigorating effects on action of selective sensitization to certain stimuli. This effect is probably an innate one, but one that can be modified by learning. For example, under controlled conditions, activity increments which one would expect as a result of deprivation may not occur. These activity increments do occur when the relevant stimuli are introduced, in the sense that the deprived organism responds more vigorously than the non-deprived one. In sex behavior, Beach (1956) suggested hormones are a necessary precondition for arousal of the male, but are not sufficient; usually a receptive female, plus physical or symbolic interaction, are also necessary for copulation to occur. The parallel mechanism, AIM, carries the main weight of the Cofer and Appley analysis. Here invigoration of behavior is enhanced by learned anticipations. These may take several forms: the incentive cues (K) which "control" anticipation, working through such processes as the r mechanism; the energy released by the states of conflict between two or more anticipations; or the ee mechanisms, suggested by Sears (1951). Where anticipation is possible the usual effects of deprivation does not appear to be operative, invigoration may occur through anticipation alone (p. 822). One major problem with the expectancy system of Hull and Spence was what to predict for low-expectancy people. Do they keep searching for the best alternative and pick that one? Does that mean that before he reaches self-fulfillment drive and habit must be abrogated? Does that mean he is less likely to behave like a high-expectancy person? Many questions remained to be asked by researchers. K as a concept did not work because it was logically possible to show that the same processes hypothesized to lead to K (behavior increase) would also lead to behavior reduction. In addition to this problem, other logical and empirical problems developed with the Hullian system, suggesting that Hull's goal of achieving an objective, physically oriented, natural-science approach to motivation did not succeed.