[Paleopsych] Reiss and Havercamp: The Sensitivity Theory of Motivations

[Premise Checker]

Steven Reiss* and Susan Havercamp: The Sensitivity Theory of
Motivations: Implications for Psychopathology
Behavioral Research and Thererapy, 34.8 (1996): 621-32

[Another published paper by Steven Reiss and his colleague, Susan 
Havercamp. I have been beating a drum for Reiss's 16 fundamental human 
motives for some time.]

Nisonger Center, The Ohio State University, 1581 Dodd Drive, Columbus,
OH 43210-1296, U.S.A.
*Author for correspondence.
(Received 4 March 1996)

Summary--Sensitivity theory holds that people differ in both the types of 
reinforcement they desire and in the amounts of reinforcement they need to 
satiate. People who crave too much love, too much attention, too much 
acceptance, too much companionship, or too much of some other fundamental 
reinforcer are at risk for aberrant behavior because normative behavior does 
not produce the desired amounts of reinforcement. People who are intolerant of 
even everyday amounts of anxiety or frustration also are at risk for aberrant 
behavior. Individual differences in desired amounts of particular reinforcers 
may predict person-environment interactions, risk factors for psychopathology, 
and the occurrence of generalized and durable therapy effects versus the 
occurrence of relapses. Parallel predictions are made for individual 
differences in tolerance of aversive stimuli. Implications are discussed for 
applied behavior analysis, the development of psychopathology, and treatment 
strategies.

INTRODUCTION

Anxiety sensitivity refers to individual differences in what people think will 
happen to them when they experience anxiety (Reiss & McNally, 1985; Reiss, 
1991). People with high anxiety sensitivity believe that the experience of 
anxiety will cause them bodily or psychological harm, whereas people with low 
anxiety sensitivity believe that anxiety is just an unpleasant but harmless 
emotion that readily dissipates. Theoretically, people with high anxiety 
sensitivity have a low capacity to cope with anxiety and are at risk of 
developing Panic Disorder, other anxiety disorders, and many ordinary fears 
(Reiss, Peterson, Gursky & McNally, 1986; Reiss, 1991; Taylor, Koch & McNally, 
1992). On the other hand, people with low anxiety sensitivity have a high 
capacity to cope with anxiety.

The concept of anxiety sensitivity (AS) has been extensively validated 
(McNally, 1994; Peterson & Reiss, 1992; Taylor, 1995). High scores on the 
Anxiety Sensitivity Index (ASI) are strongly associated with Panic Disorder 
(Cox, 1994;McNally, 1992;Peterson & Reiss, 1992), even when scores on a variety 
of alternative anxiety and fear measures are held constant. High AS has been 
found to be a risk factor for panic responses to biological challenges in 
laboratory situations (Holloway & McNally, 1987; Rapee, Brown, Antony & Barlow, 
1992; Telch & Harrington, 1994). AS is strongly related to fearfulness and only 
moderately related to frequency of anxiety and stress experiences (Reiss et 
al., 1986; McNally & Lorenz, 1987; Taylor, 1995). ASI scores at the end of 
treatment for Panic Disorder predict relapse better than alternative measures 
(Bruce, Spiegel, Gregg & Nuzzarello, 1995; Jones & Barlow, 1991). In addition 
to Panic Disorder and other anxiety disorders, AS is associated with alcoholism 
(Stewart, Knize & Phil, 1992; Stewart, 1994).

In this article, a general theory of human motivation is proposed. The idea of 
individual differences in reinforcement effectiveness, implicit in the concept 
of anxiety sensitivity, is applied to a broader list of fundamental motivators. 
Thus, this is a major expansion of previous theoretical ideas.

STATEMENT OF THEORY

The sensitivity theory of motivation may be viewed as a call for research on 
individual differences in what people want from their lives. Of course, 
psychologists already have provided considerable research on this topic. 
Sensitivity theory suggests, however, that the psychological analysis provided 
to date has been inadequate because of certain methodological limitations that 
are common in psychological research.

Psychologists generally have evaluated only one type of motivation at a time. 
Yet in everyday life people can pursue multiple types of reinforcements 
simultaneously, or they can switch from pursuing one type of reinforcement to 
another. For example, the person who reads a newspaper while eating is pursuing 
simultaneously both intellectual satisfaction and food. The person who puts 
down a newspaper to start a morning walk has switched from pursuing the 
satisfactions of intellectual activity to those of physical activity. Although 
such 'motivation switching' is a fundamental aspect of everyday behavior, there 
are few psychological efforts to account for it. Animal and human participants 
were not permitted to switch motivation in virtually all psychological 
experiments on motivation reported to date. By not studying in greater depth 
when people switch from seeking one reinforcer to another, psychologists may 
have underestimated the importance of individual differences in rates of 
satiation (individual differences in desired amounts of various reinforcers).

Another reason psychologists have underestimated individual differences in 
desired amounts of reinforcement concerns the tendency to study animals in 
deprivational states. Deprivation induces common motivation in animals who 
otherwise may have very different motivations. For example, the behavioral 
consequences of individual differences in appetite are temporarily obscured by 
deprivational procedures that make animals very hungry. The motivational 
principles that apply to starving animals may not generalize well to other 
animals who are not necessarily starving, and they may be even less applicable 
to the everyday lives of people. Whereas almost all starving people spend most 
of their time and energy searching for food, people who are not starving show 
considerable individual differences in the amount of time and energy devoted to 
the preparation and consumption of meals. Sensitivity theory reminds us that 
routinely inducing deprivational states obscures differences in how much 
reinforcement individuals want, a potentially important variable for 
understanding human motivation.

Our discussion of sensitivity theory begins with the concept of a reinforcement 
sensitivity, which is defined here as an individual difference in the 
reinforcing effectiveness of a fundamental motivator. The three key phrases in 
this definition are 'reinforcing effectiveness', 'individual difference', and 
'fundamental motivator'. The meaning of each of these phrases will be discussed 
in this article along with comments on resistance to satiation. In this 
discussion, the terms motivator and reinforcer will be used interchangably. 
Moreover, two types of reinforcers, called rewards and aversive stimuli, will 
be recognized.

The concept of reinforcement sensitivity has some similarities with the 
Hull-Spence concept of drive and with related concepts such as that of an 
'establishing operation' (Hull, 1952; Keller & Schoenfeld, 1950). The higher 
the degree of reinforcement sensitivity, the stronger is the associated 
motivational drive. A crucial difference, however, is that sensitivity refers 
to a stable individual difference, whereas drive and establishing operation 
refer to situational phenomena.

Reinforcing effectiveness

The phrase reinforcing effectiveness refers to the strength of a particular 
motivator for a particular individual (cf. Rescorla & Wagner, 1972). The more 
effective a given reinforcer, the stronger (higher) is the person's drive or 
motivational state. Deprivational states temporarily increase the effectiveness 
or motivational strength of a deprived reinforcer. For example, food 
deprivation increases the strength of the person's motivation to obtain food, 
social isolation increases the strength of the individual's motivation for 
companionship, and prolonged exposure to the same environment increases the 
drive for stimulus novelty.

Theoretically, the more effective a reward is for a particular person: (1) the 
larger is the amount of reinforcement needed to satiate the person; (2) the 
more intense and persistent is the person's seeking of reinforcement; (3) the 
more impatient the person is in waiting for reinforcement; and (4) the lower is 
the amount of reward that can function as reinforcement for instrumental 
behavior. The more effective a given aversive stimulus is for a particular 
person: (1) the lower is the person's threshold for performing coping/avoidance 
responses; (2) the more intense and persistent is the person's performance of 
coping/avoidance responses; (3) the more quickly the person will perform 
coping/avoidance responses; and (4) the lower is the amount of aversive 
stimulation that can function as negative reinforcement for instrumental 
behavior.

There are many everyday examples of these principles. Children who present 
clinically as chronically 'starved for attention' want large amounts of 
attention immediately. People with a low threshold for pain put off going to a 
dentist as long as they can and jump at the most minimal sensations when the 
dentist begins drilling. Gluttons become impatient waiting for dinner. A person 
with high anxiety sensitivity shows panic when biologically challenged 
(Holloway & McNally, 1987). People who are starving will work to obtain very 
small amounts of food if that is all that is available. People with high 
anxiety sensitivity avoid situations in which even minimal anxiety is expected.

Individual differences

Up to this point, reinforcement sensitivities have been discussed in terms of 
their similarities to states of deprivation. However, even though reinforcement 
sensitivities and states of deprivation have similar consequences (both induce 
drive), reinforcement sensitivities are not the result of states of 
deprivation. Consider the distinction between gluttony and hunger. A glutton is 
a person who habitually has a hearty appetite and overeats for pleasure--the 
dictionary indicates that gluttons are people who enjoy eating above other 
pleasures (Kipfer, 1993). Because gluttony is a personality (individual 
difference) concept, it applies to only some people. In contrast, hunger is a 
temporary situational state related mostly to how long it has been since one's 
last meal; the term hunger is not a personality factor and potentially applies 
to anyone who has not eaten in a while. Sensitivity theory is concerned with 
personality concepts such as gluttony, not with deprivational concepts such as 
hunger.

At first consideration, the idea of stable individual differences in the 
effectiveness of certain reinforcers may seem counterintuitive. After all, 
psychologists have long thought that virtually everybody seeks pleasure and 
avoids anxiety and pain. If almost everybody is motivated to obtain rewards and 
avoid aversive stimuli, the concept of 'individual differences' may seem to be 
insignificant.

Although all people are to some degree motivated to eat, the amount of time, 
effort, and persistence devoted to the pursuit of food may vary significantly 
from one individual to the next. The amount of food required for satiation 
varies considerably from one person to the next, even when deprivational 
factors are held constant. This is recognized in everyday life by references to 
some people as being 'good eaters' or having 'hearty appetites'. These phrases 
suggest recognition among lay people that there are stable individual 
differences in the motivational strength of rewards such as food. The plain 
fact is that some people just like eating much more than most people. 
Similarly, the amount of time, effort, and persistence people devote to the 
pursuit of happiness (positive mood) varies considerably from one individual to 
the next. The platitude 'everybody wants to be happy' trivializes potentially 
important individual differences in effort. Some people try to look at 
everything positively and make the most out of whatever happens. These people 
work at being happy and organize a large portion of their everyday lives to 
achieve it. Others make only token efforts to escape a life filled with 
burdens, boredom, or misery. These observations are made not as a value 
judgment on people's lives but as a factual statement that individuals differ 
considerably in the effort they make to experience positive moods.

At this point, our analysis of individual differences is descriptive rather 
than explanatory. The intent here is not to explain why some people eat more 
than others in terms of their having a stronger appetite or to explain why some 
people seek attention in terms of greater need. Rather, the intent simply is to 
observe that some people have hearty appetites, so that we may use this fact 
later to explain other psychological phenomena.

Individual differences in reinforcement sensitivity should be considered only 
with reference to a particular reward or aversive stimulus. It would be invalid 
to say something like, "Bill has higher reinforcement sensitivity than Jane". 
On the other hand, Bill might have a high reinforcement sensitivity for 
attention, and Jane might have a high reinforcement sensitivity for physical 
activity.

   Resistance to satiation. One difference between situationally-induced 
motivational states such as hunger, and motivational states associated with 
stable individual differences such as gluttony, is that satiation occurs much 
more readily in situation-induced states than in person-specific states. People 
who are deprived of food readily satiate when they eat a full meal. In 
contrast, after eating a large meal, gluttons are quick to crave food again. A 
high reinforcement sensitivity for food, as in gluttony and in other conditions 
such as Prader Willi Syndrome (a rare condition associated with hyptotonia, 
hypergonadism, extreme obesity, and sometimes mental retardation), implies 
resistance to satiation and a relatively quick reinstatement of motivational 
states following the consumption of reinforcement.

   Origins of reinforcement sensitivities. How does it happen that some people 
become gluttons and others become intellectuals? No specific hypotheses on the 
origins of reinforcement sensitivity will be advanced here. The question of 
what accounts for the individual differences in sensitivities will be left 
unanswered. It would be helpful to know what accounts for the occurrence of 
sensitivities, and why some people show much higher sensitivity than others. 
However, this knowledge is not essential to support the various research 
suggestions and hypotheses expounded in this article. It is not unusual for 
hypotheses of individual differences to be offered without hypotheses regarding 
origin. For example, the concept of intelligence has proven useful even though 
we do not fully understand the origins of individual differences in 
intelligence. It is possible that reinforcement sensitivity will prove to be a 
useful construct even though we presently do not know the origin of 
reinforcement sensitivities.

   Reinforcement preferences. Psychologists have long recognized individual 
differences in preferences for particular reinforcers. Applied behavior 
analysts routinely assess the individual's preferences for various reinforcers 
in order to select one for use in a contingency management program. Whereas one 
child might prefer to be reinforced by candy, another might prefer time in an 
enjoyable activity. To help therapists select motivators, some researchers have 
developed reinforcement checklists (Bihm, Poindexter, Kienlen & Smith, 1992).

Behavior analysts have recognized a relationship between reinforcement 
preference and reinforcement effectiveness. The primary reason for assessing an 
individual's preference for reinforcement is to maximize the effectiveness of a 
contingency management program by using the most effective reinforcer for a 
particular person. For example, if a person prefers physical activity to adult 
attention, the opportunity for physical activity may function as the more 
effective reinforcer in contingency management programs.

The Premack Principle is relevant to these comments (Premack, 1959, 1965). This 
principle suggests that reinforcement preference in a free operant situation is 
a measure of reinforcement effectiveness. The Principle holds that access to a 
more highly preferred activity can be used to reinforce time in a less 
preferred activity but not vice versa. The Principle has been supported by 
numerous studies with both animal and human Ss, although some have suggested 
alternative interpretations (Dunham, 1977).

What is new or different about the concept of reinforcement sensitivity, given 
the Premack Principle and the recognition of individual differences in 
reinforcement preferences? In the past, reinforcement preferences were 
considered technical details that were assessed in order to maximize the 
effectiveness of contingency management programs. If it were found that a child 
had a strong preference for adult attention, attention was used to reinforce 
operant behavior. However, no effort was made to modify the strength of the 
individual's attraction for attention. Rarely have psychologists assessed 
reinforcement preferences in order to modify those preferences or to modify the 
reinforcing effectiveness of a particular reward or aversive stimulus. This is 
what is called for under sensitivity theory, as explained later in this 
article.

Fundamental motivators

Fundamental motivators are defined here as conceptually distinct reinforcers 
relevant to understanding a significant amount of behavior displayed by a large 
percentage of all people. The criterion of conceptual distinctiveness means 
that fundamental motivators cannot be analyzed entirely as combinations of 
other sources of motivation. For example, swimming is not a fundamental 
motivator partially because it can be reduced to a more fundamental motivator, 
physical activity. On the other hand, moral behavior is suggested to be a 
fundamental motivator. This suggestion is based on the view that the desire to 
behave morally cannot be explained entirely in terms of avoidance of punishment 
for immoral behavior. If future research were to find that this desire is 
ultimately reducible to the desire to avoid punishment, moral behavior no 
longer would be considered to be a fundamental motivator.

Each fundamental motivator induces a drive in a great many people. Fundamental 
motivators are either rewarding or aversive for almost everybody. They are 
distinguished from nonfundamental motivators, which are rewarding or aversive 
for relatively few people. For example, sexual pleasure is suggested to be a 
fundamental motivator because it is motivating for almost everybody, whereas 
reading mysteries is not suggested to be a fundamental motivator because it is 
reinforcing for only a relatively small percentage of all people.

Fundamental motivators reinforce substantial amounts of behaviors in everyday 
lives, and/or they reinforce behaviors that generally are considered to be 
important or significant. Food is suggested to be a fundamental motivator 
because people expend a great deal of time and effort preparing and consuming 
meals and because overeating and undereating are major health and psychological 
issues. On the other hand, drinking water is not suggested to be a fundamental 
motivator because people expend little effort in their everyday lives to obtain 
water and because the associated seeking behaviors are considered relatively 
unimportant in psychology.

Sensitivity theory suggests a need for research to identify the fundamental 
motivators of humankind. What rewards are to some degree desired by virtually 
everybody? What aversive stimuli are to some degree avoided by virtually 
everybody? How do various motives relate to one another and which ones can be 
analyzed in terms of others? Although a number of personality theorists have 
attempted to address these questions, especially Murray (1938), researchers 
have produced surprisingly few efforts to provide a comprehensive listing of 
fundamental motives. Most personality theories do not provide a comprehensive 
account of fundamental motives, identifying at most only a few of them.

A preliminary list of fundamental motives is presented in Table 1. This list 
represents the authors' own views based on an analysis of the motivational 
constructs discussed in major personality theories (Hall & Lindzey, 1957). 
Since the main purpose of sensitivity theory is to identify issues for future 
research, this list is offered as an example of the sort of analysis we hope 
someday will result from empirical research. Although we have given our list 
considerable thought--this is by no means a hastily developed or arbitrary 
list--we recognize that the list is at the moment without empirical support and 
that many readers will take issue with at least some of the items on the list. 
Diverse opinions on the subject are welcome, provided we all agree that the 
issues need to be studied empirically.

Table I. Preliminary list of fundamental motives

Acceptance/Success The satisfaction associated with acceptance by
others.
Anger/Frustrative Impulses The discomfort associated with aggressive or
frustrative impulses that occur in response to a
perceived threat to self-esteem or to one's physical well-being.
Anxiety The belief that experiencing anxiety is personally harmful.
Attention The satisfaction derived from adult consideration of the
individual.
Companionship The satisfaction derived from spending time in the company
of other people.
Competence (Mastery) The satisfaction derived from competence in the
performance of a skill.
Curiosity The satisfaction derived from exploration of novel stimuli.
Dominance (Leadership) The satisfaction derived from directing or
influencing the behavior of others.
Food The satisfaction derived from eating.
Help Others The satisfaction derived from providing assistance to a
personal friend.
Help Society The satisfaction derived from contributing to
society/public welfare.
Independence The satisfaction derived from doing things on one's own
without assistance or interference from
others.
Love/Romance The satisfaction associated with romance.
Morality The satisfaction derived from behavior in accordance with a
code defining right versus wrong
conduct.
Nuturance The satisfaction derived from taking care of people, animals,
or plants.
Order The satisfaction derived from an organization of time, events, or
things into a well-defined pattern
in one's everyday life.
Physical Activity The satisfaction derived from exercise of the body.
Physical Pain The discomfort associated with bodily injury (tissue
damage), inflammation, or bodily spasms.
Positive Mood The satisfaction derived from states of positive mood, as
in happiness and optimism.
Positive Self-Regard The satisfaction associated with positive
self-esteem.
Self-Control The satisfaction associated with mastery over one's
impulses, social reactions, and emotional
reactions.
Sexual Gratification The satisfaction derived from real or fantasized
sexual intercourse or foreplay.
Social Conflict The discomfort associated with social strife such as
arguments, disagreements, and opposition.
Survival The desire to stay alive.
Vengeance The satisfaction derived from retaliation for those real or
imagined acts of others that the person
perceives to be offensive, aggressive, or threatening.


There are many aspects of Table 1 that are subject to alternative viewpoints. 
For example, some readers may question the suggestion that helping others and 
helping society are conceptually distinct reinforcers rather than substantially 
related motives. We suggest distinct motivators because we have observed people 
who seem to enjoy helping friends but pay little attention to social issues, 
and vice versa. Whether or not this view will be confirmed by factor analysis 
or other research remains to be seen.

The task of generating a comprehensive list of fundamental motivators is 
challenging. The main difficulty is to generate a list that meets the criteria 
of conceptual distinctiveness and is comprehensive in its totality. This cannot 
be accomplished by developing a list of biological drives. As suggested by the 
list presented in Table 1, some fundamental motivators are not biological 
drives. Moreover, the motivational effects of biological drives are influenced 
by culture, personality, and learning. For example, the motivational properties 
of sex for any given person reflect both the person's biological drive and the 
person's culture, attitudes, morality, and past experiences with sex. The 
motivational implications of food may be influenced considerably by 
self-perceptions of being a 'skinny' or 'fat' person. Generally, fundamental 
motivation is determined by a combination of both biological and psychological 
factors.

   Individual differences. Individuals show important differences in the 
strength of the motivations listed in Table 1. Some people are extremely 
interested in surviving, whereas alcoholics drink themselves to death. Some 
people panic in anxious situations, whereas others readily approach them. 
American football players withstand considerable physical pain that the more 
squeamish among us would avoid at all cost. To the extent that psychological 
theories of motivation have all people equally motivated to seek pleasure, or 
equally motivated to avoid anxiety, these theories are in disagreement with 
sensitivity theory.

Researchers who disagree with the assumption of individual differences should 
be careful to avoid arguing a tautology. For example, the argument that 
everybody seeks pleasure can become tautological if work is defined as pleasure 
for workaholics. If both the hedonist and the workaholic are viewed as seeking 
pleasure, important differences in motivation are trivialized. The hedonist 
attends every party possible, whereas the workaholic can hardly relax and stop 
working long enough to enjoy a single party. Rather than say that both seek 
pleasure, sensitivity theorists say that both seek reinforcement, but that what 
reinforces the two is very different. The hedonist has a high reinforcement 
sensitivity for sexual pleasure, whereas the workaholic has a high 
reinforcement sensitivity for the pleasures of intellectual activity, physical 
activity, and/or helping society.

   Set points. Under sensitivity theory every person has a set point for each 
fundamental motivator. The set point indicates either the amount of reward an 
individual desires or the strength of an aversive stimulus an individual will 
tolerate. A person with an aberrantly high or low set point is said to have a 
reinforcement sensitivity for that motivator. Consider the example of 
companionship. The hypothesis of set points implies individual differences in 
how much companionship people seek in their daily lives. People with high set 
points seek a great deal of company, whereas those with low set points prefer 
to be alone quite a bit. If the amount of companionship in one's everyday life 
is less than that indicated by the set point, the individual is temporarily 
motivated to seek additional amounts of companionship (now a positive 
reinforcer). If the amount of companionship in one's everyday life is more than 
that implied by the set point, the individual is temporarily motivated to avoid 
(decrease the amount of) companionship (temporarily a negative reinforcer). 
Depending on one's set point, a fundamental reinforcer may be pleasant or 
annoying. For example, attention is positively reinforcing for some (show-offs) 
and negatively reinforcing for others (shy people). Under sensitivity theory, 
show-offs have high set points for attention; that is, they require large 
amounts of attention before satiating. In contrast, shy people have low set 
points for attention; that is, they desire small amounts of attention and find 
large amounts annoying. Although the details will not be developed here, the 
concept of set point may be relevant to explaining motivation switching 
behavior in which a person seeks one type of reinforcer at one point and 
another a moment later. As previously noted, motivation switching is a basic 
phenomenon of behavior, but it has received surprisingly little attention. 
Theoretically, the key to such behavior for any individual may be the relative 
strength/set point of each motivation as compared with the rate at which each 
type of reward and aversive stimulus is experienced, with the greatest 
discrepancies controlling the person's behavior. Since such discrepancies 
should vary from one moment to the next, people may switch from seeking one 
reinforcer to seeking another, or they may seek both simultaneously.

   Measurement. A number of alternative methods may be applicable to assessing 
reinforcement sensitivities. One possibility is to assess free-operant 
preferences, such as the amount of time a person chooses to spend in an 
activity when many choices can be made (Premack, 1959, 1965). However, such 
assessments would need to be standardized across individuals to identify 
persons who are high and low on various reinforcer preferences.

The Motivation Assessment Scale (MAS: Durand & Crimmins, 1988) is a popular 
scale for assessing four of the most common motivations of self-injurious 
behavior (sensory, escape, attention, tangible). However, the MAS has not been 
standardized. Moreover, the psychometric properties of the MAS are limited by 
the fact that it has only four items on each of four subscales. As already 
noted, the Anxiety Sensitivity Index (ASI: Reiss et al., 1986; Peterson & 
Reiss, 1992) is a 16-item, self-report measure that has been extensively 
validated (McNally, 1994; Taylor, 1995). Researchers have also developed a 
Child ASI (Silverman, Lesig, Rabian & Peterson, 1991). The authors currently 
are developing self-report and informant-rating instruments for assessing 
reinforcement sensitivities, one for the general population and the other for 
the population of people with mental retardation. The instruments will provide 
MMPI-like, individual profiles of set points on fundamental motivators derived 
from a factor analysis.

The successful development of measures of reinforcement sensitivity may lead to 
many new research opportunities. To the extent to which people seek out what 
they want, measures of reinforcement sensitivities may be helpful in predicting 
some person-environment interactions. Additionally, psychopathology researchers 
may wish to compare standardized motivation profiles obtained from different 
diagnostic groups. For example, it would be interesting to compare motivational 
profiles associated with the various personality disorders.


IMPLICATIONS

Sensitivity theory potentially has broad implications. For the sake of brevity, 
however, our present discussion of implications will be limited to 
self-injurious behavior in persons with mental retardation, the development of 
aberrant behavior, and clinical strategies.

Functional analysis

In the 1970s a number of researchers found that different cases of 
self-injurious behavior were associated with any of four different motivations. 
Different people with self-injury are motivated by attention, internal 
stimulation ('self-reinforcing' behavior), escape from frustrative task 
demands, and tangible reinforcers (Carr, 1977; Napolitan, 1979). These findings 
stimulated the use of an assessment technique called functional analysis 
(Iwata, Dorsey, Slifer, Bauman & Richman, 1982; Kanfer & Saslow, 1969). This 
technique provides an empirical method for classifying the motivational source 
of specific examples of self-injury. As Iwata et al. (1982) put it,

Carr (1977) indicated that the behavior may be reinforced through extrinsic 
sources (e.g. through positive reinforcement such as attention, or negative 
reinforcement such as the termination of demands), or that the behavior itself 
may produce some form of intrinsic reinforcement (e.g, sensory stimulation, 
pain reduction). This conceptualization of self-injury as a multiply controlled 
operant would indicate that no single form of treatment can be expected to 
produce consistent positive results, and it suggests that one means of 
selecting a potentially effective treatment would consist of first determining 
what events are currently maintaining the behavior.

As the preceding quotation indicates, functional analysis is used to help 
select an operant treatment method. The idea is first to identify the nature of 
the motivation and then to use that information to extinguish the 
self-injurious response(s). For example, if it were found that self-injury is 
being motivated by attention, the therapist might ignore future instances of 
self-injury while attending to alternative behaviors. In this manner, the 
therapist would attempt to extinguish self-injurious behavior while 
strengthening an alternative behavior. On the other hand, if it were found that 
self-injury is being motivated by escape from frustrative task demands, the 
therapist might attempt to extinguish self-injury by never allowing escape when 
this behavior occurs.

Sensitivity theory may provide a basis for strengthening functional analysis by 
focusing attention on the role of individual differences in operant 
conditioning. Sensitivity theory suggests that what is aberrant is not only the 
self-injurious behavior but also the individual's sensitivity to reinforcing 
attention or to frustrative task demands. Thus, treatment should be aimed not 
only at extinguishing the self-injurious behavior but also at reducing the 
aberrant reinforcement sensitivities that may create predispositions for 
self-injurious behavior.

Alcohol addiction provides a helpful analogy of some of the main points to be 
made here about reinforcement sensitivity and functional analysis. Suppose that 
therapists at a substance abuse clinic determine that an individual is stealing 
money in order to support a drinking habit rather than a heroin habit or abuse 
of some other substance. In theory, the stealing could be extinguished either 
by giving the individual free alcohol or by controlling the individual's 
environment so that stealing money cannot lead to drinking. Although the 
treatment might effectively extinguish stealing behavior under circumstances 
where the therapist can control the environment, it would not treat the 
person's addiction to alcohol. As soon as the therapist's control of the 
consequences of stealing weakens, the individual will return to stealing to buy 
alcohol.

People with a high reinforcement sensitivity may be thought of as having a 
'psychological addiction' to a reinforcement. For example, people who engage in 
self-injury to obtain high amounts of attention may be considered 
'psychologically addicted' to attention. Just as people addicted to alcohol 
seek aberrant quantities of alcohol, people addicted to attention seek aberrant 
amounts of attention. Effective treatment of alcoholism requires a reduction in 
the underlying need (addiction) for alcohol. Effective treatment of 
self-injurious behavior motivated by attention sensitivity may require a 
reduction in the underlying craving (seeking) for attention. That is, effective 
treatment must do more than teach people that self-injury no longer leads to 
attention (extinction); it also should reduce the motivation for aberrant 
amounts of attention.

What does it mean to say that some people have a 'psychological addiction' for 
attention? Technically, it means that they have a high degree of reinforcement 
sensitivity for attention, as that term is defined herein. The addiction 
metaphor works partially because addictions are person-specific variables 
associated with personal suffering and indicating a need for treatment. The 
addiction metaphor also works because it implies a problem in the amount of 
substance the individual wishes to consume. The problem in alcoholism is not 
that the person drinks but that the person drinks too much. Similarly, 
sensitivity theory suggests that the problem in some examples of aberrant 
behavior is not that the person seeks attention (or some other fundamental 
reinforcement) but that the person seeks too much attention.

These theoretical issues suggest a number of possible future research issues. 
If motivations for self-injury are determined by high reinforcement 
sensitivities, the following should be true. Engaging in self-injurious 
behavior for attention (or to avoid frustrative task demands) should be part of 
a general pattern of behavior in which the individual engages in a variety of 
aberrant behaviors in order to obtain attention (or avoid frustration). The 
attention-seeking should present as a stable individual difference in 
motivation that has been true of the individual for a long time, is difficult 
to modify, and usually will last indefinitely unless specifically reduced by 
successful treatment methods. Just as anxiety sensitivity levels at the end of 
treatment have been found to predict relapse rates for patients successfully 
treated for anxiety disorder (Bruce et al., 1995; Jones & Barlow, 1991), 
reinforcement sensitivity levels at the end of treatment may predict relapse 
rates for extinguished self-injurious behaviors.

Development of behavior disorders

When a person desires an unusually large or unusually small amount of 
reinforcement, that person is said to be aberrantly motivated. Aberrant 
motivation (very high reinforcement sensitivity or very low reinforcement 
sensitivity) is assumed to result from complex interactions among biological, 
developmental, conditioning, and cognitive factors. Theoretically, aberrant 
motivation is a risk factor for aberrant behavior. That is, the amount of 
reinforcement a person seeks (how much reinforcement is required to produce 
satiation) may be the key to understanding the development of aberrant behavior 
in at least some, if not many, people. Appropriate behavior usually is followed 
by small or moderate amounts of reinforcement, whereas inappropriate behavior 
sometimes is the best strategy for obtaining high amounts of reinforcement. 
Since most people are satisfied with the amount of reinforcement that follows 
socially appropriate behavior, most people learn socially appropriate 
responses. However, people with high reinforcement sensitivity seek much higher 
amounts of reinforcement than most people. These people are not satisfied with 
the amount of reinforcement that follows socially appropriate behavior. For 
these people, socially inappropriate behavior sometimes offers the best 
strategy to obtain immediately a high amount of reinforcement.

Consider the example of a boy with a high sensitivity for attention. By 
definition, the boy should behave as if he is chronically 'starved' for 
attention. The child should show vigorous efforts to obtain as much attention 
as immediately as possible. For this child, engaging in inappropriate behavior 
may be an effective strategy to obtain quickly a large amount of reinforcement.

This theory is different from previous learning theory explanations of the role 
of reinforcement in the development of behavior disorders. Past explanations 
held that some children learned aberrant behavior when it was inadvertently 
reinforced. The pathogenic agent was in the environment in the form of 
response-reinforcement contingent relationships that inadvertently favored the 
learning and performance of aberrant behavior. In contrast, reinforcement 
sensitivity theory holds that the pathogenic agent sometimes is partially 
'within' the person, not the environment. That is, people with a high 
reinforcement sensitivity react to the same environment very differently than 
other people.

Sometimes people develop aberrant behavior because they want something too 
badly or too quickly. A person who has not eaten in a long time starts thinking 
about food constantly so that obtaining food starts to dominate the person's 
behavior. In a similar fashion, people who develop aberrant behavior may want 
to be loved too much, may need companionship too often, may need to escape 
immediately from frustration, or may desire pleasure all the time. At times 
these motivations may be extremely strong and dominate the person's behavior.

Treatment

Therapists should assess the degree to which a person's aberrant behavior may 
be related to aberrant motivation and directly treat the aberrant motivation as 
one component of an overall therapy plan. A number of direct treatment 
strategies may be attempted depending on the individual and circumstances. 
Possible strategies to be explored by future researchers include the following:

  (a) Cognitive therapy may alter reinforcement sensitivities by changing 
attitudes and beliefs about the consequences of various reinforcers. For 
example, anxiety sensitivity is reduced when people believe that the experience 
of anxiety will not harm them.

  (b) Psychotropic drugs may alter reinforcement sensitivities by affecting both 
biological processes and cognitive expectancies. These drugs should be 
evaluated for outcomes on aberrant motivation distinct from their effects on 
aberrant behavior.

  (c) Research is needed on the effects of the length of treatment on changes in 
aberrant motivation. Whereas short-term approaches sometimes may lead to 
improvements in aberrant behavior, long term trials of behavior therapy may be 
needed to change aberrant motivation. Reinforcing effectiveness may change if 
people adjust to fixed quantities of a reinforcer given over a long period of 
time. For example, if a glutton consistently ate a moderate amount of food, 
eventually the person's appetite may adjust to a lower amount of food.

  (d) It may be possible to increase the effectiveness of a reinforcer by 
pairing it with a more effective reinforcer in Pavlovian trials. For example, a 
therapist might attempt to treat social isolation behavior by pairing 
companionship with a more effective reinforcer for the person.

  (e) Desensitization and various exposure techniques sometimes may reduce high 
sensitivities for anxiety (Harrington & Telch, 1994) and perhaps frustration.

  (f) When ethical, counterconditioning may be an appropriate strategy for 
decreasing reinforcement sensitivity by pairing a reinforcer with an aversive 
stimulus (Bandura, 1969). For example, a therapist might decrease a high 
reinforcement sensitivity for sex by pairing fetish stimuli with aversive 
stimuli.

In contrast, the following strategies seem less likely to alter reinforcement 
sensitivities:

  (a) Teaching people socially appropriate alternative behaviors may have little 
impact in changing what they want. This strategy may work when skill deficits 
are assessed to be the main problem, but not when aberrant motivation is 
implicated. For example, teaching a thief appropriate job skills is of little 
benefit because there are few jobs that can produce as much money as quickly as 
robbing banks.

  (b) Extinction strategies may work well when the problem is a maladaptive 
response-reinforce- ment contingency but not when aberrant motivation is 
implicated. There is no particular reason to assume that simple extinction 
alters what people want; it only alters the behaviors that produce what they 
want.

  (c) Punishment strategies are unlikely to work when aberrant motivation is 
associated with a problem behavior. Punishment temporarily may suppress 
maladaptive behavior by creating a stronger motivation (avoidance of 
punishment), but the person's aberrant motivation remains unchanged and will 
stimulate maladaptive behavior in situations in which punishment is not 
anticipated.

Because high reinforcement sensitivities are predicted to be risk factors for 
psychopathology, aberrant motivation must be successfully treated if the 
therapy is to have durable and generalized benefits. If aberrant behavior is 
modified but aberrant motivation is left unchanged, the person is at risk for 
relapse because he/she will crave high amounts of some fundamental reinforcer. 
The person probably will not be able to obtain high amounts of reinforcement by 
behaving in socially appropriate ways and, therefore, is at risk to resort to 
aberrant behavior in an effort to obtain immediately the desired amounts of 
reinforcement. Thus, measures of aberrant motivation at the end of treatment 
may predict relapse rates. Researchers of anxiety disorders have obtained some 
data supporting this hypothesis. Reduction of anxiety sensitivity has been 
found to be an important factor in minimizing the risk of relapse in patients 
with Panic Disorder (Bruce et al., 1995; Jones & Barlow, 1991).

CONCLUSIONS

Generally, three limitations may be noted to previous research on motivation. 
First, psychologists have widely assumed but not critically examined the 
hypothesis that everybody is equally motivated to seek pleasure/happiness and 
equally motivated to avoid anxiety. People actually show wide individual 
differences in the strength of these motives. Second, psychological theories of 
motivation have been unduly influenced by the study of animals in deprivational 
states. This method has obscured the role of individual differences in desired 
amounts of various reinforcers. Whereas starving animals spend almost all of 
their time and energy seeking large quantities of food, most people spend 
widely varying amounts of time and energy pursuing a much broader range of 
reinforcers. Third, psychologists have studied only one motivation at a time. 
This has obscured the importance of individual differences in rates of 
satiation. Whereas individual differences in satiation rates may seem 
unimportant when only one motivation is considered, they seem relevant to 
explaining motivation switching in which a person changes the type of 
reinforcement he/she is pursuing.

The sensitivity theory of motivation has potentially important implications for 
future research. First, the theory identifies individual differences in desired 
amounts of reinforcement (individual differences in rates of satiation) as an 
understudied and potentially important variable. Because people spend 
considerable time and energy seeking the reinforcers they desire, these 
individual differences may predict some person-environment interactions. 
Second, the theory suggests the need for research to identify fundamental 
sources of motivation. Surprisingly, psychologists have proposed very few 
comprehensive lists of fundamental motivation. Third, a new theory on the 
development of psychopathology is suggested based on the assumption that 
different people not only desire different types of reinforcers but also desire 
widely varying amounts of each fundamental reinforcer. Aberrant motivation is 
indicated when a person wants too much of a particular type of reinforcement 
and/or is too intolerant of everyday levels of some aversive stimulus such as 
anxiety or frustration. Aberrant motivation may be a risk factor for 
psychopathology because people usually cannot obtain high amounts of 
reinforcement by behaving in socially appropriate ways. People sometimes resort 
to inappropriate behavior as the best strategy to obtain a high amount of 
immediate reinforcement. However, aberrant behavior rarely leads to enough 
reinforcement to satiate aberrant desires; people with aberrant motives rarely 
obtain what they want and are at risk for unhappiness.

Sensitivity theory has implications for practice. First, the development of 
standardized measures that produce MMPI-like profiles of motivational set 
points may help clinicians identify aberrant motivation and assess its role in 
any particular case. To date, psychologists have produced thousands of 
standardized measures, not one of which provides a comprehensive, standardized 
profile of what motivates a therapy patient. The development of standardized 
measures will help clinicians assess the extent to which a person's 
psychopathology may be related to wanting too much love or too much attention 
or too much acceptance. Second, the theory suggests reconsideration of when to 
use certain clinical strategies. These strategies must be evaluated not only 
for their effects on response-reinforcement contingencies but also for their 
effects on aberrant motivation. For example, teaching people socially 
appropriate skills to obtain desired reinforcers has been a popular strategy. 
Although this strategy may be effective when the primary problem is a lack of 
skill, it may be ineffective when an important part of the person's problem is 
aberrant motivation. Teaching thieves employment skills may not work for those 
who crave immediate riches; in such cases, it may be necessary to reduce the 
greed for a durable and generalized treatment benefit. Finally, sensitivity 
theory provides a basis for strengthening behavior analysis. In cases of 
self-injurious behavior, for example, some people are motivated to escape 
frustrative task demands, and they may be intolerant of even minor 
frustrations.

These considerations should be regarded as theoretical suggestions, not 
prescriptions for therapy. Sensitivity theory is in an early stage of 
development. Researchers are encouraged to take advantage of the theory's 
unique perspective to identify future areas of research and to test some of the 
predictions suggested in this article.

Acknowledgement--The authors express their appreciation to Ellen Langer for her 
comments on the list of fundamental motivators.

REFERENCES

Bandura, A. (1969). Principles of Behavior Modification (pp. 505-509). New 
York: Holt.

Bihm, E. M., Poindexter, A. R., Kienlen, T. & Smith, B. L. (1992). Staff 
perceptions of reinforcer responsiveness and aberrant behavior in people with 
mental retardation. Journal of Autism and Developmental Disorders, 22, 83-93.

Bruce, T. J., Spiegel, D. A., Gregg, S. F. & Nuzzarello, A. (1995). Pediatrics 
of alprazolam discontinuation with and without cognitive behavior therapy in 
panic disorder. American Journal of Psychiatry, 152, 1156--1160.

Carr, E. G. (1977). The motivation of self-injurious behavior: a review of some 
hypotheses. Psychological Bulletin, 84, 800-816.

Cox, B. (1994). Anxiety sensitivity and anxiety disorder. Paper presented at 
the annual meeting of the Association for the Advancement of Behavior Therapy, 
San Diego, CA.

Dunham, P. (1977). The nature of reinforcing stimuli. In Honig, W. K. & 
Staddon, J. E. R. (Eds), Handbook of Operant Behavior. Englewood Cliffs, N J: 
Prentice-Hall.

Durand, V. M. & Crimmins, D. B. (1988). Identifying the variables maintaining 
self-injurious behavior. Journal of the Association for the Severely 
Handicapped, 6, 17-22.

Hall, C. S. & Lindzey, G. (1957). Theories of Personality. New York: Wiley.

Harrington, P. J. & Telch, M. J. (1994). Lowering anxiety sensitivity in 
non-clinical subjects. Paper presented at the 1994 annual meeting of the 
Association for the Advancement of Behavior Therapy, San Diego, CA.

Hull, C. L. (1952). A Behavior System: An Introduction to Behavior Theory 
Coneerning the Individual Organism. New Haven: Yale Univ. Press.

Holloway, W. & McNally, R. J. (1987). Effects of anxiety sensitivity on the 
response of hyperventilation. Journal of Abnormal Psychology, 96, 330-334.

lwata, B. A., Dorsey, M. F., Slifer, K, J., Bauman, K. E. & Richman, G. S. 
(1982). Toward a functional analysis of self-injury. Analysis and Intervention 
in Developmental Disabilities, 2, 3-20.

Jones, J. C. & Barlow, D. H. (1991, November). The relationship between fear of 
fear~anxiety, diagnosis, and treatment outcomes. Paper presented at the 1991 
annual meeting of the Association for the Advancement of Behavior Therapy in 
New York, NY.

Kanfer, F. H. & Saslow, G. S. (1969). Behavioral diagnosis. In Kanfer, F. H. & 
Saslow, G. (Eds), Behavioral Therapy: Appraisal and Status (pp. 417---444). New 
York: McGraw Hill.

Keller, F. S. & Schoenfeld, W. N. (1950). Principles of Psychology. New York: 
Appleton-Century-Crofts.

Kipfer, B. A. (Ed.) (1993). Roger's 21st Century Thesaurus in Dictionary Form. 
New York: Dell.

McNally, R. J. (1992). Anxiety sensitivity distinguishes panic disorder from 
generalized anxiety disorder. Journal of Nervous and Mental Disease, 180, 
737-738.

McNally, R. J. (1994). Panic Disorder: A Critical Analysis. New York: 
Guildford.

McNally, R. J. & Lorenz, M. (1987). Anxiety sensitivity in agoraphobics. 
Journal of Behavior Therapy and Experimental psychiatry, 18, 3-11.

Murray, H. A. (1938). Explorations in Personality. New York: Oxford.

Napolitan, J. T. (1979). The classification of self-injurious behavior in 
mentally retarded children. Unpublished doctoral dissertation, Department of 
Psychology, University of Illinois at Chicago.

Peterson, R. A. & Reiss, S. (1992). Anxiety Sensitivity Index Test Manual. 
Worthington, OH: IDS Publishing Corporation.

Premack, D. (1959). Toward empirical behavior laws, I: positive reinforcement. 
Psychological Review, 66, 219-233.

Premack, D. (1965). Reinforcement theory. In Levine, D. (Ed.), Nebraska 
Symposium on Motivation. Lincoln: Univ. of Nebraska Press.

Rapee, R. M., Brown, T. A., Antony, M. M. & Barlow, D. H. (1992). Response to 
hypervcntilation and inhalation of 5.5% carbon dioxide-enriched air across the 
DSM-III-R anxiety disorders. Journal of Anxiety Disorders, 4, 171-181.

Reiss, S. (1991). Expectancy model of fear, anxiety, and panic. Clinical 
Psychology Review, II, 141-153.

Reiss, S. & McNally, R. J. (1985). Expectancy model of fear. In Reiss, S. & 
Bootzin, R. R. (Eds), Theoretical Issues in Behavior Therapy. San Diego, CA: 
Academic Press.

Reiss, S., Peterson, R. A., Gursky, D. M. & McNally, R. J. (1986). Anxiety 
sensitivity, anxiety frequency, and the prediction of fearfulness. Behavior 
Research and Therapy, 24, !-8.

Rescorla, R. A. & Wagner, A. R. (1972). A theory of Pavlovian conditioning: 
variations in the effectiveness of reinforcement and nonreinforcement. In 
Black, A. & Prokasy, W. F. (Eds), Classical Conditioning lI: Current Research 
and Theory. New York: Appleton-Century-Crofts.

Silverman, W. K., Lesig, W., Rabian, B. & Peterson, R. A. (1991). Childhood 
Anxiety Sensitivity Index. Journal of Child Clinical Psychology, 20, 162-168.

Stewart, S. H. (1994). Anxiety seasitivity and alcohol misuse. Paper presented 
at the annual meeting of the Association for the Advancement of Behavior 
Therapy, San Diego, CA.

Stewart, S. H., Knize, K. & Phil, R. O. (1992). Anxiety sensitivity and 
dependency in clinical and non-clinical panickers and controls. Journal of 
Anxiety Disorders, 6, 119-131.

Taylor, S. (1995). Anxiety sensitivity: theoretical perspectives and recent 
findings. Behaviour Research and Therapy, 33, 243-258.

Taylor, S., Koch, W. J. & McNaily, R. J. (1992). How does anxiety sensitivity 
vary across the anxiety disorders? Journal of Anxiety Disorders, 7, 249-259.

Telch, M. J. & Harrington, P. J. (1994). The role of anxiety sensitivity and 
expectedness of arousal in mediating emotional response to inhalation of 35% 
carbon dioxide. Paper presented at the annual meeting of the Association for 
the Advancement of Behavior Therapy, San Diego, CA.