FThis article is for use only of students enrolled in GPS 520: Advanced Social Psychology for purposes associated with this course and may not be retained or further disseminated.

 

Rabinowitz, V. C., Zevon, M. A., & Karuza, J. (1988).  Psychotherapy as helping: An attributional analysis. In Abramson, L. (Ed.) Social cognition and clinical psychology: A synthesis. New York, NY: Guilford Press.

 

 

Psychotherapy as helping: An attributional analysis.

Vita c. Rabinowitz

Michael A. Zevon

Jurgis Karuza, Jr.

The purpose of this chapter is to consider various psychotherapeutic systems as helping interventions and to examine the role of the therapist's attributional judgments on the process of therapy. Construing different psychotherapies as special cases of the broad category of helping allows us to describe them in terms of four basic helping models proposed by Brickman et al. (1982). We view psychotherapeutic approaches as differing importantly in the therapist's attributional judgments about the client's responsibility for the origin of his or her problem as well as for the solution to it. The implicit or explicit attributional assumptions of particular psychotherapies may cause each therapy to be more or less appropriate in treating various psychological disorders. In contrast to existing psychotherapies that rely on a single attributional perspective throughout the course of treatment, we suggest that at times it may be useful to shift perspectives in viewing the causes of and solutions to the client's problems as therapy progresses.

 

PSYCHOTHERAPY AS HELP

In its early form, psychotherapy consisted primarily of a single therapeutic approach employed in the treatment of a narrow range of specific and debilitating disorders (Freud, 1912/1958). Since this beginning, however, psychotherapy has evolved and expanded to encompass a vast number and diversity of therapeutic systems. In their contemporary form, these systems address a wide range of human problems, from specific behavioral disorders to the fostering of optimal human development and growth. As a number of reviews have noted (Corsini, 1973; Goldstein & Simonson, 197 1; Urban & Ford, 197 1), the days of psychotherapy qua psychotherapy are long past.

Despite the multiplicity of modern day therapeutic approaches, it is important not to overlook the fact that psychotherapy remains a helping process (Wills, 1982; Wolman, 1976). As Weiner (1975) has stated, psychotherapy "is an Interpersonal process in which one person communicates to another that he understands him, respects him, and wants to be of help to him" (p. 3). From within this framework, we view therapists as help givers who use their skills and training to bring about improvement in their clients' functioning. Clients are seen as recipients of help, individuals who experience and acknowledge, in some sense, an inability to resolve a particular problem or set of problems. We view this helping process as basic to psychotherapy, and see it operating across the theoretical boundaries of particular psychotherapeutic systems.

It is becoming apparent through recent empirical work, however, that the psychotherapeutic process is determined and influenced by a wide variety of factors. These elements range from nonspecific therapeutic ingredients (Goldstein & Stein, 1976), such as the interpersonal relationship between client and therapist (Fehrenbach & O'Leary, 1982; Schofield, 1964), to specific factors, such as the mediating goals and procedures inherent in a particular therapeutic orientation (Parloff, Waskow, & Wolfe, 1978). Together, these components structure the therapeutic process and determine its effectiveness. Given this multidimensional and multidetermined view of psychotherapy, no simple model of the therapeutic process is readily apparent.

Contemporary approaches to the analysis of the therapeutic process (e.g., Howard, Orlinsky, & Perilstein, 1976) have attempted to deal with this complexity by determining the relative contribution of various therapeutic components to the process and outcome of therapy. Taking their lead, we will focus on the therapists and how they translate the mediating goals and procedures of the therapeutic school they embody. Specifically, we will propose an attributional analysis of various therapeutic orientations and consider the attributional foundation of the clinical decisions made by the therapist in psychotherapy.

 

JUDGMENTS IN PSYCHOTHERAPY

Despite the admitted complexity of psychotherapeutic processes, we will focus here on the clinical decision-making basis from which the therapist makes critical decisions about the client and structures the process and pace of therapy. (Urban & Ford, 1971). On the basis of experience and training in psychotherapeutic systems, the therapist makes, implicitly or explicitly, decisions directed at issues such as whether to accept a client for therapy, the etiology of the presenting complaint, the appropriate interventions, and the criteria for judging the success of the treatment. The therapist must process information about the client and his or her complaint; in other words, the therapist must make a series of evaluations and judgments about the client.

A complicating factor in this view of psychotherapy is the variation of the therapist's role as dictated by the numerous schools of psychotherapy. For example, Gestalt, ratio nal-e motive, and psychoanalytic therapies stress the evaluative and interpretive role of the therapist. On the other hand, therapies such as Rogers's client-centered approach stress a nondirective, evaluation-free orientation. Indeed, client-centered practitioners pride themselves on their nonjudgmental approach to therapy and would perhaps bridle at any intimation that the therapist makes judgments concerning the client. It can be argued, however, on both empirical and logical grounds, that therapists of all persuasions do make judgments about clients and, further, that these judgments either implicitly or explicitly play an important role in psychotherapy. Strupp and colleagues (Strupp, 1978; Strupp & Wallach, 1973; Strupp & Williams, 1973) cogently make the point when they state that whatever the therapist's background or level of training, he or she cannot escape the necessity of (1) forming some notions or hypotheses about the patient's "problem" or difficulty, and (2) deciding what needs to be done to bring about an improvement in the patient's condition. In order to address this issue, a consideration of "judgment" is in order.

Judgments come in a variety of forms. On the one hand a judgment can involve an opinion held about another and, as such, possesses an evaluative or critical component. More pertinent to our analysis, however, is the sense of judgment that refers to an assessment or appraisal that is nonevaluative and instrumental. Judgments of this nature would seem to be endemic to psychotherapy, and, indeed, at the heart of each school of psychotherapy is a theory that defines a particular view of human nature. In attempting to explain behavior, each theory more or less explicitly guides therapists in their attempts to understand clients and their presenting problems. This guidance, in turn, sets up expectancies about clients' potentials and their ability to control their behavior. Implicitly, the theory serves as a prescriptive framework from which the therapist makes instrumental judgments and attributions about the nature of the presenting complaint and how to solve it.

While the proper role of evaluative judgments in psychotherapy may be moot, we view it as unrealistic to contend that instrumental judgments are not or should not be made. While psychotherapies may differ in the form, overtness, and centrality of these instrumental judgments, we view them as basic to the psychotherapeutic process and guided by the theoretical orientation of the therapist.

 

ATTRIBUTIONAL PROCESSES IN PSYCHOTHERAPEUTIC JUDGMENTS

With the goal in mind of ameliorating the client's presenting complaint, the therapist must make two instrumental judgments; he or she must determine the nature of the presenting complaint and its solution. Using the analysis of Brickman et al. (1982), these judgments may be seen as reflecting two strategic concerns: who or what is to blame for the cause of the presenting complaint and who or what is to have control over its solution. These concerns can best be defined in terms of two distinct attributions about the client: (1) the extent to which the client is responsible for the cause of the presenting complaint, and (2) the extent to which the client is to be held responsible for the solution to the presenting complaint.

Although several researchers and theoreticians (e.g., Ross, Rodin, & Zimbardo, 1969; Storms & McCauly, 1976; Wills, 1978) have applied attributional analysis to clinical issues, they have, in the main, avoided the consideration of the attributional dynamics that define and shape the mediating goals and procedures of therapy. We view this, at least in part, as the result of an inadequate mapping of the instrumental judgments that underlie psychotherapy as e as other helping interventions, see Karuza, Zevon, Rabinowitz, & Brickman, 1982). As Brickman et al. (1982) point out, traditional attributional analyses are predicated on a view of individuals as generally objective scientists who seek to determine the causes of events in order to achieve an accurate understanding of the world. Therapists, however, may be less interested in objectively defining causes of events than with acting on the presenting complaint.

Further, traditional attributional analyses (e.g., Jones & Davis, 1965; Kelley, 1967; Weiner et al., 1971) tend to emphasize how causality is assigned for past or current events. What is typically overlooked is how attributions are made about the cause and control of future behavior, an issue that is critical in setting the goals and direction of therapy. Because of the orientation of existing attributional theories and their language (e.g., internal or external locus of control; stable or unstable, global or specific causes) they do not clearly define the interrelationship of the strategic attributions about the client's responsibility for the cause of the presenting complaint, a past-focused attribution, and the client's responsibility for the solution of the presenting complaint, a future-focused attribution.

In the section that follows we will review the existing theoretical and empirical clinical literature that shows the impact of the attributions about the cause of and solution to the client's presenting complaint on the therapeutic enterprise.

 

ATTRIBUTIONS ABOUT THE CLIENT'S RESPONSIBILITY

A number of historical trends and ongoing debates in clinical psychology are concerned with determining the most useful attributional perspective for conceptualizing and understanding patients' presenting problems. In a historical vein, the treatment advocated by, among others, Phillipe Pinel and William H. Tuke proposed that mental patients be treated with kindness, understanding, and the belief that they (given the required environmental support) possessed the necessary capacity to effect change and experience a "cure." The decline of moral treatment and the ascendence of medical forms of treatment represented a shift in attributional perspective: The "cure" was now the responsibility of the psychiatric expert, and the patient's responsibility was reduced to a "passive, submissive, quiet, untroublesome waiting for the discovery of a cure" (Ullman & Krasner, 1975, p. 138). In a related vein, the essence of the behaviorists' challenge of psychoanalytic theory is to a large extent the question of whether deviant behavior is caused by factors in the person (an internal attributionthe psychoanalytic approach) or factors in the environment (an external attribution-the behavioral approach). We contend that therapists, as a result of their training and allegiance to a particular theory of abnormal behavior and psychotherapy, operationalize these attributions in terms of assigning responsibility for the causes of and solutions to the patient's problems. In his discussion of the difficulties in understanding abnormal behavior, Price (1972) states:

What passes for theory in the study of abnormal behavior is often mixed with large doses of ideology. Whether this is a beneficial effect is arguable. In any event, theory and ideology are often mistaken for one another, and it is sometimes unclear whether the intent of a particular viewpoint is descriptive or prescriptive (p. 5).

Past-focused Attributions

Evidence supporting our contention that therapists make attributions about the client's responsibility for the cause of problems can be found in examinations of the theoretical language of various psychotherapeutic schools. Therapeutic systems such as rational-e motive therapy (Ellis, 1962) and client-centered therapy (Rogers, 1951) adopt a view that stresses personal responsibility for one's actions, while the behavioristic therapies focus on the environment and see the causes of behavior as externally based. The psychoanalytic perspective has, likewise, implicit views on the causes of and responsibility for the client's difficulties. The major thrust of the psychoanalytic approach is a historical analysis of the client in which his or her present dysfunction is seen as caused by a past trauma. The salience of attributions for understanding psychological disturbance is more directly addressed in recent formulations of the learned hopelessness theory, formerly the learned helplessness theory, of depression, which emphasizes an attributional analysis of the causes for negative life events as key to understanding the dynamics of depression (Abramson, Metalsky, & Alloy, Chapter 2, this volume; Abramson, Seligman, & Teasdale, 1978). Increasingly, clinical researchers are recognizing the role of attributions in the development and maintenance of psychological problems, as well as the attributional aspects of psychotherapy (e.g., Harvey & Galvin, 1984; Leary & Maddux, 1987).

Further support for the prevalence of therapist-based attributions about client responsibility for the cause of problems can be inferred from investigations that illustrate the personalistic attributions made by observers for the causes of a client's problems (see Wills, 1978, for a review). An example is the study by Snyder, Shenkel, and Schmidt (1976) in which subjects were instructed to adopt the perspective of a counselor when listening to a therapy interview. Results showed that subjects under these instructions made more personalistic attributions of the cause of the client's problems as compared to subjects instructed to adopt the perspective of the client and subjects in a control group given no instructions. This tendency is echoed in Rosenhan's (1973) well-known study, which reported hospital psychiatric staff as more likely to ascribe the behavior of patients to personalistic, rather than situational, factors. As this discussion suggests, attributions regarding the cause of, and responsibility for, the client's problems are a component of all forms of psychotherapy.

Future-focused Attributions

The question of who should be responsible for solving the presenting complaint is an often overlooked but central one in psychotherapy. Only by making judgments about the client's future potentials, and the extent to which clients can be seen as responsible agents in seeking solutions to their problems, can the therapist rationally design and implement interventions.

Evidence for the salience of attributions of the client's responsibility for the solution to the presenting problem can again be found by examining the theory underlying the activity of the psychotherapist. Each theory of psychotherapy proposes intervention styles or techniques that involve a set of expectations about the client's role in, and potential for, change. Client-centered therapy, for example, sees the client as the ultimately responsible agent. The therapist's role as a change agent is minimized and consists of developing a relationship with the client based on congruence, positive regard, and empathic understanding (Rogers, 1959). Cognitively based behavioral therapies also stress the responsibility of the client for achieving a solution to the problem. The therapist's role is primarily that of a teacher who instructs the client. The client, however, is responsible for the mastery and employment of these techniques.

On the other hand, a number of therapeutic approaches exist that see the client as having little or no responsibility for the solution to the problem. Psychoanalysis is the premier example of a psychotherapy where clients are assumed to have little future responsibility for directly solving their problems. The psychoanalytic approach stresses the unconscious motivation that underlies behavior, a motivation the unanalyzed client can have little hope of controlling or even understanding directly. A further example can be found by examining the approach of therapeutic groups such as Alcoholics Anonymous. Within these therapeutic systems the client is seen as having little individual control over his or her problem. The group, its support system, and a spiritual force are held to be important regulators of behavior and the ultimate "cause" for the future adjustment of the client. Most pronounced, perhaps, is the orientation of pharmacological treatments; their practitioners assume that behavioral change and future effective functioning is largely beyond the ability of the individual. The solution to the problem and the subsequent adjustment of the client result from the administration of the external chemical intervention.

Considerable empirical evidence for the presence, in therapy, of attributions about client responsibility solutions can be found in the clinical literature, For example, Saltzman, Luetgert, Roth, Creaser, and Howard (1976) had therapists and clients rate each other on a variety of dimensions, including the perceived responsibility of the client and the therapist for solving the client's problem. The therapists' assessment of the degree of resolution of the presenting problem was found to be significantly correlated with their perceptions of the clients' acceptance of responsibility. In addition, clients' reported resolution of presenting problems was significantly correlated with their perceptions of their own responsibility for their behavior and with their therapists' perceptions of their responsibility. The fact that judgments of client responsibility were made, and that they were related to therapeutic outcome, shows the importance of considering attributions of the client's future responsibility for the solution to the problem.

As our review suggests, the therapist's attributions about responsibility for the cause of and solution to the client's problems, whether explicit or implicit, combine to affect the entire course of psychotherapy. First, these attributional judgments help define the presenting problem and its scope and therefore allow the therapist to understand the nature of the problem. Second, in assessing the cause of the problem and the potential of the client, the therapist is able to judge when, how, and where to intervene in order to ameliorate the problem. Finally, knowledge of the client's potentials and his or her responsibility for the cause of the problem provides the therapist with indications of where to look for improvement and furnishes guidelines for gauging the effectiveness of therapy.

Ideally speaking, by applying our attributional analysis to psychotherapy, a client can be viewed as having high or low responsibility for the cause of the presenting problem and at the same time be expected to have either high or low responsibility for the resolution of the problem. Thus, by crossing the two attributional dimensions, four distinct attributional combinations in psychotherapy can be derived.

 

ATTRIBUTIONS OF RESPONSIBILITY AND FOUR MODELS OF HELPING

Both theoretical (Brickman et al., 1982) and empirical (Rabinowitz, 1979) investigations have argued that each of four attributional combinations defines a particular approach to helping. The attributional combinations for presenting problems are high causal responsibility-high solution responsibility, low causal responsibility-high solution responsibility, low causal responsibility-low solution responsibility, and high causal responsibility-low solution responsibility. Each model embodies a specific helping ideology that in turn determines: (1) the characterization of clients, (2) the essential agents of change, and (3) the intervention strategy. Each of these models appears to be internally consistent and incompatible with the others. The models are described most fully in Brickman et al. (1982) and are summarized here in Table 7-1. We will first briefly describe the general characteristics of each model.

 

MORAL MODEL

Attributing responsibility for both the cause and solution of the problem to potential clients characterizes the moral model of treatment. At the beginning of the century, treatment of alcoholics and mental patients was guided by this model. These groups were generally viewed as responsible for their problems and for getting better. When help is given under moral model assumptions, it is the kind of help that motivates or reminds the person to accept responsibility.

 

TABLE 7.1:  Consequences of Attributions of Responsibility in Four Models of Helping and Coping

 

Attributions to self of

responsibility for problem

Attributions to self

of responsibility for solution

HIGH

HIGH

(Moral model)

Low

(Enlightenment model)

Perception of self

Actions expected of self

Others besides self who must act

Actions expected of others

Implicit view of human nature

Pathology

 

Lazy

Striving

Peers

Exhortation

Strong

Loneliness

 

Guilty

Submission

Authorities

Discipline

Bad

Fanaticism

 

LOW

 

(Compensatory model)

(Medical model)

Perception of self

Actions expected of self

Others besides self who must act

Actions expected of others

Implicit view of human nature

Pathology

 

Deprived

Assertion

Subordinates

Mobilization

Good

Alienation

 

Ill

Acceptance

Experts

Treatment

Weak

Dependency

 

 

for getting in and out of trouble and to take control of his or her life. Potential clients are considered to be unaware that their own actions or perceptions are at the heart of the problem, or stubborn in their refusal to relinquish these self-defeating actions or perceptions. Helpers' interventions can best be described as motivating or exhorting clients to change themselves.

The moral model is presently exemplified in the vast array of popular "how to" books, newspaper advice columns, and radio phone-in shows. Readers and listeners are generally shown how they have caused or compounded their problems and what they can do to solve or improve the problem. But the most dramatic illustration of the moral model of helping is the formerly fashionable program of Erhard Seminars Training, or est (Brewer, 1975; Frederick, 1974). The primary message to est participants is that they are "deficient" for not having realized that they are totally responsible for everything they ever have been or will be.

 

COMPENSATORY MODEL

The compensatory model of helping does not attribute responsibility to people for the cause of their problem, but expects them, after help is delivered, to take responsibility for the solution to it. The help is compensatory in the sense that it requires clients to take steps to compensate for their problems. Unlike the moral model, these problems are not seen as caused by the client. Thus, in the compensatory model, clients are seen as deprived or victimized by circumstances beyond their control, yet fully responsible for finding a solution. Helpers are expected to provide resources, training and opportunities to the client, who is expected, in turn, to use them to craft a solution to the problem. Because this kind of help is goal-directed and targeted at specific deficits, it is fixed and temporary in duration.

Examples of help in this model abound in community-action programs and government-sponsored aid to underprivileged minorities. Among psychotherapies, cognitive-behavioral treatments that incorporate such elements as self-observation, cognitive restructuring, contingency contracting, and skills training, are based on this model.

 

MEDICAL MODEL

The case in which recipients of help are not responsible for either the cause of their problem or the solution to it has come to be known as the medical model of treatment. Problems are viewed as disorders and clients as ill and unable to find solutions. The essential agents of change in this model are expert helpers, who are specially trained to solve a particular set of problems. The intervention requires skilled individuals who provide the needed treatment or service to clients who could not provide for themselves. Help in hospitals is the most familiar embodiment of the medical model.

There are now numerous biological treatments that attempt to relieve psychological disorders by altering some aspect of physiology, particularly the central nervous system. These include psychoactive drugs, electroconvulsive therapy, and the most radical of all biological treatments, psychosurgery.

 

ENLIGHTENMENT MODEL

In this model, clients are held responsible for the cause of their problem, but not for the solution of their problem. The name is derived from the essential feature of help given under this model: Clients must come to see themselves as the source of their problems, but someone or something outside of themselves as their salvation.

The agent of change in this model is a "force" or authority, often found in a select community of fellow and former sufferers. This group is uniquely suited to provide the critical resources of discipline and understanding. Having or having had the same problem as the client, the group members can sympathize with the client's distress. At the same time, their own experience with the problem gives them the knowledge and legitimacy to impose the strict rules regarded as necessary to contain and manage the problem.

Because the solution to the problem lies outside the individual, it can be maintained only so long as the relationship with the group is maintained. Thus, the treatment is relatively permanent. Past recipients of help in this model often maintain their ties to the community and thus continue to receive help and reaffirm their commitment to improvement by helping others with the same problem. With no help, individuals are seen as likely to continue their troubled behavior.

Some peer support groups for problems like alcoholism, drug addiction, and overeating are examples of enlightenment model helping. In each case, only when participants put their trust and their futures in the therapeutic community can the problem be managed.

 

UNIQUE ADVANTAGES OF THE HELPING MODEL

These moral, compensatory, medical, and enlightenment models of help represent distinct orientations. While it is ostensibly true that any of the four approaches could be applied to any pr senting problem, each of the mo dols occupies a particular domain in which it is uniquely valid and instrumental. In this section, the unique advantages associated with each del will be discussed.

There is probably no model in which greater overall potency is attributed to clients than in the moral one. Within this framework, clients are held responsible for their pasts and futures, successes and failures. By attributing this potency to clients, therapists reflect a belief in the client's capacity to improve, a belief which in itself may cause an amelioration of the presenting complaint. The extensive literature on self-fulfilling prophecy lends further credence to this premise (Rosenthal & Jacobson, 1968). Research further suggests that when clients are induced to view themselves as responsible causal agents, it is more likely that any improvement they do exhibit will be attributed to internal rather than external forces, and thereby be more enduring (Davison & Valins, 1969; Miller, Brickman, and Bolen, 1975).

In a related vein, when clients see themselves as both the source of and solution to their problems, they may avoid the tendency to adopt a "sick role" with its attendant manifestations of dependency and malingering. They may also be less likely to rely unduly on external excuses and supports. As Brickman et al. (1982) have contended, holding individuals responsible for solutions (and, often times unfairly, for the causes of problems as well) may be the most effective way to motivate them to make positive changes that they may not otherwise be inclined to make. If, as in the case of a rape victim, a certain measure of personal responsibility is assigned, the victim may be more disposed to install better locks on doors, travel with a companion at night, or move out of a dangerous neighborhood than those victims who view themselves as powerless to prevent future rapes. As Janoff-Bulman (1979) indicated, a majority of women who present themselves for rape counseling do in fact assign some responsibility to themselves for the rape, often for imprudent or unwise behavior.

According to existential philosophers, accepting full responsibility for one's acts affords an individual not only a personal dignity, but also the opportunity to discover the deepest meaning of existence. Viktor Frankl, an existentially oriented psychiatrist and founder of logotherapy, has stated (1959):

Ultimately, man should not ask what the meaning of his life is, but rather must recognize that it is he who is asked. In a word, each man is questioned by life, and he can only answer to life by answering for his own life; to life he can only respond by being responsible. Thus, logotherapy sees in responsibleness the very essence of human existence. (p. 172)

The moral model orientation offers many advantages to therapists as well as clients. If help is given, but is insufficient or falls, the failure cannot be attributed to the therapy. Further, when help does occur in this model, it tends to be offered at the discretion of the therapist, to be limited in scope and duration, and to be symbolic rather than material. The moral model is expected to achieve greatest results when applied to those clients who already possess the resources, talent, support, and opportunities to better themselves, but who require a new perspective on their problems or a motivational boost.

 

COMPENSATORY MODEL

No model takes a more auspicious view of the client's blameworthiness or prospects than the compensatory model. While clients are not seen as the sources of their own problems, they are perceived as being able to effect necessary changes in their lives. Implicit in this model is a particularly optimistic view of the client's potential: If just given the skills, resources, or opportunities, the client will be successful and self-reliant (Karuza, Zevon, Rabinowitz, & Brickman, 1982).

The attributional pattern of the compensatory model (along with the enlightenment model) asks something of both clients and therapists, and involves the two in a mutually responsive relationship. Therapists are requested to contribute some combination of time, resources, and opportunities to deserving clients in a manner that unites the therapist's feelings of competence with the fate of the clients. While the therapist is active in this form of treatment, the duration of the therapy is expected to be fixed and temporary, because it is the client's role to implement the solution.

Because this model appears to have so much to recommend it, we would expect it to work well with a large variety of clients who can play a significant role in their treatment and whose need is primarily to acquire skills.

 

MEDICAL MODEL

As with the compensatory model, the medical model orientation does not hold clients responsible for the causes of their problem and thus relieves them of the guilt and anxiety that come from wondering what they have done to bring on their difficulties. Medical ideology specifies that help is to be given to all, regardless of the circumstances surrounding the illness or injury, the personal or social characteristics of the needy, or even how effective the help is likely to be. But because in the medical model clients are also not held responsible for the solutions to their problems, there is probably no model in which clients are seen as more passive and helpless, and more needy of indefinite amounts of help.

The passivity on the part of clients, which is a natural consequence of this model's assumptions, can make it easier to treat certain disorders. If the solution to a problem requires treatment aimed at specific organs or processes rather than a, he "whole person," medical model assumptions may be ideal. Passivity on the part of the patients also makes it easier to treat large numbers of patients in a relatively short period of time, and to treat patients in institutional settings. Thus, the medical model may constitute a highly cost-effective type of treatment. The features that make patients easier to treat are, of course, advantageous to helpers in this model.

 

ENLIGHTENMENT MODEL

How can holding people responsible for problems which they are deemed incapable of solving be advantageous? For one thing, attributing responsibility to clients implies a view of them as effective and willful, if misguided, actors, a view which, we have argued, has its merits. On the other hand, denying clients responsibility for improvement encourages a perception of them as really needing, if not deserving, help. Its historic popularity and current appeal may derive from the fact that, unlike the compensatory model, the assumptions of the enlightenment model call for interventions that keep clients under tight social control.

For another, admitting one's guilt may initially be quite painful, but may ultimately bring a tremendous relief as one relaxes one's pretenses. In addition, the realization that one cannot overcome one's problems by oneself may further reduce discomfort. Encouraging clients to relinquish notions of being able to solve their problems while at the same time holding them responsible for the causes of the problem discourages clients from adopting a "sick role."

Perhaps the greatest advantage of this model lies in the special relationship of the client to the therapeutic community. In no other model is the investment of helpers in the success of the client so large. Indeed, when recipients in this model succeed, it is as if the helper's own solution (and way of life) is vindicated.

When solutions require discipline on the part of recipients, the difficult question of who has the right to impose this discipline must be addressed. When problems require empathy or understanding, the question of who can best empathize with recipients is a delicate one. Clearly, individuals who have overcome the problem seem uniquely suited to provide both discipline and understanding. In this connection, two of the most robust findings in the helping literature are that people who have close personal relationships with others cope better with their problems than those who do not (e.g., Lowenthal & Haven, 1968; Moos & Mitchell, 1982) and that similar others provide troubled people with special validation and support (cf. Gottlieb, 1983). Individuals who would be expected to profit most from the enlightenment model assumptions are those whose solutions require sustained discipline and continued support from concerned others.

 

THE MODELS AND MODES OF PSYCHOTHERAPY: AN ATTRIBUTIONAL ANALYSIS

The models discussed are seen as relatively pure and general types, strongly determined by specific attributional patterns. As such, we would not expect them to be neatly embodied in any existing form of psychotherapy. Nonetheless, as previously noted, in all modes of psychotherapy implicit and explicit assumptions about the client's responsibility for the causes and solution to the presenting problem are made. Further, these attributions seem to underlie and justify the intervention strategy.

Although there are hundreds of different types of psychotherapies, we can isolate a few dominant modes that illustrate each of the four models. First, however' a few caveats are in order. The mapping of psychotherapies onto models is bound to be somewhat imprecise because, in many cases, attributions of responsibility must be inferred from the therapy's theoretical framework. Further, many of the dominant modes of therapy have significant variations. At the very least, each type of psychotherapy may be interpreted differently by its advocates, and each therapist's unique style and approach to therapy may further contribute to variations within modes. Despite these reservations, we feel that an understanding of psychotherapies may be enhanced by considering the underlying attributional assumptions they make. For heuristic purposes, then, we will consider four ideologically distinct therapeutic approaches: rational-emotive therapy, behavior modification, psychoanalysis, and therapeutic communities.

 

RATIONAL-EMOTIVE THERAPY AS A MORAL MODEL TREATMENT

Rational-emotive therapy (RET), like several existential therapies, appears to be a clear embodiment of the moral model. Ellis (1962, 1973) rejected the medically toned term "psychotherapy" to describe his approach in favor of descriptors such as "emotional education."

This form of therapy sees clients as stubbornly fixed in their self-defeating perceptions. Their cognitive errors and irrational thoughts are seen as needing "vigorous and persistent attacks" by the therapist (Ellis, 1973). RET is usually conducted in face-to-face settings in as "efficient and rapid-fire a manner as the client can tolerate" (Ellis, 1973).

As we would expect in the case of a moral model treatment, the notion of responsibility is central to Ellis's perception of the causes and solutions to psychological problems. Ellis lays the responsibility for the client's problems squarely on the client's shoulders. He states that "It follows that feelings of worthlessness do not stem from the attitudes that an individual's parents take toward him, but from his one tendency to take these attitudes too seriously, to internalize them, then perpetuate them through the years" (Ellis, 1973, p. 34).

Ellis is just as clear on the locus of the solution to people's problems as he is on their cause. He writes:

To argue a better solution to the problem of his own worth, the individual had better see his own propensities to exaggerate the significance of others' attitudes toward him, and see clearly that he can vigorously question, challenge, change, and minimize these tendencies toward distorted thinking about himself and others.

Because, in moral model fashion, clients are seen as completely responsible for their problems, treatment is didactic, consisting mainly of persuasion attempts and candid feedback on the client's attempts at change.

BEHAVIOR MODIFICATION AS A COMPENSATORY MODEL TREATMENT

We regard the recently developed self-control and social and cognitive learning therapies as examples of compensatory model treatments. The following is Just a sampling of the therapies that have been called self-control or cognitive learning therapies (cf. Mahoney & Arnkoff, 1978): self-monitoring, the strategy of recording one's habits; self-reinforcement, the self-presentation of rewards contingent on performance of some desired response; thought stopping, the procedure designed to terminate unwanted cognitions; and coping skills therapy, a combination of procedures such as relaxation training, meditation, and preperformance rehearsal. The therapies that come under this umbrella are varied in their underlying assumptions and their intervention strategies. What they share, however, is the belief that maladaptive behavior and cognitive processes, no matter how developed, can be treated by teaching the clients skills and techniques for rearranging the environmental contingencies that affect their behavior.

Unlike radical behaviorists, who adopt the view that the individual is but a pawn of external influences (Skinner, 1972), utterly incapable of influencing his or her own actions, let alone being responsible for them, cognitively oriented behavior therapists generally advocate a position of "reciprocal determinism." This is the notion that people are, in fact, partially free because environmental contingencies are partly of their own making (Bandura, 1977). But whether a behaviorist supports the position of environmental determinism or of a reciprocal relationship between environment and behavior, virtually all behavior therapy techniques look to faulty environmental contingencies as the source of the problem.

            Consistent with this attribution, solutions are believed to follow a simple rearrangement of the contingencies. The role of the behavior modifier is critical, but limited to that of a "diagnostician-educator" (Mahoney & Arnkoff, 1978). He or she assesses the maladaptive cognitive processes and teaches the client how to alter the cognitions, behaviors, and patterns of affect that are troublesome. Once taught how to recognize and control maladaptive thoughts and behaviors, however, the client is often expected not only to monitor his or her own behavior and to compare monitored performance to goals or standards, but also to self-reward or self-criticize (Kanter, 1970).

 

PSYCHOANALYSIS AS A MEDICAL MODEL TREATMENT

  Psychoanalysis is the process by which unconscious conflicts are brought into consciousness, that is, abreaction. There they can be examined, "understood," and accepted, and the effects of events and experiences of the past can be interpreted in the light of these conflicts. The therapist's interpretations of the patient's dreams, free associations, malapropisms, and humor are traced to the influence of unconscious repressed instinctual drives and defenses. The transference of the client's beliefs and feelings from significant others in his or her life to the therapist is believed to be a necessary component of this process.

Traditional psychoanalysis holds that individuals are at the mercy of forces which they cannot identify or understand, let alone control. Human nature, the fallibility of parents, and the rigid demands of civilized society all combine to create neurotic patterns in the individual. More specifically, neuroses are seen as forced upon individuals by the peculiar circumstances of their lives; too much or too little gratification, frustration, condemnation, and so forth. It is one purpose of psychoanalysis to relieve patients of their guilt and responsibility for their past. Discussing psychoanalysis, Dollard and Miller (1950) point out:

From the patient's standpoint, the novelty of the therapeutic situation lies in its permissiveness. The therapist is understanding and friendly. He is willing, so far as he can, to look at matters from the patient's side and make the best case for the patient's view of things. (p. 243)

The therapist takes the view that what is past had to happen. The patient understands this acceptance as forgiveness, which in a sense, it is.... If the recital is followed by condemnation and punishment, we would not expect the effect of the confession to be therapeutic. (pp. 245-246)

It has been noted that simply being accepted for psychoanalytic treatment is so reassuring to clients that it is immediately followed by a decrease in symptomatology.

The solutions to the client's problems, like the causes, are seen as lying entirely outside the patient. Learning, or any real change in personality, comes about only through psychoanalysis. Since the dynamics of behavior lie hidden in the unconscious, the unanalyzed client cannot possibly know the real truth about himself or herself, despite his or her best efforts and those of significant others. In fact, the analyst generally takes the position that the client cannot and should not discuss his or her problems with anyone else, as that would weaken the transference. Talking freely with the analyst is regarded as therapeutically sufficient. Because they do not adequately understand their motives, new clients are warned that they should not, at least in the short run, make any major decisions or changes in their lives without consulting their analyst.

Psychoanalysis may be distinguished from other psychotherapies primarily by the therapist's interpretations of the client's statements, and on the presumed necessity for the therapist to uncover the client's drives and motives. Of course, the asymmetrical relationship between analyst and patient facilitates this general relinquishing of responsibility by the client. In typical medical model fashion, the therapist is viewed as an expert in an esoteric area poorly understood by the uninitiated. The behaviors of the client are viewed as "symptoms" of an underlying disorder.

 

THERAPEUTIC COMMUNITIES AS ENLIGHTENMENT MODEL TREATMENTS

As we noted earlier, we see many group therapeutic treatments as examples of help based on enlightenment model grounds. Obviously, this form of treatment appears to be much more narrow and focused than the other examples reviewed in this section. However, the model on which these groups are based can be and appears to have been applied to a variety of specific problems, including drug addiction (e.g., Daytop Village besity (e.g., Overeaters Anonymous), and compulsive gambling (e.g., Gamblers Anonymous), and to general problems in living. Two central features of all of these treatments can be identified: (1) the existence of a special community of former or current sufferers as the agents of change, specifically of emotional support and discipline, and (2) the client's acceptance of and strict adherence to enlightenment model assumptions about the causes of and solutions to the problem.

People who join group therapeutic treatments are either predisposed or socialized to accept full responsibility for their problem. Any confession by a newcomer that attempts to share the blame with family or friends is roundly denounced. At the same time that clients are required to acknowledge their responsibility for the problem, they are also required to admit that the problem is beyond their ability to solve or control. They are compelled to acknowledge that forces beyond their control determine their futures-the power embodied in the community of repentent and reformed peers. It is only with the help of the community that group members' desire to drink (or overeat or gamble, and so forth) can be overcome. The closely knit community, with its rituals of confessions and testimonials, serves to reinforce the notions of accepting responsibility for the solution.

 

IMPLICATIONS OF AN ATTRIBUTIONAL ANALYSIS OF PSYCHOTHERAPY

We have detailed an attributional analysis of the decision-making component in psychotherapy. In doing so, we have proposed that various psychotherapeutic schools and their intervention strategies are describable in terms of their underlying attributional assumptions. In this section we will consider some of the implications our attributional analysis has for the psychotherapy.

            Specific psychotherapies may be differentially efficient and effective for certain subsets of clients and disorders. We would further maintain that as an alternative to arguing for the supremacy of one psychotherapy over another, closer attention should be paid to considering the attributional assumptions of the therapeutic school and how they are translated into the activity of the therapist.

While at this juncture it is tempting to offer a series of prescriptions of which therapy would be optimally effective for particular disorders, we feel we must resist at present. Several factors lead us to this position. For one, several attempts have been made at offering prescriptive frameworks (e.g. Goldstein & Stein, 1976). In general, a mixed picture has emerged. This lack of clarity is due, in a large measure, to a realization of the complexity of the therapeutic effort. As was pointed out previously, a host of factors, ranging from client and therapist characteristics to the mediating goals of a particular therapeutic system, interact to produce the structure and pace of therapy. In this context, a therapy's attributional assumptions may be only one factor, albeit a potentially important one, that governs the efficacy of the therapeutic intervention.

A second difficulty in offering prescriptive statements is the possibility of slippage between the therapist's ideological orientation and his or her actual practice of therapy. The therapist's expectancies and approaches may not directly mirror the formal school he or she represents. In a real sense, the therapist's experiences and worldview may dilute and distort the "pure" orientation of the formal therapeutic system. Indeed, research by Fiedler (1950) and Strupp (1955; Strupp & Wallach, 1973; Strupp & Williams, 1973) has shown that the distinction among therapists of different schools blurs when one investigates therapists who have been practicing for an extended period. These therapist factors, such as their worldview or expectations about clients and the presenting complaints, may affect the attributional process and color the instrumental judgments guiding the therapy. Any attempts to offer a convincing prescriptive framework should consider these therapist factors in conjunction with the formal tenets of the psychotherapeutic system. While such a full-blown analysis is beyond the scope of this chapter, it is our hope that research that considers both the formal and informal attributional processes in therapy (especially on the consequences of having an accurate mapping of the client's responsibility for the cause of the presenting complaint and for its solution) can offer the basis for prescriptive statements about therapeutic effectiveness.

PROBLEMS IN ADOPTING AN ADEQUATE ATTRIBUTIONAL PERSPECTIVE

One important issue in this research is an adequate definition of a "proper attribution" of the client's responsibility. Should the client's beliefs about causality, the results of diagnostic instruments, or the therapist's understanding of the situation serve as the criterion? Further, the existence of causal chaining (Brickman, Ryan, & Wortman, 1975) may compound the questions of finding an adequate attributional framework. As these authors point out, there is no logical reason why attributions about the causes of events should stop at any one point. For example, an internal cause for a disorder (a disposition or trait) may itself be attributed to an external cause, (for example, the child-rearing practices of an individual's family). In a certain sense, it may be a matter of personal preference as to where to look for the causes of particular behaviors or how far one is willing to go before the "true" cause is found. The assumptions of the therapeutic school and the helping model it embodies may artificially direct the search for the causes and solutions of the problem. In practice, though, the stopping point in the causal chain may be operationally determined as the intervention strategy based on the set of attributions that first lead to a desirable change in the client. We see these questions, however, as empirical in their scope. Research exploring the relationship of client, therapist, and "objectively" based attributions to the effectiveness of the therapy is needed. On a theoretical level, however, we can specify certain factors that may cloud the true attributional picture.

First, the therapist's allegiance to a particular psychotherapeutic model may bias attributions about the cause of the client's presenting complaint and the solution for that complaint. Acting upon these attributional assumptions, the therapist may incorrectly define the etiology of the client's problem. The resulting interventions may be inappropriately related to the problem and, in the worst possible case, work against the client.

A related and complicating factor is the inherent complexity of causal analysis. Nisbett and Ross (1980) address this issue in the context of human inference; their arguments, however, directly apply to the activity of the psychotherapist. Particularly relevant is the principle of misguided parsimony, otherwise known as the "hydraulic" model of causation. Drawing upon Mill (1843/1974) and Kanouse (1972), the authors state:

The pronounced availability effects on causal attribution would appear to depend on the individual's willingness to be content when a single sufficient cause has been adduced and to forego exhaustive searches for further, potentially influential antecedents. That is, by manipulating the causal factors that the person will notice or ponder first, one can manipulate the person's ultimately preferred explanation for the event in question. (p. 128)

Our contention is that, in psychotherapy, the causal factors that are perceived first will be those most congruent with the attributional perspective of the therapist.

A further source of attributional bias in therapy may be the tendency of therapists to view their clients in a negative light (Wills, 1978). A number of studies have shown that therapists and professional help givers view clients as less adjusted (Wills, 1978), less capable of improvement (Batson, O'Quin, & Pych, 1982), and more in need of help as contrasted with individuals not seeking treatment and with the perceptions of the clients themselves. This negative bias may shape and direct the therapist's attributional judgments and lead him or her to see the client as less responsible both for the cause of the presenting complaint and for its solution. In terms of the analysis presented above, this lack of responsibility for cause and solution is characteristic of the medical model orientation. In other words, the negativity bias may cause the therapeutic endeavor to regress toward this medical model orientation, which views the client as sick and dependent on the therapist for amelioration of the problem. This negativity bias is a therapist-specific factor that operates independently of any formal theoretical orientation.

Wills suggests several causes for this bias. Of particular importance is the fact that the clinician's training and experiences may lead to a selective focus on the weakness and deficiencies of clients. Large caseloads and their resulting time pressures may force the therapist to concentrate on the problematic aspects of the client's behavior "because these are most immediately relevant for ameliorating the client's presenting problem" (Wills, 1978, p. 987). Certain personal or social characteristics of clients, such as advanced age, appear to elicit among professional helpers and observers alike unfavorable assumptions about the clients' prospects for solving their problems or controlling their futures (Karuza, Zevon, & Rabinowitz, 1986; Zevon, Karuza, & Brickman, 1982). In addition, the therapist's training may predispose him or her to identify negative or unpleasant facts in the client's background as playing an important etiological role (Meehl, 1973).

This negativity bias may be particularly deleterious when one considers that it may work against psychotherapies that reflect the moral, compensatory, or enlightenment models. The therapist may be caught between the attributional assumptions reflected in the therapeutic orientation and his or her personal attributional bias in regard to the client's responsibility. The negative bias may become prepotent for the therapist and result in the therapy drifting toward medical model attributional assumptions and associated intervention strategies.

Also contributing toward this drift is a corresponding therapist "burnout" effect. The initial image of the client as a troubled and overwhelmed individual stands in contrast to the image of the therapist as a responsible, competent expert. If the therapist were not perceived in this way, the client would presumably not seek his or her help, and the therapist would not inspire the confidence needed for a successful therapeutic interaction (Frank 1973). At least relative to the clients, then, therapists may initially view themselves as being in control of their lives, responsible for their problems and solutions. With competence attributed to them and deficits attributed to clients, they may also view themselves as responsible for clients and the course of therapy.

However, all of the factors that make it difficult for clients in psychotherapy to improve, including the simple intractability of some problems, make it easy for therapists to change their perceptions of themselves as well as their clients. When help is wholly or partly unsuccessful, there is some evidence to suggest that therapists tend to minimize their own responsibility for clients and heighten their sense of client's responsibility for themselves (Maslach, 1978; Wills, 1978). The therapist who has devoted much time and effort to an unsuccessful case may come to believe that the client is at fault for the failure to help because he or she is stubborn, uncooperative, irresponsible, or incorrigible. At the same time, the therapist may come to feel frustrated and helpless for being ineffective (Maslach, 1978). In the end, especially with unsuccessful cases, the attributional orientation of the therapist is likely to be the reverse of what it was at the start of therapy. Therapists may see clients as responsible for the causes of their problems and their failure to improve. On the other hand, they may see themselves as not responsible for their client's problems or for solving those problems.

 

 

PROCESS REDEFINED

 

In the beginning, we viewed psychotherapy as a problem-solving process in which the therapist's attributional judgments played an important role in the dynamic process of defining the client's presenting complaint, setting the goals of therapy, and crafting a solution for the client's problem. In our considerations of the attributional bases of various psychotherapeutic systems, much of this process view was absent. For the sake of expediency, we intentionally adopted a static compartmentalized view of these components of psychotherapy. We did so for two reasons: first, to isolate and highlight the particular attributional orientation of each therapy, and, second, to illustrate how the attributional orientation defines and is mirrored in the actual intervention strategy of the therapist. In vivo, psychotherapy is an ongoing process that ideally leads the client through a series of cognitive/ emotional changes. As therapy progresses, changes occur in the client-therapist relationship, the client's self-disclosure (Jourard, 1964), self-perceptions, self-examination, and behavior. For each psychotherapeutic school, the attributional assumptions not only define the nature of the presenting complaint, but offer the therapist a criterion, or definition of a fully functioning, adjusted client. The thrust of the therapeutic endeavor is to lead the client to this goal. In this traditional sense, the process of therapy is theoretically canalized.

At this point, we would like to reconsider the process notion in therapy from an admittedly speculative perspective. As the therapy progresses and interventions are implemented, changes may occur in the client's behavior and his or her verbal reports, changes that may be indicative of a shift in the client's view of his or her responsibility for past, present, and future actions. The ability and tendency of the therapist to recognize these changes and reevaluate the client's responsibility for his or her actions may be limited by the directional momentum of the therapy. As indicated above, the therapist may adhere to a specific set of attributions about the client that are dictated by the therapist's theoretical orientation. Thus, the therapist may fail to recognize change in the client, incorrectly attribute causes for a particular behavior of the client, distort the potentials of the client, or continue with interventions that are ineffective or irrelevant. When clients do not fit the attributional profile dictated by the therapeutic system, therapists may label the client as uncooperative or incompetent, or the therapists may start to feel that they are failing. This may set the stage for the negativity bias in viewing clients or, in the long run, to therapist "burn-out."

Conversely, if the client is initially "cooperative" and responds to the therapy early on with an alleviation of symptoms, the therapist may be seduced into accepting the attributional assumption of the therapy as an absolute. Thus, the therapist may set up expectancies for the client that mirror the attributional ideology of the therapeutic system, expectancies that may be unrealistically accepted and embodied by the client. Further, the therapist, upon seeing the preliminary success of the treatment, may be tempted to continue treating the client with the same intervention strategy that brought about the initial change, even though it is based on erroneous assumptions and is unlikely to be helpful in the long run.

Erroneous attributions about the client's ability to take responsibility for solving a problem may underlie some current approaches to the treatment of alcoholism. A controversial issue in the treatment of alcohol abuse is whether alcoholics can learn to control their drinking so that they can drink in moderation. The prospect of being able to "drink socially" is appealing to many alcoholics and attracts them to treatment (Pomerleau, Pertschuk, Adkins, Brady, 1978). Several multimodal cognitive-behavior therapies have been developed to promote controlled drinking among alcoholics. These therapies embody the characteristically optimistic compensatory model assumptions that once clients acquire the knowledge, skills, and resources to control their drinking, they can become responsible "social drinkers." To this end, clients receive education about the effects of alcohol and how to discriminate their blood alcohol level, group therapy, self-management skills, job-seeking and interpersonal skills training, social drinking practice sessions and relaxation training. Despite reports of short-term success with such programs, there is no evidence to suggest that they have any long-term success (Armor, Polich, & Stambu, 1976; Foy, Nunn, & Rychtarik, 1984) or that their orientation to alcoholism is valid over time.

In redefining the "process" in psychotherapy, we are contending that therapy naturally affects the client's and therapist's view of the client's responsibility. We propose that this fluidity of attributional perspective may be adaptive, and offer it as an alternative to the more intransient attributional stance associated with existing theories of psychotherapy. Specifically, we see that each of the four attributional views of the client may be valid at various steps in the course of therapy. As clients change, the old set of attributions guiding the therapist's activities may no longer be relevant, while new assumptions of the cause of the client's present and future behavior may become valid. Instead of locking the client into a fixed set of attributions, we are arguing that therapists should be flexible, reevaluating clients periodically and implementing interventions that are in keeping with the changes in the clients' perceived responsibility for their behavior and in their newly developing potentials. In other words, we are proposing a guided eclecticism in which the attributional orientations and interventions of each therapeutic model are employed. As therapy progresses, instead of approaching the client from a particular therapeutic vantage point, a shifting perspective may better serve the goals of psychotherapy.

To illustrate this point, let us consider a rational-emotive approach, which reflects the moral model of helping. A client, when he or she first approaches the therapist, may indeed feel helpless and dependent, more typical of someone who fits a medical model orientation. The first task of the therapist, rather than preaching self-reliance, might be to help organize the client's life, offering suggestions and directions on how to take care of some of the pressing problems of the client (e.g., "Whom should I see for a divorce?" or "How can I avoid confronting my mother?"). As the client gets on firmer footing, the therapist may then shift to a compensatory approach, training the client in skills needed to cope with the demands of life (e.g., assertiveness training) and reinforcing the client's sense of responsibility for his or her future. After these skills have been acquired, the therapist may be better able to implement the rational-emotive approach, that is, to encourage and motivate the client to take control over his or her life.

An example of this can be found by considering the "paradox of depression." According to Abramson and Sackeirn (1977), depressed people blame themselves for their own unhappiness and that which they inflict on others, yet feel helpless to do anything about the situation. From the vantage point of the models, the paradox disappears. Feelings of responsibility for problems and solutions, for pasts and futures, need not be correlated. It seems entirely possible that people might feel responsible for the origin of the problem, but not for the solution.

When people become stuck in this attributional set, relief may come simply by moving them out of this set-to any of the others. If it is recognized that a central problem with depression is precisely this pattern of attributions, then helping might take the form of changing people's attitudes about the origin or the solution to the Problem. The recent popularity of cognitively oriented therapies such as Beck's (1976) for problems like depression may indicate the virtues  of demanding more active and responsible behavior on the part of the client.

Increasingly, it is becoming clear that more than half of all visits to physicians are for conditions with no known physical basis (Cummings & Follette, 1976). Many of the complaints brought to medical doctors are stress related illnesses or physical symptoms caused by psychological problems. Doctors and patients who pursue biological explanations for what are essentially emotional problems and look to a pill, a shot, or an operation to cure the malady may be incorrectly applying the medical model to the presenting problem. Mounting evidence suggests that simply inducing these patients to discuss their problems and feelings and enter short- or long-term psychotherapy can have lasting positive effects on their physical health (Cummings & Follette, 1976; Jones & Vischi, 1979). One of the most commonly offered explanations for this consistent finding is that the psychotherapeutic and behavioral treatments make patients more active and responsible participants in their own care. Again, it appears that guiding clients to adopt a new orientation to long-standing problems can be quite beneficial.

By following a more fluid or cyclic approach, the advantages of each therapy can be used to build a more comprehensive and effective therapeutic effort. At the same time, the problems associated with a unidirectional attributional approach may be avoided. What we are proposing is theoretical in nature, and perhaps anathema to theoretical purists. However, we hope that a consideration of the attributional foundations of psychotherapy and the dynamic changes in attributions inherent in therapy will stimulate research in social and clinical psychology that, in the long run, will inform and improve the therapeutic process.