The Theory and Practice of Group Psychotherapy (53 page)

Detailed inquiry should be made into the client’s interpersonal and group relationships, relationships with early chums, closest prolonged friendships, and degree of intimacy with members of both sexes. Many of Harry Stack Sullivan’s interview techniques are of great value in this task.
⁵⁰
It is informative, for example, when inquiring about friendships to ask for the names of best friends and what has become of them. It is valuable to obtain a detailed history of formal and informal groups, childhood and adult cliques, fraternities, club memberships, gangs, teams, elected offices, and informal roles and status positions. I find it valuable to ask the client to give a detailed description of a typical twenty-four hours and to take particular note of the way the client’s life is peopled.

The predictive power of this type of interview has yet to be determined empirically, but it seems to me far more relevant to subsequent group behavior than does the traditional intake clinical interview. This interview approach has become a standard assessment component in interpersonal therapy (IPT) and cognitive behavioral analysis system psychotherapy (CBASP).
⁵¹

Fifty years ago, Powdermaker and Frank described an interpersonal relations interview that correctly predicted several patterns of subsequent group therapy behavior, such as “will dominate the group by a flood of speech and advice”; “will have considerable difficulty in showing feelings but will have compulsion to please the therapist and other members”; “will be bland and socially skillful, tending to seek the leader’s attention while ignoring the other members”; “will have a wait-and-see attitude”; or “will have a sarcastic, superior ‘show-me’ attitude and be reluctant to discuss his problems.”
⁵²
Contemporary psychotherapists have made an important addition to this approach: they emphasize the client’s beliefs and expectations about relationships, which give form to the client’s interpersonal behavior. This behavior in turn pulls characteristic responses from others.
⁵³

Such a sequence is illustrated in the following vignette, which also illustrates the necessity of the therapist attending to his own emotional and behavioral reactions and responses to the client.

• Connie, a woman in her forties, was referred by her family physician for group therapy because of her social anxiety, dysthymia, and interpersonal isolation. Immediately on entering the office she told me she had a “bone to pick” with me. “How could you leave a message on my answering machine calling me Connie and yourself Doctor So-and-so? Don’t you understand the power imbalance that perpetuates? Haven’t you heard of feminism and empowerment? Do you treat all the women you know like this, or only your clients?”

I was at first stunned, and then felt threatened and angry. After a few moments’ reflection I considered that she indeed had a point, and I acknowledged my carelessness.

Later in the session I asked whether we might explore the extent of her anger, and we soon began discussing her expectation that she would be silenced and devalued in this process, as she had been so many times in the past. I told her that she had, in a sense, presented a powerful test to me—hoping, perhaps, that I would not take the bait, that I would not confirm her expectations about how her world always treats her, a pattern that often resulted in her feeling rebuked, attacked, and shut down. I suggested that she no doubt came to these beliefs honestly and that they reflected her experiences in life. She may well initially relate to the group members in the same way that she did with me, but she did have a choice. She could make the group experience yet another in a series of angry rejections, or she could begin a process of learning and understanding that could interrupt this self-fulfilling prophecy.

Summary

Group behavior can be predicted from a pretherapy encounter. Of all the prediction methods, the traditional intake individual interview oriented toward establishing a diagnosis appears the least accurate, and yet it is the most commonly used. An individual’s group behavior will vary depending on internal psychological needs, the manner of expressing them, the interpersonal composition and the norms of the group. A general principle, however, is that
the more similar the intake procedure is to the actual group situation, the more accurate will be the prediction of a client’s behavior
. The most promising single clinical method may be observation of a client’s behavior in an intake, role-play, or waiting-list group. If circumstances and logistics do not permit this method, I recommend that group therapists modify their intake interview to focus primarily on a client’s interpersonal functioning.

PRINCIPLES OF GROUP COMPOSITION

To return now to the central question: Given ideal circumstances—a large number of client applicants, plenty of time, and a wealth of information by which we can predict behavior—how then to compose the therapy group?

Perhaps the reason for the scarcity of interest in the prediction of group behavior is that the information available about the next step—group composition—is even more rudimentary. Why bother refining tools to predict group behavior if we do not know how to use this information? Although all experienced clinicians sense that the composition of a group profoundly influences its character, the actual mechanism of influence has eluded clarification.
⁵⁴
I have had the opportunity to study closely the conception, birth, and development of more than 250 therapy groups—my own and my students’—and have been struck repeatedly by the fact that some groups seem to jell immediately, some more slowly, and other groups founder painfully and either fail entirely or spin off members and emerge as working groups only after several cycles of attrition and addition of members. It has been my impression that whether a group jells is only partly related to the competence or efforts of the therapist or to the number of “good” members in the group. To a degree, the critical variable is some as yet unclear blending of the members.

A clinical experience many years ago vividly brought this principle home to me. I was scheduled to lead a six-month experiential group of clinical psychology interns, all at the same level of training and approximately the same age. At the first meeting, over twenty participants appeared—too many for one group—and I decided to break them into two groups, and asked the participants simply to move in random fashion around the room for five minutes and at the end of that time position themselves at one or the other end of the room. Thereafter, each group met for an hour and a half, one group immediately following the other.

Although superficially it might appear that the groups had similar compositions, the subtle blending of personalities resulted in each having a radically different character. The difference was apparent in the first meeting and persisted throughout the life of the groups. One group assumed an extraordinarily dependent posture. In the first meeting, I arrived on crutches with my leg in a cast because I had injured my knee playing football a couple of days earlier. Yet the group made no inquiry about my condition. Nor did they themselves arrange the chairs in a circle. (Remember that all were professional therapists, and most had led therapy groups!) They asked my permission for such acts as opening the window and closing the door. Most of the group life was spent analyzing their fear of me, the distance between me and the members, my aloofness and coldness.

In the other group, I wasn’t halfway through the door before several members asked, “Hey, what happened to your leg?” The group moved immediately into hard work, and each of the members used his or her professional skills in a constructive manner. In this group I often felt unnecessary to the work and occasionally inquired about the members’ disregard of me.

This “tale of two groups” underscores the fact that the composition of the groups dramatically influenced the character of their subsequent work. If the groups had been ongoing rather than time limited, the different environments they created might eventually have made little difference in the beneficial effect each group had on its members. In the short run, however, the members of the first group felt more tense, more deskilled, and more restricted. Had it been a therapy group, some members might have felt so dissatisfied that they would have dropped out of the group. The group was dominated by what Nitsun describes as “antigroup” forces (elements present in each group that serve to undermine the group’s work).
⁵⁵
Because of their narrower range of experience in the group, they learned less about themselves than the members of the other group did.

A similar example may be drawn from two groups in the Lieberman, Yalom, and Miles group study.
⁵⁶
These two short-term groups were randomly composed but had an identical leader—a tape recording that provided instructions about how to proceed at each meeting (the Encountertape Program). Within a few meetings, two very different cultures emerged. One group was dependably obedient to the taped instructions and faithfully followed all the prescribed exercises. The other group developed a disrespectful tone to the tape, soon referring to it as “George.” It was common for these members to mock the tape. For example, when the tape gave an instruction to the group, one member commented derisively, “That’s a great idea, George!” Not only was the culture different for these groups, but so was the outcome. At the end of the thirty-hour group experience—ten meetings—the irreverent group had an appreciably better outcome.

Thus, we can be certain that composition affects the character and process of the group. Still, we are a long way from concluding that a given method X composes a group more effectively than method Y does. Group therapy outcome studies are complex, and rigorous research has not yet defined the relationship between group composition and the ultimate criterion: therapy outcome. Despite some promising work using the personality variables reviewed earlier in this chapter, we still must rely largely on nonsystematic clinical observations and studies stemming from nontherapy settings.

Clinical Observations

The impressions of individual clinicians on the effects of group composition must be evaluated with caution. The lack of a common language for describing behavior, the problems of outcome evaluation, the theoretical biases of the therapist, and the limited number of groups that any one clinician may treat all limit the validity of clinical impressions in this area.

There appears to be a general clinical sentiment that heterogeneous groups have advantages over homogeneous groups for
long-term intensive interactional group therapy
.†
⁵⁷
Homogeneous groups, on the other hand, have many advantages if the therapist wishes to offer support for a shared problem or help clients develop skills to obtain symptomatic relief over a brief period.
⁵⁸
Even with these groups, however, composition is not irrelevant. A homogeneous group for men with HIV or women with breast cancer will be strongly affected by the stage of illness of the members. An individual with advanced disease may represent the other members’ greatest fears and lead to members’ disengagement or withdrawal.
⁵⁹

Even in highly specialized, homogeneous, manual-guided group therapies, such as groups for individuals dealing with a genetic predisposition to developing breast or colorectal cancer, the therapist can expect composition to play a substantial role.
⁶⁰
Like the group of psychology interns described earlier, some therapy groups quickly come together, whereas others stumble along slowly, even with the same leader.

In general, though,
homogeneous groups jell more quickly, become more cohesive, offer more immediate support to group members, are better attended, have less conflict, and provide more rapid relief of symptoms
. However, many clinicians believe that they do not lend themselves to long-term psychotherapeutic work with ambitious goals of personality change. The homogeneous group, in contrast to the heterogeneous group, has a tendency to remain at superficial levels and is a less effective medium for the altering of character structure.

The issue becomes clouded when we ask, “Homogeneous for what?” “Heterogeneous for what?” “For age?” “Sex?” “Symptom complex?” “Marital status?” “Education?” “Socioeconomic status?” “Verbal skills?” “Psychosexual development?” “Psychiatric diagnostic categories?” “Interpersonal needs?” Which of these are the critical variables? Is a group composed of women with bulimia or seniors with depression
homogeneous
because of the shared symptom, or
heterogeneous
because of the wide range of personality traits of the members?

A number of authors seek to clarify the issue by suggesting that the group therapist strive for
maximum heterogeneity in the clients’ conflict areas and patterns of coping, and at the same time strive for homogeneity of the clients’ degree of vulnerability and capacity to tolerate anxiety.
For example, a homogeneous group of individuals who all have major conflicts about hostility that they dealt with through denial could hardly offer therapeutic benefit to its members. However, a group with a very wide range of vulnerability (loosely defined as ego strength) will, for different reasons, also be retarded: the most vulnerable member will place limits on the group, which will become highly restrictive to the less vulnerable ones. Foulkes and Anthony suggest blending diagnoses and disturbances to form a therapeutically effective group. The greater the span between the polar types, the higher the therapeutic potential.
⁶¹
But the head and tail of the group both must stay connected to the body of the group for therapeutic benefit to emerge.