Psychotherapy Process Research Paper

1. Psychotherapy

The forms of psychotherapy most commonly include individual therapy (one therapist with one patient), couple therapy (one therapist with two married or emotionally committed patients), family therapy (one or more therapists with members of a multigenerational family), and group therapy (one or two therapists with several patients who were previously strangers).

The settings in which psychotherapy is practiced include community mental health centers, private professional offices, university or college counseling centers, inpatient or outpatient psychiatric facilities, social work agencies, specialized schools, correctional facilities, consulting firms, and church-affiliated counseling programs.

The practitioners of psychotherapy are not members of a single profession, but instead are individuals who have typically been trained in one of the following professions: psychiatry, clinical or counseling psychology, clinical social work, psychiatric nursing, marital counseling or other counseling specialties, or the religious ministry. The level of involvement of different professions in psychotherapeutic practice, and their relative standing, varies from country to country.

Psychotherapeutic practitioners also differ in terms of the theoretical models they use to guide their work with patients. Broadly, these include analytic and psychodynamic theories, behavioral theories, cognitive and cognitive-behavioral theories, experiential and humanistic, and systemic or family-systems theory. Traditionally, these theories have formed the bases for rival ‘schools’ of psychotherapy. More recently, many therapists have adopted an eclectic or integrative attitude, seeking to broaden their effectiveness by combining the strengths of several models. According to the theory followed, the treatment setting and format, and the professional identity of the therapist, these varied mental health practices may be known as psychotherapy, therapeutic counseling, psychoanalysis, behavior modification, personal training, growth facilitation, or some other term. Their practitioners may be referred to as psychotherapists, counselors, psychoanalysts, or simply by the name of the profession in which they were trained (e.g., psychologist, psychiatrist, social worker). The persons who seek help from these practitioners may be called patients, clients, counselees, or analysands. For simplicity in discussion, I refer to all those treatments as psychotherapies, to the persons who practice them as psychotherapists, and to the persons who experience them as patients.

2. Psychotherapy Research

Scientific research on the psychotherapies emerged as a field of inquiry in the middle of the twentieth century, following World War II, through the work of teams led by investigators such as C. Rogers, J. Frank, D. Shakow, M. Lorr, and others (Freedheim 1992, Chaps. 9–12, Russell 1994, Chap. 7). In this field, a distinction soon grew up between studies that focused mainly on treatment effectiveness, or the impact of therapy on patients, which were called ‘outcome’ research, and studies concerned primarily with the interactions of patients and therapists during the course of therapy, which were called ‘process’ research. Even from the outset, however, some studies included measures both of process and outcome, and in the years that followed process–outcome research has continued to be a vital focus of inquiry (Orlinsky et al. 1994, Orlinsky and Howard 1986).

Although this simple distinction between process and outcome served adequately for three decades, an increasing volume and complexity of studies and findings led eventually to the proposal of more finely differentiated frameworks for research. Based on extensive reviews of process–outcome, one of these (Orlinsky and Howard 1986) offered an overview of the field based on two distinctions: first, between therapeutic process (the observable events of psychotherapy) and its functional contexts (the social, cultural, and psychological circumstances in which therapy takes place); second, between context states that exist before and after the process events being studied, which influence and are influenced by them.

The influences of prior contextual states on events in therapy constitute determinants of therapeutic process. These would include, most immediately, the personal problems and psychological characteristics of the patient, and the therapeutic model, professional skills, and personal characteristics of the therapist. However, they also (if indirectly) include the organizational characteristics of the treatment setting, the nature and state of the mental health service delivery system, current societal pressures and stressors, and prevailing cultural standards of individual effectiveness, maturity, and well-being.

The reciprocal impacts of therapeutic events on subsequent contextual conditions comprise the consequences of therapeutic process. Researchers have been most directly concerned with the impacts of therapeutic events on the patient’s life and personality (i.e., clinical ‘outcome’). However, this broader framework also directs attention to the impact of therapeutic events on the professional and personal life of the therapist, and to the aggregate influence of patients’ and therapists’ experiences on cultural values and social institutions.

Between the contextual determinants of psychotherapy and its contextual consequences lies the set of psychological interactions and transformations that comprise the psychotherapeutic process. Researchers who would observe and analyze therapeutic process must, like photographers, first make a number of decisions about framing and focusing their studies. Framing a psychotherapy process study entails decisions about the location of the phenomenon (where to look), the temporal unit of observation (how long to look), and the observational perspectives to be employed (from which angle to look). Focusing the study involves making decisions about the facets and phases of therapeutic process to assess (which ‘variables’ or changes to measure).

3. Process Research Frames

3.1 Location

While therapeutic process may be defined abstractly in terms of psychological interactions and transformations, process researchers have differed over where, concretely, they can be best observed. Many researchers view therapeutic process as evidently consisting of the interactions and communications that take place between patient and therapist during the regular meetings in therapy sessions. On this view, therapeutic process includes all of the events that can be observed and recorded during therapy sessions. However, other researchers define therapeutic process in terms of the changes which patients undergo as they improve. This view directs observations primarily to the patient’s experience and activity in and outside of session, and regards much that occurs during therapy sessions as therapeutically inert or inessential. To distinguish them clearly, these contrasting but somewhat overlapping concepts will be referred to as therapy session processes and therapeutic change processes, respectively. Investigators who frame their studies in terms of session process generally take a descriptive, theoretically nonpartisan, approach to defining therapy. By comparison, investigators who frame their research in terms of change processes usually hold, and seek to test, some particular theoretical conception of therapy (e.g., Dahl et al. 1988, Rice and Greenberg 1984).

3.2 Temporal Units

The concept of process refer most generally to activities and events that proceed or evolve over time. However, some activities and events proceed or evolve so rapidly, or so slowly, that they normally cannot be noticed. Examples of the former are interactions that physicists observe among subatomic particles; or, more familiarly, the rapid projection of still photographs that we do not notice separately but experience sequentially as ‘moving’ pictures. Examples of the latter include the gradual cooling of romantic affairs into stable companionate relationships through years of shared routine; or, more remotely, the imperceptible erosion of mountains by millennia of wind and rain. The point is commonplace but basic and relevant.

Observations of process must be framed within the temporal span of human consciousness, but there is still a range of noticeable rates at which events occur. In observing therapeutic process, the researcher may choose to examine very brief events, like changes in facial expression or vocal intonation; or somewhat longer events, like the statement by one person of a coherent line of thought involving several sentences, or dialog between patient and therapist on a particular topic that includes successive statements by each; still longer events, like the quality of whole therapy sessions; or even longer events, like the opening, middle, and closing phases of an extended course of treatment. The temporal units or frames of reference that researchers select for study inevitably determine the kind of therapeutic events they can observe, and the descriptions of therapeutic process they give (Orlinsky et al. 1994, pp. 275–6).

3.3 Observational Perspectives

Any complex event can be viewed from several angles, and each observational perspective has the potential to reveal a different aspect of the event. In social interactions, including the interactive events of psychotherapy, the observational perspectives include those of the participants (called ‘participant observers’) and of immediate or remote bystanders (nonparticipant observers). In psychotherapy, there are two distinct participant observational perspectives, that of patients and that of therapists, who experience the events of therapy directly but from different vantage points. (The simplest case is that of individual therapies, which constitute the most frequent form of treatment; the same principle becomes more complex in practice when the treatment format includes more than one patient or more than one therapist, as in group or family therapies.)

To the patient’s and therapist’s natural observational perspectives, psychotherapy research has added that of nonparticipant observers by using clinical raters who are not part of the events being observed, and whose view of therapeutic process is generally mediated by audio or video recordings of therapy sessions or transcriptions of the verbal content of sessions. (Nonparticipant observations are sometimes also made by mechanical means, e.g., by using clocks to time speech durations.) Participant observers have a ‘first hand’ and ‘inside’ but also fleeting view of the events they experience. They have direct awareness of their own thoughts, feelings, and intentions, and immediate impressions of their own and others’ behaviors, but these are typically organized around their goals in action. By contrast, nonparticipant observers just gain a ‘second hand’ knowledge of therapeutic process, and can only infer the covert internal states of the participants, but they have the advantage over participant observers of being able to review the recorded events many times, without the situational pressures to respond that participants experience.

Researchers may, and have, used one, two, or all of these observational perspectives in their studies, depending on their interests and the questions they investigate. However, since patients, therapists, and outside observers have different perspectives on the events of therapy, they are bound to see somewhat different aspects of process. It should not seem strange to researchers, or an indication of methodological weakness, if observers who view therapeutic phenomena from different perspectives sometimes produce discrepant observations, since those discrepancies may validly reflect the inherent complexity of social phenomena.

4. Process Research Focus

When undertaking a study of therapeutic process, researchers understand that they are dealing with a set of events that is too complex to be observed in its entirety, and must simplify the focus of the study if it is to be successful. However a study is framed in terms of session process or change process, temporal units, and observational perspectives, the researcher must focus its observations on a few specific facets of therapeutic process: the ‘variables’ that are to be qualitatively assessed or quantitatively measured. For the simplification to be meaningful, these variables must be defined in terms of some conceptual model of therapeutic process, and to be widely accepted by other researchers that model must be generally accepted within the research community.

4.1 Research Variables

Traditionally, the variables examined in psychotherapy process studies have been differentially defined according to whether they focus on the instrumental adaptive (goal-directed, or task) aspect of therapist and patient behavior, or the social-emotional (expressive-interpersonal, or relational) aspect of therapist and patient behavior. Therapist instrumental or task oriented behaviors involve the technical interventions, methods, or procedures that various clinical theories indicate therapists should employ to help patients. Research variables include the nature, timing, skillfulness, and appropriateness with which therapeutic techniques are used. Examples of specific technical interventions in psychoanalytic therapy are use of free associations to interpret dreams, interpretation of patient resistances, and transference interpretations. A typical therapist intervention in client-centered therapy would be empathic reflection of feeling; in behavioral therapy, relaxation training to assuage phobic anxiety; in gestalt therapy, the ‘two-chair’ technique to explore and resolve internal conflicts; and so on. Patient instrumental-adaptive or task-oriented variables generally focus on the patient’s response to technical interventions, such as cooperativeness or resistance, as well as specific effects of interventions such as enhanced insight following an interpretation, deeper self-exploration and ‘experiencing’ in response to therapist empathy, or decreased anxiety when facing previously frightening stimuli as a result of relaxation training. Major research methods focused on instrument aspects of therapeutic process include those described in Greenberg and Pinsof (1986, Chaps. 2–8, 14).

A second traditional type of process variable that has been studied by psychotherapy researchers is defined in terms of the social-emotional or expressive- interpersonal aspect of therapist and patient behavior, and their respective contributions to the formation and development of the therapeutic relationship. Relationship variables generally focus on the ways that patient and therapist feel and behave toward one another: the manner or style of their relating, complementary to the content of their interactions. Again, specific qualities of relationships have been emphasized by different theoretical orientations. Researchers influenced by psychoanalytic thought have focused on two issues related to the clinical concept of transference: core-conflictual themes in relationships and therapeutic alliance. Others, influenced by client-centered theory, have focused on therapist-offered relationship conditions such as genuineness, empathy, and positive regard. Social psychological theories of interpersonal behavior have provided another conceptual approach to research on relationships, emphasizing the reciprocal influences between patient and therapist, focusing on dimensions of behavior such as acceptance vs. rejection and control vs. autonomy. Useful research methods for studying therapeutic relationships are described by their proponents in Greenberg and Pinsof (1986, Chaps. 9–12, 15).

A more detailed, systematic classification that has emerged from repeated reviews of the research literature relating process to outcome (Orlinsky et al. 1994) distinguishes the following categories of process variables: therapeutic contract, therapeutic operations, therapeutic bond, participant self-relatedness, in-session impacts, and sequential flow. These are aspects of process that can be observed simultaneously in any unit of process, viewed from any observational perspective.

The therapeutic contract includes variables indicating the goals, tasks, and norms of patient and therapist roles, as defined by the therapist’s treatment model; the number and qualifications of persons occupying each role; and the terms, setting, and schedule of their meetings. Therapeutic operations refer to the specific instrumental actions that patients and therapists perform in implementing the therapeutic contract, including the patient’s presentation of a problematic condition (e.g., symptoms), the therapist’s expert understanding of that condition (diagnosis), the therapist’s ameliorative interventions (techniques), and the patient’s response to those interventions (e.g., cooperation). The therapeutic bond is the personal connection that develops between the participants in therapy as they perform their respective roles as patient or therapist more or less in accordance with their therapeutic contract: their ‘personal compatibility’ (mutual feelings, and expressive attunement or rapport), and their ‘teamwork capacity’ (personal commitment to their roles, and interactive coordination). Participant self-relatedness reflects the intrapersonal aspect of patients’ and therapists’ involvement in the therapeutic process, including variables of self-awareness(e.g.,openness’s .defensiveness), self-efficacy (e.g., confidence vs. embarrassment), and self-control (e.g., expressiveness vs. constriction). The in-session impact of these interdependent and reciprocally influential aspects of therapeutic process consists of the patient’s ‘therapeutic realizations’ (e.g., hope, relief, understanding, courage) as well as the therapist’s reinforcements (satisfactions and frustrations). The sequential flow of these process aspects within the therapy hour is indicated by variables reflecting session development, and the flow over longer periods of time is indicated by variables reflecting the course of treatment.

4.2 Research Methods

The events of psychotherapy occur in a series of meetings between patient and therapist called therapy sessions. The most directly observable aspects of those events consist of conversation. In some types of therapy, conversation may be supplemented by inviting patients to engage in certain activities or exercises designed to provide significant learning or corrective emotional experiences for the patient. Given the prominence of verbal and nonverbal communications, researchers have relied heavily on audio and video tapes of therapy sessions to record the events of therapy. Recording and replay technology have provided investigators with voluminous records for detailed examination from an external or nonparticipant observational perspective, but have also constrained them to concentrate on temporal units that have generally been briefer than whole sessions, and sometimes no longer than one complete statement or interchange between a patient or therapist. Examples of relevant methods can be found in Greenberg and Pinsof (1986, Chaps. 2–8).

But conversation is just the outward side of therapeutic events. The inner side of process is comprised of patients’ and therapists’ experiences of themselves and one another during their sessions, and during the imagined conversations or fantasy interactions that they may have in the times between sessions (e.g., Greenberg and Pins of 1986, Chap. 13). This aspect of therapeutic process has been studied mainly through the use of postsession questionnaires. These retrospective ratings utilize the participant observational perspectives of patients and therapists to provide research data, and necessarily take the whole session as the temporal unit of observation.

Chapters providing further discussion of process research methods can be found in Bergin and Garfield (1994), Greenberg and Pinsof (1986), Horvath and Greenberg (1994), Kiesler (1973), and Russell (1994).

5. Main Process–Outcome Findings

A half century of research on psychotherapy has cumulatively demonstrated that qualities of the therapeutic bond or relationship, observed during the course of therapy from varied observational perspectives, robustly predict treatment outcomes as assessed by patients, therapists, clinical judges, and psychometric measures, although there is still some debate about whether these relationship qualities produce that outcome or merely reflect its on-going attainment. Global ratings of relationship quality are demonstrably linked to outcome from all observational perspectives. In terms of specific relationship qualities, therapists are better able to distinguish the patient’s contribution to therapeutic process, whereas patient as well as therapist process contributions are associated with outcome from the distinct observational perspectives of patients, clinical judges, and psychometric measures. The patient’s personal investment, collaborativeness, expressiveness, emotional responsiveness, and positive feelings towards the therapist are particularly important, as is the therapist’s empathy and credibility. This has led to a widespread view of the therapeutic bond or alliance as the most salient aspect of therapeutic process, and the ‘common factor’ that enables different types of psychotherapy to be more or less equivalent in effectiveness.

With regard to therapeutic operations, some types of technical interventions, when skillfully focused on patients’ life problems, have also been shown persuasively to be associated with positive outcomes (e.g., interpretation, experiential confrontation, and paradoxical intention), although this too is evident from the observational perspectives of patients and clinical judges rather than therapists. In terms of participant self-relatedness, assessments of patient openness (vs. defensiveness) have been very consistently associated with outcome, from all observational perspectives. Likewise, patients’ positive in-session impacts (‘therapeutic realizations’) have consistently been found to predict positive treatment outcomes.

6. Process Research And Practice

Psychotherapy process research thus far has proven itself a scientifically successful field of inquiry. Initially, it had to overcome an exclusion rule, stated by Freud (1953, p.185), that ‘the ‘‘analytic situation’’ allows the presence of no third party. … An unauthorized listener who hit upon a chance one of them would as a rule form no useful impression; he would be in danger of not understanding what was passing between the analyst and the patient.’ Thus in the earliest days, investigators were mainly concerned to demonstrate the feasibility of systematic and methodologically objective research on therapeutic process. Once that had been established (e.g., Rogers and Dymond 1954), another 15 years or so were devoted primarily to devising reliable measures and rigorous methods of process research, during which time some therapists who had taken an interest in the field were disappointed by the methodological preoccupations of researchers and the paucity of clinically applicable results (e.g., Talley et al. 1994, Chaps. 2, 4, 5). However, the rate of research on process and process–outcome patterns increased steadily from about 1970 onwards. As the 1990s approached, a considerable body of well-replicated findings had been amassed, from which implications for therapeutic practice could begin to be drawn (e.g., Talley et al. 1994, Chap. 6). Current pressures for accountability and quality assurance from managed care have forced therapists of all schools to become interested in outcome research, in order to demonstrate the effectiveness of their treatments. Yet the knowledge that most psychotherapists have of process and process– outcome research findings lags behind the times. As progress in process research continues, a major challenge will be to increase clinical awareness of the relevance and potential usefulness of these research findings.

Bibliography:

1. Bergin A E, Garfield S L (eds.) 1994 Handbook of Psychotherapy and Behavior Change, 4th edn. Wiley, New York
2. Dahl H, Kaechele H, Thomae H (eds.) 1988 Psychoanalytic Process Research Strategies. Springer-Verlag, Berlin
3. Freedheim D K (ed.) 1992 History of Psychotherapy: A Century of Change. American Psychological Association Press, Washington DC
4. Freud S 1953 The question of lay analysis. In: Strachey A (ed.) The Standard Edition of the Complete Psychological Works of Sigmund Freud. Hogarth Press and the Institute of Psychoanalysis, London, Vol. 20
5. Greenberg L S, Pinsof W M (eds.) 1986 The Psychotherapeutic Process: A Research Handbook. Guilford, New York
6. Horvath A O, Greenberg L S (eds.) 1994 The Working Alliance: Theory, Research, and Practice. Wiley, New York
7. Kiesler D J (ed.) 1973 The Process of Psychotherapy: Empirical Foundations and Systems of Analysis. Aldine, Chicago
8. Orlinsky D E, Grawe K, Parks B K 1994 Process and outcome in psychotherapy—nochveinmal. In: Bergin A E, Garfield S L (eds.) Handbook of Psychotherapy and Behavior Change, 4th edn. Wiley, New York
9. Orlinsky D E, Howard K I 1986 Process and outcome in psychotherapy. In: Garfield S L, Bergin A E (eds.) Handbook of Psychotherapy and Behavior Change, 3rd edn. Wiley, New York
10. Rice L N, Greenberg L S 1984 Patterns of Change: Intensive Analysis of Psychotherapy Process. Guilford, New York
11. Rogers C R, Dymond R F (eds.) 1954 Psychotherapy and Personality Change. University of Chicago Press, Chicago
12. Russell R L (ed.) 1994 Reassessing Psychotherapy Research. Guilford, New York
13. Talley P F, Strupp H H, Butler S F (eds.) 1994 Psychotherapy Research and Practice: Bridging the Gap. Basic Books, New York