Controversies in Psychotherapy Research: Epistemic Differences in Assumptions about Human Psychology

Introduction

It is the thesis of this paper that differences in philosophical assumptions about the appropriate subject matter and methods for human psychology have contributed to the controversy about the external validity of empirically supported therapies (ESTs). The appeal of establishing a scientific psychology based on the methods and principles of the natural sciences has been the priority of most academic psychologists since the latter half of the nineteenth century. Since the nineteenth century critics have expressed concern about the potential limitations of a human psychology based entirely on the natural sciences model, yet these concerns are rarely voiced in contemporary psychology textbooks. The concern asks whether there are fundamental differences between human beings and other objects and nonhuman species that require a psychology unrestricted by the epistemic assumptions and methodologies of the natural sciences. In order to understand the differences in epistemic models that are the basis for this controversy, it is worthwhile to review briefly the core assumptions that are implicit in the design of existing EST studies.

There are several core principles at the foundation of the natural science approach to knowledge development in psychology; among these the principles of empiricism and verifiability are central. Empiricism holds that only those propositions having direct implications for sensory experience are cognitively meaningful and within the domain of science (Friedman, 1999). Sense-based observations are assumed to be both necessary and sufficient to provide the basis for the formulation of probabilistic laws and principles about the psychological phenomena under study. The corollary principle of verifiability asserts that a scientific proposition is meaningful only if there is a finite experimental process for conclusively determining whether the proposition is true or false. A related movement in the philosophy of science, positivism,refers to an early twentieth century movement in physics known as logical positivism. Positivism asserts that the ideal scientific concept is one that adheres in practice as closely as possible to the theoretical definitions and the experimental operations that researchers used to obtain the observations relevant to proving the concept. Despite adoption of a positivistic model of science by most psychologists since the mid-twentieth century (Ayers, 1952), the assumptions of positivism were rejected as naïve by physicists nearly 50 years earlier for several reasons, most notably of which was that scientific constructs are implicitly defined by their place in the theory, and most observations are theory laden (Carnap, 1936/1937; Feest, 2005; Hanson, 1958). A related bias in scientific psychology is reductionism,the view that general levels of observation are most informative and valid when they can be folded into events that occur at more fundamental levels (e.g., behaviors, automatic thoughts, genetics and neurochemistry).

In summary, academic psychologists have long attempted to develop a scientific psychology by adopting the assumptions and methods of the natural sciences as the only legitimate framework for all psychological knowledge. This approach to human psychology is based on the assumption that, as with all phenomena in the natural world, there are underlying causal mechanisms operating beneath the surface phenomena of actions and experiences and these can be observed directly or indirectly and represented as law, like regularities. These laws or principles must in turn be based on sensory observations discernible to all researchers and must be capable of being verified as true or false, preferably in experiments. The advantages of the natural science approach to psychology are considerable, both in terms of the elevated status of the natural sciences and the enhanced access to resources and funding.

Freud accepted the knowledge framework of the natural sciences and struggled with how to integrate his clinical observations and insights into his understanding of early twentieth century science. He attempted to resolve this dilemma by developing a meta-psychology based on the assumed existence of basic drives that he believed could explain how and why psychic events occur. Freud’s meta-psychology was soon rejected by most natural scientists, however, as largely metaphorical and unable to meet the epistemic and methodological criteria at the foundation of the natural sciences. Cognitive-behavioral theorists (CBT) also have attempted to broaden the content of scientific human psychology to include mental events while maintaining the boundaries of normative science. This group tried to achieve this goal by construing mental events as cognitive processes (or the consequences of the process); the potentially observable events were conceptualized as discrete, internal mental representations or symbols, the rules of which might be described by algorithms related to specific neural mechanisms and functional circuits. Cognitive behavioral therapists have developed strategies for changing problematic human thoughts and behaviors while attempting to remain within the assumptive framework of the natural sciences by limiting their observations to self-reports that allow for the formulation of general principles and therapeutic strategies that can be tested in experiments or quasi-experimental studies. Given the epistemic framework adopted by CBT clinicians it is not surprising that the majority of therapies on lists of ESTs are behavioral or cognitive-behavioral therapies.

Empirically Supported Psychotherapies

Since the 1990s policy makers and third-party providers have demanded evidence that psychotherapy is both efficacious and cost effective and they have limited support and reimbursement for services to therapies that have been demonstrated efficacious in randomized control studies. In 1995 these pressures prompted publication of a white paper call for action by a Task Force of the American Psychological Association (APA, 1995). The APA Call for Action (Template for Developing Guidelines: Interventions for Mental Disorders and Psychosocial Aspects of Physical Disorders) emphasized the need to identify evidence-based psychotherapy practices if psychotherapists are going to continue to compete in the mental health market place. The APA white paper emphasized that psychotherapists must provide evidence of proven efficacy for specific disorders in the form of lists of empirically supported therapies ([ESTs] Chambless and Holton, 1998). The criteria for identifying an established EST were described in detail by Chambless and Holton (1998) as follows: “a treatment should be manualized and demonstrated to be more effective than other treatments or placebo, or equivalent to an established EST in at least two randomized control group studies or in a number of single case design experiments conducted by different researchers” (Chambless and Holton, 1998, p. 9). Since the publication of the white paper, scores of therapy manuals have been published that meet criteria as empirically validated or supported. Given the constraints implicit in requiring therapy manualization and limiting outcome indicators to symptom reduction, lists of ESTs are presently dominated by behavioral and cognitive-behavioral (CBT) therapies.

As lists of ESTs have proliferated, advocates have become increasingly critical of practitioners of “unscientific” meaning-based forms of psychotherapy that do not lend themselves to manualization and do not limit their goals to symptom reduction. These therapies are referred to as “less essential and outdated” (Calhoun, Moras, Pilkonis, and Rehm, 1998, p. 151), and articles in the popular press suggest that these practitioners are outdated and uninformed, e.g., “Is Your Therapist a Little Behind the Times?” (Baker, McFall, and Shorham, 2009), and “Ignoring the Evidence: Why Do Psychologists Reject Science” (Begley, 2009). Empirically supported therapies advocates have proposed that the APA mandate training in and use of ESTs and “impose stiff sanctions, including expulsion if necessary” against clinicians who do not comply with EST practice guidelines (Lohr, Fowler, and Lilienfeld, 2002, p. 8). Others have suggested that the knowledge accumulated on the basis of EST studies indicates that it is time to differentiate between “psychological treatments” and “psychotherapy” (Barlow and Carl, 2011). They argue that we should use the term “psychological treatments” to refer to those established empirically based practices that are designed to address specific diagnoses within a health-care context. The term “psychotherapy” would, in turn, be restricted to references to the processes and procedures designed to resolve issues, such as: interpersonal and relationship problems, concerns related to the meaning of life, and ideas about personal growth and actualization concerns. This would be administered outside the health-care system. This proposal is, of course, compatible with the priorities and values of empiricist psychology, descriptive psychiatry, and managed care companies, but it ignores important methodological limitations and conceptual concerns about the external validity of EST efficacy studies (Shean, 2012; Wampold, 2005, Westen and Morris, 2001). Additional validity issues related to questions about long-term outcome, relapse rates, symptom substitution, and the interpersonal impacts of decontextualized approaches to symptom reduction, must also be considered. There is evidence that these concerns are appropriate, despite their being ignored or dismissed in the EST literature (Doesschate, Bockting, Koeter, Maarten, and Schener 2010; Hafner, 1976, 1983, 1984; Keltner, 1987; Ottenbacher and Ottenbacher, 1981; Shea, Elkin, Imber, Sotsky, and Watkins, et al., 1992). The old fable of the men using different sensory systems to describe and elephant applies here, in that theories both help us to focus and to see more clearly and to blind us.

There is a form of circularity inherent in the design of psychotherapy efficacy studies in which psychological problems are reframed as problematic behaviors and thoughts to be treated by manualized therapy techniques. This circularity is evident in the manner in which problems, treatments, outcome criteria, and indicators of effectiveness are defined, implemented and measured (Butler and Strupp, 1986; Russell, 1994; Stancombe and White, 1998; Strupp, 1996). Randomized control trial psychotherapy efficacy studies are structured and privileged by the epistemic views, values, and moral understandings of those conducting the studies (Cushman and Gilford, 2000; Luborsky, Diguer, Seligman, Rosenthal, Krause, and Johnson, et al., 1999; Martin and Thompson, 2003). Circularity occurs because psychological theories do not simply describe or explain events that are independent of them, but they are integrated with what constitutes the objects or topics of study (Richardson, Fowers, and Guignon, 1999). Many EST proponents have failed to acknowledge this circularity and do not recognize the inferential distance between efficacy studies with selected, patient samples and effectiveness studies conducted in naturalistic practice settings (Seligman, 1996). There is, for example, much writing on the effectiveness of idiographic psychotherapies that is ignored in the EST literature (Leichsenring, 2001, 2005; Shedler, 2010; Westen, and Harnden-Fischer, 2001; Wampold, 2007). In addition the role of process variables in psychotherapy outcome has been ignored by most efficacy studies.

Psychotherapy process research indicates that the quality of the therapeutic relationship accounts for about 30% of the variance in outcome; therapist techniques about 15%; expectancy, therapist credibility about 15%; and environmental and patient characteristics (e.g., readiness for change, openness, engagement, active participation, ability to verbalize feelings) about 40% (Asay and Lambert, 1999). Evidence also indicates that therapist characteristics have a significant effect on outcome so that therapists within treatments vary significantly in their outcomes (Horvath and Bedi, 2002; Kim, Wampold, and Bolt, 2006; Wampold and Brown, 2005). These effects have been ignored in the design and conduct of RCT studies; as a consequence treatment effects are likely to be inflated (Wampold and Serlin, 2000), and progress in understanding the relationship between process and outcome is obscured (Baldwin, Wampold & Imel, 2007; Kasper, Hill, & Kivlighan, 2008). We do not understand, for example, the effects of moderator variables on psychotherapy effectiveness, including levels of distress and impairment, co-occurring problems, dispositional traits, self-reflectiveness, openness to experience, access to social supports, or coping skills, nor do we understand how these moderators interact with therapeutic approaches. Empirically supported therapies efficacy studies have ignored these variables. Research is needed to identify when, how, in what context and with whom strategies are most likely to be effective, and whether or not these strategies can be effective across theoretical approaches. Given this brief review of the limitations of efficacy studies it would be wise to remain circumspect about generalizations and conclusions based on the assumptions of the external validity of lists of ESTs.

Meaning-Based Psychotherapies

Many practicing psychotherapists find the literature on empirically supported therapies to be an informative but limited resource for understanding and addressing the range and complexity of problems encountered in actual practice. These practitioners tend to favor a range of individual, meaning-based, and narrative approaches that foster enhanced understanding in psychotherapy (e.g., gestalt, humanistic-existential, psychodynamic/object-relations). For want of a better term I refer to this group of approaches as meaning-based therapies. Meaning-based therapists share the view that the diversity of problems encountered in the consulting room—and therapeutic effectiveness—cannot be adequately understood in the context of the epistemic assumptions of the natural sciences. They emphasize the importance of the therapeutic effects of process variables and the importance of interpersonal engagement in an individualized (and relatively unstructured) process intended to generate improved understanding of experiences, actions, feelings, memories and relationships and their associated meanings. During this process various strategies are implemented to facilitate the likelihood that individual experiences and meanings will be examined, rekindled, explored, deconstructed, reconstructed, and re-connoted in ways that are broadly intended to facilitate new, more fulfilling ways of living and relating (Kaye, 1995). Meaning-based therapy is understood as a process between a therapist and a patient, involving a unique series of semistructured interactions and related experiences that take place over time. Both therapist and patient come with unique characteristics, and the interaction that takes place in a particular context is expected to result in a reciprocal process with unpredictable emergent results (Wampold, Hollon, and Hill, 2011). Not everyone seeking psychotherapy services is interested in, capable of, or motivated to engage in narrative therapy, similarly not everyone seeking therapy presents is primarily interested in therapies that focus on symptom reduction, and many individuals seeking psychotherapy services do not present with prominent symptom-based complaints. Many are interested in changing the trajectory of their lives in a broad sense. There are obvious limits to the degree to which meaning-based therapies can be described in manuals, as these therapies are based on reciprocal processes and responses to interventions and must accommodate the unique qualities of the therapeutic relationship as well as the background, potential, and circumstance of each individual. Meaning-based approaches may include (but typically extend beyond) symptom control (which involves help in connecting diverse sources of information and experiences, such as memories, emotions, omissions, recurring patterns and ways of relating) with the goal of living a more satisfying and effective life, having richer and more satisfying relationships, and finding new, more effective ways of having patients become more effective agents in their involvements (Martin and Thompson, 2003). Meaning-based therapies are structured and intended to facilitate engagement in a process of exploration and mutual understanding, in ways that help reduce problems and reveal the truth in what has been previously been misunderstood, distorted, and concealed so that opportunities for new experiences are increased. Strategies are timed and applied selectively on a case-by-case basis to allow the client’s experiences to unfold through a relatively unstructured process involving openness, spontaneity, and uninterrupted communication (Frie, 2010). The goals are to engage and influence the self-understanding of clients in a manner that is both historically and culturally grounded and idiographic, reflexive, evaluative, and related to issues related to understanding and living within appropriate limits of control and freedom, with emphasis on agency and the interpersonal and embodied basis of experience (Frie, 2003; Woolfolk, 1998). Meaning-based therapies are founded on the assumption that there are few facts about human problems that that can be understood apart from the meaning-laden narratives that form and shape our lives. Actions change as a result of shifts in understandings that define who we are (Guignon, 1998). Human actions are assumed not only to be past shaped, but also to be understood in relationship to the accomplishment of specific aims and goals within the individual’s culture and life story.

Despite extensive research, evidence for the effectiveness of meaning-based therapies, they have been dismissed and described as unscientific and rejecting science because they cannot be evaluated in the context of RCT designs (Begley, 2009). It is not that meaning-based therapists “reject science” so much as that they do not accept the epistemic limits of what is considered to be appropriate evidence characteristic in the natural-science approach to psychology. The problem for these therapists: how are claims to knowledge to be substantiated in ways that minimize the risks of relativism, self-deception, and doctrinaire authoritarianism?

Hermeneutics

Hermeneutics is a branch of philosophy that establishes the validity of understandings when constructs are interpretive and do not refer to things that literally exist in nature. Hermeneutics emerged during the seventeenth century as a discipline to interpret of biblical texts. Wilhelm Dilthey (1894/1977) was among the first to suggest that hermeneutics can provide an approach to knowledge development in psychology, an approach that transcends the limits of a natural science model. Dilthey argued there is a fundamental distinction between the natural sciences, which observe as separate the world of natural objects in order to derive causal laws, and understanding of the meanings and experiences of human beings living in the world, in which facts cannot be separated from context and values (or observations) from theory. He proposed applying the hermeneutic circle as an alternative approach to ascertaining the validity of understanding and interpretation through ongoing discourse and judgment which is based on the logical ordering and harmony of the parts with the whole. The hermeneutic method, as applied to psychotherapy, requires that descriptions, interpretations, and analogical understandings be capable of convincing both experts and clients that the interpretation can stand up against other ways of understanding.

Heidegger (1962/1960) was an important contributor to the development of hermeneutics. Heidegger focused his philosophical work on the question of what it means to be human. He concluded that as human beings we care, and because we care we are immersed in involvements with other entities in our lives in an immediate, unreflective manner. In contrast to the Cartesian view, we do not stand apart and observe but are immersed in our activities and take for granted the context and background to our understandings and experiences. It is not detached knowing but practical engagement in activity, discourse, and interaction that is characteristic of our mode of being. It is only when our expectations and routines are disrupted by some sort of failure that we may detach, stand back, and attempt to understand or go beyond and question the assumptions, conventions, and beliefs implicit in our day-to-day experiences and practices (Martin and Thompson, 2003). Heidegger maintained that our mode of being-in-the-world always involves a context within which experience occurs; this context provides a clearing or horizon of experience that consists of the customs, practices, language, and culture in which we develop and live; it is a clearing or horizon that cannot be separated into subject and object.

Gadamer (1995/1960) built on Heidegger’s philosophy and attempted to identify what is common to all forms of human understanding. He views the fundamental mode of human being-in-the-world as that of understanding and interpreting, and he argues there can be no knowledge free of presuppositions. This background or “horizon of intelligibility,” described by Gadamer (1960) guides all understanding, including how we pose questions and the manner in which we evaluate which questions and answers make sense so that there is a circularity between past and present in any interpretive inquiry. Understanding occurs when there is a fusion of horizons of intelligibility, whether applied to interpreting historical texts or understanding others. Gadamer’s approach to understanding of truth in art and text suggests that the process of hermeneutic inquiry develops from and fulfills what is implicit in the reality in what was previously disorganized or inchoate. His views can be applied to the disorganized and disruptive life stories of many psychotherapy clients (Guignon, 1998). In this sense, meaning-based psychotherapy does not represent a straightforward uncovering and assembling of the nature of specific thoughts and behaviors that increase or decrease by the application of psychological science; meaning-based psychotherapy involves processes that foster engagement involving the informed selection, interpretation, organization, and arrangement of emotions, memories, thoughts, and perceptions from many possibilities that shape our lives (Woolfolk, 1998). What matters in terms of outcome is how these new experiences and understandings play out in terms of fostering improved understanding of the continuity and coherence of our lives in ways that can be applied to an improved satisfaction, meaning and purpose. The goal of meaning-based therapies is understanding that is not scientific comprehension (episteme) or technical knowledge (techne) but the form of wisdom that Aristotle referred to as phronesis or practical knowledge. Phronesis involves being able to understand situations and to choose means and ends appropriate to the situation (Bernstein, 1983; Woolfolk, 1998).

Meaning-based psychotherapy allows the client to become aware that the particular interpretive horizon in which he is living admits only selected experiences and ways of being, which are inchoate and dysfunctional. The goals are identified through a process of mutual encounter and dialogue intended to facilitate the probability that this horizon will shift and broaden so that different, more fulfilling, and effective ways of feeling, behaving, and thinking become possible (Cushman, 1995). These approaches do not deny the impact of unresolved childhood conflicts on current experience and ways of relating, but they often replace the repression and unconscious with an emphasis on the embodiment of childhood experiences generalized to the present (Chessick, 1990; Merleau-Ponty, 1962). Hermeneutics provides a method to validate the interpretations and effects of narrative therapies that does not transform people into objects. It helps us to avoid the dangers of relativism and doctrinaire authoritarianism inherent in distortions that result from uniform rejection of the epistemic assumptions and methods of the natural science model of human psychology.

Conclusion

In the human sciences theory transforms self-understanding and alters the constitutive features of practices and vice versa, in a reciprocal manner (Richardson and Slife, 2011). A psychology based entirely on the assumptions and methods of the natural sciences cannot escape self-defining conclusions about what is worthwhile and how it is to be studied, nor can it reach the degree of certainty associated with the natural sciences. This model of psychology assumes that beneath surface phenomena there are common causal laws and principles that are discernible by systematic observation and testable by experiments or quasi-experimental studies. Meaning-based therapists tend to view the epistemic assumptions of the natural science model as too limited to allow for a complete understanding of the complexity of human psychology. They tend to emphasize the importance of interpersonal engagement in relatively unstructured interactive processes that are expressed in sharing of patterns of experiences, actions, feelings, memories, and concerns about relationships and their associated meanings; these processes increase the likelihood that previous and current experiences and meanings will be examined, rekindled, explored, deconstructed, reconstructed, and reconnoted in ways that facilitate growth and recovery (Kaye, 1995). As a consequence, meaning-based psychotherapists emphasize the salience of a whole socially and historically constituted person who transcends diagnostic categories. These therapists emphasize the importance of interventions that improve agency and increase experience-based awareness of the interpersonal and embodied basis of attitudes and actions (Frie, 2003). The goals of meaning-based psychotherapy are to help patients achieve understandings that help them live in a manner that is more faithful to the integrity of their life stories as they become more effective moral agents in their personal and interpersonal engagements (Martin and Thompson, 2003). Of course, there must be a means of ascertaining whether interpretations, understandings, and changes that are observed are sound and valid, even though they cannot be final. Truth in this sense is not static but fluid because new understandings and changes are based on the individual’s life narrative, and projected outcomes are constantly being revised in light of new circumstances and developments. This understanding of psychotherapy requires a different model for evaluating outcome, one that is not reducible to a static relationship between interpretation and fact (Guignon, 1998). Hermeneutic philosophy provides a process by which one can establish a form of validity that is applicable when constructs do not refer to things that literally exist in nature. From a hermeneutic perspective the validity of understandings and interpretations are considered on the logical ordering and harmony of the parts with the whole and on the manner in which they are taken up and applied to new situations. These changes should also be capable of convincing experts (as well as patients) that the interpretation can stand up against other ways of understanding (Gadamer, 1995). These desired changes are not necessarily arbitrary or unscientific: they must foster realizations and understandings of meanings and structures that were implicit but not adequately understood in the individual’s life trajectory.

If lists of ESTs are adopted as required forms of therapy for accreditation, credentialing, and third-party reimbursement, these policies will result in changes that are not in the best interests of therapists or individuals seeking mental health services. Hermeneutic philosophy provides an approach to understanding psychotherapy outcome that allows therapists to regard the client as an important source of truth, to take the role of developmental and cultural contexts seriously, to appreciate the complexity of human psychology, to resist reductionistic and technical explanations, and to appreciate the role of one’s own situation in shaping our ability to understand another person (Orange, 2010). There is much more to be learned about the processes and effects of different psychotherapy relationships as well as approaches. Implementation of proposals to restrict certification, reimbursement and research funding based on lists of ESTs will predictably stifle innovation and creativity in the field and deny access to psychotherapy services to many individuals who seek assistance for problems that do not meet DSM-IV or DSM-V descriptive criteria (Howard, Comille, Lyons, Vessey, and Lueger, et al., 1996).

Skilled meaning-based psychotherapy typically requires extensive training, supervision, and practice experiences beyond those provided by advanced degree programs. There is obviously an important place in every therapist’s competencies for knowledge and training in EST listed therapies, and many individuals are not candidates for narrative psychotherapies for any number of reasons. But calls for policies that exclude non-EST (aka meaning-based psychotherapies) from certification and reimbursement are premature and based on a narrow model of human psychology that is not sufficient to understand, study, and conceptualize many aspects of human experience and suffering that are addressed in meaning-based psychotherapies.