Davis, Sean D

Some researchers have hypothesized that factors common across therapy models are large- ly responsible for change. In this study we conducted semi-structured, open-ended qual- itative interviews with three different MFT model developers (Dr. Susan M. Johnson, Emotionally Focused Therapy; Dr. Frank M. Dattilio, Cognitive-Behavioral Therapy; and Dr. Richard C. Schwartz, Internal Family Systems Therapy), Dr. Johnson and Dr. Schwartz’s former students, and each of their former clients who had terminated ther- apy successfully. We examined possible common factors in our qualitative data analysis. Common factors fell into two main categories of model-dependent factors and model- independent factors. This article-the first of two-reviews the model-dependent common factors, common elements found across three distinct therapies. They include common conceptualizations, common interventions, and common outcomes, each with several sub- categories. We discuss the clinical, training, and research implications of the results.

For decades, marriage and family therapy (MFT) researchers have attempted to discover if MFT works, and if so, why. Recent meta-analytic reviews of decades of comparative effectiveness research reveal two consistent findings. First, MFT does work. The average person receiving marital therapy is better off than 84% of the untreated sample (for marital and family therapy it is 65%), and improvement is generally sustained over a long period of time (Shadish & Baldwin, 2002). In fact, marital therapy has a larger effect size (the usual standardized measure of outcome of treatments) than coronary bypass surgery for angina and electroconvulsive therapy for depression (Shadish & Baldwin, 2002). Second, these results generally hold true regardless of the therapeutic model employed or, with a few exceptions, the presenting problem being treated (Shadish & Baldwin, 2002). These MFT findings generally parallel the individual psychotherapy literature.

Though several hypotheses exist to explain these findings, one prevalent hypothesis has been that “in their natural clinical form, there is considerable overlap among the major models of… therapy” (Jacobson & Addis, 1993, p. 88). While models of therapy differ widely in their conceptualization of problems, the problems they are describing and what the therapist actually does in therapy to ameliorate those problems may be remarkably similar among models. Perhaps what Asay and Lambert (1999) said of psychotherapy in general applies equally to MFT: “Different therapies embody common factors [italics added] that are curative, though not emphasized by the theory of change central to any one school” (p. 29).

In the larger field of psychotherapy, research aimed at uncovering the common factors of therapy models which account for change has blossomed over the past 15 years (Wampold, 2001), while similar research in MFT is still in its infancy (Davis & Butler, 2004). We know very little about common factors of change unique to MFT.

The next steps for MFT researchers interested in furthering our understanding of common factors include generating hypotheses about what components of different models are common factors responsible for change (Blow & Sprenkle, 2001; Sprenkle & Blow, 2004a; Sprenkle, Blow, & Dickey, 1999; Wampler, 1997), determining if they are necessary and/or sufficient by empirically relating these factors to outcome, understanding how and when they are implemented across various models (and if how and when they are implemented makes a difference), and formulating a theory or model of these components that provides adequate clinical, training, and research guidance. The end result could be a meta-theory of change in MFT which would point clinicians, researchers, and educators to what needs to happen when helping people change. Finding a core set of common factors responsible for change could greatly simplify MFT training and therapy. Such a theory could improve MFT training, research, and practice. We are a long way from complete development of any one of these goals, however, let alone all of them.

The small common factors literature in MFT has been met with recent challenges (Sexton & Ridley, 2004; Sexton, Ridley, & Kleiner, 2004). Sexton et al. (2004) begin their argument by acknowledging that “the basic premise of the position is probably correct: There are central and common factors that contribute to successful outcomes that cut across seemingly different theoretical and practice models” (p. 134). Despite their agreement with the core hypothesis of common factors. Sexton et al. (2004) and Sexton and Ridley (2004) raise the following concerns with the movement: (a) Is there research support for common factors? (b) Do common factors integrate research into practice? (c) Do common factors provide an adequate theoretical or conceptual foundation to explain the processes or mechanisms of change? (d) Do common factors advance theory development? (e) Do common factors provide the guidance necessary for clinical work? and (f) Can common factors serve as the basis of clinical training? (pp. 135-141).

Sexton et al. (2004) and Sexton and Ridley (2004) contend that the answer to all of the above questions is “no.” Although they offer different arguments for each point, a central theme of each argument is that, in their current articulation, common factors are too simplistic to move forward our understanding of the change process.

The current articulation of common factors in MFT consists of less than 10 articles (Davis & Butler, 2004), almost all of which are characteristic of the early stages of theory formation. As expected in these early stages, the current MFT common factors literature consists primarily of lists of proposed common factors (Blow & Sprenkle, 2001; Davis & Butler, 2004; Sprenkle & Blow, 2004a; Sprenkle et al., 1999), articles providing overviews of the common factors literature in individual psychology, and calls for MFT researchers to devote more attention to common factors (Duncan, Miller, & Sparks, 2003; Miller, Duncan, & Hubble, 1997; Wampler, 1997).

Thus, Sexton et al. (2004) and Sexton and Ridley (2004) are probably right. A student with no clinical training other than in the current MFT common factors literature would probably not have a sufficient road map to sit down with a client family and do as good of a job as a student trained in a well-developed therapy model. However, that is a reflection of the current emerging state of the MFT common factors research and practice, not the overall promise of the research should it be further developed. As Sprenkle and Blow (2004b) note in their reply to Sexton et al. (2004), “We never suggested the common factors approach to be a finished or completed product, as they seem to imply, and we hope to contribute to the on-going development of the approach” (p. 153). Should the common factors movement in MFT be further developed, it would likely provide the benefits that Sexton et al. (2004) acknowledge it would:

Without question, finding a common core of factors to explain successful therapy would be a major breakthrough. This finding would simplify practice, training, and research. It would unify the theoretical schools of MFT, which often compete against one another and find themselves in contentious struggles. In essence, it would serve as a shorthand explanation for the complexity of practice and the diversity of clients, settings, and the sometimes disparate research findings, (p. 131)

Thus we come to the purpose of the present study, which is to take the next step in common factors theory development by detailing an inductively derived, empirically based model of common factors in MFT. To do this, we utilized grounded theory techniques to search for commonalities that different models may share but are articulated differently across models (i.e., narrow factors). We also investigated broad common factors related to such things as therapist and client variables, expectancy, and the therapeutic relationship, which is the focus of our companion article. We synthesized narrow and broad common factors into a common factor meta-model of change in MFT. Both findings come from interviews conducted with three prominent MFT couple therapy model developers who work primarily from a specific theory, their former clients, their former trainees, and the trainee’s former clients. The model developers are Dr. Susan M. Johnson (one of the founders and current leader of emotionally focused therapy; EFT), Dr. Frank M. Dattilio (a current leader and refiner of cognitive-behavioral marital therapy, CBMT, and a former student of Aaron Beck, the founder of cognitive therapy), and Dr. Richard C. Schwartz (the primary founder of the internal family systems model; IFS). In order to provide a context within which to interpret the findings, the next section will focus briefly on what each of these clinicians believe leads to change.


Emotionally Focused Therapy

An EFT therapist (Greenberg & Johnson, 1988; Johnson, 2004) views relationships through an attachment theory lens (Bowlby, 1988). According to attachment theory, every person has a strong need to be accepted and loved, and they seek a secure relationship in which each person is emotionally accessible and responsive to these needs. These needs are referred to as attachment needs (Johnson, 2004). According to EFT (Johnson, 2004), if the attachment bond is threatened a person will respond with a display of emotions in an attempt to restore the attachment bond. Problems arise when – out of a perceived lack of emotional safety – a partner expresses secondary, harsh emotions such as anger rather than softer, primary emotions such as fear or loneliness. Secondary emtions tend to distance others, getting the individual less of what he or she needs – emotional closeness.

The target of intervention is the attachment-based emotional experience, expression, and responsiveness of each member of the couple. An EFT therapist intervenes experientially by having partners identify and express their primary (rather than secondary) emotions to each other and respond to those of their partner, thereby shifting the negative interactional cycle and creating a secure attachment in which each partner feels safe being emotionally vulnerable.

Cognitive- Behavioral Marital Therapy

A cognitive-behavioral marital therapist (CBMT) is concerned with behavior, the consequences that maintain that behavior, and the cognitions associated with that behavior (Dattilio, 2001). Behavioral aspects of CBMT have their roots in classical conditioning (Pavlov, 1932), in which an unconditioned stimulus like food, which leads to an unconditioned response like salivating, is paired with a conditioned stimulus. Over time, the conditioned stimulus alone results in the unconditioned response. Principles of operant conditioning (Skinner, 1953) also had a large influence on the development of CBMT. In operant conditioning, behavior is maintained by its consequences. Behaviors that elicit positive reinforcement are likely to be repeated; those that are punished or ignored are likely to be extinguished.

Cognitive aspects of CBMT have their roots in cognitive therapy (Beck, 1976; Ellis, 1962). Cognitive therapy is concerned with automatic thoughts based on inferences about others and one’s own behavior. These automatic thoughts are shaped by a person’s schémas – core beliefs about how the world functions (Dattilio, 2001).

A couple must learn both the behavioral and cognitive skills necessary to achieve healthy functioning, and it is the therapist’s responsibility to ensure they learn them. The couple must be able to understand the importance of CBMT concepts and be able to implement them on their own. The therapist’s role is to explain the concepts so the clients understand them. Then, the therapist employs numerous techniques through in-session practice and homework to ensure that the couple sees their behaviors differently and responds to each other more positively (Dattilio, 2002).

Internal Family Systems Theory

The IFS model was developed primarily by Dr. Richard Schwartz (1995) in an attempt to merge systems thinking with intrapsychic processes and larger cultural and political issues (Breunlin, Schwartz, & Mac Kune-Karrer, 2001; Goulding & Schwartz, 1995; Nichols & Schwartz, 2001). According to IFS, the human mind is divided into several different “parts” that follow the same systemic rules of relating to each other that groups of people do (Breunlin et al., 2001). Schwartz (1995) uses terms such as “exiles,” “managers,” and “firefighters” to represent these internal “parts.”

Additionally, everyone’s mind also contains a self that is different from the parts (Breunlin et al., 2001). Once differentiated from the parts (i.e., separated from their feelings and thoughts), the self is capable of standing meta to the parts and leading with compassion, confidence, and vision. The self is fully equipped to lead once it has been differentiated from the parts; it does not need further development by the therapist. The person’s internal as well as external systems both strive for balance, harmony, and leadership (Breunlin et al., 2001). The same rules that govern the person’s internal systems govern their external systems, so change in one will trigger change in the other.

According to Schwartz (1995), a healthy person has parts that understand and accept each other, and fully allow each other to perform their unique roles when needed. The self is fully in charge of directing the parts. The goal of treatment is to help the parts communicate with, understand, and accept each other without behaving in an extreme way. The therapist borrows techniques from many different schools of family therapy to achieve this, from experiential “empty chair” techniques to narrative restorying to paradoxical interventions.

Narrow and Broad Common Factors

Common factors are typically discussed as being either narrow or broad (Sprenkle & Blow, 2004a). Narrow factors (Lambert, 1992) are those elements of therapy directly informed by a clinician’s model. Narrow factors (the focus of this article) include common goals of therapy shared by different models such as shifting each partner’s cognition, affect, or behavior. Though called by different names, all are thought to be common goals of different models. Broad factors (Hubble, Duncan, & Miller, 1999; Tallman & Bohart, 1999), which we discuss in the companion article, are elements of therapy inherent in the nature of therapy itself; they are independent of a clinical model. Broad factors include the therapeutic relationship, client and therapist variables, and expectancy/placebo variables. The common factors position is often critiqued as not offering a model that integrates broad and narrow factors into a model of change (Sexton et al., 2004).

The results of our overall study generally fell into narrow and broad aspects of therapy. This is the first study on common factors to outline in detail both narrow and broad aspects of therapy, as well as to describe how the two combine to influence change in therapy. Because of the space necessary to adequately describe the narrow and broad variables in our study and the relationship between them, we have divided the results of our study into two articles. The present article will address the narrow or “model-dependent” variables (those variables that appear across models) we discovered, as well as implications of these variables for future MFT research, practice, and training. The second article will discuss the broad or “model-independent” variables, such as therapist and client variables, as well as how the model-dependent and modelindependent variables combine to influence change. The second article will also discuss the implications the model-independent common factors have for MFT training, research, and practice.


Research Questions Guiding This Study

The following research questions guided this study:

Client Perspective on Change

1 . What do the clients say helped them change?

2. Do the clients of different therapists report that similar things helped them change, different things, or some combination of the two?

3. If there are similarities in what the clients of different therapists say helped them change, do their comments follow the current common factors framework? If so, how?

4. If the client’s comments do not follow the current common factors framework, in what ways do they differ from the current framework?

5. If the clients report that different things helped them change, are these differences model-specific, random, or some combination of the two?

6. Are there differences between the model developers and their students in terms of what their clients say helped them change? If so, what are those differences?

Therapist Perspective on Change

7. What do the therapists say helped their clients change?

8. When discussing what helped their clients change, do different therapists report similar things, different things, or some combination of the two?

9. Do the therapists’ comments suggest evidence of common factors concepts intermingled with their theoretical conceptualizations of change? If so, what particular elements get mentioned?

Co-construction of Change

10. How did the therapist and client make sense of the change together?

11. Does the therapist have a sense of what the client would say helped him or her change? If so, how did he or she arrive at these opinions?

12. Does the client have a sense of what the therapist would say helped him or her change? If so, how did he or she arrive at these opinions?

Using Modified Grounded Theory Methods to Study Common Factors in MFT

Grounded theory methodology (Glaser, 1993, 2001; Glaser & Strauss, 1967) is well suited for a study of common factors because the notion of common factors suggests that there is a theory within the existing theories that begs to be discovered. Something is common among the theories of therapy that helps people recover from a multitude of problems. This methodology offers a systematic approach to inductively analyze raw data from clients discussing what helped them change. Unlike most qualitative approaches, however, grounded theory also deductively tests the tentative conclusions reached during the research process. That is, the researcher weighs new data with existing categories inductively derived from existing data to see if the categories continue to fit. A grounded theorist gathers a descriptive account of change in an effort to draw theoretical linkages between different conceptually linked categories derived from the description. This unique mix of deductive and inductive inquiry makes grounded theory methodology a good fit for studying an area where there is a promise of a theory-in-waiting.

A researcher using pure grounded theory (Strauss & Corbin, 1998) starts with no assumptions about the data – he or she starts from the ground up. In this study, however, we have several assumptions about the data (i.e., there are factors common across models at least partially responsible for change). This precluded us from using grounded theory in its purest form. Still, we used several data collection and analysis techniques from grounded theory (e.g., constant comparative method, analytic induction, open coding, and axial coding) to answer our research questions. We will refer to the grounded theory methods we used hereafter as a “modified grounded theory.”

Rationale for Selection of Participants

The main objective of our study was to discover whether or not there are similarities in the change process as reported by clients who received theoretically distinct therapy. Such an objective makes critical case sampling (Patton, 2002) an ideal method for recruiting a sample. Patton (2002) describes a critical case as “a statement to the effect that ‘if it happens there, it will happen anywhere,’ or vice versa, ‘if it doesn’t happen there, it won’t happen anywhere” (p. 236). In other words, if common factors are found in the experiences of clients receiving theoretically distinct therapy, then they will likely be found in clients of other therapists utilizing the same models.

In critical case sampling, critical cases are purposefully sought using whatever means of recruitment is most likely to secure participation in the study. We contacted six MFT model developers, and three agreed to participate. The model developers were asked to provide the names of former students whom they believed would participate, and they were contacted in a similar manner via email. All model developers located former students who were willing to participate.

Client Selection

Drs. Schwartz, Dattilio, Johnson, and one of their former students were each asked to recruit a couple to participate in the study. The therapists were given details about the study to aid in recruiting former clients. They also filled out an online consent form. Once clients agreed to participate we contacted them to set up an interview appointment, fill out an online consent form, and sign a release of information form.

Initially, we asked therapists to recruit a couple who had terminated successfully within the past year, who did not seek the therapist because of his or her model, and with whom they had primarily used the model the therapist is associated with. All of the therapists except for Dr. Schwartz and Ms. O’Neil, MSW (Dr. Schwartz’s former student), were able to locate a couple who fit that criteria. Dr. Schwartz and Ms. O’Neil were only able to locate an individual who had been primarily working on couple issues. Additionally, Dr. Schwartz’s client was still occasionally coming in for therapy about once a month. As working with an individual on couple issues is an approach theoretically consistent with Dr. Schwartz’s model, IFS, we proceeded with those interviews. Dr. Dattilio was unable to recruit any couples who fit the criteria exactly, so the first author interviewed a couple whom Dr. Dattilio was still seeing about once every 3 weeks and would soon be finished with therapy. Dr. Dattilio referred two former students. However, those students were unable to locate any clients who were willing to participate in the study, so they were excluded from the study. This exclusion may affect the results in unforeseen ways.


Three couples and two individuals working primarily on relationship issues participated in the study. Each participant chose a pseudonym by which they were referred in the interviews and all subsequent publications. Demographic data for each of the couples is presented in Table 1.

The client sample was relatively homogenous. Therefore, readers should exercise caution when transferring the findings from this study to a more diverse clientele.

Participant Nomenclature

A nomenclature was developed to facilitate identification of participant quotes during the coding and writing process. Nomenclature for each client is in Table 1 . The first three letters of the nomenclature identify the modality (e.g., EFT). The second two letters, preceded by a dash, indicate model developer (DV) or student of a model developer (ST). The third two-letter set indicates that the person is a client (CL). The final one-letter set indicated husband (H) or wife (W). So, EFT~DV~CL~W indicates Cassandra, the wife in Dr. Johnson’s client couple. As the IFS clients were individuals, there is no fourth letter set. For example, Bridgette, Ms. O’Neil’s (Dr. Schwartz’s former trainee) client is IFS~ST~CL.


Therapists were selected based on their either being an MFT model developer or leader or a former student of the same. Model developers were chosen because it was assumed that their clients would receive theoretically distinct therapy. Former students of model developers were chosen because it was assumed that their practice may be more transferable to an average clinician than the model developer’s practice.

Ms. Beth O’Neil, MSW, is Dr. Schwartz’s former student. Dr. Judy Makinen is Dr. Johnson’s former student. Additionally, Marianne Dattilio served as a co-therapist with Dr. Frank Dattilio periodically throughout therapy with their clients, but was not interviewed.

Trustworthiness, Credibility, and Transferability

Primary concerns many researchers have with qualitative research include issues of trustworthiness, credibility, and transferability (Erikson, 1986). To what extent do the data actually reflect the client’s experience? How do researchers account for the impact of their biases on the data? How do readers know if the study’s findings relate to the circumstances of others? These issues were addressed in part by using the constant comparative method of data analysis described above (Strauss & Corbin, 1998). Several additional steps (see Table 2) were taken throughout the procedure to ensure trustworthiness, credibility, and transferability.

I further sought to establish trustworthiness, credibility, and transferability through providing numerous quotes to clarify each category and subcategory so readers can decide for themselves whether or not they agree with my conclusions. In order for a category to be included, at least one therapist from each model had to mention it. Space limits the number of quotes that can be used in this article.1 Additionally, using the constant comparative method and triangulation – the process of using more than one source (e.g., researcher) to analyze the data – helped ensure the minimization of researcher bias.

Data Collection

Each couple and therapist were asked to participate in a 30- to 60-min open-ended audiotaped telephone interview. Each client was interviewed individually. Thirteen interviews (i.e., eight clients and five therapists) were completed. Questions used to guide the initial interviews are found in Appendix A (clients) and Appendix B (therapists). The interviews were guided by the broad research questions discussed previously. In general, the therapists were asked the same questions as the clients. In the constant comparative method of data collection (Strauss & Corbin, 1998), data are simultaneously collected and analyzed. I (the first author) kept a journal of potential themes that seemed to emerge during each interview. In a recursive process, the tentative themes from initial interviews were used to refine the questions asked in subsequent interviews, and data from subsequent interviews refined or disconfirmed existing themes. I also remained open to the possibility of new themes emerging in later interviews.

I (the first author) transcribed approximately one third of the interviews in preparation for data analysis. After signing a confidentiality agreement and completing training, four undergraduate students transcribed the remaining audiotapes.


First, I (the first author) independently read the 177 pages of transcripts and journal entries in an open coding procedure. Strauss and Corbin (1998) describe open coding as “the analytic process through which concepts are identified and their properties and dimensions are discovered in data” (p. 101). In other words, I “broke apart” the data and put it back together as I read through the transcripts and searched for broad categories of what seemed to contribute to change. This initial open coding was done on paper copies of the transcripts, with categories being noted in the margins. This initial coding took approximately 40 hours. As I read more transcripts, I deductively searched for information to support the categories (i.e., themes) that were emerging from the data, and inductively searched for overlooked data to form new categories. I completed the second round of open coding by going through each interview again, this time entering the data into NVivo, a qualitative analysis software program. Through this program I placed quotes from each interview into categories and tentative subcategories. This process took about 30 hours. During this process I also made notes about possible relationships between categories and subcategories for the next stage of coding.

I reached saturation-the “point of redundancy” (Crabtree, 1999, p. 41) when no new categories or subcategories emerged from the data – after about 10 interviews. I discovered considerably more themes at the beginning and middle of data collection and analysis than I did at the end, suggesting that I’d reached saturation. I continued coding the remaining three interviews for the purpose of further differentiating existing categories and to give equal representation to each model.

After I finished open coding I began axial coding (Strauss & Corbin, 1998). Axial coding involves further refining the broad categories by defining their subcategories and explaining how the subcategories are linked to the categories. I spent about 20 hours reading through the codes during this stage. The coding in this stage was done exclusively in NVivo.

At this point I addressed trustworthiness by having two master’s-level MFT students read 30% of the transcripts in an open coding procedure. Each of the students read the same interviews, and both found generally the same themes that I did. Slight differences in wording served primarily to further refine the existing subcategories.

Following this stage, I returned to the data to make relational statements. Strauss and Corbin (1998) said the following about relational statements: “We call these initial hunches about how concepts relate hypotheses because they link two or more concepts, explaining the what, why, where, and how of phenomena” (p. 135). Relational statements specify how the categories are linked theoretically. I continued in this endeavor until I reached saturation, which took about 30 hours.

Throughout the coding, I looked for themes as well as differences within couple dyads, within therapist/couple dyads, and across all couple and therapist/couple dyads. Since my focus was on identifying themes common among models, a primary focus of coding was across models. RESULTS FOR MODEL-DEPENDENT THEMES

Overall, the data fell into two broad categories: model-dependent themes and model-independent themes. The terms “model-dependent” and “model-independent” are used instead of the traditional terms “narrow” and “broad,” respectively, in order to make the relationship of the variables in each category to the therapy model more apparent. The model-dependent themes are the focus of this article; model-independent themes (i.e., themes inherent in the nature of therapy itself) will be discussed in the second article. Model-dependent themes are elements of therapy that are directly informed by the therapist’s model. Model-dependent themes would not exist if the therapist was not using a model. An example of a model-dependent theme would be the way the therapist conceptualizes the case. A therapist’s model informs the therapist what he or she should look for in the couple relationship. If a therapist did not have a model, he or she would have no way of conceptualizing the case. Model-dependent themes fell into the following two categories: (a) common conceptualizations and (b) common interventions. Each category had several subcategories.

Therapy seemed to roughly progress through the stages of conceptualization, intervention, and then outcome. Model-dependent and model-independent themes were found throughout each of those three stages of therapy, though we will discuss those stages primarily in this model-dependent article.

Model-Dependent Themes

Model-dependent themes, again, are common themes that appear across different therapies. They are themes that are directly informed by the therapist’s theory, but that appear to have common elements.

Therapist’s Common Conceptualizations

All therapists had conceptualizations of their client’s presenting problem that were informed by their model. They had a clear view of what the problem was and how it came to be. While much of the conceptualization was model-specific, there were three main areas of overlap: (a) family of origin influences on their current relationship; (b) current interactional cycles that resulted from prior family of origin experiences; and (c) affective, behavioral, and cognitive dysfunctions learned in the family of origin that perpetuated the current interactional cycle.

Family of origin influences on current behavior. At least one therapist from each model viewed his or her client’s problems as stemming from experiences in the family of origin. Dr. Dattilio described the following:

A lot of my time was focused on helping Geller deal with his feelings about being rejected and that went back to his childhood, and the [things] his wife did that sometimes reminded him of the way his father [acted].

Dr. Johnson mentioned that “[My clients] grew up in an Eastern-European culture which I know quite well and where I don’t think they had a very supportive environment to grow up in.” She went on to describe how, because of their family of origin and cultural issues, her clients did not learn to be emotionally available and responsive to each other.

Interactional cycles: cognitive, behavioral, and affective elements. Most therapists looked at how each spouse’s behavior, thoughts, and emotions influenced those of their partner’s, and vice versa. For example, Dr. Dattilio described how the wife’s passive behavior influenced her husband’s behavior: “One of the things that she learned is that she unwittingly was enabling her husband to be more and more like her mother all the time and wasn’t even aware of it.”

Dr. Johnson described her client’s interactional cycle this way:

They just want[ed] their partner to change and their partner to be more loving and accepting . . . They would sometimes both attack, but basically their negative cycle is that he would sort of shoot and run and withdraw.

Most therapists also identified maladaptive affect, behavior, and cognitions/beliefs as the three elements that perpetuated the dysfunctional interactional cycles. The following quote from Ms. O’Neil (IFS~ST) is characteristic of cognitive aspects of her client’s cycle:

. . . she was interpreting behaviors in a negative [way]. Her perception out of the lens of these parts [perceived] behaviors in a way that reinforced those beliefs of those negative parts. [This] didn’t really allow . . . her to have any access to his [or her] self . . .

Not surprisingly, there were some model-specific differences in which of the three elements different therapists emphasized. With EFT, clinicians focused more on affective elements while the CBT therapist focused more on behavioral and cognitive elements. IFS therapists seemed to focus on all three elements equally. Interestingly, however, most of these differences were relatively slight; therapists from each model frequently mentioned affective, behavioral, and cognitive/beliefs elements as contributing to their client’s current dysfunctional interactional cycle despite their model’s emphasis on one element over the others. For example, despite her model’s focus on affective elements, Dr. Makinen (EFT-ST) mentions her focus on the cognitive and behavioral elements that perpetuated her client’s cycle:

I think he needed to hear about the impact that his ranting, as they both called it, had on her and [how it] resulted in her withdrawal and distancing from him . . .

Even though CBT emphasizes cognitive and behavioral aspects of interactional cycles, Dr. Dattilio acknowledges also paying attention to affective elements of the cycle:

A lot of my time was focused on helping Geller deal with his feelings about being rejected [which] went back to his childhood, and the way his wife did [things] that sometimes reminded him of the way his father [acted]. He needed to process that emotionally.

The fact that the different therapists focus on elements outside the usual scope of their model is somewhat of a surprise, as EFT strongly emphasizes affective elements of interactional cycles (Johnson, 2004) and CBT strongly emphasizes cognitive and behavioral elements (Dattilio & Epstein, 2003). The IFS literature seems to slightly favor the emotional and cognitive elements over the behavioral (Breunlin et al., 2001).

Therapists’ Model-Specific Conceptualizations

Other than family of origin influences on current interactional cycles, there was little overlap between how therapists utilizing different models conceptualized their cases. All of the therapists spoke the language of their model very clearly as they described how they conceptualized their clients’ problems. The EFT therapists spoke at length about attachment, the CBT therapist spoke about the interplay between cognitions and behaviors, and the IFS therapists talked about parts.

Clients’ Conceptualizations

Clients’ conceptualization of the presenting problem varied widely. Practically all of the clients mentioned something quite narrow and vague, such as “my husband is too controlling,” “we just can’t communicate,” or other similar issues familiar to therapists. In fact, the lack of any common theme to their initial complaints was in and of itself the only consistent theme. When thinking about change, the lack of any commonalities to initial client complaints is meaningful, as it may evidence the very reason they are in therapy: they do not have an adequate way of thinking (i.e., a model) to explain how they got into their problem and what they can do to get out of it.

The Yellow Brick Road Map: Adopting a Model

The extent to which a therapist provides a model that clients view as a credible explanation for their problems (i.e., the model fits their experience) is another common factor in this study. The client’s adoption of the therapist’s problem conceptualization – the yellow brick road map, if you will – was evident as therapy progressed. When asked to reflect on their view of the problem when they first sought help, the clients’ responses were varied and often vague. However, clients tended to adopt the therapist’s conceptualization of their problem after they had been exposed to it in therapy. This process seemed to consist of the therapist offering an explanation for their problem, and, among our participants, the client accepting it. The effects of wholeheartedly adopting the therapist’s model are illustrated by IFS~DV~CL when he said, “There’s absolutely no way that I would have been in a place of this kind of internal clarity … if I hadn’t been working with this kind of model or something like it as [fervently] as I had.”

Once the clients adopted the therapist’s conceptualization of their problem (a willingness to do so is a common factor discussed in our companion article), the clients’ conceptualization followed the same themes as outlined above for the therapists (i.e., family of origin influences, interactional cycles, model-specific conceptualizations). For example, after being exposed to the therapist’s conceptualization of the problem, CBT~DV~CL~W described how her family of origin experience affected her current interactional cycle with her husband:

[I learned] quite a few things about [how] my upbringing . . . affected … my ability to communicate . . . and how I react to different things. [For example,] my husband has a very strong personality. How I react to [his personality] was one of the issues. [Through therapy] I became much more aware of my tendencies as it related to the marriage and interacting with Geller.

Prior to beginning therapy, she described her goals more vaguely:

… we were obviously experiencing some marital issues. I would say the primary goal was to improve our communication; to reduce the level of stress in the marriage.

EFT~DV~CL~H illustrates how initial vague goals were replaced with a model-specific conceptualization of their problem once therapy had begun. When asked to describe their initial goals, he said they wanted to “[be] able to see things from different perspectives. Our ultimate goal was to better the relationship. That was the ultimate goal.” Later, he showed his familiarity with EFT and emotional elements of interactional cycles by stating that in order for them to improve, they needed to become more vulnerable, “because … the other party will not shoot at a fragile person, at a person who is . . . openly willing to share emotions.”

Common Interventions

The second model-dependent category is common interventions. The therapists in this study used their model-informed conceptualization of the problem to drive their interventions. Therefore, the areas of overlap in their interventions are quite similar to the areas of overlap in their conceptualization. There was, however, a great deal more overlap in their interventions than in their conceptualization. Many of their interventions were couched in the conceptual language of their model but, pragmatically, looked very similar across models.

Raising Awareness of the Cycle and Each Individual’s Role in It

The interactional cycle – one of the common conceptualizations – was also a primary focus of intervention for therapists in each of the three models. Therapists’ intervention into the interactional cycle seemed to follow roughly the same pattern across the three models: (a) raise awareness of the interactional cycle by slowing down the process; (b) help partners be aware of their role in it by learning to stand meta to themselves; and (c) and help them adopt a healthier stance by encouraging personal responsibility. The temporal sequencing of these events was not always clear, nor did it seem that each step happened in isolation. At times it seemed that each step co-occurred with the others. Additionally, dysfunctional aspects of the clients’ family of origin seemed to be used throughout each of these steps to provide a context for their current difficulties. Rather than dwelling on the past, each model seemed to focus on altering patterns learned in their family of origin in the present.

Slowing down the process. The first step in altering the interactional cycle was to slow down the process. This was done in several ways, including structuring the amount of time each client talked, helping each person listen to his or her partner rather than jumping to conclusions, and helping clients see their partners differently. This increased awareness seemed to help couples slow down and begin to explore other possibilities for their current difficulties than what they came in with.

Standing meta. Once clients had slowed down the process, therapists helped them stand meta to, or outside of, themselves. Therapists used several different techniques to facilitate this process, such as reframing their partner’s behavior or intent. One of the most common methods was to encourage the clients to explore alternative explanations for their partner’s actions, thus allowing them to ascribe more benevolent intent to those actions. One creative intervention designed to help a client stand meta to him- or herself was used by Dr. Makinen (EFT~ST). The husband of the couple she was working with tried very hard to convince his wife to stop taking antidepressants. He talked a lot about this and would not let his wife share her feelings on the matter. When Dr. Makinen tried to interrupt him, he thought she was siding with the wife and got upset. Here is how Dr. Makinen handled it:

So I said to him . . . “We could tape the sessions and then what you can do is you can take the tape home and listen to it. You might find that useful.” . . . That was very powerful because he got to hear himself and the impact he had on her and how she shut down. And when they came back [for] the next session I asked them what that was like and he was just blown away by that … I think that helped make a big, big difference.

In other words, learning to stand meta helped the client realize his own role in the relationship problems. Therapists from each model gave similar examples.

Encourage personal responsibility. Slowing down the process and helping clients stand meta alone was not enough. Once clients had done this, therapists helped clients take personal responsibility for their role in the cycle. There was no one technique used to do this; instead, it was more directly or indirectly encouraged in the language that the therapists used with their clients. Dr. Dattilio, for example, told a client, “You’re a big boy now and you’re married, you’re a dad and have a wife and you need to knock that off.” Similarly, Ms. O’Neil (IFS~ST) talked about people needing to “own their own parts.”

Use of Metaphor

Several clients reported that their therapists provided a metaphor of the cycle and their role in it. Dr. Johnson’s clients spoke of being tanks “shooting at each other,” or “warriors hiding behind armor and shooting at each other.” They mentioned that this helped them keep the cycle and their role in it in mind as they focused on becoming more vulnerable. Dr. Dattilio’s clients spoke of a “table” metaphor that helped the wife remember that she was responsible for doing her part (i.e., being two of the four legs), but she couldn’t do her husband’s part (i.e., the other two legs). This helped her to remember to be responsible for her own emotions in an argument rather than her husband’s.

Family of Origin as a Context for the Cycle

At least one therapist and client from each model frequently referenced dysfunctional affect, behaviors, or cognitions learned in the client’s family of origin or previous relationships as the source of his or her role in the cycle. Current interactional cycles were explained as being the fruit of these early influences. However, time was not spent blaming the past. Rather, clients of therapists from all three models were encouraged to explore new ways of interacting in the here and now. The family of origin was simply used as a context to understand the origin of their current difficulties. This seemed to provide a credible explanation for what was happening in their relationship and a backdrop for their work at altering the cycle. For example, CBT~DV~CL~H states:

She comes from one background and one way of being brought up . . . and I’m coming from a different scenario and it’s like night and day. She’s working on her demons and I’m working on my demons and then together . . . it’s better.

Altering the Cycle

The methods used by each therapist to alter the cycle were largely model-specific. All therapists focused on helping their clients understand what was “underneath” or “driving” their stance in the cycle, but they largely focused on model-specific elements as they did this. For example, EFT therapists focused largely on emotional elements perpetuating the cycle, while the CBT therapist focused largely on cognitive and behavioral elements of the cycle. IFS therapists seemed to focus on all three aspects equally (though they seemed to slightly favor emotional and behavioral aspects as they talked about parts feeling and behaving in extreme ways). Though each therapist primarily favored the element that his or her model focused on, each focused on all three elements – behavioral, cognitive, and emotional – to some degree.

Emotional regulation. Using model-specific language, at least one therapist from each model commented on the importance of his or her client’s ability to learn to recognize and regulate his or her emotions in order to alter the cycle. In model-specific language, they all mentioned that clients need to shift from being emotionally reactive to being emotionally responsive to their partners, but each therapist had different ways of getting clients there.

The EFT therapists focused directly on processing primary (as opposed to secondary) emotions as a means of altering the cycle. One of Dr. Johnson’s clients stated:

It was learning a new technique – a new life approach actually, [learning] how to deal with your own feelings instead of expressing anger and frustration. I tried to talk deeper about how I feel in certain situations.

Her husband similarly stated, “[I learned to] … talk about how [I] feel – that I feel hurt, that I feel whatever – [to talk about] my feelings without accusing the other person . . .”

Dr. Dattilio also focused on regulating emotion as a means of altering the cycle, though he went about it differently. He believed that change would come more quickly if you focused on the cognitions that preceded the emotions:

[I] don’t ignore emotion; emotion is very important. [I just] don’t feel that emotional processing in and of itself is enough to get people to change and [I] think you can move a little quicker if you can get right to the antecedents, which is a lot of the cog- nition. A lot of my time was focused on helping Geller (the client) deal with his feelings about being rejected and that went back to his childhood and the way his wife did things that sometimes reminded him of the way his father [acted]. He needed to process that emotionally. Mediating it cognitively helped him really turn that around rather than focusing on the affect.

Cognitive reframing. Each of the therapists focused on reframing attributions that each partner made about the other that perpetuated the cycle. The intent that one partner attributed to the other’s actions was often seen as contributing to the dysfunctional interactional cycle and thus became the focus of intervention. Interventions aimed at altering these cognitions include exploring other alternatives to their spouse’s behavior, as discussed earlier. Consistent with his model, Dr. Dattilio also had his clients monitor their self-talk outside of the session, recognizing and challenging those beliefs that perpetuated the cycle.

Dr. Makinen (EFT~ST) deliberately changed her client’s interpretations of his own behavior. The client was making sexual advances toward his wife late at night when his wife was trying to sleep, which was causing a lot of distress in the marriage. In an effort to help him change his role in the cycle, Dr. Makinen reframed his desire as a temptation to be dealt with, as he knew the time of his advances would not be acceptable to his wife. As he was religious, viewing his behavior as a temptation provided him with a frame for his behavior that he could accept and deal with. Similarly, Ms. O’Neil (IFS~ST) used an activity she called “parts mapping” that drew attention to her client’s cognitions. She essentially had her client identify the statements (i.e., cognitions) coming from each of her parts. In other words, though each attempted to alter cognitions, therapists utilizing different models did it differently. The EFT therapists seemed to challenge their clients’ cognitions experientially in the course of an enactment or by using a reframe, and usually in the context of emotional processing. The CBT therapist directly challenged clients’ cognitions either in session or with homework. The IFS therapists, on the other hand, called attention to the cognition and let the clients choose what to do with it.

Behavioral shifts. As expected, Dr. Dattilio focused on directly altering behavior that perpetuated the dysfunctional cycle more than the IFS or EFT therapists. He spent some time teaching communication skills. Despite this, altering behavior was still an area of interventive overlap among the three theories. Therapists from each model tried in some fashion to get their clients to behave differently. Dr. Makinen (EFT~ST) had her clients structure their at-home interaction by setting a time limit on how long they could talk about an issue. One of the ways Dr. Johnson encouraged her clients to behave differently was by “taking more risks with each other at home.” This intervention was aimed at shifting emotions, but was directed at changing their behavior as a means of changing their emotional experience. Ms. O’Neil (IFS~ST) saw that her client got stressed easily; this led her to have a short temper with her husband, which set in motion a negative interactional cycle. Ms. O’Neil gave the wife homework designed to alter this cycle by engaging in self-care when she started to get stressed.

CL: [Ms. O’Neil’s exercises] . . . helped a lot . . . [She would say,] . . . “Ok, when you’re washing the dishes and you’re worrying about what’s out there that needs to be done and what about this application and what about this form, what about this messy house.” When it’s . . . like that, as hard as it is to do, [it] is the best time to climb in bed and read a book or take a bath, because then you gain perspective.

Model-Specific Interventions

It was easy to tell what model a client’s therapist used by listening to the client’s language. The EFT clients spoke about identifying, accessing, and expressing primary emotions. The CBT clients spoke about maladaptive cognitions and behaviors. The IFS clients spoke about parts (in fact, each IFS client had a very difficult time when I asked them to explain how they changed to a layperson unfamiliar with IFS language).

Common Outcomes

All of the clients in this study ended up experiencing remarkably similar changes at the conclusion of therapy. We will discuss the “common outcomes” category here for conceptual continuity with the “common conceptualizations” and “common interventions” categories, though the common outcomes were not a model-dependent variable.


A statement from EFT~DV~CL~H summarized a shift that most of the clients experienced when he said that he learned to “just give up [his] pride and show [his] softer side.” CBT~DV~CL~H said it this way: “It’s just time to … step outside [myself] and think of her.” IFS~ST~CL said, “We both stopped trying so hard to change the other person.” Clients reported a softening toward themselves and their partners. They voluntarily abandoned their previously harsh, critical view of their partners (and in some instances, themselves) in favor of being more patient and loving. When faced with an event that previously would have triggered a harsh response, angry emotions, and critical thoughts, they instead acted, felt, and thought more “softly.” They were able to exit on their own what previously would have become a negative interactional cycle.

It is one thing to soften; it is another for clients to know what to do once they have softened. Clients reported not only having a change of heart, so to speak, but also reported knowing what to do once they had softened to carry those changes into actual different ways of interacting with their partners.

Though we discuss changes in affect, behavior, and cognition separately, in reality changes in one aspect co-occurred with changes in the other two aspects. In other words, one person did not experience a softening of cognition without also experiencing a softening of behavior and emotion. Clients emphasized shifts in some elements more than others depending on which model they received, but they all mentioned a shift in each element.

Clients’ awareness of the cycle and their own role in it. A softening in behavior, affect, and cognition seemed to be both preceded and accompanied by an awareness of the cycle and the client’s role in it and a willingness to take responsibility for changing that role. In order to abandon their stance once they had softened, clients had to be aware of what that stance was and how it was contributing to the relationship problems. CBT~DV~CL~W states:

But now … the awareness level is so much greater that we can almost see it coming. And when you can see it coming and you’re aware that you’re headed in a direction that you don’t want to go it’s much easier to intercept . . . We have to step back and say, “Okay, if this is not a good time, let’s calm down and come back to it, table it, whatever.” It’s the awareness of it.

EFT~ST~CL~H also became aware of his role in the cycle in which he would pursue sexual intimacy and his wife would withdraw as a result of his pursuing, which would increase his pursuing and so on. He stated, “It concerns me that I was a good part of the problem . . .”

Softening of thoughts. As would be expected from the CBT model’s focus on the role of cognition in marital problems, Dr. Dattilio’s clients mentioned shifts in thinking more than the other clients. However, at least one client from each model mentioned that his or her thoughts had changed as a result of therapy. When a client’s thinking shifted, shifts in behavior and affect either co-occurred or soon followed. Cognitions were shifted using several interventions, including reframes (used in all models), direct didactic challenges of thoughts deemed irrational (used primarily in CBT), indirect experiential challenges (used primarily in EFT) and nondirec- tive, inductive processing (used primarily in IFS). Shifts in thought included attributing different intent to their partner’s actions, as evidenced by EFT~DV~CL~H’s mention that “the way [my] partner behaves is not necessarily how she wants to behave … so I give her the benefit of the doubt.” EFT~ST~CL~W mentions that her partner’s perception of her behavior has shifted: “He doesn’t raise the issues that we have hammered to death and sort of agreed to disagree on. Or if he does and I say, ? don’t feel strong enough to talk about this,’ he respects that and doesn’t see me as putting [him] off or being defensive.” This “softer” attribution to the intent behind their partner’s actions – the refusal to take their partner’s actions at face value – allowed them to treat each other more kindly, which shifted the interactional cycle.

Similarly, IFS~ST~CL mentioned that her thoughts changed as she repeated the questions she had heard her therapist repeat over and over in therapy (e.g., “Who is that? How old is that part? When did [you] first remember having this feeling?”). IFS~ST~CL mentioned that when she interacted with her husband, “I was able to actually [say], ‘Okay, this is a part of his,’ and actually sit back with curiosity and try to learn more about that part, rather than just respond rapid-fire back with one of my parts.” In other words, when an argument started with her husband she had explored alternative beliefs about the conflict enough to be able to exit the cycle on her own without becoming emotionally reactive.

Softened behavior. For each client, harsh, defensive behavior was replaced with softer, more loving behavior. This was one of the most pervasive findings, perhaps because the very notion of change implies that your behavior will be different in some way. Rather than simply behaving differently, this shift in behavior seemed to reflect a deeper, second-order shift in the way the clients felt about the relationship. Each partner abandoned his or her critical stance in the interactional cycle for a softer, more nurturing stance. Again, this change did not occur in isolation; it was accompanied by shifts in cognition and affect.

EFT~ST~CL~W states, “I think I am slower to jump to conclusions. I’m a better listener … I am able to . . . not become defensive and not react emotionally – I just listen.” CBT~DV~CL~H similarly states, “When I think she’s hostile or nasty now I don’t come back and blast her. [Instead] now I ask what’s wrong. And then she’ll tell me ‘Hey you didn’t have to say that or do that.’ Then I say, ‘Geez I didn’t even realize, I’m sorry.’ Then it seems to deflate the whole thing.”

Softened affect/emotional regulation. Another key way in which couples exited their negative interactional cycles was by shifts in the way they expressed emotion. Before therapy was successful couples tended to match harshness with harshness. Success began to be marked when soft expressions of emotions elicited the same from their partner. Couples were better able to regulate their emotions; they were less emotionally reactive, which made it easier for their spouse to be the same, replacing defensiveness with emotional accessibility.

Not surprisingly, this change was more readily visible in EFT clients, as EFT therapists made the focus on emotional processing their entry point into altering the cycle. EFT~DV~CL~H, for example, stated, “I think what was crucial was that [Dr. Johnson] was able to make us share our inner emotions. That shows [our] vulnerability, [so] then the other person takes a look back and maybe shares her vulnerability. That would work for us.”

CBT and IFS clients also mentioned an increased ability to regulate their negative emotions and express their softer emotions as a significant event in rebuilding their relationships. Using IFS “parts” language, IFS~DV~CL states:

. . . over time, I did start to see the difference between any part and self. . . . self has an energy and a calm about him that none of the parts have, because all of the parts . . . have & … prevailing emotion . . .

In nonmodel-specific terms, his process of finding his “self was accompanied by learning to regulate his emotions. The focus on finding his “self seemed to help him stand back from himself – to stand meta to himself in the cycle – and choose a calmer response to a situation that previously would have made him anxious.

The process of viewing anger as a cue to respond with love and concern is a hallmark of EFT, and is congruent with the goals of IFS and CBT. An EFT therapist accomplishes this goal by accessing and processing primary emotions. A CBT therapist helps clients regulate their emotions by focusing on altering cognitions associated with the affect. An IFS therapist does it by encouraging the calmer “self to respond to a partner’s emotionally reactive “part.” Since clients from all three models reached similar outcomes, perhaps which aspect a therapist chooses to focus on is more a matter of therapist preference and client fit as long as the therapist helps the client regulate his or her emotions.

Making Space for the Other

The second major category of outcome is making space for the other. Several clients emphasized the point that EFT~ST~CL~W did when asked what she learned in therapy about nurturing relationships. She simply said, “Give each other space.” Maturana (1992) described love as “opening space for the existence of another,” which occurred in each therapy to one degree or another.

Shifts in this area seemed to be more significant to clients than shifts in their interactional cycles. Shifts in their interactional cycles could be thought of as first-order change (Watzlawick, Weakland, & Fisch, 1974), or a change in the way their relationship “looked.” Shifts in the “making space for the other” category could be thought of as second-order change (Watzlawick et al., 1974), or shifts at a deeper level. In other words, clients did not only shift their outward behavior, but they also shifted who they were as a person in their relationships. The subcategories we describe below related to “making space for the other” are interdependent; change in one co-occurs with change in another.

Support of partner’s autonomy. Several clients (usually the husband) in this study began therapy as controlling. As therapy progressed and clients began to see their partners and themselves differently, they replaced their controlling stance with a more accommodating stance. They made space for their partners to grow according to their own desires. CBT~DV~CL~H discusses the change he experienced:

[I] was … a very jealous type of person that [was] insecure. I want[ed] to . . . keep her down or keep her with me. But I found that my wife needed to grow and needed to do several different things … to fulfill herself. I guess not only to say it’s ok to do; really it’s none of my business to say no anyway.

He mentioned how he decided to change. Dr. Dattilio had told him that “if he didn’t get with the program, she’s going to leave.” This was CBT~DV~CL~H’s response: “I was afraid of losing her. . . . It’s just time to [get with the program] and step outside [my]self and think of her.”

His wife acknowledged his change, mentioning that he was a lot more accepting of her feedback and her wishes.

IFS~ST~CL similarly stated, “We both stopped trying so hard to change the other person. Also … we both developed [a] greater acceptance [of each other].”

Confidence. As the controlling spouse eased up and the submissive spouse re-engaged, both partners seemed to enjoy an increased sense of self-confidence. The submissive spouse was able to freely say what was on his or her mind without fear of retribution from his or her partner. This seemed to lend legitimacy to the client’s opinions, which bolstered his or her self-confidence. CBT~DV~CL~W described the ripple effect of her increased confidence:

… the beauty of it is [the confidence is] not just with my husband. It’s with my friends, and I’m stronger in what I feel and what I’m thinking. I’m not afraid to voice [my opinions] and I don’t bottle things up and get annoyed as much. So the beauty of it is that not only is it great in the marriage but it’s helping me with other people. . . . I don’t worry as much now about whether or not people are going to like it. … I’m not as eager to please as I used to be and therefore I’m not as stressed out.

IFS~ST~CL described her increased self-confidence in parts language. She stated: “[I learned to] forgive [myself] and . . . instead of hating one of [my] parts [I] developed] an understanding of where . . . that part came from and [that my part is] really neat. She’s not as horrible as you think.” She noticed that as she relaxed and stopped trying to control her husband both she and her husband gained more confidence.

Slowing down. The process of making space for their partner involved both partners slowing down their process. Before entering therapy, most clients were trying to change their situation. The harder they tried to change, the worse things seemed to become. Consequently, each client’s efforts to make things change had a certain hurriedness or anxiety to it. Some clients were trying to force their partner to change, some were engaging in compulsive behaviors, and so on. At the end of therapy, all therapists mentioned that their clients had slowed down and relaxed in their efforts to change things outside of them. Using EFT terms, Dr. Johnson mentioned that her clients went from attacking their partners to “a more secure attachment than they had ever had in . . . their lives. They were able to look at their vulnerabilities [and] move into a place where they could tell each other their deepest needs and feelings and be responsive to each other.” Ms. O’Neil (IFS~ST) said that her client was “just more free to enjoy her life and not have so many worries that really weren’t very grounded.” Dr. Schwartz said he thinks his client is “much less anxious in general, less stressed . . .” Clients from each model mentioned similar changes.

The clients seemed to abandon their attacking stance for a more curious stance toward their partners. EFT~ST~CL~H, who had previously been controlling, mentioned that after therapy he was able to “listen … I mean really listen. [I] don’t sort of look off into space. [I] look into the person’s eyes, try to feel their feeling, understand why they’re saying what they’re saying.”

Personal responsibility. Making space for their partner involved clients taking personal responsibility for their stance in the relationship. Previously, they were caught up in blaming their partner for the relationship problems. As they grew in self-confidence and slowed down, they also adopted a greater responsibility for their experience in the relationship. They stopped demanding that their partner change and instead focused on how they could change, regardless of what their partner did.

IFS~ST~CL uses parts language to describe how she took personal responsibility and began to open up space for Mohammed, her husband, when I asked her how she is different with him now:

I could lighten up easier. … I could sit there patiently and listen instead of reacting.

Basically when his parts would come up, I could be more in self. I really worked at . . . being more centered and in self. And I could just kind of sit and listen and nod instead of jumping.

She went on to say that Mohammed began to open space for her once she did for him.

EFT~ST said that when the husband saw the effects of his controlling behavior on his wife, and his wife started to re-engage with her husband, he “[took] responsibility for his blaming everyone else, the doctors, the medication, [etc.].” CBT~DV~CL~H said that now when he is involved in an argument, he “[tries] to look back and think, ‘It obviously starts with you.’ [I try] to look at [myself] before [I] worry about her.”


Clinical Implications

This study has several clinical implications for couple therapy. First, this study highlights the importance of having a model for doing therapy. Therapy was helpful, we believe, largely because the client’s chaos was replaced with the therapist’s order (i.e., their model). The client came into therapy seemingly unable to organize all that was happening to him or her into anything he or she could use as a guide out of his or her difficulties. As Frank and Frank (1991) noted, clients coming into therapy are “conscious of having failed to meet their own expectations or those of others, or of being unable to cope with some pressing problem . . . [and] feel powerless to change the situation or themselves” (p. 35). We believe the therapists provided that order with their models. A model will likely be effective, we believe, if it (a) orients the therapist to credible aspects of dysfunction; (b) provides a clear definition of a healthy relationship; and (c) provides a clear operational map for how to help a client from dysfunction to health.

If a therapist does not have a model that provides a relevant definition of dysfunction and health and how to help clients from one to the other to guide conceptualization and intervention, he or she might only add to the confusion. Without a model to guide the therapist, the therapist will not know what to attempt to change, how to change it, or how to know when therapy is complete. Replacing one person’s chaos with another’s will not help. We acknowledge that some experiential therapists, such as Carl Whitaker (Napier & Whitaker, 1978), eschew models in favor of supporting spontaneity and direct experience that leads to a reenactment of family issues symbolically in the therapy room. This spontaneous experience, and the connections that come from it, result in awareness and existential shifts. However, we maintain that this rejection of a formal model and the resulting experiential interactions in the therapy room itself represent a coherent model of change.

Proposing that a model is necessary to therapy is in contrast to some proponents of common factors, who claim that, as all models seem to work the same, models are largely irrelevant (Duncan & Miller, 2000). Rather, our thinking (based on the present findings) is consistent with Sprenkle and Blow (2004b), who proposed that models are necessary as they are the vehicle, or framework, through which common factors operate. As long as the model fits the above-mentioned criteria, it may not matter which model is used, as they all reached similar ends. (At least that was the case in this study.)

Similarly, the model-specific interventions in this study were mentioned by the clients when they were asked what it was about therapy that helped them change. Clients described modelspecific things that their therapist did that helped them. Interestingly, however, all of the clients said that the model-specific interventions helped them in similar ways. This supports the systemic concept of equifinality (Bertalanffy, 1968), or the idea that the same end can be reached through several different means.

Though this study suggests that a therapist be familiar with credible models of therapy, there may be some wisdom in being flexible within those models if one’s current model is not working well with a client. That is, we believe that therapists, should not lose sight of the ends (i.e., the definition of health) as they focus on the means (i.e., the interventions), and should be flexible in applying the model’s interventions if they are not helping the clients reach desired ends.

Research Implications

This study has several implications for future couple therapy research. Comparative efficacy research is at the heart of the common factors debate (Sprenkle, 2002). Proponents of such research claim that it will ultimately show that one model is more effective than the others (American Psychological Association, 1993). Common factors researchers claim that such an approach “screams of scientific or theoretical arrogance” (Asay & Lambert, 1999, p. 23), and that what works about therapy is largely not found in the model per se. To date, meta-analytic reviews of the comparative efficacy literature have failed to show any significant differences among tested treatments (Wampold, 2001). Some common factors researchers cite these findings as evidence that comparative efficacy research should stop (Duncan & Miller, 2000). Others (Sprenkle & Blow, 2004a, 2004b) suggest that this research could continue, but should be expanded to include a study of variables broader than just model-specific factors (e.g., researchers could use the same study to focus on the relationship between therapist or client variables and outcome).

We believe that this study primarily supports the latter common factors claim. As clients in this study consistently claimed that aspects of the model did help them, models should not be ignored. Instead, successful therapy seems to be a combination of model-independent (e.g., therapist, client) and model-dependent (e.g., model-driven interventions) variables. Future comparative efficacy research could be useful, but only if it is expanded to include more independent variables than just the model-specific variables. Comparative efficacy research could also include measuring variables such as the client and therapist factors we identify in our companion article as well as in-session process (Butler & Wampler, 1999) and other variables as they relate to outcome.

Training Implications

Students’ learning of models should go hand in hand with their learning of couple function and dysfunction. This way, therapists would be able to form a picture of what they think constitutes a dysfunctional relationship, what constitutes a healthy relationship, and how they can help their clients move from one to the other (i.e., the model). Neglecting an understanding of health and dysfunction could leave a therapist with a quiver full of arrows, but no knowledge about what to hunt or when to stop hunting.

Furthermore, whether or not a model “works” or whether it fits empirical concepts of relational dysfunction and health are not the only criteria that should be used to judge the usefulness of a theory. Ethical and moral issues such as sensitivity to issues of diversity, and the propensity of the model to reinforce harmful stereotypes should also be taken into account when evaluating the usefulness of a model. For example, does a model encourage the therapist to value the experiences of each family member equally? Is the model flexible enough to find use with diverse cultures?

Once therapists understand concepts of relational dysfunction and health, it may be helpful to train therapists in general techniques that help clients go from one to the other. For example, learning relaxation or mindfulness techniques to help clients regulate their emotions is a technique easily adapted to most models.

Relevance of Findings with MFT Models Not Represented in This Study

Though the themes in this study were derived from three distinct models of MFT, evidence of similar themes exists in other MFT models. For example, several MFT theories other than those in this study emphasize family of origin experiences to conceptualize cases (e.g., Boszormenyi-Nagy, 1987; Kerr & Bowen, 1988; Scharff& Scharff, 1987). And certainly others, such as structural therapy (Minuchin & Fishman, 1981), focus to one degree or another on transgenerational issues when working with clients.

The notion of interactional cycles is as old as family therapy itself (Watzlawick, Beavin, & Jackson, 1967). Practically every MFT model incorporates some aspect of interactional cycles into their conceptualization and intervention. For example, brief therapy approaches such as strategic (Haley, 1976), solution-focused (deShazer, 1988), and Milan systemic (Palazzoli, Boscolo, Cecchin, & Prata, 1978) theories, as well as others, rely heavily on viewing one partner’s behavior in the context of that of the other partner.

Raising awareness of the cycle through slowing down the process, standing meta to oneself and one’s partner, and encouraging personal responsibility are also common across MFT theories not included in this study. For example, narrative externalization of the problem (White & Espton, 1990) serves to help clients slow down their internal and interpersonal processes by helping them battle the problem rather than themselves or each other. The solution-focused therapist (deShazer, 1988) helps clients slow down by helping them search for exceptions to their current problem and encouraging them to re-create those exceptions. From a Bowenian perspective (Kerr & Bowen, 1988), differentiation could be seen as slowing down, and standing meta to one’s self in order to take personal responsibility in relationships by becoming more proactive than reactive to one’s emotions.

And depending on one’s perspective, it is not difficult to see emotional, cognitive, and behavioral components in such diverse therapies as narrative (White & Espton, 1990), Bowen (Kerr & Bowen, 1988), strategic (Haley, 1976), systemic (Palazzoli et al., 1978), and even experiential therapies (Napier & Whitaker, 1978).


In this article, the first of two companion articles, we outlined the model-dependent results of a qualitative study aimed at discovering common factors of change across different MFT models. Model-dependent variables are those variables directly informed by the therapist’s model. Across three different models, we discussed common conceptualizations, interventions, and outcomes, as well as clinical, research, and training implications of our findings. In the second article, we will discuss the model-independent variables we found, as well as how model-dependent and model-independent variables appear to work together to produce change.


1 Readers may contact the first author for more detailed quotes relating to each category if they wish.


American Psychological Association. (1993). Task force report on promotion and dissemination of psychological practices. Washington, DC: Author.

Asay, T. P., & Lambert, M. J. (1999). The empirical case for the common factors in therapy: Quantitative findings. In M. A. Hubble, B. L. Duncan, & S. D. Miller (Eds.), The heart and soul of change: What works in therapy (pp. 23-55). Washington, DC: American Psychological Association.

Beck, A. T. (1976). Cognitive therapy and the emotional disorders. New York: New American Library.

Bertalanfly, L. (1968). General system theory: Foundation, development, applications. London: Allen Lane.

Blow, A. J., & Sprenkle, D. H. (2001). Common factors across theories of marriage and family therapy: A modified Delphi study. Journal of Marital and Family Therapy, 27, 385-401.

Boszormenyi-Nagy, I. (1987). Foundations of contextual therapy: Collected papers of Ivan Boszormenyi-Nagy. New York: Brunner/Mazel.

Bowlby, J. (1988). A secure base. New York: Basic Books.

Breunlin, D. C, Schwartz, R. C, & Mac Kune-Karrer, B. (2001). Metaframeworks: Transcending the models of therapy. San Francisco: Jossey-Bass.

Butler, M. H., & Wampler, K. S. (1999). Couple-responsible therapy process: Positive proximal outcomes. Family Process, 38, 27-54.

Crabtree, B. F. (1999). Doing qualitative research. Thousand Oaks, CA: Sage.

Dattilio, F. M. (2001). Cognitive-behavior family therapy: Contemporary myths and misconceptions. Contemporary Family Therapy, 23, 3-18.

Dattilio, F. M. (2002). Homework assignments in couple and family therapy. Psychotherapy in Practice, 58, 535-547.

Dattilio, F. M., & Epstein, N. B. (2003). Cognitive-behavior couple and family therapy. In T. L. Sexton, G. R. Weeks, & M. S. Robbins (Eds.), Handbook of family therapy (pp. 147-173). New York: Brunner-Routledge.

Davis, S. D., & Butler, M. H. (2004). Enacting relationships in marriage and family therapy: A conceptual and operational definition of an enactment. Journal of Marital and Family Therapy, 30, 319-333.

deShazer, S. (1988). Clues: Investigating solutions in brief therapy. New York: Norton.

Duncan, B. L., & Miller, S. D. (2000). The heroic client: Doing client directed, outcome-informed therapy. San Francisco: Jossey-Bass.

Duncan, B. L., Miller, S. D., & Sparks, J. A. (2003). Interactional and solution-focused brief therapies: Evolving concepts of change. In T. L. Sexton, G. R. Weeks, & M. S. Robbins (Eds.), Handbook of family therapy: The science and practice of working with families and couples (pp. 101-124). New York: Brunner-Routledge.

Ellis, A. (1962). Reason and emotion in psychotherapy. Sacramento, CA: Citadel Press.

Erikson, F. (1986). Qualitative methods in research on teaching. In M. C. Wittrock (Ed.), Handbook of research on teaching (3rd ed., pp. 119-161). New York: MacMillan.

Frank, J. D., & Frank, J. B. (1991). Persuasion and healing: A comparative study of psychotherapy (3rd ed.). Baltimore: Johns Hopkins University Press.

Glaser, B. G. (1993). Examples of grounded theory: A reader. Mill Valley, CA: Sociology Press.

Glaser, B. G. (2001). Doing grounded theory. Grounded Theory Review, 2, 1-18.

Glaser, B. G., & Strauss, A. (1967). The discovery of grounded theory. Chicago: Aldine.

Goulding, R. A., & Schwartz, R. C. (1995). Mosaic mind: Empowering the tormented selves of child abuse survivors. New York: Norton.

Greenberg, L. S., & Johnson, S. M. (1988). Emotionally focused therapy for couples. New York: Guilford Press.

Haley, J. (1976). Problem-solving therapy. San Francisco: Jossey-Bass.

Hubble, M. A., Duncan, B. L., & Miller, S. (Eds.). (1999). The heart and soul of change: What works in therapy. Washington, DC: American Psychological Association.

Jacobson, N. S., & Addis, M. E. (1993). Research on couple therapy: What do we know? Where are we going? Journal of Consulting and Clinical Psychology, 61, 85-93.

Johnson, S. M. (2004). The practice of emotionally focused marital therapy: Creating connection (2nd ed.). New York: Brunner/Mazel.

Kerr, M. E., & Bowen, M. (1988). Family evaluation: An approach to Bowen theory. New York: W. W. Norton.

Lambert, M. J. (1992). Psychotherapy outcome research: Implications for integrative and eclectic therapists. In J. C. Norcross & M. R. Goldfried (Eds.), Handbook of psychotherapy integration (pp. 94-129). New York: Basic Books.

Maturana, H. R. (1992). Biology, emotions and culture [Videotape 6]. Calgary, Canada: Vanry & Associates.

Miller, S. D., Duncan, B. L., & Hubble, M. A. (1997). Escape from Babel: Toward a unifying language for psychotherapy practice. New York: Norton.

Minuchin, S., & Fishman, H. C. (1981). Family therapy techniques. Cambridge, MA: Harvard University Press.

Napier, A. Y., & Whitaker, C. A. (1978). The family crucible. New York: Harper & Row.

Nichols, M. P., & Schwartz, R. C. (2001). Family therapy: Concepts and methods (5th ed.). Boston: Allyn & Bacon.

Palazzoli, M. S., Boscolo, L., Cecchin, G, & Prata, G. (1978). Paradox and counterparadox: A new model in the therapy of the family in schizophrenic transaction. New York: Jason Aronson.

Patton, M. Q. (2002). Qualitative research and evaluation methods (3rd ed.). Thousand Oaks, CA: Sage.

Pavlov, I. P. (1932). Neuroses in man and animals. Journal of the American Medical Association, 99, 1012-1013.

Scharff, D., & Scharff, J. (1987). Object relations family therapy. New York: Jason Aronson.

Schwartz, R. C. (1995). Internal family systems therapy. New York: Guilford Press.

Sexton, T. L., & Ridley, C. R. (2004). Implications of a moderated common factors approach: Does it move the field forward? Journal of Marital and Family Therapy, 30, 159-164.

Sexton, T. L., Ridley, C. R., & Kleiner, A. J. (2004). Beyond common factors: Multilevel-process models of therapeutic change in marriage and family therapy. Journal of Marital and Family Therapy, 30, 131-150.

Shadish, W. R., & Baldwin, S. A. (2002). Meta-analysis of MFT interventions. In D. H. Sprenkle (Ed.), Effectiveness research in marriage and family therapy (pp. 339-370). Alexandria, VA: American Association of Marriage and Family Therapy.

Skinner, B. F. (1953). Science and human behavior. New York: Macmillan.

Sprenkle, D. H. (Ed.). (2002). Editor’s introduction. In Effectiveness research in marriage and family therapy (pp. 9-26). Alexandria, VA: American Association of Marriage and Family Therapy.

Sprenkle, D. H., & Blow, A. J. (2004a). Common factors and our sacred models. Journal of Marital and Family Therapy, 30, 113-130.

Sprenkle, D. H., & Blow, A. J. (2004b). Common factors are not islands – they work through models: A response to Sexton, Ridley, and Kleiner. Journal of Marital and Family Therapy, 30, 151-158.

Sprenkle, D. H., Blow, A. J., & Dickey, M. H. (1999). Common factors and other non-technique variables in marriage and family therapy. In M. A. Hubble, B. L. Duncan, & S. D. Miller (Eds.), The heart and soul of change: What works in therapy (pp. 329-359). Washington, DC: American Psychological Association.

Strauss, A. L., & Corbin, J. (1998). Basics of qualitative research: Techniques and procedures for developing grounded theory (2nd ed.). Thousand Oaks, CA: Sage.

Tallman, K., & Bohart, A. C. (1999). The client as a common factor: Clients as self-healers. In M. A. Hubble, B. L. Duncan, & S. D. Miller (Eds.), The heart and soul of change: What works in therapy (pp. 91-132). Washington, DC: American Psychological Association.

Wampler, K. (1997). Systems theory and outpatient mental health treatment: Twelve priorities for MFT research. Paper presented at the Inaugural AAMFT Research Conference, Santa Fe, NM.

Wampold, B. E. (2001). The great psychotherapy debate: Models, methods, and findings. Mahwah, NJ: Erlbaum.

Watzlawick, P., Beavin, J. H., & Jackson, D. D. (1967). Pragmatics of human communication: A study of interactional patterns, pathologies, and paradoxes. New York: Norton.

Watzlawick, P., Weakland, J., & Fisch, R. (1974). Change: Principles of problem formation and problem resolution. New York: Norton.

White, M., & Espton, D. (1990). Narrative means to therapeutic ends. New York: Norton.

Sean D. Davis

Alliant International University-Sacramento Campus

Fred P. Piercy

Virginia Polytechnic Institute and State University

Sean D. Davis, PhD, Marital and Family Therapy Program, Alliant International University-Sacramento Campus; Fred P. Piercy, PhD, Department of Human Development, Virginia Polytechnic Institute and State University.

The authors wish to thank Drs. Frank M. Dattilio, Susan M. Johnson, Richard C. Schwartz, Judy Makinen, Ms. Beth O’Neil, MSW, and their clients for their participation in this study. We also thank Drs. Douglas H. Sprenkle, Scott W. Johnson, Lenore M. McWey, Anna Beth Benningfield, and Margaret L. Keeling for their support throughout various stages of this project.

This project was completed while Sean Davis was a doctoral student at Virginia Tech. Portions of this project were presented at the 2005 annual meeting of the American Association of Marriage and Family Therapy, Kansas City, Missouri. This study was supported with monies from the 2005 AAMFT Graduate Student Research Award and the 2006 AAMFT Dissertation Award.

Address correspondence to Dr. Sean Davis, Marital and Family Therapy Program, Alliant International University, 425 University Avenue, Suite 201, Sacramento, California 95825; E-mail: sdavis2(nalliant.edu

Copyright Blackwell Publishing Jul 2007

Provided by ProQuest Information and Learning Company. All rights Reserved

You May Also Like





Practice Magic

Practice Magic Smith, Craig W Practice Magic. (1995). (Practice management software for billing, check depositing, and calendaring.)…

Parents’ and children’s perceptions of family processes in inner-city families with delinquent youths: A qualitative investigation

Parents’ and children’s perceptions of family processes in inner-city families with delinquent youths: A qualitative investigation Madden-…