Show Summary Details
Page of

(p. 245) Empathy 

(p. 245) Empathy
Chapter:
(p. 245) Empathy
Author(s):

Robert Elliott

, Arthur C. Bohart

, Jeanne C. Watson

, and David Murphy

DOI:
10.1093/med-psych/9780190843953.003.0007
Page of

date: 21 July 2019

Empathy has a long and sometimes stormy history in psychotherapy. Proposed and codified by Rogers and his followers in the 1940s and 1950s, it was initially widely portrayed as primarily a therapist trait and put forward as the foundation of helping skills training programs popularized in the 1960s and early 1970s. Claims concerning its trait-like status and universal effectiveness led to skepticism and then came under intense scrutiny by psychotherapy researchers in the late 1970s and early 1980s. After that, research on empathy went into relative eclipse, resulting in a dearth of research between 1981 and 2000.

Since the late-1990s, however, empathy has again become a topic of scientific interest in clinical, developmental, and social psychology (e.g., Bohart & Greenberg, 1997; Ickes 1997), particularly because empathy, now reconceptualized as an interactional variable, has come to be seen as a key element of the new field of social neuroscience (e.g., Decety & Ickes, 2009). This development has helped relegitimize empathy as a central element of psychotherapy, which has led to an explosion of empathy research in the past 20 years. In fact, interest in empathy has recently rippled into related disciplines such as medicine, where it is now an active topic of investigation in a wide range of medical interventions (from anesthesiology to acupuncture) using a quite diverse array of measures (Pedersen, 2009).

In this chapter, we begin by reviewing definitions and measures of therapist empathy, including the conceptual problem of separating empathy from other relationship variables. We follow this with clinical examples illustrating different forms of therapist empathy and empathic response mode. We move on to describe a range of landmark studies of therapist empathy, including both quantitative and qualitative investigations, followed by a brief review of previous systematic reviews of therapist empathy. The core of our review, however, is a meta-analysis of 82 studies of the relation between therapist empathy and client outcome. In addition to reporting overall effect sizes, we examine a range of moderator variables, such as theoretical orientation and measurement perspectives. After looking at possible mediator variables, we offer a framework (p. 246) for assessing the causal status of the empathy–outcome connection and use this framework to review the evidence. Finally, we explore the role of client contributions and the limitations of the existing evidence base, touching on diversity issues, before moving on to the implications for training and therapeutic practice.

Definitions

The first problem with researching empathy in psychotherapy is that there is no consensual definition (Batson, 2009; Bohart & Greenberg, 1997; Duan & Hill, 1996; Pedersen, 2009). The problem is compounded when trying to talk about empathy across different fields such as nursing, education, medicine, and psychology. For instance, Bloom (2016) recently wrote a book titled Against Empathy, in which he argues that empathy can be destructive in personal relationships, as well as in making moral decisions on a societal level. However, this argument is based entirely on Bloom’s definition of empathy, which emphasizes sentimental emotional identification with another.

We begin by synthesizing a range of contemporary dictionary definitions in order to provide a useful working definition:

  1. 1. Empathy is interpersonal and unidirectional, provided by one person to another person.

  2. 2. Empathy is conceptualized primarily as an ability or capacity and only occasionally as an action.

  3. 3. Empathy involves a range of related mental abilities/actions, including

    1. a. Primarily: Understanding the other person’s feelings, perspectives, experiences, or motivations

    2. b. But also: Awareness of, appreciation of, or sensitivity to the other person

    3. c. Achieved via: Active entry into the other’s experience, described variously in terms of vicariousness, imagination, sharing, or identification.

Several features of this definition can be criticized, for example, that it portrays empathy in outmoded trait-like terms, that it ignores the role of the recipient, that it is too broad, and that it involves a mysterious or potentially misleading process of identification (cf. Bloom, 2016).

Recent neuroscience research on empathy begins to clarify the conceptual confusion. Researchers have made concerted efforts to use a variety of methods ranging from performance tasks, self-report, and neuropsychological assessment to functional magnetic resonance imaging and transcranial stimulation. Research examining the brain correlates of subprocesses of empathy (Decety & Ickes, 2009) extended the initial discovery of “mirror neurons” in the motor cortex of macaque monkeys (e.g., Gallese et al., 1996) to a broader range of affective and perspective-taking components of empathy in humans (Decety & Lamm, 2009). The result of this research has been to deepen and clarify our understanding of therapist empathic processes (Watson & Greenberg, 2009).

(p. 247) The current general view (e.g., Eisenberg & Eggum, 2009) is that empathy can be roughly separated into three major subprocesses, each with specific neuroanatomical correlates. First, there is a more or less automatic, intuitive emotional simulation process that mirrors the emotional elements of the other’s bodily experience with brain activation centering in the limbic system (amygdala, insula, anterior cingulate cortex) and elsewhere (Decety & Lamm, 2009; Goubert et al., 2009). Second, a more deliberate, conceptual, perspective-taking process operates, particularly localized in medial and ventromedial areas of prefrontal cortex and the temporal cortex (Shamay-Tsoory, 2009). Third, there is an emotion-regulation process that people use to reappraise or soothe their personal distress when vicariously experiencing the other person’s pain or discomfort, allowing them to mobilize compassion and helping behavior for the other (probably based in the orbitofrontal cortex, as well as in the prefrontal and right inferior parietal cortex; Decety & Lamm, 2009; Eisenberg & Eggum, 2009). Both psychotherapy and neuroscience research have repeatedly found that conceptual and emotional elements of empathy do not correlate highly (Duan & Kivlighan, 2005, Hein & Singer, 2010), although in skilled therapists they are likely to coordinate seamlessly and holistically.

Interestingly, the two therapeutic approaches that have most focused on empathy—client-centered therapy and psychoanalytic—have emphasized its cognitive or perspective-taking (Selman, 1980) aspects. That is, they have focused on empathy as connected knowing (Belenky et al., 1986), understanding the client’s frame of reference or way of experiencing the world. In fact, Carl Rogers talked about empathic understanding, not empathy (Shlien, 1997). By some accounts, 70% or more of Rogers’ responses were to felt meaning rather than to feeling, despite the fact that his mode of responding is typically called “reflection of feeling” (Brodley & Brody, 1990; Hayes & Goldfried, 1996; Tausch, 1988). However, understanding clients’ frames of reference does include understanding their affective experiences. In addition, empathy and sympathy have typically been sharply differentiated, with therapists such as Rogers disdaining sympathy but prizing empathy (Shlien, 1997). In affective neuroscience terms, this means that therapists in these traditions have often emphasized conscious perspective-taking processes over the more automatic, bodily based emotional simulation processes.

Nevertheless, it is easy to see both processes in Rogers’s (1980) definition of empathy:

the therapist’s sensitive ability and willingness to understand the client’s thoughts, feelings and struggles from the client’s point of view. [It is] this ability to see completely through the client’s eyes, to adopt his frame of reference. [p. 85] . . . It means entering the private perceptual world of the other . . . being sensitive, moment by moment, to the changing felt meanings which flow in this other person. . . . It means sensing meanings of which he or she is scarcely aware. (p. 142)

Defined this way, empathy is a higher order category, under which different subtypes, aspects, and modes can be nested. There are different ways one can put oneself into the shoes of the other: emotionally, cognitively, on a moment-to-moment basis, or (p. 248) by trying to grasp an overall sense of what it is like to be that person. Within these subtypes, different aspects of the client’s experience can become the focus of empathy (Bohart & Greenberg, 1997). Similarly, there are many ways of expressing empathy, including empathic reflections, empathic questions, experience-near interpretations, empathic conjectures, responsive use of other therapeutic procedures, and a wide range of responsive and carefully tuned nonverbal expressions. Accordingly, empathy is best understood as a complex construct consisting of different acts used in multiple ways.

We distinguish between three main modes of therapeutic empathy: empathic rapport, communicative attunement, and person empathy. First, for some therapists empathy is the establishment of empathic rapport and support; this is the definition favored in cognitive-behavioral therapy (CBT). The therapist exhibits a benevolent compassionate attitude toward the client and tries to demonstrate that he or she understands the client’s experience, often to set the context for effective treatment. A second mode of empathy consists of an active, ongoing effort to stay attuned on a moment-to-moment basis with the client’s communications and unfolding experience. Humanistic and person-centered experiential therapists are most likely to emphasize this form of empathy. The therapist’s communicative attunement may be expressed in many ways but most likely in empathic responses. The third mode, person empathy (Elliott et al., 2003) or experience-near understanding of the client’s world, consists of a sustained effort to understand the kinds of experiences the client has had, both historically and presently, that form the background of the client’s current experiencing. The question is: How have the client’s experiences led him or her to see/feel/think/act as he or she does? This is the type of empathy emphasized by psychodynamic therapists. Of course, empathic rapport, communicative attunement, and person empathy are not mutually exclusive, and their differences are a matter of emphasis.

Many other definitions for empathy have been advanced: as a trait or response skill (Egan, 1982; Truax & Carkhuff, 1967), as an identification process of “becoming” the experience of the client (Mahrer, 1997), and as a hermeneutic interpretive process (Watson, 2001). Many definitions, particularly in the fields of social, developmental, and personality psychology, explicitly include elements of caring, concern, and compassion.

Perhaps the most practical conception, and one that we draw on in our meta-analysis, is Barrett-Lennard’s (1981) operational definition of empathy in terms of three different perspectives: (a) the therapist’s empathic resonance with the client, (b) the observer’s perception of the therapist’s expressed empathy, and (c) the client’s experience of received therapist empathy.

Measures

Reflecting the complex, multidimensional nature of empathy, a confusing welter of measures has been developed across disciplines. During the initial stages of our literature search, many of the abstracts we encountered were of studies on empathy in medicine, nursing, and social and developmental psychology. One of us counted 17 different measures of empathy within the first 100 abstracts reviewed. Following (p. 249) Barrett-Lennard (1981), most measures of therapist empathy in psychotherapy research fall into three categories: expressed empathy rated by nonparticipant raters, client-rated received empathy, and therapist-rated own empathic resonance. To these can be added a fourth category: empathic accuracy, defined as congruence between therapist and client perceptions of the client (Ickes, 1997; e.g., Duan & Hill, 1996).

In our meta-analysis we have utilized measures of empathy that went beyond rating the mere presence of supposedly empathic therapist response modes such as reflection or paraphrases of the client’s words. There is a literature correlating frequency of reflections with outcome, with disappointing results (Orlinsky et al., 2003; Orlinsky & Howard, 1986). Instead, we looked for measures that assessed the quality of therapist empathy. Thus we have focused on studies in which therapists rated their own empathic quality, clients rated their perceptions of how empathic the therapist was, or raters rated the quality of therapists’ empathy or studies that attempted to assess empathic accuracy. We note that the scales used to rate the fidelity of motivational interviewing (Moyers et al., 2015) make the same distinction: empathy is rated independently of the number of reflections provided by the therapist.

Observer Ratings

Some of the earliest observer measures of empathy were those of Truax and Carkhuff (1967) and Carkhuff and Berenson (1967). The earliest of these scales asked raters to decide if the content of the therapist’s response detracted from the client’s response, was interchangeable with it, or added to it; more recent measures focus more globally on activeness, consistency, and depth of empathy. Typically, trained raters listened to 2- to 15-minute samples from session tapes. Samples are usually drawn from the beginning, middle, and/or the end of therapy. Scales such as these do not fully reflect the client-centered conception of empathy as an attitude because they focus narrowly on a particular kind of response, often empathic reflections. Furthermore, the equation of a particular response with empathy has also made these scales less appropriate for measuring empathy in approaches other than client-centered (Lambert et al., 1978). More recent observer empathy measures reflect broader understandings of forms of empathic responding. Watson and Prosser (2002) developed an observer-rated measure of empathy that assesses therapists’ verbal and nonverbal behavior and shows convergence validity with client ratings on the Barrett-Lennard Relationship Inventory (BLRI).

In addition, the therapist’s general empathy can be rated by others who know or have supervised the psychotherapist. For instance, therapists’ empathic capacities can be rated by their supervisors (Gelso et al., 2002). For purposes of our meta-analysis, we grouped together all measures that used external observers.

Client Ratings

The most widely used client-rated measure of empathy is the empathy scale of the BLRI: Other to Self version. Alternative client rating measures have been developed (e.g., Hamilton, 2000; Lorr, 1965; Saunders et al., 1989) as well. Rogers (1957) (p. 250) hypothesized that clients’ perceptions of therapists’ facilitative conditions (positive regard, empathy, and congruence) predict therapeutic outcome. Accordingly, the BLRI, which measures clients’ perceptions, is an operational definition of Rogers’ hypothesis. In several earlier reviews, including our meta-analysis in the previous editions of this book, client-perceived empathy predicted outcome better than observer- or therapist-rated empathy (Barrett-Lennard, 1981; Elliott et al., 2011; Gross & DeRidder, 1966; Gurman, 1977; Orlinsky et al., 2003; Orlinsky & Howard, 1978, 1986).

Therapist Ratings

Therapist empathy self-rating scales are not as common, but perhaps the most widely used is the BLRI: Myself to Other version. Earlier reviews (Barrett-Lennard, 1981; Gurman, 1977) found that therapist-rated empathy neither predicted outcome nor correlated with client-rated or observer-rated empathy. However, we previously found that therapist-rated empathy did predict outcome but at a lower level than client or observer ratings (Elliott et al., 2011).

Empathic Accuracy

A fourth type of empathy measure uses measures of therapist–client perceptual congruence, commonly referred to as “empathic accuracy” (Ickes, 1997). These typically consist of therapists rating or describing clients as they think the clients would see themselves on personality scales, symptom lists, or within-session experiences and then comparing these to how clients actually rated or described themselves. For instance, one study compared how therapists rated clients on Kelly’s Role Repertory Test grid with how clients rated themselves (Landfield, 1971). The measure of empathy is the degree of congruence between therapist and client ratings. This can also be referred to as predictive empathy, because the therapist is trying to predict how clients will rate themselves. This is closer to a measure of the therapist’s ability to form a global understanding of what it is like to be the client (person empathy) than it is to a process measure of ongoing communicative attunement.

Recent work on empathic accuracy provides a predictive measure of accuracy of communicative attunement (Ickes, 1997, 2003). This line of research typically employs a tape-assisted recall procedure in which therapists’ or observers’ moment-to-moment empathy is measured by comparing their perceptions of client experiences to clients’ reports of those experiences. For example, researchers (Kwon & Jo, 2012) asked clients to listen to tapes of a session and report on what they were thinking and feeling at various times. Therapists independently listened to the tapes and tried to predict what the clients were thinking and feeling. The degree to which therapist and client agreed was used as a measure of empathic accuracy. Empathic accuracy correlated with counseling outcome .63 (p < .001).

(p. 251) Correlations among Empathy Measures

Intercorrelations of empathy measures have generally been modest. Low correlations have been reported between cognitive and affective measures (Gladstein et al., 1987) and between predictive measures and the Barrett-Lennard Relationship Inventory (Kurtz & Grummon, 1972). Other research has found that tape-rated measures correlate only moderately with client-perceived empathy (Gurman, 1977).

These low positive correlations are not surprising when one considers what the different instruments are supposed to be measuring. Trying to predict how a client will fill out a symptom checklist turns out to be quite different from responding sensitively in the session, demonstrating subtle understanding of what the client is communicating, while checking and adjusting one’s emerging understanding with that of the client. Similarly, client ratings of therapist understanding may be based on many other things than the therapist’s particular skill in empathic reflection. Accordingly, we should not expect different measures of this complex construct to correlate (Gladstein et al., 1987) and could even hypothesize on this basis that using more empathy measure in a study will “dilute” the average association between empathy and outcome. However, Watson and Prosser’s (2002) observer measure of empathy did correlate with the BLRI: Other to Self client measure at .66 (p < .01).

Confounding between Empathy and Other Relationship Variables

A related concern is the distinctiveness of empathy from other facets of the therapeutic relationship. One early review of more than 20 studies primarily using the BLRI found that, on average, empathy correlated .62 with congruence and .53 with positive regard (Gurman, 1977). Factor analysis of scale-level scores found that one global factor typically emerged, with empathy loading on it along with congruence and positive regard (Gurman, 1977). Others have reported that the empathy scale loaded .93 on a global BLRI factor, with Positive Regard loading .87 and Congruence loading .92 (Blatt et al., 1996). More recently, empathy and the other relationship conditions loaded on a global BLRI factor that accounted for 69% of the variance (Watson & Geller, 2005). Such results suggest that clients’ perceptions of empathy are not clearly differentiated from their perceptions of other relationship factors.

On the other hand, empathy did emerge as a separate dimension in reviews of factor analytic studies where specific items were analyzed rather than scale scores (Gurman, 1977). In addition, empathy tends to correlate more highly with the bond component of the therapeutic alliance than with the task and goal components (Horvath & Greenberg, 1986). Thus there is evidence both for and against the hypothesis that the Rogerian triad of empathy, unconditional positive regard, and congruence are distinct variables.

In this regard, we found that many measures of empathy create conceptual confusion by including aspects of both empathy and positive regard. For example, a well-known empathy scale (Burns & Nolen-Hoeksema, 1992), included in our previous (p. 252) meta-analyses, has more items on it dealing with positive relationship qualities in general than it does specific empathy items. We decided in this meta-analysis to include only studies where the “empathy” scale included at least 50% of items we could clearly identify as empathy. Accordingly, we have excluded the Burns and Nolen-Hoeksema instrument from this meta-analysis.

Finally, there is both conceptual and measurement overlap between empathy and other recent relationship constructs such as compassion (e.g., Strauss et al. 2016), presence (Geller et al., 2010), and responsiveness (Elkin et al., 2014). Empathy is typically included as a component in both conceptual and operational definitions of these constructs. Furthermore, in a study of therapists’ experience of empathy, compassion was a major component of empathy (Greenberg & Rushanski-Rosenberg, 2002).

In conclusion, empathy is at least partially distinct conceptually from other relationship constructs. However, operationally, things are less clear, with overlap between it and other constructs such as positive regard, congruence, compassion, presence, and responsiveness. In practice, separating empathy out from other relational qualities is a reductionistic fiction, treating relationships as if they were the constituent elements of chemical compounds. Ultimately, we think that it is more useful to treat empathy (and other relationship constructs) as components of a higher order therapeutic relationship.

Clinical Examples

In this section we provide clinical examples of some different types of empathy and specific empathic response modes used to promote in-session client processes and taught in contemporary empathy training (e.g., Elliott et al., 2003; Johnson et al., 2005). We use a running case example to illustrate these.

“Rick” was a 30-year-old unmarried man from a family of unsympathetic high achievers; he had been struggling since his early 20s to break into the movie business. He presented saying that he was anxious and worried much of the time, and at his first appointment he was clearly agitated. At the beginning of treatment, Rick’s therapist focused on building rapport and trust using empathic understanding responses. Empathic reflection responses convey understanding of clients’ experiences while empathic affirmations are attempts by the therapist to validate the client’s perspective. For example:

c1: I’m really in a panic (anxious and looking plaintively at the therapist). I feel anxious all the time. Sometimes it seems so bad, I really worry that I’m on the verge of a psychotic break. I’m afraid I’ll completely fall apart. Nothing like this has ever happened to me before. I always felt in charge of myself, but now I can’t seem to get any control over myself at all.

t1: So feeling really, really anxious as if you might break down [empathic reflection]—it is just so hard to control and manage it [empathic affirmation].

(p. 253) c2: Yes! I don’t know myself anymore. I feel so lost. The anxiety’s like a big cloud that just takes over, and I can’t even find myself in it anymore. I don’t even know what I want, what to trust. . . . I’m so lost.

t2: So you feel so lost, like you don’t even know yourself or what you want and need. No wonder you feel lost if it takes over like that. Anxiety can do that, ambushing us and taking over [empathic reflection and empathic affirmation]

c3: (Client tearing up:) Yes, I do feel ambushed and confused (sadly and thoughtfully).

The therapist’s empathic recognition provided the client with a sense of being understood, building rapport and fostering a sense of safety that gradually helped the client move from agitation into reflective sadness. To facilitate this the therapist began using more exploratory reflections, which attempted to get at that which was implicit in the client’s narratives and help him focus on information that had been in the background but not yet fully articulated, including emerging client experiences. For example:

t 3: And I hear that this leaves you feeling, sort of, sad?

c4: Yes, this is such a familiar feeling. . . . I always felt lost as a kid. Everyone was always so busy—there was no place for me. My siblings were focused on their sports and academic achievements. I was the youngest so I was expected to tag along to their activities even though I hated it. It was so boring!

t4: It sounds almost as if you felt like the odd one out in your family, like you didn’t quite fit in somehow? [exploratory reflection]

c5: Yes, very much so. There was so much going on. Mum was always busy with her activities or driving my siblings somewhere. I used to escape with my books and my music.

To further amplify the client’s experience the therapist next uses evocative reflections, attempts to bring the client’s experience alive in the session using rich, evocative, concrete, connotative language, often with a probing, tentative quality. For example:

t5: So you felt forgotten somehow? I have an image of you as a little boy sitting alone in a corner curled up with your book as the people around you rushed to and fro?

c6: Yes, I used to hide away and try to disappear. (Client’s voice breaks)

The therapist here is trying to communicatively attune in the moment to the client’s experience and feel her way into what it must have felt like to be a child in that busy, high-achieving household. The therapist continues to facilitate the exploration of the client’s inner experience using process reflections and empathic conjectures, which go beyond exploratory reflections in that they are attempts to guess or infer what clients might be feeling but have not said out loud yet. They are presented tentatively as hunches grounded in what the client has shared. For example, while watching Rick the therapist noticed that her client’s voice shifted and that he looked very sad.

(p. 254)

t6: I noticed your voice changed just then [process reflection]. You look very sad; are you? [empathic conjecture] What is happening inside as you recall the busy household? [exploratory question]

c7: I feel like I can’t live up to their expectations. Even though I know I’ve got all this potential, I always feel there is something wrong with me.

t7: So you feel like a failure? [empathic conjecture] Like there is something wrong with you? You are not quite right? [exploratory reflection]

At other times therapists can use empathic refocusing responses to offer clients the opportunity to return to an experience they had distanced themselves from. For example:

c8: Yes, that’s how it feels when I go home and then it’s worse as they pester me about not being married and poke fun at me about my impractical career goals and marginal job waiting table.

t8: So when you’re at home, I guess it’s just so difficult, but a minute ago I also heard something in you that hangs onto your ambitions and sees yourself as talented and having a lot to offer, is that right? [empathic refocusing response, referring back to C7, delivered as an empathic conjecture]

c9: Yes, that’s in there somewhere also, but then it just dissolves.

These examples demonstrate how therapists work to remain communicatively attuned to their clients on a moment to moment basis in the session. And as clients continue to share and explore their experiences, therapists begin to develop a sense of person empathy, providing a more holistic understanding of their clients’ experiences. This more holistic understanding provides some guidance in terms of attending to markers that could become the focus of treatment, for example, unresolved painful relationships or the client’s self-criticism.

Landmark Studies

Quantitative Studies

One of the oldest and most influential of studies on empathy and outcome was conducted by Barrett-Lennard (1962), largely because it was the origin of the most widely used empathy measure, the Empathy subscale of the BLRI. The author adapted and refined an earlier Q-sort measure into parallel client and therapist-perspective self-report measures that included therapist empathic understanding. He provided initial validity and reliability data, using a sample of 35 clients from the University of Chicago Counseling Center. The BLRI was administered to both therapists and their clients after Sessions 5, 15, and 25 and at the end of treatment; outcome was assessed using therapist posttherapy global ratings of degree of client change and an integration of client improvement self-ratings on several pre–post measures of adjustment. All told, six different process–outcome correlations could be calculated, based on (p. 255) dichotomizing clients as more versus less improved; these associations were all in the positive direction and ranged from .16 to .49, with a mean of .42.

Miller and colleagues (1980), in a study of focused versus broad-spectrum behavior therapy for problem drinkers, randomly assigned clients to therapists. Therapists were nine student trainees who received three months of intensive training in basic listening and counseling skills, along with training in the treatment modalities to be delivered. Three raters observed the psychotherapy sessions through one-way mirrors and rated therapists’ accurate empathy utilizing a modification of the Truax scales. Ratings were done before outcome data were collected or analyzed. Mean therapist skill ratings were compared to therapists’ success rates across clients, and these two were highly related (r = .82) in a positive direction. Higher therapist empathy led to greater reductions in problem drinking.

This study also bears on the causality issue, at least on the temporal order of empathy and outcome. The ratings were performed before the outcome data were available. Since clients were randomly assigned to therapists, it seems somewhat implausible that the top-rated therapist on empathy, for instance, just happened to receive 100% of clients who went on to improve the most. Thus this study provides evidence that empathy rated during treatment, or some unmeasured covariate of empathy, occurred before and probably contributed causally to outcome.

Another landmark study in our meta-analysis is Moyers et al. (2016), which had by far the largest number of clients (700) and employed sophisticated multilevel modeling to separate the effects of empathy on outcome both within and between therapists. Drawing on data from the Project COMBINE trial on the effects of motivational interviewing on alcohol problems, outcomes were assessed using drinking behaviors after treatment. Empathy was assessed using an observer measure based on transcripts of session recordings. Between-therapist effects accounted for 11% of the variance in empathy, requiring multilevel modelling to separate out between- and within-therapist components of the association between empathy and outcome.

The effects for empathy were comparatively small. The within-therapist effect picked up differences between clients with whom the therapist was more or less empathic than usual for them; it was statistically significant and equivalent to r = .15. The between-therapist effect looked at more trait-like differences in therapist empathy level and was equivalent to r = .05, which was not statistically significant. Averaged together in our analysis, these two effects amounted to r = 10. Our meta-analysis includes several recent large studies with small effects, mostly but not exclusively from the motivational interviewing literature. These studies involve very brief interventions (often only two sessions) and often include only observer measures of empathy; they may thus represent a different population of studies from the rest of the meta-analysis.

Qualitative Studies

In contrast to the large number of quantitative studies reviewed here, there is relatively little qualitative research on empathy. In the first of a series of three linked qualitative studies, Myers (2000) examined the experience of five female psychotherapy clients. (p. 256) The focus of the studies was on the process of empathic listening, and interviews were conducted after the therapy had ended. The clients were selected because they had formed a good relationship with their therapist and because they were particularly articulate about their experiences. Myers explicitly acknowledges that no claims of generality can be made from her findings.

Clients participated in a humanistic empathy-based therapy for at least 20 sessions. They were interviewed and also provided written narrative accounts to a series of questions. Interview and written data were content-analyzed for themes. In the first study, three factors characterized empathy: the experience of being understood in contrast to the negative experience of being misunderstood by others, getting feedback from the therapist, and development of a feeling of safety. All clients interviewed reported being listened to as an essential element of their relationship with their therapists.

In the second study (Myers, 2003), the impacts of being empathically listened to and understood were identified. These prominently included an increased sense of personal agency, a redefined sense of self, a renewed sense of being-in-the-world, increased self-acceptance, and increased self-empathy. In the third study (Myers & White, 2010), the clients were interviewed 10 years later. They continued to attribute personal change and growth to being empathically listened to and understood. In particular they noted how they had continued to grow based on their treatment. They emphasized an enhanced sense of personal agency in the form of an increased sense of self-efficacy and greater skill at emotional regulation.

In another landmark qualitative study, MacFarlane and colleagues (2017), Martin and Sterne (1976) interviewed nine clients about their experiences of empathy in psychotherapy. They examined at three features of the empathic experience: (a) clients’ phenomenology of empathy, (b) clients’ interpretations of the psychotherapists’ empathic communication, and (c) clients’ perceptions of the utility (e.g., benefits and consequences) of empathy. The nine clients were seen by eight therapists, who were doctoral students in clinical psychology. As in the Myers (2000, 2003) studies, clients were nominated by therapists based on their verbal skills and their ability to report on their experiences. Eight clients were interviewed within 24 hours of a therapy session; the other client was interviewed within 48 hours.

All clients engaged in a two-hour Interpersonal Process Recall interview based on videotapes of their session. Data were analyzed utilizing grounded theory. Three types of empathy emerged: clients’ perceptions of therapists’ cognitive empathy (a complex combination of therapist attentiveness, asking pertinent questions, reflections, and trying to see things from the client’s perspective), emotional empathy (perceptions of therapists’ emotional attunement), and client empathy (clients’ attunement to the therapist). Two classes of empathy benefits, as perceived by clients, were frequently reported: process-related benefits (how empathy facilitated the therapy process, including impacts like pacing of the session, facilitating openness, helping overcome demographic differences between therapist and client, and facilitating trust) and client-related benefits (improvements in self-understanding, therapist undivided attention as a corrective experience, feeling better, feeling happier). Both of these qualitative (p. 257) studies point to the value of including the client perspective in empathy research and also show how an intensive qualitative examination of the empathic process can begin linking it to client outcome.

Results of Previous Meta-Analyses

Major reviews of the empathy literature have occurred since the 1970s. The first major review of the association between therapist facilitative conditions (including empathy) and outcome (Truax & Mitchell, 1971) was later strongly criticized (Lambert et al., 1978) for selective citations of results and numerous methodological failings. Two reviews in the 1970s (Gurman, 1977; Parloff et al., 1978) focused on client perceptions of facilitative conditions and were fairly positive but expressed concern about possible confounds among client-perceived empathy, other facilitative conditions, and client-rated outcome. These focused more broadly on facilitative conditions without attempting to separate out therapist empathy and were either narrative reviews or used box scores to summarize results. Subsequently, Orlinsky and colleagues (1994, 2003) separated out therapist–client mutual empathic resonance and reported strong results using a box score method.

The first meta-analyses to focus specifically on the empathy–outcome literature were the two previous versions of this chapter in 2002 (Bohart et al., 2002) and 2011 (Elliott et al., 2011). Both meta-analyses reported a moderately positive but variable relation between therapist empathy and client outcome. The 2011 meta-analysis was conducted on 57 studies (224 effects) and encompassed a total of 3,599 clients. The average weighted correlation between empathy and outcome was r = .31.

Meta-Analytic Review

In this section we report the results of an updated meta-analysis conducted on available research relating empathy to psychotherapy outcome. We addressed the main question of the overall association between therapist empathy and client outcome. Additionally, we investigated multiple potential moderators of that association: (a) Do different forms of psychotherapy yield different levels of association between empathy and outcome? (b) Does the perspective used to measure empathy predict the level of association between empathy and outcome? (c) Does broad type of client-presenting problem matter for the level of empathy–outcome association? and (d) What other study or sample characteristics predict the level of association between empathy and outcome (i.e., year of publication, sample size, treatment setting, therapy modality and length, therapist experience, type of outcome measure, unit of process)?

Search Strategy

We started with the studies we used in our meta-analyses (Bohart et al., 2002; Elliott et al., 2011) in the two previous editions of this book, which included studies gathered (p. 258) from a wide variety of sources. We then did an inclusive search of PsycInfo for all years, using the search terms:

  • “empathy” or “empathic”

  • AND “psychotherapy”, “counseling” OR “counselling”

  • AND “change” OR “outcome*” OR “improvement”

  • AND methods: empirical study, quantitative study, treatment outcome, or clinical trial

Screening and Rating of Articles

This search produced 2,222 potential sources, which were then screened systematically as documented in Table 7.1. The inclusion criterion was process outcome research studies relating measured therapist empathy to psychotherapy outcome. The studies (p. 259) needed to report a correlation or sufficient information to calculate one. (Results of applying the exclusion criteria are presented in Table 7.1.) The abstracts of the potential sources were screened by the four co-authors, with a sample of 200 studies to assess reliability. The interrater reliability between original rater and checking rater was kappa = .61, which reflects the difficulty of identifying and agreeing on low baserate events (less than 2% of potential sources were eventually retained for analysis). Accordingly, disagreements were retained for retrieval and further screening. Agreements (n = 13) usually ended up being retained in the final sample used in the meta-analysis (57%), whereas disagreements (n = 14) were almost always dropped (8% retained). This process resulted in 148 sources being retained. Screening for duplicates resulted in dropping 15 sources, 14 of which had been included in the previous version of this meta-analysis, which covered sources through 2008 (Elliott et al., 2011). This process resulted in 133 sources, of which we were able to locate full texts for 99 (most of the dropped sources were doctoral dissertations).

Table 7.1. Prisma Information for Empathy–Outcome Meta-Analysis

Stage

N of Studies Included

N of Studies Excluded

Notes

1. Search Result:

2,222

Date of search: March 3, 2017

2. Abstract screening stage: Possible Empathy → Outcome studiesa

148

2,074

Not empathy → outcome studies

3. Screening for duplicates

133

15

Including 14 from Elliott et al. (2011)

4. Full text retrieval

99

34

Unable to retrieve full text

5. Full text review and analysis, plus studies carried over from Elliott et al. (2011)

24 + 58 = 82

75

Failed exclusion criteria at full text review or analysis

Note. Results of applying exclusion criteria:

  1. 1. Not about psychotherapy/counseling (e.g., training or supervision or empathy in nontherapist populations such healthcare workers in medical encounters) (step 2: 900; step 5: 8)

  2. 2. Not research on therapist empathy (e.g., theory articles, clinical case studies; research on client empathy as an outcome) (step 2: 579; step 5: 15)

  3. 3. Not process-outcome research on therapist empathy (e.g., measure therapist empathy but do not relate it to client outcome; includes helpful factors studies) (step 2: 433; step 5: 41)

  4. 4. n < 5 (e.g., case studies) (step 2: 48; step 5: 0)

  5. 5. Analogue studies (i.e., do not involve actual sessions with real clients) (step 2: 54; step 5: 1)

  6. 6. Empathy not specifically/explicitly measured (e.g., global Barrett-Lennard Relationship Inventory scores) (step 2: 19; step 5: 16)

  7. 7. Sessions <2 (step 2: 9; step 5: 11)

  8. 8. Effect size (r or d) not reported or calculable (step 2: 1; step 5: 0)

  9. 9. Reviews and meta-analyses (step 2: 12; step 5: 0)

Additional duplicates identified at step 5: 3

The exclusion criteria were again applied to these 99 full text sources, which were each evaluated by two of the co-authors. Interrater reliability was low (kappa = .45) because of ongoing confusion about whether “therapy” and “outcome” were being studied, whether therapist global facilitative conditions could be counted as measures of empathy, and a lack of clarity in the written reports about the length of brief interventions such as motivational interviewing. All disagreements were therefore discussed to reach consensus, resulting in 24 studies being retained in the analysis. These 24 studies were added to 58 studies that were carried over from Elliott et al. (2011), for a total 82 studies. (See Supplemental Table S1 for summary information about these studies.)

Supplemental Table S1. Table of Studies, Sample Sizes, and Effects

Authors

Year

Meta-Analysis Version

N of Clients

N of Effects

r

Barnicot et al.a

2014

3

157

2

.010

Barrett-Lennard

1962

1

35

6

.418

Bergin & Jasper

1969

1

24

1

.050

Beutler et al.

1972

1

31

1

.190

Brouzos et al.

2015

3

40

4

.000

Brug et al.

2007

3

142

6

.046

Buckley et al.

1981

1

71

6

.177

Bugge et al.

1985

1

274

1

.574

Bulllmann et al. [PCT]

2004

2

85

1

.330

Bulllmann et al. [CBT]

2004

2

86

1

.410

Cartwright & Lerner

1965

1

28

2

.173

Clark & Culbert

1965

1

10

1

.360

Cooley & Lajoy

1980

1

54

10

.316

Cramer & Takens

1992

2

37

6

.328

Dicken et al.

1977

3

43

9

.080

Dormaar et al.

1989

1

135

3

.000

Duan & Kivlighan

2002

3

57

2

.290

Filak & Abeles

1984

1

50

2

.339

Free et al.a

1985

3

57

9

.038

Fretz

1966

1

17

3

.320

Fuertes & Brobst

2002

3

85

1

.570

Fuertes et al.

2006

2

51

2

.529

Fuertes et al.

2007

2

59

2

.214

Gabbard et al.

1986

1

42

12

.218

Garfield & Bergin

1971

1

38

10

.016

Gelso et al.

2002

2

63

2

.235

Gillispie et al.

2005

3

121

2

.050

Goldman et al.

2000

1

38

4

.117

Goodman et al.

2015

3

5

1

.540

Greenberg & Webster

1982

1

31

4

-.046

Gross & DeRidder

1966

1

8

1

.750

Guydish et al.a

2014

3

151

2

.238

Hall & Davis

2000

1

162

2

.170

Hamilton

2000

1

132

1

.730

Hansen et al.

1968

1

70

2

.532

Hoffart et al.

2002

2

35

6

.045

Horvath & Greenberg

1981

1

29

3

.141

Kasarabada et al.

2002

3

511

1

.004

Kim et al.

2009

3

61

3

.580

Kolden

1996

3

121

2

.320

Kurtz & Grummon

1972

1

31

42

.069

Kwon & Jo

2012

3

48

1

.613

Lafferty et al.

1989

1

60

1

.640

Langhoff et al.

2008

2

55

4

.295

Lerner

1972

1

30

4

.165

Lesser

1961

1

22

2

-.292

Lorr

1965

1

320

3

.267

Malin & Posa

2015

3

30

1

.260

Marshall et al.

2002

2

39

2

.400

Marshall et al.

2003

2

41

3

.121

Martin & Sterne

1976

1

143

4

.150

Melnick & Pierce

1971

1

18

1

.450

Miller et al.t

1980

1

41

1

.819

Mitchell et al.

1973

1

120

1

.000

Moyers et al.a

2016

3

700

2

.100

Muller & Abeles

1971

1

36

1

.410

Murphy & Cramer

2014

3

62

2

.190

Orlinsky & Howard

1967

1

37

4

.386

Pantalon et al.

2004

2

16

2

.400

Payne et al.

2007

2

6

3

.693

Peake

1979

1

55

2

.196

Rabavilas et al.

1979

1

36

1

.506

Ritter et al.

2002

3

88

3

.250

Roback & Strassberg

1975

1

12

1

.070

Saltzman et al.

1976

1

55

4

.153

Sandberk & Akbaş

2015

3

20

2

.470

Sapolsky

1965

1

16

1

.377

Saunders

2000

1

114

5

.191

Spohr et al.

2016

3

40

2

.230

Staples et al. [Behavior Therapy]

1976

1

30

3

.050

Staples et al. [Psychodynamic Therapy]

1976

1

30

3

.003

Strupp et al.

1969

1

44

1

.330

Thrasher et al.

2006

3

30

1

.480

Truax

1966

1

80

3

.365

Truax et al.

1965

1

40

1

.198

Truax et al.

1966

1

40

5

.332

Truax & Wittmer

1971

1

40

5

.259

Truax et al.

1971

1

116

3

.510

Watson et al.

2014

3

55

7

.370

Wiprovnick et al.a

2015

3

59

1

.310

Wisconsin Project (Barrington, 1967; Kiesler et al., 1967; Van der Veen, 1967)

1967

1

12

12

.128

Woodin et al.

2012

3

25

2

.110

Characteristics of the Studies

Table 7.2 summarizes the 82 studies used in the meta-analysis. For measures of outcome, we included a study as long as there was some assessment of the effects of therapy, even if only at the session level (immediate outcome). There is some conceptual overlap between feeling understood and client satisfaction/session helpfulness, which was used in 19% of studies; we subsequently examined type of outcome measure as a moderator variable. The resulting sample consisted of 290 separate tests of the empathy–outcome association, aggregated into 82 different samples (from 80 studies) and encompassing a total of 6,138 clients, who were seen for an average of 25 sessions.

Table 7.2. Selected Study Characteristics

Parametric Characteristics:

N

M

SD

Range

Sample size: study/samples

Sample size: clients:

82 samples, 6,138 clients

74.9

(median: 42.5)

101.1

5–700

Length of therapy (sessions)

61

25.2

37.6

2–228

Effects per study

81

3.6

5.0

1–42

Categorical Characteristics:

Selected Categories

%

Year of publication

82

1961–1980

1981–2000

2001–2016

43

18

39

Theoretical orientation

82

Mixed, eclectic, or unknown

Humanistic-experiential

Cognitive-behavioral therapy

Psychodynamic

60

17

15

9

Treatment format

82

Individual

75

Treatment setting

81

Outpatient

83

Client presenting problem

82

Mixed, unspecified

54

Therapist experience level

73

Recent PhD or MD

Master’s level

34

34

Outcome assessment time point

80

Posttreatment

58

Main outcome perspective

69

Client

55

Outcome measure type

61

Symptom ratings

Improvement/recovery

Client satisfaction

24

21

19

Main empathy perspective

69

Client (mostly Barrett-Lennard, 1962)

Observer (mostly Truax & Carkhuff, 1967)

51

39

Empathy measurement unit

80

Therapy to date

53

Estimation of Effect Size

We used Pearson correlations as our main effect size metric in a random effect model. Our strategy was to extract all possible effects. Therefore, we used the following conventions (extensions of those used in Smith et al., 1980) to estimate r: First, if we had a significance level, we converted it to r. If the result was nonsignificant but we had enough information to calculate a t and then convert, we did so. If we had no other information than that the effect was nonsignificant, we set r at zero. If the (p. 260) (p. 261) (p. 262) authors indicated a “nonsignificant trend” but did not report a correlation (e.g., Kiesler et al. [1967] indicated several trends on Minnesota Multiphasic Personality Inventory scales), we estimated the trend by conservatively assigning an effect size of half the size of a significant r.

Coding Procedure and Analyses

As summarized in Table 7.2, we coded multiple features of each study. The following variables were coded: therapy format (individual or group), theoretical orientation, experience level of therapists, treatment setting (inpatient, outpatient), number of sessions (typically the mean), type of problems (mixed/unspecified, depression or anxiety, mild problems, and severe problems such as psychosis or incarceration), source of (p. 263) outcome measure (therapist rating, client rating, objective, and other measures), when outcome was measured (e.g., postsession, posttherapy, follow-up), type of outcome measured (e.g., symptom change, improvement, global), source of empathy measure (objective ratings, therapist, client, therapist–client congruence, trait measure), and unit of measure (e.g., 2- to 5-minute samples, session, therapy to date).

We conducted two sets of analyses: by effects and by studies. First, we analyzed the 290 separate effects to examine the impact of perspective of empathy measurement and type of outcome. Second, study-level analyses averaged individual effects within client samples before further analysis, thus avoiding problems of nonindependence and eliminating bias due to variable numbers of effects reported in different studies (Lipsey & Wilson, 2001). For analyses across studies, including overall effects and moderator variable analyses, we used Fisher’s r to z transformation, weighted studies by inverse error (n – 3), and analyzed for heterogeneity of effects using Cochrane’s Q, and a restricted maximum likelihood (REML) random effects model using Wilson’s (2006) macros for SPSS. We also calculated I2, an estimate of the proportion of variation due to true variability as opposed to random error (Higgins et al., 2003) and fail-safe numbers (vs. r = .2).

Results

Probably the single best summary value, as shown in Table 7.3, is the study-level random effects weighted r of .28 (95% confidence interval [CI] of .23–.33), a medium effect size (equivalent to d = .58). For analyses of the 290 nonindependent separate effects, average effects were somewhat smaller, at .21 (95% CI = .18–.24; equivalent to d = .43). These values were very similar to our previous reviews (Bohart et al., 2002; Elliott et al., 2011) and indicate that empathy generally accounts for about 9% of the variance in therapy outcome. This effect size is on the same order of magnitude as analyses of the relation between the alliance in individual therapy and treatment outcome (i.e., Flückiger et al., Chapter 2, this volume; d = .57). Overall, empathy typically accounts for more outcome variance than do specific treatment methods (compare Wampold’s [2015] estimate of d = .2 for intervention effects).

Table 7.3. Empathy–Outcome Correlations: Overall Summary Statistics

Effect Level (N = 290)

Study Level (N = 82)

N

M (95% CI)

M (95% CI)

Weighted Mean r

.21* (.18–.24)

.28* (.23–.33)

Cochrane’s Q

1039.2*

348.68*

I2

72.2%

76.8%

Fail-safe number

13

33

Note. 95% CI = confidence interval.

* p < .001.

We also assessed the likelihood of bias, either due to studies with negative effects not being published or to smaller studies with weaker methods producing more favorable results. First, we calculated the fail-safe number, that is, the number of studies with r = 0 results required to reduce the weighted effect to a minimum clinically interesting value of r = .2 (see Table 7.3). This value was based on 33 studies for the study-level effect of .28; the comparable number for effect level effects was only 12, which means that the effect-level effect is not particularly robust and would drop below the r = .2 threshold with only a small number of underreporting null results. Second, we created a funnel plot of the relation between effect size and level of standard error of r (which is a joint function of sample size and effect size). The correlation between standard error of r and effect size was .06, indicating an absence of bias deriving from smaller studies with less precise effects producing larger effects. Further, as can be seen in Figure 7.1, the funnel plot is quite symmetrical, making it unlikely that the (p. 264) overall effect would be shifted negatively if a larger number of more powerful studies were to be carried out.

Figure 7.1 Funnel plot of empathyoutcome effect by standard error of r.

Figure 7.1 Funnel plot of empathyoutcome effect by standard error of r.

Nevertheless, the .28 value conceals statistically significant variability in effects, as indicated by a study-level Cochrane’s Q of 348.68 (p < .001); in addition, I2 was 72%, a large value. Figure 7.1 also attests to the wide variability of effects, even in studies with reasonably large samples and small standard errors. These findings mean that a further examination of possible moderators of the empathy–outcome association is essential (Lipsey & Wilson, 2001). Our sample was large enough to include internal replication of the empathy–outcome association across a range of theoretical orientations in the .21–.30 range (Table 7.4), as well as across observer and client measurement perspectives (Table 7.5) and several client-presenting problems or populations (i.e., depression/anxiety, severe/incarcerated, and mixed/unspecified; Table 7.6). (p. 265)

Table 7.4. Mean Study-Level Effects across Theoretical Orientation

Theoretical Orientation

n

Mean Weighted r

Within Group Q

I2

Experiential/ Humanistic

14

0.24**

15.32

15.1%

Cognitive-Behavioral

12

0.30**

61.12**

82.0%

Psychodynamic

7

0.21**

5.62

0

Other/Unspecified

49

0.30**

257.08**

81.3%

Between groups Q

1.41 (df = 3, 78) (NS)

Note. Mean correlations and significance tests (vs. null hypothesis r = 0) for subgroups calculated using Fisher’s z scores and a restricted maximum likelihood random effects model using Wilson’s (2006) macros for SPSS. Cochrane’s Q tests for heterogeneity were evaluated as a chi-square test under a fixed effect model.

* p < .05;

** p < .01.

Table 7.5. Mean Within-Study Effects across Empathy Measurement Perspectives

Measurement Perspective

n

Mean Weighted r

Within Group Q

I2

Observer

102

.21a**

347.25**

70.9%

Client

117

.27a**

459.35**

74.8%

Therapist

37

.19a**

69.19**

48.0%

Empathic Accuracy

34

.01b

75.94**

56.5%

Between groups Q

26.11** (df = 3, 286)

Note. See note for Table 7.4. Mean weighted r values that share superscripts do not differ significantly (p ≥.05).

** p < .001.

Table 7.6. Mean Between-Study Effects across Grouped Client Problem Populations

Client Problem Population

N

Mean Weighted r

Within Group Q

I2

Severe, chronic, or incarcerated

15

.32**

57.89**

75.8%

Mixed or unspecified

44

.30**

181.5**

76.5%

Depression or anxiety

10

.26**

17.07*

47.3%

Mild, normal, or physical problems

6

.17

7.07

29.3%

Self-damaging activity

7

.19*

15.00*

60.0%

Between groups Q

70.15** (df = 4, 76)

Note. See note for Table 7.4. Mean weighted r values do not differ significantly (p ≥.05).

* p < .05;

** p < .01.

Moderators and Mediators

We divide this section on moderators and mediators into two parts: meta-analytic analyses of potential moderators and therapist mediators.

Moderator Analyses

Although the significant Q and large I2 statistics point to important moderator variables or sources of heterogeneity, they do not specify what these are. In our previous meta-analyses, we hypothesized that the empathy–outcome association would be larger in humanistic-experiential therapies, such as person-centered. However, we have again found tantalizing evidence that empathy might prove more important to outcome in cognitive-behavioral therapies or other/unspecified treatments than in humanistic or psychodynamic therapies (p < .05). None of the pairwise comparisons were statistically significant, possibly because the larger effects are marked by large Q and I2 statistics indicating other sources of unexplained variability.

We have also replicated the differences we previously found among the empathy measurement perspectives using effect-level analyses (Table 7.5; between-groups (p. 266) Q significant at p < .001). Specifically, client measures predicted outcome the best (mean weighted r = .27; n = 117), slightly better than observer rated measures (r = .21; n = 102) and therapist measures (r = .19; n = 37); each of these mean effects was significantly greatly than zero (p < .001) but did not differ significantly from each other. In contrast, empathic accuracy measures were unrelated to outcome (r = .01; n = 34, ns), and were statistically smaller than effects for each of the other three measures (p < .05). However, most perspectives were again characterized by large (I2 > 50%), statistically significant amounts of nonchance heterogeneity. Clarification of the source of this heterogeneity awaits further research; however, for now it seems fair to say that clients’ feelings of being understood and observer ratings (and, to a lesser extent, therapist impressions) appear to carry significant weight as far as outcome goes but that empathic accuracy measures do not, in spite of their intuitive appeal.

A new feature here was the analysis of separate effects for client populations grouped into five broad headings (Table 7.6). The largest empath–outcome association was for severe/chronic incarcerated populations, mixed/unspecified, and depressed/anxious. Smaller effects were found for mild/normal/physical problem and self-damaging activities (e.g., substance misuse). (Although the overall between-groups Q showed statistically significant differences, none of the paired comparisons were significant.)

Finally, in Table 7.7, we examined several other variables that might account for some of the heterogeneity of the effect sizes: year of publication, number of effects analyzed in a study, sample size, setting (outpatient vs. inpatient), treatment format (individual vs. group), length of therapy in sessions, therapist experience level, client severity (estimated by scaling client presenting problem/population), globalness of outcome measures (individualized to satisfaction ratings), and size of empathy unit (5 min segment to whole therapy). Using weighted correlations (random effects, REML analyses), only the number of effects reported in a study (r = –.23), outcome globality (r = .32), and outpatient setting (r = –.14) were statistically associated with outcome. Using backwards stepwise regression (random effects, REML), four of these continuous variables significantly predicted effect size (p < .05): analyzing fewer effects in a study (β‎= –.30), having a smaller sample of clients (β‎ = –.24), outpatient setting (β‎ (p. 267) = –.31), and outcome globality (β‎ = .28); together, these four variables accounted for 33% of the variance in effect size.

Table 7.7. Weighted Bivariate Associations between Study-Level Effect Size and Selected Continuous Moderator Variables

Predictor

k

r

Year of publication

82

.01

N of effects analyzed in study

82

–.23*

N of clients in study

82

–.15

Setting (1 = outpatient; 2 = inpatient)

81

–.14*

Format (1 = individual; 2 = group)

72

.04

Length of therapy (in sessions)

61

.10

Therapist experience level (6-point scale)

73

–.16

Level of client problem severity (4-point scale)

64

.09

Outcome globality (6-point scale: individualized to satisfaction ratings)

62

.32*

Size of empathy unit (6-point scale)

80

.04

Note. All analyses weighted effects using inverse error (i.e., n – 3); random effects analyses followed restricted maximum likelihood model.

* p < .05.

To sum up our moderator results: First, we found a “more is less” effect. That is, more ambitious (in terms of number of clients) and wider ranging studies (in terms of range of measures) produced smaller effects. Second, for the second time in our series of meta-analyses, we found tantalizing trends that the empathy–outcome association might be larger for CBT than for experiential and psychodynamic clinicians. Third, the client’s perception of empathy fares better in predicting outcome than the therapist’s. Fourth, using global outcome variables like client satisfaction resulted in larger associations between therapist empathy and outcome, possibly because conceptual confounding between client perceptions of empathy and client ratings of posttherapy satisfaction. Fifth, there was some evidence that the empathy–outcome association was stronger for clients in outpatient settings. Sixth, therapist empathic accuracy did not predict outcome.

In contrast, we found the empathy–outcome association held across a wide range of other variables, both substantive (client-presenting problem/severity, therapy modality) and methodological (year of publication, level/size of unit at which empathy was measured). Findings for therapist experience level and length of therapy were equivocal.

Therapist Mediators

As noted earlier, affective neuroscience researchers have proposed that empathy involves three interlinked skills: affective simulation, perspective-taking, and (p. 268) regulation of one’s own emotions (Decety & Jackson, 2004). Supporting this, research has found a relation between various measures of cognitive complexity, such as those of perspective-taking or abstract ability, and empathy in both developmental psychology and psychotherapy (Eisenberg & Fabes, 1990; Henschel & Bohart, 1981; Watson, 2001). With respect to affective simulation and emotion regulation, therapists open to conflictual, countertransferential feelings were perceived as more empathic by clients (Peabody & Gelso, 1982).

The degree of similarity between therapist and client (Duan & Hill, 1996; Gladstein & Associates, 1987; Watson, 2001) also influences the level of empathy. Similarity and familiarity between the target of empathy and the empathizer have been found to be important modulators of empathy in neuroscience research on mirror neurons (Watson & Greenberg, 2009). Another vital factor is therapist nonlinguistic and paralinguistic behavior. This encompasses therapists’ posture, vocal quality, ability to encourage exploration using emotion words, and relative infrequency of talking too much, giving advice, and interrupting (Duan & Hill, 1996; Watson, 2001). Other research has shown that responses that are just ahead of the client seem to be more effective than responses that are either at the same level as the client or at a more global level (Sachse, 1990a, 1990b; Tallman et al., 1994; Truax & Carkhuff, 1967). In a qualitative study of clients’ experience of empathy, interrupting, failing to maintain eye contact, and dismissing the client’s position while imposing the therapist’s own position were all perceived as unempathic (Myers, 2000). Conversely, being nonjudgmental, attentive, open to discussing any topic, and paying attention to details were perceived as empathic.

Evidence for Causality

In this chapter we have focused largely on quantitative process–outcome correlational studies in which in-session empathy was used to predict client outcome, generally posttherapy. We have assembled a large and quite broad body of correlational evidence that clearly shows small to medium covariation between these two variables.

What does this covariation mean from a causal point of view? We do not claim that this evidence by itself is sufficient to justify strong and generalizable causal inference that therapist empathy causes client outcome. On the other hand, we would similarly doubt any other single kind of evidence that might be brought to bear on the causal link between these two variables, including randomized controlled trials (RCTs). For us, saying any single kind of evidence is sufficient for making causal inferences is dogma rather than science.

What we advocate instead is a general strategy that logically considers the kind of evidence needed to make generalizable causal inferences. We recommend combining two frameworks for doing so (Cooke & Campbell, 1979; Haynes & O’Brien, 2000). Together, these provide an integrative model of causal inference, consisting of six conditions, all of which we hold are necessary for generalizable causal inference; these fall into three sets of paired conditions (see Box 7.1). Using this framework, we (p. 269) next consider what knowledge claims are currently justified by the available evidence on therapist empathy as a cause of client posttherapy change:

  1. 1. Precedence: This condition is satisfied when studies measure (a) client outcome at or after the end of therapy and (b) therapist empathy at some earlier point during therapy, which helps us rule out or minimize reverse causation, where for example client early improvement enables therapists to be more empathic. Path analytic or panel correlational designs help us to do this statistically. Assessing therapist empathy independently (e.g., by observer or therapist self-ratings) from client outcome also helps. In our review, we found statistically significant effects when the precedence condition was satisfied, although they were generally smaller.

  2. 2. Plausibility: Regardless of the empirical data, scientists do not accept causal relationships in the absence of a plausible theory linking cause and effect. The best known such theory for the empathy–outcome connection was proposed by Rogers (1957) in his “process equation,” which described therapy empathy as one of six conditions needed for personality change. If Rogers’ theory now seems overly general, qualitative research studies, such as those we have sampled from here, begin to provide what qualitative researchers (e.g., Corbin & Strauss, 2008) refer (p. 270) to as a “grounded theory” of the specific ways in which therapist empathy can lead to client outcome. Based on current theory and emerging qualitative research, we regard this condition as well satisfied.

  3. 3. Statistical conclusion validity: This is a necessary but not sufficient condition for causal inference. The credo we were taught, “Correlation does not imply causation,” is actually an oversimplification. It would be more accurate to say, “Correlation does not establish causation . . . but it certainly helps.” It is also true to say, “Correlation proves causation” in the old-fashioned sense of the word “prove,” which originally meant “to put to the test.” Most accurately, we can say, “Correlation implies that causation is possible and worth investigating further”; conversely, we can say, “In the absence of attenuating factors and suppressor variables, the absence of correlation rules out causation.” Assessing the statistical covariation between therapist empathy and client outcome was the main focus of our review and is well-supported by our data.

  4. 4. Internal validity: Ruling out alternative causes is what RCTs are supposed to be good at; however, the ethical codes that govern psychotherapists forbid them from deliberately practicing in an unempathic manner. Thus alternative strategies must be used. First, it is possible to contrast treatments that explicitly highlight therapist empathy, such as humanistic-experiential psychotherapies (HEPs), with no treatment or other therapies that do not emphasize empathy. A recent meta-analysis (Elliott et al., 2013) of 31 RCTs involving HEPs did just this, finding a weighted standardized difference (dw) of .81 in comparison to no-treatment or wait-list control groups. In a related meta-analysis, the same authors also analysed 17 RCTs directly comparing Person-centered therapy to CBT, finding statistical equivalence (dw = -.10) between the two treatments. Thus from the experimental evidence we can say that treatments that emphasize therapist empathy are (a) better than no treatment and (b) as effective as CBT. However, this evidence does not tell us what actually brought about change in the treatments studied (see “Construct validity”).

    That means that a second work-around strategy is needed, as is used in other observational sciences such as astronomy: analyzing and assessing realistic alternative causes (Cook & Campbell, 1979). The most plausible of these internal validity threats are the following:

    • Reverse causation: Clients who are getting better elicit more empathy from therapists; establishing precedence (see condition 1) can certainly rule this out, as path analysis allows us to do.

    • Third variable causation/selection bias: Clients who bring more personal resources to therapy are more likely both to improve and to elicit more therapist empathy. Assessing plausible third variables, such as client openness to experience, self-awareness, and greater empathy for the therapist, makes it possible to rule this out.

    • Mortality: Clients with less empathic therapists may be more likely to drop out of the study, so that their posttherapy data are lost, suppressing possible empathy–outcome effects.

    • (p. 271) Compensatory equalization: Therapists who are less empathic may try to compensate by providing a wider range of more technical interventions, such as self-help resources and homework, also suppressing empathy effects.

    The other known internal validity threats (Cook & Campbell, 1979) are unlikely to feature in correlational research or will create conservative biases by increasing error variance; these include instrumentation (rater drift, practice effects), history, maturation, statistical regression, interactions with selection, diffusion/imitation of treatments, compensatory rivalry, and resentful demoralization. This suggests two ideas: (a) In general, internal validity threats are as likely to suppress empathy–outcome associations as to inflate them, and (b) reverse and third variable causation are the only really credible internal validity threats that need to be controlled for in process-outcome research on therapist empathy. This speaks to the importance of moving to more sophisticated causal modeling research; in this update we identified only six such studies, not enough to allow trustworthy conclusions. We advocate delaying the making of strong claims for therapist empathy as a cause of client outcome until there are enough well-designed causal modeling studies.

  5. 5. Construct validity: Although internal validity looms large in debates over correlational data, in our experience the construct validity of the cause and effect variables has proven to be a much more challenging causal condition to satisfy. How do we know that the key causal factor is therapist empathy and not therapist positive regard or even client satisfaction (a weak kind of outcome)? Just because a researcher has labeled his or her instrument “empathy” does not mean that it is in fact what someone else would call empathy as opposed to positive regard, psychological presence, or responsiveness. As we have indicated in this chapter, separating out therapist empathy from other relational conditions is particularly problematic, leading us to conclude (in the spirit of Rogers): Empathy is a key component of the broader set of therapist offered qualities that clients use to cause themselves to change. In short, we are conceding any claim that therapist empathy uniquely predicts (let alone causes) client outcomes.

  6. 6. External validity: The empathy–outcome association has been demonstrated across a wide range of real-world settings, including different theoretical orientations, treatment settings, client-presenting problems, and therapist experience levels.

Client Contributions

As noted, clients contribute to both the experience of empathy and its effects in psychotherapy in several ways. Empathy may be at least as much a client variable as it is a therapist variable. Clients’ perceptions of therapist empathy correlate slightly better with outcome than do therapists’ or observers’ ratings.

Furthermore, there is no particular set of therapist behaviors or techniques that clients identify as “empathic” (Bachelor, 1988). From clients’ perspectives, empathy often reflects a global relationship ambiance. This suggests that in part it is (p. 272) clients’ construction of what happens in therapy that constitutes empathy (Bohart & Byock, 2005).

It is probably more accurate to say that empathy is interactionally constructed (Brodley, 2002; Wynn & Wynn, 2006), which can happen in different ways. First, it matters how clients and therapists mutually perceive one another. In a recent study (Murphy & Cramer, 2014), researchers determined that when therapists and clients were mutually perceiving as experiencing high levels of therapeutic facilitative conditions (including empathy), there was a stronger correlation with outcome. Second, clinical and research experience suggests that the amount of therapist empathy may partly depend on the client and the client’s behavior. Early studies (Kiesler et al., 1967) found that levels of empathy were higher with clients who had less pathology and who were brighter but yet were lower in self-esteem. Therefore, the client him- or herself almost certainly influences therapist empathy. Client revelation is an essential link in the cycle of empathy (Barrett-Lennard, 1981). Clients who are more open to and communicate their inner experiencing will be easier to empathize with.

On the other hand, not all clients respond favorably to explicit empathic expressions. One set of reviewers (Beutler et al., 1986) cited evidence that highly sensitive, suspicious, and oppositional patients perform relatively poorly with therapists who are particularly empathic and involved. Another study (Mohr & Woodhouse, 2000) found that some clients prefer business-like rather than warm, empathic therapists. Of course, when therapists are truly empathic they attune to their clients’ needs and adjust how and how much they express empathy.

More broadly, different types of empathy may prove hindering or helpful to clients at different times. In alliance ruptures, for example, it is probably useful for therapist empathy to be accompanied and deepened by genuine warmth, openness, and concern for the clients’ feelings, rather than defending oneself and blaming the client (also see Safran, Muran, & Eubanks, Chapter 16, this volume).

Keeping in mind the notion of empathy as not only getting inside the skin of the client but getting inside the skin of the relationship (O’Hara, 1984), in some cases the therapist is probably more empathic by not expressing empathy. Martin (2000) notes: “Think of the insensitive irony of a therapist who says, ‘I sense the sadness you want to hide. It seems like you don’t want to be alone right now but you also don’t want somebody talking to you about your sadness’ ” (pp. 184-185). This response might technically seem empathic, but in fact, at a higher level, it is unempathic and intrusive, because it violates the client’s need for interpersonal distance. Variations among clients in desire for and receptivity to different expressions of empathy need further research.

In conclusion, empathy represents a mutual process of shared communicative attunement (Orlinsky et al., 2003). In the sense of the title of this book—“psychotherapy relationships that work”—it may well be that we need to think of empathy more as a part of genuine relationship and study it as such in its association with outcome, rather than solely as a therapist or client variable.

(p. 273) Limitations of the Research

Beyond the difficulties in making causal inferences from process–outcome correlations, many reviewers (e.g., Lambert et al., 1978; Patterson, 1984; Watson, 2001) have described a range of problems with research on empathy. In addition to the well-known difficulty of inferring causality from correlational data, these entail (a) the questionable validity of some outcome measures (e.g., client satisfaction); (b) the lack of sensitive outcome measures; (c) the restricted range of predictor and criterion variables; (d) confounds among variations in time of assessment, experience of raters, and sampling methods; and (e) incomplete reporting of methods and results. In fact, these and other problems are not restricted to empathy research but are common to all process–outcome research (Elliott, 2010).

The restricted range of predictor and criterion variables is particularly a problem. In an early and widely cited study (Mitchell et al., 1973), for instance, most of the therapists scored below the minimum level of empathy considered to be effective, and outcome was only modest to moderate in the study. It is not surprising that no significant correlations were found. Furthermore, in a few cases, results were reported as either significant in the positive direction or nonsignificant, possibly disguising weak negative effects. Such reporting practices prove problematic for calculating effect sizes based on limited information, introducing error into calculations.

In this iteration of this meta-analysis, we noted that number of tests of the empathy–outcome association, number of clients, and outcome specificity (vs. global satisfaction) all lowered effect size. This pattern of results suggests a methodologically-driven “more is less” effect (more tests, more participants, more specificity), which needs further investigation.

In the meantime, we believe it is time to move beyond research that simply correlates empathy (and other relationship conditions) with outcome to look at more differentiated questions concerning the measurement of empathy, its manifestation, and its mechanisms of change. It is our view that a mix of research methods is needed to do this, including careful qualitative interview research, detailed discourse analysis on within-session empathy change processes, systematic single-case studies using rich case records, and sophisticated quantitative research using time series and multilevel modeling methods on which to base sound causal inferences.

In attempting to establish a causal relation between empathy and outcome, we advocate a number of statistical procedures that researchers can perform in future studies that offer an alternative to the true experiment with randomization. These include using structural equation modeling and path analysis that each allow tests for variance within causal models in which empathy is represented as an exogenous variable. In structural equation models, empathy is considered a latent construct that is measured by its effect on manifest variables such as outcome. A second alternative is the use of longitudinal data sets with multiple observations of empathy and outcome across the psychotherapeutic process. Panel design using cross-lagged correlations is a method proposed as an adequate test of spuriousness (Kenny, 1975; e.g., Cramer & Takens, 1992). One concern with the cross sectional design in much of current research is (p. 274) that it is not possible to determine whether higher empathy leads to better outcome or vice versa (reverse causation); testing the cross-lagged correlations over time can offer greater security in the claims about the causal nature of the relationship variables on outcomes. By adding at least one additional time point for the assessment of the relationship variables that is synchronized to outcome assessment, the question of causality could be more rigorously addressed in the literature.

Diversity Considerations

Few studies have examined diversity or multicultural competence and psychotherapists’ empathy. The development of multicultural competence is required of mental health professionals as reflected in training and accreditation guidelines. It is important for therapists working with diverse populations to be empathic to their clients’ specific circumstances as well as the complexities inherent in their social and political locations (Fuertes et al., 2006). This in-depth understanding of clients’ specific location within society includes sensitivity to race, oppression, socioeconomic status, gender, sex, and religion as well as other sociopolitical forces. Competent therapists working with diverse populations display high levels of person empathy as well as relational and moment-to-moment empathy for their clients’ in-session experiences.

The few studies that have looked at empathy with diverse clients in psychotherapy have examined its relations with working alliance, real relationship, client satisfaction, and intention to follow treatment recommendations. One study highlighted the association between therapists’ empathy and clients’ satisfaction with the mental health services they were receiving (Gillispie et al., 2005). Clients’ overall satisfaction was related to how empathic they perceived their therapists to be and was also positively related to their intent to make use of aftercare. The authors suggest that clients from diverse groups may have a greater need for therapists to be understanding, nonjudgmental, and emotionally supportive during treatment to ensure their participation.

One intriguing study (Fuertes et al., 2007) investigated the relation between multicultural competence, the real relationship (see Chapter 6), and BLRI-E subscale. In this study with AfricanAmerican clients, a positive relation was found between clients’ ratings of the real relationship and their experiences of their therapists as empathic. However, it is important not to conflate empathy with multicultural competence, as some studies have shown that the latter is important above and beyond general therapeutic competence and empathy for ethnic-minority clients.

Few studies have looked at gender, socioeconomic status, religion, or oppression specifically with respect to empathy. Nevertheless, researchers have looked at the impact of motivational interviewing on reducing violent abusive behavior in groups of men and women (Woodin et al., 2012). These researchers reported a reduction in both groups’ hostile/violent behavior when therapists used more reflections. In addition, a trend toward a reduction in abusive behavior was found with women when counselors showed greater empathy and used open-ended questions. This finding warrants follow-up, as it raises the question of whether men and women respond differently to various (p. 275) types of empathic behaviors. Given the few studies in the area, future research must examine the role of empathy (and other relationship variables) in working with clients from diverse groups.

Training Implications

In the 1940s and 1950s, Carl Rogers and colleagues conducted empathy training in experiential workshops (Kirschenbaum, 2007). They did not just offer didactic instruction on the concept of empathy but conducted live demonstrations and skill practice; over time they also added personal development work. In the 1970s, as part of the initial uncritical enthusiasm for helping skills training, a range of micro-skills programs emerged, with an emphasis on specific therapist behaviors (Danish & Hauer, 1973; Goodman; 1978; Ivey & Gluckstern, 1974). These skill training packages made certain assumptions about the mapping between particular therapist response modes (especially therapist reflection) and higher order concepts such as empathy. As we noted in the Therapist Mediators section, these assumptions soon proved simplistic as research found (a) reflections do not always convey empathy and (b) a wide range of other (especially nonverbal) behaviors can convey empathy.

A meta-analysis of 76 studies of empathy training in the helping professions (Dexter, 2012) produced an overall large (but highly heterogeneous) standardized mean difference effect of .99. That indicates the general effectiveness of a wide range of empathy training methods. Even didactic-only training produced reliable medium-sized effects (standardized mean difference) = .40); however, the combination of didactic and experiential methods produced the largest reliable effect (1.93). Adding feedback and modeling also produced large effects (1.76). In addition, there was a trend toward longer trainings (e.g., longer than 26 hours) having larger effects than briefer trainings (e.g., <5 hrs). These results are consistent with the humanistic-experiential therapy tradition, where substantial amounts of empathy training are a key element of the first year of a training program. This tradition also highlights the value an element not studied in the meta-analysis: personal development work to help students learn about their empathy blocks and blind spots.

In any case, this training meta-analysis points to the value of an extensive, multipronged approach to training empathic therapists. We regret that this level of empathy training is no longer the norm in the training of mental health professionals and are keenly aware of the problems posed by inadequate prior empathy training for therapists trying to learn advanced therapeutic methods. Without the foundational skill and capacity of empathy, much subsequent psychotherapy training rests on shaky ground.

Therapeutic Practices

The most robust meta-analytic evidence is that clients’ perceptions of feeling understood by their therapists relate to outcome. As we have shown, empathy is a (p. 276) medium-sized predictor of outcome in psychotherapy. It also appears to be a general predictor across theoretical orientation, treatment formats, and client severity levels. This repeated finding, in dozens of individual studies and now in multiple meta-analyses, leads to a series of clinical recommendations:

  • Psychotherapists continuously work to understand their clients and to demonstrate this understanding through responses that address the most poignant aspects of the clients’ experience to facilitate clients’ tracking of their inner experience in terms of the questions or concerns that they are exploring in therapy.

  • The primary task is to be empathically attuned and to understand the import or impact of clients’ experiences as opposed to being focused on words and content. Empathic therapists do not parrot clients’ words back or reflect only the content of those words; instead, they understand their clients’ goals overall as well as their moment-to-moment experiences in the session. Empathy entails capturing the nuances and implications of what people say and reflecting these back to them for their consideration.

  • Empathic responses show that therapists continually adjust their assumptions and understandings, attending to the leading edge of client experience so as to facilitate awareness of feelings and perspectives.

  • Our meta-analysis determined that clients’ reports of therapist empathy best predict eventual treatment outcome. Thus regularly assessing and privileging the client’s experience of empathy, instead of trying to intuit whether one’s behavior is empathic or not, can be helpful in treatment.

  • The meta-analysis also documented that observer ratings of accurate empathy predict outcome. Research has identified a range of useful empathic responses, several of which we presented earlier in the Clinical Examples section.

  • Empathy is not only something that is “provided” by the therapist as if it were a medication but is a co-created experience between a therapist who is trying to understand the client and a client who is trying to communicate with the therapist and be understood. Empathy is shown as much in how well the therapist receives, listens, respects, and attends to the client as in what the therapist does or says.

  • Empathic therapists assist clients to symbolize their experience in words and track their emotional responses so that clients can deepen their experience and reflexively examine their feelings, values, and goals. Therapists can help clients focus on this inner experience and access as much internal information as possible. To this end, therapists can attend to that which is not said or that which is at the periphery of awareness as well as that which is said and is in focal awareness (Watson, 2001).

  • Empathy entails individualizing responses to particular patients. We found significant heterogeneity in the empathy–outcome association, pointing to the value of personalization and clinical judgment. For example, certain fragile clients may find the usual expressions of empathy too intrusive, while hostile clients may find empathy too directive; still other clients may find an empathic focus on feelings too foreign (Kennedy-Moore & Watson, 1999). Effective therapists know when—and when not—to respond empathically. When clients do not want therapists to be (p. 277) explicitly empathic, truly empathic therapists will use their perspective-taking skills to provide an optimal therapeutic distance (Leitner, 1995), respecting their clients’ boundaries.

  • There is no evidence that accurately predicting clients’ own views of their problems or self-perceptions is effective. Therapists should neither assume that they are mind readers nor that their experience of the client will be matched by the client’s experience of self. Empathy is best offered with humility and held lightly, ready to be corrected.

  • Finally, because research has shown empathy to be inseparable from the other relational conditions, therapists are advised to offer empathy in the context of positive regard and genuineness. Empathy will probably not prove effective unless it is grounded in authentic caring for the client. Any one of the conditions without the others would provide a distinctly different interpersonal climate and relationship. We encourage psychotherapists to value empathy as both an “ingredient” of a healthy therapeutic relationship as well as a specific, effective response that promotes strengthening of the self and deeper exploration.

References

References marked with an asterisk indicate studies included in the meta-analyses.

Bachelor, A. (1988). How clients perceive therapist empathy—a content-analysis of received empathy. Psychotherapy, 25, 227–240. https://www.doi.org/10.1037/h0085337Find this resource:

*Barnicot, K., Wampold, B., & Priebe, S. (2014). The effect of core clinician interpersonal behaviours on depression. Journal of Affective Disorders, 167, 112–117. https://www.doi.org/10.1016/j.jad.2014.05.064Find this resource:

*Barrett-Lennard, G. (1962). Dimensions of therapist response as causal factors in therapeutic change. Psychological Monographs, 76(43), 1–36.Find this resource:

Barrett-Lennard, G. T. (1981). The empathy cycle: Refinement of a nuclear concept. Journal of Counseling Psychology, 28, 91–100.Find this resource:

*Barrington, B. L. (1967). The differential effectiveness of therapy as measured by the Thematic Apperception Test. In C. R. Rogers (Ed.), The therapeutic relationship and its impact (pp. 337–352). Madison: University of Wisconsin Press.Find this resource:

Batson, C. D. (2009). These things called empathy: Eight related but distinct phenomena. In J. Decety & W. Ickes (Eds.), The social neuroscience of empathy (pp. 3–15). Cambridge, MA: MIT Press.Find this resource:

Belenky, M. F., Clinchy, B. M., Goldberger, N. R., & Tarule, J. M. (1986). Women’s ways of knowing: The development of self, voice, and mind. New York, NY: Basic Books.Find this resource:

*Bergin, A. E., & Jasper, L. G. (1969). Correlates of empathy in psychotherapy: A replication. Journal of Abnormal Psychology, 74, 477–481.Find this resource:

Beutler, L. E., Crago, M., & Arizmendi, T. G. (1986). Research on therapist variables in psychotherapy. In S. L. Garfield & A. E. Bergin (Eds.), Handbook of psychotherapy and behavior change (3rd ed., pp. 257–310). New York, NY: Wiley.Find this resource:

*Beutler, L. E., Johnson, D. T., Neville, C. W., & Workman, S. N. (1972). “Accurate empathy” and the AB dichotomy. Journal of Consulting and Clinical Psychology, 38, 372–375.Find this resource:

Blatt, S. J., Zuroff, D. C., Quinlan, D. M., & Pilkonis, P. A. (1996). Interpersonal factors in brief treatment of depression: Further analyses of the National Institute of Mental Health (p. 278) Treatment of Depression Collaborative Research Program. Journal of Consulting and Clinical Psychology, 64, 162–171.Find this resource:

Bloom, P. (2016). Against empathy: The case for rational compassion. New York, NY: Ecco.Find this resource:

Bohart, A. C., & Byock, G. (2005). Experiencing Carl Rogers from the client’s point of view: A vicarious ethnographic investigation. I. Extraction and perception of meaning. The Humanistic Psychologist, 33, 187–212.Find this resource:

Bohart, A. C., Elliott, R., Greenberg, L. S., & Watson, J. C. (2002). Empathy. In J. Norcross (Ed.), Psychotherapy relationships that work (pp. 89–108). New York, NY: Oxford University Press.Find this resource:

Bohart, A. C., & Greenberg, L. S. (1997). Empathy: Where are we and where do we go from here? In A. C. Bohart & L. S. Greenberg (Eds.), Empathy reconsidered: New directions in psychotherapy (pp. 419–450). Washington, DC: American Psychological Association.Find this resource:

Bohart, A. C., & Tallman, K. (1999). How clients make therapy work: The process of active self-healing. Washington, DC: American Psychological Association.Find this resource:

Brodley, B. T. (2002). Observations of empathic understanding in two client-centered therapists. In J. C. Watson, R. N. Goldman, & M. S. Warner (Eds.), Client-centered and experiential psychotherapy in the 21st century: Advances in theory, research and practice (pp. 182–203). Ross-on-Wye, England: PCCS Books.Find this resource:

Brodley, B. T., & Brody, A. F. (1990, August). Understanding client-centered therapy through interviews conducted by Carl Rogers. Paper presented at the annual convention of the American Psychological Association, Boston, MA.Find this resource:

*Brouzos, A., Vassilopoulos, S. P., & Baourda, V. C. (2015). Members’ perceptions of person-centered facilitative conditions and their role in outcome in a psychoeducational group for childhood social anxiety. Person-Centered & Experiential Psychotherapies, 14, 32–46. https://www.doi.org/10.1080/14779757.2014.965843Find this resource:

*Brug, J., Spikmans, F., Aartsen, C., Breedveld, B., Bes, R., & Fereira, I. (2007). Training dietitians in basic motivational interviewing skills results in changes in their counseling style and in lower saturated fat intakes in their patients. Journal of Nutrition Education & Behavior, 39, 8–12. https://www.doi.org/10.1016/j.jneb.2006.08.010Find this resource:

*Buckley, P., Karasu, T. B., & Charles, E. (1981). Psychotherapists view their personal therapy. Psychotherapy: Theory, Research and Practice, 18, 299–305.Find this resource:

*Bugge, I., Hendel, D. D., & Moen, R. (1985). Client evaluations of therapeutic processes and outcomes in a university mental health center. Journal of American College Health, 33, 141–146.Find this resource:

*Bullmann, F., Horlacher, K. D., & Kieser, B. (2004). Clarifying as a mediator-variable in person-centered psychotherapy and therapists´ gender and therapy-school differences in empathy and positive regard (Unpublished study). Psychological Institute, University of Heidelberg, Heidelberg, Germany.Find this resource:

Burns, D. D., & Nolen-Hoeksma, S. (1992). Therapeutic empathy and recovery from depression in cognitive-behavioral therapy: A structural equation model. Journal of Consulting and Clinical Psychology, 60, 441–449.Find this resource:

Carkhuff, R. R., & Berenson, B. (1967). Beyond counseling and therapy. New York, NY: Holt, Rinehart & Winston.Find this resource:

*Cartwright, R. D., & Lerner, B. (1965). Empathy, need to change, and improvement in psychotherapy. Journal of Consulting Psychology, 27, 138–144.Find this resource:

*Clark, J. V., & Culbert, S. A. (1965). Mutually therapeutic perception and self-awareness in a T-group. Journal of Applied Behavioral Science, 1, 180–194.Find this resource:

(p. 279) Cook, T. D., & Campbell, D. T. (1979). Quasi-experimentation: Design and analysis issues for field settings. Chicago: Rand McNally.Find this resource:

*Cooley, E. J., & La joy, R. (1980). Therapeutic relationship and improvement as perceived by clients and therapists. Journal of Clinical Psychology, 36, 562–570.Find this resource:

Corbin, J., & Strauss, A. (2008). Basics of qualitative research: Techniques and procedures for developing grounded theory (3rd ed.). Thousand Oaks, CA: Sage.Find this resource:

*Cramer, D., & Takens, R. (1992). Therapeutic relationship and progress in the first six sessions of individual psychotherapy: A panel analysis. Counselling Psychology Quarterly, 5, 25–36.Find this resource:

Danish, S. J., & Hauer, A. L. (1973). Helping skills: A basic training program. New York, NY: Behavioral Publications.Find this resource:

Decety, J., & Ickes, W. (Eds.). (2009). The social neuroscience of empathy. Cambridge, MA: MIT Press.Find this resource:

Decety, J., & Jackson, P.L. (2004). The functional architecture of human empathy. Behavioral and Cognitive Neuroscience Reviews, 3, 71–100.Find this resource:

Decety, J., & Lamm, C. (2009). Empathy versus personal distress: Recent evidence from social neuroscience. In J. Decety & W. Ickes (Eds.), The social neuroscience of empathy (pp. 199–213). Cambridge, MA: MIT Press.Find this resource:

Dexter, V.J. (2012). Research synthesis with meta-analysis of empathy training studies in helping professions (Doctoral dissertation). New York University. Retrieved from https://media.proquest.com/media/pq/classic/doc/2696560791/fmt/ai/rep/NPDF?_s=NihQg7jRRnne7H%2FjhvJ2RzN7ybE%3D

*Dicken, C., Bryson, R., & Kass, N. (1977). Companionship therapy: A replication in experimental community psychology. Journal of Consulting and Clinical Psychology, 45, 637–646. https://www.doi.org/10.1037/0022–006X.45.4.637Find this resource:

*Dormaar, J. M., Dijkman, C. I., & de Vries, M. W. (1989). Consensus in patient-therapist interactions: A measure of the therapeutic relationship related to outcome. Psychotherapy and Psychosomatics, 51, 69–76.Find this resource:

Duan, C., & Hill, C. E. (1996). A critical review of empathy research. Journal of Counseling Psychology, 43, 261–274.Find this resource:

*Duan, C., & Kivlighan, D. M. Jr. (2002). Relationships among therapist presession mood, therapist empathy, and session evaluation. Psychotherapy Research, 12, 23–37. https://www.doi.org/10.1093/ptr/12.1.23Find this resource:

Egan, G. (1982). The skilled helper (2nd ed.). Monterey, CA: Brooks/Cole.Find this resource:

Eisenberg, N., & Eggum, N. D. (2009). Empathic responding: Sympathy and personal distress. In J. Decety & W. Ickes (Eds.), The social neuroscience of empathy (pp. 71–83). Cambridge, MA: MIT Press.Find this resource:

Eisenberg, N., & Fabes, R. A. (1990). Empathy: Conceptualization, assessment, and relations to prosocial behavior. Motivation and Emotion, 14, 131–149.Find this resource:

Elkin, I., Falconnier, L., Smith, Y., Canada, K. E., Henderson, E., Brown, E. R., & McKay, B. M. (2014). Therapist responsiveness and patient engagement in therapy. Psychotherapy Research, 24, 52–66.Find this resource:

Elliott, R. (2010). Psychotherapy change process research: Realizing the promise. Psychotherapy Research, 20, 123–135.Find this resource:

Elliott, R., Bohart, A. C., Watson, J. C., & Greenberg, L. S. (2011). Empathy. In J. Norcross (Ed.), Psychotherapy relationships that work (2nd ed., pp. 132–152). New York, NY: Oxford University Press.Find this resource:

(p. 280) Elliott, R., Watson, J., Goldman, R., & Greenberg, L. S. (2003). Learning emotion-focused therapy: The process-experiential approach to change. Washington, DC: American Psychological Association.Find this resource:

*Filak, J., & Abeles, N. (1984). Posttherapy congruence on client symptoms and therapy outcome. Professional Psychology: Research & Practice, 15, 846–855.Find this resource:

*Free, N. K., Green, B. L., Grace, M. C., Chernus, L. A., & Whitman, R. M. (1985). Empathy and outcome in brief focal dynamic therapy. American Journal of Psychiatry, 142, 917–921. https://www.doi.org/10.1176/ajp.142.8.917Find this resource:

*Fretz, B. R. (1966). Postural movements in a counseling dyad. Journal of Counseling Psychology, 13, 335–343.Find this resource:

*Fuertes, J. N., & Brobst, K. (2002). Clients’ ratings of counselor multicultural competency. Cultural Diversity & Ethnic Minority Psychology, 8, 214–223. https://www.doi.org/10.1037/1099-9809.8.3.214Find this resource:

*Fuertes, J. N., Mislowack, A., Brown, S., Gur-Arie, S., Wilkinson, S., & Gelso, C. J. (2007). Correlates of the real relationship in psychotherapy: A study of dyads. Psychotherapy Research, 17, 423–430.Find this resource:

*Fuertes, J. N., Stracuzzi, T. I., Bennett, J., Scheinholtz, J., Mislowack, A., Hersh, M., & Cheng, D. (2006). Therapist multicultural competency: A study of therapy dyads. Psychotherapy: Theory, Research, Practice, Training, 43, 480–490.Find this resource:

*Gabbard, C. E., Howard, G. S., & Dunfee, E. J. (1986). Reliability, sensitivity to measuring change, and construct validity of a measure of counselor adaptability. Journal of Counseling Psychology, 33, 377–386.Find this resource:

Gallese, V., Fadiga, L., Fogassi, L., & Rizzolatti, G. (1996). Action recognition in the premotor cortex. Brain, 119, 593–609.Find this resource:

*Garfield, S. L., & Bergin, A. E. (1971). Therapeutic conditions and outcome. Journal of Abnormal Psychology, 77, 108–114.Find this resource:

Geller, S. M., Greenberg, L. S., & Watson, J. C. (2010). Therapist and client perceptions of therapeutic presence: The development of a measure. Psychotherapy Research, 20, 599–610.Find this resource:

*Gelso, C. J., Latts, M. G., Gomez, M. J., & Fassinger, R. E. (2002). Countertransference management and therapy outcome: An initial evaluation. Journal of Clinical Psychology, 58, 861–867.Find this resource:

*Gillispie, R., Williams, E., & Gillispie, C. (2005). Hospitalized African American mental health consumers: Some antecedents to service satisfaction and intent to comply with aftercare. American Journal of Orthopsychiatry, 75, 254–261. https://www.doi.org/10.1037/0002-9432.75.2.254Find this resource:

Gladstein, G., A., & Associates. (1987). Empathy and counseling: Explorations in theory and research. New York, NY: Springer-Verlag.Find this resource:

*Goldman, R., Greenberg, L., & Angus, L. (2000, June). The York II Psychotherapy Study on Experiential Therapy of Depression. Paper presented at the annual meeting of the Society for Psychotherapy Research, Chicago, IL.Find this resource:

Goodman, G. (1978). SASHAtapes: Self-led automated series on help-intended alternatives. Los Angeles: California Self-Help Center, Department of Psychology, University of California.Find this resource:

*Goodman, G., Edwards, K., & Chung, H. (2015). The relation between prototypical processes and psychological distress in psychodynamic therapy of five inpatients with borderline personality disorder. Clinical Psychology & Psychotherapy, 22, 83–95. https://www.doi.org/10.1002/cpp.1875Find this resource:

(p. 281) Goubert, L., Craig, K. D., & Buysee, A. (2009). Perceiving others in pain: Experimental and clinical evidence of the role of empathy. In J. Decety & W. Ickes (Eds.), The social neuroscience of empathy (pp. 153–165). Cambridge, MA: MIT Press.Find this resource:

Greenberg, L. S., & Rushanski-Rosenberg, R. (2002). Therapists’ experience of empathy. In J. C. Watson, R. N. Goldman, & M. S. Warner (Eds.), Client-centered and experiential psychotherapy in the 21st century: Advances in theory, research and practice (pp. 204–220). Ross-on-Wye, England: PCCS Books.Find this resource:

*Greenberg, L. S., & Webster, M. (1982). Resolving decisional conflict by means of two-chair dialogue: Relating process to outcome. Journal of Counseling Psychology, 29, 468–477.Find this resource:

*Gross, W. F., & DeRidder, L. M. (1966). Significant movement in comparatively short-term counseling. Journal of Counseling Psychology, 13, 98–99.Find this resource:

Gurman, A. S. (1977). The patient’s perception of the therapeutic relationship. In A. S.Gurman & A. M. Razin (Eds.), Effective psychotherapy: A handbook of research (pp. 503–543). New York, NY: Pergamon.Find this resource:

*Guydish, J., Campbell, B. K., Manuel, J. K., Delucchi, K. L., Thao, L., Peavy, K. M., & McCarty, D. (2014). Does treatment fidelity predict client outcomes in 12-step facilitation for stimulant abuse? Drug & Alcohol Dependence, 134, 330–336. https://www.doi.org/10.1016/j.drugalcdep.2013.10.020Find this resource:

*Hall, J. A., & Davis, M. H. (2000). Dispositional empathy in scientist and practitioner psychologists: Group differences and relationship to self-reported professional effectiveness. Psychotherapy, 37, 45–56.Find this resource:

*Hamilton, J. C. (2000). Construct validity of the core conditions and factor structure of the Client Evaluation of Counselor Scale. The Person-Centered Journal, 7, 40–51.Find this resource:

*Hansen, J. C., Moore, G. D., & Carkhuff, R. R. (1968). The differential relationships of objective and client perceptions of counseling. Journal of Clinical Psychology, 24, 244–246.Find this resource:

Hayes, A. M., & Goldfried, M. R. (1996). Rogers’ work with Mark: An empirical analysis and cognitive-behavioral perspective. In B. A. Farber, D. C. Brink, & P. M. Raskin (Eds.), The psychotherapy of Carl Rogers (pp. 357–374). New York, NY: Guilford.Find this resource:

Haynes, S. N., & O’Brien, W. O. (2000). Principles of behavioral assessment: A functional approach to psychological assessment. New York: Plenum.Find this resource:

Hein, G., & Singer, T. (2010). Neuroscience meets social psychology: An integrative approach to human empathy. In M. Mikulincer & P. R. Shaver (Eds.), Prosocial motives, emotions, and behavior: The better angels of our nature (pp. 109–126). Washington, DC: American Psychological Association.Find this resource:

Henschel, D. N., & Bohart, A. C. (1981, August). The relationship between the effectiveness of a course in paraprofessional training and level of cognitive functioning. Paper presented at annual conference of the American Psychological Association, Los Angeles, CA.Find this resource:

Higgins, J. P. T., Thompson, S. G., Deeks, J. J., & Altman, D. G. (2003). Measuring inconsistency in meta-analyses, British Journal of Medicine, 327, 557–560.Find this resource:

*Hoffart, A., Versland, S., & Sexton, H. (2002). Self-understanding, empathy, guided discovery, and schema belief in schema-focused cognitive therapy of personality problems: A process-outcome study. Cognitive Therapy and Research, 26, 199–219.Find this resource:

Horvath, A. O., & Greenberg, L. S. (1986). The development of the Working Alliance Inventory. In L. S. Greenberg & W. M. Pinsof (Eds.), The psychotherapeutic process: A research handbook (pp. 529–556). New York, NY: Guilford.Find this resource:

*Horvath, A. O., & Greenberg, L. S. (1989). Development and validation of the working alliance inventory. Journal of Counseling Psychology, 36, 223–233.Find this resource:

(p. 282) Ickes, W. (2003). Everyday mind reading: Understanding what other people think and feel. Amherst, NY: Prometheus.Find this resource:

Ickes, W. (Ed.). (1997). Empathic accuracy. New York, NY: Guilford.Find this resource:

Ivey, A. E., & Gluckstern, N. B. (1974). Basic attending skills: Participant manual. North Amherst, MA: Microtraining Associates.Find this resource:

Johnson, S. M., Bradley, B., Furrow, J., Lee, A., Palmer, G., Tilley, D. G., & Woolley, S. R. (2005). Emotionally focused marital therapy workbook. New York, NY: Brunner Routledge.Find this resource:

*Kasarabada, N. D., Hser, Y., Boles, S. M., & Huang, Y. C. (2002). Do patients’ perceptions of their counselors influence outcomes of drug treatment? Journal of Substance Abuse Treatment, 23, 327–334. https://www.doi.org/10.1016/S0740-5472(02)00276-3Find this resource:

Kennedy-Moore, E., & Watson, J. C. (1999). Expressing emotion: Myths, realities, and therapeutic strategies. New York, NY: Guilford.Find this resource:

Kenny, D. A. (1975). Cross-lagged panel correlation: A test for spuriousness. Psychological Bulletin, 82, 887–903.Find this resource:

*Kiesler, D. J., Klein, M. H., Mathieu, P. L., & Schoeninger, D. (1967). Constructive personality change for therapy and control patients. In C. R. Rogers (Ed.), The therapeutic relationship and its impact (pp. 251–294), Madison: University of Wisconsin Press.Find this resource:

*Kim, B. S. K., Ng, G. F., & Ann, A. J. (2009). Client adherence to Asian cultural values, common factors in counseling, and session outcome with Asian American clients at a university counseling center. Journal of Counseling & Development, 87, 131–142. https://www.doi.org/10.1002/j.1556-6678.2009.tb00560.xFind this resource:

Kirschenbaum, H. (2007). The life and work of Carl Rogers. Ross-on-Wye, England: PCCS Books.Find this resource:

*Kolden, G. (1996). Effective processes in early sessions of dynamic psychotherapy. Journal of Psychotherapy Practice & Research, 5, 122–131.Find this resource:

*Kurtz, R. R., & Grummon, D. L. (1972). Different approaches to the measurement of therapist empathy and their relationship to therapy outcomes. Journal of Consulting and Clinical Psychology, 39, 106–115.Find this resource:

*Kwon, K. I., & Jo, S. Y. (2012). The relationship among counselor experience level, empathic accuracy, and counseling outcome in the early phase of counseling. Asia Pacific Education Review, 13, 771–777. https://www.doi.org/10.1007/s12564-012-9235-8Find this resource:

*Lafferty, P., Beutler, L. E., & Crago, M. (1989). Differences between more and less effective psychotherapists: A study of select therapist variables. Journal of Consulting and Clinical Psychology, 57, 76–80.Find this resource:

Lambert, M. J., DeJulio, S. J., & Stein, D. M. (1978). Therapist interpersonal skills: Process, outcome, methodological considerations, and recommendations for future research. Psychological Bulletin, 85, 467–489.Find this resource:

Landfield, A. W. (1971). Personal construct systems in psychotherapy. Chicago, IL: Rand McNally.Find this resource:

*Langhoff, C., Baer, T., Zubraegel, D., & Linden, M. (2008). Therapist-patient alliance, patient-therapist alliance, mutual therapeutic alliance, therapist-patient concordance, and outcome of CBT in GAD. Journal of Cognitive Psychotherapy, 22, 68–79.Find this resource:

Leitner, L. M. (1995). Optimal therapeutic distance: A therapist’s experience of personal construct psychotherapy. In R. A. Neimeyer & M. J. Mahoney (Eds.), Constructivism in psychotherapy (pp. 357–370). Washington, DC: American Psychological Association.Find this resource:

*Lerner, B. (1972). Therapy in the ghetto. Baltimore, MD: Johns Hopkins University Press.Find this resource:

*Lesser, W. M. (1961). The relationship between counseling progress and empathic understanding. Journal of Counseling Psychology, 8, 330–336.Find this resource:

(p. 283) Lipsey, M. W., & Wilson, D. B. (2001). Practical meta-analysis. Thousand Oaks, CA: SAGE.Find this resource:

*Lorr, M. (1965). Client perceptions of therapists: A study of therapeutic relation. Journal of Consulting Psychology, 29, 146–149.Find this resource:

Mahrer, A. R. (1997). Empathy as therapist-client alignment. In A. C. Bohart & L. S. Greenberg (Eds.), Empathy reconsidered: New directions in psychotherapy (pp. 187–216). Washington, DC: American Psychological Association.Find this resource:

*Malin, A. J., & Pos, A. E. (2015). The impact of early empathy on alliance building, emotional processing, and outcome during experiential treatment of depression. Psychotherapy Research, 25, 445–459. https://www.doi.org/10.1080/10503307.2014.901572Find this resource:

*Marshall, W. L., Serran, G. A., Fernandez, Y. M., Mulloy, R., Mann, R. E., & Thornton, D. (2003). Therapist characteristics in the treatment of sexual offenders: Tentative data on their relationship with indices of behaviour change. Journal of Sexual Aggression, 9, 25–30.Find this resource:

*Marshall, W. L., Serran, G. A., Moulden, H., Mulloy, R., Fernandez, Y. M., Mann, R. E., & Thornton, D. (2002). Therapist features in sexual offender treatment: Their reliable identification and influence on behaviour change. Clinical Psychology and Psychotherapy, 9, 395–405.Find this resource:

Martin, D. G. (2000). Counseling and therapy skills (2nd ed.). Prospect Heights, IL: Waveland Press.Find this resource:

*Martin, P. J., & Sterne, A. L. (1976). Post-hospital adjustment as related to therapist’s in-therapy behavior. Psychotherapy: Theory, Research and Practice, 13, 267–273.Find this resource:

MacFarlane, P., Anderson, T., & McClintock, A. S. (2016), Empathy from the client’s perspective: A grounded theory analysis. Psychotherapy Research, 27, 227–238.Find this resource:

*Melnick, B., & Pierce, R. M. (1971). Client evaluation of therapist strength and positive-negative evaluation as related to client dynamics, objective ratings of competence and outcome. Journal of Clinical Psychology, 27, 408–410.Find this resource:

*Miller, W., Taylor, C., & West, J. (1980). Focused versus broad spectrum behavior therapy for problem drinkers. Journal of Consulting and Clinical Psychology, 48, 590–601.Find this resource:

*Mitchell, K. M., Truax, C. B., Bozarth, J. D., & Krauft, C. C. (1973, March). Antecedents to psychotherapeutic outcome. NIMH Grant Report 12306. Hot Springs: Arkansas Rehabilitation Research and Training Center.Find this resource:

Mohr, J. J., & Woodhouse, S. S. (2000, June). Clients’ visions of helpful and harmful psychotherapy: An approach to measuring individual differences in therapy priorities. Paper presented at the 31st annual meeting of the Society for Psychotherapy Research, Chicago, IL.Find this resource:

*Moyers, T. B., Houck, J., Rice, S. L., Longabaugh, R., & Miller, W. R. (2016). Therapist empathy, combined behavioral intervention, and alcohol outcomes in the COMBINE research project. Journal of Consulting and Clinical Psychology, 84, 221–229. https://www.doi.org/10.1037/ccp0000074Find this resource:

Moyers, T. B., Manuel, J. K., & Ernst, D. (2015). Motivational interviewing treatment integrity coding manual 4.2.1 (Unpublished manuscript). Retrieved from https://casaa.unm.edu/download/MITI4_2.pdf

*Muller, J., & Abeles, N. (1971). Relationship of liking, empathy and therapists’ experience to outcome in psychotherapy. Journal of Counseling Psychology, 18, 39–43.Find this resource:

*Murphy, D., & Cramer, D. (2014). Mutuality of Rogers’s therapeutic conditions and treatment progress in the first three psychotherapy sessions. Psychotherapy Research, 24, 651–661. https://www.doi.org/10.1080/10503307.2013.874051Find this resource:

(p. 284) Myers, S. (2000). Empathic listening: Reports on the experience of being heard. Journal of Humanistic Psychology, 40, 148–173.Find this resource:

Myers, S. A. (2003). Relational healing: To be understood and to understand. Journal of Humanistic Psychology, 43, 86–104.Find this resource:

Myers, S. A., & White, C. M. (2010). The abiding nature of empathic connections: A 10-year followup study. Journal of Humanistic Psychology, 50, 77–95.Find this resource:

O’Hara, M. M. (1984). Person-centered gestalt: Towards a holistic synthesis. In R. F. Levant & J. M. Shlien (Eds.), Client-centered therapy and the person-centered approach: New directions in theory, research and practice (pp. 203–221). New York, NY: Praeger.Find this resource:

Orlinsky, D. E., Grawe, K., & Parks, B. K. (1994). Process and outcome in psychotherapy—noch einmal. In A. E. Bergin & S. L. Garfield (Eds.), Handbook of psychotherapy and behavior change (4th ed., pp. 270–378). New York, NY: Wiley.Find this resource:

*Orlinsky, D. E., & Howard, K.I. (1967). The good therapy hour: Experiential correlates of patients’ and therapists’ evaluations of therapy sessions. Archives of General Psychiatry, 12, 621–632.Find this resource:

Orlinsky, D. E., & Howard, K. I. (1978). The relation of process to outcome in psychotherapy. In S. L. Garfield & A. E. Bergin (Eds.), Handbook of psychotherapy and behavior change (2nd ed., pp. 283–330). New York, NY: Wiley.Find this resource:

Orlinsky, D. E., & Howard, K. I. (1986). Process and outcome in psychotherapy. In S. L. Garfield & A. E. Bergin (Eds.), Handbook of psychotherapy and behavior change (3rd ed., pp. 311–384). New York, NY: Wiley.Find this resource:

Orlinsky, D. E., Rønnestad, M. H., & Willutzki, U. (2003). Process and outcome in psychotherapy. In M. J. Lambert (Ed.), Bergin and Garfield’s handbook of psychotherapy and behavior change (5th ed., pp. 307–389). New York, NY: Wiley.Find this resource:

*Pantalon, M. V., Chawarski, M. C., Falcioni, J., Pakes, J., & Schottenfeld, R. S. (2004). Linking process and outcome in the community reinforcement approach for treating cocaine dependence: A preliminary report. American Journal of Drug and Alcohol Abuse, 30, 353–367.Find this resource:

Parloff, M. B., Waskow, I. E., & Wolfe, B. E. (1978). Research on therapist variables in relation to process and outcome. In S. L. Garfield & A. E. Bergin (Eds.), Handbook of psychotherapy and behavior change (2nd ed., pp. 233–282). New York, NY: Wiley.Find this resource:

Patterson, C. H. (1984). Empathy, warmth, and genuineness: A review of reviews. Psychotherapy, 21, 431–438.Find this resource:

*Payne, A., Liebling-Kalifani, H., & Joseph, S. (2007). Client-centred group therapy for survivors of interpersonal trauma: A pilot investigation. Counselling & Psychotherapy Research, 7, 100–105.Find this resource:

Peabody, S. A., & Gelso, C. J. (1982). Countertransference and empathy: The complex relationship between two divergent concepts in counseling. Journal of Counseling Psychology, 29, 240–245.Find this resource:

*Peake, T. H. (1979). Therapist-patient agreement and outcome in group therapy. Journal of Clinical Psychology, 35, 637–646.Find this resource:

Pedersen, R. (2009). Empirical research on empathy in medicine—A critical review. Patient Education & Counseling, 76(3), 307–322. https://www.doi.org/10.1016/j.pec.2009.06.012Find this resource:

*Rabavilas, A. D., Boulougouris, J. C., & Perissaki, C. (1979). Therapist qualities related to outcome with exposure in vivo in neurotic patients. Journal of Behaviour Therapy and Experimental Psychiatry, 410, 293–294.Find this resource:

*Ritter, A., Bowden, S., Murray, T., Ross, P., Greeley, J., & Pead, J. (2002). The influence of the therapeutic relationship in treatment for alcohol dependency. Drug & Alcohol Review, 21, 261–268. https://www.doi.org/10.1080/0959523021000002723Find this resource:

(p. 285) *Roback, H. B., & Strassberg, D. S. (1975). Relationship between perceived therapist-offered conditions and therapeutic movement in group psychotherapy with hospitalized mental patients. Small Group Behavior, 6, 345–352.Find this resource:

Rogers, C. R. (1957). The necessary and sufficient conditions of therapeutic personality change. Journal of Consulting Psychology, 21, 95–103.Find this resource:

Rogers, C. R. (1980). A way of being. Boston, MA: Houghton Mifflin.Find this resource:

Sachse, R. (1990a). Concrete interventions are crucial: The influence of the therapist’s processing proposals on the client’s intrapersonal exploration in client-centered therapy. In G. Lietaer, J. Rombauts, & R. Van Balen (Eds.), Client-centered and experiential psychotherapy in the nineties (pp. 295–308). Leuven, Belgium: Leuven University Press.Find this resource:

Sachse, R. (1990b). The influence of therapist processing proposals on the explication process of the client. Person-Centered Review, 5, 321–344.Find this resource:

Sachse, R., & Elliott, R. (2001). Process-outcome research in client-centered and experiential therapies. In D. Cain & J. Seeman (Eds.), Humanistic psychotherapies: Handbook of research and practice (pp. 83–115). Washington, DC: American Psychological Association.Find this resource:

*Saltzman, C., Leutgert, M. J., Roth, C. H., Creaser, J., & Howard, L. (1976). Formation of a therapeutic relationship: Experiences during the initial phase of psychotherapy as predictors of treatment duration and outcome. Journal of Consulting and Clinical Psychology, 44, 546–555.Find this resource:

*Sanberk, I., & Akbaş, T. (2015). Psychological counseling processes of prospective psychological counsellors: An investigation of client-counsellor interactions. Educational Science: Theory & Practice, 15, 859–878.Find this resource:

*Sapolsky, A. (1965). Relationship between patient-doctor compatibility, mutual perceptions, and outcome of treatment. Journal of Abnormal Psychology, 70, 70–76.Find this resource:

*Saunders, S. M. (2000). Examining the relationship between the therapeutic bond and the phases of treatment outcome. Psychotherapy, 37, 206–218.Find this resource:

Saunders, S. M., Howard, K. I., & Orlinsky, D. E. (1989). The Therapeutic Bond Scales: Psychometric characteristics and relationship to treatment effectiveness. Psychological Assessment: A Journal of Consulting and Clinical Psychology, 1, 323–330.Find this resource:

Selman, R. I. (1980). The growth of interpersonal understanding. Orlando, FL: Academic Press.Find this resource:

Shamay-Tsoory, S. (2009). Empathic processing: Its cognitive and affective dimensions and neuroanatomical basis. In J. Decety & W. Ickes (Eds.), The social neuroscience of empathy (pp. 215–232). Cambridge, MA: MIT Press.Find this resource:

Shlien, J. (1997). Empathy in psychotherapy: A vital mechanism? Yes. Therapist’s conceit? All too often. In A. C. Bohart & L. S. Greenberg (Eds.), Empathy reconsidered: New directions in psychotherapy (pp. 63–80). Washington, DC: American Psychological Association.Find this resource:

Smith, M. L., Glass, G. V., & Miller, T. I. (1980). The benefits of psychotherapy. Baltimore, MD: Johns Hopkins University Press.Find this resource:

*Spohr, S. A., Taxman, F. S., Rodriguez, M., & Walters, S. T. (2016). Motivational interviewing fidelity in a community corrections setting: Treatment initiation and subsequent drug use. Journal of Substance Abuse Treatment, 65, 20–25. https://www.doi.org/10.1016/j.jsat.2015.07.012Find this resource:

*Staples, F. R., Sloane, R. D., Whipple, K., Cristol, A. H., & Yorkston, N. (1976). Process and outcome in psychotherapy and behavior therapy. Journal of Consulting and Clinical Psychology, 44, 340–350.Find this resource:

Strauss, C., Taylor, B. L., Gu, J., Kuyken, W., Baer, R., Jones, F., & Cavanagh, K. (2016). What is compassion and how can we measure it? A review of definitions and measures. Clinical Psychology Review, 47, 15–27.Find this resource:

(p. 286) *Strupp, H. H., Fox, R. E., & Lessler, K. (1969). Patients view their psychotherapy. Baltimore, MD: Johns Hopkins University Press.Find this resource:

Tallman, K., Robinson, E., Kay, D., Harvey, S., & Bohart, A. (1994, August). Experiential and non–experiential Rogerian therapy: An analogue study. Paper presented at the American Psychological Association Convention, Los Angeles, CA.Find this resource:

Tausch, R. (1988). The relationship between emotions and cognitions: Implications for therapist empathy. Person-Centered Review, 3, 277–291.Find this resource:

*Thrasher, A. D., Golin, C. E., Earp, J. A. L., Tien, H., Porter, C., & Howie, L. (2006). Motivational interviewing to support antiretroviral therapy adherence: The role of quality counseling. Patient Education and Counseling, 62, 64–71. https://www.doi.org/10.1016/j.pec.2005.06.003Find this resource:

*Truax, C. B. (1966). Therapist empathy, warmth, and genuineness and patient personality change in group psychotherapy: A comparison between interaction unit measures, time sample measures, and patient perception measures. Journal of Clinical Psychology, 22, 225–229.Find this resource:

Truax, C. B., & Carkhuff, R. R. (1967). Toward effective counseling and psychotherapy: Training and practice. Chicago, IL: Aldine.Find this resource:

*Truax, C. B., Carkhuff, R. R., & Kodman, F. Jr. (1965). Relationships between therapist-offered conditions and patient change in group psychotherapy. Journal of Clinical Psychology, 21, 327–329.Find this resource:

Truax, C. B., & Mitchell, K. M. (1971). Research on certain therapist interpersonal skills in relation to process and outcome. In A. E. Bergin & S. L. Garfield (Eds.), Handbook of psychotherapy and behavior change (1st ed., pp. 299–344). New York, NY: Wiley.Find this resource:

*Truax, C. B., Wargo, D. G., Frank, J. D. Imber, S. D., Battle, C. C., Hoehn-Saric, R., . . . Stone, A. R. (1966). Therapist empathy, genuineness and warmth and patient therapeutic outcome. Journal of Consulting Psychology, 30, 395–401.Find this resource:

*Truax, C. B., & Wittmer, J. (1971). The effects of therapist focus on patient anxiety source and the interaction with therapist level of accurate empathy. Journal of Clinical Psychology, 27, 297–299.Find this resource:

*Truax, C. B., Wittmer, J., & Wargo, D. G. (1971). Effects of the therapeutic conditions of accurate empathy, nonpossessive warmth, and genuineness on hospitalized mental patients during group therapy. Journal of Clinical Psychology, 27, 137–142.Find this resource:

*Van der Veen, F. (1967). Basic elements in the process of psychotherapy: A research study. Journal of Consulting Psychology, 31, 295–303.Find this resource:

Wampold, B. E. (2015), How important are the common factors in psychotherapy? An update. World Psychiatry, 14, 270–277. https://www.doi.org/10.1002/wps.20238Find this resource:

Watson, J. C. (2001). Re-visioning empathy. In D. Cain & J. Seeman (Eds.), Humanistic psychotherapies: Handbook of research and practice (pp. 445–471). Washington, DC: American Psychological Association.Find this resource:

Watson, J. C., & Geller, S. M. (2005). The relation among the relationship conditions, working alliance, and outcome in both process-experiential and cognitive-behavioral psychotherapy. Psychotherapy Research, 15, 25–33.Find this resource:

Watson, J. C., & Greenberg, L. S. (2009). Empathic resonance: A neuroscience perspective. In J. Decety & W. Ickes (Eds.), The social neuroscience of empathy (pp. 125–138). Cambridge, MA: MIT Press.Find this resource:

Watson, J. C., & Prosser, M. (2002). Development of an observer rated measure of therapist empathy. In J. C. Watson, R. Goldman, & M Warner (Eds.), Client-centered and experiential (p. 287) psychotherapy in the 21st century: Advances in theory, research and practice (pp. 303–314). Ross on Wye, England: PCCS Books.Find this resource:

*Watson, J. C., Steckley, P. L., & McMullen, E. J. (2014). The role of empathy in promoting change. Psychotherapy Research, 24, 286–298. https://www.doi.org/10.1080/10503307.2013.802823Find this resource:

Wilson, D. B. (2006). Meta-analysis macros for SAS, SPSS, and Stata. Retrieved from http://mason.gmu.edu/~dwilsonb/ma.html

*Wiprovnick, A. E., Kuerbis, A. N., & Morgenstern, J. (2015). The effects of therapeutic bond within a brief intervention for alcohol moderation for problem drinkers. Psychology of Addictive Behaviors, 29, 129–135. https://www.doi.org/10.1037/a0038489Find this resource:

*Woodin, E. M., Sotskova, A., & O’Leary, K.D. (2012). Do motivational interviewing behaviors predict reductions in partner aggression for men and women? Behaviour Research & Therapy, 50, 79–84. https://www.doi.org/10.1016/j.brat.2011.11.001Find this resource:

Wynn, R., & Wynn, M. (2006). Empathy as an interactionally achieved phenomenon in psychotherapy: Characteristics of some conversational resources. Journal of Pragmatics, 38, 1385–1397.Find this resource: