An Integrative Review of Therapeutic Empathy

The Complex Nature of Therapeutic Empathy

Therapeutic empathy has long been identified as a particularly robust predictor of outcome (e.g., Elliot et al., 2018; Lafferty et al., 1989; Luborsky et al., 1988), yet its complexity has made it difficult to operationalize. Historically, some theorists have emphasized the sensory-emotional components (Kohut, 1959; Titchener, 1915), while others have emphasized the cognitive-rational components (Rogers, 1959). In a paradigm shift, however, contemporary theories have pointed to the empathic process (e.g., Rogers, 1975). The most emergent theories, among these, have emphasized the therapist’s capacity to dialectically shift between states of emotional resonance and co-regulation (Holmes & Slade, 2018). Psychotherapy research has yet to validate what appears to be a therapeutic “empathic dialectic,” though social neuroscience research has acknowledged the complexity of empathy, pointing to neurobiological events undergirding its component parts (e.g., Dana, 2018; Decety & Lamm, 2009). The purpose of this paper is to review definitions of therapeutic empathy, emphasizing those that have been supported by contemporary theory and research.

The History of Therapeutic Empathy

Empathy as an affective state

The concept of empathy originated with the German word einfühlung, defined as the projection of oneself into the objects of one’s perception, which gives way to a certain form of aesthetic appreciation (Lipps, 1905, as cited in Wispé, 1987). Lipps (1905) used einfühlung to describe how one comes to know another person, highlighting the role of motor mimicry, and the body’s resultant afferent feedback (as cited in Wispé, 1987). Borrowing and translating this term, Titchener (1915) coined the term empathy, as “the natural tendency to feel ourselves into what we perceive or imagine… [empathic ideas] are the converse of perceptions; their core is imaginal, and their context is made up of sensations” (p. 198). The stance of these earliest theorists, who emphasized the role of affective resonance in empathy, was partially commensurate with two psychotherapy theorists who rose to prominence thereafter: Heinz Kohut and Carl Rogers.

Therapeutic empathy as a developmental capacity

Kohut (1959) posited that empathy is a mode of psychoanalytic observation that involves “vicarious introspection,” or imagining what it would be like to be the patient as the patient reflects on their experience. Kohut also claimed that empathy is “as basic an endowment as… vision, hearing, touch, taste, and smell” (1977, p. 144), and this view of empathy was criticized for capturing only a “primitive form of empathic understanding” (Feshbach, 1987, p. 275). Kohut (2010) rebutted this and other criticisms, elucidating what he felt he had already made clear in his seminal paper from 1959: that the type of empathy therapists are capable of experiencing is contingent on their developmental level. Kohut did not go into much greater detail to describe empathy from the therapist’s perspective, since he ultimately remained tethered to a one-person model of psychotherapy, emphasizing the therapist’s role in uncovering the patient’s unconscious thoughts (Greenberg & Mitchell, 1983).

Therapeutic empathy as a multidimensional state or process

As opposed to Kohut, Rogers did not have the same allegiance to classical psychoanalysis and was able to break free from the one-person theory of psychology that confined Kohut. Rogers (1959) thus highlighted the therapists’ subjectivity in therapeutic empathy, emphasizing how important it is that therapists acknowledge the emotional separation between themselves and their patients, in order to avoid over-identifying with them. Later, Rogers (1975) contended that therapeutic empathy is best described as a multi-dimensional process, which means “being sensitive, moment by moment, to the changing felt meanings which flow in this other person… and sensing meanings of which he or she is scarcely aware (p. 142).” At the time, this perspective was radical because it initiated a focus away from the one-person model of psychotherapy and toward a two-person approach, where the therapist was considered another subject in the relationship with the client-subject (Aron, 1992; Benjamin, 1992).

Contemporary relational theorists have embraced this intersubjective viewpoint, which has enabled them to address the unconscious field between the therapist and the patient, or what Ehrenberg (1992) has called “the intimate edge.” From this perspective, Buechler (2008) has described the empathic process, suggesting that the therapist first “feel into” or affectively resonate with the patient’s experience, and then “feel out of” that experience, becoming aware of the patient’s emotions as distinct from their own. Like Ehrenberg (1992), Buechler (2008) described the tenuous boundary that can exist between self and other, and emphasized the importance of therapists’ self-regulation (i.e., self-reflection and interpersonal boundaries), which allows them to “emerge with unusual readiness to hear… the [patient’s] material” (p. 45). 

Theories of Psychotherapy Process

Theories focused on the working alliance, define this as “agreements on the therapeutic goals; consensus on the tasks that make up therapy; and the bond between the patient and therapist” (Horvath et al., 2011, p. 10). This emphasizes a therapists’ capacity to respond flexibly to the needs of each patient, and in doing so, they highlight aspects of therapists’ self-regulation. 

For example, Greenberg (2015) recommends that therapists’ use empathy to follow their patients’ emotional experience to build the alliance in the early phases of treatment, and that therapists use their emotional awareness (an aspect of self-regulation) to challenge patients’ emotional schemas in the later phases. Yet, as illustrated by Buechler (2008) and other theories, it seems therapeutic empathy involves both following and challenging patients.

Throughout the psychotherapy process, and particularly when the therapist challenges the patient, ruptures – defined as “negative shifts in the quality of the therapeutic alliance or ongoing problems in establishing one” (Safran, 1993, p. 34) – are thought to be inevitable. To effectively (i.e., sensitively) address ruptures, Safran and Muran (2000) suggest that therapists consider themselves participant-observers, shifting their attention to the moment-to-moment intersubjective negotiations between themselves and their patients. By using “mindfulness-in-action” to explore these relational enactments, therapists can then, in turn, repair ruptures in the alliance.

Attachment theorists, Holmes and Slade (2018), embrace a similar view of the psychotherapy process. They emphasize therapists’ mentalization or capacity to understand and perhaps, put words to the states underlying the interpersonal experiences of themselves and their patients. Similar to mindfulness-in-action, mentalization serves to coregulate the patients’ arousal. These theories of the psychotherapy process point to an empathic dialectic: therapists’ capacity to shift from emotional resonance to co-regulation, depends on their own self-regulatory skills (i.e., mindfulness-in-action and mentalization), and works to co-create new relational possibilities with (and for) their patients. 

Empirical Validation for the Empathic Dialectic

Empirical research has validated aspects of the empathic dialectic in several studies, which have used different measures of empathy. The Interpersonal Reactivity Index (i.e., IRI; Davis, 1983a) is a self-report measure which includes affective and cognitive dimensions. The affective dimensions include empathic concern, or the tendency to experience other-oriented feelings of warmth and compassion for others, and personal distress empathy, or the tendency to experience self-oriented feelings of discomfort in reaction to other’s distress. Referencing these affective dimensions, Davis (1983b) found that compared to personal distress empathy, empathic concern is more likely to lead to helping behavior. 

Researchers have since compared therapists to non-therapists on these affective dimensions of empathy. In a matched sample of 19 therapists and 19 control subjects, Hassenstab et al. (2007) found therapists are less likely to experience personal distress empathy, compared to non-therapists. Hall et al. (2012) later administered the IRI to a sample of 290 psychologists, divided into practitioners and scientists. Their findings suggested that practitioners were more likely to endorse empathic concern, and (again) less likely to endorse personal distress empathy. Since personal distress empathy represents an over-identification with another’s state, it makes sense that it would be less common among therapists, whose training involves acknowledging the emotional separation between themselves and their patients (per Rogers, 1959). Yet, to claim the other extreme; that personal distress empathy is completely unrelated to patient care (Hojat et al., 2005) seems misguided. 

Indeed, research has shown that personal distress empathy cannot be extricated from other forms of empathy, methodologically or in practice. Chrysikou and Thompson’s (2016) factor analysis failed to differentiate the cognitive and affective dimensions of the IRI. Drawing on these results, they suggested:

Items that are considered to capture affective empathy… require the individual to use cognitive empathy to put herself in a situation before responding. Essentially, in this scale, cognitive empathy acts as a gatekeeper to the accurate measurement of affective empathy (p. 6).

It follows that the cognitive components of empathy (perspective-taking) coexist with the emotional components (personal distress empathy). Therefore, a therapist who is likely to engage in perspective-taking, is also likely to resonate with their patients’ internal states (i.e., to experience personal distress empathy).

Social neuroscience research supports the idea that empathy is a multidimensional process and that a certain degree of affective resonance (or personal distress empathy) is likely to be triggered alongside other forms of empathy. Administering functional magnetic resonance imaging (fMRI), behavioral, and self-report measures of empathy to non-therapists, Decety and Lamm (2009) found that several discrete neural networks are involved in empathizing with another’s pain. Some are linked to automatic self-oriented processes marked by emotional contagion (e.g., personal distress empathy) and others are linked to deliberate other-oriented processes marked by perspective-taking (e.g., empathic concern). 

Commensurate with contemporary psychotherapy theories (Buechler, 2008; Holmes & Slade, 2018; Safran & Muran, 2000), Decety and Lamm (2009) similarly contend that emotion regulation is indispensable, because it tones down self-oriented states like emotional contagion (i.e., personal distress empathy) so that other-oriented states of empathy and compassion (i.e., empathic concern) may emerge. In turn, this facilitates co-regulation as the same neurobiological systems that control self-regulation are also responsible for sending out “cues of safety” (e.g., a genuine smile, or compassionate response) (Dana, 2018). Through co-regulation, the brains join as a single healing system (Hasson et al., 2012).    

Conclusions, Implications, and Recommendations

Historically, psychotherapy theorists have emphasized the sensory-emotional (Kohut, 1959) or cognitive-rational components of empathy (Rogers, 1959), while theorists now define therapeutic empathy as a process (e.g., Buechler, 2008; Holmes & Slade, 2018). Drawing on contemporary theories and research, the current paper suggested the phrase “empathic dialectic” to refer to therapists’ ability to shift from states of emotional resonance and co-regulation, in order to effectively co-create corrective emotional experiences with (and for) their patients. 

Consistent with American Psychological Association’s aim to develop “clinical competencies” for the accreditation of clinical psychology doctoral programs (Fouad et al., 2009), we recommend the empathic dialectic be added to the larger literature on clinical competencies. As this paper demonstrated, therapists’ can enhance their capacity to effectively navigate the empathic dialectic by developing more efficient self-regulation. Therefore, we advocate for therapists’ participation in relational supervision (Sarnat, 2012) and personal therapy (Orlinsky et al., 2005), which tend to enhance self-regulatory skills.