A tendency in which represented aspects of important and formative relationships (e.g., with parents) are attributed to other people. This process of transference can be conscious but is typically unconscious (Levy and Scala 2012). Originally conceived by Sigmund Freud, transference is believed to be a central component in the therapeutic relationship between therapist and patient, and although there may be real aspects in the relationship, transference represents a distortion or cognitive-affective bias. In psychodynamic and psychoanalytic therapies, the tracking and/or analysis of the transference is believed to be important in the treatment. Within any interaction, there are individual differences in transference in terms of the degree, extent, rigidity, and awareness of transference. Transference can be reality based in that it is based on aspects of the individual or the situation that can pull for transference. It can also be evoked, that is, people can act in ways to elicit behaviors from others that are consistent with their transference, and the amount of transference can vary as a function of the individual, the target, and the situation. Lastly, an important feature of transference is that some aspects are not only unconscious but are related to conflicts and defensive processes.
Introduction and Brief Historical Overview
The concept of transference has been central within psychoanalysis (see “Psychoanalysis”) and psychodynamic psychotherapy (see “Psychodynamic Perspective”) since Sigmund Freud’s earliest writings and increasingly has been utilized broadly across multiple psychotherapy orientations. However, transference, and the use of transference interpretations in psychotherapy, has also been a highly controversial topic that has evolved over time. Transference first appeared in Freud’s neurological writings in 1888 (Freud 1888) and evolved over years of creative synthesis rooted in the discourses of Freud’s time. Freud initially conceptualized transference as “displaceable energies” to indicate the transfer of strong feelings developed within a particular relationship to another person who was independent of the origin of those feelings and then later, in Case Studies in Hysteria (Breuer and Freud 1895), as a “false connection” where the patient transferred unconscious ideas about a figure from the past onto the person of the physician. Freud noted that these ideas often melted away with the conclusion of treatment. Though the concept continued to evolve throughout Freud’s life, by 1900, transference was ready to stand as it does today at the core of psychoanalytic theory.
In his discussion of the famous case of Dora, Freud (1905) noted that Dora was constantly comparing him with her father, even at a conscious level, and he began to recognize that there may be real aspects of the figure that the transference is projected onto, the physician or the psychotherapist in this case, that might actually pull for or allow for the experience of transference. Later, Freud (1912) noted that some aspects of transference are conscious and changeable, whereas others are unconscious and relatively impervious to development or change. Freud also posited that the most serious difficulties the psychotherapist must grapple with lie in the management of transference and that the resolution of transference was synonymous with the resolution of neurosis. Just prior to his death, Freud described transference as the central mechanism of therapeutic change and saw the central task in psychoanalysis as the establishment, interpretation, and resolution of transference. The concept of transference continued to evolve after Freud’s death and into its present-day conception with contributions from Melanie Klein (1952), Gill (1982), and others.
Empirical Evidence for the Concept of Transference
There are two essential areas of research that have produced evidence for the concept of transference: basic cognitive and social psychological research and psychotherapy process research.
Basic Cognitive and Social Psychological Research
Much of the basic cognitive and social psychological research on transference has centered around the work of Susan Andersen and colleagues, who have developed an elegant paradigm for assessing transference phenomena (for details, see Andersen et al. 2012). This research has consistently shown that significant-other mental representations are activated and applied, or transferred, to a novel target (e.g., new person) in everyday social perception. That is, people tend to view others in ways that are consistent with preexisting significant-other mental representations. Further, the social cognitive research on transference indicates that these significant-other mental representations are chronically accessible and in a continual state of readiness for use and do not require priming. These mental representations are applied to new people even when there is no concrete similarity between them and the effect is enhanced or amplified when primed and persists and is exacerbated over time.
From an attachment theory perspective (see “Attachment Theory”), there is evidence for a general transference process wherein people apply their attachment representations of past romantic partners both to those that resemble their past partner as well as to those that do not resemble their past partner. However, people do so to a greater degree when the others resemble their past partner, indicating a specific transference process. Importantly, research suggests that people tend to feel more anxious and less avoidant, or less defensive, toward others that resemble their past partners.
Taken together, these findings suggest that transference is influenced by aspects of the person, that some of it is reality based, and that there is both a general and specific transference process. But most importantly these findings, as well as others, indicate that individual differences in attachment-based defensive processes are related to transference phenomena in ways consistent with a psychodynamic transference process model, and not simply a cognitive- or information-processing model. Specifically, working models of attachment relationships are not only transferred to new relationships in both general and specific ways that are consistent with a social cognitive or information-processing model, but there are individual differences as a function of attachment anxiety, avoidance, and security that are indicative of conflict and defensive processes (see “Defense Mechanisms”) suggestive of a psychodynamic transference process model.
Psychotherapy Process Research
A number of research projects have noted the relationship between the narratives of others and the narratives of therapists. These studies generally find similarity between the two sets of narratives and that this relationship becomes stronger over time. However, the effect associated with these findings is rather modest, it is only present for a subset of patients, and even when present, there is a high degree of variability in the amount of transference shown. Further research has shown that conflicts about parents are worked out in the psychotherapy process through one’s sense or representation of the therapist, and that this may be in part because the therapist is seen as the safest person to do so with. Furthermore, as these conflicts are worked through, the patient can tolerate more accurate representations of their self and important others and see people, including themselves, more accurately. This research, while not definitive, would also suggest a psychodynamic transference process.
Analysis of Transference and Transference Interpretations
The focus on transference phenomena within the therapeutic relationship in psychoanalysis has led to an emphasis on the analysis of the transference and the use of transference interpretations in this context. Transference interpretations focus on connecting the patient’s feelings and behaviors that are occurring in the here and now of the therapy with regard to the therapist with the patient’s pre-conceived representational models of significant others. Most prototypically, a transference interpretation is a tactful comment that clarifies and links the patient’s experience of others outside of therapy with that of the therapist in therapy and to the patient’s experience of past relationships with caregivers.
The use of transference interpretations in psychotherapy, though initially considered the sine qua non of the psychodynamic approach, has been controversial. Over the years, technical developments within psychoanalysis have increasingly stressed the importance of the therapeutic relationship, corrective emotional experience, and implicit relational knowing as agents of change. Further, early correlational research suggested that the use of transference interpretations was related to poorer outcome, particularly for the most seriously disturbed patients with personality pathology (see “Personality Disorder”). These findings led many psychodynamic theorists to deemphasize the use and importance of transference interpretations. Some clinical writers have gone even further, arguing that transference interpretations are often experienced by patients, particularly those with personality pathology and disturbed relatedness, as hostile and attacking. Nonetheless other clinicians continued to see transference interpretations as a valuable clinical tool.
Transference Interpretations in Psychotherapy Research
A number of studies have examined transference interpretations, including correlational studies that have looked at narratives, process research, outcome research, as well as quasi experimental studies and experimental studies using randomized controlled trials (RCT). Overall, the results from these studies suggest a complicated relationship between the use of transference interpretations and outcome. Transference interpretations may result in substantial increases in the patient’s ability to collaborate and positive outcomes, but they also may produce marked decreases in collaboration and lead to worse outcomes. For this reason, transference interpretations have been described as a high-risk high-yield intervention. Clinical theory suggests that transference interpretations are most effective when a series of preparatory interventions by the therapist create a climate in which the patient can accept the therapist’s observation. These interventions are thought to provide a buffer for the patient from the sharpness of a transference interpretation. Consistent with these clinical observations, the data from the First Experimental Study of Transference (FEST; Høglend et al. 2006) indicate that patients with high levels of quality object relations (see “Object Relations Theory”) may benefit from low to moderate levels of transference interpretations but have difficulty with high levels, whereas patients with low levels of quality object relations may tolerate only low levels of transference interpretations. Furthermore, the accuracy of the interpretation as well as the therapeutic process (e.g., accepting/nonjudgmental, non-hostile) in which it is embedded is also important and related to outcome, and this may be true particularly when the therapeutic alliance is weak. The accepting/nonjudgmental, non-hostile attitude and the use of preparatory comments and the tactful delivery of an accurate interpretation may foster the alliance such that poor alliances become stronger alliances. Finally, the mechanism by which a transference-based treatment may be making an impact, and leading to better outcome, may be through the development of insight (see “Insight”).
Transference-Focused Psychotherapy (TFP)
As noted, transference interpretations may be a highly useful treatment tool, especially for patients who were thought to be most negatively affected by them, those with poor quality object relations and personality disorders. TFP (see “Transference-Focused Psychotherapy” and “Personality Disorder Treatment”; Clarkin et al. 2006) is a modified psychodynamic psychotherapy designed specifically for individuals with personality disorders, particularly borderline personality disorder. It is a structured, twice weekly, outpatient treatment based on Otto Kernberg’s object relations model and has been found to be efficacious in multiple independent studies.
Good experimental evidence has emerged from both social psychological findings and psychotherapy process and outcome research that supports the concept of transference. The conceptualization of transference outlined here is consistent with what is known about schemas and pattern matching, implicit memory processes, and other concepts from cognitive and neuroscience. However, there is also evidence that transference is not just a cognitive-information bias or process but that it is also a dynamic process related to attachment and defensive processes. Nevertheless, these findings are preliminary and need to be confirmed in future research. For a number of years now, the prevailing view has been that transference interpretations are potentially harmful, that one needs to establish a strong working alliance first, and that they are not useful early in treatment. Most importantly, transference interpretations have been seen as particularly problematic for those with low quality of object relations or personality disorders. However, recent experimental data using RCT designs suggest that low to moderate levels of transference interpretation and transference-based treatments are highly effective and that these treatments can lead to structural change or change in social cognitive processing. In fact, this is especially true for those with low quality object relations and personality disorders. Despite continued controversy, most psychodynamic theorists recognize the importance of transference and attending to it in a therapeutic context.
- Breuer, J., & Freud, S. (1895). Studies on hysteria. In J. Strachey (Ed.), The standard edition of the complete psychological works of Sigmund Freud (Vol. 2). London/ England: Hogarth Press.Google Scholar
- Clarkin, J. F., Yeomans, F., & Kernberg, O. F. (2006). Psychotherapy of borderline personality: Focusing on object relations. Washington, DC: American Psychiatric Press.Google Scholar
- Freud, S. (1888). Hysteria. In J. Strachey (Ed.), The standard edition of the complete psychological works of Sigmund Freud (pp. 41–57). London: Hogarth Press Inc.Google Scholar
- Freud, S. (1905/1963). Dora: An analysis of a case of hysteria. New York: Simon & Schuster, Inc..Google Scholar
- Freud, S. (1912). The dynamics of transference. In S. J. Ellman (Ed.), Freud’s technique papers: A contemporary perspective (pp. 35–50). Northvale: Jason Aronson Inc..Google Scholar
- Gill, M. M. (1982). Centrality of the analysis of transference. In M. M. Gill (Ed.), Analysis of transference (Vol. 1, pp. 43–58). New York: International University Press.Google Scholar
- Klein, M. (1952). The origins of transference. International Journal of Psychoanalysis, 26, 11–33.Google Scholar