Send email for updates
Peluso P.R., Freund R.R.
Psychotherapy: 2018, 55(4), p. 461–472.
Unable to obtain a copy by clicking title? Try asking the author for a reprint by adapting this prepared e-mail or by writing to Dr Peluso at firstname.lastname@example.org.
Research findings amalgamated for the American Psychological Association show that the outcomes of psychotherapy are substantially and significantly better the more the therapist or especially the client display emotional arousal during therapy – though among substance use clients, focusing on emotions has not been universally helpful.
Summary [Though not specific to clients with drug and alcohol problems, the principles derived from this review of psychotherapy studies are likely to be applicable, partly because severe substance use problems generally form part of a broader complex of psychosocial problems. Addictions work may not necessarily best be conceptualised as psychotherapy, but there is a therapeutic element to it which makes these findings relevant to keyworkers and counsellors.]
Research findings amalgamated for the American Psychological Association related outcomes of psychotherapy to the degree to which therapist or client displayed or reported emotional arousal during therapy.
For both therapist and client the relationship was statistically significant, and for the client especially, substantial, such that greater emotional expressiveness was associated with better outcomes.
Whether these relationships reflected a causal link (suggesting that therapists should promote emotional expression) could not be established by the types of studies included in the analyses, and among substance use clients, focusing on emotions has not been universally helpful.
The featured review is one of several in a special issue of the journal Psychotherapy devoted to features of the therapist–client relationship related to effectiveness, based on the work of a task force established by the American Psychological Association. This particular review synthesised findings on the links between outcomes and the degree to which therapists or clients expressed emotion during therapy.
From a bio-evolutionary perspective, emotions serve a critical survival purpose by providing information about personally meaningful circumstances, and are more intense the greater the perceived personal significance. This information is used to stimulate and guide action to promote one’s self-care.
In psychotherapy, facilitating emotional expression has been seen as an important way to help clients with their emotional problems. Research relating psychotherapy outcomes to emotional expression has only recently been developed, yet convincingly indicates that emotion substantially contributes to clinical outcomes and is a foundation for clinical efficacy. A productive therapy relationship can train clients in adaptive means of experiencing and expressing emotion. Through and within the therapy relationship, clients vicariously experience a model of emotional regulation. Psychotherapists of all theoretical orientations work toward creating productive emotional environments that foster corrective emotional experiences. Unfold the supplementary text for definitions of emotion and related terms including ‘mood’ and ‘affect’, the main focus of the review, seen as a relatively short-lived burst of emotion focused on a specific trigger.
Emotional experiences in therapy may relate to the therapist or the client and may be assessed either on the basis of their responses to questionnaires or by observers. An example of the former is the Positive and Negative Affect Schedule. It consist of 10 adjectives reflecting positive affect (including “active”, “enthusiastic” and “interested”) and 10 reflecting negative affect (including “scared”, “upset” and “irritable”). Clients or therapists rate the intensity of their feelings by assigning each adjective a value from 1 (not at all) to 5 (very much).
An example of an observer-coded system is the Emotional Facial Action Coding System. Certain combinations of movements of parts of the face are designated as expressions of emotions, including happiness, sadness, surprise, fear, anger, disgust, and contempt. For example, happiness is recognised by the raising of the upper eyelids plus raising of the cheeks. These actions are also rated for intensity.
The featured review incorporated two meta-analyses amalgamating findings on the relationship between emotional expressions during therapy and the outcomes of that therapy – one relating to the therapist’s emotions, the other those of the client. Reviewers sought formally published studies of ‘real’ (rather than simulated) individual (rather than group) therapies whose results had been published in English, and which reported on the emotion/outcomes link in a way which enabled their findings to be aggregated with those from other studies. The aim was to assess the overall strength of the links between emotional arousal and outcomes, and to explore possible influences on the strength of the links found in different studies. Strength was calculated as a correlation coefficient, an expression of the degree to which outcomes co-varied with emotional arousal. The chosen metric ranges from -1 (perfect negative co-variation, meaning that as one side of the link gets larger the other diminishes to the same degree) to +1 (perfect positive co-variation, meaning that as one side of the link gets larger so does the other, and to the same degree). The strength of the relationships between emotions and outcomes was also expressed as an effect size. Effectively these metrics indicate how influential emotional arousal had been if causally linked to outcomes.
Both therapist and client emotional expressiveness were significantly and positively related to better outcomes. The client-focused link was particularly strong, especially when clients’ emotional expressions were rated by observers. Details below.
Across the 13 articles reporting on the relationship between therapist affect and outcomes, the link between the two amalgamated to a statistically significant correlation of 0.28, equating to a medium effect size of 0.56. This link indicates that better outcomes are recorded when the therapist has been seen by themselves or by observers as emotionally aroused. In two ways the finding was robust. Firstly, the individual studies did not vary significantly in their findings; secondly, another 23 missed studies which recorded a zero correlation would have been needed before the aggregated result dipped below practical significance.
Across the 42 articles reporting on the relationship between client affect and outcomes, the link between the two amalgamated to a statistically significant correlation of 0.40, equating to a large effect size of 0.85. This link indicates that better outcomes are recorded when the client has been seen by themselves or by observers as emotionally aroused. The finding was robust in the sense that another 138 missed studies which recorded a zero correlation would have been needed before the aggregated result dipped below practical significance. However, individual studies varied significantly in the strengths of the links they recorded.
At 0.45, the correlation between emotional arousal and outcomes was greater when observers had rated emotion based on videos or transcripts. At 0.20, it remained statistically significant but substantially and significantly weaker when in retrospect the client had rated their own emotional arousal.
The results of the two meta-analyses indicate that expression of affect during psychotherapy sessions by the therapist or by the client is at least moderately related the outcomes of that therapy. Whether these links are due to a causal relationship cannot be established by the types of studies included in the analyses. However, the results provide a reason to investigate causality.
Outcomes were more strongly associated with the client’s emotional expression than that of the therapist, suggesting that client affect is more important in relation to treatment outcomes. It has been found that clients who do not process or ‘get in touch’ with their emotions tend to elicit negative therapist reactions and have worse clinical outcomes.
The much weaker correlation between outcomes and the client’s emotional arousal as rated by the client rather than by an observer, must be seen in the context of the way these ratings are collected – typically in retrospect after the session has ended. By this stage clients may be unaware of the scope of their emotional expressions, or have processed the emotion so it no longer has the same relevance or power. In contrast, trained observers watching a video recording may pick up salient cues (facial displays, tone of voice, and elements of speech) not apparent to clients. These cues are potentially important for therapists to attend and respond to.
Limitations of the research include the exclusion of studies of group, couple, and family therapies, of unpublished work, and of non-English language articles, though at least half the studies were conducted by European teams. Perhaps the biggest limitation was the relative lack of research on how therapists might be able to improve outcomes for their clients by facilitating productive expression and processing of emotion.
Amalgamated research findings show that client or therapist expression of emotion during psychotherapy sessions is strongly predictive of good outcomes. Even without hard evidence of a causal link, the following conclusions and practices safely be advanced.
• Emotion matters. Clients benefit when practitioners find opportunities to facilitate client expression and processing of emotion in therapy rather than trying to control or even discourage it. Findings reinforce the idea of psychotherapy as a ‘crucible’ that contains emotionally charged reactions without being compromised by them.
• Recent findings suggest that suppressing emotions adversely affects therapeutic outcomes. Together with the featured analyses, this suggests therapists should avoid expressions (such as criticism, dogmatic interpretations, or inflexibility) which provoke defensive emotional reactions in clients.
• Neither should therapists themselves avoid displaying emotion. Such reactions facilitate the therapeutic relationship and are predictive of good treatment outcomes. Perhaps having learnt this, experienced therapists tend to display emotion more than those less experienced.
• Therapists can consider preparing their clients to experience emotions, placing such experiences in a productive context. Researchers have found that “clients of therapists who emphasized affect experienced greater affect”. Given the featured review’s findings, this strategy is an important consideration for therapists.
• Therapists can learn and practice coaching as opposed to dismissing emotion. In therapy, emotion needs to be focused on, validated, and worked with directly to promote emotional change.
• Therapists can work toward fostering productive and corrective emotional experiences with clients. In the context of a safe, trusting relationship, the skills to understand and resolve an emotional experience can become internalised by the client into strategies to regulate the experience and expression of their emotions. In this context, facilitating emotional expression becomes one of the therapeutic tasks, and eliciting meaning from and resolving emotional reactions become therapeutic goals.
• Therapists can construct processes for getting accurate, real-time feedback on emotion in psychotherapy. This information can then be used to create a feedback loop to practitioners (during or immediately after a session) to guide them in tailoring therapy or focusing on certain affective elements which signal progress.
commentary Relative to studies of interventions themselves, or of other facets of the therapeutic relationship analysed in the series of which the featured article forms a part (listed below), the association between observed emotional expressiveness of clients during psychotherapy and how well they progress is unusually strong, while that between therapist expressiveness and outcomes rivals that of several other facets. In their recommendations, the reviewers clearly consider it prudent to assume that these links arise from a causal relationship – that facilitating emotional expression causes better outcomes. Such a link is not only supported by the strength of the findings but also by its theoretical and common-sense credentials; a bottled-up client, emotionally withdrawn from therapy, seems less likely to benefit than one who offers emotional material and insights into their condition which the therapist can work with.
However, the reviewers also cautioned that their findings could not establish causality, merely that they are consistent with this hypothesis. If in reality there is no causal link, the practice recommendations would be nullified. It seems possible, for example, that clients who before therapy more freely express their emotions would do better in any event, regardless of the therapist’s attempts to elicit or dampen their reactions. The same patients may also elicit more emotional reactions in their therapists. In these scenarios, emotional expressiveness of both client and therapist would remain associated with better outcomes, but not because expressiveness helped cause these improvements. Without effectively random allocation of patients to more or less emotionally provoking therapies or therapists, alternative explanations of a link between expressiveness and outcomes cannot be eliminated. Even then, it would seem difficult to construct therapies or to find/train therapists differing only in their provoking of client emotions, and not also in other ways which might instead have been the active ingredients in affecting outcomes.
Underlying the overall averages reported in the featured review are variations in the therapeutic value of arousing different kinds of emotions in different kinds of situations, and differences in how different people react to therapies which heighten or focus on emotions. The featured review itself warns therapists against provoking defensive emotional reactions in clients, because these will block the expression of more productive emotions.
In the substance use sector we also have evidence that for some people, focusing on emotion in therapy is counterproductive. From the little we know, it seems that high levels of depressive symptoms or low levels of emotional distress call for therapies which defocus from emotions, while patients at the opposite ends of these dimensions may do better in more emotion-focused therapies. However, rather than simply matching one dimension on which clients vary to one dimension of therapy, dimensions of therapy such as directiveness and the degree to which the focus is on feelings rather than actions interact with multiple client variables, demanding an approach which matches a multidimensional client profile to a multidimensional therapeutic mix. Unfold the supplementary text for more on three relevant studies.
Listed below are analyses of the other reviews commissioned by the American Psychological Association task force.
Overall conclusions based on the other 16 reviews
The ‘real’, person-to-person relationship
Alliance in couple and family therapy
Alliance in child and adolescent therapy
Cohesion in group therapy
Goal consensus and collaboration
Therapist self-disclosure and ‘immediacy’
Repairing ruptured alliances between therapists and clients
Treatment outcome expectations
Feeding back client progress data therapists
Last revised 28 March 2019. First uploaded 06 March 2019
Give us your feedback on the site (two-minute survey)
Open Effectiveness Bank home page
Add your name to the mailing list to be alerted to new studies and other site updates
REVIEW 2018 Congruence/genuineness: a meta-analysis
REVIEW 2018 Cohesion in group therapy: a meta-analysis
REVIEW 2018 Psychotherapy relationships that work III