TY - JOUR
T1 - What do these sounds tell us about the therapeutic alliance
T2 - Acoustic markers as predictors of alliance
AU - Nof, Aviv
AU - Amir, Ofer
AU - Goldstein, Pavel
AU - Zilcha-Mano, Sigal
N1 - Publisher Copyright:
© 2020 John Wiley & Sons, Ltd.
PY - 2021/7/1
Y1 - 2021/7/1
N2 - Predicting the trajectories of alliance formation that the patient is likely to establish with the therapist during treatment, even before their first meeting, can help prevent the potentially harmful consequences of deterioration in alliance, such as poor outcome and premature dropout. The present study aimed to examine the ability of four pretreatment acoustic markers to predict the alliance that is likely to be formed in the course of treatment: F0 span, speech rate, pause proportion and jitter. Data from 560 observations of 38 patients were collected as part of an ongoing randomized clinical trial of short-term psychotherapy for major depressive disorder. The acoustic markers were measured using high-quality recordings at baseline, before the patient and therapist ever met or had any type of communication. A multilevel model was used to examine the ability of the four acoustic markers to predict the slopes of alliance formation in the course of treatment, all markers being introduced in the same model. The clinical utility of the acoustic markers was explored in two case studies. The model explained 22% of the variance in alliance formation. Higher levels of both jitter and pause proportion at baseline predicted less strengthening of the alliance in the course of treatment. The findings, which should be replicated in larger samples, suggest that much of the therapeutic alliance can be predicted based on the acoustic characteristics of the patient's voice in the first 3 min of their intake, before they even meet their therapist.
AB - Predicting the trajectories of alliance formation that the patient is likely to establish with the therapist during treatment, even before their first meeting, can help prevent the potentially harmful consequences of deterioration in alliance, such as poor outcome and premature dropout. The present study aimed to examine the ability of four pretreatment acoustic markers to predict the alliance that is likely to be formed in the course of treatment: F0 span, speech rate, pause proportion and jitter. Data from 560 observations of 38 patients were collected as part of an ongoing randomized clinical trial of short-term psychotherapy for major depressive disorder. The acoustic markers were measured using high-quality recordings at baseline, before the patient and therapist ever met or had any type of communication. A multilevel model was used to examine the ability of the four acoustic markers to predict the slopes of alliance formation in the course of treatment, all markers being introduced in the same model. The clinical utility of the acoustic markers was explored in two case studies. The model explained 22% of the variance in alliance formation. Higher levels of both jitter and pause proportion at baseline predicted less strengthening of the alliance in the course of treatment. The findings, which should be replicated in larger samples, suggest that much of the therapeutic alliance can be predicted based on the acoustic characteristics of the patient's voice in the first 3 min of their intake, before they even meet their therapist.
KW - acoustic markers
KW - jitter
KW - pause proportion
KW - psychotherapy processes
KW - therapeutic alliance
UR - http://www.scopus.com/inward/record.url?scp=85098082811&partnerID=8YFLogxK
U2 - 10.1002/cpp.2534
DO - 10.1002/cpp.2534
M3 - ???researchoutput.researchoutputtypes.contributiontojournal.article???
C2 - 33270316
AN - SCOPUS:85098082811
SN - 1063-3995
VL - 28
SP - 807
EP - 817
JO - Clinical Psychology and Psychotherapy
JF - Clinical Psychology and Psychotherapy
IS - 4
ER -