Objectives: Most studies comparing face-to-face to computer-based interventions focused on their overall efficacy, neglecting to explore the difference between common and specific factors that compose such interventions. Our study is exploratory, designed to focus on these basic therapeutic factors by comparing the performance of an internet-based software which simulates therapeutic interventions with a trained human therapist. Method: Two client-therapist interactions, one with a real therapist and one with a computer agent (the "Eliza program"), were both rated by 138 real-life professionals by use of a survey. The survey tapped into aspects relating to both performance and the quality of the therapeutic relationship. Results and conclusions: The perceived difference between the "Eliza" program and the human therapist seemed to lie in the quality of the performances, and not in some intrinsic features of either of the two. The evaluators predominantly found the human therapist to perform better on all the dimensions taken into account. Interestingly enough, what seemed to have a selective impact was the form of therapy they declared to practice. In this regard, the therapists that considered themselves CBT practitioners discriminated more clearly the internet-based program from the trained human therapist on dimensions related to specific factors (e.g., correct approach of the problem) and the overall performance, but not on dimensions related to common factors (e.g., empathy).
Can you tell the difference? comparing face-to-face versus computer-based interventions. The "Eliza" effect in psychotherapy
Cristea I. A.;
2013
Abstract
Objectives: Most studies comparing face-to-face to computer-based interventions focused on their overall efficacy, neglecting to explore the difference between common and specific factors that compose such interventions. Our study is exploratory, designed to focus on these basic therapeutic factors by comparing the performance of an internet-based software which simulates therapeutic interventions with a trained human therapist. Method: Two client-therapist interactions, one with a real therapist and one with a computer agent (the "Eliza program"), were both rated by 138 real-life professionals by use of a survey. The survey tapped into aspects relating to both performance and the quality of the therapeutic relationship. Results and conclusions: The perceived difference between the "Eliza" program and the human therapist seemed to lie in the quality of the performances, and not in some intrinsic features of either of the two. The evaluators predominantly found the human therapist to perform better on all the dimensions taken into account. Interestingly enough, what seemed to have a selective impact was the form of therapy they declared to practice. In this regard, the therapists that considered themselves CBT practitioners discriminated more clearly the internet-based program from the trained human therapist on dimensions related to specific factors (e.g., correct approach of the problem) and the overall performance, but not on dimensions related to common factors (e.g., empathy).Pubblicazioni consigliate
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.