The First Step Towards Robotic Counselors

I wrote a year or so ago about early trials with virtual therapists to support people with PTSD.  It found that soldiers were more likely to open up when speaking with a virtual avatar than when talking to a human, or indeed when taking a survey. The researchers believe the avatar provides the advantages of anonymity whilst also providing a degree of social connectivity.

A recent study from the University of Plymouth suggests similar results could be achieved by a physical robot.  As with the PTSD research, users reported finding the NAO robot to be non-judgemental as it delivered the motivational interview, and was therefore effective in supporting behavioral change.

Motivational interviews (MI) are a commonly used technique to support behavioral change.  It’s designed to encourage the patient to talk about their need and reasons for changing, and therefore the primary role of the counsellor is to evoke that conversation.

Articulating your goals

The results suggest that the robot was very effective in that role as it encouraged participants to talk about their goals and dilemmas for increasing their physical activity.  It achieved this via a relatively simple script it was programmed with to lead the conversation during the hour long session.

“We were pleasantly surprised by how easily the participants adapted to the unusual experience of discussing their lifestyle with a robot,” the researchers say. “As we have shown for the first time that a motivational interview delivered by a social robot can elicit out-loud discussion from participants.”

What’s more, most of the participants reported that the interaction was both enjoyable and helpful.  They revealed that it was especially useful to hear themselves talk about their behavior out loud, and the fact that the robot didn’t interrupt was another distinct plus point.

There were also strong concerns expressed by the group about being judged by a human interviewer.  The relatively impersonal manner of the robot therefore was actually a positive, which suggests such technology could be particularly useful for discussions on sensitive issues.

“The next stage is to undertake a quantitative study, where we can measure whether participants felt that the intervention actually increased their activity levels,” the team conclude.