No Danger Here, Will Robinson – Robot Therapist Shows Promise in Speech Therapy
Could a robot deliver speech therapy? It might sound a little like science fiction, but robot-delivered speech therapy is possible and could soon become a reality in clinical practice. Researchers at the University of Massachusetts, Amherst, are among the first to investigate the potential of humanoid therapists for rehabilitation patients in speech and physical therapy.
New insights into the potentials of robot-delivered therapy were derived from a single case study using an adult stroke patient. The aim of the study was to assess how therapy interventions in speech, affected interventions in another area of rehabilitation (in this case, physical therapy) in two different delivery scenarios. Following therapy with a robot therapist, the client made positive gains in verbal expression as well as physical improvement in his range of motion. Researchers also claim the therapy resulted in significant improvement in quality of life.
The robot, uBot, is a child-sized unit with arms and a computer screen. You might think artists designed the robot to look humanoid with 3D modeling, but the screen serves as the human interface. This screen is where the therapists interact with the client. It was developed as part of a grant from the American Heart Association to investigate the effect of stroke rehabilitation delivered by a humanoid robot.
Not only did they measure outcomes in speech and physical movement, but they examined the client’s service delivery schedule. Researchers found greater improvement when therapies (speech versus physical) were separated, rather than delivered back to back. This might have given clients a boost in making the brain connections needed to re-gain skills without being too taxing and demanding. While the study was done with only one adult client, these results should urge all service providers to consider the frequency and therapy schedule that patients receive- perhaps “back to back” isn’t the most effective for some children. Or perhaps service delivery in the classroom session has it’s own set of challenges.
There is no question as to the potential of robots used in future speech therapy projects. Prior to this study researchers had shown that robots have the potential to learn basic language after interactions with a human. Further, we know experiments using robots specifically for children with autism has shown promise in therapy for language learning.
But more research into just how this technology can be incorporated into the clinical setting is needed. Of course there is no shortage of practical questions including: how does insurance reimbursement work? Do machines help reach rural or under served populations? How much preparation and training into setting up the robot are needed before it can be use? What kind of emotional response, social skills or bond can clients build with a human-like, artificial therapist? Could a robot manage other elements of working with a patient, such as behavior management or reinforcement? And of course, what is the cost?