Robotics Research Lab
USC Computer Science
USC Engineering
/ Research / Projects / Using Hands-Off Assistive Robotics for Educational Intervention for Children with Attention Deficit/Hyperactivity Disorder (AD/HD)

Overview Approach Evaluation Videos Contact Details Past Contributors


Attention deficit/hyperactivity disorder (AD/HD) is a condition that affects 3% to 5% of school-age children or approximately 2 million children in the United States. It is estimated that at least one child will have AD/HD in a class size of 25 to 30 students. Children with AD/HD suffer from low self esteem, social and academic failure, substance abuse, and a possible increase in the risk of antisocial and criminal behavior. Early intervention will reduce these risks and provide long-term effects to improve the children's lives. A combination treatment of medical, educational, behavioral and psychological intervention strategy has been shown to be the most effective in treating children with AD/HD. If an educational, behavioral and psychological intervention strategy would prove effective, prescription medication could be minimized or eliminated. This reduces the undesirable side effects of medication.

This work aims to develop a socially assistive mobile robot capable of providing an effective educational intervention strategy for elementary school students with AD/HD. It also aims to assist teachers in helping the students with AD/HD. Teaching children with AD/HD requires special training and individualized attention from the teachers. However, due to the shortage of special education teachers and students with AD/HD being placed in the same classrooms with other students, students receive significantly less attention than they need in order to excel. The human-robot interaction is not only going to provide one-on-one attention needed by the student with AD/HD, but also monitor the student's progress and increase his or her attention and academic productivity.


The main goals of this project are to implement a robot called Eddie to do the following:

  1. Provide an educational intervention strategy through human-robot interaction.
  2. Provide one-on-one attention needed by a student with AD/HD.
  3. Supervise the student's progress in the absence of a teacher.
  4. Determine the effectiveness of the proposed educational intervention strategy in terms of improving the student's scores.
Human-Robot InteractionThe beneficiary population for this study consists of elementary school students diagnosed with AD/HD. Each student can be doing some exercises from a book, a computer, or other alternative forms of learning. For the purpose of this project, educational software that is specially developed to supplement an elementary school Math curriculum will be used because of its automated scoring process, as well as its ability to keep track of the length of time consumed by the student. The student with AD/HD doing his or her work may sometimes have difficulty following instructions or paying attention. He or she may not feel comfortable with seeking assistance from the teacher. It is suggested that using an educational toy robot to interact with the student on a one-on-one basis can indirectly help the student build his or her self-esteem. This is accomplished by engaging the student to do some physical exercises or answer relevant questions such as elementary Math problems like addition, subtraction, multiplication, and division. In this case, the robot is serving as a socially competent peer and mentor for the student with AD/HD. The socially assistive mobile robot will monitor the student's progress by making sure he or she is on task. At random intervals, the robot will initiate contact with the student by asking to play a game. The student will also have the opportunity to initiate contact with the robot by using fluorescent colored flash cards to get the robot's attention; this element of physical interaction will provide a subtle way for the student with AD/HD to develop social skills. Most importantly, the series of exchanges taking place in all parties culminate as solutions in improving the student's educational and social well-being.

A pilot study has already been performed in a controlled environment at the University of Southern California's Interaction Laboratory. Each of the experiments last for 30 minutes. The data collected for both the computer and robot interactions have been analyzed off-line to determine the effectiveness of the AD/HD educational intervention strategy. Comparison of the percentage of number of problems solved correctly between the human-computer interactions are made. To determine how much the student's improvement is as a result of interacting with the robot, comparison of the percentage of number of problems solved correctly between the human-robot interactions are used. These comparisons serve as evidence whether the student is improving or not. The length of time in the logs further indicate whether the student's attention span on the computer has increased after interacting with the robot. The number of human-robot interactions indicate how often the student initiated the contact with the robot or vise versa. Finally, a survey is conducted at the end of the experiment to show whether the student found this intervention strategy useful. Videos of the preliminary results are included in the Video section. These results will be used in a fuller study that is being designed.


Preliminary results:


Prof. Maja Mataric' : mataric (at)

Past Contributors

Jenny Chang : changjen (at)