We report results of a study in which a low cost sociable robot was immersed at an Early Childhood Education Center for a period of 2 weeks. The study was designed to investigate whether the robot, which operated fully autonomously during the intervention period, could improve target vocabulary skills of 18-24 month of age toddlers. The results showed a 27 % improvement in knowledge of the target words taught by the robot when compared to a matched set of control words. The results suggest that sociable robots may be an effective and low cost technology to enrich Early Childhood Education environments.

Participatory user interface design with adolescent users on the autism spectrum presents a number of unique challenges and opportunities. Through our work developing a system to help autistic adolescents learn to recognize facial expressions, we have learned valuable lessons about software and hardware design issues for this population. These lessons may also be helpful in assimilating iterative user input to customize technology for other populations with special needs.

Localizing facial features is a critical component in computer vision applications such as person identification and expression recognition. Practical applications require detectors that operate reliably under a wide range of conditions, including variations in illumination, ethnicity, gender, age, and imaging hardware. One challenge for the development of such detectors is the inherent tradeoff between robustness and precision. Robust detectors provide poor localization and detectors sensitive to small shifts, which are needed for precise localization, generate a large number of false alarms. Here we present an approach to this tradeoff based on context dependent inference. First robust detectors are used to detect contexts in which target features occur, then precise detectors are trained to localize the features given the context. This paper describes the approach and presents a thorough empirical examination of the parameters needed to achieve practical levels of performance, including the size of the training database, size of the detector's receptive fields, and methods for information integration. The approach operates in real time and achieves, to our knowledge, the best performance to-date reported in the literature.

Individuals diagnosed with Autism Spectrum Disorders (ASD) often have challenging behaviors (CB's), such as self-injury or emotional outbursts, which can negatively impact the quality of life of themselves and those around them. Recent advances in mobile and ubiquitous technologies provide an opportunity to efficiently and accurately capture important information preceding and associated with these CB's. The ability to obtain this type of data will help with both intervention and behavioral phenotyping efforts. Through collaboration with behavioral scientists and therapists, we identified relevant design requirements and created an easy-to-use mobile application for collecting, labeling, and sharing in-situ behavior data in individuals diagnosed with ASD. Furthermore, we have released the application to the community as an opensource project so it can be validated and extended by other researchers.

Labeling videos for affect content such as facial expression is tedious and time consuming. Researchers often spend significant amounts of time annotating experimental data, or simply lack the time required to label their data. For these reasons we have developed VidL, an open source video labeling system that is able to harness the distributed people-power of the internet. Through centralized management VidL can be used to manage data, custom label videos, manage workers, visualize labels, and review coders work. As an example, we recently labeled 700 short videos, approximately 60 hours of work, in 2 days using 20 labelers working from their own computers.