The workshop was organized within the MobileHCI 2016 conference in Florence, Italy, September 6th, 2016.

Workshop theme and goals

Gaze tracking in psychological, cognitive, and user interaction studies has recently evolved toward mobile solutions, as they enable direct assessment of users’ visual attention in natural environments. Eye movements gather information for action planning, so they provide information about users’ intentions and next actions. However, gaze behavior in natural, unstructured tasks is quite complex and cannot be satisfactorily explained using models that have been created in controlled laboratory environments. Gaze behavior involves an interplay of cognitive processes. The modeling of these processes computationally is difficult, not least because of the unknowns involved: it is a challenge to construct an experimental setup where the “ground truth” is known, and can be used to train, e.g., a machine learning model. In addition, data from tracking experiments is seldom shared, obstructing interdisciplinary approaches in analyzing and utilizing gaze data. As a result, individual contributions to gaze-based inference stay isolated, and do not form a larger whole for advancing understanding on gaze-action behavior.

The issue of inferring user action with mobile gaze tracking is highly multidisciplinary, requiring deep understanding of a variety of research fields, including the functioning of human visual system, mathematical modeling, computer vision, machine learning, information technology, cognitive processes, user interaction, and psychology. To evolve the inference of gaze-action behavior in natural environments, an interdisciplinary approach bringing together a number of disciplines from cognitive sciences to machine learning is needed.

The goal of the workshop is to bring together a cross-domain group of individuals to

(i) discuss and contribute to the problem of using mobile gaze tracking for inferring user action,

(ii) advance the sharing of data and analysis algorithms as well as device solutions, and

(iii) increase understanding of behavioral aspects of gaze-action sequences.

The workshop proposes an interdisciplinary gathering for recognizing potential synergies, mapping solved and unsolved problems, and creating a research roadmap for the future applying the strengths of each contributing field.

This workshop is a full-day workshop, aiming to bring together people from industry and academia. Paper submissions should be made via an electronic submission system (instruction are coming up). A paper should have a length of 2 to 4 pages in the SIGCHI Extended Abstracts format and will be reviewed by the program committee. Successful submissions will have the potential to raise discussion, provide insights for other attendees, and illustrate open challenges and potential solutions. All accepted publications will be published on the workshop website and in the ACM Digital Library. At least one author of each accepted paper needs to register for the conference and the workshop itself. During the workshop, each paper will be given 10-20 minutes for an oral presentation. In addition, there will be room for demonstrations and hands-on sessions.