Quantitative and Visual Cognition and Human Performance Divisions

My central research theme concerns geometric and dynamic models of psychological phenomena. My research ranges from the theoretical*conceptualization and implementation of new mathematical models*to the substantive*the application of these models and tests to problems in perception, cognition, and other areas.
Mathematical Models. I have developed a suite of statistical tests which can reveal and characterize geometric properties of proximity data.
Application to Substantive Areas. Intimately tied to the idea of a geometric model is the notion of trajectories. Many dynamic processes can be well-described as proceeding along paths in a geometric space. This approach has proven to be especially fruitful in application to the study of emotional facial expressions. In recent work, we have created an interactive face model that allows a subject to create real-time expressions using a joystick. By sampling the movements of the joystick hundreds of times a second, we record the resulting real-time expression as a trajectory. In our current work we have demonstrated statistically that emotional expressions previous lumped together as *happiness* can actually be differentiated. We have further demonstrated that observers can reliably differentiate these expressions when presented with dynamic displays, but not static displays. The end result of the work is that the commonly-accepted six basic emotions will need to be expanded, since the sole remaining roadblock to the acceptance of the differentiated *happiness* emotions was the lack of a known facial signal. This result is being prepared for submission to the Proceedings of the National Academy of Sciences.

Representative Publications:

• Polk, J., Spencer-Smith, J. DiBerardino, L. Ellis, D., Downend, M., Rosengren, K. S. (In press). Quantifying variability in phase portraits: Application to gait ontogeny. Infant Behavior and Development.
• Spencer-Smith, J. & Townsend, J. T. Tests for curvature in pairwise proximity data. To be submitted to Journal of Mathematical Psychology.
• Milanak, M.E., Berenbaum, H. & Spencer-Smith, J. (To be submitted). Race, gender, and emotion effects on facial affect recognition. Emotion and Cognition.
• Spencer-Smith, J. & Milanak, M.E. (Completed, to be submitted) Making Faces Updated: Dynamic Facial Affect Models set (DFAMS). Behavioral Research Methods.
• Spencer-Smith, J. A reinterpretation of the additive constant. To be submitted to Psychometrika.

Classes Recently Taught:

• P406 Statistical Methods I
• P435 Mathematical Psychology
• P509 Multidimensional Scaling