Search: This Site | People | Departments | Penn State
Topology/Geometry Seminar
Friday, May 8, 2009
4:00 pm,
260 Hawthorn
Eric Charles
Penn State Altoona
Mapping the topology of emotion space.
Abstract: The research objective of this project is to test the hypotheses that emotion attributions are responses to particular types of motion, and that the motions that elicit different attributions are related in a specific topological fashion. The project will lead to fundamental insights about the nature the causes of emotional attributions, and develop novel research methods that use modern mathematical methods to provide more robust solutions to old psychological questions.
Intro Part 1: Psychologists do not often invoke topological models, but on occasion they do so, usually in a somewhat arbitrary manner, and without intention to explore the implications of their models fully. One example that has gained considerable traction in the literature is the Circumplex Model of Affect (“affect” being an unnecessarily technical term for “emotion”). Proponents of such models, for example, may propose emotions to exist at various degrees such as: pleasure (0), excitement (45), arousal (90), distress (135), displeasure (180), depression (225), sleepiness (270), and relaxation (315). However, a circumplex implies much more, and many other models could accommodate a similar sense of adjacency.
Intro Part 2: Researchers studying aspects of perception-action linkage have long contended that individual people’s behavior can be described as a constant function of environmental properties. Traditionally “environmental properties” have been conceived as algebraic functions of stimulation, e.g. one can intercept a fly ball by moving so as to cancel any optical acceleration. However, there is evidence that a more broad conception is needed, especially in terms of the math involved. This seems obvious when you begin to talk about perceiving emotions. There are many minimal displays to which subjects attribute emotional content. For example, if shapes (triangles, circles, etc.) move the right way, people attribute emotional content to them (e.g., “the triangle is mad at the circle”). These displays are made intuitively, and no one knows much about what moving the “right way” entails. A formula that described such motions would be very useful to researchers. We anticipate that such formula will be best expressed in topological language, rather than using simple algebraic formulae.
Objective and Methods: We plan to create formula to describe the necessary motions for emotion attribution in minimalist displays. We plan on doing this by having shapes interact in various ways and then having participants rate them in terms of emotional content and in terms of similarity to other displays. The similarity judgments will allow us to map out the minimal dimensionality of the parameter space. It will also allow comparison of the spaces that constitute different emotion attributions. We are interested in testing implications of the circumplex model: Is the space continuous or discrete? Does there appear to be a hole in the middle? etc. We have a rough version of this task already, running in NetLogo. If the project is successful we will have a program that can produce novel displays that elicit standardized emotional attributions.
We meet weekly on Friday afternoons at Penn State Altoona (and sometimes at Penn State University Park). If you are interested in giving a talk in our seminar please contact one of the coordinators for the Spring 2009 semester: Wojtek Dorabiala and Aissa Wade.
