Why I'd like to participate and what I've been working on that's related.

I've been interested in and done research in several topics related to those of the meeting:

• Graphic displays: how they communicate, universals in how people construct them; how they are perceived and misperceived, remembered and misremembered; how they are interpreted; how to better design them.
• Spatial cognition: distortions in cognitive maps; descriptions and depictions of the visuospatial world; geographic categories

Mireille Betrancourt (a former post-doctoral visitor from INRIA-Grenoble) and Julie B. Morrison (a current graduate student) and I have been conducting a literature search on attempts to improve learning or performance through animation. Mireille and I did a set of experiments using minimal animation, that is, successive presentation of organized (or random) elements of the whole. We found that successive presentation of organized parts lead to faster learning and to a relevant mental model as compared with random presentation. However, successive presentation was not superior to static whole presentation, in accordance with previous work on part-whole learning.

Our review of the literature has revealed that in many cases, animation does not improve outcomes. This has come as a surprise to many of the researchers, who fully expected animation to be beneficial. We developed an analysis of the failures of animation, attributing it to both perceptual and conceptual difficulties. This analysis suggests when and where animation may be successful. For example, the oft-said, represent space with space and time with time seems to be too simplistic. It is not the space or time itself that is critical, it is how space or time is conceptualized. The work of Hegarty and Zacks' work described below show that there are events that occur continuously over time that are conceived of discretely, not continuously. Julie B. Morrison's dissertation research will explore cases, both concrete and abstract, where according to our analysis animation is expected to facilitate or not to facilitate.

Jeff Zacks has been working with me on event perception, categorization, and description. We selected familiar and unfamiliar events based on a norming study. The familiar events are making a bed and doing the dishes and the unfamiliar ones are fertilizing a plant and putting together a saxophone. Events unfold over time and are inherently dynamic. Nevertheless, we have found that people parse them into discrete units, which they break down into finer units. People's descriptions also follow the hierarchical organization. Jeff's next project is to investigate various ways of teaching the events, some of which are dynamic and some of which use stills chosen to be compatible or incompatible with people's parsing of the events.

Brief bio:

1969 Ph.D. University of Michigan Cognitive Psychology
1967-1977 Department of Psychology, Hebrew University, Jerusalem, Israel
1977-present Department of Psychology, Stanford University, Stanford, Ca Currently Professor

Some relevant papers:

Bryant, D. J. and Tversky, B. (In press). Mental representations of spatial relations from diagrams and models. Journal of Experimental Psychology: Learning, Memory and Cognition.

Levy, E., Zacks, J., Tversky, B. and Schiano, D. (1996). Gratuitous graphics: Putting preferences in perspective. Human factors in computing systems: Conference proceedings (pp. 42-49). NY: ACM.

Tversky, B. (1995). Cognitive origins of graphic conventions. In F. T. Marchese (Editor). Understanding images. Pp. 29-53. New York: Springer-Verlag.

Tversky, B. (1997). Memory for pictures, environments, maps, and graphs. In D. Payne and F. Conrad (Editors), Intersections in basic and applied memory research. Pp. 257-277. Mahwah, NJ: Erlbaum.

Tversky, B. (1998). Three dimensions of spatial cognition. In M. A. Conway, S. E. Gathercole, and C. Cornoldi (Editors). Theories of memory II. Pp. 259-275. Hove, East Sussex: Psychological Press.