Bridging the gap between HCI and DHM: The modeling of spatial awareness within a cognitive architecture

Abstract

In multiple investigations of human performance on natural tasks in three-dimensional (3D) environments, we have found that a sense of space is necessary for accurate modeling of human perception and motor planning. In previous work, we developed ACT-R/DHM, a modification of the ACT-R cognitive architecture with specific extensions for integration with 3D environments. ACT-R/DHM could leverage existing extensions from the ACT-R community that implement the spatial sense, but current research seems to indicate that an "egocentric-first" approach is most appropriate. We describe the implementation of a custom spatial module in ACT-R/DHM, which allows for the consideration of spatial locations by adding a single ACT-R module that performs a very small set of operations on existing location information. We demonstrate the use of the 3D, egocentric-first spatial module to simulate a machine interaction task. © 2009 Springer Berlin Heidelberg.