The tendency to underestimate egocentric distances in immersive virtual environments (VEs) is not well understood. However, previous research (A. R. Richardson & D. Waller, 2007) has demonstrated that a brief period of interaction with the VE prior to making distance judgments can effectively eliminate subsequent underestimation. Here the authors examine the mechanism underlying the effect of VE interaction and the conditions that may give rise to it. In Experiment 1, after interacting with an immersive VE, participants tended to overestimate distances in the physical world, indicating that the interaction involved a recalibration of the perceptual-motor system. Experiment 2 demonstrates that visual information is not necessary (and that body-based information is necessary) during the interaction period for it to have a positive effect on subsequent distance estimation accuracy. Experiment 3 illustrates that the interaction task does not need to be goal directed in order to be effective.

Potential or actual applications of this research include the improvement of VE systems requiring accurate spatial awareness.

It is often noted that distances are significantly underestimated in computer-simulated (virtual) environments. Two experiments examine observers' ability to use error corrective feedback to improve the accuracy of judgments of egocentric and exocentric distances. In Experiment 1, observers viewed objects in an immersive virtual environment and estimated their distance through a blindfolded walking task. Different groups received feedback on either egocentric, exocentric or none of these judgments. Receiving feedback improved observers' ability to estimate only those distances for which feedback was provided. These effects persisted for at least 1 week. In Experiment 2, observers estimated egocentric distance by means of both a direct and indirect walking task. Receiving feedback on the direct walking task predominantly improved direct estimates and not indirect estimates. These findings suggest that although feedback training offers a relatively straightforward and immediate way of overcoming problems of distance estimation, its effects are specific to both the type of judgment and the type of response.

No abstract

In The ArchiTecTure MAchine (1970), nICholas neGroponte postulates the DevelopMent oF DesIGn MaChInes whereIn the "DesIGn proCess, ConsIDereD as evolutIonary, Can be presenteD to a MaChIne, also ConsIDereD as evolutIonary, anD a Mutual traInInG, resIlIenCe, anD Growth Can be DevelopeD." the book, dedicated to "the first machine that can appreciate the ges ture," argues for developing machines with human like quali ties. this paper aims to develop an alternative trajectory to the "evolutionary" architecture machine, this time not towards anthropomorphism but responsiveness. the aim on one level is the same: to create machines that appreciate the gesture. however our approach is tied to more modest aims and means that bring current thinking on evolutionary processes and the forming of materials together. the reconfigurable mold (rCM) is an architecture machine that produces parts that can be combined to create more complex organizations. the molds are simple analog computers that employ various continuous scales like volume, weight and heat to develop their unique components. parametric alterations are made possible by affecting these measures in the process of fabrication. an underlying material that is instrumental in the molds is rub ber, whose variable elasticity provides unique possibilities for indexing the gesture that remains elusive for industrial pro cesses. Reconfigurable Molds as Architecture MachinesProceedings 287Materials and Craft Inspired By NatureReconfigurable Molds as Architecture Machines FIGure 1. (toP RIght) the RCM-J FIGure 2. (BottoM left) DIffeReNt ReCIPeS FIGure 3. (toP left) RCM-D