In the present study an alternative analysis to tau was considered that was based on perceived speed and size and that assumed constant deceleration for the detection of collision events. Observers were presented with displays simulating a 3-D environment with obstacles in the path of observer motion. During the trial, observer motion decelerated at a constant rate and was followed by a blackout prior to the end of the display. Observers had to detect which trials resulted in a collision. The results indicate that collision detection varied as a function of the size of the obstacles, observer speed, and edge rate--findings not predicted by an analysis of tau. The results suggest that observers use an analysis based on speed and size information. A model that assumes constant deceleration is proposed for braking control.

We examined the minimum latency required to locate and identify a visual target (visual search) in a two-alternative forced-choice paradigm in which the visual target could appear from any azimuth (0 degree to 360 degrees) and from a broad range of elevations (from 90 degrees above to 70 degrees below the horizon) relative to a person's initial line of gaze. Seven people were tested in six conditions: unaided search, three aurally aided search conditions, and two visually aided search conditions. Aurally aided search with both actual and virtual sound localization cues proved to be superior to unaided and visually guided search. Application of synthesized three-dimensional and two-dimensional sound cues in the workstations are discussed.

Observers were presented with displays simulating a 3-D environment with obstacles in the path of motion. During the trial, observer motion decelerated at a constant rate and was followed by a blackout prior to the end of the display. On some trials the rate of deceleration resulted in stopping before the collision, whereas on other trials the rate of deceleration resulted in a collision with the obstacles. The observer's task was to detect which trials simulated an impending collision. Proportion of collision judgments was greater for older as compared with younger observers when a collision was not simulated. Older observers showed less sensitivity to detect collisions than younger observers did, particularly at high speeds. The age-dependent results are discussed in terms of analyses based on tau and constant deceleration. The results suggest that increased accident rates for older drivers may be due to an inability to detect collisions at high speeds.

Four subjects were tested in a two-alternative, forced-choice, three down-one up adaptive paradigm in which two 200-ms signals were presented sequentially with a 200-ms interstimulus interval. The subject's task was to indicate whether the second stimulus was to the right or left of the first. Tests were conducted with 57 dB (A-weighted), 1.0-kHz high-pass noise, the minimum audible angle (MAA) task, and with lights emitting 620 nm at a luminance level of 200 mL, the minimum visible angle (MVA) task. Localization performance in the MAA task was equal to or better than that obtained in the MVA task for all regions of the frontal field with only one exception, presentations at 0 degrees azimuth. The implications of these results are discussed.

The experiments described in this report provide baseline performance measures of aurally directed detection and search for visual targets in an observer's immediate space. While the simple target detection task was restricted to the frontal hemi-field (extending 180 degrees in azimuth and 150 degrees in elevation), visual search performance (discrimination of which of two light arrays was present on a given trial) was evaluated for both the frontal and rear hemi-fields. In both tasks, the capacity to process information from the visual channel was improved substantially (a 10-50 percent reduction in latency) when spatial information from the auditory modality was provided concurrently. While performance gains were greatest for events in the rear hemi-field and in the peripheral regions of the frontal hemi-field, significant effects were also evident for events within the subject's central visual field. The relevance of these results to the development of virtual 3-D sound systems is discussed.