We have previously reported that detection performance of monkey subjects during active visual search is constrained by object density. Here we report similar results for human subjects. Methods. Three subjects searched for a trial-by-trial cued T or L target in arrays of rotated (60 deg increments) T's & L's. Array set sizes of 6, 12, 24, or 48 items were randomly placed within a 35.5 × 26.5 deg display area. Each stimulus segment was 1.0 × 0.25 deg. Stimuli were either all green or all red, varied by trial. The task was to find and fixate the target for 600 ms. Target detection was defined as having occurred when the next saccade captured the target by landing within 1.1 deg of the target center. Eye position measurement was made using an SMI Eyelink system. Detection probability was measured as a function of the V1 cortical image separation of the target and its nearest neighbor. Separations were measured on a 3D model of the curved V1 surface constructed using current estimates of a human cortical magnification factor. Results. In humans, as well as in monkeys, active search proceeded via sequential fixations of stimuli (75% of fixations within 2 deg of nearest stimulus). Target detection probability can be described as a threshold function of the mm separation of the V1 cortical image representations of the target and its nearest distractor. To achieve comparable performance humans require about twice (2X) the cortical separation as monkeys. The similarity of this result to the known doubling of the size of the ocular dominance column widths in humans versus monkeys, suggests that the scale of spatial interactions between objects is linked to the scale of hypercolumns in V1.

We propose a datamining based method for automated reverse engineering of search strategies during active visual search tasks. The method uses a genetic program (GP) that evolves populations of fuzzy decision trees and selects an optimal one. Previous psychophysical observations of subjects engaged in a simple search task result in a database of stimulus conditions and concomitant measures of eye gaze information and associated psychophysical metrics that globally describe the subjects search strategies. Fuzzy rules about the likely design properties of the components of the visual system involved in selecting fixation location during search are defined based on these metrics. A fitness function that incorporates both the fuzzy rules and the information in the database is used to conduct GP based datamining. The information extracted through the GP process is the internal design specification of the visual system vis-à-vis active visual search.

Our previous research examined the effects of target eccentricity and global stimulus density on target detection during active visual search in monkey. Here, eye movement data collected from three human subjects on a standard single-color Ts and Ls task with varying set sizes were used to analyze the probability of target detection as a function of local stimulus density. Search performance was found to exhibit a systematic dependence on local stimulus density around the target and as a function of target eccentricity when density is calculated with respect to cortical space, in accordance with a model of the retinocortical geometrical transformation of image data onto the surface of V1. Density as measured by nearest neighbor separation and target image size as calculated from target eccentricity were found to contribute independently to search performance when measured with respect to cortical space but not with standard visual space. Density relationships to performance did not differ when target and nearest neighbor were on opposite sides of the vertical meridian, underscoring the hypothesis that such interactions were occurring within higher visual areas. The cortical separation of items appears to be the major determinant of array set size effects in active visual search.

Virtual environments will be naturally integrated with high performance computing (HPC) since: (a) advanced grand challenge type simulations will benefit from dynamic virtual environment front-ends, and (b) advanced virtual environments will require simulation engines based on high performance computing resources. Such an integration minimally requires interoperability and standardized interfaces between both components. The discussion of these requirements is extended and it is shown how relevant design issues can be elegantly addressed by the multilashing object-oriented visual interactive environment (MOVIE) system, currently under development at North Parllel Architectures Center as an infrastructure layer for dynamic interactive HPC.< >

MOVIE (multitasking object-oriented visual interactive environment) is the new software system for high performance distributed computing (HDPC), currently in the advanced design and implementation stage at Northeast Parallel Architectures Center (NPAC), Syracuse University. MOVIE System is structured as a multiserver network of interpreters of high-level object-oriented programming language MovieScript. MovieScript derives from PostScript and extends it in the C++ syntax based object-oriented interpreted style towards 3D graphics, high performance computing and general purpose high level communication protocol for distributed and MIMD-parallel computing. The authors describe the overall open systems based MOVIE design and itemize currently implemented, developed and planned components of the system