Related Experiment Video
Updated: Mar 17, 2026

Investigating Object Representations in the Macaque Dorsal Visual Stream Using Single-unit Recordings
Published on: August 1, 2018
Modeling Visual Exploration in Rhesus Macaques with Bottom-Up Salience and Oculomotor Statistics
Seth D König1, Elizabeth A Buffalo2
1Wallace H. Coulter Department of Biomedical Engineering at the Georgia Institute of Technology and Emory UniversityAtlanta, GA, USA; Yerkes National Primate Research CenterAtlanta, GA, USA; Graduate Program in Neuroscience, University of WashingtonSeattle, WA, USA; Washington National Primate Research CenterSeattle, WA, USA.
Abstract:
There is a growing interest in studying biological systems in natural settings, in which experimental stimuli are less artificial and behavior is less controlled. In primate vision research, free viewing of complex images has elucidated novel neural responses, and free viewing in humans has helped discover attentional and behavioral impairments in patients with neurological disorders. In order to fully interpret data collected from free viewing of complex scenes, it is critical to better understand what aspects of the stimuli guide viewing behavior. To this end, we have developed a novel viewing behavior model called a Biased Correlated Random Walk (BCRW) to describe free viewing behavior during the exploration of complex scenes in monkeys. The BCRW can predict fixation locations better than bottom-up salience. Additionally, we show that the BCRW can be used to test hypotheses regarding specific attentional mechanisms. For example, we used the BCRW to examine the source of the central bias in fixation locations. Our analyses suggest that the central bias may be caused by a natural tendency to reorient the eyes toward the center of the stimulus, rather than a photographer's bias to center salient items in a scene. Taken together these data suggest that the BCRW can be used to further our understanding of viewing behavior and attention, and could be useful in optimizing stimulus and task design.

