Determining subprocesses of visual feature search with reaction time models

Area of Science:

• Cognitive Psychology
• Visual Perception
• Computational Neuroscience

Background:

• The traditional serial/parallel dichotomy in visual search mechanisms is increasingly questioned.
• Existing reaction time slope analyses lack sufficient detail to differentiate intermediate search strategies.
• Overt feature search paradigms require nuanced understanding of attentional deployment.

Purpose of the Study:

• To investigate visual search mechanisms between the extremes of 'pop-out' and 'strictly serial search'.
• To develop and validate a novel reaction time analysis technique for detailed modeling of search processes.
• To characterize the dynamics of attention, including dwell and shift durations, during visual search.

Main Methods:

• Development of a novel reaction time analysis technique modeling individual reaction times as sums of search step durations.
• Numerical fitting of several model variants to empirical reaction times in an overt feature search task.
• Direct measurement and validation of the 'range of attention' parameter in a second experiment.

Main Results:

• The best-fitting model indicated that multiple items can be processed within a single fixation.
• Attention movement was found to be abrupt, not continuous, even in 'pop-out' conditions.
• Explicit attentional shifts to the target were frequently observed, even in visually salient search scenarios.

Conclusions:

• The developed model explains variations in reaction time functions without relying on a strict parallel/serial distinction.
• Visual search involves complex attentional dynamics, including discrete movements and parallel processing capabilities.
• The findings challenge simplistic models of visual search and highlight the nuanced nature of attentional deployment.