Attention switching in depth using random-dot autostereograms: attention gradient asymmetries

Area of Science:

• Cognitive Psychology
• Neuroscience
• Visual Perception

Background:

• Investigating attentional mechanisms is crucial for understanding visual processing.
• Random-dot autostereograms (RDASs) offer a controlled method to study depth perception and attention.
• High perceptual load tasks can reveal limitations and characteristics of attentional allocation.

Purpose of the Study:

• To examine attention shifts along the sagittal plane in distractor-free, high perceptual load tasks.
• To determine how target set characteristics influence the observation of depth gradients.
• To test the hypothesis of a viewer-centered asymmetric attention gradient dependent on perceptual load.

Main Methods:

• Utilized random-dot autostereograms (RDASs) to create stimuli with varying depths.
• Employed a same/different comparison task under high perceptual load conditions.
• Conducted three experiments manipulating target set similarity (five similar vs. two dissimilar objects).

Main Results:

• A robust asymmetric depth gradient was observed when the target set comprised five similar objects.
• No depth gradient was detected when the target set consisted of two dissimilar objects.
• The presence and shape of the depth gradient were contingent on target discriminability.

Conclusions:

• Attention gradients in the depth plane are asymmetric and viewer-centered.
• Perceptual or attentional load, defined by target-set discriminability, significantly modulates these attention gradients.
• Findings support a model where attentional allocation adapts to task demands and stimulus properties.