Oculomotor target selection is mediated by complex objects.

Area of Science:

• Neuroscience
• Cognitive Science
• Visual Perception

Background:

• Oculomotor target selection involves visual feature discrimination, but the role of oculomotor substrates in this process is debated.
• One hypothesis suggests oculomotor vector representations are reweighted by external task relevance, while another proposes local feature analysis within oculomotor modules.

Purpose of the Study:

• To investigate whether the onset latency of saccadic interference from complex stimuli is longer than from simple stimuli.
• To determine if oculomotor vector representations encode features or are feature-invariant during saccadic target selection.

Main Methods:

• Behavioral experiment measuring human saccade metrics (curvature, endpoint deviations, frequency, error) over time after distractor onset.
• Used novel, visually complex distractors requiring discrimination from targets to guide saccades.
• Manipulated visual similarity between targets and distractors to assess oculomotor vector modulation.

Main Results:

• The earliest saccadic interference latency observed was approximately 110 ms, longer than in previous studies with simple stimuli.
• Initial oculomotor vector representations encoded features, showing increased modulation when distractors were highly similar to targets.
• Oculomotor vector modulation decreased over the experimental time course.

Conclusions:

• Saccadic interference latency is scaled by task difficulty and feature complexity, suggesting feature computation occurs outside the oculomotor system.
• The oculomotor system receives processed visual information, not raw features, after sufficient cognitive processing.
• Initial oculomotor vector representations are not feature-invariant and encode task relevance.