How the range effect contaminates control scores in studies of visual illusions

Area of Science:

• Psychology
• Perception
• Psychophysics

Background:

• Alignment illusions are a common area of study in visual perception.
• Control scores are typically used as a baseline in psychophysical experiments.
• Within-subjects designs are frequently employed in psychological research.

Purpose of the Study:

• To examine the relationship between mean control scores and mean experimental scores in alignment illusion studies.
• To investigate the influence of the range effect on control scores.
• To assess the implications for psychophysical studies and experimental design.

Main Methods:

• Analysis of data from 23 experiments on alignment illusions.
• Statistical examination of mean control scores versus mean experimental scores.
• Evaluation of the impact of the range effect on score bias.

Main Results:

• Control scores were found to be significantly biased towards experimental scores.
• The range effect was identified as the mechanism causing this bias.
• The degree of bias was substantial in the examined experiments.

Conclusions:

• Control scores in alignment illusion research may be unreliable due to range effect bias.
• The findings raise concerns about the validity of control scores in psychophysical studies.
• Within-subjects designs require careful consideration, especially when extreme values influence outcomes.