A model for the time-order error in contrast discrimination.

Area of Science:

• Psychophysics
• Cognitive Psychology
• Sensory Perception

Background:

• Two-interval forced-choice discrimination experiments involve comparing stimulus magnitudes across sequential intervals.
• Signal detection theory predicts order independence (balance condition), but time-order errors (systematic underestimation) are consistently observed.
• Time-order errors, described by Fechner, challenge standard psychophysical models.

Purpose of the Study:

• To investigate sensory and procedural factors contributing to time-order errors in discrimination tasks.
• To develop a formal model explaining how presentation order influences performance despite a single underlying mechanism.
• To test the hypothesis that time-order errors arise from specific sensory and procedural contaminants.

Main Methods:

• Analysis of sensory factors like temporary desensitization.
• Examination of procedural factors such as short interstimulus/intertrial intervals and response bias.
• Derivation of a formal model to predict performance variations based on presentation order.
• Experimental validation of the model and the failure of the balance condition.

Main Results:

• Empirical data consistently violate the balance condition, confirming the presence of time-order errors.
• The proposed model successfully explains how sensory and procedural factors lead to order-dependent performance.
• Experimental results support the hypothesis that time-order errors result from contamination by identified factors.

Conclusions:

• Time-order errors are not artifacts but systematic effects explained by specific sensory and procedural influences.
• The derived model provides a framework for understanding and potentially mitigating order effects in psychophysical tasks.
• This research refines our understanding of perceptual comparisons and the limitations of traditional signal detection models.