Maximum likelihood estimation of difference scaling functions for suprathreshold judgments.

Area of Science:

• Cognitive Psychology
• Psychophysics
• Mathematical Psychology

Background:

• Maximum likelihood estimation (MLE) is a standard method for creating perceptual scales from binary judgments.
• Existing MLE methods rely on Thurstone's theory but do not account for diminishing returns in suprathreshold difference perception.
• Diminishing returns describe the overestimation of perceived differences when summing smaller ones.

Purpose of the Study:

• To develop and validate a novel MLE approach for perceptual scaling that incorporates diminishing returns.
• To model the perception of differences under Thurstone's theory, allowing for the possibility of diminishing returns.
• To evaluate the feasibility and accuracy of this new method with simulated and empirical data.

Main Methods:

• Developed an adapted MLE model based on Thurstone's theory to include diminishing returns.
• Validated the model using Monte Carlo simulations with experimental triads.
• Applied the method to empirical datasets to assess its practical application in perception research.

Main Results:

• The proposed method successfully models diminishing returns, its absence, and the opposite effect.
• Accuracy was demonstrated in both known and unknown perceptual scale scenarios.
• The method proved feasible for analyzing empirical perception data.

Conclusions:

• The new MLE approach provides more accurate modeling of suprathreshold difference judgments.
• It offers a more complete understanding of the perceptual processes involved in comparative judgments.
• This method allows for a robust evaluation of Thurstone's theory of difference judgments.