EVALUATION OF THE EQUIVALENCE OF TWO DIFFERENT METHODS OF THRESHOLD ESTIMATIONS WITH ADAPTIVE PSYCHOPHYSICAL PROCEDURES
Because of its higher efficiency, adaptive psychophysical procedures are becoming increasing popular compared to traditional non-adaptive procedures. The present study evaluates the equivalence of two different methods for threshold estimation based on single trials and so- called mid-run estimates, respectively. For this purpose, 286 participants performed four psychophysical temporal tasks: duration discrimination of filled and empty intervals in the range of milliseconds, duration discrimination of filled intervals in the range of seconds, and rhythm perception. All tasks employed an adaptive rule based on the weighted-up-down procedure (Kaernbach, 1991). As a psychophysical indicator of performance, 75%-difference thresholds were determined on the basis of both single trials and mid-run estimates. As a measure of congruence, Pearson correlations were computed across individual threshold estimates obtained with both methods. Correlation coefficients ranging from rxy = .95 to rxy = .98 suggest a highly reliable level of equivalence for both methods of threshold estimation.