Stimulus frequency in the identification of line segments

Summary In a complete identification experiment line segments with 0, 45, 90, and 135 degrees orientation were exposed tachistoscopically as stimuli. The four stimuli were identified under three successive frequency conditions: In the first condition the stimuli occurred with equal frequency, in the second condition the stimulus probabilities were 0.4, 0.3, 0.2, and 0.1, and in the third condition the stimulus probabilities were 0.2, 0.1, 0.4, and 0.3. A 4×4 stimulus-responses confusion matrix was obtained from each condition and subject. The hit rates obtained for a more frequent stimulus (P=0.4 or 0.3) were higher than the hit rates obtained for a less frequent stimulus, and the same relation held between the proportions of an incorrect response corresponding to a more frequent stimulus and a less frequent stimulus. The confusion matrices were analysed according to the perceptual all-or-none model which allows separation of the effects of stimulus frequency on sensory and on decision processes in identification. Results showed marked effects of stimulus frequency on the decision processes yet no systematic effects on the sensory processes. Attempts to decompose the effects of stimulus frequency on the decision processes failed.Paper presented at the Symposium: Heinrich Düker — Impulse seiner Forschung. 20. Tagung experimentell arbeitender Psychologen, Marburg, 1978