Relative signal frequency imbalance does not affect perceptual encoding in choice reactions

There is conflicting evidence in the literature with regard to the question whether relative signal frequency affects perceptual processing of signals in choice reactions. The present study addresses this question by an experimental paradigm in which two stimuli, subtending a 100° visual angle, are simultaneously presented. At presentation the left signal is always fixated, followed by a saccadic eye shift and a subsequent fixation of the right signal, finally followed by a same/different response. A special property of this paradigm is that the fixation time of the left signal appears to represent an index of encoding unconfounded by response selection. Hence, if relative signal frequency affects encoding it should affect the fixation time of the left signal. Four experiments are described in which increasingly extreme signal frequencies and numbers of alternative stimuli were used. In addition, stimulus quality was varied in all studies. In no case was an effect of relative signal frequency imbalance observed. Stimulus quality consistently affected the fixation time of the left signal but did not interact with variations in relative signal frequency. These results cast serious doubt on the conclusion of Stanovich and Pachella (1977) that stimulus encoding is influenced by stimulus probability. Furthermore the results do not support a response set notion which assumes that encoding represents a top-down internal process than can be preset on the basis of expectancy.