Saccadic selectivity without identification: unconscious perception?

Eye movements were recorded in a visual search task in order to investigate unconscious perception. The study was designed to substantiate the problems related to the classical dissociation paradigm and to demonstrate the methodological advantages of the relative sensitivity approach (Reingold & Merikle, 1988). Participants had to search for a target (O) among similar (Q) and dissimilar distractors (X) indicating whether the target was on the left or right side of the display. The left-right decisions served as the direct measure of (conscious/ unconscious) perception. Saccadic selectivity towards similar distractors served as the indirect measure. In the experiment, saccadic selectivity occurred before accuracy in left-right decisions, i.e., the indirect measure was more sensitive than the direct measure. The result pattern is compatible with both the indirect-without-direct data pattern required by the dissociation paradigm as well as the indirect-greater-than-direct data pattern required by the relative sensitivity approach. The latter approach is to be preferred since it is based on a minimal working assumption and not flawed by the methodological problems of the dissociation paradigm (validity problem and null sensitivity problem). Whether our direct and indirect measures are truly comparable, as required by the relative sensitivity approach, remains debatable.