Automaticity of unconscious response inhibition: comment on Chiu and Aron (2014).

Abstract

A recent study (Chiu & Aron, 2014) suggested that unconscious response inhibition is maintained when subliminal stimuli are mixed with supraliminal stimuli that are associated with response inhibition (mixed session), but it is abolished when they are presented alone (single session). However, awareness of the subliminal stimuli is likely to differ in the 2 sessions because of priming of awareness--awareness for subliminal stimuli is elevated (e.g., no longer subliminal) when mixed with supraliminal stimuli (Lin & Murray, 2014a). Here, in a novel design, we measured the awareness level in both sessions and found that the session-dependent effect was due to an awareness difference: The effect disappeared when awareness was comparable and emerged only when awareness was different. Arguments based on the lack of correlation between awareness and unconscious effects are refuted because typical correlation analysis underestimates the true correlation because of range restriction and it speaks only about individual differences that cannot explain within-subject effects (e.g., stimulus context here). Our findings also point to an attention-based mechanism underlying priming of awareness: Supraliminal trials are less attention-demanding, allowing for more attentional resources for subliminal trials in the mixed than single sessions. We discuss 2 implications. First, unconscious effects depend on top-down task sets and bottom-up stimulus strength. Second, to properly demonstrate unconscious processing, we stress the importance of having equivalent trial sequences between the main and awareness tests, promote a conjunction method that can strengthen inference, and discuss establishing a limit for equivalence between observed and chance performance.