Abstract
It is often argued that humans can discriminate stimuli outside consciousness. In a typical paradigm a stimulus (the ‘prime’) is masked such that participants are hardly aware of it. In a ‘direct task’, participants discriminate the prime and are close to chance, suggesting that they do not perceive the prime consciously. In an ‘indirect’ task, participants discriminate a ‘target’ stimulus following the prime. Here, the prime has congruency effects on reaction times (participants are faster if prime and target belong to the same category). This is seen as evidence for better discrimination in the indirect task and preserved unconscious processing. However, this reasoning is only correct if the congruency effect is converted to an appropriate discrimination performance and then compared to the discrimination in the direct task (Franz & Luxburg, Psychological Science, 2015). Because this is rarely done, the literature on priming effects needs reconsideration. We performed two studies replicating and extending well established effects (cf. the behavioral experiments of Dehaene et al., Nature, 1998): Primes and targets were digits and participants discriminated their numerical size. In experiment 1 (N=18), we found a congruency effect of 11.3 ms (t(17)=4.2, p< 0.001) in the indirect task, replicating the literature. However, the discrimination performance in the indirect task (55.3 %) did not exceed the direct task performance (54.1 %, t(17)=1.1, p=0.3). In experiment 2 (N=20), prime visibility was varied using eight different prime contrasts. We again found congruency effects between 13 and 29 ms (all ps < 0.02), but no better discrimination in the indirect task. These results indicate that --- at least for our stimuli --- there is no better or preserved unconscious processing of the numerical value of digits in cases where conscious discrimination is close to chance.
Acknowledgement: This project was supported by a grant to VHF by the Deutsche Forschungsgemeinschaft (DFG) within the CRC 1233