Abstract:

Negative Compatibility Effect (NCE) is the surprising result that masked prime arrows inhibit the responses to compatibility arrow targets and facilitate the responses to the opposite direction arrow targets at the prime-target Interstimulus Interval of around 100 ms in the masked prime paradigm. NCE has been hotly investigated in terms of its subconsciousness and being counterintuitive. But whether NCE can be observed with semantic stimuli is still unknown. Traditionally, it is believed that subliminal information processing is automatic and is not down-regulated by top-down cognitive control process. However, recent studies challenged this conventional view. As a kind of subliminal perception effect, whether NCE is affected by top-down cognition control is still unknown. The present study combined masked-prime paradigm and Go/No-go paradigm to examine whether there was semantic NCE and whether task demands modulated semantic NCE. Eighteen subjects participated in the study. All stimuli were black against white background presented in the center of the screen. In masked-prime paradigm, Primes were Chinese character “左” (Left) or “右” (Right), targets were character or double arrows in go condition, either compatibility or incompatibility to the prime. Participants were instructed to judge the direction of the arrow or the character by pressing the one of left and right arrow keys. There were no responses to no-go trials. Reaction times were entered into 2 (compatibility) × 2 (task demand) analyses of variance (ANOVA). The results showed that, the main effect of compatibility and task demand were both significant; the interaction between compatibility and task demand was also significant. Further analysis demonstrated that NCE was significant only in the condition when the prime was related to task demands; there was no significant effect when the prime was not related to task demands. The behavioral data supported the existence of semantic NCE, but it was only observed when the prime was related to task demands. ERP waveform analysis revealed that P3 latency is longer only when prime was related to the task demand in Go trials. In No-go trials, N2 mean amplitude has significant difference between two character primes conditions in the related condition; P3 mean amplitude showed a significant difference between the related and unrelated conditions. The results suggest that P3 latency is an effective component to measure semantic NCE, especially in the Go trials. The findings indicated the existence of semantic NCE. In addition, task demands influenced semantic NCE through top-down cognitive control process.

Key words: task demands, negative compatibility effect, subliminal semantic priming