Exploring similarities and differences in reaction time distributions for social and non-social stimuli: a distributional analysis of spatial interference tasks.

Using distributional analysis, we examined five datasets from spatial interference tasks that included gaze and arrow targets (Marotta et al., 2018; Marotta et al., 2019; Hemmerich et al., 2022) to explore similarities and differences between social and non-social stimuli. Two of these tasks implemented a block design, with separate blocks for arrow and eye-gaze stimuli, while the remaining three tasks employed an intrablock design, where one of them added whole-face gaze and words as social and non-social stimuli, respectively. To conduct the analyses, RT distributions were divided into five bins using quantile-averaging, ranging from the fastest to slowest RTs. The conditional accuracy functions (CAFs) exhibited a clear dissociation between social and non-social stimuli based on response speed. The standard congruency effect was observed on response accuracy for both stimulus types in trials with slowest RTs, while opposite congruency effects were seen in trials with fastest RTs. For delta plots analyses, quadratic regressions were fitted for each participant. The results showed that the majority of our analyses exhibited significantly higher quadratic terms and lower linear terms compared to zero. This indicates a negative-going delta function for faster responses and a positive-going function for slower responses, implying a reduction in congruency effect with non-social and a more negative effect with social stimuli for medium RTs. It seems that irrelevant spatial information fades with arrow and word stimuli to a certain extent. However, the reversed congruency effect observed with social stimuli resulted in a less negative effect on both faster and slower RTs.