Hierarchical drift‐diffusion modelling uncovers differences of valenced self‐evaluation

Differences in valenced self‐evaluation refer to positive and negative coexistence in the process of self‐evaluation, while there is a clear difference in cognitive processes. The present study aimed to uncover the differences in the latent cognitive parameters (e.g., processing speed) in valenced self‐evaluation using the hierarchical drift‐diffusion model in two independent experiments. A self‐referential decision‐making task was applied in both experiments, and a self‐descriptiveness task plus the rating of related emotions (e.g., pride and shame) were also used but only in Experiment 2. Results of Experiments 1 & 2 showed a faster processing speed for accepting positive attributes and longer times for encoding and response execution in negative self‐evaluation. Moreover, Experiment 2 found cognitive parameters had predictive effects on subsequent decisional outcomes such as self‐descriptiveness and self‐related emotions via Bayesian inference. The current study provided findings that help to understand the cognitive mechanism behind self‐positivity and self‐accuracy biases.