Instructed speed and accuracy affect binding

Abstract

In the past few decades, binding and retrieval mechanisms have gained increased interest in research on human action control. Recent studies show that these mechanisms also play a role in the control of multiple independent actions. Here, two or more successively executed responses seem to be bound to each other so that repeating one of them can retrieve the other, affecting performance in this second response and resulting in so-called response–response binding effects. Binding effects are typically found in the response time data and, somewhat less reliably, also in the error rates. Whether binding effects show in the response times, the error rates, or both, is likely influenced by the current speed–accuracy settings of the participants, with binding effects more likely showing in error rates under a speed setting, while more likely showing in RTs under an accuracy setting. Alternatively, different speed–accuracy settings might also entail changes in executive control, affecting the size of observed binding effects. In this study, we tested these assumptions by comparing binding effects under different speed–accuracy settings that were induced via instructions focusing on speed, accuracy, or both (ambivalent). Binding effects were observed in response times independent of instructions, while in error rates, they only showed under speed or ambivalent instructions. These findings indicate that binding effects can be affected by instructions regarding speed and accuracy.