Dissecting the mistake: Investigating separable motor and cognitive processes of post-error slowing

People typically slow down following a mistake, a phenomenon known as post-error slowing. The causes of such slowing are not fully understood, and a key question is whether it is merely incidental, maladaptive, or strategically useful for producing subsequent correct responses. Both non-strategic (e.g., motor biases, perceptual distraction) and strategic/cognitive processes (e.g., increased response caution, variability in bias) have been proposed to contribute to post-error slowing. However, it is difficult to delineate these processes when relying on traditional response time measures (e.g., keypresses) that provide a single measure that encompasses both non-strategic and cognitive processes. The current study employed a “touch-and-swipe” response measure to better understand how individuals adjust their response timing post-error. In an object sorting task embedded in an airport security mobile game (Airport Scanner, Kedlin Co.), participants tap on objects (“touch time”) and then swipe them to the top or bottom of the screen (“swipe time”). Previous work with this paradigm (Kramer et al., 2021) suggests that touch time relates more to motor responses and swipe time more to cognitive processes. Here, both touch and swipe time revealed post-error slowing—both subcomponents of total response time showed longer responses following an error. Likewise, both subcomponents were influenced by whether the swipe direction (upward or downward) was consistent with the previous trial with faster responses for repeated swipe directions. Notably, the swipe direction effect was larger for the touch than the swipe time. Finally, the swipe time—but not the touch time—demonstrated an interaction between previous trial accuracy and swipe direction wherein larger post-error slowing occurred when swipe motion repeated. The different outcomes for the touch and swipe response time subcomponents highlight the multifaceted nature of post-error slowing and reveal a distinct cognitive and strategic driver of the slowing.