How to measure post-error slowing: The case of pre-error speeding

Anna Foerster,Roland Pfister

doi:10.3758/s13428-021-01631-4

Abstract

Post-error slowing is one of the most widely employed measures to study cognitive and behavioral consequences of error commission. Several methods have been proposed to quantify the post-error slowing effect, and we discuss two main methods: The traditional method of comparing response times in correct post-error trials to response times of correct trials that follow another correct trial, and a more recent proposal of comparing response times in correct post-error trials to the corresponding correct pre-error trials. Based on thorough re-analyses of two datasets, we argue that the latter method provides an inflated estimate by also capturing the (partially) independent effect of pre-error speeding. We propose two solutions for improving the assessment of human error processing, both of which highlight the importance of distinguishing between initial pre-error speeding and later post-error slowing.

Highlights

Research on human performance has a natural tendency to address situations in which actions go awry
These observations might be taken to support the Δpost ∣ E − 1 method of computing post-error slowing because the trial preceding the error might be seen as the best estimate for the current response time level
It would be advisable to gauge for each dataset whether it is possible to use a narrower window to increase the number of trials that can be used for analysis, or whether it might even be advisable to focus on relatively distant pre-error and post-error response times when

Summary

Introduction

These observations might be taken to support the Δpost ∣ E − 1 method of computing post-error slowing because the trial preceding the error might be seen as the best estimate for the current response time level. That is, using the trial immediately preceding an error as a baseline to compute post-error slowing likely incurs an over-estimation of the effect, and this systematic confound can be observed for different experimental paradigms.

Results

Conclusion