The Stroop Effect From a Mixture of Reading Processes: A Fixed-Point Analysis

Abstract

For the last 80 years, the Stroop task has been used to test theories of attention and cognitive control and it has been applied in many clinical settings. Most theories posit that the overwhelming power of written words overcomes strict instructions to focus on print color and ignore the word. Recent evidence suggests that trials in the Stroop task could in fact be a mixture of reading trials and non-reading trials. Here we conduct a critical test of this mixture hypothesis, where a mixture of processes should satisfy the xed-point property (Falmagne, 1968).

An Empirical Exploration of Fast and Slow Errors at Group and Subject Levels in Speeded Decision-making

“…In a broader sense, we believe that our study necessitates a field-wide discussion on what exactly makes an effect a "qualitative benchmark", and what standards should a pattern of data have to meet to become a benchmark that we base theories and models on. For instance, the Stroop effect (see Stroop, 1935;Tillman, Eidels, & Finkbeiner, 2016;Tillman, Howard, Garret, & Eidels, 2017) has been replicated at the group level for the last 80 years (MacLeod, 1991) and is also consistent across individuals (Haaf & Rouder, 2017, suggesting that it is a strong qualitative benchmark for developing theories. In the speeded decision making literature, a reliable effect is the positive skew of RT distributions, which is generally considered to be an ubiquitous trend (Evans, Hawkins, Boehm, Wagenmakers, & Brown, 2017).…”