Validity of the Worst Performance Rule as a Function of Task Complexity and Psychometric g: On the Crucial Role of g Saturation

Abstract

Within the mental speed approach to intelligence, the worst performance rule (WPR) states that the slower trials of a reaction time (RT) task reveal more about intelligence than do faster trials. There is some evidence that the validity of the WPR may depend on high g saturation of both the RT task and the intelligence test applied. To directly assess the concomitant influence of task complexity, as an indicator of task-related g load, and g saturation of the psychometric measure of intelligence on the WPR, data from 245 younger adults were analyzed. To obtain a highly g-loaded measure of intelligence, psychometric g was derived from 12 intelligence scales. This g factor was contrasted with the mental ability scale that showed the smallest factor loading on g. For experimental manipulation of g saturation of the mental speed task, three versions of a Hick RT task with increasing levels of task complexity were applied. While there was no indication for a general WPR effect when a low g-saturated measure of intelligence was used, the WPR could be confirmed for the highly g-loaded measure of intelligence. In this latter condition, the correlation between worst performance and psychometric g was also significantly higher for the more complex 1-bit and 2-bit conditions than for the 0-bit condition of the Hick task. Our findings clearly indicate that the WPR depends primarily on the g factor and, thus, only holds for the highly g-loaded measure of psychometric intelligence.

A meta-analysis of the worst performance rule

“…The worst performance rule has been conceptually replicated in many studies with different reaction time tasks, different samples and different age groups, e.g. in school children (Fernandez, Fagot, Dirk, & de Ribaupierre, 2014), undergraduate students (e.g., Diasco & Brody, 1992;Fernandez et al, RUNNING HEAD: A META-ANALYSIS OF THE WPR 7 2014; Kranzler, 1992;Leite, 2009;Schmiedek et al, 2007;Schmitz, Rotter, & Wilhelm, 2018;Unsworth et al, 2010), and age-heterogeneous community samples (e.g., Fernandez et al, 2014;Frischkorn, Schubert, Neubauer, & Hagemann, 2016;Rammsayer & Troche, 2016;Schmitz & Wilhelm, 2016). In addition, it has been suggested that the size of the worst performance effect, i.e.…”

“…However, except for these two publications, none of the studies on the worst performance rule formally tested the worst performance rule as regression. Only one study statistically tested the difference of correlations between slowest and fastest reaction times (Rammsayer & Troche, 2016), and only a few studies calculated regressions of cognitive ability test performance on multiple RT quantiles to test if slowest reaction times predicted any variance in cognitive ability test performance beyond fastest and/or median reaction times (Fernandez et al, 2014;Salthouse, 1993;1998). The vast majority of studies on the worst performance rule instead only reported the course of correlations between reaction times and cognitive ability test performance across quantiles of the RT distribution and employed no formal test of the worst performance rule.…”

“…However, there are hardly any systematic empirical studies on the role of task complexity and g-loading. One notable exception is a study by Rammsayer and Troche (2016), who demonstrated that the worst performance rule only emerged for a highly g-saturated measure of intelligence, but not for a low g-saturated measure of intelligence. In addition, they found that a manipulation of task complexity, which consisted of varying the number of choice alternatives in a computerized Hick task, moderated the strength of the worst performance rule.…”

“…In addition, it has been suggested that the size of the worst performance effect, i.e. the slope with which the correlation between reaction times and cognitive abilities increases over quantiles of the reaction time distribution, becomes larger with greater task complexity and greater g-loading of the cognitive ability measure (Coyle, 2003a;Kranzler, 1992;Rammsayer & Troche, 2016). Together, these results have influenced process models of intelligence and cognitive development that accounted for bottlenecks in information-processing as an important explanandum (Kovacs & Conway, 2016) or considered changes in information-processing consistency as a central catalyst of cognitive development (Coyle, 2017).…”

Neural anticipatory mechanisms predict faster reaction times and higher fluid intelligence

“…A second group of studies has focused more on the role of attentional or executive lapses in explaining RT‐ability effects and has shown that higher and lower ability participants differ not only in their median RTs but also in their proportion of long‐latency responses. This finding gives rise to the so‐called worst performance rule (Larson & Alderton, ), whereby RT‐ability correlations are often strongest for the slowest RTs (Frischkorn et al, ; McVay & Kane, ; Rammsayer & Troche, ; Schmiedek, Oberauer, Wilhelm, Süß, & Wittmann, ; Unsworth, Redick, Lakey, & Young, ). Finally, it should be noted that the speed of processing and executive control theories are not necessarily mutually exclusive and can both be represented via diffusion models, which construe RT as the outcome of a still lower‐order process of evidence accumulation (McVay & Kane, ; Ratcliff, Schmiedek, & McKoon, ; van Ravenzwaaij, Brown, & Wagenmakers, ; Schmiedek et al, ).…”