The strong link between fluid intelligence and working memory cannot be explained away by strategy use

Interindividual Differences in Cognitive Variability Are Ubiquitous and Distinct From Mean Performance in a Battery of Eleven Tasks

Our performance on cognitive tasks fluctuates: the same individual completing the same task will differ in their response's moment-to-moment. For decades cognitive fluctuations have been implicitly ignored - treated as measurement error - with a focus instead on aggregates such as mean performance. Leveraging dense trial-by-trial data and novel time-series methods we explored variability as an intrinsically important phenotype. Across eleven cognitive tasks with over 7 million trials, we found highly reliable interindividual differences in cognitive variability in every task we examined. These differences are both qualitatively and quantitatively distinct from mean performance. Moreover, we found that a single dimension for variability across tasks was inadequate, demonstrating that previously posited global mechanisms for cognitive variability are at least partially incomplete. Our findings indicate that variability is a fundamental part of cognition - with the potential to offer novel insights into developmental processes.

Should Intelligence Tests Be Speeded or Unspeeded? A Brief Review of the Effects of Time Pressure on Response Processes and an Experimental Study with Raven's Matrices

Intelligence tests are often performed under time constraints for practical reasons, but the effects of time pressure on reasoning performance are poorly understood. The first part of this work provides a brief review of major expected effects of time pressure, which includes forcing participants to skip items, convoking a mental speed factor, constraining response times, qualitatively altering cognitive processing, affecting anxiety and motivation, and interacting with individual differences. The second part presents data collected with Raven's matrices under three conditions of speededness to provide further insight into the complex effects of time pressure, with three major findings. First, even mild time pressure (with enough time available for all participants to complete the task at a leisurely pace) induced speeding throughout the whole task, starting with the very first item, and participants sped up more than was actually required. Second, time pressure came with lower confidence and poorer strategy use and a substantial decrease of accuracy (d = 0.35), even when controlling for response time at the item level-indicating a detrimental effect on cognitive processing beyond speeding. Third, time pressure disproportionately reduced response times for difficult items and participants with high ability, working memory capacity, or need for cognition, although this did not differentially affect ability estimates. Overall, both the review and empirical sections show that the effects of time pressure go well beyond forcing participants to speed or skip the last few items and make even mild time constraints inadvisable when attempting to measure maximal performance, especially for high-performing samples.

The Development of Relational Reasoning: An Eyetracking Analysis of Strategy Use and Adaptation in Children and Adults Performing Matrix Completion

Relational reasoning is a key component of fluid intelligence and an important predictor of academic achievement. Relational reasoning is commonly assessed using matrix completion tasks, in which participants see an incomplete matrix of items that vary on different dimensions and select a response that best completes the matrix based on the relations among items. Performance on such assessments increases dramatically across childhood into adulthood. However, despite widespread use, little is known about the strategies associated with good or poor matrix completion performance in childhood. This study examined the strategies children and adults use to solve matrix completion problems, how those strategies change with age, and whether children and adults adapt strategies to difficulty. We used eyetracking to infer matrix completion strategy use in 6- and 9-year-old children and adults. Across ages, scanning across matrix rows and columns predicted good overall performance, and quicker and higher rates of consulting potential answers predicted poor performance, indicating that optimal matrix completion strategies are similar across development. Indices of good strategy use increased across childhood. As problems increased in difficulty, children and adults increased their scanning of matrix rows and columns, and adults and 9-year-olds also shifted strategies to rely more on consulting potential answers. Adapting strategies to matrix difficulty, particularly increased scanning of rows and columns, was associated with good overall performance in both children and adults. These findings underscore the importance of both spontaneous and adaptive strategy use in individual differences in relational reasoning and its development.

Cognitive strategies in matrix-reasoning tasks: State of the art

Cognitive strategies in matrix-reasoning tasks have been investigated for the last decade and a half. Several steps were made since the first paper in the field, but the advances have been sparse and with little connection. Here we present a review of the state of the art in this subject. We introduce how this topic was born and how to measure these strategies, covering the diverse methods and measures that are presented in the literature to progress on this subject and the applications that were developed, as well as the knowledge that resulted from these applications. Furthermore, the future directions are discussed with the intention to engage new researchers in this topic, as well as to bring awareness to limitations that were found given the available scientific literature.

Subject-performed task effect on working memory performance in children with autism spectrum disorder

A previous study found that children with autism spectrum disorder (ASD) have better recall when they perform instructions (subject-performed task [SPT]) than when they passively hear instructions (verbal task [VT]) in a working memory task for instructions, an effect that is called the SPT effect. This study explored whether the SPT effect exhibited by ASD children is caused by the movement component or by processing materials twice. More importantly, this study explored whether intelligence influences the SPT effect exhibited by ASD children. ASD children with three levels of intelligence (N = 56) and a control group, children with intellectual disability (ID) who had low intelligence but did not have ASD (N = 21), were asked to perform working memory tasks for instructions under VT, SPT and repeated (hearing the instruction twice) conditions. No significant difference in performance was observed between the VT and repeated conditions, regardless of the child's level of intelligence. ASD children with lower-middle intelligence exhibited a smaller SPT effect than ASD children with upper-middle intelligence. Critically, while ASD children with low intelligence did not exhibit the SPT effect, an ID group with equivalent low intelligence did show this effect. Therefore, these results show that the SPT effect for ASD children is caused by the movement component and is uniquely associated with a certain level of intelligence, namely, lower middle and higher levels of intelligence. LAY SUMMARY: In ASD children, the benefit of physically performing instructions for working memory performance is uniquely associated with a certain level of intelligence. Only ASD children with lower-middle intelligence (and higher) benefit from physically performing instructions, and higher intelligence increases this benefit; ASD children with low intelligence do not show this benefit. This benefit in ASD children is attributed to the additional motoric code generated by physical performance.

Age, Education and Intellectual Quotient Influences: Structural Equation Modeling on the study of Benton Visual Retention Test (BVRT)

This study searched for sociodemographic influences on visual memory and visuoconstructive ability in healthy and clinical samples evaluated with Benton Visual Retention Test (BVRT) in two studies. In Study 1, we searched for changes related to age in children, adolescents, adults and elderly on the performance of the BVRT. In Study 2, we investigated the relations among age, years of education and intellectual quotient (IQ) on the performance of the BVRT using Structural Equation Modeling (SEM). Participants were 624 individuals aged between six and 89 years old (M = 25.40; SD = 22.34) from the normatization and evidence validity studies at Brazil. We used a sociodemographic questionnaire, BVRT and IQ measure was estimated. Study 1 has shown a performance similar to the developmental graphics with a U-inverted pattern in relation to age: An increase of the visual memory ability in the children and adolescent groups as age increases, a tendency of a decrease in the performance in the adult group that intensifies in the elderly group. Study 2 found that the model for the BVRT performance tested by SEM denoted satisfactory goodness-of-fit indexes, χ2/gl = 2.67, p < .001; CFI = .92; TLI = .93; RMSEA = .004, 90% CI = [.03, .05];WLSMV = 1.79, and corroborated the theoretical assumption. The SEM model confirmed in this study highlight the strong role of years of education in the prediction of BVRT scores.

Scoring Alternatives for Mental Speed Tests: Measurement Issues and Validity for Working Memory Capacity and the Attentional Blink Effect

Research suggests that the relation of mental speed with working memory capacity (WMC) depends on complexity and scoring methods of speed tasks and the type of task used to assess capacity limits in working memory. In the present study, we included conventional binding/updating measures of WMC as well as rapid serial visual presentation paradigms. The latter allowed for a computation of the attentional blink (AB) effect that was argued to measure capacity limitations at the encoding stage of working memory. Mental speed was assessed with a set of tasks and scored by diverse methods, including response time (RT) based scores, as well as ex-Gaussian and diffusion model parameterization. Relations of latent factors were investigated using structure equation modeling techniques. RT-based scores of mental speed yielded substantial correlations with WMC but only weak relations with the AB effect, while WMC and the AB magnitude were independent. The strength of the speed-WMC relation was shown to depend on task type. Additionally, the increase in predictive validity across RT quantiles changed across task types, suggesting that the worst performance rule (WPR) depends on task characteristics. In contrast to the latter, relations of speed with the AB effect did not change across RT quantiles. Relations of the model parameters were consistently found for the ex-Gaussian tau parameter and the diffusion model drift rate. However, depending on task type, other parameters showed plausible relations as well. The finding that characteristics of mental speed tasks determined the overall strength of relations with WMC, the occurrence of a WPR effect, and the specific pattern of relations of model parameters, implies that mental speed tasks are not exchangeable measurement tools. In spite of reflecting a general factor of mental speed, different speed tasks possess different requirements, supporting the notion of mental speed as a hierarchical construct.