This study examined emotion recognition ability (ERA) as a predictor of positive and negative affect in two Australian and one German-speaking samples (total N = 469) during the first 2 weeks of major public life restrictions in the COVID-19 pandemic in March/April 2020. Individuals with higher ERA did not report more positive affect, but they felt less burdened and reported less negative affect. This association was fully mediated by lower COVID-19-related media consumption and less negative affect after reading an eyewitness report from an Italian city with a high COVID-19 death toll. However, higher ERA was also related to arguing more with close others. For low-to-medium ERA, an adaptive cognitive emotion regulation style predicted lower media consumption and for medium-to-high ERA, a maladaptive regulation style marginally increased the perceived likelihood of experiencing a similar situation as in Italy, suggesting that regulation style may moderate the ERA–affect relationship.

One of the best-established findings in intelligence research is the pattern of positive correlations among various intelligence tests. Although this so-called positive manifold became the conceptual foundation of many theoretical accounts of intelligence, the very nature of it has remained unclear. Only recently, Process Overlap Theory (POT) proposed that the positive manifold originated from overlapping domain-general, executive processes. To test this assumption, the functional relationship between different aspects of executive attention and the positive manifold was investigated by re-analyzing an existing dataset (N = 228). Psychometric reasoning, speed, and memory performance were assessed by a short form of the Berlin Intelligence Structure test. Two aspects of executive attention (sustained and selective attention) and speed of decision making were measured by a continuous performance test, a flanker task, and a Hick task, respectively. Traditional structural equation modeling, representing the positive manifold by a g factor, as well as network analyses, investigating the differential effects of the two aspects of executive attention and speed of decision making on the specific correlations of the positive manifold, suggested that selective attention, sustained attention, and speed of decision making explained the common but not the unique portions of the positive manifold. Thus, we failed to provide evidence for POT’s assumption that the positive manifold is the result of overlapping domain-general processes. This does not mean that domain-general processes other than those investigated here will not be able to show the pattern of results predicted by POT.

Braver's (2012) dual mechanisms of cognitive control differentiate between proactive control (PMC; i.e. early selection and maintenance of goal-relevant information) and reactive control (RMC; i.e. a late mobilization of attention when required). It has been suggested that higher cognitive capacities (as indicated by reasoning ability as a major characteristic of fluid intelligence) facilitate using the more resource-demanding PMC. We propose the following alternative explanation: engagement in PMC during the completion of reasoning tests leads to better test performance because gained knowledge (i.e. rules learned) during completion of early items is better maintained and transferred to later items. This learning of rules during the completion of a reasoning test results in an item-position effect (IPE) as an additional source of individual differences besides reasoning ability. We investigated this idea in a sample of 210 young adults who completed the AX-Continuous Performance Task (AX-CPT) and the Vienna Matrices Test (VMT). Using fixed-links modeling, we separated an IPE from reasoning ability in the VMT. Based on reaction time (RT) patterns across AX-CPT conditions, we identified three different groups by means of latent-profile analysis. RT patterns indicated engagement in PMC for Group A, mixed PMC and RMC for Group B, and RMC for Group C. With the consideration of the IPE, groups did not differ in their reasoning abilities. However, Group A (engaging in PMC) had a more pronounced IPE than Group C (engaging in RMC). Therefore, we conclude that PMC contributes to a stronger IPE, which in turn leads to higher scores in reasoning tests as measures of fluid intelligence.

Despite the high similarity of reasoning ability items, research indicates that individuals apply different strategies when solving them. The two distinct strategies are response elimination and constructive matching. The latter, frequently showing a positive correlation with reasoning ability, entails the individual systematically investigating the presented problem matrix of an item before scanning the response alternatives. To further understand the sources of individual differences in strategy use during test taking, three different eye-tracking metrics were investigated in participants (N = 210) solving the Raven’s Advanced Progressive Matrices (APM). Relying on the fixed-links modeling approach, bifactor models were fit to the data. The latent model approach revealed, in line with other research, a positive correlation between reasoning ability and constructive matching. The results further indicated that a change in strategy use was correlated with the item-position effect and not reasoning ability. The former exhibited a different direction of effect, depending on the eye-tracking metric analyzed. When investigating the toggle rate, the participants used more constructive matching towards the end of the APM. The proportional time to first fixation on response alternatives indicated less constructive matching as the test progressed, and the proportional time on the problem matrix exhibited no distinct pattern regarding a change in strategy use. These diverging results point towards the possibility of a more nuanced problem-solving behavior than previously assumed. By including the item-position effect in the analyses, the increasing individuals differences in problem-solving behavior can be taken into account, which could be a necessary step in attaining a more comprehensive understanding of problem-solving behavior.