Individual differences in cognition, affect, and performance: behavioral, neuroimaging, and molecular genetic approaches

Abstract

We describe the use of behavioral, neuroimaging, and genetic methods to examine individual differences in cognition and affect, guided by three criteria: (1) relevance to human performance in work and everyday settings; (2) interactions between working memory, decision-making, and affective processing; and (3) examination of individual differences. The results of behavioral, functional MRI (fMRI), event-related potential (ERP), and molecular genetic studies show that analyses at the group level often mask important findings associated with sub-groups of individuals. Dopaminergic/noradrenergic genes influencing prefrontal cortex activity contribute to inter-individual variation in working memory and decision behavior, including performance in complex simulations of military decision-making. The interactive influences of individual differences in anxiety, sensation seeking, and boredom susceptibility on evaluative decision-making can be systematically described using ERP and fMRI methods. We conclude that a multi-modal neuroergonomic approach to examining brain function (using both neuroimaging and molecular genetics) can be usefully applied to understanding individual differences in cognition and affect and has implications for human performance at work.

Figure 1

Probability of detection, P(D), as a function of time on task, t, in a 40-minute visual vigilance task. Open circles show the mean detection rates for a group of 20 participants. Fitted line: P(D) = K (e^−at + 1/(1 + e^−bt)), where K is the asymptotic level of performance and a and b are free parameters.

Figure 2

Figure 2A. Group-averaged ERPs (for 16 participants) at three midline electrode sites for attended and unattended stimuli. Reprinted from Figure 3 in Fu et al., NeuroImage, 39, 2008, 1349. Reprinted with permission from Elsevier Inc. Figure 2B. Amplitudes of the P1 component (uV) at the Pz electrode site for attended and unattended stimuli for the 16 individual participants.

Figure 2

Figure 2A. Group-averaged ERPs (for 16 participants) at three midline electrode sites for attended and unattended stimuli. Reprinted from Figure 3 in Fu et al., NeuroImage, 39, 2008, 1349. Reprinted with permission from Elsevier Inc. Figure 2B. Amplitudes of the P1 component (uV) at the Pz electrode site for attended and unattended stimuli for the 16 individual participants.

Figure 3

Individual differences in performance and fMRI responses during a 2-back working memory task. Participant A (left) showed activation in bilateral inferior frontal regions, typically related to working memory performance. An age- and gender-matched participant B (right image) was slower in reaction times and less accurate compared to A. Additional regions including anterior and posterior cingulate (part of the default mode network) were engaged in participant B during the 2-back working memory task.

Figure 4

Top eight versus bottom eight performers in a delayed match-to-sample working memory task. Better performers on the working memory task showed deactivation of posterior precuneus/cingulate (top) during the working memory task, whereas poor performers failed to show the deactivation pattern (bottom).

Figure 5

Speed of lexical decision correlates negatively with fractional anisotropy (FA) in left inferior frontal white matter. (A) Example of regions of interest (ROIs), shown on a representative participant’s FA images. ROIs included white matter of the left inferior frontal (LIF) region, the posterior limbs of the internal capsule (IC), and the optic radiations (OPR). (B) A significant negative correlation was observed between lexical decision reaction time and FA in white matter of the LIF region. Modified from Gold et al. (2007).

Figure 6

Simulated command and control task requiring decision-making under time pressure. Display shows a “sensor to shooter targeting system” with a terrain view containing enemy units, (red), friendly artillery and battalion units (green), and a headquarters unit (blue). Participants had 10 s to identify the most threatening enemy target and select a corresponding friendly unit to engage it, based on military rules of engagement.

Figure 7

A model of individual differences in preferred processing style. A “verbal” individual prefers words and a more automatic information processing style. Hence when materials or emotional stimuli are presented in verbal form, reaction times are faster than when presented as images. The emotional center of the brain (amygdala in red) is proposed to be reached faster via the visual cortex in a “visual” individual. Modified from Childers and Jiang (2008).

Figure 8

Inferior frontal activity of low sensation seekers, and lack of frontal responses in high sensation seekers during repeated visual experience (see circled area). Low-resolution current density reconstructions (CDRs) based on the sLORETA model using a color scale for CDRs as a source of designated time points from 444–500ms, which revealed the largest group differences in ERP responses. Modified from Jiang et al. (2009).