Intelligence-related Differences Research Articles

A chronometric model of the relationship between frontal midline theta functional connectivity and human intelligence.

Individual differences in cognitive control have been suggested to act as a domain-general bottleneck constraining performance in a variety of cognitive ability measures, including but not limited to fluid intelligence, working memory capacity, and processing speed. However, owing to psychometric problems associated with the measurement of individual differences in cognitive control, it has been challenging to empirically test the assumption that individual differences in cognitive control underlie individual differences in cognitive abilities. In the present study, we addressed these issues by analyzing the chronometry of intelligence-related differences in midfrontal global theta connectivity, which has been shown to reflect cognitive control functions. We demonstrate in a sample of 98 adults, who completed a cognitive control task while their electroencephalogram was recorded, that individual differences in midfrontal global theta connectivity during stages of higher-order information-processing explained 65% of the variance in fluid intelligence. In comparison, task-evoked theta connectivity during earlier stages of information processing was not related to fluid intelligence. These results suggest that more intelligent individuals benefit from an adaptive modulation of theta-band synchronization during the time-course of information processing. Moreover, they emphasize the role of interregional goal-directed information-processing for cognitive control processes in human intelligence and support theoretical accounts of intelligence, which propose that individual differences in cognitive control processes give rise to individual differences in cognitive abilities. (PsycInfo Database Record (c) 2021 APA, all rights reserved).

Efficient hubs in the intelligent brain: Nodal efficiency of hub regions in the salience network is associated with general intelligence

Intelligence-related differences in the intrinsic functional organization of the brain were studied with a graph-theoretical approach, comparing effects on nodal measures of brain network efficiency (concerning specific nodes of the network) and global measures (concerning the overall brain network). Functional imaging data acquired for 54 healthy adult participants during wakeful rest were modeled as graphs representing individual functional brain networks. Nodal and global measures of efficient network organization (i.e., nodal efficiency and global efficiency) were correlated with intelligence scores (IQ from the Wechsler Abbreviate Scale of Intelligence, WASI). While global efficiency showed no significant association with intelligence, the nodal efficiency was significantly associated with intelligence in three brain regions. Participants with higher IQ scores showed higher nodal efficiency in right anterior insula (AI) and dorsal anterior cingulate cortex (dACC), two hub regions of a functional brain network previously described as salience network. Furthermore, higher IQ was associated with lower nodal efficiency in the left temporo-parietal junction area (TPJ). Distinct connectivity profiles were observed for brain regions showing a positive versus negative correlation between IQ and nodal efficiency. Our analyses suggest that intrinsic (i.e., task-independent) connectivity profiles of brain regions that have previously been associated with salience processing (AI and dACC) and the filtering of irrelevant information from higher-level processing (TPJ), play a role in explaining individual differences in intelligence. Based on these intelligence-related effects in resting-state fMRI data, we discuss the potential relevance of processing salient information for the explanation of differences in cognitive performance and intelligence.

Intelligence-related differences in the asymmetry of spontaneous cerebral activity.

Recent evidence suggests the spontaneous BOLD signal synchronization of corresponding interhemispheric, homotopic regions as a stable trait of human brain physiology, with emerging differences in such organization being also related to some pathological conditions. To understand whether such brain functional symmetries play a role into higher-order cognitive functioning, here we correlated the functional homotopy profiles of 119 healthy subjects with their intelligence level. Counterintuitively, reduced homotopic connectivity in above average-IQ versus average-IQ subjects was observed, with significant reductions in visual and somatosensory cortices, supplementary motor area, rolandic operculum, and middle temporal gyrus, possibly suggesting that a downgrading of interhemispheric talk at rest could be associated with higher cognitive functioning. These regions also showed an increased spontaneous synchrony with medial structures located in ipsi- and contralateral hemispheres, with such pattern being mostly detectable for regions placed in the left hemisphere. The interactions with age and gender have been also tested, with different patterns for subjects above and below 25 years old and less homotopic connectivity in the prefrontal cortex and posterior midline regions in female participants with higher IQ scores. These findings support prior evidence suggesting a functional role for homotopic connectivity in human cognitive expression, promoting the reduction of synchrony between primary sensory regions as a predictor of higher intelligence levels.

Emotional intelligence and oscillatory responses to emotional facial expressions

Emotional intelligence-related differences in oscillatory responses to emotional facial expressions were investigated in 48 subjects (26 men and 22 women) in age 18-30 years. Participants were instructed to evaluate emotional expression (angry, happy and neutral) of each presented face on an analog scale ranging from -100 (very hostile) to + 100 (very friendly). High emotional intelligence (EI) participants were found to be more sensitive to the emotional content of the stimuli. It showed up both in their subjective evaluation of the stimuli and in a stronger EEG theta synchronization at an earlier (between 100 and 500 ms after face presentation) processing stage. Source localization using sLORETA showed that this effect was localized in the fusiform gyrus upon the presentation of angry faces and in the posterior cingulate gyrus upon the presentation of happy faces. At a later processing stage (500-870 ms) event-related theta synchronization in high emotional intelligence subject was higher in the left prefrontal cortex upon the presentation of happy faces, but it was lower in the anterior cingulate cortex upon presentation of angry faces. This suggests the existence of a mechanism that can be selectively increase the positive emotions and reduce negative emotions.

Intelligence related differences in EEG-bandpower

Several studies on the relationship between event-related desynchronization/synchronization (ERD/ERS) and cognitive performance revealed contradictory results particularly for the alpha band. Studies from our laboratory have shown that good performers show a larger upper alpha ERD (interpreted in terms of larger cortical activation) than bad performers. In contrast, other researchers found evidence for the neural efficiency hypothesis, which states that more intelligent subjects exhibit a smaller extent of cortical activation, which is assumed to be reflected by a smaller upper alpha ERD. Here we address the question whether these divergent results may be due to differences in general task difficulty. Using a modified version of the RAVEN, individually divided into easy and difficult tasks, a group of average and a group of highly intelligent subjects (IQ− and IQ+) have been investigated. While in the theta frequency IQ+ subjects generally exhibited a significantly stronger activation, we found a significant interaction of task difficulty and IQ group in the upper alpha band, indicating both, a weaker activation for the high IQ group during the easy tasks, and a significant increase from easy to difficult tasks for IQ+ only.

Intelligence related differences in induced brain activity during the performance of memory tasks

Abstract The aim of this study was to investigate the conflicting results in event-related desynchronization/synchronization (Pfurtscheller, 1999) in the upper alpha and theta bands related to intelligence. Two groups of individuals (high intelligent M IQ =128, and average intelligent M IQ =93), solved two memory tasks while their EEG was recorded. The data were analyzed with the ERD/ERS method and low resolution brain electromagnetic tomography (LORETA). In the upper alpha band high intelligent individuals displayed more desynchronization than did average intelligent individuals. In the theta band a minor but reverse effect was observed. The LORETA analysis indicated that average intelligent individuals displayed much higher signal to noise ratios (SNR) than did high intelligent individuals. The EEG source location for the high intelligent group was more anterior, whereas for the average intelligent group a more posterior EEG source location was observed. The findings were explained in the light of the neural efficiency model.

Implicit and explicit learning in young adults with mental retardation.

We examined intelligence-related differences in explicit and implicit learning using an artificial grammar paradigm. Young adults with and without mental retardation completed a sequence-learning and identification task. For some participants, sequences were constructed following an artificial grammar; for others, sequences were random. Explicit learning was determined by ability to learn and later identify random sequences. Implicit learning was determined by the tendency to incorrectly identify new grammatical sequences as seen before, relative to new nongrammatical sequences. Participants with mental retardation did more poorly than participants without mental retardation on explicit learning but just as well on implicit learning. Results suggest that learning of complex materials, when accomplished through implicit processing, is functionally equivalent in individuals with and without mental retardation.

Detection of changes in naturalistic scenes: comparisons of individuals with and without mental retardation.

Abilities of individuals with and without mental retardation to search for and detect salient changes to naturalistic scenes were investigated using the flicker paradigm. Located in areas of central or marginal interest, changes involved an object's color, shape, or presence. Individuals with mental retardation required more time to detect changes of all types, and the magnitude of the group difference was more pronounced for marginal-interest changes. Supplemental eye-tracking data from 6 participants suggested that the basis of this effect was that individuals with mental retardation tended to maintain gaze in the region of central interest for longer periods of time prior to sampling the other areas of the scene. Implications for intelligence-related differences in visual attention are discussed.

Is the phonological loop responsible for intelligence-related differences in forward digit span?

Children with mild mental retardation were compared with CA-matched and verbal-age-matched children without mental retardation on a forward digit span task. Forward digit span is typically used as a measure of the phonological loop (Baddeley, 1986). We hypothesized that intelligence-related differences in forward digit span are largely due to central executive functioning rather than phonological loop functioning. Results showed that the highly significant group difference in forward digit span was reduced to nonsignificance when a measure of central executive functioning was covaried out.

Age-Related and Intelligence-Related Differences in Implicit Memory: Effects of Generation on a Word-Fragment Completion Test

In two experiments, subjects either read a bracketed word in a sentence or generated a word in response to its definition. A word-fragment completion test was then carried out. In Experiment 1, children's priming under the generate condition was substantial, as compared with baseline performance, but was significantly lower than that under the read condition, whereas there was no difference in adults’ priming between these two conditions. Furthermore, prior generation induced an age-related increase in priming despite no age difference under the read condition. In Experiment 2, mentally retarded persons exhibited a profile similar to that of children. These results suggest that there are two different components in implicit memory, one that shows no developmental difference and heavily relies on perceptual processing and the other that shows an age-related or intelligence-related increase and heavily relies on conceptual processing.

Interrelationships among auditory and visual cognitive tasks: An Event-Related Potential (ERP) study

The two studies presented here used event-related potentials (ERPs) to investigate relationships among three elementary cognitive tasks (visual and auditory inspection time and pitch discrimination). This article focuses particularly on contrasts between subgroups with high and low threshold. In one study the participants ( n = 29) performed both auditory inspection time (IT) and pitch discrimination tasks. In the second study, participants ( n = 35) performed a visual IT task. ERPs were obtained during performance of these tasks, and related to measures of the participant's IT or pitch threshold (PT) and intelligence (Raven's APM or Alice Heim 5 Test). The article gives details of a new bootstrap procedure to obtain confidence intervals for ERPs from subgroups differing in threshold or IQ. The ERP waveform contained regions in which individual differences could be related to IT or pitch thresholds. For both visual and auditory IT, the threshold-related part of the waveform fell around 150 to 160 ms after stimulus onset, in the rising phase of the P200 component. In the pitch discrimination task threshold-related differences occurred later, from the peak of the P200 component onwards (200–300 ms after stimulus onset). The ERP results imply that, despite a substantial correlation (+.45) between auditory IT and PT thresholds, different aspects of the stimulus-processing mechanism are critical in these two auditory tasks. In contrast, both IT tasks—which were united in emphasising speed of intake, despite the differences in the stimuli involved—were linked by the ERP results. Some investigators have concluded that pitch sensitivity rather than speed of stimulus intake is tapped by the auditory IT task. The ERP data contradict this view, and lend support to an analysis by Deary (1994b) that distinguished “speed” and “pitch sensitivity” factors in auditory tasks. Intelligence-related differences were found in the VIT-task ERPs around 100 and 150 ms after stimulus onset, but individual differences the auditory-task ERPs showed little relationship to RAPM score. Several factors may underlie this difference in the results for auditory and visual tasks.

Visual search in unidimensional arrays: A comparison between subjects with and without mental retardation

Through the utilization of unidimensional visual search tasks with homogeneous backgrounds, 16 individuals with mental retardation and 16 individuals without mental retardation were tested to determine whether the efficiency of search would vary as a function of dimension (color, form, size, and line orientation, ) and/or level of intelligence. Results indicated group by set size interactions for the dimensions of form and size, indicating intelligence-related differences in search efficiency for these dimensions. In addition, pronounced performance variability within the group of subjects with mental retardation was evident. Results demonstrated no difference in efficiency of search between groups for the dimensions of color and line orientation, thus indicating that the processes involved in speeded visual search were intact for these dimensions. These findings could lead to a greater understanding of the functional parameters of visual information processing within the population of individuals with mental retardation.

Global/local processing in mentally retarded and nonretarded persons

This study examined intelligence-related differences in perceptual processing in a global/local task. Both mentally retarded and nonretarded persons processed global information faster than local information. Furthermore, both groups showed a similar global precedence effect. The effects of extensive practice at naming local features was also examined. Local processing times became faster for both groups as a result of local naming practice. The implication of the results are discussed in the context of global/local processing models as well as processing in other interference tasks (e.g., the Stroop task; Stroop, 1935). These results point to the need to assess task and processing parameters when extending findings of intelligence-related differences across apparently similar tasks.

Intelligence-related differences in the learning, maintenance, and transfer of problem-solving strategies

Intuitive (e.g., Sternberg, Conway, Ketron, & Bernstein, 1981) and informed (e.g., Brown & Campione, 1984; Campione & Brown, 1984; Resnick & Glaser, 1976) perspectives on intelligence emphasize the importance of strategy learning, maintenance, and transfer. There is, however, surprisingly little relevant data, and they conflict (cf. Campione & Brown, 1987; Campione, Brown, & Ferrara, 1982). The purpose of this study was to examine whether intelligence-related differences in strategy learning, maintenance, and transfer exist. Intellectually gifted, average, and mentally retarded children were taught strategies on a version of the balance-scale problem, and were then given additional training until they transferred these strategies to a related version of the problem. Maintenance was assessed 2 weeks after the completion of strategy training. We found that mentally retarded children needed more training to learn and transfer balance-scale strategies, and were less likely to maintain strategies that either gifted or average children. The implications of these findings are discussed.

Detection of stimulus organization: Evidence of intelligence-related differences

Utilizing checkerboard-like stimulus patterns varying with respect to symmetry and number of adjacencies, the effects of stimulus structure on the performances of mentally retarded and nonretarded subjects on a rapid-presentation match-to-sample task were examined. The disparity between target and distractor stimuli was manipulated across high, low, and moderate levels of stimulus organization. Consistent with previous findings (i.e., Caruso & Detterman, 1983) , detection rates for the two groups were equivalent at both high and low levels of distractor organization. Of particular importance, however, was that intelligence-related performance differences were evident only with distractors of intermediate organization. The influence of stimulus structure on stimulus detectability, and the implications of these findings for explaining intelligence-related differences across a wide range of tasks are discussed.

Sustained attention and intelligence

The sustained attention of retarded and nonretarded adults was assessed during two 60-min vigilance tests that differed in memory demand. Subjects compared visually presented digit sets that were separated by 3 or 6 s; consecutive matching digit sets served as the target event. A significant interaction between IQ and time period was obtained for target detection and detectability, p( A) , measures; the performance of retarded subjects declined more rapidly during the latter stages of the vigil than did the performance of nonretarded subjects. Memory load had no effect on target detection, but the 6-s delay resulted in a significant increase in false alarm rate by both IQ groups. Intelligence-related differences in sustained attention appear to be related to the availability of attentional resources during a vigilance task.

Breakdowns in flexible use of information: Intelligence-related differences in transfer following equivalent learning performance

Groups of retarded and nonretarded children were investigated as they learned three rules underlying problems adapted from the Raven Progressive Matrices Test: rotation, imposition, and subtration. The interest was in evaluating group differences in the ease with which the original set of rules was learned and the flexibility with which these rules were transferred to modified contexts. In the case of either learning or transfer, the metric of performance was the amount of aid, specifically the number of computer-generated graduated hints, needed for each subject to solve the problems. The groups were matched on their entering competence to deal with the rules under scrutiny, and also on mental age. No group differences were obtained as the subjects learned the original set of rules. However, on subsequent occasions when required to make flexible use of those rules, clear group differences emerged; and those differences increased as the similarity of the training and test contexts decreased. The results were seen as consistent with theories that emphasize transfer flexibility as one potential source of individual and comparative differences in intelligence.

Differences in intelligence and automatic memory processes

Intelligence-related differences in memory have been viewed as dependent upon control rather than structural memory processes. A test of this notion was made by comparing retarded and nonretarded persons on a subject-paced continuous recognition memory task which was judged to be nonstrategic. The stimuli were computer-generated faces and words. Analyses of pacing rates and posttest interviews provided no consistent evidence for strategy use. Differences in memory between the retarded and nonretarded subjects were attributed to differences in automatic aspects of encoding.

Automatic and conscious processing in retarded and nonretarded adolescents

Educable mentally retarded and nonretarded adolescents participated in incidental learning tasks that emphasized the utilization of processes that were consciously controlled but not deliberately aimed at memory (Experiment 1). Retarded individuals' performance on a standard recognition test was equivalent to that of nonretarded subjects following phonetic encoding and nonstrategic encoding, but was deficient following semantic encoding. Retarded subjects also demonstrated a lower level of performance on a rhyme recognition task. In Experiment 2, retarded subjects provided a pattern of responding identical to that of nonretarded subjects on a picture-word interference task designed to assess automatic processing. The two groups produced equivalent levels of semantic activation. It was argued that the results of the two experiments indicate deficient semantic processing on the part of retarded individuals relative to that of nonretarded individuals that cannot be accommodated by a structural-deficiency model, a developmental-lag model, or a hypothesis that predicts intelligence-related differences only when the task involves the use of deliberate mnemonic strategies.

Intelligence-related differences in semantic processing speed

Educable mentally retarded and nonretarded adolescents verified superordinate and basic level descriptions of common objects. Obtained category decision times, as measured by superordinate-basic level difference scores, were longer for retarded than for nonretarded subjects, indicating fundamental group differences in semantic processing speed. Additional data were interpreted as indicating that retarded adolescents may be slower in making basic level as well as superordinate level decisions, suggesting that they have difficulty in making semantic classification decisions in general. Other results, from a semantic priming task performed by the same subjects, suggested that the obtained group differences in semantic processing speed were related to the active (deliberate) processes required for verification, rather than to category knowledge per se or more passive processes associated with the activation of stored semantic knowledge.