Reaction times and speed-of-processing: Their relationship to timed and untimed measures of intelligence

Eighty-one university students were given a battery of reaction time tests, measuring the speed with which they performed various elementary cognitive processes, and a group test of intelligence which yielded timed and untimed scores. Multiple regression analyses indicated that speed of information-processing was an equally good predictor of timed and untimed intelligence test performance, and that various combinations of speed-of-processing variables accounted for as much as 48% of the variance in intelligence test scores. Different speed-of-processing variables varied relatively systematically in the extent to which they correlated with the individual subtests and scales of the intelligence test.     

Is speed of information processing related to fluid or to crystallized intelligence?

Speed of information processing was examined in relation to the theory of fluid and crystallized intelligence. Sixty sixth grade children completed the Standard Progressive Matrices test and the Mill Hill Vocabulary scale, both untimed, as measures of fluid and crystallized intelligence, and three RT tasks: Memory Scanning, Picture Identification, and Sentence-Picture Comparison. As predicted, RT parameters, with the exception of slope on two tasks, were negatively correlated with scores on both tests. There were no consistent differences between the two tests in the size of correlations with RT, and partial correlations, eliminating the effects of fluid and crystallized intelligence respectively, failed to support any causal relationship between fluid and crystallized intelligence in either direction. Some limitations of the research, specifically in the possible effect of error rates on the speed-intelligence relationship, and in the application to individuals of task parameters derived from group performance, are noted.     

Sex differences in visual-spatial ability: The role of performance factors

Two studies were designed to explore the role of performance factors as sources of the frequently noted higher male scores on visual-spatial ability tests. In the first study, the mental rotations test (MRT) was administered to male and female college students of equally high quantitative ability (based on SAT math scores). Although males had significantly more correct responses on the test than did females, their advantage was eliminated when the ratio of correct responses to items attempted was used as the dependent measure. In the second study, the same test was administered to new groups of male and female college students. In this sample, the males had significantly higher SAT math scores. The MRT was administered under standard, timed conditions and under untimed conditions. Both raw and ratio scores were calculated. With SAT math score as the covariate, analyses of covariance indicated that males demonstrated higher performance in the timed, raw-score condition but not in the untimed or in the ratio-score conditions. The theoretical and social policy implications of these studies are discussed.     

Comparing subtest profiles of g loadings and correlations with RT measures

It is argued that if a response time (RT) measure correlates with psychometric test (PT) scores because it shares variance in common with general intelligence, g, then the profile of g loadings for a set of PTs would be predictable from the profile of correlations between the RT measure and the PT scores. On the other hand, if an RT correlates with PTs through variance not shared with g, the g loadings of the PTs should be unrelated to the correlations between RT and PTs. The profiles of g loadings and correlations with RT were compared for a set of 6 RT measures and 8 PT scores from Smith and Stanley (1983). To demonstrate statistical significance, standard errors of the statistics were generated by Efron's bootstrap technique (Efron, 1979). It was clearly shown that the profile of the PT's g loadings could be well predicted from the RT-PT correlations for four of the RT measures. It can be concluded that RT tasks do measure general intelligence. Analysis of the errors in prediction suggested that the RTs may correlate more with fluid than crystallized intelligence.     

Speed of information processing and general intelligence

This study investigates the relationship between a number of measures of speed of cognitive information-processing and intelligence test scores. One hundred university students were given five tests of speed-of-processing, measuring their speed of encoding, short-term memory scanning, long-term memory retrieval, efficiency of short-term memory storage and processing, and simple and choice reaction time or decision-making speed. They were also given the Wechsler Adult Intelligence Scale and the Raven Advanced Progression Matrices. A number of multiple regression analyses show that the cognitive processing measures are significantly related to IQ scores. Other analyses indicate that this relationship cannot be attributed to the common content shared by the reaction time and the intelligence tests, nor to the fact that parts of the WAIS are timed. It is concluded that the reaction time tests measure basic cognitive operations which are involved in many forms of intellectual behavior, and that individual differences in intelligence can be attributed, to a moderate extent, to variance in the speed or efficiency with which individuals can execute these operations.     

Clocking g: Relating intelligence and measures of timed performance

The relationships between intelligence test scores and measures derived from reaction time (RT) and perceptual speed procedures were investigated in 137 twelve-year-old students with IQs ranging from 59 to 142. A range of intelligence tests were used and the scores factor analyzed to produce general, spatial and verbal factors. Test and factor scores were correlated with perceptual speed and with measures taken from 2, 4, and 8 choice RT tasks using a response keyboard upon which the subject's fingers directly rested, thus avoiding interpretive problems associated with a “home key.” Inspection time correlated poorly with intelligence. Only three of the RT measures produced correlations greater than .25 with the general factor. These were the slope of Hick's law, B, (correlation −.28), the 8 choice mean RT, RT8, (−.33) and the 8 choice standard deviation, SD8 (−.41), compared with the average intercorrelation between the intelligence measures of .40. Test-retest correlations of the RT measures, taken over a year for half the subjects, were low as reliability measures, with .35 for B, .52 for RT8, and .48 for SD8. Correlations of RT measures with spatial scores were not significantly greater than with verbal scores, suggesting that whatever relationship exists is with a general factor rather than only a spatial one.     

THE BRITISH ABILITY SCALES SPEED OF INFORMATION PROCESSING SUBTEST: WHAT DOES IT MEASURE?

Summary . Recent theory and research has focused on the relationship between speed of performing elementary cognitive operations and general intelligence. The developers of the British Ability Scales (BAS) included the Speed of Information Processing (SOIP) subtest as a measure of mental processing speed. To test the validity of the SOIP subtest, a group of 12-year-old children were given the BAS short-form, including the SOIP subtest, and a series of electronically timed reaction time (RT) tasks. Correlations between RT and SOIP were higher than correlations between RT and other BAS subtests, suggesting convergent and discriminant validity. A hierarchical factor analysis of the data shows that the BAS SOIP subtest is related more to speed of apprehension (RT) than to speed of making the motor response, or movement time (MT).     

The comparability of intelligence test results: Group- and individual-level comparisons of seven intelligence tests

A significant body of research has demonstrated that IQs obtained from different intelligence tests substantially correlate at the group level. Yet, there is minimal research investigating whether different intelligence tests yield comparable results for individuals. Examining this issue is paramount given that high-stakes decisions are based on individual test results. Consequently, we investigated whether seven current and widely used intelligence tests yielded comparable results for individuals between the ages of 4–20 years. Results mostly indicated substantial correlations between tests, although several significant mean differences at the group level were identified. Results associated with individual-level comparability indicated that the interpretation of exact IQ scores cannot be empirically supported, as the 95% confidence intervals could not be reliably replicated with different intelligence tests. Similar patterns also appeared for the individual-level comparability of nonverbal and verbal intelligence factor scores. Furthermore, the nominal level of intelligence systematically predicted IQ differences between tests, with above- and below-average IQ scores associated with larger differences as compared to average IQ scores. Analyses based on continuous data confirmed that differences appeared to increase toward the above-average IQ score range. These findings are critical as these are the ranges in which diagnostic questions most often arise in practice. Implications for test interpretation and test construction are discussed.     

Implications of the Cattell-Horn-Carroll theory on ethnic differences in IQ

The Cattell-Horn-Carroll theory of intelligence and the Spearman-Jensen hypothesis were applied to help interpret mean IQ differences among ethnic groups. The Cattell-Horn-Carroll theory classifies cognitive abilities on three strata, which differ by magnitude of generality including specific, broad, and general abilities. The Spearman-Jensen hypothesis advances the belief that mean IQ differences between African Americans and Euro-Americans occur as a function of the test's g loadings. Although IQ differences between African Americans and Euro-Americans occur, the differences should be larger on tests with higher g loadings and smaller on tests with lower g loadings. The analysis described in this paper may serve as the catalyst to encourage all intelligence test publishers to supply test users with data regarding mean IQ differences among various ethnic groups.     

Investigating Retesting Effects in a Personnel Selection Context

Repeat applicants to a fire‐fighter position undertook the same cognitive ability and situational judgment tests on multiple occasions and the resultant practice effects were investigated. Practice effects of approximately two fifths of a standard deviation were observed between Times 1 and 2 on timed tests of Abstract Reasoning and Mechanical Comprehension. Smaller practice effects were observed, however, on a timed test of numerical comprehension ability and the untimed Teamwork Knowledge Skills and Abilities situational judgment test. For all four tests, the additional practice effect observed at Time 3 were smaller and nonsignificant; a result supplemented by latent growth curve analyses.     

Eliminating the IQ-RT correlation by eliminating an experimental confound

Three experiments investigated the relation between visual scanning demands, reaction time (RT), and psychometrically defined intelligence (IQ). Prior studies have shown reliable correlations between RT and IQ in the range of −.20 to −.80. However, these studies have confounded the number of possible stimuli (stimulus uncertainty) with the size of the area in which the stimuli may appear (visual angle). Experiment 1 replicated these studies retaining this confound. As the number of stimuli increased from one to eight, the visual angle was permitted to increase as well (from 0° to 30°). The results showed that RT varied in accord with Hick's (1952) law, and a median correlation between IQ and six RT parameters (subjects' mean RTs and standard deviations at three levels of stimulus uncertainty) of −.47 was observed. Experiment 2 removed the confound, varying only stimulus uncertainty, and the median IQ-RT correlation declined to −.02. Experiment 3 held stimulus uncertainty constant at 1 bit (two stimuli) and varied visual angle; a median correlation of −.19 was observed. It was concluded that many of the previously reported correlations may not have hinged on speed of information processing alone, but at least in part on subjects' abilities to scan the display across which the stimuli appeared.     

The broad factor of working memory is virtually isomorphic to fluid intelligence tested under time pressure

How much overall variance in fluid intelligence (Gf) can be predicted by four working memory (WM) functions: storage capacity, attention control, relational integration, and updating was investigated under time pressured Gf testing. Confirmatory factor analysis indicated that the broad WM factor, which was subsumed by these four WM functions, shared 83.4% of variance with Gf tested under pressure, whereas a reanalysis of previous data with the same model showed that only 58.2% variance was shared with virtually untimed Gf tests. Moreover, in timed Gf tests, only the easiest, early items contributed to the WM-Gf correlation, whereas in untimed tests also the hardest, late items were linked with Gf. These results suggest that the measurement of “fast” intelligence primarily taps the functions of WM, whereas “slow(er)” intelligence depends also on some other cognitive processes beyond WM.     

Effects of divided attention in the word-fragment completion task with unique and multiple solutions

A study with 114 young adults investigated the correlations of intelligence factors and working-memory capacity with reaction time (RT) tasks. Within two sets of four-choice RT tasks, stimulus–response compatibility was varied over three levels: compatible, incompatible, and arbitrary mappings. Two satisfactory measurement models for the RTs could be established: A general factor model without constraints on the loadings and a nested model with two correlated factors, distinguishing compatible from arbitrary mappings, with constraints on the loadings. Structural models additionally including factors for working memory and intelligence showed that the nested model with correlated factors is superior in fit. Working-memory capacity and fluid intelligence were correlated strongly with the nested factor for the RT tasks with arbitrary mappings, and less with the general RT factor. The results support the hypothesis that working memory is needed to maintain arbitrary bindings between stimulus representations and response representations, and this could explain the correlation of working-memory capacity with speed in choice RT tasks.     

Personal tempo, personality, cognitive speed and cognitive performance

One hundred young males filled in an untimed personality questionnaire. The time needed was taken to be the subjects' personal tempo. They also worked twice on a timed substitution test; their score on this test was taken to be their cognitive speed. It was hypothesized that personal tempo was related to personality and cognitive speed to performance on cognitive tests. The results confirmed the hypothesis. Those whose personal tempo was faster had higher Extraversion and lower Neuroticism and Lie-scores. Those whose cognitive speed was faster, however, did better on four out of five cognitive tests and had a higher Lie-score. It seems that the two speed factors, as hypothesized by Spearman, are indeed two separate and distinct factors, each correlated with different variables: cognitive and personality.     

THE EFFECT OF PERFORMANCE TIME AND RETESTING UPON THE FACTOR STRUCTURE OF INTELLIGENCE TESTS

Changes in the factor structure of intelligence tests between early and later stages of performance and between testing and retesting were studied. In addition to factor analyses, correlations between the test scores of various time periods and the final factor scores were computed. The principal findings were the powerful influence of the verbal ability in the initial stages of most tests, the gradual 'purification' of the factors, and the occurrence of a shift of the highest factor loadings of some tests from one factor to another, particularly in tests with increasing item difficulty.     

Do clever old people have earlier and richer first memories?

In a study of 377 people, from 54 to 83 years old, it was shown that the age of earliest memories was unrelated to current age but was related to intelligence. People with high IQ test scores remembered first and second memories at an earlier age than did those with lower IQ test scores. Intervals between first and second memories were briefer for those with high scores. The content of early memories was unrelated to age or intelligence.     

Word identification in adults with mild mental retardation: does IQ influence reading achievement?

OBJECTIVE: To assess the relationship between word identification and intelligence in adults with mild mental retardation (IQ < 80). METHOD: A standardized evaluation was administered to 67 adults with mild mental retardation. The evaluation included a psychiatric interview, the WAIS-R, and a 2-h interview with a speech therapist and reading tests. RESULTS: Causes of mental retardation were diverse in the sample, and IQ scores ranged from 43 to 79 (mean score = 64). All subjects exhibited reading impairment, including 69% with severe impairment. No subject with an IQ score under 65 was able to perform adequately in the word identification tasks. Word identification was correlated with total and verbal IQ, but not with performance IQ. CONCLUSION: Our results suggest that, in contrast to subjects with normal intelligence, IQ score is correlated with reading in subjects with mild mental retardation. Finally, remediation should be preferentially implemented for subjects with IQ score greater than 65.     

Measures of timed performance and intelligence

The question of an association between IQ and measures of timed performance derived from inspection time and reaction time was examined in a sample of 182 adults and by reanalyzing data involving 48 adults from a previously published study. Multiple regression analysis found that measures of timed performance accounted for as much as 25% of IQ variance in the normal population, but that the inclusion of borderline and mildly retarded subjects resulted in much higher correlation coefficients because of the markedly less efficient performance of these persons in tasks of this kind. This outcome raised doubts about the validity of combining data from retarded and nonretarded subjects. Results ran counter to claims that tasks of the kind used are largely uninfluenced by cognitive variables, so that findings are not necessarily explained satisfactorily in terms of a mental speed factor. It was concluded that these measures of timed performance do not, at this time, provide a basis from which a reliable culture-fair measure of intelligence might be devised.     

IQ stability: the relation between child and young adult intelligence test scores in low-birthweight samples

The stability of IQ from childhood to adulthood in low-birthweight subjects was measured in two independent samples with follow-up intervals of approximately 14 and 9.5 years. In both samples, intelligence was assessed with the WISC at a mean age of 9.5. Twenty-six subjects were retested with the WAIS at a mean age of 23.5, and 78 subjects with the BPP (the Danish Military Draft Board Intelligence Test) at the age of 19.1. Both samples obtained childhood and adult test scores below the expected means. For the Wechsler Verbal, Performance and Full-Scale IQs, the stability quotients were 0.86, 0.86, and 0.89 in the WAIS sample, and the retest correlations for the three IQs with the BPP score were 0.66, 0.65, and 0.74. Thus, the majority of children showed stable patterns of intellectual development from middle childhood to young adulthood.     

Recurrent severe hypoglycemia, intelligence, and speed of information processing

In a previous study, the cumulative effect of recurrent severe hypoglycemia on cognitive function was examined in 100 patients with insulin-treated diabetes and a significant correlation was observed between the frequency of severe hypoglycemic episodes and the difference between their current IQ and estimated premorbid IQ (Langan, Deary, Hepburn, & Frier, 1991). The study here extended this model for examination of the role of biological environmental influences on intelligence. Eighty-five of the original cohort of 100 patients were recalled and tested on Wechsler Performance IQ, a test of premorbid IQ, the Hick Reaction Time task, the Sternberg Memory Scan task and a test of Rapid Visual Information Processing (RVIP). The patients' frequency of severe hypoglycemia was assessed using a structured interview. Both IQ-decrement estimates and frequencies of severe hypoglycemia proved reliable after 18 months (correlations > .75) and the corrected correlation between these two variables was about .4. Therefore, repeated episodes of acute severe neuroglycopenia are associated with a lowering of IQ. Frequency of severe hypoglycemia was significantly related to Hick decision times and to response thresholds in the RVIP task. Factor analysis suggested that hypoglycemia affects decision or response-initiation processes rather than encoding, storage, comparison, or classification processes in short-term memory. Indices of these short-term memory processes were highly related to IQ-type test scores. This study also confirms, in a sample with a near-normal distribution of IQ, that measures of psychometric intelligence are significantly correlated with indices from elementary cognitive tests, especially RVIP task, which is based on signal-detection theory.