Constructive replication of the visual–perceptual-image rotation model in Thurstone's (1941) battery of 60 tests of mental ability

doi:10.1016/j.intell.2004.12.001

Intelligence

Volume 33, Issue 4, July–August 2005, Pages 417-430

https://doi.org/10.1016/j.intell.2004.12.001 Get rights and content

Abstract

We recently evaluated the relative statistical performance of the Cattell–Horn fluid–crystallized model and the Vernon verbal–perceptual model of the structure of human intelligence in a sample of 436 adults heterogeneous for age, place of origin, and educational background who completed 42 separate tests of mental ability from three test batteries. We concluded that the Vernon model's performance was substantively superior but could be significantly improved. In so doing, we proposed a four-stratum model with a g factor at the top of the hierarchy and three factors at the third stratum. We termed this the Verbal–Perceptual-Image Rotation (VPR) model. In this study, we constructively replicated the model comparisons and development of the VPR model using the data matrix published by Thurstone and Thurstone (1941) [Thurstone, L. L., & Thurstone T. G. (1941). Factorial studies of intelligence. Chicago: University of Chicago Press]. The data matrix was generated by scores of 710 Chicago eighth graders on 60 tests of mental ability.

Section snippets

Sample

For this study, we made use of the data matrix of 60 mental ability tests published by Thurstone and Thurstone (1941) in their Factorial Studies of Intelligence. It was clear that a large battery of tests was necessary in order to generate sufficient detail to replicate our original observations (Johnson & Bouchard, 2005); at the same time, the larger the battery, the stronger the test, as there was less likelihood that we would merely reproduce commonly agreed-upon content factors. The sample

Results

For the fluid–crystallized model (Fig. 1), we obtained a chi-square of 6501.82 on 1654 df, p<0.00001, with RMSEA=0.064 and BIC=−4357.13. The correlation between the two third-stratum factors, representing original fluid and crystallized intelligence, was 0.99, indicating that, in this battery, fluid and crystallized intelligence were effectively equivalent. In addition, the loading of the second-stratum fluid factor on the third-stratum original fluid factor was 0.99, the loading of the

Discussion

The primary goal of this study was to replicate constructively our findings (Johnson & Bouchard, 2005) (1) that Vernon's verbal–perceptual model offers a better description of the structure of intellect than does Cattell and Horn's fluid–crystallized model and (2) that substantial improvement in descriptive power can be gained by explicitly recognizing mental image rotation ability at a level comparable to that of verbal and perceptual ability. In spite of the use of two very demographically

References (15)

M. Hegarty et al.
A dissociation between mental rotation and perspective-taking spatial abilities
Intelligence
(2004)
M. Browne et al.
Alternative methods of assessing model fit
Sociological Methods & Research
(1992)
J.B. Carroll
Human cognitive abilities: A survey of factor-analytic studies
(1993)
R.B. Cattell
Abilities: Their structure, growth, and action
(1971)
J.L. Horn
Models of intelligence
W. Johnson et al.
Testing the grand old models of the structure of human intelligence: It's verbal, perceptual, and visualization (VPZ), not fluid and crystallized
Intelligence
(2005)
D.T. Lykken
Statistical significance in psychological research
Psychological Bulletin
(1968)

There are more references available in the full text version of this article.

Cited by (42)

The notion of validity in experimental crowd dynamics
2023, International Journal of Disaster Risk Reduction
As experimental methods become increasingly popular in crowd dynamics, the question of experimental validity becomes more critical than ever. In fact, there will be no question more fundamental than what makes a crowd experiment valid. For the purposes of design, evaluation and interpretation of crowd experiments, it is paramount that we distinguish between different dimensions of their validity. Here, I differentiate between four categories of experimentation in crowd dynamics based on their underlying purpose of investigation. This includes experiments whose main aim is (i) testing a behavioural theory/hypothesis, (ii) exploring empirical behavioural regularities; (iii) assessing collective (system-level) performance rather than individual behaviour (iv) calibrating and/or validating pre-existing models (including partial/local or full/global model calibration). While delineating various dimensions of external and internal validity as they relate to crowd experiments and enumerating factors that could compromise their validity, I argue how the importance of different validity dimensions varies depending on the investigation's purpose (i.e., the category of experiment). This could provide clarity about circumstances where artificiality/simplicity of laboratory settings is justified as opposed to experiments for which higher levels of contextual fidelity/realism and ecological validity should be given a higher priority. It is critical that investigators (a) be conscious of the trade-off between controllability and realism in their experiment design and (b) identify potential factors that can jeopardise validity of their experiments and take measures to mitigate such threats. Furthermore, it is argued that inferences made from field data do not inherently supersede those of experiments on the issue of validity. Depending on the purpose of investigation, maintaining internal validity sometimes necessitates laboratory settings and justifies artificiality. Naturality of observations per se does not guarantee validity. In the absence of adequate control over the causal relationship, the question of external validity will be futile, as there will be no internally valid inference to be generalised in the first place. Therefore, striking a reasonable balance between internal and external validity of crowd experiments should be made a major design consideration. Overall, the matter of validity is more complicated than the mere creation of a realistic context in the experiment. Studies should not be dismissed merely on the account of ecological validity (contextual realism). The purpose of the design (category of experiment) should be taken into account in making that determination, although the compromise of contextual realism should always be kept to a minimum, and unnecessary compromise should be avoided.
Spearman's hypothesis tested comparing Sudanese children and adolescents with various other groups of children and adolescents on the items of the Standard Progressive Matrices
2016, Intelligence
Spearman's hypothesis tested at the level of items states that differences between ethnic groups on the items of an IQ test are a function of the g loadings of these items, such that there are small differences between ethnic groups on items with low g loadings and large differences between ethnic groups on items with high g loadings, and it has been confirmed in a fairly substantial number of studies. Most biological variables show correlational patterns that are more similar to the pattern in group differences than non-biological variables. In this paper, we tested Spearman's hypothesis, comparing a group of Sudanese children and adolescents (N = 7226) with other groups of children and adolescents from Denmark, Cyprus, Croatia, Bosnia, South Africa, Estonia, Ukraine, Ireland, Russia, and Chile (total N = 13,105). The analyses were carried out on 19 comparisons between the Sudanese children and the other children. Aggregating all data points from the present study showed that Spearman's hypothesis was strongly confirmed with a sample-size–weighted r with a value of .70; it is argued that applying corrections for statistical artifacts would make this value substantially larger. These outcomes suggest that biological variables, including brain-related variables, are more important than non-biological variables in explaining group differences.
Cognitive deficits at age 22 years associated with prenatal exposure to methylmercury
2016, Cortex
Prenatal exposure to mercury has been associated with adverse effects on child neurodevelopment. The present study aims to determine the extent to which methylmercury-associated cognitive deficits persist into adult age. In a Faroese birth cohort originally formed in 1986–1987 (N = 1,022), prenatal methylmercury exposure was assessed in terms of the mercury concentration in cord blood and maternal hair. Clinical examinations of 847 cohort members at age 22 years were carried out in 2008–2009 using a panel of neuropsychological tests that reflected major functional domains. Subjects with neurological and psychiatric diagnoses were excluded from the data analysis, thus leaving 814 subjects. Multiple regression analysis included covariates previously identified for adjustment. Deficits in Boston Naming Test (BNT) and other tests of verbal performance were significantly associated with the cord-blood mercury concentration. Deficits were also present in all other tests applied, although most were not statistically significant. Structural equation models were developed to ascertain the possible differences in vulnerability of specific functional domains and the overall association with general intelligence. In models for individual domains, all of them showed negative associations, with crystallized intelligence being highly significant. A hierarchical model for general intelligence based on all domains again showed a highly significant negative association with the exposure, with an approximate deficit that corresponds to about 2.2 IQ points at a 10-fold increased prenatal methylmercury exposure. Thus, although the cognitive deficits observed were smaller than at examinations at younger ages, maternal diets with contaminated seafood were associated with adverse effects in this birth cohort at age 22 years. The deficits affected major domains of brain functions as well as general intelligence. Thus, prenatal exposure to this marine contaminant appears to cause permanent adverse effects on cognition.
Comparison of methods for factor extraction for cognitive test-like data: Which overfactor, which underfactor?
2016, Intelligence
Research published in Intelligence showed that the number of factors measured by individual intelligence tests has increased dramatically over time (Frazier & Youngstrom, 2007). When “gold standard” methods (parallel analysis based on principal components analysis, PA-PCA, and minimum average partial, MAP) were applied to these same tests fewer factors emerged, leading the authors to conclude that tests were inappropriately overfactored, and that modern tests are measuring far fewer underlying constructs than they are intended to measure. The article was influential, with the findings cited commonly and used to guide subsequent analyses. Here, we tested a key assumption of Frazier and Youngstrom and subsequent research that has used these methods: whether MAP and PA-PCA are accurate in recovering the correct number of factors with cognitive-test-like data. MAP and PA-PCA were compared to other exploratory and confirmatory factor analytic techniques in their ability to recover the correct number of factors in simulated data. Data conformed to common values found in intelligence literature, and varied based on the number of tests per factor, magnitude of factor loadings and factor correlations, and sample size. Results showed that MAP and PA-PCA, in fact, underfactored under many realistic data conditions, meaning that they recovered too few factors. Confirmatory methods were more accurate. Among exploratory methods PA based on principal axis factoring (not principal components analysis) was most accurate, although all methods underfactored with few tests per factor and high factor correlations. These findings suggest that PA-PCA and MAP are not “gold standard” methods for determining the number of factors underlying intelligence data and that other methods are more accurate. We argue for the importance of formal and informal theory in factor analytic investigations of intelligence tests.
Intelligence is multidimensional: Theoretical review and implications of specific cognitive abilities
2015, Human Resource Management Review
Human resource management researchers typically treat cognitive ability as a unidimensional construct. The current paper reviews possible rationales for this choice, including practical convenience, the parsimony of Spearman's theory of general mental ability (g), positive manifold among cognitive tests, and empirical evidence of only modest incremental validity of specific cognitive abilities for predicting job and training performance over and above g. In contrast to HR researchers' dominant practice of treating cognitive ability as unidimensional, we recommend a renewed interest in narrower, second-stratum cognitive abilities. The renewed focus on multiple dimensions of intelligence is supported by several arguments, including superior empirical fit of hierarchical and oblique multifactor models over unidimensional models of cognitive test data, the Cattell–Horn–Carroll theoretical model and Carroll's large-scale empirical support for a hierarchical model of intelligence with several second-stratum factors (i.e., specific cognitive abilities), empirical evidence of modest incremental validity (typically at or above 2%) of specific cognitive abilities predicting job performance beyond g, the notion of a compatibility principle of the cognitive ability–job performance relationship in which specific abilities should predict specific criteria but not broad criteria, application of bifactor and relative importance methodologies to predict job performance via g and specific abilities simultaneously, evidence that adverse impact in hiring can be partly curtailed by differentially weighting specific cognitive abilities, and theoretical models of reciprocal causation among specific cognitive abilities which can explain positive manifold in the absence of g. After arguing for multidimensional models of intelligence, we review a variety of second-stratum cognitive abilities that have been described under the Cattell–Horn–Carroll model, highlighting similarities and differences among specific abilities.
A Janus View: J. P. Das's Ideas in Retrospect and Prospect
2015, Cognition, Intelligence, and Achievement: A Tribute to J. P. Das
A Festschrift is an opportunity to reflect on and honor the achievements of a respected academic scholar. J. P. Das is certainly such a person. To understand his contributions, one needs to put them in a historic context. Accordingly, we first comment on the contributions of two major figures who provided the basis for Das’s programmatic research agenda: Alexander Luria (1902–1977) and Lev Vygotsky (1896–1934). We follow this with illustrations of Das’s work on the assessment of cognitive functions and educational intervention/remediation programs, drawing on work that one of us (JC) and Das did collaboratively. Finally, we consider what Das’s research focus and the PASS model might look like if it were just now being developed.

View all citing articles on Scopus

View full text

Constructive replication of the visual–perceptual-image rotation model in Thurstone's (1941) battery of 60 tests of mental ability

Abstract

Section snippets

Sample

Results

Discussion

Intelligence

Alternative methods of assessing model fit

Sociological Methods & Research

Human cognitive abilities: A survey of factor-analytic studies

Abilities: Their structure, growth, and action

Models of intelligence

Testing the grand old models of the structure of human intelligence: It's verbal, perceptual, and visualization (VPZ), not fluid and crystallized

Intelligence

Statistical significance in psychological research

Psychological Bulletin