The heritability of fluid and crystallized intelligences: By the mava design and oses analysis

R. B. Cattell, J. M. Schuberger, F. M. Ahern, V. Kameoka

Research output: Contribution to journalArticlepeer-review

4 Scopus citations


The present research has three distinguishing characteristics among behaviour genetic researches in this area: (1) use of the MAVA rather than the Twin Method, with advantages regarding assumptions, (2) psychologically, the separate measurement of fluid and crystallized general intelligences, gf and gc, instead of an undifferentiated g, and (3) uses of the OSES (Overlapping Simultaneous Equation Sets) in analysis for solutions. The data is on 470 brothers reared together, 94 identical twins reared together, 124 fraternal twins reared together, and 1973 boys (fluid intelligence) and 2973 (crystallized intelligence) unrelated. All were in the 12‐18 years age range. The MAVA method used employed the equations for the six concrete variances that could be used together. Only one six set and four sets of five from among the six finally proved algebraically and numerically soluble, in terms of the required abstract variances. The alternative of least squares fit solution is being presented also elsewhere. The conclusions are (1) that past work by the twin method has been seriously in error in equating inter‐sib (or fraternal twin) to inter‐identical twin environmental differences, the latter here turning out to be about half the former; (2) that Burt's heritability of 0.80 is an upper limit, applying to the twin method estimates of environmental difference only ‐ the heritability of both fluid and crystallized intelligence across the general population turns out to be around .60; (3) that the genothreptic correlations, not previously soluble, are substantially negative and support the law of coercion to the bio‐social mean; (4) that a theory of duo‐threpsis, pointing to differential action of two kinds of environment in producing threptic variance, best fits the finding that the within family heritability is no higher for fluid than crystallized intelligence whereas the between family heritability is decidedly higher; and (5) that sampling errors and some contrasts with least squares solutions, indicate that it would be desirable to seek still larger samples if heritabilities are to be determined finely enough to permit further theoretical advance. 1981 Australian Psychological Society

Original languageEnglish (US)
Pages (from-to)355-374
Number of pages20
JournalAustralian Journal of Psychology
Issue number3
StatePublished - Dec 1981

All Science Journal Classification (ASJC) codes

  • General Psychology


Dive into the research topics of 'The heritability of fluid and crystallized intelligences: By the mava design and oses analysis'. Together they form a unique fingerprint.

Cite this