Arthur R. Jensen (2000) Behavioral and Biological Phenomena Equally "real" and Related. Psycoloquy: 11(018) Intelligence g Factor (31)

Volume: 11 (next, prev) Issue: 018 (next, prev) Article: 31 (next prev first) Alternate versions: ASCII Summary
PSYCOLOQUY (ISSN 1055-0143) is sponsored by the American Psychological Association (APA).
Psycoloquy 11(018): Behavioral and Biological Phenomena Equally "real" and Related

Reply to Partridge on Jensen on Intelligence-g-Factor

Arthur R. Jensen
Educational Psychology
School of Education
University of California
Berkeley, CA 94720-1670


Partridge's criticisms labor under his misconception that I consider biological phenomena to be more "real" than behavioral or psychological phenomena, and he appears to be unaware of the recent research on the physical correlates of IQ and g.


behavior genetics, cognitive modelling, evoked potentials, evolutionary psychology, factor analysis, g factor, heritability, individual differences, intelligence, IQ, neurometrics, psychometrics, psychophyiology, skills, Spearman, statistics
1. I have not claimed in "The g Factor' (Jensen, 1998; 1999) or elsewhere that I consider biological structures or variables to be any more real than behavioral phenomena, systematically observed or measured. Both domains are equally valid matter for scientific study. But I am not a philosophic dualist, as some behavioral scientists appear to be, at least implicitly. Behavior in an environment is mediated by biological structures, existing mostly in the brain. Nor do I accept the tabula rasa notion that individual differences in abilities, including cognitive abilities, are wholly shaped by the individual's environmental circumstances.

2. One of the main points of my book is that individual differences in abilities, particularly general mental ability, or g, as I have defined it, must ultimately be explained in terms of physical processes. A strictly psychological explanation of the striking phenomenon of individual differences in terms of conditioning and learning in different environments is far from adequate and, in fact, is absolutely contradicted by much evidence. A large proportion of the variance in cognitive abilities, especially when distilled as psychometric g, is attributable to genetic variance, which necessarily implies that biological structures, mostly in the brain, are a large part of the causal basis of individual differences, even though at present we have not yet identified specifically the neural processes, brain chemistry, or whatever, that cause individual differences in various cognitive functions and the fact that they all are correlated to some extent. They all share a common source of variance which is psychometrically defined as g.

3. The chief aim of future research on g theory is to discover the physiological factors that account for g. Psychological or psychometric theorizing, can only describe g in terms of the characteristics of tests that are the most highly g loaded -- what Spearman (1927) referred to as defining g by site rather than by nature -- or merely making statements about g rather than explaining its cause, which Spearman believed must be sought in terms of the brain, "whereby physiology will achieve the greatest of all its triumphs" (p. 407).

4. I beg to differ with Partridge's (1999, par.# 2 & 3) opinion that cranial volumes and brain volumes are unreliably measured. External measures of the skull are somewhat more attenuated estimates of brain size than are direct measurements of the brain via magnetic resonance imaging (MRI). But imperfect reliability of measurement means that the obtained correlations of head-size or brain-size with IQ are lower than would be expected if the measurements were perfectly reliable. Unreliability can only diminish the correlation. So if one knows precisely the reliability coefficient of the brain measurements, then the correlation with IQ can be divided by the square root of the reliability coefficient to get a more accurate estimate of the true correlation between brain and IQ. Yet even without this correction, the correlation between IQ and brain size as measured in vivo by MRI averages about +.40 (controlling for body size) in the dozen or so published studies. I don't know of any other physical characteristic that is more highly correlated with IQ. But .40 is considered a substantial correlation in psychological research; it is about the same as the correlation between school children's IQs and the socioeconomic status or occupational level of their parents. (The correlation between adult IQ and occupational level is closer to .70.)

5. Of course, no one has claimed that processes are entities. At each higher level of complexity there are emergent properties that are absent at lower levels of complexity. I have elaborated on this point elsewhere (Jensen, 1997). But this information is revealed in the course of a proper reductionist analysis of a phenomenon. Reductionism in the physical and biological sciences, as well as in psychology, has such an outstanding batting average as compared to holistic thinking that it would be foolish to abandon it in the study of g. There is nothing in Partridge's (par.#5 &6) argument about integrated levels of neural organization being responsible for the emergence of g (i.e., correlated individual differences in a host of cognitive abilities) that contradicts anything I have ever written. I am simply advocating that we try to discover just what these hierarchical levels of integration consist of at the neurological level. And here, I believe, it may pay, as a heuristic research strategy, to distinguish between understanding the neural mechanisms involved in various specialized cognitive functions (learning, memory, etc.), on the one hand, and the basis of individual differences in measures of these functions and their intercorrelations, on the other.

6. The important advances in modern neuroscience could be characterized as reductionist. (For an excellent review of just how this approach has advanced our knowledge of the neural processes involved in memory, see McGaugh [2000].) But this obviously doesn't imply that a neurophysiological explanation of human intelligence can be discovered by learning more and more about single neurons. It could well be the case, however, that biochemical differences in the neurons in certain areas of the brain, or their density, or amount of dendritic arborization, or conduction velocity, or glucose metabolic rates, or concentrations of neurotransmitters, or any combination of these variables, would account for some substantial part of the individual differences variance in the various cognitive abilities we call "intelligence" and in their communality or g. My book (Jensen, 1998) points out enough relationships of g to a number of brain phenomena such as these to warrant the claim that such empirically tractable brain phenomena are an entering wedge for exploring the nature of g in physical terms and indeed merit further research, although certainly not to the exclusion of other as yet unexpected brain correlates of psychometric g.


Jensen, A.R. (1997). The neurophysiology of g. In C. Cooper, & V. Varma (Eds.), Processes in individual differences (pp. 108-125). London: Routledge.

Jensen, A.R. (1998). The g factor: The science of mental ability. Westport, CT: Praeger.

Jensen, A.R. (1999). Precis of: "The g Factor: The Science of Mental Ability" PSYCOLOQUY 10(23). psyc.99.10.023.intelligence-g-factor.1.jensen

McGaugh, J.L. (2000). Memory: A century of consolidation. Science, 287, 248-251.

Partridge, T.(1999). The g factor and the role of biology in complex behavior. PSYCOLOQUY 10(60) psyc.99.10.060.intelligence-g-factor.10.partridge

Spearman, C.E. (1927). The abilities of man. London: Macmillan.

Volume: 11 (next, prev) Issue: 018 (next, prev) Article: 31 (next prev first) Alternate versions: ASCII Summary