Schoenemann, P. Thomas, 2000, "The effects of different brain indices on brain/behavior relationships: a within-species study of humans," (abstract) American Journal of Physical Anthropology,  Supplement 30:273.


It is well known that brain size scales with body size.  Various indices have been proposed that allow for the estimation of brain size independent of body size (e.g., Jerison’s "encephalization quotient" or EQ).  An implicit assumption is sometimes made that these body size corrected brain measures are a better measure of inherent behavioral differences than are simple differences in absolute brain size.  This may partly be due to the fact that, for both absolute brain size and brain-size-as-a-percentage-of-total-body-weight, Homo sapiens do not exceed other species, even though we are clearly extremely successful mammals on purely behavioral grounds.  Some research, however, is consistent with the idea that absolute brain size, in and of itself, has important behavioral implications.  In addition, some comparative brain/behavior studies have used indices which scale one part of the brain with the some other part (or the brain as a whole).  In any case, which brain indices are the most important behaviorally is an empirical question that can only be decided with actual behavioral data: a given brain index might be more highly associated with some behavioral dimensions than it is with others, and some brain indices might be more highly related to particular behavioral dimensions than are other indices.

This raises the question of whether brain/behavior correlations in humans might differ significantly depending on how brain variables are scaled.  The present study addresses this issue by assessing whether different brain indices vary with respect to their associations with different behavioral measures.  Brain size (obtained from high resolution MRI) and behavioral variability (assessed through a battery of cognitive and behavioral tasks) were obtained from 72 healthy human females.  Correlations were then calculated using different brain indices.  The results suggest that the use of different scaling measures does not significantly affect the strength of brain/behavior associations within humans.  The implications of these findings for understanding hominid brain evolution will be discussed.