Endocasts are made from the inside of a reassembled fossil skull. Unfortunately this procedure does not capture the convolutions or fissures in the surface of the brain. These greatly expand the brain surface within a given volume, and it is the surface area that is the neurologically active part of the cortex or cerebrum.

The erectus brain could be shown slightly larger to adjust for the relative difference in body size between the two species; the human brain might be shown larger to signify the more convoluted and larger brain surface in modern humans. The arbitrariness of these adjustments suggests why we should not put too much emphasis on brain volume as a direct indicator of functional "intelligence."

The figure shows the endocast of a Homo erectus brain (blue) superimposed on that of a Homo sapiens (red), aligned horizontally on the brain stem under the cerebellum (C), and vertically along the bottom margin of the temporal lobe (T).

Brain function is best inferred from the relative size and form of different brain areas. The erectus brain shows the characteristic "football" shape of hominid brains from Homo ergaster on up. This shape arises principally from a tandem expansion of the frontal (F) and occipital (back, O) lobes in relation to the rest of the brain.

Increases in the frontal lobe appear in Australopithecus africanus and all subsequent hominid brains. This expansion signals a radical change in frontal lobe function, away from olfactory analysis toward complex abstract processing.

Terrence Deacon proposes the frontal lobe as the developmental and cognitive key to human language ability. If so, frontal lobe expansion implies that hominid language abilities may be quite old, perhaps predating the toolmaking abilities that appear in stone artifacts at least 2.4 million years old.

The modern brain shows its greatest expansion in the parietal lobe (P). This expansion accounts for the rounded shape of human skulls in contrast to the flattened "football" form of skulls in earlier species, including Neandertals. Technological, abstract and computational thinking seems to arise in the parietal lobe, and this is the area of greatest relative difference between the two outlines. We might associate this parietal expansion with the appearance of remarkably diverse and refined tool cultures about 90,000 years ago.

The separate evolutionary trajectories of the frontal-occipital and parietal lobes seem genetically based: in modern humans some brain abnormalities affect either frontal/occipital or parietal brain functions, but not both. (It's suggestive to find a similar pattern in australopithecine skulls: A. africanus brains show clear frontal-occipital lobe expansion over A. afarensis, while A. robustus presents an enlarged parietal lobe after that.)

Deacon presents compelling evidence against a "unitary" model of language abilities (as proposed by Chomsky and Pinker) in favor of language as the result of many interdependent processing activities located throughout the brain. Language arising from such a fragmentary architecture would have had an erratic or variable evolutionary history, perhaps appearing gradually as incremental grammars and less generalized systems of reference -- some spoken, some not.

This model is similar to the most recent cognitive theories of "intelligence," which view it not as a unitary mental capacity but as a composite of many discrete cognitive components. If both language and intelligence arise from distributed components -- and are functionally interdependent -- then the evolutionary story of the human mind is more dynamic than the linear increase in raw brain volume implies.

Conclusions based on endocasts are tentative, but linguistic and technical abilities may have had somewhat separate evolutionary histories, responded to different evolutionary pressures, and amplified (through their loose coupling in evolution) the range of evolutionary adaptations open to our ancestral lines.

Information provided by: http://www.handprint.com