"Is it not reasonable to anticipate that our understanding of the human mind would be greatly aided by knowing the purpose for which it was designed?"
George C. Williams[1]
"Even I will admit that at some point, in the story of human evolution, circumstances conspired to encourage mentality at our modern level."
Steven J. Gould[2]
The evolution of the human brain has been one of the most significant events in the evolution of life, yet many fundamental questions remain to be answered. The fossil record, in concert with a comparative neuroanatomical analysis of closely related species, shows that the hominid brain increased in size more than 3-fold over a period of around 2.5 million years. Other important qualitative changes occurred as well, though at least some would appear to be simple correlated responses to changes in overall size. In principle there are at least three ways to explain the increase: 1) unintentional side-effects of selection on other aspects of hominid biology (some form of pleiotropic effect), 2) random change (genetic drift), or 3) selection for some sort of increased behavioral ability. However, given what we know about vertebrate biology and the evolutionary process in general, and about the evolutionary costs of neural tissue in particular, an adaptive explanation for these changes is the only likely answer.
If this dramatic increase in brain size occurred because of selection for behavioral ability, then what kind of behavioral ability was selected for? This is the underlying question driving the research in this study. Suggestions such as hunting, tool-making, throwing, language, 'intelligence', and 'behavioral complexity' are often found in introductory anthropology textbooks, but without any direct empirical support. While it is true that all of these abilities differ in Homo sapiens as compared to other primates, all of them are quite complex in terms of the basic cognitive abilities they require. They are composed of a number of simpler, more basic cognitive processes, and are not properly thought of as self-contained, independent wholes. For example, hunting and throwing both involve 'spatial abilities' of some kind, and certainly involve the ability to form accurate visual representations in the brain. Tool making requires memory of a particular sequential order of actions. Language requires hierarchical thinking, associative memory, and a host of other basic cognitive abilities. Furthermore, it is unlikely that each of these higher-level cognitive abilities requires a unique set of lower-level processes. That is, success at tool making might well depend on processes which also lend themselves to success at language, for example. Given this, we should expect that selection for a complex behavioral ability (like language) would result in better functioning lower-level processes.
How might one delineate these simpler cognitive processes experimentally? Beginning with the work of Francis Galton (a contemporary and cousin of Charles Darwin), the discipline of differential psychology has focused on the delineation of individual differences in various cognitive dimensions. One salient finding of this line of inquiry has been that all tests that can be considered 'cognitive' (meaning that they are not obviously related to physical strength, sensory acuity, or other simple physical capacities) correlate with each other much more often than can be explained by chance with negative correlations being very rare, and never very large (Jensen 1987). Furthermore, various subsets of these cognitive tests form natural clusters (i.e., are highly correlated with one another). The most parsimonious interpretation of these findings is that different tests are tapping the same set of underlying cognitive abilities, and these are then given descriptive labels (e.g., 'spatial ability') based on the types of cognition they seem to require. There are of course other interpretations, but this is a reasonable working assumption.
While higher-level abilities, such as complex cooperative hunting, may well have been the immediate selective sieve underlying hominid brain size evolution, it is unlikely that there evolved a whole new set of brain nuclei devoted exclusively to this ability (the possible exception being language, but see below). Instead, a more plausible evolutionary explanation would suggest that basic cognitive processes improved with increasing brain size, albeit under the selective sieve of one or more of these complex behaviors, thereby improving a whole range of higher cognitive functions. Thus a great amount of behavioral ability and complexity could have evolved without direct and independent selection for each of the complex behaviors like language and hunting.
The actual genetic, physiological mechanisms by which brain size increase occurred are not known. One intriguing possibility is that point mutations increased the efficiency, speed, reliability, etc. of basic physiological processes that play a key role in synaptic transmission and plasticity. For example, N-methyl-D-aspartate (NMDA) receptors have been implicated in various forms of long-term learning and memory (Morris and Davis 1994). Because of differences both in the types of learning that NMDA receptors are associated with, as well as in the neuroanatomical locations where this learning occurs, it would appear that NMDA receptors function simply by detecting associations between presynaptic and postsynaptic activity (Morris and Davis 1994). Since there is evidence that synaptic connections even in mature brains are use-dependent (see, e.g., Artola and Singer 1994), and further, since neuron cell death (which prunes back early developmental neuronal overproduction) is also dependent on some form of use-dependent feedback (see Purves 1988), it is plausible that even small changes in the efficacy of basic physiological pathways involved in synaptic transmission (such as those in which NMDA receptors play a role) could result in increases in overall neural volume. In other words, by making the feedback more effective, fewer neurons will be pruned in the first place and existing connections will spur even more synaptic growth between existing neurons.[3]
In any case, no matter how brain size increase was accomplished, and no matter how complex behaviors are processed in the brain, it is clear that if selection for one or more behavior is the explanation for the increase in brain size in hominids (and/or any of the other neuroanatomical differences seen in modern humans) we should expect to find associations between variation in these neuroanatomical features and variation in cognitive abilities associated with these complex behaviors. The reason for this follows from the logic of evolutionary change in the face of reproductive costs: this kind of change occurs only when a statistical reproductive advantage accrues to individuals who are more similar to the modern human condition for the trait in question. Only in those generations in which this reproductive advantage accrues can there be any change in the population characteristics of this trait. Furthermore, as the trait changes over time, its associated evolutionary costs will usually increase as well. When the trait in question no longer has any benefit, there will be selection against it, thereby making any threshold effects (i.e., in which brain size increases beyond some critical point have no effect on behavior) unlikely. This will be discussed in detail in the next chapters.
All of this suggests that there should be an association between the various neuroanatomical differences in modern humans and some sort of increased behavioral ability, and it is this question that is the focus of the present study. One obvious place to start looking for correlates of neuroanatomical variability is in the basic dimensions of cognitive ability that have already been delineated by psychologists, but a comparative perspective on brain-behavior relationships suggests additional possibilities. A number of studies that have addressed various aspects of this question will be reviewed, and the unresolved issues that form the rationale for this study will be highlighted.
The present study extends the search for behavioral correlates in three main ways. First, it addresses the question of whether brain size has an actual causal effect on behavioral differences, instead of just being correlated with them because of common environmental influences, or because of various mating patterns. Second, the range of behavioral dimensions addressed has been expanded to include areas of specific interest to theories of behavioral evolution (including cognitive tasks that activate pre-frontal cortical tissue and basic components of linguistic processing). Finally, the possible effect of rate-of-maturation as a counterbalancing 'cost' of the evolution of brain size is addressed. In addition, measurement of various neuroanatomical variables was accomplished through magnetic resonance imaging (MRI), using a protocol which results in an almost 4-fold improvement in a spatial resolution compared to the next-highest resolution MRI/behavior study published to date.
Specifically, I have investigated the within- and between-family relationships in normal individuals between: 1) various aspects of brain anatomy derived from high resolution magnetic resonance imaging (MRI), and 2) a wide variety of cognitive abilities. Precise measurements of overall brain volume, volume of the cerebrum, volume of the gray and white matter regions of the cerebrum, volume of the brain stem and cerebellum, as well as a measure of gray/white matter differentiation, were extracted from MRI brain scans performed on 36 pairs of sisters. The interrelationships between these neuroanatomical variables and a diverse array of cognitive/behavioral dimensions were calculated using a variety of statistical techniques. The results suggest that there are no simple answers to the basic questions of hominid neuroanatomical evolution.