ABSTRACT Why do modern humans have larger brains than earlier people such as Homo erectus? Since large brains cause obstetrics, infancy and locomotion problems, the advantage they offer must be substantial. One might link this advantage to increased IQ. But there is a problem: individuals exist with psychometrically normal IQ but Homo erectus sized brains (MRI volumetric surveys outliers, microcephaly and hemispherectomy). Why did evolution increase brain size (with all its complications) when humans (as these individuals demonstrate) can have normal IQ without bigger brains? Here I argue the advantage links to increased capacity for expertise, an aspect of intelligent behaviour not measured by IQ tests, that both links to brain size and would have been critical to the survival of our simple hunter-gatherers ancestors.
KEY WORDS: brain size, IQ, expertise, MRI volumetrics, microcephaly, hemispherectomy, individual differences, Homo erectus, human evolution.
I. THE COGNITIVE PUZZLE OF OUR BIG BRAINS
1.Human brains expanded roughly by half between Homo erectus and modern Homo sapiens sapiens. The brains of modern people are roughly 1371 cc (male), and 1216 cc (female). In contrast, those of Homo erectus were roughly 930 cc (SD 130 cc, range 736 cc to 1172 cc, based upon the 28 cranial remains of Homo erectus intact enough to allow estimates of brain volume [Stanyon, Consigliere & Morescalchi, 1993; cranial to brain volume adjusted with equation 4 from Aiello & Dunbar, 1993]). What aspect of early cognition drove this increase in size?
2.First, we should note that brain expansion beyond that of Homo erectus size causes neonatal, obstetric and female location handicaps (reviewed below). Thus, whatever selected for brain size increase must have offered a very strong compensating benefit. Second, we should note that we know from clinical sources that modern people can have brains no larger than Homo erectus yet have normal IQ scores. Thus, the compensating benefit offered by large brains is unlikely to be intelligence as measured by IQ, since why should evolution have increased brain size with all its associated problems for something that Homo erectus sized brains can possess without expansion?
3.Have we here an anomaly --or are these reports of individuals with small brains but normal intelligence upon closer examination incorrect? I shall argue neither -- what we have here is a scientific window to understand the nature of the cognition for which evolution increased early human brain size. This is because psychometrically measured IQ leaves out many cognitions relevant to intelligent behaviour. In particular, IQ tests by design ignore one well studied area of intelligence, known to be independent of IQ scores: expertise (Ackerman, 1996; Ericsson & Lehmann, 1996; Ceci & Liker, 1986). I argue, (i) the capacity for expertise correlates absolutely with brain size -- to have more expertise capacity, a brain must be bigger (expertise capacityreflects the number of 'information chunks' that can be created and processed --something that links to the number of the cortical columns in the brain). Further, since (ii) expertise is reported by anthropologists to play a critical role in the survival of simple hunter-gatherers, that early humans would have been under a natural selective pressure to increase their capacity for expertise. We, thus have an empirically testable window into the past as to whether it was expertise capacity (rather than IQ) that selected for increased brain size.
4.This conjecture has probably been previously overlooked because it crosses three different (and largely non-communicating) areas of science. First, the palaeobiology of the problems posed by the enlargement of the Homo brain. Second, the medical and MRI evidence that individuals with brain size as low as Homo erectus can exist with psychometrically measured normal IQ -- and indeed live otherwise unremarkable normal nonretarded lives. (A fact, in view of the presumed link between brain size and IQ, many palaeoanthropologists refuse to believe). Third, evidence that expertise is a component of intelligent behaviour that is (i) independent of the cognitions measured by IQ tests, (ii) holds an absolute link to brain size, and (iii) plays a key role (as recorded by anthropologists) in the survival and so fitness of simple hunter-gatherers. The primary focus of this paper is to review and bring together in one place these three separate areas of evidence. Hopefully, the electronic nature of Psycoloquy will bring scholars in these various areas together.
II. LARGE BRAINS, LARGE PROBLEMS.
5.The size of most but not all body parts can be increased without causing problems. For instance, if it should have advantaged earlier hominids that their legs or arms should increase or decrease 50% in length, no strong disadvantage (apart from changed nutrition needs for their physiological development and maintenance) would impede the natural selection of such different body proportions. This, however, is not the case with the hominid brain. The expansion of the brain to modern dimensions, from those of Homo erectus is known to have had severely disadvantageous knock-on effects upon infancy, birth and female locomotion.
6.(A). Enlarging Homo erectus brains required developmentally extending brain growth into the first postnatal year. As a result, a modern human infant's initial 12 months of life resemble that of a foetus outside the womb. This period of extreme immaturity would have substantially increased the burden and complexity of parenting infants (Martin, 1982).
7.(B). Not all brain expansion can be delayed into the postnatal period. The human brain in order to achieve its later size is already compared to other apes highly enlarged before birth. In consequence, the human infant's head at birth is anatomically a tight fit in the birth canal. This creates considerable obstetric complications (Rosenberg, 1992). These obstetric problems become particularly significant when brain size increased with Homo erectus beyond 801 cc (Martin, 1982; I have changed his 850 cc figure for cranial volume into a brain volume to aid comparability with other volumes; Aiello and Dunbar, 1993, equation 4 [given in note 1]).
8.(C). To minimise these problems, the female pelvis has increased the dimensions of the birth canal. Such anatomical adaptations however have been achieved at the cost of a female's pelvis being nonoptimal (though still functional) for efficient bipedal locomotion -- (a phenomena that can be seen in the different performance of females compared to men in athletic events dependent upon lower limbs but not upper ones [such as discus throwing]) (Day, 1992).
9.These disadvantages of increasing brain size argue that some component of cognition increased by brain size must have outweighed them. Given that intelligence is central to the behaviour of our species, and that intelligence is traditionally linked to its measurement by psychometric IQ tests, we would expect brains size to link to normal IQ.
10. Indeed, moderate correlations exist between IQ and brain size (Wickett, Vernon, & Lee, 1994). However, outliers to this moderate IQ brain size link exist in the clinical and MRI volumetric literature (see below) and these cause theoretical problems. Namely, if small brains can exist in the modern world with modern levels of intelligence, why not a million years ago? Would not such small but bright brains have been selected in preference to large bright ones to avoid the above knock-on problems associated with big brains? This anomaly however only arises if we assume that IQ tests measure the cognitions that advantaged brain size increase in early people. As we shall see, an overlooked aspect of intelligent cognition, expertise capacity, exists that is a much better candidate. This is (i) not measured by IQ (indeed it is independent of IQ scores), (ii) central to early human survival, and (iii) directly linked to brain size. Thus, instead of an anomaly, we have a new way of empirically investigating the cognition central to human evolution. First, I will review the MRI volumetric and clinical evidence for normal IQed individuals with very low brain size. Then I will review expertise capacity and its importance for the survival of early humans.
III. SMALL BRAINED BUT NONRETARDED.
11. Research of three kinds suggests that small brained people can have normal IQs: (i) a recent MRI survey upon brain size (Giedd et al,. 1996), (ii) reports upon individuals born with microcephaly -- defined as head circumference 2 SD below the mean --(general review of microcephaly and normal intelligence, Dorman, 1991); and (iii) case studies upon individuals with a dysfunctional cerebral hemisphere that was removed when young -- hemispherectomy (Smith & Sugar, 1975; Griffith & Davidson, 1966; Vining et al., 1993).
III.1 VOLUMETRIC SURVEYS UPON BRAIN SIZE
12. Jay Giedd and colleagues of the National Institute of Mental Health, Child Psychiatry Branch received 624 responses to a newspaper advertisement for a MRI brain scanning study of 4-18 year-olds (1996). This group was carefully screened with psychometric tests, and a psychiatric interview. Those with a learning disorder (or family members with one) were excluded. Of the 624 responses only 112 meet their stringent criteria for 'normality'. After MRI scans, volumes for various brain areas were measured. Striking variance was found. Of the 104 individuals which successfully completed their scans, volume for the cerebral hemispheres ranged from 735 cc (a 10 year-old male) to 1470 cc (a 14 year-old male) (taken from scatter diagram, Fig. 4). Unfortunately, Giedd has not supplied total brain volumes but these can be inferred. The cerebral cortex makes up only 86.4 % of brain volume when measured by MRI (Filipek, Richelme, Kennedy & Caviness, 1994) so the total brain volume of the 10-year old would be larger at 850.7 cc. Brains at 10-years are about 4.4% smaller than adult size (Dekaban & Sadowsky, 1978) suggesting that as an adult his brain will grow to 888 cc. Even using the lower figure of 80% cerebrum to brain ratio derived from anatomical studies suggests a figure of only 960 cc.
The finding of normal people with small brains is not without precedence in earlier cranial circumference surveys.
13. In a group of 1,006 school aged children, Clifford Sells of the University of Washington School of Medicine (1977) found 19 -- nearly 2% -- were microcephalics. On tests of Academic Achievement (measured by the Comprehensive Test of Basic Skills) nine performed above 50% (one individual scored at 81%). Of the 19, 12 had also been measured on standardised IQ tests, (seven students were in school districts that did not do intelligence tests); of these 12, seven had average or above age IQs -- one indeed had an IQ of 129.
14. Of 188 children with microcephaly (of mixed origins including Down syndrome) studied by Edward Sassaman and Ann Zartler at the Child Development Center at Rhode Island Hospital (1982), 60 -- 39.1% --were not retarded and 13 -- 7% -- were of average IQ. In two subgroups, there were more nonretarded individuals: amongst those with normal body growth, 48%, and amongst those with growth failure and head breadth deficiency, 44.4%, were nonretarded. The breakdown of the last group is interesting since in this group of 18 individuals, the Stanford-Binet Intelligence Scale, Wechsler Intelligence Scale for Children-Revised and the Bayley Scales of Infant Development (Mental Scale) showed four had normal intelligence (22%), two had below average intelligence, two were borderline and ten retarded (five mild, three moderate, none severe or profound, plus two 'untestable'). Thus while microcephaly is strongly associated with retardation and low intelligence, a small but significant percentage -- 7% overall, and 22% in one subgroup -- show average intelligence in spite of their small brains.
II.2. MICROCEPHALIC INDIVIDUALS
15. In the late 1960s, several paleoanthropologists (Holloway, 1968; Lenneberg, 1967) noted that microcephalic individuals could speak in spite of possessing very small brains. All such cases however were mentally retarded. More recently, clinical reports have shown microcephalics need not be retarded and often possess normal IQs.
16. Livia Rossi and colleagues (1987) of the University Medical School of Milan reported a genetic study covering 21 autosomal dominant microcephalic adults and children of six families. Psychometric tests were given to 13 of them and (except for one individual) found them to have nonretarded IQs. (Several microcephalic individuals, it should be noted, refused to do IQ tests, as the research observed, 'because they perceived themselves as perfectly normal' and resented the implication that they were not (p. 656)!) Of the 13 which allowed their IQ to be measured, a 29 year old mother, referred to as C2, is notable for both the extreme smallness of her brain (4.7 SD below a control group) given her psychometric above average mental abilities.
17. C2, indeed, had the smallest brain amongst the 21 microcephalics. In spite of this, she had a Wechsler IQ measured at 112 and a performance upon Raven's matrix task described as 'good'. Her school record notes she was good in mathematics though inadequate in 'italian literature'. At present her employment is described as semiskilled but this is typical for people in her area.
18. Unfortunately, though she has a head 4.7 SD smaller than average, the exact size of her brain is unknown. Only her cranial volume and head circumference were measured together with those of a control group of 15 women. This found her to have a head circumference of 48.8 cm, 4.7 SDs below that of a control group range of 53.3 and 56.2 cm (with deviation from the mean between -1 and +1.4 SD). For comparison, reference data for US children finds that the average female child at two years has a head circumference of 48.6 cm and 54.4 cc at 18 years (Roche, Mukherjee, Guo & Moore, 1987). C2's cranial volume was estimated by Moesch's procedure (displaced water with head immersed to the upper rim of the orbit and occipital protuberance [which as result includes the volume of a person's hair]). The volume of the heads by this measure of the control of 15 women was 2118ñ158 cc, C2's was 61% of this at 1300 cc. On the surface this may seem an inadequate guide to brain volume given the average female brain has a volume only of about 1216 cc. But head size includes skull, and meninges surrounding the brain; these are usually described as filling up roughly one centimetre around the brain (as can be seen in any brain scan which includes the surrounding skull), and, of course, this volume also includes hair volume. This would account for apparent discrepancies: for instance the volume of a sphere one centimetre radius smaller than one which would give a volume of 2118 cc is 1416 cc, if 1.345 centimetres smaller it would be 1216 cc (the average volume of a female brain). Comparison with the control group (assuming they have average brain volume, and that the measure of head volume parallels brain volume) would suggest C2 has a brain volume somewhere near 741 cc. An alternative approach would be to note that the average volume of a female brain is 1216 cc with a 2 SD of 194 cc which would suggest a brain at 4.7 SD would be 760 cc. It would thus seem unlikely that her brain approached anywhere near normal size and was possibly on the low side for Homo erectus.
19. Medline searches under microcephaly and normal intelligence (or normal IQ) finds 21 other reports of microcephaly and normal intelligence, though unfortunately most papers do not report psychometric or volumetric data.
20. One report of normal intelligence exists in a brain smaller than Homo erectus. According to Burt Wilder (1911), Daniel Lyon was a nonretarded white watchman who worked for 20 years at the end of the nineteenth century in New York at the Pennsylvania Railway Terminal. He could read, write, and according to legal representatives of the company that employed him 'there was nothing defective or peculiar about him, either mentally or physically'. He was of average weight 65.8 kg though of below average high 1.55 m. After he died aged 46 in 1907 from bronchitis his brain was removed and subject to a professional autopsy with 'accurate scales'. This found it weighed just 680 grams (624 cc assuming a specific gravity of 1.09 for fresh brain).
21. Upon examination Burt Wilder could not attribute the small size of Daniel Lyon's brain to either pathology or atrophy. Indeed, the only unusual feature he noted was that the cerebellum (part of the brain attached to the brainstem) was near normal size. This suggests that the volume of Daniel Lyon's cerebral hemispheres might have been small even for his already small brain. Indeed, the total size of his cerebral hemispheres, 371 cc is 128 cc less than the 499 cc -- 80% -- which would be expected of a normally proportioned brain of 624 cc.
22. All these cases have been average or only slightly above average in intelligence. Perhaps small brains prevent extremely high levels of intelligence. But distinguished modern people have had brains within the size range of Homo erectus. For instance, the brain of the Noble Prize winning novelist Anatole France (1844-1924) weighted 1017 grams (933 cc) at post mortuum (Gould, 1981, p. 92). One qualification to this figure is that brains shrink slightly with age and Alzheimer's disease. Though a 80-year old, he was non-demented (he had married one Mademoiselle Emma Laprevotte only a few months before he died). Estimates for brain shrinkage over the 20 to 80 year age range for men vary in six studies between 3.9% and 8.6% (Dekaben & Sadowsky, 1978) suggesting his brain when young could have been as high as 1013 cc -- but still well within the SD range of Homo erectus, 930 cc ñ 130 cc.
23. A rare treatment for brain pathology is hemispherectomy --removal of one of the two cerebral hemispheres. Only those with severely diseased brains have such operations, thus (as might be expected) while the surgery does not make them any worse, it leaves most in a poor state. A few, however, recover to lead surprisingly normal lives and show psychometrically normal or above normal IQs.
24. Aaron Smith and Oscar Sugar of the University of Illinois School of Medicine (1975) report the case of a male who started to become paralysed on the right side at five months, having seizures at three years and at five and half years of age had his left hemisphere removed. In spite of this, he went to a normal school, learned to play the baritone horn and even became a member of his high school band. At 26, he was working as an 'industrial executive (traffic controller)' having completed 'his senior year at a prominent Midwestern university with a dual major in sociology and business adminstration'. His full IQ at that time was 116 (WAIS verbal IQ, 126; performance IQ, 102).
25. Huw Griffith and May Davidson (1966) of the Radcliff Infirmary, Oxford, report the case of P.G. who at the age of 10 started having fits followed a year later by paralysis on the left side of his body. At the age of 19 he had his right cerebral hemisphere removed. Postoperatively, his IQ by the revised Stanford Binet was 142 points; by the Wechsler test: verbal, 118, and performance, 94. 15 years later, at 34, he was retested and the Wechsler verbal had increased to 121 and the Wechsler performance had remained relatively unchanged at 91. Since his operation according to Griffith and Davidson, he 'having obtained a university diploma after [the] operation has [now] a responsible administrative position with a local authority'.
26. Eileen Vining and colleagues at the John Hopkins University School of Medicine (1993) report the outcomes of 12 children at an average of 9 years follow up. They give extended details upon five of them. One case '13', a female, is of interest. She started having seizures at five, by the age seven she was having up to 20 a day. At seven and half, she had a left hemispherectomy remaining in coma for six weeks. Two years later she had a full scale IQ of 98. Three years afterwards, though she needed some help in mathematics, she was in seventh grade gaining grades of A and B.
27. The brain mass following hemispherectomy in these three individuals can only be estimated. The human cerebral cortex makes up 80% of the total brain (using the percentage found by post mortem rather than MRI), as a result hemispherectomy will reduce it by around 40%. Assuming their initial brains were average, they would have been 1,371 cc, of which 1097 cc would be cerebral cortex and associated white matter. Thus they would have lost around 548 cc of cortical tissue leaving them with a brain of around 823 cc. Similar calculations for the average volume of a female brain would suggest a brain after hemispherectomy of 730 cc.
28. As a group, the above cases provide evidence that some brains can possess normal cognition by psychometric tests while having brains within the range or smaller than the average Homo erectus one.
IV.1 SMALL BRAIN SIZE NORMALLY LINKS TO RETARDATION.
29. The above reported individuals are exceptions: the presence of a small brain in the overwhelming number of individuals (whether due to early brain removal or microcephaly) links to retardation. However, this would be expected. Even if microcephalic brains were capable of normal levels of intelligence, the pathological factors responsible for their abnormal brains not only reduce their size but will usually have affects (independent of this) upon neural integrity. In the case of hemispherectomy, the prior pathology in the original brain might have already effected the remaining normal brain. Thus, it will be relatively uncommon that brain mass is reduced without a parallel but separate impairment of cognition.
IV.2 BRAIN INJURIES DESTROY COGNITION.
30. Following accidents and strokes people show dramatic loss of cognitive abilities. This, it might be argued, suggests that the brain cannot work with reduced brain mass. However, the damage in these cases links to its suddenness. The brain can easily recover from injuries made in small stages, but not the same one made all at once (Feeney & Baron, 1986). This is in part because sudden brain injuries are followed by the release of excitotoxins which cause further neuron loss. In contrast, brain loss which happens slowly can pass without notice. For instance, asymptomatic meningioma can grow (at an average of 2.4 millimetres a year) into the size of plums -- five centimetres diameter across -- without noticeable effects (Olivero, Lister & Elwood, 1995). Interestingly, even when brain damage is sudden as with post-traumatic atrophy or focal damage, it can happen with minimal change in measured IQ (Bigler, 1995).
IV.3 IQ CORRELATES WITH BRAIN SIZE.
30.1. While IQ correlates with brain-size, the correlation is modest with the best recent estimate being around 0.40 (Wickett, Vernon, & Lee, 1994). The percentage contribution of a 0.40 correlation is 16% -- thus leaving 84% of intelligence to be explained by things other than brain size. Even higher reports of up to 0.51 correlation (Andreason, Flaum, Swayze et al., 1993; Willerman, Schultz, Rutledge & Bigler, 1991) would still leave 74% unexplained. Moreover, such IQ brain size correlations derive from studies upon educated people with a restricted range of IQ and brain size -- methodically problems exist questioning their generality (Peters, 1993). It is not known whether likewise the 0.40 correlation found between IQ and brain size holds in extreme cases outside the range tested; indeed, the evidence reviewed above suggests it does not. Further, size would not be expected to more than modestly increase cognitive competence as larger brains (due to their size) have larger neurons and more myelinated axons connecting them (Deacon, 1990). Perhaps because this, people with larger than normal brains are not necessarily brighter (DeMyer, 1986); indeed, there appears to be a plateau and perhaps a decease in IQ in the largest brains (Reiss, Abrams, Singer, Ross, & Denckla, 1996).
31. The 0.40 correlation, moreover, does not by itself rule out the existence of outliers. A statistical correlation is, after all, only a mathematical means of describing a data set. Thus, the 0.40 correlation between IQ and brain size is in itself not evidence against the existence of the cases reviewed above.
V EXPERTISE, IQ AND EVOLUTIONARY COGNITION
V.1 PSYCHOMETRIC IQ
32. Perhaps, surprisingly, given its status, psychometric measured intelligence leaves out by design some aspects of cognition known to link to intelligent behaviour. First, Binet and Simon, the founders of IQ testing at its start, wished to separate 'natural intelligence from instruction' by 'disregarding, in so far as possible, the degree of instruction which the subject possesses (Binet & Simon, 1908/1961, p. 93). As a result, from its beginnings IQ has excluded that important aspect of intelligence that concerns a person's capacity over a prolonged period to learn a skill in depth (which would require measuring skills that might have depended upon 'instruction'). Second, IQ tests for pragmatic reasons are restricted to measuring skills that can be tested with easily administered and standardised tasks. The result of this is that IQ tests are biased to measure those skills which can measured in terms of performance against time (it is easy to time performance) while ignoring a person's capacity to expertly know a specialised domain in depth (it is difficult to create standardise measurements for them). Reflecting these factors, while IQ tests show high correlations with tasks that measure reaction speeds such as the time to judge whether two lines are different lengths (Deary & Stough, 1996; Kranzler & Jensen, 1989), they show no or very moderate correlations with people's ability to acquire expertise (Ackerman, 1996; Ceci & Liker, 1986; Doll & Mayr, 1987; Ericsson & Lehmann, 1996; Shuter- Dyson & Gabriel, 1981).
V.2. EXPERTISE CAPACITY
33. Experts bring to a problem a store of knowledge acquired over a long period that shapes how they analyse and approach its solution. Generally, novices view a problem situation in terms of its salient features (they lack knowledge of how to better analyse it), while experts (with their extensive knowledge) retrieve part of the solution as they comprehend it (Ericsson & Lehmann, 1996). Thus while, for instance, non-chess players see merely chess pieces, chess masters to see possible future moves and potential strategies. Such in depth perception arises from people acquiring and being able to actively use a large numbers of information 'chunks' when analyzing a problem. Estimates put the amount of such knowledge, for instance, in chess masters at 50,000 chucks of information (Gobet & Simon, 1996). Such information processing chunks take many years to acquire. After reviewing performance in sport, medicine, chess and music, Ericsson and Lehmann (1996) propose that before people can show expertise in any domain they must have spent at least several hours of daily practice for a minimum of 10-years.
V.3 EXPERTISE AND BEHAVIOURAL INTELLIGENCE
34. In the modern world, in depth knowledge is not required for nonretarded and successful existence nor for being measured with a normal IQ. In general, skills dependent upon in depth knowledge are limited to only a few occupations or task domains which require a person to process a large variety of possibilities, variables or options (Ericsson & Lehmann, 1996). One example which has been studied in the context of IQ (and thus enables its linkage to IQ to be examined) is the ability of those with many years of daily experience in studying racing statistics to predict the odds of horses in races (Ceci & Liker, 1986). This requires making complex inferences and trade-offs between statistics concerning horses, jockeys, the physical state of different race tracks. Reasoning among experts can be as complex as considering 'the need of horse A to avoid challenging horse B at the first quarter mile mark, if horse C is to have any change of beating horse D'. Stephan Ceci and Jeffrey Liker (1986) measured the IQs using the Wechsler Adult Intelligence Scale of experts in predicting form. Of 12 experts, their measured IQs ranged between 81 and 128 (four were between 80 and 90, three between 90 and 100, two between 100 and 110 and only three above 120) (Ceci & Liker, 1986: table 6). Ceci and Liker (1986, p. 265) comment 'whatever it is that an IQ test measures, it is not the ability to engage in cognitively complex forms of multivariate reasoning'. The expert performance in other domains such as chess (Doll & Mayr, 1987) and music (Shuter-Dyson & Gabriel, 1981) likewise show expertise that correlates poorly, or not at all, with IQ.
V.4. EXPERTISE AND EARLY HOMINID FITNESS
35. While the capacity to develop expertise is not necessary for normal life in a modern industrial society such skills would have been critical for the survival of early hunter-gatherers. Richard Lee (1979) notes, the !Kung when hunting use in depth expert knowledge and reasoning. For instance, hunters can spot of a particular track, the animal which made it, its sex, age, whether alone or ill, what it was eating, and to within 15 minutes how long ago since the animal made it. They can do this by reading the shape, depth and condition of tracks, whether they are alone, how they are spaced and located. For instance, tracks found to the east side of a tree might suggest it rested there in the morning shade (Lee 1979, pp. 212-213). Such knowledge like other forms of expertise takes many years, indeed decades, to learn. Moreover, it is known to be more important than physical skill to hunting success: the individuals most successful at hunting are those in the over 39 age group (with decades of experience tracking), not the more physical able (but less experienced) individuals under this age (Lee 1979, pp. 242- 244). Indeed, an old man in his fifties or sixties might go with a young man (usually his son) interpreting expertly the tracks while the young man does the hunting.
36. This example suggests that in populations of early humans the individuals possessing the brains with the greatest capacity to acquire expertise would out survive those without their extensive knowledge. The reasons for this would not only include expertise in hunting but expertise in many other activities such as gathering, tool manufacture and social communication. Over time the advantages given by success in these activities would result in the natural selection of brains with increased capacity for expertise.
37. One factor that increases a person's expertise capacity is likely to have been brain size. Expertise, as noted, links to the number of information chunks a person can hold and actively process. Therefore, a person's capacity for expertise will link to the number of their cortical columns able to specialise neural networks in representing and processing them -- and through this their cerebral mass. Harry Jerison (1991) has noted that brain size through increasing the numbers of columns increases the information processing power of the brain. Further, there is some recent evidence that the increased information processing requirements of expertise leads to skill expansion over large areas of the cortex. For instance, expertise in violin playing depends upon fine coordination of the left hand fingers and accurate coordination between their two hands. If expertise links to increased cortical area devoted to finger coordination, we would expect expert violinists to devote more of their brain to finger coordination. This is indeed the case. Expert violinists have two to three times the area of cortex devoted to their left fingers than nonviolinists (Elbert, Pantev, Wienbruch, Rochstroh & Taub, 1995). Further, the need of expert violinists to coordinate their two hands leads them to develop a larger link between two sides of their brain dealing with motor coordination than nonviolinists (Schlaug, JĄncke, Huang, Staiger & Steinmetz, 1995). Thus, there is not only theoretically but some empirically evidence that expertise needs large amounts of brain to neurally store and actively process its information chunks. This suggests that a strong link should exist between for the capacity for acquiring expertise skills and brain mass.
38. Thus, in conclusion, the paradox of individuals with Homo erectus sized brains but normal IQ need not be viewed as necessarily being anomalous. Intelligence may link to IQ tests, but the intelligence which was central to the evolution of our brains could link to something IQ tests do not measure -- expertise capacity. Though the example given above concerns hunting, no reason exists why expertise in this should have been the sole advantage of larger brains. Expertise in gathering, scavenging, social relationships, language, tool making and passing on acquired skills and knowledge could equally have been critical.
VI FUTURE RESEARCH
39. If expertise is the missing link between brain size and human intelligence then opportunities exist for cognitive scientists and clinicians to contribute directly to understanding human evolution. The above evidence of small brains and normal IQ needs to be researched in much more detail. Do individuals with normal IQs but small brains show limited expertise as might be predicted? As noted, expertise would have been important not only in hunting (the only one discussed here due to space and evidence) but also such activities as social skills and the motor control needed in tool making. The role of expertise in these areas still needs detailed study by social psychologists and motor control scientists.
40. Recently, volumetric studies of the brain have found considerable variation in the size of brain areas. Raz and colleagues (1997), for instance, have found many-fold variation -- three-fold for the inferior parietal cortex and nearly four-fold for the primary visual cortex (1997: diagram 5B). Does a person's capacity to develop an expertise depending upon the abilities of circuits in a particular area vary with its size? Intuitively it should. But at present we do not know whether individuals with unusually large or small areas have different capacities to develop expertise using them.
41. Psychologically orientated anthropologists, further, need to provide us with field data about the role of expertise in simple hunter-gather bands, especially in regard to nonhunting activities such as gathering. One of the reviewers of this papers, for instance, questioned whether tracking as done by the !Kung actually required expertise. There is at present no anthropology of the expertise of such activities with which to answer him. In conclusion, palaeoanthropology and many areas of psychology by engaging in an interdisciplinary 'colloquy' over expertise could do much to further the science of our origins.
 The formula derived by Aiello and Dunbar (1993) for linking hominid skull volume to brain volume is Log10(B) = 3.015 + 0.986 Log10(C), where B is the total brain size in mm3, and C is the internal cranial capacity measured in cc.
Ackerman, P. L. (1996). A theory of adult intellectual development: Process, personality, interests, and knowledge. Intelligence, 22, 227-257.
Aiello, L. & Dunbar, R. (1993). Neocortex size, group size and the evolution of language. Current Anthropology, 34, 184-192.
Andreason, N. C., Flaum, M., Swayze, V., O'Leary, D. S., Alliger, R., Cohen, G., Ehrhardt, J. & Yuh, W. T. (1993). Intelligence and brain structure in normal individuals. American Journal of Psychiatry, 150, 130-134.
Binet, A. & Simon, T. (1908/1961). The development of intelligence in children. In J. J. Jenkins & D. G. Paterson (Reprint Eds.), Studies in individual differences: the search for intelligence (pp. 96-111). New York: Appleton-Century-Crofts.
Bigler, E. D. (1995). Brain morphology and intelligence. Developmental Neuropsychology, 11, 377-403.
Ceci, S. J., & Liker, J. K. (1986). A day at the races: a study of IQ, expertise, and cognitive complexity. Journal of Experimental Psychology: General, 115, 255-266.
Day, M. H. (1992). "Posture and childbirth," In The Cambridge encyclopedia of human evolution. Edited by S. Jones, R. Martin and D. Pilbeam, p. 88. Cambridge: University of Cambridge Press.
Deacon, T. (1990). Fallacies of progression in theories of brain-size evolution. International Journal of Primatology, 11, 193-236.
Deary, I. J. & Stough, C. (1996). Intelligence and inspection time: achievements, prospects, and problems. American Psychologist, 51, 599-608.
Dekaban, A. S. & Sadowsky, D. (1978). Changes in brain weights during the span of human life: Relations of brain weights to body heights and body weights. Annals of Neurology, 4, 345-356.
DeMyer, W. (1986). Megalencephaly: Types, clinical syndromes, and management. Pediatric Neurology, 2, 321-328.
Doll, J., & Mayr, U. (1987). Intelligenz und Schacleistung [Intelligence and achievement in chess]. Psychologische BeitrĄge, 29, 270- 289.
Dorman, C. (1991). Microcephaly and intelligence. Developmental Medicine and Child Neurology, 33, 267-272.
Elbert, TH., Pantev, C., Wienbruch, C. Rochstroh, B. & Taub, E. (1995). Increased cortical representation of the fingers of the left hand in string players. Science, 270, 305-307.
Ericsson, K. A., & Lehmann, A. C. (1996). Expert and exceptional performance: Evidence of maximal adaption to task constraints. Annual Review of Psychology, 47, 273-305.
Feeney, D. & Baron, J-C. (1986). Diaschisis. Stroke, 17, 817-830.
Filipek, P. A., Richelme, C., Kennedy, D. N. & Caviness, V. S. (1994). The young adult human brain: An MRI-based morphometric analysis. Cerebral Cortex, 4, 344-360.
Giedd, J.N., Snell, J. W., Lange, N., Rajapakse, J. C., Casey, B. J., Kozuch, P. L., Vaituzis, A. C., Vauss, Y. C., Hamburger, S. D., Kaysen, D. & Rapoport, J. L. (1996). Quantitative magnetic resonance imaging of human brain development: Ages 4-18. Cerebral Cortex, 6, 551-560.
Gobet, F., & Simon, H. A. (1996). Recall of random and distorted chess position: Implication for the theory of expertise. Memory and Cognition, 24, 493-503.
Gould, S. J. (1981). Mismeasure of man. New York: Norton.
Griffith, H. & Davidson, M. (1966). Long-term changes in intellect and behaviour after hemispherectomy. Journal of Neurology, Neurosurgery and Psychiatry, 29, 571-576.
Holloway, R. L. (1968). The evolution of the primate brain: some aspects of quantitative relations. Brain Research, 7, 121-172.
Jerison, H. (1991). Brain size and the evolution of mind. Fifty-ninth James Arthur lecture on the evolution of the human brain. New York: American Museum of Natural History.
Kranzler, J. H., & Jensen, A. R. (1989). Inspection time and intelligence: A meta-analysis. Intelligence, 13, 329-347.
Lee, R. B. (1979). The !Kung San: Men, women, and work in a foraging society. Cambridge: Cambridge University Press.
Lenneberg, E. H. (1967). Biological foundations of language. New York: Wiley.
Martin, R. D. (1982). Human brain evolution in an ecological context: Fifty-second James Arthur lecture on the evolution of the human brain. New York: American Museum of Natural History.
Olivero, W. C., Lister, J. R. & Elwood, P. E. (1995). The natural history and growth rate of asymptomatic meningiomas. Journal of Neurosurgery, 83, 222-224.
Peters, M. (1993). Still no convincing evidence of a relation between brain size and intelligence in humans. Canadian Journal of Experimental Psychology, 47, 751-756.
Reiss, A. L., Abrams, M. T., Singer, H. S., Ross, J. L. & Denckla, M. B. (1996). Brain development, gender and IQ in children: A volumetric imaging study. Brain, 119, 1763-1774.
Raz, N., Gunning, F. M., Head, D., Dupuis, J. H., McQuin, J., Briggs, S. D., Loken, W. J., Thornton, A. E. & Acker, J. D. (1997). Selective aging of the human cerebral cortex observed in Vivo: Differential vulnerability of the prefrontal grey matter. Cerebral Cortex, 7, 282- 282.
Roche, A. F., Mukherjee, D., Guo, S. & Moore, W. M. (1987). Head circumference reference data: Birth to 18 years. Pediatrics, 79, 706- 712.
Rosenberg, K. R. (1992). The evolution of modern human childbirth. Yearbook of Physical Anthropology, 35, 89-124.
Rossi, L. N., Candini, G., Scarlatti, G., Rossi, G., Prina, E. & Alberti, S. (1987). Autosomal dominant microcephaly without mental retardation. American Journal of Diseases of Children, 141, 655-659.
Sassaman, E. A. & Zartler, A. S. (1982). Mental retardation and head growth abnormalities. Journal of Pediatric Psychology, 7, 149-156.
Schlaug, G., JĄncke, L., Huang, Y., Staiger, J. F. & Steinmetz, H. (1995). Increased corpus callosum size in musicians. Neuropsychologia, 33, 1047-1055.
Sells, C. J. (1977). Microcephaly in a normal school population. Pediatrics, 59, 262-265.
Shuter-Dyson, R., & Gabriel, C. (1981). The psychology of musical ability. London: Methuen.
Smith, A. & Sugar, O. (1975). Development of above normal language and intelligence 21 years after left hemispherectomy. Neurology, 25, 813- 818.
Stanyon, R., Consigliere, S. & Morescalchi, M. A. (1993) Cranial capacity in hominid evolution. Human Evolution, 8, 205-216.
Vining, E. P., Freeman, J. M., Brandt, J., Carson, B. S. & Uematsu, S. (1993). Progressive unilateral encephalopathy of childhood (Rasmussen's syndrome): A reappraisal. Epilepsia, 34, 639-650.
Wickett, J. C., Vernon, P. A. & Lee, D. H. (1994). In vivo brain size, head perimeter, and intelligence in a sample of healthy adult females. Personality and Individual differences, 16, 813-838.
Wilder, B. G. (1911). Exhibition of, and preliminary note upon, a brain of about one-half the average size from a white man of ordinary weight and intelligence. Journal of Nervous and Mental Diseases, 30, 95-97.
Willerman, L., Schultz, R., Rutledge, J. N. & Bigler, E. D. (1991). In vivo brain size and intelligence. Intelligence, 15, 223-228.