Provided for non-commercial research and educational use only.

Not for reproduction, distribution or commercial use.

This chapter was originally published in the book Progress in Brain Research,
Vol. 195, published by Elsevier, and the attached copy is provided by Elsevier for the
author's benefit and for the benefit of the author's institution, for non-commercial
research and educational use including without limitation use in instruction at your
institution, sending it to specific colleagues who know you, and providing a copy to
your institution’s administrator.

All other uses, reproduction and distribution, including without limitation commercial
reprints, selling or licensing copies or access, or posting on open internet sites, your
personal or institution’s website or repository, are prohibited. For exceptions,
permission may be sought for such use through Elsevier's permissions site at:
http://www.elsevier.com/locate/permissionusematerial

From: P. Thomas Schoenemann, Evolution of brain and language. In Michel A.

Hofman, Dean Falk, editors: Progress in Brain Research, Vol. 195, Amsterdam:
The Netherlands, 2012, pp. 443-459.
ISBN: 978-0-444-53860-4
© Copyright 2012 Elsevier B.V.
Elsevier
 Author's personal copy

M. A. Hofman and D. Falk (Eds.)

Progress in Brain Research, Vol. 195
ISSN: 0079-6123
Copyright Ó 2012 Elsevier B.V. All rights reserved.

CHAPTER 22

Evolution of brain and language

P. Thomas Schoenemann*

Department of Anthropology, Indiana University, Bloomington, IN, USA

Abstract: In this chapter evolutionary changes in the human brain that are relevant to language are
reviewed. Most of what is known involves assessments of the relative sizes of brain regions. Overall
brain size is associated with some key behavioral features relevant to language, including complexity
of the social environment and the degree of conceptual complexity. Prefrontal cortical and temporal
lobe areas relevant to language appear to have increased disproportionately. Areas relevant to
language production and perception have changed less dramatically. The extent to which these
changes were a consequence specifically of language versus other behavioral adaptations is a good
question, but the process may best be viewed as a complex adaptive system, whereby cultural learning
interacts with biology iteratively over time to produce language. Overall, language appears to have
adapted to the human brain more so than the reverse.

Keywords: coevolution; conceptual complexity; communication; cortex; Broca’s area; Wernicke’s area;
comparative primate; brain scaling.

Introduction Comparative studies of the brains of humans

and other animals, combined with an under-
Among all the behavioral changes made possible standing of the different functions of specific
by human brain evolution, language is arguably brain regions, and considered within a realistic
the most critical to defining the human condition. evolutionary perspective, allow a reasonable
Other animals communicate, but none do so with sketch of the evolution of brain and language.
the richness and complexity of human language. Languages must be learnable by the brains
This means that there must be some important of children in each generation. Thus, langu-
differences between the brains of humans and age change (a form of cultural evolution) is
other animals in areas relevant to communication. constrained by the existing abilities of brains in
each generation. Because language is critical to an
*Corresponding author. individual’s adaptive fitness, language also likely
Tel.: þ1-812-8558800; Fax: þ1-812-8554358 had a fundamental influence on brain evolution.
E-mail: [email protected] Humans are particularly socially interactive

DOI: 10.1016/B978-0-444-53860-4.00022-2 443

 Author's personal copy

444

creatures, which makes communication central The evolution of brain circuits cannot be under-
to our existence. Two interrelated evolutio- stood independent of the evolution of language,
nary processes therefore occurred simultaneously: and vice versa.
language adapted to the human brain (cultural Because the evolutionary benefits of language
evolution), while the human brain adapted to to an individual would always have been depen-
better subserve language (biological evolution). dent on the preexisting cognitive abilities of others,
This coevolutionary process resulted in langu- language evolution is inherently constrained.
age and brain evolving to suit each other New genetic variants enhancing the perception of
(Christiansen, 1994; Christiansen and Chater, linguistically relevant signals would have been
2008; Deacon, 1992). favored only to the extent that they increase the
The coevolution of language and brain can be individual’s ability to perceive and rapidly process
understood as a complex adaptive system the acoustic signals already used by others for
(Beckner et al., 2009). Complex adaptive systems language. Similarly, changes affecting the produc-
are characterized by interacting sets of agents tion of linguistically relevant signals would be
(which can be individuals, neurons, etc.), where favored only to the extent that they could be under-
each agent behaves in an individually adaptive stood by the preexisting perceptual abilities
way to local conditions, often following very of others. Signals too complicated or subtle for
simple rules. The sum total of these interactions others to process would not be adopted, and hence
nevertheless leads to various kinds of emergent, mutations influencing them would not likely
system-wide order. With respect to the coevolution spread.
of brain and language, a number of language- It is possible for some new variant to be adaptive
relevant neural systems interact with and influence strictly at the individual level (and therefore
each other in important ways. Syntax depends spread) even if it was not immediately useful for
fundamentally on the structure of semantics, since language, but this could only be true if they were
the function of syntax is to code higher-level beneficial for some other reason. In this case, how-
semantic information. Semantics in turn depends ever, it would not be a “language” variant. If it
on the structure of conceptual understanding, spread wide enough (for nonlinguistic reasons), it
which—as will be reviewed below—is a function might later be co-opted for language. This would
of brain structure. These structures are in turn the not result in the evolution of highly language-
result of biological adaptation: circuits that result specific circuits.
in conceptual understanding that is useful to a For these reasons, any adaptive changes in lan-
given individual’s environmental realities will be guage circuitry occurring during a given sequence
selected for. of the evolutionary process will be biased toward
In general, the selective environment for pri- slight modifications of preexisting circuits, and
mate species is largely a social one. The adap- away from major changes in the ways communi-
tiveness (reproductive benefit) of an individual’s cation is processed by the brain (Schoenemann,
particular behavior at any given moment in time 2005). Language circuits should show extensive
depends crucially on the flexible responses of homologies with preexisting systems in closely
others in the group, who are also attempting to related animals. These hijacked circuits would,
behave in an adaptive manner in response. Under- by definition, be domain general (contra Tooby
standing language evolution ultimately involves and Cosmides, 1992).
understanding how the repeated complex commu- Inferences about evolutionary changes in the
nicative interactions of individuals result in cultural brain relevant to language are derived from
change in languages, and how these changes in knowledge of how language is processed in the
turn influence biological change in the long term. brain, combined with knowledge of how our
 Author's personal copy

445

brains are different from those of our closest evo- size (Blinkov and Glezer, 1968), and premotor cor-
lutionary relatives. To the extent that a particular tex (Brodmann area 6) and primary visual
area known to be relevant to language appears to cortex (Brodmann area 17) appear to have lagged
have also changed significantly, we are justified significantly behind the increase in overall brain
in inferring that this area was important for lan- size (Deacon, 1997; Schoenemann, 2006). Thus,
guage evolution, though whether it increased brain evolution is remarkably plastic over the long
specifically for language will be difficult to deter- term (contra Finlay et al., 2001). It is true that brain
mine. Evolutionary inferences will also involve size correlates with body size across major groups
thinking about the interplay of different behav- of animals, which in turn has led to the use of rela-
ioral abilities over our history. Both an evolution- tive brain size measures when comparing species
ary perspective and a complex adaptive systems (e.g., the encephalization quotient or EQ; Jerison,
approach predict that language evolution 1973; see also Chapter 20). However, this associa-
occurred hand in hand with the evolution of other tion does not require that relative brain size is
aspects of cognition. Language processing behaviorally relevant. It likely just represents an
depends heavily on the integration of a large inevitable tradeoff between the utility of brains
number of abilities that are processed in widely and their disproportionate metabolic costs: larger
dispersed circuits across the brain (Damasio and brains may always be useful, but only larger
Damasio, 1992; Mueller, 1996). Assessing the animals can afford to pay for them metabolically
coevolution of language and brain therefore (e.g., Martin, 1981). In fact, absolute brain size is
requires a broad focus on a number of brain empirically a much better predictor of species
regions. differences in behavior than is relative brain size
(Deaner et al., 2007; Gibson et al., 2001; for a dis-
cussion see Chapter 15). For this reason, any
Evolutionary changes in the brain changes in brain anatomy are therefore potentially
relevant to language of behavioral importance.
One caveat that needs to be stated; however,
Knowing how different language-relevant brain we do not know the details of exactly how differ-
areas have changed over our evolutionary history ent areas of the brain actually process informa-
is central to understanding the coevolution of brain tion. We lack the neural equivalent of a circuit
and language. However, it is not clear what counts diagram for anything larger than the 302 neuron
as significant change: increases relative to brain brain of the worm Caenorhabditis elegans (White
size, body size, or simply absolute size independent et al., 1986; Chapter 17). Because of this, we can-
of either brain or body? Because of the evolu- not be sure that a fourfold increase in one area
tionary costs to increasing the absolute numbers has greater behavioral implications than a twofold
of neurons (Hofman, 1983), changes in absolute increase in some other area. Nevertheless, any
size of an area independent of body or brain increase would seem to be important, given the
increases are likely to be behaviorally relevant. costs of maintaining excess neurons.
Further, there are many examples of changes that
appear to be unrelated to either body or brain size
increases. The olfactory bulb (responsible for sense Overall brain size
of smell) has actually decreased in size (being half
that found in chimpanzees; Stephan et al., 1981), Human brains are about three times larger than
the primary motor cortex (Brodmann area 4) that found in our closest relatives, the African apes
has apparently not increased at all in absolute (even taking body size into account; Jerison, 1973).
 Author's personal copy

446

Exactly what behavioral implications this has is shown to be associated with various measures of
a matter of debate, but given the importance of brain size (including absolute brain size) in
language to the human condition, it is reasonable primates (Fig. 1; Dunbar, 1995; Reader and
to believe that at least part of this increase is Laland, 2002). Social group size is a reasonable
due to language. The idea that the evolution of proxy for the complexity of one’s social existence.
brain size and language are related is an old Human social networks appear to be particularly
one going back at least to Darwin (1882), who complex, and given that language is an inherently
argued there was likely a “. . .relation between the social activity, the selective value of language is
continued use of language and the development likely greatest for humans.
of the brain. . .” (p. 87). Because different parts of The size of the neocortex, which plays a key role
the brain have changed in different ways, focusing in conscious awareness generally as well as
solely on overall brain size is an oversimplification. mediating a number of complex cognitive functions
However, there are some interesting correlates including language, appears to be strongly
of overall brain size that are likely relevant to associated with overall brain size (Hofman, 1985).
language evolution. The neocortex makes up over 80% of the entire
First, as pointed out above, absolute brain size human brain, which is the highest value among all
is empirically behaviorally relevant. Overall brain primates. The corresponding values for apes (who
size also correlates strongly with both length of have the next largest brains among primates) range
maturation (Harvey and Clutton-Brock, 1985) from 76% to 73%, while particularly small
and overall lifespan (Allman et al., 1993; Hofman, brained monkeys range down to 59%, and the
1993). This means that the bigger the brain, the smallest brained primate of all, the mouse lemur
greater the potential for behavioral learning to (Microcebus murinus), has a neocortex that takes
be a central part of the organism’s behavioral rep- up only 44% of its brain (Hofman, 1985, 1988;
ertoire. Larger brained animals rely on learning
more than do smaller brained animals (Deacon,
1997), and larger brained primates do better at a 2
variety of experimental learning tasks (Deaner
et al., 2007).
A great deal of modern human behavior (includ-
Log (mean group size)

ing language) depends critically on learning. While

learning can be biased in particular ways by
evolved innate influences, human behavioral evo- 1
lution is better characterized by increasing behav-
ioral flexibility rather than greater numbers of
hardwired, innate circuits. Learning language obvi-
ously depends on being able to understand chang-
ing, fluid contingencies between constituents and
meaning. The increasing behavioral flexibility and 0
reliance on learning made possible by the increase
in brain size, therefore, made language increas-
0 1 2 3
ingly possible, if not inevitable.
Log (brain volume in cc)
Primates as a group are particularly interactively
social, and interactive sociality is a particularly Fig. 1. The relationship between brain volume and mean
complicated niche (Holloway, 1975; Humphrey, group size in primate species. N¼36, r¼0.75, p<0.0001. Data
1984). The size of the typical social group has been from Dunbar (1995).
 Author's personal copy

447

Stephan et al., 1981). This would suggest that as (texture)]), most concepts require the integration
brains get larger, conscious behavior becomes of more than one sense. For example, the concept
increasingly important. “coffee” typically invokes not just a particular
In addition, as the neocortex increases, areas taste but also a smell, a visual image of a mug, the
of it that are devoted specifically to integrating sensation of warmth, and so forth (Damasio and
different types of information (so-called associa- Damasio, 1992). For these sensory impressions
tion areas) increase disproportionately, at the to be bound in some way into the concept “coffee,”
expense of areas devoted either to the processing the different areas that process these impressions
of sensory information from a single modality must be connected. A complete list of areas
(such as the primary visual or primary auditory that are relevant to just the basic features of con-
cortices) or to the conscious control of muscle ceptual awareness would be very long, involving
movement (Fig. 2). The larger these “association all the visual (color, shape, motion, etc.), spatial,
areas” are, the greater the likely potential for auditory, temporal organization, olfactory, taste,
increasingly complex types of integrative pro- somatosensory, and limbic system (emotion)
cessing (Schoenemann, 2010). areas. These are processed using extensive regions
Further, as these association areas expand, of the parietal, occipital, and temporal lobes
they appear to evolve increasing numbers of rel- (Fig. 3b).
atively specialized processing areas. Larger Given that conceptual awareness forms the
brains have greater numbers of identifiably dis- very foundation of language (Hurford, 2003a),
tinct cortical areas (Changizi and Shimojo, and given that larger brains appear to give rise
2005; Northcutt and Kaas, 1995). This turns out to more complex conceptual universes (and hence
to be a predictable consequence of increasing more interesting things to communicate about),
brain size: any given area of the neocortex will and given that humans are intensely socially inter-
tend to be less directly connected to other areas active, increasing brain size itself should be seen
in larger brains compared to smaller brains as an excellent proxy for language evolution
(Ringo, 1991). This means that areas are able (Gibson, 2002; Schoenemann, 1999, 2005).
to carry out tasks increasingly independent of
each other, leading predictably to increasing
functional localization. Classical language areas
An fundamentally important consequence of
this that it allows for the formation of richer, Broca’s and Wernicke’s areas were the first cortical
more complex, and more subtle conceptual under- regions to be associated with specific linguistic
standing (Gibson, 2002; Schoenemann, 1999, abilities. Broca’s aphasics display nonfluent, effort-
2005). Much of the brain appears to be relevant ful, and agrammatical speech, whereas Wernicke’s
to concept formation (Barsalou, 2008; Damasio aphasics display grammatical but meaningless
and Damasio, 1992; Schoenemann, 2005). When a speech in which the wrong words (or parts of
subject imagines an object that is not actually pres- words) are used (Bear et al., 2007; Damasio and
ent, similar areas of their brain are activated Damasio, 1992). Broca’s area is located in the
as when the object is being viewed (Damasio posterior–inferior frontal convexity of the neocor-
et al., 1993; Kosslyn et al., 1993). Different kinds tex, while Wernicke’s area is localized to the gen-
of basic sensory input—visual, auditory, olfactory, eral area where parietal, occipital, and temporal
taste, and somatosensory (touch, temperature, lobes meet (Fig. 3d). For most people, these areas
pain, body position)—are processed in different are functional for language primarily in the left
areas. While some basic concepts involve only a hemisphere. However, it turns out that Broca’s
single sensory modality (e.g., [green] or [smooth and Wernicke’s aphasias (the specific types of
 Author's personal copy

448

(a) Primary motor

Somatosensory
Premotor

Primary
visual

Cerebellum
Primary auditory

Brainstem 1 cm

(b) Somatosensory
Primary motor

Premotor Primary
visual

Cerebellum

Insula

1 cm
Primary auditory Brainstem

(c) Somatosensory Primary

visual Primary auditory
Motor
Cerebellum

Insula Brainstem 1 cm

Fig. 2. Size of “association” cortex in mammals of different brain size. Lateral line drawings of the cortex of human (a), galago
(prosimian primate) (b), and hedgehog (non-primate mammal) (c). Images are not to scale (note 1cm scale bars for each image;
 Author's personal copy

449

language deficits) are not exclusively associated aphasia, including aspects of syntactic processing
with damage to Broca’s and Wernicke’s cortical (Lieberman, 2000). It is clear that a simple model
areas (Dronkers, 2000). Damage to the caudate of language being processed solely in Broca’s and
nucleus, putamen, and internal capsule (structures Wernicke’s areas is too simplistic (Poeppel and
of the cerebral hemispheres that are deep to the Hickok, 2004), though these areas are clearly
cortex) also appear to play a role in Broca’s relevant.

(a) (b)

Parietal

Frontal

Occipital
Temporal

Cerebellum

Brainstem

(c) (d)

Arcuate fasciculus

Prefrontal
Wernicke’s area
Broca’s area

Fig. 3. Major regions of the brain. (a) Lateral view of the brain. (b) Lobes of the cortex, with cerebellum and brainstem indicated.
(c) Prefrontal cortex. (d) Classical language areas. The arcuate fasciculus is a connective pathway deep to the cortex. The brain
image is a 3D rendering of an average of 27 MRIs of the same individual from Holmes et al. (1998), used with permission.

corresponding volumes: hedgehog: 3ml, galago: 10ml, human: 1350ml). The white regions represent cortical areas that are
devoted to processing information other than primary sensory or motor (muscle movement) information. They function to
integrate information in various interesting ways. Hedgehog brains have essentially no association cortex, whereas human brains
have significantly more association cortex than other primates, both absolutely and proportionately. The human insula is not
visible on the surface, being buried deep to the Sylvian fissure (which separates the temporal lobe from the frontal and anterior
parietal lobes). The human brain was drawn from a 3D rendering of an average of 27 MRIs of the same individual (Holmes
et al., 1998, used with permission). Galago and hedgehog brains were drawn from images at http://www.brains.rad.msu.edu, and
http://brainmuseum.org, used with permission. Functional areas are mapped following Nieuwenhuys et al. (2008).
 Author's personal copy

450

The evolutionary histories of Broca’s and Brodmann area 45 was 6.0 times larger on the left
Wernicke’s areas are quite intriguing, since and 5.0 times larger on the right (Schenker et al.,
homologs to both of these areas have been 2009). For comparison, overall brain size was 3.6
identified in nonhuman primate brains (Striedter, times larger for this sample, thus suggesting that
2005). These species lack human language there have been disproportionate increases in
capabilities, of course, so these areas must have Broca’s area—particularly, in the left hemi-
evolved originally for other purposes. An evolu- sphere—during human evolution. Quantitative
tionary perspective would predict that in nonhu- comparisons of Wernicke’s area have not been
man primates these areas likely process reported, though it does appear that it is signifi-
information in ways that would be useful to lan- cantly bigger in both absolute and relative terms
guage (Schoenemann, 2005), with language then in humans as compared to macaque monkeys
naturally making use of them (i.e., language (Petrides and Pandya, 2002; Striedter, 2005).
adapting to the human brain: Christiansen and Given that Broca’s and Wernicke’s areas play
Chater, 2008). The presence of these “language” different but complementary roles in language
areas in nonlinguistic animals undermine models processing, they must be connected in some way.
that imply the evolution of completely new kinds A tract of nerve fibers known as the arcuate fas-
of circuits (e.g., Bickerton, 1990; Pinker, 1995). ciculus (Fig. 3d) directly connects these areas
Some intriguing findings suggest that these areas (Geschwind, 1974). It tends to be larger on the
do function in primates in ways that would predis- left side than the right in humans, consistent with
pose them to human language processing. The the lateralization of expressive language pro-
homolog of Broca’s area in monkeys has been cessing to the left hemisphere for most people
shown to contain neurons that fire both when a (Nucifora et al., 2005). In addition, it appears to
monkey performs a specific action as well as when have been elaborated in human evolution. The
it hears a sound related to that same action (“mir- homolog of Wernicke’s area in macaque monkeys
ror neurons”; Kohler et al., 2002), which may projects to prefrontal regions that are close to the
form the basis for the ability to attach meaning homolog of Broca’s area, but apparently not
to sounds. Stimulation of the Broca’s area homo- directly to it (Aboitiz and Garcia, 1997). Instead,
log in macaque monkeys results in orofacial projections directly to their homolog of Broca’s
movements (Petrides et al., 2005), which are area originate from a region just adjacent to their
foundational to human speech. Hearing species- homolog of Wernicke’s area (Aboitiz and Garcia,
specific calls has been shown to activate Broca’s 1997). This would suggest that there has been an
and Wernicke’s areas in monkeys (Gil-da-Costa elaboration and extension of projections to more
et al., 2006). In chimpanzees, communicative sig- closely connect Broca’s and Wernicke’s areas
naling (begging) has activated the homolog of over the course of human (or ape) evolution.
Broca’s area (Taglialatela et al., 2008). Further Recent work using diffusion tensor imaging
exploration of the function of these areas in non- (which delineates approximate white matter axo-
humans will allow a better idea of how and why nal tracts in vivo) suggests that both macaques
they became co-opted for human language. and chimpanzees have tracts connecting areas in
Detailed quantitative data on the size of these the vicinity of Wernicke’s area to regions in the
areas have been reported only for humans and vicinity of Broca’s area (Rilling et al., 2007).
chimpanzees so far. For the two areas that com- However, connections between Broca’s area and
prise Broca’s area, one study reported that the middle temporal regions (important to seman-
Brodmann area 44 in the left hemisphere is 6.6 tic processing—see below) are only clear in
times larger and in the right 4.1 times larger in chimpanzees and humans, and are even more
humans as compared to chimpanzees, while extensive in humans (Rilling et al., 2007). These
 Author's personal copy

451

changes are certainly relevant to language evolu- memory for serial order and temporal information,
tion, though knowing whether they evolved spe- and attention (references in Schoenemann, 2006).
cifically for language, rather than to support Teasing apart the relative contributions of these
more general cognitive operations involving con- various behavioral abilities to the evolution of pre-
ceptual understanding, is unknown. frontal areas will likely be very difficult.
The prefrontal cortex itself has many compo-
nents, not all of which have changed to the same
Prefrontal cortex extent. Broca’s area has increased disproportion-
ately, particularly on the left side, as discussed
Areas in the prefrontal cortex (in addition to above (Schenker et al., 2009). Brodmann area
Broca’s area; Fig. 3c) appear to be involved in a 13, located in the posterior orbitofrontal cortex
variety of linguistic tasks, including various (posteroinferior frontal cortex, above the eyes),
semantic aspects of language (Gabrieli et al., seems to have lagged behind the increase in over-
1998; Kerns et al., 2004; Luke et al., 2002; Maguire all brain size, being only 1.5 times larger than
and Frith, 2004; Noppeney and Price, 2004; the average ape (Semendeferi et al., 1998). This
Thompson-Schill et al., 1998), syntax (Indefrey area is involved in assessing emotional aspects of
et al., 2004; Novoa and Ardila, 1987), and higher social interactions. Given the increasing impor-
level linguistic processing, such as understanding tance of the social context during human evolu-
the reasoning underlying a conversation (Caplan tion, its increase might seem relatively small
and Dapretto, 2001). (though not trivial). However, its specific rele-
There appears to have been a significant elabora- vance to language is unclear, except insofar as it
tion of the prefrontal cortex during human evolu- presumably contributes to conceptual under-
tion, with cytoarchitectural data pointing to an standing of social relationships, which form the
approximately twofold increase for the entire pre- basis for caring about communication in the first
frontal cortex over what would be predicted for a place (see discussion below about conceptual
primate brain as large as ours (Brodmann, 1909; understanding and language evolution).
Deacon, 1997). MRI studies generally support Brodmann area 10, by contrast, is 6.6 times
these conclusions though some debate remains larger than the corresponding areas in pongids
(reviewed in Schoenemann, 2006; see also Smaers (Semendeferi et al., 2001; Chapter 9). This
et al., 2011). Using a proxy for prefrontal cortex, increase is actually close to what one would
we found that connective tracts (white matter areas expect given the peculiar way in which area seems
composed mostly of axons) seem to account for a to scale with overall brain size in primates
greater portion of the increase (Schoenemann (Holloway, 2002). Nevertheless, because this area
et al., 2005). This makes sense given that prefrontal is specifically active in linguistic tasks that require
areas generally have an oversight role, modifying selection of appropriate words given a specific
activity in other posterior areas of the brain. semantic context (Gabrieli et al., 1998; Luke
Because prefrontal areas mediate a number of et al., 2002), it seems likely its increase is relevant
important behaviors besides language, language to language evolution.
evolution may not be the primary driving force
behind these changes. A variety of higher-order
behavioral abilities that were likely crucial for Concepts and semantic processing
human evolution are known to be mediated there,
including planning, maintaining behavioral goals, As discussed above, language depends critically
processing social information, temporary storage/ on a foundation of conceptual understanding of
manipulation of information (working memory), the world, which in turn appears to depend on a
 Author's personal copy

452

wide network of many different areas of the Areas involved in auditory processing—the pri-
brain. Humans are particularly biased toward mary auditory cortex of the temporal lobe
visual information, which, as a consequence, (Brodmann areas 41 and 42) and immediately
forms an important component of conceptual adjacent auditory association areas—appear to
understanding for most people (blind people be slightly larger than predicted based on overall
being an exception). Visual information pro- brain size (primary auditory by 6%, auditory
cessing starts in the retina of the eye and is trans- association by 17%; Deacon, 1997). In absolute
ferred through intermediate nuclei to the primary terms, these areas would still be more than
visual cortex, located in the occipital lobe (Figs. 2a approximately three to four times larger than
and 3b), where it becomes available to conscious the equivalent area in apes, suggesting the
awareness (Bear et al., 2007). From here, visual increases reflect enhancements with respect to
information is processed along two major pat- auditory processing.
hways: the dorsal stream (extending up into the In addition to areas involved in auditory pro-
parietal lobe), which processes information cessing, and conceptual and semantic processing,
regarding the location and motion of an object, the temporal lobe also contains the hippocampus
and the ventral stream (extending to the anterior (which plays a key role in memory formation)
tip of the temporal lobe), which processes infor- and the amygdala (which is central to marking
mation regarding the characteristics of the objects the emotional valence of events). Both of these
themselves (e.g., shape, color, etc.; Bear et al., structures lie on the medial surface of the tempo-
2007). The dorsal stream can therefore be ral lobes (hidden from lateral view). In absolute
thought of as the “where” pathway, and the ven- volume, the hippocampus is about 2.7 times larger
tral stream as the “what” pathway (Bear et al., in humans compared to chimpanzees (N¼1 for
2007). These two pathways consequently corre- each species, however; Stephan et al., 1981). In
spond at least broadly to the networks involved Stephan et al.’s (1987) data, the human amygdala
in conceptualizing objects (which get mapped as was 3.7 times larger than their chimpanzee’s (also
nouns) versus actions/orientations/directions N¼1), which may indicate that marking the emo-
(which are central to concepts generally mapped tional content and/or importance of events was
as verbs) (cf., Hurford, 2003b). particularly important.
The understanding of proper nouns appears to The planum temporale, located just posterior
depend on anterior and medial areas of the tem- to the primary auditory cortex, has been well
poral lobe, whereas understanding common studied comparatively. The planum temporale is
nouns appears to depend on the lateral and infe- larger on the left side in humans (Geschwind
rior temporal lobes (Damasio and Damasio, and Levitsky, 1968; Sommer et al., 2008), which
1992). In a comparative perspective, the human was assumed to reflect a functional anatomical
temporal lobe as a whole is 23% larger than pre- correlate of language evolution. However, apes
dicted given our brain size, or about four times show a similar asymmetry in this region (Gannon
larger in absolute terms (though this is similar to et al., 1998), suggesting that the lateralized size
overall brain size difference in this sample; data differences are not language specific. Details of
from Rilling and Seligman, 2002). The human dis- neuronal cell spatial organization have been
proportion in the temporal lobe appears to be shown to be asymmetric in the planum temporale
greatest with respect to white matter axonal of humans but not other primates, however, with
tracts, which represent long distance connections the left hemisphere displaying wider minicolumns
with other parts of the brain. This suggests a sig- than the right (Buxhoeveden et al., 2001). The
nificant increase in the dense network of intercon- functional significance of this cytoarchitectural
nectivity underlying conceptual understanding. asymmetry is unknown, though presumably it
 Author's personal copy

453

has some behavioral relevance. It may be that the area, again favoring the left (Gannon et al., 1998;
planum temporale has a role in auditory informa- Schenker et al., 2009). The left prefrontal overall
tion for communication generally (not just for has been shown to be particularly large relative
language). If so, it would be yet another example to the right, in humans compared to other
of language co-opting preexisting language- primates (Smaers et al., 2011). Although this
friendly circuits. appears to be largely predicted by primate scaling
With respect to the “where” pathway, which trends for these regions, it nevertheless likely has
likely grounds concepts central to most verbs, behavioral implications. Curiously, in humans, the
there is some indication of an evolutionary expan- whole right hemisphere is actually very slightly
sion of this area during human evolution, based larger overall than the left (1%; Allen et al.,
on analyses of brain endocasts of fossil hominins 2002). The author is not aware of studies compar-
(Bruner, 2004). Detailed comparative neuroana- ing the volumes of entire hemispheres in other
tomical studies of the parietal lobe have not been primates, however. Given that the right hemi-
reported, however. The semantic generation of sphere is very close to the left in overall size in
verbs (the actual words themselves) seems to also humans, it has probably increased approximately
involve Broca’s area (Damasio and Damasio, threefold along with the cortex as a whole. Also,
1992; Posner and Raichle, 1994) which also since Broca’s area (at least) is larger on the left
appears to have evolved disproportionately (see hemisphere, by definition other areas of the right
above). must therefore be larger than their corresponding
areas on the left, given that the right hemisphere
is slightly larger as a whole. Certainly, there is
Right hemisphere nothing to suggest that the right hemisphere has
particularly lagged behind the left overall during
Although the cortical language areas discussed so human evolution, though additional detailed com-
far are localized to the left hemisphere in most parative asymmetry studies remain to be done.
people, the right hemisphere also appears to be
involved in language. The right hemisphere unde-
rstands short words (Gazzaniga, 1970) and also Basal ganglia
entertains alternative possible meanings for par-
ticular words (Beeman and Chiarello, 1998), Although the cortex is heavily involved in language
suggesting it is involved in interpreting multiple processing, a group of interconnected nuclei deep
intended meanings of a given linguistic communi- to the cortex, collectively known as the basal
cation. The right hemisphere also plays a greater ganglia, appear to play important roles also. They
role in spatial processing in most people (Tzeng participate in an important circuit loop that
and Wang, 1984; Vallar, 2007), thus presumably functions in the selection and initiation of willed
grounding the semantics of spatial terms. The movements generally (Bear et al., 2007). A variety
right frontal lobe mediates aspects of prosody of studies have implicated these circuits not just in
(Alexander et al., 1989), which is critically impor- language production but also in language compre-
tant to understanding intended meaning. hension (see references in Hochstadt et al., 2006).
As discussed above, the planum temporale has Important symptoms of Parkinson’s and
been shown to be asymmetric in both apes and Huntington’s diseases, which affect the basal
humans (Gannon et al., 1998), but only humans ganglia in particular, include problems understand-
seem to show differences in neuronal spatial orga- ing complicated syntax (e.g., center-embedded
nization, favoring the left (Buxhoeveden et al., clauses), as well as processing semantic informa-
2001). Humans also show asymmetries in Broca’s tion (Hochstadt et al., 2006). Broca’s aphasia
 Author's personal copy

454

typically requires disruption to areas other than 2003), which have undergone a significant
Broca’s area, likely including circuits involving increase in relative proportion in both apes and
the basal ganglia (Lieberman, 2002). humans. With respect to overall body weight,
Comparative studies of the relative size of the human lateral cerebellar hemispheres are 2.9
basal ganglia in humans suggest that these nuclei times larger than predicted (MacLeod et al.,
are only about 65% as large as predicted for a pri- 2003) and therefore are not explained by any pos-
mate brain as large as ours (Schoenemann, 1997; sible differences in muscle mass. Given its role in
Stephan et al., 1981). However, they are still language processing, this increase may therefore
about twice as large in absolute terms as pre- be relevant to language evolution.
dicted based on body size. Since humans do not
appear to have significantly more sophisticated
willed movements than apes (except possibly Vocalization
those related to vocalization and hand manipula-
tion), it seems possible these increases indicate The muscles responsible for vocalization are
an important role supporting higher cortical directly innervated by nuclei in the brainstem,
functions including language. which relay signals from midbrain and higher cor-
tical areas. The muscles of the vocal folds, respon-
sible for changing the pitch of the primary vocal
Cerebellum signal, are innervated by the nucleus ambiguous.
The vocal signal is filtered in various ways by
The cerebellum is not only involved in mod- manipulations of the tongue, lower jaw, and lips,
ulating motor signals from the cortex (Bear thereby producing the sound changes we inter-
et al., 2007) but has also been implicated in higher pret as speech (Denes and Pinson, 1963). The
cognitive functions, including goal organization tongue is innervated by the hypoglossal nucleus
and planning, aspects of memory and learning, and nucleus ambiguous, the lower jaw by the tri-
attention, visuospatial processing, modulating geminal nucleus, and the lips by the facial motor
emotional responses, and language (Baillieux nucleus (all in the brain stem). This whole system
et al., 2008). The cerebellum appears to play a depends on the maintenance and manipulation of
role in speech production and perception, as well air pressure in the lungs, using the muscles of
as both semantic and grammatical processing the chest and abdomen, innervated by anterior
(Ackermann et al., 2007; Baillieux et al., 2008; horn areas of the spinal cord (Carpenter and
De Smet et al., 2007). The cerebellum also seems Sutin, 1983).
to play a role in timing mechanisms generally Even though vocal production is clearly impor-
(Ivry and Spencer, 2004). Given the importance tant to language, these brainstem nuclei do not
temporal information plays in language produc- appear to be particularly enlarged in humans.
tion and perception, this may explain why the The hypoglossal nucleus in humans is large but
cerebellum is implicated in studies of language shows substantial overlap with apes, while the
brain function. human trigeminal and facial motor nuclei
The human cerebellum is slightly smaller than completely overlap in size with those of the great
one would predict based on brain size (Rilling apes (Sherwood et al., 2005). Thus, there is little
and Insel, 1998) but is 2.9 times larger based evidence that language evolution substantially
on body size (the largest increase of all brain modified these nuclei.
regions outside the neocortex). The higher cogni- Because conscious muscle movement originates
tive functions appear to be localized to the lateral in the cerebral cortex, intentional speech utilizes
hemispheres of the cerebellum (MacLeod et al., direct connections from cortex to the relevant
 Author's personal copy

455

brainstem nuclei. Humans also have two indirect the brain that are relevant to language processing.
pathways, the first involving connections routed The increase in overall brain size paved the way
through the reticular formation of the brainstem for language both by encouraging localized corti-
(for tongue, lower jaw, and lip movement) and cal specialization and by making possible increas-
the nucleus retroambiguus (for muscles of respi- ingly complicated social interactions, which
ration; Deacon, 1997; Striedter, 2005), and a sec- provided the central usefulness for language,
ond through the cingulate gyrus of the limbic thereby driving its evolution. Specific brain areas
system (Jürgens, 1994), which mediates involun- relevant to language increased disproportion-
tary vocal responses to pain or strong emotions ately: the prefrontal cortex (areas relevant to
(Deacon, 1997; Striedter, 2005). semantics and syntax) and the temporal lobe
Comparative studies show that nonhuman (particularly, areas relevant to connecting words
primates have these same indirect pathways, to meanings and concepts). Broca’s and
allowing them a variety of emotionally mediated Wernicke’s areas, and the arcuate fasciculus con-
vocalizations, but have only weak direct connections necting them, were not specially evolved for lan-
controlling the tongue and respiration muscles, and guage but do appear to have been elaborated.
appear to completely lack direct connections to the Other areas that participate in language pro-
larynx (Jurgens, 2002; Jurgens and Alipour, 2002). cessing, including the basal ganglia and cerebel-
This suggests that at least some new direct cortical lum, are larger than predicted based on overall
pathways to the brainstem evolved for deliberate body weight, though they have not increased as
conscious vocalization in humans. much relative to overall brain size. Finally, the
auditory processing pathways do not appear to
have changed much, though pathways allowing
Auditory perception for enhanced conscious control of speech have
been either added (to the larynx) or
Conscious awareness of sound depends on signals strengthened.
from the cochlea (where sound is translated into These make excellent starting points for future
neural signals) reaching the primary auditory cor- research, which hopefully will clarify exactly how
tex located in the temporal lobe through a series circuitry in these areas support language and
of intermediate nuclei in the brainstem and mid- may have been modified by language evolution.
brain. Comparative data suggest that primates A better understanding of the complex
have somewhat smaller intermediate auditory interactions between areas will allow us to test
nuclei than expected for their brain weights some of the more interesting speculative ideas in
(though the primate sample was small; the literature (e.g., Deacon, 1997). Because lan-
Glendenning and Masterton, 1998). Both in abso- guage was not the only behavioral change in
lute terms and in relation to body size, human human evolution, unraveling the interplay
auditory nuclei appear to be reasonably large, between behavioral domains will be critical.
though not dramatically so. Overall, this suggests Assessing nonlanguage functions of “language”
only modest evolutionary changes in these nuclei. areas will also be crucial. All these findings are
consistent with the view that language adapted
to the brain more than the reverse (Christiansen
Conclusion and Chater, 2008).
Three major factors seem to have conspired to
Our current understanding of brain/language drive the evolution of language: first, the general
coevolution is based on comparative gross elaboration of—and increasing focus on—the
anatomical assessments of differences in parts of importance of learned behavior; second, a
 Author's personal copy

456

significant increase in the complexity, subtlety, References

and range of conceptual understanding that was
possible; and third, an increasingly complex, Aboitiz, F., & Garcia, V. R. (1997). The evolutionary origin of
the language areas in the human brain. A neuroanatomical
socially interactive existence. Because language
perspective. Brain Research. Brain Research Reviews, 25(3),
itself facilitates thinking and conceptual aware- 381–396.
ness, language evolution would have been a Ackermann, H., Mathiak, K., & Riecker, A. (2007). The con-
mutually reinforcing process: increasingly compli- tribution of the cerebellum to speech production and speech
cated brains led to increasingly rich and varied perception: Clinical and functional imaging data. Cerebellum
(London, England), 6(3), 202–213.
thoughts, driving the evolution of increasingly
Alexander, M. P., Benson, D. F., & Stuss, D. T. (1989). Frontal
complicated language, which itself facilitated even lobes and language. Brain and Language, 37, 656–691.
more complex conceptual world that these brains Allen, J. S., Damasio, H., & Grabowski, T. J. (2002). Normal
would then want to communicate (Savage- neuroanatomical variation in the human brain: An MRI-
Rumbaugh and Rumbaugh, 1993; Schoenemann, volumetric study. American Journal of Physical Anthropol-
ogy, 118, 341–358.
2009a,b). This process highlights the usefulness
Allman, J., McLaughlin, T., & Hakeem, A. (1993). Brain
of thinking about language evolution as a com- weight and life-span in primate species. Proceedings of
plex adaptive system. The extent to which the National Academy of Sciences of the United States of
increasing conceptual complexity itself might America, 90(1), 118–122.
have driven language evolution represents an Baillieux, H., De Smet, H. J., Paquier, P. F., De Deyn, P. P., &
Marien, P. (2008). Cerebellar neurocognition: Insights into
intriguing research question for the future.
the bottom of the brain. Clinical Neurology and Neurosur-
gery, 110(8), 763–773.
Barsalou, L. W. (2008). Grounded cognition. Annual Review
of Psychology, 59, 617–645.
Acknowledgments Bear, M. F., Connors, B. W., & Paradiso, M. A. (2007). Neuro-
science: Exploring the brain (3rd ed.). Baltimore;
This review has been adapted from Schoenemann Philadelphia, PA: Lippincott Williams & Wilkins.
Beckner, C., Blythe, R., Bybee, J., Christiansen, M. H., Croft, W.,
(2009a), which grew out of a contribution to the
Ellis, N. C., et al. (2009). Language is a complex adaptive
IIAS International Seminar on Language, Evolu- system: Position paper. Language Learning, 59(Suppl. 1), 1–26.
tion, and the Brain held in Kyoto, Japan in April Beeman, M. J., & Chiarello, C. (1998). Complementary right-
2007, organized by William Wang and Junjiro and left-hemisphere language comprehension. Current
Kanamori, as well as from Schoenemann Directions in Psychological Science, 7(1), 2–8.
Bickerton, D. (1990). Language and species. Chicago: Univer-
(2009b), which grew out of a workshop on lan-
sity of Chicago Press.
guage evolution at the Santa Fe Institute in Blinkov, S. M., & Glezer, I. I. (1968). The human brain in
March 2007, organized by John Holland. I wish figures and tables (B. Haigh, Trans.). New York: Plenum
to thank Michel Hofman for thoughtful comments Press.
and for pointing out errors in an earlier draft. Brodmann, K. (1909). Vergleichende lokalisationslehre der
grosshirnrinde in ihren prinzipien dargestellt auf grund des
Any errors that remain are of course my own.
zellenbaues. Leipzig: Johann Ambrosius Barth Verlag.
My thinking on language evolution has benefited Bruner, E. (2004). Geometric morphometrics and
from various discussions with William Wang, paleoneurology: Brain shape evolution in the genus homo.
James Minett, Vince Sarich, Jim Hurford, and Journal of Human Evolution, 47, 279–303.
Morten Christiansen. The galago and hedgehog Buxhoeveden, D. P., Switala, A. E., Litaker, M., Roy, E., &
Casanova, M. F. (2001). Lateralization of minicolumns in
brains in Fig. 2a and b were adapted with permis-
human planum temporale is absent in nonhuman primate
sion from images at http://www.brains.rad.msu. cortex. Brain, Behavior and Evolution, 57, 349–358.
edu, and http://brainmuseum.org, supported by Caplan, R., & Dapretto, M. (2001). Making sense during con-
the US National Science Foundation. versation: An fMRI study. Neuroreport: An International
 Author's personal copy

457

Journal for the Rapid Communication of Research in Neuro- Gazzaniga, M. S. (1970). The bisected brain. New York:
science, 12(16), 3625–3632. Appleton-Century-Crofts.
Carpenter, M. B., & Sutin, J. (1983). Human neuroanatomy Geschwind, N. (1974). Selected papers on language and the
(8th ed.). Baltimore, MD: Williams & Wilkins. brain. Boston: Reidel Publishing Co.
Changizi, M. A., & Shimojo, S. (2005). Parcellation and area- Geschwind, N., & Levitsky, W. (1968). Human brain: Left-
area connectivity as a function of neocortex size. Brain, right asymmetries in temporal speech region. Science (New
Behavior and Evolution, 66(2), 88–98. York, NY), 161(837), 186–187.
Christiansen, M. H. (1994). Infinite languages, finite minds: Gibson, K. R. (2002). Evolution of human intelligence: The
Connectionism, learning and linguistic structure. Unpublished roles of brain size and mental construction. Brain, Behavior
PhD dissertation, University of Edinburgh, Edinburgh, and Evolution, 59(1–2), 10–20.
Scotland. Gibson, K. R., Rumbaugh, D., & Beran, M. (2001). Bigger is
Christiansen, M. H., & Chater, N. (2008). Language as shaped better: Primate brain size in relationship to cognition. In
by the brain. The Behavioral and Brain Sciences, 31, D. Falk & K. Gibson (Eds.), Evolutionary anatomy of the
489–509. primate cerebral cortex (pp. 79–97). Cambridge: Cambridge
Damasio, A. R., & Damasio, H. (1992). Brain and language. University Press.
Scientific American, 267(3), 89–95. Gil-da-Costa, R., Martin, A., Lopes, M. A., Munoz, M.,
Damasio, H., Grabowski, T. J., Damasio, A., Tranel, D., Fritz, J. B., & Braun, A. R. (2006). Species-specific calls acti-
Boles-Ponto, L., Watkins, G. L., et al. (1993). Visual recall vate homologs of Broca’s and Wernicke’s areas in the
with eyes closed and covered activates early visual cortices. macaque. Nature Neuroscience, 9(8), 1064–1070.
Society for Neuroscience Abstracts, 19, 1603. Glendenning, K. K., & Masterton, R. B. (1998). Comparative
Darwin, C. (1882). The descent of man and selection in relation morphometry of mammalian central auditory systems: Vari-
to sex (2nd ed.). London: John Murray. ation in nuclei and form of the ascending system. Brain,
De Smet, H. J., Baillieux, H., De Deyn, P. P., Marien, P., & Behavior and Evolution, 51(2), 59–89.
Paquier, P. (2007). The cerebellum and language: The story Harvey, P. H., & Clutton-Brock, T. H. (1985). Life history var-
so far. Folia Phoniatrica et Logopaedica, 59(4), 165–170. iation in primates. Evolution, 39, 559–581.
Deacon, T. W. (1992). Brain-language coevolution. In J. A. Hochstadt, J., Nakano, H., Lieberman, P., & Friedman, J.
Hawkins & M. Gell-Mann (Eds.), The evolution of human (2006). The roles of sequencing and verbal working memory
languages (Vol. XI, pp. 49–83). Redwood City, CA: in sentence comprehension deficits in Parkinson’s disease.
Addison-Wesley. Brain and Language, 97(3), 243–257.
Deacon, T. W. (1997). The symbolic species: The co-evolution Hofman, M. A. (1983). Energy metabolism, brain size, and
of language and the brain. New York: W.W. Norton. longevity in mammals. The Quarterly Review of Biology,
Deaner, R. O., Isler, K., Burkart, J., & van Schaik, C. (2007). 58, 495–512.
Overall brain size, and not encephalization quotient, best Hofman, M. A. (1985). Size and shape of the cerebral cortex
predicts cognitive ability across non-human primates. Brain, in mammals. I. The cortical surface. Brain, Behavior and
Behavior and Evolution, 70(2), 115–124. Evolution, 27, 28–40.
Denes, P. B., & Pinson, E. N. (1963). The speech chain. Hofman, M. A. (1988). Size and shape of the cerebral cortex
Garden City, NY: Anchor Press/Doubleday. in mammals. II. The cortical volume. Brain, Behavior and
Dronkers, N. F. (2000). The pursuit of brain-language Evolution, 32, 17–26.
relationships. Brain and Language, 71(1), 59–61. Hofman, M. A. (1993). Encephalization and the evolution
Dunbar, R. I. M. (1995). Neocortex size and group size in of longevity in mammals. Journal of Evolutionary Biology,
primates: A test of the hypothesis. Journal of Human Evolu- 6, 209–227.
tion, 28, 287–296. Holloway, R. L. (1975). The role of human social behavior in
Finlay, B. L., Darlington, R. B., & Nicastro, N. (2001). Devel- the evolution of the brain. New York: American Museum
opmental structure in brain evolution. The Behavioral and of Natural History.
Brain Sciences, 24, 263–308. Holloway, R. L. (2002). Brief communication: How much
Gabrieli, J. D., Poldrack, R. A., & Desmond, J. E. (1998). The larger is the relative volume of area 10 of the prefrontal cor-
role of left prefrontal cortex in language and memory. Pro- tex in humans? American Journal of Physical Anthropology,
ceedings of the National Academy of Sciences of the United 118(4), 399–401.
States of America, 95(3), 906–913. Holmes, C. J., Hoge, R., Collins, L., Woods, R., Toga, A. W.,
Gannon, P. J., Holloway, R. L., Broadfield, D. C., & & Evans, A. C. (1998). Enhancement of mr images using
Braun, A. R. (1998). Asymmetry of chimpanzee planum registration for signal averaging. Journal of Computer
temporale: Humanlike pattern of Wernicke’s brain language Assisted Tomography, 22(2), 324–333.
area homolog. Science, 279(5348), 220–222.
 Author's personal copy

458

Humphrey, N. (1984). The social function of intellect. In Con- Maguire, E. A., & Frith, C. D. (2004). The brain network
sciousness regained (pp. 14–28). Oxford: Oxford University associated with acquiring semantic knowledge. NeuroImage,
Press. 22(1), 171–178.
Hurford, J. R. (2003a). The language mosaic and its evolution. Martin, R. D. (1981). Relative brain size and basal metabolic
In M. H. Christiansen & S. Kirby (Eds.), Language evolu- rate in terrestrial vertebrates. Nature, 293, 57–60.
tion (pp. 38–57). Oxford: Oxford University Press. Mueller, R.-A. (1996). Innateness, autonomy, universality?
Hurford, J. R. (2003b). The neural basis of predicate-argument Neurobiological approaches to language. The Behavioral
structure. The Behavioral and Brain Sciences, 26(3), 261–283 and Brain Sciences, 19(4), 611–675.
discussion 283–316. Nieuwenhuys, R., Voogd, J., & van Huijzen, C. (2008). The
Indefrey, P., Hellwig, F., Herzog, H., Seitz, R. J., & human central nervous system (4th ed.). Berlin: Springer-
Hagoort, P. (2004). Neural responses to the production Verlag.
and comprehension of syntax in identical utterances. Brain Noppeney, U., & Price, C. J. (2004). Retrieval of abstract
and Language, 89(2), 312–319. semantics. NeuroImage, 22, 164–170.
Ivry, R. B., & Spencer, R. M. (2004). The neural representa- Northcutt, R. G., & Kaas, J. H. (1995). The emergence
tion of time. Current Opinion in Neurobiology, 14(2), and evolution of mammalian neocortex. Trends in
225–232. Neurosciences, 18(9), 373–379.
Jerison, H. J. (1973). Evolution of the brain and intelligence. Novoa, O. P., & Ardila, A. (1987). Linguistic abilities
New York: Academic Press. in patients with prefrontal damage. Brain and Language,
Jurgens, U. (2002). Neural pathways underlying vocal control. 30, 206–225.
Neuroscience and Biobehavioral Reviews, 26(2), 235–258. Nucifora, P. G., Verma, R., Melhem, E. R., Gur, R. E., &
Jürgens, U. (1994). The role of the periaqueductal grey in Gur, R. C. (2005). Leftward asymmetry in relative fiber den-
vocal behaviour. Behavioural Brain Research, 62(2), sity of the arcuate fasciculus. Neuroreport, 16(8), 791–794.
107–117. Petrides, M., Cadoret, G., & Mackey, S. (2005). Orofacial
Jurgens, U., & Alipour, M. (2002). A comparative study on the somatomotor responses in the macaque monkey homologue
cortico-hypoglossal connections in primates, using biotin of Broca’s area. Nature, 435(7046), 1235–1238.
dextranamine. Neuroscience Letters, 328(3), 245–248. Petrides, M., & Pandya, D. N. (2002). Comparative
Kerns, J. G., Cohen, J. D., Stenger, V. A., & Carter, C. S. cytoarchitectonic analysis of the human and the macaque
(2004). Prefrontal cortex guides context-appropriate ventrolateral prefrontal cortex and corticocortical connec-
responding during language production. Neuron, 43(2), tion patterns in the monkey. The European Journal of Neu-
283–291. roscience, 16(2), 291–310.
Kohler, E., Keysers, C., Umilta, M. A., Fogassi, L., Gallese, V., Pinker, S. (1995). Facts about human language relevant to its
& Rizzolatti, G. (2002). Hearing sounds, understanding evolution. In J.-P. Changeux & J. Chavaillon (Eds.), Origins
actions: Action representation in mirror neurons. Science, of the human brain (pp. 262–283). Oxford: Clarendon.
297(5582), 846–848. Poeppel, D., & Hickok, G. (2004). Towards a new functional
Kosslyn, S. M., Alpert, N. M., Thompson, W. L., anatomy of language. Cognition, 92(1–2), 1–12.
Maljkovic, V., Weise, S. B., Chabris, C. F., et al. (1993). Posner, M. I., & Raichle, M. E. (1994). Images of mind. New
Visual mental imagery activates topographically organized York: W.H. Freeman.
visual cortex: Pet investigations. Journal of Cognitive Neuro- Reader, S. M., & Laland, K. N. (2002). Social intelligence,
science, 5, 263–287. innovation, and enhanced brain size in primates. Pro-
Lieberman, P. (2000). Human language and our reptilian brain: ceedings of the National Academy of Sciences of the United
The subcortical bases of speech, syntax, and thought. States of America, 99(7), 4436–4441.
Cambridge, MA: Harvard University Press. Rilling, J. K., Glasser, M. F., Preuss, T. M., Ma, X., Zhang, X.,
Lieberman, P. (2002). On the nature and evolution of the neu- Zhao, T., et al. (2007). A comparative diffusion tensor imag-
ral bases of human language. American Journal of Physical ing (DTI) study of the arcuate fasciculus language pathway
Anthropology, 45, 36–62. in humans, chimpanzees and rhesus macaques. American
Luke, K. K., Liu, H. L., Wai, Y. Y., Wan, Y. L., & Tan, L. H. Journal of Physical Anthropology, 132(S44), 199–200.
(2002). Functional anatomy of syntactic and semantic pro- Rilling, J. K., & Insel, T. R. (1998). Evolution of the cerebel-
cessing in language comprehension. Human Brain Mapping, lum in primates: Differences in relative volume among
16(3), 133–145. monkeys, apes and humans. Brain, Behavior and Evolution,
MacLeod, C. E., Zilles, K., Schleicher, A., Rilling, J. K., & 52(6), 308–314.
Gibson, K. R. (2003). Expansion of the neocerebellum Rilling, J. K., & Seligman, R. A. (2002). A quantitative mor-
in hominoidea. Journal of Human Evolution, 44(4), phometric comparative analysis of the primate temporal
401–429. lobe. Journal of Human Evolution, 42(5), 505–533.
 Author's personal copy

459

Ringo, J. L. (1991). Neuronal interconnection as a function of Sherwood, C. C., Hof, P. R., Holloway, R. L., Semendeferi, K.,
brain size. Brain, Behavior and Evolution, 38, 1–6. Gannon, P. J., Frahm, H. D., et al. (2005). Evolution of the
Savage-Rumbaugh, E. S., & Rumbaugh, D. M. (1993). The brainstem orofacial motor system in primates: A compara-
emergence of language. In K. R. Gibson & T. Ingold tive study of trigeminal, facial, and hypoglossal nuclei. Jour-
(Eds.), Tools, language and cognition in human evolution nal of Human Evolution, 48(1), 45–84.
(pp. 86–108). Cambridge: Cambridge University Press. Smaers, J. B., Steele, J., Case, C. R., Cowper, A., Amunts, K.,
Schenker, N. M., Hopkins, W. D., Spocter, M. A., & Zilles, K. (2011). Primate prefrontal cortex evolution:
Garrison, A. R., Stimpson, C. D., Erwin, J. M., et al. Human brains are the extreme of a lateralized ape trend.
(2009). Broca’s area homologue in chimpanzees (Pan Brain, Behavior and Evolution, 77, 67–78.
troglodytes): Probabilistic mapping, asymmetry, and com- Sommer, I. E., Aleman, A., Somers, M., Boks, M. P., &
parison to humans. Cerebral Cortex, 20(3), 730–742. Kahn, R. S. (2008). Sex differences in handedness, asymme-
Schoenemann, P. T. (1997). An MRI study of the relationship try of the planum temporale and functional language lateral-
between human neuroanatomy and behavioral ability. ization. Brain Research, 1206, 76–88.
Unpublished dissertation, University of California, Stephan, H., Frahm, H., & Baron, G. (1981). New and revised
Berkeley, Berkeley. data on volumes of brain structures in insectivores and
Schoenemann, P. T. (1999). Syntax as an emergent character- primates. Folia Primatologica, 35, 1–29.
istic of the evolution of semantic complexity. Minds and Stephan, H., Frahm, H. D., & Baron, G. (1987). Comparison
Machines, 9, 309–346. of brain structure volumes in insectivora and primates.
Schoenemann, P. T. (2005). Conceptual complexity and VII. Amygdaloid components. Journal für Hirnforschung,
the brain: Understanding language origins. In W. S.-Y. 28(5), 571–584.
Wang & J. W. Minett (Eds.), Language acquisition, Striedter, G. F. (2005). Principles of brain evolution. Sun-
change and emergence: Essays in evolutionary linguistics derland, MA: Sinauer Associates.
(pp. 47–94). Hong Kong: City University of Hong Kong Taglialatela, J. P., Russell, J. L., Schaeffer, J. A., & Hopkins, W. D.
Press. (2008). Communicative signaling activates ‘Broca’s’ homolog
Schoenemann, P. T. (2006). Evolution of the size and func- in chimpanzees. Current Biology, 18(5), 343–348.
tional areas of the human brain. Annual Review of Anthro- Thompson-Schill, S. L., Swick, D., Farah, M. J.,
pology, 35, 379–406. D’Esposito, M., Kan, I. P., & Knight, R. T. (1998). Verb
Schoenemann, P. T. (2009a). Brain evolution relevant to lan- generation in patients with focal frontal lesions: A neuropsy-
guage. In J. Minett & W. S.-Y. Wang (Eds.), Language, evo- chological test of neuroimaging findings. Proceedings of
lution, and the brain. Hong Kong: City University of Hong the National Academy of Sciences of the United States of
Kong Press. America, 95(26), 15855–15860.
Schoenemann, P. T. (2009b). Evolution of brain and language. Tooby, J., & Cosmides, L. (1992). The psychological
Language Learning, 59(Suppl. 1), 162–186. foundations of culture. In J. H. Barkow, L. Cosmides & J.
Schoenemann, P. T. (2010). The meaning of brain size: The Tooby (Eds.), The adapted mind: Evolutionary psychology
evolution of conceptual complexity. In Human brain and the generation of culture (pp. 19–136). New York:
evolving: Papers in honor of Ralph Holloway. Bloomington, Oxford University Press.
Indiana: Stone Age Institute Press. Tzeng, O. J. L., & Wang, W. S.-Y. (1984). Search for a
Schoenemann, P. T., Sheehan, M. J., & Glotzer, L. D. (2005). common neurocognitive mechanism for language and
Prefrontal white matter volume is disproportionately movements. American Journal of Physiology. Regulatory,
larger in humans than in other primates. Nature Neuroscience, Integrative and Comparative Physiology, 246(15),
8(2), 242–252. R904–R911.
Semendeferi, K., Armstrong, E., Schleicher, A., Zilles, K., & Vallar, G. (2007). Spatial neglect, Balint-Homes’ and
Van Hoesen, G. W. (1998). Limbic frontal cortex in Gerstmann’s syndrome, and other spatial disorders. CNS
hominoids: A comparative study of area 13. American Jour- Spectrums, 12(7), 527–536.
nal of Physical Anthropology, 106(2), 129–155. White, J. G., Southgate, E., Thomson, J. N., & Brenner, S.
Semendeferi, K., Armstrong, E., Schleicher, A., Zilles, K., & (1986). The structure of the nervous system of the nematode
Van Hoesen, G. W. (2001). Prefrontal cortex in humans Caenorhabditis elegans. Philosophical Transactions of the
and apes: A comparative study of area 10. American Journal Royal Society of London B, Biological Sciences, 314(1165),
of Physical Anthropology, 114(3), 224–241. 1–340.