The evolution of the cortico-cerebellar complex in primates: anatomical connections predict patterns of correlated evolution

https://doi.org/10.1016/S0047-2484(02)00162-8Get rights and content

Abstract

Investigations into the evolution of the primate brain have tended to neglect the role of connectivity in determining which brain structures have changed in size, focusing instead on changes in the size of the whole brain or of individual brain structures, such as the neocortex, in isolation. We show that the primate cerebellum, neocortex, vestibular nuclei and relays between them exhibit correlated volumetric evolution, even after removing the effects of change in other structures. The patterns of correlated evolution among individual nuclei correspond to their known patterns of connectivity. These results support the idea that the brain evolved by mosaic size change in arrays of functionally connected structures. Furthermore, they suggest that the much discussed expansion of the primate neocortex should be re-evaluated in the light of conjoint cerebellar expansion.

Introduction

Compared to other mammals of similar body size, primates have evolved an unusually large brain (Passingham, 1982, Deacon, 1990). This large size is not thought to be due to the production of new structures, but to the modification and expansion of existing ones (Simpson, 1967, Jacob, 1982, Preuss, 1995). Hence, the brains of primates and other mammals are similar in terms of the types of structures that they contain, but different in the precise form, arrangement and relative size of these structures. Differences can vary from simple changes in size to more complex reorganization of neurons and their connections. However, research on brain evolution has focused primarily on simple size changes in overall brain size or individual brain structures, particularly the neocortex (e.g., Dunbar, 1992, Barton and Dunbar, 1997). While such research is certainly justified, it is limited by the fact that brain structures do not function in isolation, but rather as parts of distributed systems (Van Essen et al., 1992, Young et al., 1994). Such systems consist of arrays of brain components whose interconnections serve to coordinate the processing of a particular type of information. Within a system, each component is specialized for a particular aspect of the systems information processing: these are functional systems with a division of labour.

Investigations into the evolution of the primate brain should take account of these systems by looking at interconnected structures rather than focusing solely on individual brain regions. At present, there is controversy over whether the brain has tended to evolve as a coordinated whole (Finlay & Darlington, 1995) or whether individual structures or systems evolved independently of changes in other parts of the brain (mosaic evolution, e.g., Barton & Harvey, 2000). If mosaic change has occurred, this should be most apparent at the level of integrated functional systems. This paper tests the mosaic change hypothesis by analysing patterns of correlated evolution among interconnected structures. Given that functional systems are distributed across multiple structures, it is predicted that individual structures within a functional system will exhibit correlated evolution independent of evolutionary change in other systems. However, brain structures may, and often do, participate in more than a single functional system (most notably the neocortex which is involved in numerous different systems). This means that changes in the size of one structure may be related to a large number of functional systems, not just one. For this reason, it is desirable to look, as far as it is possible, at functionally specific sub-regions or nuclei within gross brain structures. The extent to which this can be done is, however, limited by the anatomical resolution of the available data (see below).

In this paper we focus on interrelationships among the cerebellum, neocortex, the relays between them (pons and thalamus) and the vestibular system. Cerebellar systems are of interest because recent evidence suggests that the cerebellum has expanded in some groups of primates (Rilling & Insel, 1998). In addition, preliminary evidence indicates conjoint expansion of the cerebellum with a structure that has been the subject of intense scrutiny, the neocortex (Barton & Harvey, 2000). Finally, volumetric comparative data are available on individual cerebellar and vestibular nuclei (Matano & Hirasaki, 1997).

Section snippets

Cerebellar systems

The cerebellum is a highly heterogeneous structure that has been implicated in the planning, execution and control of motor actions as well as, more controversially, in a number of cognitive functions (Fiez et al., 1992, Leiner et al., 1993, Fabbro et al., 2000). The cerebellum receives input from the neocortex (via the pons), the vestibular system (lateral vestibular nucleus) and the spinal cord. The output structures of the cerebellum are the cerebellar nuclei, which send projections to the

Methods

We used the method of independent contrasts, which enables the assessment of correlated evolution in comparative data sets (Felsenstein, 1985, Harvey and Pagel, 1991). The method works by calculating standardized contrasts between sister taxa in the phylogeny. Hence, a contrast score represents the evolutionary change that has occurred since the common ancestor of the sister taxa. These contrasts can then be subjected to standard methods of correlation and regression. The particular computer

Results

In each case the significance level is set at P<0.05. The graphs are provided only for those correlations reaching significance.

Discussion

These results corroborate the suggestion that, during the radiation of the primate order, the neocortex and cerebellum have undergone correlated evolution (Barton and Harvey, 2000, Barton, 2002). They additionally show that other related structures, the pons, thalamus and vestibular complex, have also changed in concert. Furthermore, we have shown that, at a finer scale, the patterns of correlated evolution are to a great extent predictable from information on anatomical connectivity. Hence

Conclusions and summary

Previous work on the evolution of the primate brain has generally focused on changes in individual structures. Brain structures, however, do not function in isolation, but rather contribute to distributed functional systems. The present analyses demonstrate correlated evolution among neocortex, cerebellum, vestibular complex and relay stations (pons and thalamus) and, at as fine a scale as allowed by the available data, the patterns of correlated evolution reflect functional connectivity.

References (26)

  • B.L. Finlay et al.

    Linked regularities in the development and evolution of mammalian brains

    Science

    (1995)
  • P.H. Harvey et al.

    The Comparative Method in Evolutionary Biology

    (1991)
  • F. Jacob

    The Possible and the Actual

    (1982)
  • Cited by (108)

    • Emotional disorders and the cerebellum: Neurobiological substrates, neuropsychiatry, and therapeutic implications

      2021, Handbook of Clinical Neurology
      Citation Excerpt :

      The dentate nucleus conveys the cerebellar posterior lobe efferents to these higher-order regions of the cerebral cortex (Middleton and Strick, 1994). The ventral and lateral parts of the dentate nucleus (Dum and Strick, 2003), the portions most involved in these projections, have expanded through evolution, along with the cerebellar posterior lobe and the cerebral cortical association areas (Dow, 1974; Leiner et al., 1986; Whiting and Barton, 2003). Trans-synaptic viral tracer studies in monkey reveal that the motor cortex in the precentral gyrus is linked with cerebellar lobules V, VI, and VIII, whereas dorsolateral prefrontal cortex area 46 is linked with crus I, crus II, and lobule IX in the cerebellar posterior lobe (Fig. 6.9; Strick et al., 2009).

    • Brain size expansion in primates and humans is explained by a selective modular expansion of the cortico-cerebellar system

      2019, Cortex
      Citation Excerpt :

      When comparing brain region sizes across a broad sample of species, neuroanatomical information is scarce and comprises a lower level of anatomical resolution than can be obtained from model species. For example, early studies on the co-evolution of the cortico-cerebellar system consisted of evolutionary correlations between whole neocortex and whole cerebellar cortex (Whiting & Barton, 2003). While studies comparing brain organization across a broad sample of species currently still suffer from lower anatomical resolution than studies using model species, continued efforts to overcome this drawback are of paramount importance because of the substantial benefits of addressing neuroanatomical questions with a high comparative resolution.

    • The human connectome from an evolutionary perspective

      2019, Progress in Brain Research
      Citation Excerpt :

      If the advanced cognitive capacities of humans cannot be explained simply by a disproportionately sized neocortex, then which other adaptations underlie human brain function? Accumulating evidence suggests that evolutionary changes to specific brain components play a role beyond overall volume increases of the neocortex (Miller et al., 2019; Smaers and Soligo, 2013; Whiting and Barton, 2003). One influential hypothesis postulates that, as the brain expanded, association areas became progressively untethered from the strong molecular gradients that constrain the development of primary areas, allowing for more plasticity of association areas (Buckner and Krienen, 2013) (Fig. 2, top).

    • The cerebellum and cognition

      2019, Neuroscience Letters
    View all citing articles on Scopus
    View full text