Learned birdsong and the neurobiology of human language

Abstract

Vocal learning, the substrate for human language, is a rare trait found to date in only three distantly related groups of mammals (humans, bats, and cetaceans) and three distantly related groups of birds (parrots, hummingbirds, and songbirds). Brain pathways for vocal learning have been studied in the three bird groups and in humans. Here I present a hypothesis on the relationships and evolution of brain pathways for vocal learning among birds and humans. The three vocal learning bird groups each appear to have seven similar but not identical cerebral vocal nuclei distributed into two vocal pathways, one posterior and one anterior. Humans also appear to have a posterior vocal pathway, which includes projections from the face motor cortex to brainstem vocal lower motor neurons, and an anterior vocal pathway, which includes a strip of premotor cortex, the anterior basal ganglia, and the anterior thalamus. These vocal pathways are not found in vocal non-learning birds or mammals, but are similar to brain pathways used for other types of learning. Thus, I argue that if vocal learning evolved independently among birds and humans, then it did so under strong genetic constraints of a pre-existing basic neural network of the vertebrate brain.

FIGURE 1

Family trees of living mammalian and avian orders. The mammalian tree is derived from the morphological analysis by Novacek;, horizontal lines indicate extant of geologic evidence from fossils. The avian tree is derived from DNA-DNA hybridization analysis by Sibley and Alquist (page 838). The Latin name of each order is given along with examples of common species. Passeriformes are divided into its two suborders, suboscine and oscine songbirds. The vertical line down the trees indicates the cretaceous-tertiary boundary, the time of the dinosaur extinction; MYA=millions of years ago. Open and closed circles show the minimal ancestral nodes where vocal learning could have either evolved independently or been lost independently. Independent losses would have at least required one common vocal learning ancestor, located by the right-facing arrows. Within primates, there would have to have been least seven independent losses (tree shrews, prosimians, new and old world monkeys, apes, and chimps) followed by the regaining of vocal learning in humans (assuming that all non-human primates are vocal non-learners). Both trees are modified here from the original sources– such that they include updated information and are stylistically more comparable. The trees are not meant to present the final dogma of mammalian and avian evolution, as there are many differences of opinion among scientists. Rather, the trees presented here are those that lead to more conservative interpretations of the evolution of vocal learning. For example, one alternative view of avian evolution posted on <http://tolweb.org/tree?group=Neornithes&contgroup=Aves> as of 2/15/2004 would have resulted in the interpretation of 16 independent losses of vocal learning instead of three.

FIGURE 2

Proposed comparable vocal and auditory brain areas among vocal learning birds and humans: (A) parrot, (B) hummingbird, (C) songbird, and (D) human. Left hemispheres are shown, as this is the dominant side for human language. White regions and black arrows indicate proposed posterior vocal pathways; dark grey regions and white arrows indicate proposed anterior vocal pathways; dashed lines show connections between the two vocal pathways; light grey indicates auditory regions. For simplification, not all connections are shown. The globus pallidus in the human brain, also not shown, is presumably part of the anterior pathway as in non-vocal pathways of mammals. Basal ganglia, thalamic, and midbrain (for the human brain) regions are drawn with dashed-line boundaries to indicate that they are deeper in the brain relative to the anatomical structures above them. The anatomical boundaries drawn for the proposed human brain regions involved in vocal and auditory processing should be interpreted conservatively and for heuristic purposes only. Human brain lesions and brain imaging studies do not allow one to determine functional anatomical boundaries with high resolution. Scale bar: ~7 mm. Abbreviations are in Table 1.

FIGURE 3

Comparative and simplified connectivity of posterior and anterior vocal pathways in (A) songbirds, (B) parrots, and (C) mammals. Dashed lines: connections between posterior anterior pathways; inputs and outputs are labeled relative to anterior pathways. Output from songbird MAN to HVC and RA is not from the same neurons; medial MAN neurons project to HVC, lateral MAN neurons project to RA. ○, excitatory neurons; • inhibitory neurons; +, excitatory glutamate neurotransmitter release; − inhibitory GABA release. MSp, medium spiny neurons. GPn, globus pallidus-like neuron in songbird Area X and parrot MMSt. Only the direct pathway through the mammalian basal ganglia (St to GPi) is shown as this is the one most similar to Area X connectivity (MSp to GPn). X-p, X-projecting neuron of HVC. RA-p, RA-projecting neuron of HVC. PT-5, pyramidal tract neuron of motor cortex layer 5. IT-3, Intratelencephalic projecting neuron of layer 3. Connections that need validation for this modelto be correct are whether collaterals of the same neurons of mMAN project to mArea X and to HVC, as opposed to different neurons, whether input from HVC into Area X is onto the Aera X MSp neurons, whether the microcircuitry in parrot MMst is the same as in songbirds, whether the collaterals of single IT-3 neurons of mammal cortex send branches to both layers 3 and 5 of motor cortex or just to one layer per IT-3 neuron. Abbreviation are in Table 1.

FIGURE 4

Comparative and simplified connectivity among auditory pathways in reptiles, mammals, and birds, placed in order from left to right of the most recently evolved. The connectivity from CM to CSt in birds needs verification by retrograde tracing. Abbreviations are in Table 1.

FIGURE 5

Human brain areas where damage results in speaking and/or hearing deficits. (A) Surface view of the left side of a human brain. (B) Frontal section cut through the prefrontal areas that show verbal aphasias and brain activation when speaking. Also highlighted are the face motor cortex (FMC) and auditory areas. (C) Saggital section highlighting anterior, cortical, basal ganglia, and thalamic areas that when damaged appear to lead to aphasia deficits. The arrows indicate proposed connectivity based upon that found in non-human mammals. The exact anatomical boundaries drawn for the proposed brain regions involved in vocal and auditory processing should be interpreted conservatively and for heuristic purposes only. Human brain lesions and brain imaging studies do not allow one to determine functional anatomical boundaries with high resolution. The image in A is from John W. Sundsten of the Digital Anatomist Project <http://www9.biostr.washington.edu/da.html>; Those in B and C, are from S. Mark Williams at the Duke University Medical Center. All images are used with permission. Abbreviations are in Table 1.

FIGURE 6

Some of the brain areas that show hearing and vocalizing-driven ZENK gene expression in vocal learning birds: songbirds (canary), parrots (budgerigar), and hummingbirds (sombre). Each image is a sagittal section each from a different bird of a different behavioral condition for 30 minutes; white label is ZENK mRNA gene expression, accumulated from over 30-minutes; grey background is cresyl violet stained cells. Quiet control animals did not hear songs or sing for an extended period of time. Hearing only animals heard playbacks of their species-specific songs. Hearing & Vocalizing animals heard similar playbacks and sang. The quiet control canary was actively moving around, causing ZENK expression in areas around vocal nuclei. It is not possible to have all activated brain regions in one section. In parrots, the HVC analogue, NLC (not shown), is situated more laterally, and it is also in hummingbirds (VLN shown). The auditory activated areas NCM and CMM in hummingbirds are shown without the rest of the brain. In the vocalizing canary, MO is not distinctly revealed by ZENK, whereas it is in zebra finches. The MO analogue in hummingbirds, VAH, is flat and small. Area X, MMSt, and VAS of songbirds, parrots and hummingbirds respectively are all in the same area of the striatum, but have different vocalizing-driven gene expression levels and different shapes. Songbird images are from ref. , parrot from ref. , and hummingbird from ref. . Scale bar 1mm.