ArticlePDF Available

Abstract and Figures

Significance Birds are remarkably intelligent, although their brains are small. Corvids and some parrots are capable of cognitive feats comparable to those of great apes. How do birds achieve impressive cognitive prowess with walnut-sized brains? We investigated the cellular composition of the brains of 28 avian species, uncovering a straightforward solution to the puzzle: brains of songbirds and parrots contain very large numbers of neurons, at neuronal densities considerably exceeding those found in mammals. Because these “extra” neurons are predominantly located in the forebrain, large parrots and corvids have the same or greater forebrain neuron counts as monkeys with much larger brains. Avian brains thus have the potential to provide much higher “cognitive power” per unit mass than do mammalian brains.
Content may be subject to copyright.
Birds have primate-like numbers of neurons in
the forebrain
Seweryn Olkowicz
a
, Martin Kocourek
a
, Radek K. Luˇcan
a
, Michal Porteš
a
, W. Tecumseh Fitch
b
,
Suzana Herculano-Houzel
c,d,1
, and Pavel N
emec
a,2
a
Department of Zoology, Faculty of Science, Charles University in Prague, CZ-12844 Prague, Czech Republic;
b
Department of Cognitive Biology, University
of Vienna, 1090 Vienna, Austria;
c
Instituto de Ciências Biomédicas, Universidade Federal do Rio de Janeiro, CEP 21941-902, Rio de Janeiro, Brazil;
and
d
Instituto Nacional de Neurociência Translacional, Ministério da Ciência e Tecnologia/Conselho Nacional de Pesquisas, CEP 04023-900, São Paulo, Brazil
Edited by Dale Purves, Duke University, Durham, NC, and approved May 6, 2016 (received for review August 27, 2015)
Some birds achieve primate-like levels of cognition, even though
their brains tend to be much smaller in absolute size. This poses a
fundamental problem in comparative and computational neuro-
science, because small brains are expected to have a lower
information-processing capacity. Using the isotropic fractionator
to determine numbers of neurons in specific brain regions, here
we show that the brains of parrots and songbirds contain on
average twice as many neurons as primate brains of the same
mass, indicating that avian brains have higher neuron packing
densities than mammalian brains. Additionally, corvids and parrots
have much higher proportions of brain neurons located in the
pallial telencephalon compared with primates or other mammals
and birds. Thus, large-brained parrots and corvids have forebrain
neuron counts equal to or greater than primates with much larger
brains. We suggest that the large numbers of neurons concen-
trated in high densities in the telencephalon substantially contrib-
ute to the neural basis of avian intelligence.
intelligence
|
evolution
|
brain size
|
number of neurons
|
birds
Many birds have cognitive abilities that match or surpass
those of mammals (1). Corvids and parrots appear to be
cognitively superior to other birds, rivalling great apes in many
psychological domains (13). They manufacture and use tools (4,
5), solve problems insightfully (6), make inferences about causal
mechanisms (7), recognize themselves in a mirror (8), plan for
future needs (9), and use their own experience to anticipate future
behavior of conspecifics (10) or even humans (11), to mention just
a few striking abilities. In addition, parrots and songbirds (in-
cluding corvids) share with humans and a few other animal
groups a rare capacity for vocal learning (12), and parrots can
learn words and use them to communicate with humans (13).
Superficially, the architecture of the avian brain appears very
different from that of mammals, but recent work demonstrates
that, despite a lack of layered neocortex, large areas of the avian
forebrain are homologous to mammalian cortex (1416), con-
form to the same organizational principles (15, 17, 18), and play
similar roles in higher cognitive functions (14, 19), including
executive control (20, 21). However, bird brains are small and
the computational mechanisms enabling corvids and parrots to
achieve ape-like intelligence with much smaller brains remain
unclear. The notion that higher encephalization (relative brain
size deviation from brainbody allometry) endows species with
improved cognitive abilities has recently been challenged by data
suggesting that intelligence instead depends on the absolute
number of cerebral neurons and their connections (2225). This
is in line with recent findings that absolute rather than relative
brain size is the best predictor of cognitive capacity (2628).
However, although corvids and parrots feature encephalization
comparable to that of monkeys and apes, their absolute brain
size remains small (29, 30). The largest average brain size in
corvids and parrots does not exceed 15.4 g found in the common
raven (29) and 24.7 g found in the hyacinth macaw (30), re-
spectively. Do corvids and parrots provide a strong case for re-
viving encephalization as a valid measure of brain functional
capacity? Not necessarily: it has recently been discovered that
the relationship between brain mass and number of brain neu-
rons differs starkly between mammalian clades (31). Avian
brains seem to consist of small, tightly packed neurons, and it is
thus possible that they can accommodate numbers of neurons
that are comparable to those found in the much larger primate
brains. However, to date, no quantitative data have been avail-
able to test this hypothesis.
Here, we analyze how numbers of neurons compare across
birds and mammals (3239) of equivalent brain mass, and de-
termine the cellular scaling rules for brains of songbirds and
parrots. Using the isotropic fractionator (40), we estimated the
total numbers of neuronal and nonneuronal cells in the cerebral
hemispheres, cerebellum, diencephalon, tectum, and brainstem
in a sample of 11 parrot species, 13 vocal learning songbird species
(including 6 corvids), and 4 additional model species representing
other avian clades (Figs. S1 and S2). Because most of the cited
mammalian studies analyzed cellular composition of only three
brain subdivisions, namely the pallium (referred to as the cerebral
cortex in those papers), the cerebellum, and rest of brain, we
divided the avian brain identically to ensure an accurate com-
parison of neuronal numbers, densities, and relative distribution
of neurons in birds and mammals. Specifically, the avian pallium
(comprising the hyperpallium, mesopallium, nidopallium, arcopallium,
and hippocampus) was compared with its homologthe mam-
malian pallium (comprising the neocortex, hippocampus, olfac-
tory cortices such as piriform and entorhinal cortex, and pallial
Significance
Birds are remarkably intelligent, although their brains are
small. Corvids and some parrots are capable of cognitive feats
comparable to those of great apes. How do birds achieve im-
pressive cognitive prowess with walnut-sized brains? We in-
vestigated the cellular composition of the brains of 28 avian
species, uncovering a straightforward solution to the puzzle:
brains of songbirds and parrots contain very large numbers of
neurons, at neuronal densities considerably exceeding those
found in mammals. Because these extraneurons are pre-
dominantly located in the forebrain, large parrots and corvids
have the same or greater forebrain neuron counts as monkeys
with much larger brains. Avian brains thus have the potential
to provide much higher cognitive powerper unit mass than
do mammalian brains.
Author contributions: S.O., M.K., S.H.-H., and P.N. designed research; S.O., M.K., R.K.L., M.P., and
P.N. performed research; R.K.L. and M.P. collected experimental animals; S.O., M.K., S.H.-H., and
P.N. analyzed data; and S.O., M.K., W.T.F., S.H.-H., and P.N. wrote the paper.
The authors declare no conflict of interest.
This article is a PNAS Direct Submission.
Freely available online through the PNAS open access option.
1
Present address: Department of Psychology and Department of Biological Sciences,
Vanderbilt University, Nashville, TN 37240.
2
To whom correspondence should be addressed. Email: pgnemec@natur.cuni.cz.
This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10.
1073/pnas.1517131113/-/DCSupplemental.
www.pnas.org/cgi/doi/10.1073/pnas.1517131113 PNAS
|
June 28, 2016
|
vol. 113
|
no. 26
|
72557260
NEUROSCIENCE
amygdala) (1416, 41). The avian subpallium (formed by the
striatum, pallidum, and septum), diencephalon, tectum, and brain-
stem were pooled and compared with the same regions of
mammalian brains that are referred to as the rest of brain.The
cerebellum is directly compared between the two clades. The
results of our study reveal that avian brains contain many more
pallial neurons than equivalently sized mammalian brains.
Results
Total Numbers of Neurons. We found that the bird brains have
more neurons than mammalian brains and even primate brains
of similar mass (Fig. 1 Aand B), and have very high neuronal
densities (Fig. 2 Band C). Among the songbirds studied, weighing
between 4.5 and 1,070 g, brain mass ranges from 0.36 to 14.13 g,
and total numbers of neurons in the brain from 136 million to 2.17
billion (Fig. S3 and Table S1; for complete data see Datasets S1
and S2). In the parrots studied, body mass ranges between 23 and
1,008 g, brain mass from 1.15 to 20.73 g, and numbers of brain
neurons from 227 million to 3.14 billion. Interestingly, the re-
lationship between brain mass and the number of brain neurons
can be described by similar power functions in these two bird
# non-neuronal cells
1
10
Brain mass (g)
10
2
10
3
GG
CL
TA
DN
10
8
10
9
10
10
# neurons
10
11
1
10
Brain mass (g)
10
2
10
3
DN
CL
GG TA
BC
DE
10
8
10
9
DN
GG
CL
TA
Body mass (g)
10
10
10
11
# neurons
CL
GG
DN
TA
Body mass (g)
10 10
2
10
3
10
4
10
5
10
6
1
10
Brain mass (g)
10
2
10
3
Mouse
Rat
Marmoset
Galago
1.86 g
8.36 g
10.1 g
0.42 g
1.80 g
7.78 g
10.2 g
71
483 200
1,509
2,122
636
936
x106x106
0.36 g
A
164
Goldcrest
Starling
Rook
Sulphur-crested Cockatoo
Songbirds
Parrots
Primates
Artiodactyls
Rodents
Other birds
Corvid songbirds
Non-corvid songbirds
Parrots
Primates
Artiodactyls
Rodents
Other birds
10
8
10
9
10
10
10
11
10 10
2
10
3
10
4
10
5
10
6
Birds Mammals
Fig. 1. Cellular scaling rules for brains of songbirds and parrots compared with
those for mammals. (A) Avian and mammalian brains depicted at the same scale.
Numbers under each brain represent brain mass (in grams) and total number of
brain neurons (in millions). Notice that brains of songbirds (goldcrest, starling, and
rook) and parrots (cockatoo) contain more than twice as many neurons as rodent
(mouse and rat) and primate (marmoset and galago) brains of similar size. (Scale
bar: 10 mm.) (B) Brain mass plotted as a function of total number of neurons. Note
that allometric lines for songbirds (green line) and parrots (red line) do not differ
from each other, but they do differ from allometric lines for mammals (for sta-
tistics, see SI Results). (C) Brain mass plotted as a function of total number of
nonneuronal cells. (D) Brain mass plotted as a function of body mass. (E)Total
number of brain neurons plotted as a function of body mass. Allometric lines for
the taxa examined are significantly different (for statistics, see SI Results). Each
point represents the average values for one species. Data points representing
noncorvid songbirds are light green, and data points representing corvid songbirds
are dark green. The fitted lines represent reduced major axis (RMA) regressions
and are shown only for correlations that are significant [coefficient of determi-
nation (r
2
) ranges between 0.831 and 0.997; P0.021 in all cases]. Because non-
neuronal scaling rules are very similar across the clades analyzed, the regression
lines are omitted in C. Data for mammals are from published reports (for details,
see Methods). CL, pigeon (Columba livia); DN, emu (Dromaius novaehollandiae);
GG, red junglefowl (Gallus gallus); TA, barn owl (Tyto alba).
110
Brain mass (g)
10
6
10
5
10
4
Neuronal density (N/mg)
110
Brain mass (g)
10
6
10
5
10
4
Non-neuronal density (NN/mg)
110
Brain mass (g)
10
6
10
5
10
4
Non-neuronal density (NN/mg)
CL GG TA
DN
110
Brain mass (g)
10
6
10
5
10
4
Neuronal density (N/mg)
CL
GG TA
DN
CB
ED
telencephalon
cerebellum
diencephalon
tectum
brainstem
Parrots Songbirds
and other birds
Color code:
Telencephalon
Cerebellum
Brainstem
Tec tu m
A
Diencephalon
Fig. 2. Cellular densities in avian brains. (A) Lateral view of the starling brain
showing the brain regions analyzed (for details, see SI Methods and Fig. S2).
Neuronal (Band C) and nonneuronal cell density (Dand E) plotted as a
function of brain mass. Data points representing noncorvid songbirds are light
green,anddatapointsrepresentingcorvidsongbirdsaredarkgreen.All
graphs are plotted using the same y-axis scale for comparison. Note that
neuronal density varies greatly among principal brain divisions and decreases
significantly with increasing brain mass in all divisions but the telencephalon,
whereas nonneuronal cell density is similar across brain divisions and species,
but lower in the telencephalon (for statistics, see SI Results). The fitted lines
represent RMA regressions and are shown only for correlations that are sig-
nificant (r
2
ranges between 0.410 and 0.962; P0.030 in all cases).
7256
|
www.pnas.org/cgi/doi/10.1073/pnas.1517131113 Olkowicz et al.
groups (Table S2). Thus, songbirds and parrots with similar brain
masses also have similar total numbers of brain neurons, as shown
in Fig. 1B. Because the scaling exponents are significantly higher
than 1.0 in both groups, any gain in number of brain neurons is
accompanied by an even more pronounced gain of mass: a 10-
fold increase in the number of neurons results in a 16.9- and
14.0-fold larger brain in songbirds and parrots, respectively. With
their higher neuronal densities (Fig. 3 AC), songbird and parrot
brains accommodate about twice as many neurons as primate
brains of the same mass and two to four times more neurons than
rodent brains of equivalent mass (Fig. 1B). Songbirds and parrots
also show a large brain mass for their body mass compared with
nonprimate mammals (Fig. S4 Aand B). Consequently, they
have many more neurons than a nonprimate mammal of the
same body size (Fig. 1E). For instance, the goldcrests body mass
is 9-fold smaller than the mouse, but its brain has 2.3-fold
more neurons. Large corvids and parrots possess the largest
avian brains, harboring the highest absolute numbers of neurons
(Fig. 1 Dand Eand Fig. S4C). Their total numbers of neurons
are comparable to those of small monkeys or much larger un-
gulates (Fig. S5).
Relative Distribution of Mass and Neurons. The bird/mammal com-
parison becomes even more striking when the relative distribu-
tion of neurons among the major brain components is taken into
consideration. In the birds examined, the telencephalon mass
fraction increases with brain size at the expense of all other brain
components, ranging from 63% to 80% in songbirds, and from
71% to 85% in parrots (Fig. 4 Aand Band Table S3); the rel-
ative proportion of the telencephalon resembles that reported
for primates (42) (primates, 74 ±5%; songbirds, 72 ±6%; parrots,
78 ±5%). The cerebellar mass fraction decreases from 11% to
8% in songbirds, and from 11% to 6% in parrots. Besides this,
telencephalon mass scales approximately isometrically with the
number of neurons, whereas all other brain components hyper-
scale in mass as they gain neurons (Table S2), because neuronal
densities decrease and average neuronal sizes increase signifi-
cantly as brains get larger within all brain parts but the telen-
cephalon (Fig. 2 Band C). Thus, in contrast to mammals, larger
brains of songbirds and parrots contain increasing proportions of
neurons in the telencephalon, and correspondingly decreasing
proportions of brain neurons in the cerebellum and other brain
regions (Fig. 4 Cand D). Neuronal densities in the avian pallium
exceed those observed in the primate pallium by a factor of 34
(Fig. 3A). Hence, the telencephalon houses 3862% of all brain
neurons in songbirds and 5378% in parrots (Fig. 4C); the pallium
houses 3355% in songbirds and 4661% in parrots (Fig. 3Dand
Table S4). This markedly contrasts with the situation found in
mammals, in which the pallium accounts for most of total brain
volume, but the cerebellum houses a large majority of brain neu-
rons (3239) (Fig. 3 DF). Notably, the human pallium contains a
mere 19% of brain neurons but represents 82% of brain mass (38).
Thus, when avian and mammalian brains of equivalent size are
compared, avian pallial neurons greatly outnumber those observed
in the mammalian pallium (Fig. 3Gand Fig. S5). For instance, the
goldcrest has 64 million pallial neurons, almost five times more
than the mouse pallium. The raven or the kea have 1.2 billion
pallial neurons, more than in the pallium of a capuchin monkey,
and the blue-and-yellow macaw has 1.9 billion pallial neurons,
more than in the pallium of a rhesus monkey.
Subpallium. Although once believed to constitute almost the
entire avian telencephalon (14), the subpallium (basal ganglia
homolog) accounts only for 1022% of total telencephalon
volume in songbirds and for 1518% in parrots, and houses
only 916% of telencephalic neurons in songbirds and 1424%
in parrots (Tables S3 and S4). In songbirds, both the relative
mass of the subpallium and the fraction of telencephalic
Rest of brain mass (g)
10
5
10
4
Neuronal density (N/mg)
11010
2
Cerebellum mass (g)
0.1
10
5
10
6
Neuronal density (N/mg)
10 10
2
10
3
Pallium/Cerebral cortex mass (g)
1
10
5
10
4
Neuronal density (N/mg)
TA
GG
CL
Songbirds
Parrots
Primates
Artiodactyls
Rodents
DN
ABC
70
5
Brain mass (g)
TA
DN
30
10
CL
GG
% of brain neurons
Brain mass (g)
TA
50
90
20
DN
CL
GG
% of brain neurons
Brain mass (g)
TA
10
1
30
DN
CL
GG
% of brain neurons
DEF
70
50
Pallium Cerebellum Rest of brain
x106
2.85 g 10.62 g 10.20 g 39.18 g 14.38 g 69.83 g
Owl Monkey Capuchin Monkey Macaque monkey
529 442 1,204 1,140 1,917 1,710
G
# of neurons
GG DN
TA
CL CL
TA
GG
DN
10 10
2
10
3
1
10 10
2
10
3
110 10
2
10
3
1
10 10
2
10
3
1
Eurasian Jay Raven Blue-and-yellow Macaw
Fig. 3. Neuronal densities and relative distribution
of neurons in birds and mammals. (AC) Neuronal
densities in the pallium (A), cerebellum (B), and rest
of the brain (C). Note that neuronal densities are
higher in parrots and songbirds than in mammals (for
statistics, see SI Results). (DF) Average proportions of
neurons contained in the pallium (D), cerebellum (E),
and rest of the brain (F). Note that increasing pro-
portions of brain neurons in the rest of the brain in
parrots are attributable specifically to increasing
numbers of neurons in the subpallium (Fig. 5). Data
points representing noncorvid songbirds are light
green, and data points representing corvid songbirds
are dark green. The fitted lines represent RMA re-
gressions and are shown only for correlations that are
significant (r
2
ranges between 0.389 and 0.956; P
0.033 in all cases). (G) Brains of corvids (jay and raven),
parrots (macaw), and primates (monkeys) are drawn at
the same scale. Numbers under each brain represent
mass of the pallium (in grams) and total numbers of
pallial/cortical neurons (in millions). Circular graphs
show proportions of neurons contained in the pal-
lium (green), cerebellum (red), and rest of the brain
(yellow). Notice that brains of these highly intelligent
birds harbor absolute numbers of neurons that are
comparable, or even larger than those of primates
with much larger brains. (Scale bar: 10 mm.) Data for
mammals are from published reports (for details, see
Methods). CL, pigeon; DN, emu; GG, red junglefowl;
TA, barn owl.
Olkowicz et al. PNAS
|
June 28, 2016
|
vol. 113
|
no. 26
|
7257
NEUROSCIENCE
neurons contained within it decrease with increasing telenceph-
alon size (Fig. 5 Band C). In parrots, in contrast, the relative
mass remains constant and neuronal fraction increases with
telencephalon size. Therefore, large-brained parrots have a
relatively larger subpallium within the telencephalon that ac-
commodates relatively more telencephalic neurons than that
of large-brained songbirds (Fig. 5 BD), implying that parrots
have evolved a specific, previously unrecognized cerebrotype (43)
distinguished by a higher number of neurons allocated to the
subpallium. Because subpallial structures play an important role
in sensory and motor learning and execution of motor behavior
(15, 44), we suggest that the relatively enlarged subpallium in
large parrots is likely associated with their greater learning skills,
including vocal learning, and enhanced foot and beak dexterity
(5, 6, 13, 45).
Nonneuronal Scaling Rules. Although neuronal scaling rules for avian
brains differ from those for mammalian brains (Fig. 1B), non-
neuronal scaling rules are shared between the two vertebrate classes
(Fig. 1Cand Table S2). In line with data from all mammals analyzed
so far (3239), the densities of nonneuronal (glial and endothelial)
cells remain similar across bird species in all brain structures, except
for the telencephalon, where nonneuronal cell density appears to be
distinctively lower (Fig. 2 Dand E).Thelattermaybeaspecific
avian feature, as it has not been observed in mammals (31).
Glia/Neuron Ratio. Neurons outnumber nonneuronal cells in both
bird groups examined (Fig. S6Aand Table S5). The proportion of
nonneuronal cells in the brain ranges between 21% and 37% in
songbirds and from 31% to 41% in parrots. Hence, the maximal
glia/neuron ratio (if all nonneuronal cells were glial cells) for the
whole brain ranges from 0.27 to 0.59 in songbirds and from 0.44 to
0.69 in parrots. Like in mammals (3239, 46), the proportion of
nonneuronal cells is very small in the cerebellum, varying between
12% and 19% in songbirds and between 14% and 19% in parrots,
but, in contrast to mammals, nonneuronal cells also constitute a
minor cellular fraction in the telencephalon, representing 2140%
of cells in songbirds and 3143% of cells in parrots (Fig. S6B).
Nonneuronal cells predominate in the remaining brain regions
analyzed, representing in songbirds and parrots, respectively, 60
90% and 7994% of all cells in the diencephalon, 2870% and
5271% of all cells in the tectum, and 7695% and 8595% of all
cells in the brainstem (Fig. S6B). The fact that neurons constitute
an extremely small cellular fraction in the diencephalon of many
avian species is an unexpected finding. Given that nonneuronal
cell densities are similar to those found in most other brain di-
visions investigated (Fig. 2 Dand E), this is unlikely to be due to
a technical error. The numeric preponderance of neurons over
nonneuronal cells in the bird brain as a whole is therefore due to
the disproportionately large numbers of neurons in the telenceph-
alon and cerebellum.
Corvid Brain as a Scaled-Up Songbird Brain. When considering the
numbers of neurons and nonneuronal cells and their allocations
to the major brain divisions, the same scaling rules apply to the
brains of corvids and noncorvid songbirds (Figs. 15 and Table
S2). Thus, it is not cellular composition but encephalization that
sets corvids apart from other songbirds. Technically, residual
brain mass calculated from regressions for all songbirds is sig-
nificantly larger in corvids than in noncorvid songbirds [species
examined in this study: t
(2,11)
=2.542, P=0.03, Fig. 1D; species
collated from literature: t
(2,848)
=7.55, P<10
6
,Fig. S4C].
Because corvid brains tend to be larger than brains of noncorvid
songbirds for any given body size (Fig. 1Dand Fig. S4C), corvids
have larger total numbers of neurons than noncorvid songbirds of
the same body size (Fig. 1E). We suggest that corvid brains are
scaled-up songbird brains, just as humans brains are to brains of
nonhuman primates (38, 47), and that large absolute numbers of
neurons endow corvids with superior cognitive abilities.
Comparison with Other Birds. The similarity of neuronal scaling
rules between songbirds and parrots is not too surprising, con-
sidering their close phylogenetic relationship (4851). The ex-
amination of outgroup taxa, however, suggests that, as in mammals
(31), different neuronal scaling rules apply to various bird lineages.
The closest relative to songbirds and parrots of the species sam-
pled, the barn owl (Fig. S1)(4851) resembles songbirds and
parrots in terms of encephalization (Fig. 1D), relative telencephalon
size (Fig. 4A), and neuronal densities in the telencephalon and di-
encephalon (Fig. 2C), but has a proportionally smaller subpallium
(Fig. 5B) and lower neuronal densities in the tectum and cerebel-
lum (Fig. 2C). The emu, the red junglefowl, and the pigeon, all
species representing more basal bird lineages (Fig. S1), share
lower degree of encephalization (Fig. 1D), a proportionally smaller
telencephalon (Fig. 4A), small telencephalic and dominant cer-
ebellar neuronal fractions (Fig. 4C), generally lower neuronal
densities (Fig. 2C), and larger glia/neuron ratios (Fig. S6).
% of brain non-neurons
70
50
30
10
10 20
Brain mass (g)
% of brain mass
80
40
10
10 20
Brain mass (g)
% of brain non-neurons
14
8
4
10
10 20
Brain mass (g)
% of brain mass
6
2
10
14
80
70
60
50
40
30
10 20
Brain mass (g)
10 20
Brain mass (g)
10
8
6
4
2
% of brain neurons
% of brain neurons
CD
GG
CL
TA
DN
10 20
Brain mass (g)
12
8
4
12
6
TA
TA
DN
DN
CL
CL
CL
GG
GG
GG
TA
DN
DN
TA
GG
CL
CL
GG
TA
DN
CL
CL
GG
GG
TA
TA
DN
DN
DN
TA
GG
CL
E
AB
F
telencephalon
cerebellum
diencephalon
tectum
brainstem
Fig. 4. Relative distribution of mass and cells in avian brains. Average per-
centages of mass (Aand B), number of neurons (Cand D), and number of
nonneuronal cells (Eand F) contained in the principal brain divisions relative to
the whole brain in each species, plotted against brain mass. Data points rep-
resenting noncorvid songbirds are light green, and data points representing
corvid songbirds are dark green. The fitted lines represent RMA regressions
and are shown only for correlations that are significant (r
2
ranges between
0.389 and 0.956; P0.023 in all cases). Note that both telencephalon mass
fraction and proportions of neuronal and nonneuronal cells contained in the
telencephalon increase with brain size.
7258
|
www.pnas.org/cgi/doi/10.1073/pnas.1517131113 Olkowicz et al.
Therefore, their brains harbor much smaller absolute numbers of
neurons than brains of equivalently sized songbirds or parrots.
For instance, although a red junglefowl is 50-fold heavier than
a great tit, both birds have approximately the same number of
brain neurons (Fig. 1Eand Fig. S3). Remarkably, even in these
basal birds, neuronal densities in the pallium are still comparable
to those observed in the primate cortex (Fig. 3A). Thus, high
neuronal density in the telencephalon appears characteristic of
all birds. This means that neuronal densities in the primate
pallium are matched by those of chicken and emu, but surpassed
by those of songbirds and parrots.
Discussion
Assuming that brains of parrots and songbirds have diverged
from the presumptive ancestral avian pattern found in all rep-
resentatives of basal bird lineages examined and characterized by
a mammal-like numerical preponderance of cerebellar neurons,
we suggest that birds generally have higher neuronal densities
than mammals, and further that parrots and songbirds have ac-
quired an expanded telencephalon with increased neuronal den-
sities. Two proximate, synergistic mechanisms likely contributed to
this evolutionary process. First, just like the expansion of neocortex
in primates (52), the expansion of the telencephalon in parrots and
songbirds is associated with delayed and protracted neurogenesis,
an expanded subventricular zone, and delayed neuronal matura-
tion (5355). It has been suggested that extensive posthatching
neurogenesis and brain maturation promote learning from con-
specifics and may have facilitated the emergence of specialized
circuits that mediate vocal learning and possibly also other flexible
and innovative behaviors (56). Second, analyses of brain gene ex-
pression profiles strongly suggest that songbirds and parrots in-
dependently evolved vocal learning pathways by duplication of
preexisting, surrounding motor circuits (57, 58). Intriguingly, par-
rot pallial song nuclei underwent a further duplication event to
evolve a unique additional circuit, the so-called shell song system,
which seems to be particularly well developed in large-brained
parrots (45). What ultimate mechanisms drive the evolution of
the enlarged, neuron-rich telencephalon, which sets parrots and
songbirds apart from the more basal birds we examined, remains
poorly understood. We suggest that this expansion has been due
to simultaneous selective pressures on cognitive enhancement
and an evolutionary constraint on brain size, which may stem
from the constraints on body size imposed by active flight.
Altriciality and the extended parental care that has developed in
avian ancestors simultaneously relaxed constraints on the dura-
tion of ontogenesis, a precondition for telencephalic expansion
by the mechanisms described above (56). Moreover, a short neck
relative to many other bird lineages may have reduced biophys-
ical constraints on head size (cf. ref. 59).
Our finding of greater than primate-like numbers of neurons
in the pallium of parrots and songbirds suggests that the large
absolute numbers of telencephalic neurons in these two clades
provide a means of increasing computational capacity, support-
ing their advanced behavioral and cognitive complexity, despite
their physically smaller brains. Moreover, a short interneuronal
distance, the corollary of the extremely high packing densities of
their telencephalic neurons, likely results in a high speed of in-
formation processing, which may further enhance cognitive
abilities of these birds. Thus, the nuclear architecture of the
avian brain appears to exhibit more efficient packing of neurons
and their interconnections than the layered architecture of the
mammalian neocortex.
Further comparative studies on additional species are required
to determine whether the high neuronal densities and preferential
allocation of neurons to the telencephalon represent unique fea-
tures of songbirds, parrots, and perhaps some other clades like
owls, or have evolved multiple times independently in large-brained
birds. More detailed quantitative studies should assess the distri-
bution of neurons among various telencephalic regions involved in
specific circuits subserving specific functions. The results, combined
with behavioral studies, will enable us to determine the causal re-
lationships between neuronal numbers and densities and percep-
tual, cognitive, and executive/motor abilities, and greatly advance
our understanding of potential mechanisms linking neuronal den-
sity with information-processing capacity.
Methods
Experimental procedures were all approved by the Institutional Animal Care and
Use Committee at Charles University in Prague. Altogether, 73 birds belonging
to 28 species were used in this study (Table S1). Animals were killed by an
overdose of halothane and perfused with 4% (wt/wt) paraformaldehyde. Brains
were removed, postfixed for an additional 721 d, and dissected into the ce-
rebral hemispheres, cerebellum, diencephalon, tectum, and brainstem. In one
individual per species, one hemisphere was dissected into the pallium and the
subpallium. In these brain components, the total numbers of cells, neurons, and
nonneuronal cells were estimated following the procedure of isotropic frac-
tionation described earlier (40). The reduced major axis regressions to power
functions were calculated to describe how structure mass, numbers of cells, and
densities are interrelated across species. Analysis of covariance was used to
compare scaling among groups (taxonomic orders or brain regions). To
compare relative brain size between corvid and noncorvid songbirds, we
computed ttest on the residuals of a loglog regression of brain mass
against bo dy mass (residual brain mass, hereafter). For the comparison with
cellular scaling rules reported previously for mammals, the reduced major
axis re gressions we re calculated from quantitative data published for primates
(33, 37, 38), rodents excluding the naked mole-rat (32, 39), and artiodactyls
(36). In addition, the published quantitative data for Eulipothyphla (34)
and Afroth eria (35) were used for comparison in Fig. S5. Further details are
provided in Supporting Information.
10
20
110
Telencephalon (g)
% of telencephalon mass
110
Telencephalon (g)
10
20
% of telencephalon
neurons
BCD
10
8
10
9
10
8
10
7
DN
TA
GG
CL
Pallial neurons
Subpallial neurons
CL
GG
TA
DN
Songbirds
Parrots
7
CL
GG
DN
TA
5
A
Pallium
Subpallium
Fig. 5. Subpallium in avian telencephalon. (A) Diagram of sagittal section through the zebra finch brain showing relative position and size of the pallium and
subpallium. (Band C) Average percentages of mass (B), number of neurons (C) contained in the subpallium relative to the whole telencephalon in each
species, plotted against telencephalon mass. (D) Relationship between numbers of subpallial and pallial neurons. Note that, in parrots, the number of neurons
in the subpallium increases faster than in the pallium (scaling exponent =1.19 ±0.13), whereas an opposite trend is observed in songbirds (scaling exponent =
0.91 ±0.1). The fitted lines represent RMA regressions and are shown only for correlations that are significant (r
2
ranges between 0.379 and 0.981; P0.025 in
all cases). Songbirds shown in green (data points representing noncorvids are light green, and data points representing corvids are dark green), parrots in red,
and other birds in black. CL, pigeon; DN, emu; GG, red junglefowl; TA, barn owl.
Olkowicz et al. PNAS
|
June 28, 2016
|
vol. 113
|
no. 26
|
7259
NEUROSCIENCE
ACKNOWLEDGMENTS. We thank O. Güntürkün, H. J. ten Donkelaar, T. Bugnyar,
N. C. Bennett, M. Prevorovsky, and K. Kverkova for reading of the manuscript and
discussions; V. Miller and T. Hajek for logistic support; Y. Zhang and V. Blahova for
their assistance with experiments; Z. Pavelkova and B. Strakova for collecting data
on avian and mammalian brain and body mass from the literature; L. Kratochvil
for methodological advice; P. Benda and J. Mateju for help with acquiring animal
experiment approvals; R. Vodicka for assistance with anaesthesia of the emu; and
P. Benda and J. Mlikovsky for providing access to dissection facilities of the
National Museum of the Czech Republic. This project was funded by Czech Science
Foundation (14-21758S) (to P.N.), Grant Agency of Charles University (851613) (to
M.K.), Specific Research Grant from Charles University in Prague (SVV 260 313/
2016) (to M.K.), the European Social Fund and the state budget of the Czech
Republic (CZ.1.07/2.3.00/30.0022) (to S.O.)., the Brazilian National Council for
Scientific and Technological Development (to S.H.-H.), Fundação de Amparo à
Pesquisa do Estado do Rio de Janeiro (to S.H.-H.), and the James S. McDonnell
Foundation (to S.H.-H.).
1. Emery NJ (2006) Cognitive ornithology: The evolution of avian intelligence. Philos
Trans R Soc Lond B Biol Sci 361(1465):2343.
2. Emery NJ, Clayton NS (2004) The mentality of crows: Convergent evolution of in-
telligence in corvids and apes. Science 306(5703):19031907.
3. Clayton NS, Emery NJ (2015) Avian models for human cognitive neuroscience: A
proposal. Neuron 86(6):13301342.
4. Weir AA, Chappell J, Kacelnik A (2002) Shaping of hooks in New Caledonian crows.
Science 297(5583):981.
5. Auersperg AMI, Szabo B, von Bayern AMP, Kacelnik A (2012) Spontaneous innovation
in tool manufacture and use in a Goffins cockatoo. Curr Biol 22(21):R903R904.
6. Huber L, Gajdon GK (2006) Technical intelligence in animals: The kea model. Anim
Cogn 9(4):295305.
7. Taylor AH, Miller R, Gray RD (2012) New Caledonian crows reason about hidden
causal agents. Proc Natl Acad Sci USA 109(40):1638916391.
8. Prior H, Schwarz A, Güntürkün O (2008) Mirror-induced behavior in the magpie (Pica
pica): Evidence of self-recognition. PLoS Biol 6(8):e202.
9. Raby CR, Alexis DM, Dickinson A, Clayton NS (2007) Planning for the future by
western scrub-jays. Nature 445(7130):919921.
10. Emery NJ, Clayton NS (2001) Effects of experience and social context on prospective
caching strategies by scrub jays. Nature 414(6862):443446.
11. Bugnyar T, Schwab C, Schloegl C, Kotrschal K, Heinrich B (2007) Ravens judge com-
petitors through experience with play caching. Curr Biol 17(20):18041808.
12. Jarvis ED (2004) Learned birdsong and the neurobiology of human language. Ann N Y
Acad Sci 1016:749777.
13. Pepperberg IM (1999) The Alex Studies: Cognitive and Communicative Abilities of
Grey Parrots (Harvard Univ Press, Cambridge, MA).
14. Jarvis ED, et al.; Avian Brain Nomenclature Consortium (2005) Avian brains and a new
understanding of vertebrate brain evolution. Nat Rev Neurosci 6(2):151159.
15. Jarvis ED, et al. (2013) Global view of the functional molecular organization of the
avian cerebrum: Mirror images and functional columns. J Comp Neurol 521(16):
36143665.
16. Pfenning AR, et al. (2014) Convergent transcriptional specializations in the brains of
humans and song-learning birds. Science 346(6215):1256846.
17. Shanahan M, Bingman VP, Shimizu T, Wild M, Güntürkün O (2013) Large-scale net-
work organization in the avian forebrain: A connectivity matrix and theoretical
analysis. Front Comput Neurosci 7:89.
18. Calabrese A, Woolley SM (2015) Coding principles of the canonical cortical microcir-
cuit in the avian brain. Proc Natl Acad Sci USA 112(11):35173522.
19. Kirsch JA, Güntürkün O, Rose J (2008) Insight without cortex: Lessons from the avian
brain. Conscious Cogn 17(2):475483.
20. Güntürkün O (2005) The avian prefron tal cortexand cog nition. Curr Opin Neurobiol
15(6):686693.
21. Veit L, Nieder A (2013) Abstract rule neurons in the endbrain support intelligent
behaviour in corvid songbirds. Nat Commun 4:2878.
22. Striedter GF (2005) Principles of Brain Evolution (Sinauer Associates, Sunderland, MA).
23. Roth G, Dicke U (2005) Evolution of the brain and intelligence. Trends Cogn Sci 9(5):
250257.
24. Herculano-Houzel S (2011) Brains matter, bodies maybe not: The case for examining
neuron numbers irrespective of body size. Ann N Y Acad Sci 1225:191199.
25. Dicke U, Roth G (2016) Neuronal factors determining high intelligence. Philos Trans R
Soc Lond B Biol Sci 371(1685):20150180.
26. Deaner RO, Isler K, Burkart J, van Schaik C (2007) Overall brain size, and not en-
cephalization quotient, best predicts cognitive ability across non-human primates.
Brain Behav Evol 70(2):115124.
27. MacLean EL, et al. (2014) The evolution of self-control. Proc Natl Acad Sci USA 111(20):
E2140E2148.
28. Stevens JR (2014) Evolutionary pressures on primate intertemporal choice. Proc Biol
Sci 281(1786):20140499.
29. Mlikovsky J (2003) Brain size and forearmen magnum area in crows and allies (Aves:
Corvidae). Acta Soc Zool Bohem 67(1-4):203211.
30. Iwaniuk AN, Dean KM, Nelson JE (2005) Interspecific allometry of the brain and brain
regions in parrots (Psittaciformes): Comparisons with other birds and primates. Brain
Behav Evol 65(1):4059.
31. Herculano-Houzel S, Manger PR, Kaas JH (2014) Brain scaling in mammalian evolution
as a consequence of concerted and mosaic changes in numbers of neurons and av-
erage neuronal cell size. Front Neuroanat 8:77.
32. Herculano-Houzel S, Mota B, Lent R (2006) Cellular scaling rules for rodent brains.
Proc Natl Acad Sci USA 103(32):1213812143.
33. Herculano-Houzel S, Collins CE, Wong P, Kaas JH (2007) Cellular scaling rules for
primate brains. Proc Natl Acad Sci USA 104(9):35623567.
34. Sarko DK, Catania KC, Leitch DB, Kaas JH, Herculano-Houzel S (2009) Cellular scaling
rules of insectivore brains. Front Neuroanat 3:8.
35. Neves K, et al. (2014) Cellular scaling rules for the brain of afrotherians. Front
Neuroanat 8:5.
36. Kazu RS, Maldonado J, Mota B, Manger PR, Herculano-Houzel S (2014) Cellular scaling
rules for the brain of Artiodactyla include a highly folded cortex with few neurons.
Front Neuroanat 8:128.
37. Gabi M, et al. (2010) Cellular scaling rules for the brains of an extended number of
primate species. Brain Behav Evol 76(1):3244.
38. Azevedo FA, et al. (2009) Equal numbers of neuronal and nonneuronal cells make the
human brain an isometrically scaled-up primate brain. J Comp Neurol 513(5):532541.
39. Herculano-Houzel S, et al. (2011) Updated neuronal scaling rules for the brains of
Glires (rodents/lagomorphs). Brain Behav Evol 78(4):302314.
40. Herculano-Houzel S, Lent R (2005) Isotropic fractionator: A simple, rapid method for
the quantification of total cell and neuron numbers in the brain. J Neurosci 25(10):
25182521.
41. Reiner A, et al.; Avian Brain Nomenclature Forum (2004) Revised nomenclature for
avian telencephalon and some related brainstem nuclei. J Comp Neurol 473(3):
377414.
42. Clark DA, Mitra PP, Wang SSH (2001) Scalable architecture in mammalian brains.
Nature 411(6834):189193.
43. Iwaniuk AN, Hurd PL (2005) The evolution of cerebrotypes in birds. Brain Behav Evol
65(4):215230.
44. Reiner A, Medina L, Veenman CL (1998) Structural and functional evolution of the
basal ganglia in vertebrates. Brain Res Brain Res Rev 28(3):235285.
45. Chakraborty M, et al. (2015) Core and shell song systems unique to the parrot brain.
PLoS One 10(6):e0118496.
46. Herculano-Houzel S (2014) The glia/neuron ratio: How it varies uniformly across brain
structures and species and what that means for brain physiology and evolution. Glia
62(9):13771391.
47. Herculano-Houzel S (2009) The human brain in numbers: A linearly scaled-up primate
brain. Front Hum Neurosci 3:31.
48. Hackett SJ, et al. (2008) A phylogenomic study of birds reveals their evolutionary
history. Science 320(5884):17631768.
49. Jetz W, Thomas GH, Joy JB, Hartmann K, Mooers AO (2012) The global diversity of
birds in space and time. Nature 491(7424):444448.
50. Jarvis ED, et al. (2014) Whole-genome analyses resolve early branches in the tree of
life of modern birds. Science 346(6215):13201331.
51. Prum RO, et al. (2015) A comprehensive phylogeny of birds (Aves) using targeted
next-generation DNA sequencing. Nature 526(7574):569573.
52. Lui JH, Hansen DV, Kriegstein AR (2011) Development and evolution of the human
neocortex. Cell 146(1):1836.
53. Charvet CJ, Striedter GF (2008) Developmental species differences in brain cell cycle
rates between northern bobwhite quail (Colinus virginianus) and parakeets (Melop-
sittacus undulatus): Implications for mosaic brain evolution. Brain Behav Evol 72(4):
295306.
54. Charvet CJ, Striedter GF (2009) Developmental origins of mosaic brain evolution:
Morphometric analysis of the developing zebra finch brain. J Comp Neurol 514(2):
203213.
55. Charvet CJ, Striedter GF (2011) Causes and consequences of expanded subventricular
zones. Eur J Neurosci 34(6):988993.
56. Charvet CJ, Striedter GF (2011) Developmental modes and developmental mecha-
nisms can channel brain evolution. Front Neuroanat 5:4.
57. Feenders G, et al. (2008) Molecular mapping of movement-associated areas in the
avian brain: A motor theory for vocal learning origin. PLoS One 3(3):e1768.
58. Chakraborty M, Jarvis ED (2015) Brain evolution by brain pathway duplication. Philos
Trans R Soc Lond B Biol Sci 370(1684):20150056.
59. Taylor MP, Wedel MJ (2013) Why sauropods had long necks; and why giraffes have
short necks. PeerJ 1:e36.
60. Peterson AT, Brisbin IL (1998) Genetic endangerment of wild red junglefowl Gallus
gallus?Bird Conserv Int 8(04):387394.
61. Mullen RJ, Buck CR, Smith AM (1992) NeuN, a neuronal specific nuclear protein in
vertebrates. Development 116(1):201211.
62. Mezey S, et al. (2012) Postnatal changes in the distribution and density of neuronal
nuclei and doublecortin antigens in domestic chicks (Gallus domesticus). J Comp
Neurol 520(1):100116.
63. Bohonak AJ, van der Linde K (2004) RMA: Software for reduced major axis regression,
Java version. Available at www.kimvdlinde.com/professional/rma.html. Accessed
January 5, 2015.
64. Nixdorf-Bergweiler B, Bischof H-J (2007) A Stereotaxic Atlas of the Brain of the Zebra
Finch, Taeniopygia guttata, with Special Emphasis on Telencephalic Visual and Song
System Nuclei in Transverse and Sagittal Sections (National Library of Medicine, Na-
tional Center for Biotechnology Information, Bethesda, MD). Available at www.ncbi.
nlm.nih.gov/books/NBK2356/. Accessed March 3, 2016.
7260
|
www.pnas.org/cgi/doi/10.1073/pnas.1517131113 Olkowicz et al.
Supporting Information
Olkowicz et al. 10.1073/pnas.1517131113
SI Methods
Animals. Three individuals per species were collected with some
exceptions for large parrots, two songbird species, and the emu, in
which only one or two birds were examined. The following species
were purchased from local breeders: all species of parrots, zebra
finch, azure-winged magpie, common hill myna, raven, emu, red
junglefowl, and barn owl. According to some authors, the genetic
integrity of the red junglefowl Gallus gallus is endangered due to
hybridization with domestic or feral chickens at the edge of
fragmented forests (60). Although we thus cannot exclude ad-
mixture of genes from domestic or feral chicken, the red junglefowl
used in this study appeared to have a pure wild phenotype. The
remaining birds were wild-caught in Czech Republic (Permission
No. 00212/CS/2013 and 446/2013). All birds were sexually ma-
ture or at least had adult-like size and plumage coloration. We
determined the sex of all animals upon dissection and found that
we had included both males and females in the analysis. The
sample sizes were too small to analyze sex differences.
Animals were killed by an overdose of halothane. They were
weighed and immediately perfused transcardially with warmed
PBS containing 0.1% heparin followed by cold phosphate-buff-
ered 4% (wt/wt) paraformaldehyde solution. Skulls were partially
opened and postfixed for 3060 min, after which brains were
dissected and weighed. Brains were postfixed for additional
721 d and then dissected. All procedures were approved by
Institutional Animal Care and Use Committee at Charles Uni-
versity in Prague, Ministry of Culture (Permission No. 47987/
2013) and Ministry of the Environment of the Czech Republic
(Permission No. 53404/ENV/13-2299/630/13).
Dissection. Brains were dissected into distinct components using
the Olympus SZX 16 stereomicroscope. The cerebral hemi-
spheres were detached from the diencephalon by a straight cut
separating the subpallium from the thalamus. The tectum (optic
lobe) was bilaterally excised from the surface of the brainstem.
The excised parts included most of the tectal gray, optic tectum,
and torus semicircularis. Both left and right tectum were pro-
cessed together. The cerebellum was cut off at the surface of the
brainstem. Finally, the remaining structures were dissected into
diencephalon (rostral part) and brainstem (caudal part) along the
plane connecting the posterior commissure dorsally and hypo-
thalamusmesencephalon boundary ventrally. For most individ-
uals, only one cerebral hemisphere was processed, because in our
preliminary studies we detected negligible differences between
left and right hemisphere mass and cell numbers. In one indi-
vidual per species, the second hemisphere was dissected into the
pallium and the subpallium. These hemispheres were embedded
in agarose and sectioned on a vibratome at 300500 μm (depending
on size of a hemisphere) in the coronal plane. Under oblique
transmitted light at the stereomicroscope and with the use of a
microsurgical knife (Stab Knife Straight; 5.5 mm; REF 7516;
Surgical Specialties Corporation), we manually dissected the
pallium from subpallium on each section by cutting along the
pallial-subpallial lamina, as defined by Reiner et al. (41).
The subpallium included all major subpallial cell groups enu-
merated therein; the remaining parts of the telencephalon con-
stituted the pallium. The dissected structures were dried with
paper towel, weighed, incubated in 30% (wt/wt) sucrose solution
until they sank, then transferred into antifreeze (30% glycerol, 30%
ethylene glycol, 40% phosphate buffer), and frozen for further
processing.
Isotropic Fractionator. We estimated total numbers of cells, neu-
rons, and nonneuronal cells following the procedure of isotropic
fractionation described earlier (40). Briefly, each dissected brain
division was homogenized in 40 mM sodium citrate with 1%
Triton X-100 using Tenbroeck tissue grinders (Wheaton). When
turned into an isotropic suspension of isolated cell nuclei, ho-
mogenates were stained with the florescent DNA marker DAPI,
adjusted to a defined volume, and kept homogenous by agitation.
The total number of nuclei in suspension, and therefore the total
number of cells in original tissue, was estimated by determining
density of nuclei in small fractions drawn from a homogenate. At
least four 10-μL aliquots were sampled and counted using a
Neubauer improved counting chamber (BDH; Dagenham) with
an Olympus BX51 microscope equipped with epifluorescence
and appropriate filter settings (Olympus filters U-MWU2 for
DAPI and U-MWG2 for Alexa Fluor 546-conjugated secondary
antibodies); additional aliquots (typically two to five) were as-
sessed when needed to reach the coefficient of variation among
counts 0.15. Once the total cell number was known, the pro-
portion of neurons was determined by immunocytochemical
detection of neuronal nuclear marker NeuN (61). This neuron-
specific protein was detected by the mouse monoclonal antibody
anti-NeuN (clone A60; Chemicon; dilution, 1:800), which was
recently characterized by Western blotting with chick brain
samples and shown to react with a protein of the same molecular
weight as in mammals (62), indicating that it does not cross-react
with other proteins in birds. The binding sites of the primary
antibody were revealed by Alexa Fluor 546-conjugated goat anti-
mouse IgG (Life Technologies; dilution, 1:500). An electronic
hematologic counter (Alchem Grupa) was used to count simul-
taneously DAPI-labeled and NeuN-immunopositive nuclei in the
Neubauer chamber. A minimum of 500 nuclei was counted to
estimate percentage of double-labeled neuronal nuclei. Numbers
of nonneuronal cells were derived by subtraction.
Data Analysis. All analyses were performed using average values
for each species; variables were log-transformed before the sub-
sequent statistical analyses. Correlations between variables were
assessed using nonparametric Spearman rank test. If a signifi-
cance criterion of P<0.05 was reached, the reduced major axis
regressions were calculated to describe how structure mass,
numbers of cells, and densities are interrelated across species.
Analysis of covariance (ANCOVA) was used to compare scaling
among groups (taxonomic orders or brain regions). The signifi-
cant interaction between categorical and continuous predictors
in the full-factorial ANCOVA demonstrates statistically differ-
ent slopes of the regression lines among groups and precludes
the direct comparison of the magnitude of differences among
groups based just on the differences in intercepts. In these cases,
the group responsible for the significant interaction was excluded
from the ANCOVA model, and, subsequently, the effect of
categorical predictor was tested across groups with statistically
homogenous slopes, and their differences were compared based
on differences in the intercepts. The planned comparisons of
least-squares means was used to examine significant pairwise
differences. To compare relative brain size between corvid and
noncorvid songbirds, we computed ttest on the residuals of a
loglog regression of brain mass against body mass (residual
brain mass, hereafter). For the comparison with cellular scaling
rules reported previously for mammals, the reduced major axis
regressions were calculated from quantitative data published for
primates (33, 37, 38), rodents excluding the naked mole-rat (32,
Olkowicz et al. www.pnas.org/cgi/content/short/1517131113 1of13
39), and artiodactyls (36). In addition, the published quantitative
data for Eulipothyphla (34) and Afrotheria (35) were used for
comparison in Fig. S4.
The regressions were calculated using RMA for JAVA 1.21
(63); ANCOVA and ttest, using Statistica 10.0 (Stat Soft); and
all other analyses were performed in JMP 10.0 (SAS Institute).
SI Results
The results of the ANCOVA are summarized below for selected,
important comparisons among taxonomic orders and brain re-
gions. They are listed in order, in which they appear in the figures.
Ad Fig. 1. (B) Allometric lines for songbirds (green line) and
parrots (red line) do not differ from each other [full-factorial
ANCOVA, slopes: F
(1,20)
=0.537, P=0.47; intercepts: F
(1,20)
=0.580,
P=0.46], but they do differ from allometric lines for mammals
[slopes: F
(4,40)
=4.290, P=0.006; intercepts: F
(4,40)
=3.595, P=
0.014; post hoc analyses indicate that the regression line for rodents
has a different slope and that parrots and songbirds have signifi-
cantly smaller brains for a given number of neurons than primates
and artiodactyls, P<10
6
for all planned comparisons].
(E) Allometric lines for the taxa examined are significantly
different [slopes: F
(5,38)
=3.653, P=0.009; intercepts: F
(5,38)
=
2.558, P=0.043; post hoc analyses indicate that the regression
line for primates has a different slope and that parrots and
songbirds have a significantly higher number of neurons for a
given body mass than rodents and artiodactyls, P<0.001 for all
planned comparisons].
Ad Fig. 2. (Band C) Neuronal density varies significantly among
principal brain divisions in both parrots [slopes: F
(4,45)
=16.2, P<
10
6
;intercepts:F
(4,45)
=233.0, P<10
6
] and songbirds [slopes:
F
(4,55)
=14.4, P<10
6
;intercepts:F
(4,55)
=523.9, P<10
6
].
(Dand E) Comparison of the telencephalon with data pooled
for the all other structures examined indicate that nonneuronal
cell density is significantly lower in the telencephalon than in the
remaining brain divisions in both parrots [slopes: F
(1,51)
=0.00,
P=0.995; intercepts: F
(1,51)
=58.94, P<10
6
] and songbirds
[slopes: F
(1,61)
=0.0, P=0.838; intercepts: F
(1,61)
=238.0, P<10
6
].
Ad Fig. 3. (A) Pallial neuronal densities are significantly higher in
parrots and songbirds than in mammals [slopes: F
(4,41)
=5.948,
P=0.0007; intercepts: F
(4,41)
=75.688, P=<10
6
; post hoc
analyses indicate that the regression line for rodents has a dif-
ferent slope and that parrots and songbirds have significantly
higher telencephalic neuronal densities than primates and ar-
tiodactyls, P<10
6
for all planned comparisons].
(B) Cerebellar neuronal densities tend to be higher in parrots
and songbirds than in mammals [slopes: F
(4,40)
=7.84, P<10
4
;
intercepts: F
(4,40)
=24.71, P=<10
6
; post hoc analyses indicate
that the regression line for primates has a different slope and
that parrots and songbirds have significantly higher cerebellar
neuronal densities than rodents and artiodactyls, P<10
4
for all
planned comparisons].
(C) Neuronal densities in the rest of brain are significantly
higher in parrots and songbirds than in mammals [slopes: F
(4,41)
=
4.876, P=0.003; intercepts: F
(4,41)
=86.875, P=<10
6
;posthoc
analyses indicate that the regression line for parrots and for ro-
dents differ in slope from other regression lines and that song-
birds have significantly higher neuronal densities than primates
and artiodactyls, P<10
6
for all planned comparisons].
Ad Fig. 5. (C) Allometric lines for songbirds (green line) and
parrots (red line) differ significantly in slope [F
(1,20)
=17.232,
P=0.0005].
Olkowicz et al. www.pnas.org/cgi/content/short/1517131113 2of13
Starling (Sturnus vulgaris)
Hill Mynah (Gracula religiosa)
Blackbird (Turdus merula)
Goldcrest (Regulus regulus)
Zebrafinch (Taeniopygia guttata)
Great Tit (Parus major)
Blackcap (Sylvia atricapilla)
Raven (Corvus corax)
Rook (Corvus frugilegus)
Jackdaw (Corvus monedula)
Magpie (Pica pica)
Eurasian Jay (Garrulus glandarius)
Azure-Winged Magpie (Cyanopica cyanus)
Blue and Yellow Macaw (Ara ararauna)
Green-Rumped Parrotlet (Forpus passerinus)
Monk Parakeet (Myiopsitta monachus)
Grey Parrot (Psittacus erithacus)
Budgerigar (Melopsittacus undulatus)
Eastern Rosella (Platycercus eximius)
Alexandrine Parakeet (Psittacula eupatria)
Sulphur-Crested Cockatoo (Cacatua galerita)
Goffin's Cockatoo (Cacatua goffini)
Cockatiel (Nymphicus hollandicus)
Kea (Nestor notabilis)
Barn Owl (Tyto alba)
Rock Pigeon (Columba livia)
Emu (Dromaius novaehollandiae)
Red Junglefowl (Gallus gallus)
Corvids
PasseriformesPsittaciformes
Strigiformes
Columbiformes
Galliformes
Struthioniformes
Fig. S1. Phylogenetic relationships among the 28 species examined. The tree was constructed using birdtree.org/; its topology follows recent studies (4649).
Note that songbirds and parrots are sister groups and together with the distantly related barn owl belong to the clade core landbirds (Telluraves); the pigeon
represents the Columbea, a basal clade of the Neoaves; the red junglefowl represents the Galloanseres, a sister group of Neoaves and the most basal clade of
Neognathae; and the emu represents Paleognathae (tinamous and flightless ostriches), the most basal clade of extant birds (48). Also note that all passerine
birds examined were vocal learners belonging to the clade Oscines.
Olkowicz et al. www.pnas.org/cgi/content/short/1517131113 3of13
Fig. S2. Brain dissection and labeling of neurons and nonneuronal cells. (Aand B) Brain of the raven before and after the dissection. (A) Ventral side of the
brain showing approximate lines of dissection of the brainstem and tectum. (B) Brain dissected into parts used for isotropic fractionation. (C) NeuN-im-
munolabeled transverse section of the zebra finch brain depicting the line of dissection of the tectum from the rest of the mesencephalon. (DF) Dissection of
the telencephalon into pallium and subpallium. NeuN-immunolabeled transverse sections of the zebra finch brain at rostral (D), intermediate (E), and caudal
(F) telencephalic levels. Lines of dissection follow the pallial-subpallial lamina and divide the telencephalon into pallium (dorsal part) and subpallium (ventral
part). Coordinates anterior to the Y point are indicated in millimeters at Bottom Left (64). (GI) High-power micrographs showing a sample of homogenate
from the telencephalon of the Eurasian jay; dissociated nuclei stained with DAPI (G) and immunolabeled with NeuN antibody (H), dual-fluorescence merge
image (I). Note that neurons are double-labeled, whereas the nonneuronal cells are devoid of anti-NeuN immunoreactivity. [Scale bars: 10 mm (Aand B); 1 mm
(Cand F); 50 μm(I).]
Olkowicz et al. www.pnas.org/cgi/content/short/1517131113 4of13
Cerebellum Telencephalon
Brainstem
Tectum Diencephalon
Fig. S3. Brain size, morphology, and number of neurons for the avian species examined. Dorsal and lateral views of representative brains are accompanied by
information concerning total number of brain neurons (yellow), number of pallial neurons (blue), and brain mass (red). M, million. (Scale bar, 10 mm.)
Olkowicz et al. www.pnas.org/cgi/content/short/1517131113 5of13
1
10
Brain mass (g)
1
10
Brain mass (g)
B
C
Songbirds
Parrots
Primates
Non -primates mammals
Body mass (g)
10 10
2
10
3
Corvids
Non-corvids songbirds
Parrots
Body mass (g)
10 10
2
10
3
10
4
10
5
10
6
10
7
1
10
Brain mass (g)
10
2
10
3
A
Songbirds
Parrots
Primates
Non -primates mammals
Body mass (g)
10 10
2
10
3
Fig. S4. Brainbody scaling in birds and mammals. (Aand B) Taxonomic differences in relative brain size among songbirds (including both Oscines and
Suboscines), parrots, primates, and nonprimate mammals. Inset in Acorresponds to the magnified view shown in B. Note that allometric lines for these
taxonomic groups are significantly different [full-factorial ANCOVA, slopes: F
(3,2618)
=78.43, P<10
6
; intercepts: F
(3,2618)
=7.44, P<10
4
; post hoc analyses
indicate that the regression line for primates has a different slope (P<0.001 for all pairwise comparisons) and that parrots and songbirds have significantly
larger brains for a given body mass than nonprimate mammals (P<10
6
for both planned comparisons)]. (C) Relative brain size differences among parrots,
corvids, and noncorvid songbirds. Note that allometric lines for these taxonomic groups are significantly different [slopes: F
(2,996)
=4.24, P=0.014; intercepts:
F
(2,996)
=5.99, P=0.003; post hoc analyses indicate the regression line for songbirds has a different slope (P0.045 for both pairwise comparisons) and that
parrots have significantly larger brains for a given body mass than corvids (P<10
6
)]. Mean brain mass versus mean body mass for species are plotted; the
fitted lines represent reduced major axis regressions. The relationship between brain mass and body mass can be described by the following power functions:
songbirds, M
BR
=0.087 ×M
BO0.737
,r
2
=0.953; noncorvid songbirds, M
BR
=0.096 ×M
BO0.698
,r
2
=0.92; corvids, M
BR
=0.097 ×M
BO0.725
,r
2
=0.952; parrots, M
BR
=
0.123 ×M
BO0.716
,r
2
=0.954; primates, M
BR
=0.061 ×M
BO0.823
,r
2
=0.925; nonprimate mammals, M
BR
=0.055 ×M
BO0.730
,r
2
=0.977; all values of P<0.0001. The
data on body mass and brain mass were collated from the literature (for references, see Dataset S3); cetaceans were excluded from the dataset.
Olkowicz et al. www.pnas.org/cgi/content/short/1517131113 6of13
African elephant (4x106)
Giraffe (4.7x105)
Greater kudu (2.2x105)
Pig (1x105)
Human (7x104)
Damaliscus (6x104)
Capybara (4.8x104)
Emu (3.3x104)
Springbok (2.5x104)
Bonnet monkey (8,013)
Baboon (8,000)
Long-tailed macaque (5,700)
Rabbit (4,600)
Rhesus monkey (3,900)
Capuchin monkey (3,340)
Agouti (2,843)
Rock hyrax (2,517)
Prairie dog(1,515)
Western tree hyrax (1,150)
Raven (1,070)
Blue&Yellow Macaw (1,008)
Galago (946.7)
Owl monkey (925.0)
Red Junglefowl (861.3)
Squirrel monkey (858.8)
Kea (708.0)
Squirrel (500.0)
Grey Parrot (453.5)
Sulphur-Cr. Cockatoo (430.0)
Rook (429.3)
Barn Owl (369.7)
Marmoset (361.0)
Rock Pigeon (322.5)
Rat (315.1)
Guinea pig (311.0)
Hill Mynah (262.1)
Goffin's cockatoo (244.8)
Proechimys (223.5)
Alexandrine Parakeet (223.0)
Jackdaw (209.7)
Magpie (178.6)
Tree shrew (172.5)
Hamster (168.1)
Eurasian Jay (160.0)
Four-toed elep. shrew (132.5)
Eastern Rosella (102.0)
Cockatiel (101.0)
Eastern mole (95.30)
Monk Parakeet (93.96)
Blackbird (85.02)
Azure-Winged Magpie (84.11)
Golden mole (79.00)
Starling (73.07)
Mouse lemur (60.00)
Elephant shrew (45.08)
Hairy-tailed mole (42.70)
Star-nosed mole (41.40)
Mouse (40.40)
Budgerigar (35.33)
Green-Rump. Parrotlet (23.23)
Zebra Finch (17.39)
Great Tit (17.12)
Blackcap (16.58)
Short-tailed shrew (16.16)
Smoky shrew (7.77)
Goldcrest (4.52)
African elephant (4,661)
Human (1,509)
Giraffe (539)
Greater kudu (314)
Damaliscus (155)
Baboon (151)
Springbok (107)
Rhesus monkey (87.4)
Capybara (76.0)
Pig (65.0)
Bonnet monkey (61.5)
Capuchin monkey (52.2)
Long-tailed macaque (46.2)
Squirrel monkey (30.2)
Emu (21.8)
Blue&Yellow Macaw (20.7)
Agouti (18.4)
Rock hyrax (17.1)
Owl monkey (15.8)
Raven (14.1)
Kea (13.6)
Western tree hyrax (12.8)
Galago (10.2)
Sulphur-Cr. Cockatoo (10.1)
Rabbit (9.29)
Grey Parrot (8.83)
Rook (8.36)
Goffin's cockatoo (8.27)
Marmoset (7.79)
Jackdaw (6.02)
Squirrel (5.76)
Alexandrine Parakeet (5.69)
Barn Owl (5.62)
Magpie (5.43)
Prairie dog (5.32)
Eurasian Jay (4.59)
Guinea pig (3.76)
Hill Mynah (3.67)
Monk Parakeet (3.42)
Azure-Winged Magpie (3.39)
Tree shrew (2.85)
Red Junglefowl (2.82)
Eastern Rosella (2.72)
Four-toed elep. shrew (2.60)
Cockatiel (2.21)
Proechimys (2.21)
Rock Pigeon (2.10)
Blackbird (1.89)
Starling (1.86)
Mouse lemur (1.83)
Rat (1.80)
Budgerigar (1.32)
Green-Rump. Parrotlet (1.15)
Eastern mole (1.15)
Elephant shrew (1.09)
Hamster (1.02)
Great Tit (0.94)
Golden mole (0.86)
Star-nosed mole (0.85)
Hairy-tailed mole (0.80)
Blackcap (0.77)
Zebra Finch (0.49)
Mouse (0.42)
Short-tailed shrew (0.37)
Goldcrest (0.36)
Smoky shrew (0.19)
African elephant (257,951)
Human (86,060)
Baboon (10,950)
Giraffe (10,775)
Rhesus monkey (6,391)
Greater kudu (4,948)
Bonnet monkey (3,780)
Capuchin monkey (3,691)
Long-tailed macaque (3,439)
Squirrel monkey (3,246)
Blue&Yellow Macaw (3,136)
Damaliscus (3,060)
Springbok (2,736)
Pig (2,229)
Raven (2,171)
Kea (2,149)
Sulphur-Cr. Cockatoo (2,122)
Capybara (1,601)
Grey Parrot (1,566)
Rook (1,509)
Owl monkey (1,478)
Emu (1,335)
Goffin's cockatoo (1,161)
Alexandrine Parakeet (1,096)
Eurasian Jay (1,085)
Jackdaw (968)
Galago (936)
Hill Mynah (906)
Magpie (897)
Rock hyrax (777)
Agouti (753)
Azure-Winged Magpie (741)
Monk Parakeet (697)
Barn Owl (690)
Eastern Rosella (642)
Marmoset (638)
Rabbit (513)
Western tree hyrax (505)
Starling (483)
Squirrel (479)
Prairie dog (474)
Cockatiel (453)
Blackbird (379)
Budgerigar (322)
Rock Pigeon (310)
Tree shrew (274)
Mouse lemur (262)
Guinea pig (240)
Eastern mole (238)
Green-Rump. Parrotlet (227)
Great Tit (226)
Red Junglefowl (221)
Proechimys (211)
Rat (200)
Four-toed eleph. shrew (170)
Goldcrest (164)
Blackcap (157)
Star-nosed mole (142)
Hairy-tailed mole (140)
Elephant shrew (139)
Zebra Finch (136)
Hamster (90.0)
Mouse (70.9)
Golden mole (67.1)
Short-tailed shrew (58.8)
Smoky shrew (39.49)
Human (16,340)
African elephant (5,593)
Baboon (2,880)
Blue&Yellow Macaw (1,917)
Giraffe (1,730)
Rhesus monkey (1,710)
Bonnet monkey (1,660)
Squirrel monkey (1,340)
Kea (1,281)
Raven (1,204)
Capuchin monkey (1,140)
Sulphur-Cr. Cockatoo (1,135)
Grey Parrot (850)
Rook (820)
Long-tailed
macaque
(801)
Greater kudu (763)
Goffin's cockatoo (599)
Alexandrine Parakeet (575)
Damaliscus (571)
Eurasian Jay (529)
Jackdaw (492)
Magpie (443)
Owl monkey (442)
Emu (439)
Barn Owl (437)
Hill Mynah (410)
Azure-Winged Magpie (400)
Springbok (397)
Monk Parakeet (396)
Eastern Rosella (333)
Pig (307)
Capybara (307)
Cockatiel (258)
Marmoset (245)
Starling (226)
Galago (226)
Rock hyrax (198)
Budgerigar (149)
Blackbird (136)
Agouti (111)
Green-Rump. Parrotlet (103)
Western tree hyrax (99.0)
Great Tit (83.0)
Squirrel (77.3)
Rock Pigeon (71.9)
Rabbit (71.5)
Goldcrest (64.2)
Red Junglefowl (60.7)
Tree shrew (60.4)
Zebra Finch (55.2)
Prairie dog (53.8)
Blackcap (52.2)
Guinea pig (43.5)
Four-toed eleph. shrew (33.9)
Rat (31.0)
Eastern mole (28.7)
Proechimys (26.1)
Elephant shrew (25.9)
Mouse lemur (22.3)
Golden mole (21.5)
Star-nosed mole (17.3)
Hamster (17.1)
Hairy-tailed mole (15.7)
Short-tailed shrew (15.4)
Mouse (13.7)
Smoky shrew (9.73)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
Rank Body mass (g) Brain mass (g) Brain neurons (x106) Pallium neurons (x106)
Ordered by
Body mass
1539 50 31 432
57
Brain mass
1332 40 36 533
60
Brain neurons
924 32 45 12 47
59
Pallium neurons
1118 27 52 19 48
63
Afrotheria
Artiodactyla
Eulipotyphla
Songbirds
Primates
Parrots
Rodents
DCBA
E
Fig. S5. Quantitative data currently available for the avian and mammalian species examined with the isotropic fractionator. (AD) Species ranked in descending
order from the largest to the smallest body mass (A), brain mass (B), total number of brain neurons (C), and total number of pallial neurons (D). The mean values of
these variables are given in brackets. (E) Median ranks for the avian and mammalian clades examined. Data for mammals are from published reports (3239).
Olkowicz et al. www.pnas.org/cgi/content/short/1517131113 7of13
Brain mass (g)
110
70
30
% of non-neuronal cells
90
20
40
50
60
80
10
2.34
0.43
Glia/neuron ratio
9.10
0.25
0.67
1.00
1.50
4.00
0.11
CL GG TA DN
DN
TA
GG
CL
50
30
% of non-neuronal cells
55
25
35
40
45
1.00
0.43
Glia/neuron ratio
1.22
0.33
0.54
0.67
0.82
Brain mass (g)
110
B
A
Color code:
Songbirds, Parrots, Pigeon (CL), Red junglefowl (GG), Barn owl (TA), Emu (DN).
telencephalon
cerebellum
diencephalon
tectum
brainstem
Fig. S6. Glia/neuron ratios for the avian species examined. Each point represents the average proportion of nonneuronal cells (left axis) and the glia/neuron
ratio (right axis) for one species, plotted against the average brain mass for that species. Songbirds are shown in green, parrots in red, and other birds in black.
(A) The overall glia/neuron ratio in the brain. Note the higher proportion of nonneuronal cells in all outgroup taxa. (B) Variation in the glia/neuron ratio
among the principal brain divisions investigated. Note that nonneuronal cells constitute a minor cellular fraction in the telencephalon of all species except
three representatives of basal bird lineagesthe emu, the red junglefowl, and the pigeon. Also note the high proportion of nonneuronal cells in the brainstem
and the diencephalon.
Olkowicz et al. www.pnas.org/cgi/content/short/1517131113 8of13
Table S1. Cellular composition of the brains of 28 bird species
Species nBody mass, g Brain mass, g Total neurons, ×10
6
Total nonneurons, ×10
6
Parrots
Green-rumped parrotlet 3 23.2 ±0.7 1.146 ±0.042 227.20 ±3.81 135.00 ±4.70
Budgerigar 3 35.3 ±4.6 1.317 ±0.041 321.82 ±10.62 176.05 ±4.26
Cockatiel 3 101.0 ±4.6 2.205 ±0.132 452.77 ±44.51 234.66 ±10.25
Eastern rosella 3 102.0 ±3.4 2.716 ±0.032 641.88 ±79.00 318.17 ±47.28
Monk parakeet 3 94.0 ±1.7 3.420 ±0.168 696.77 ±75.26 393.49 ±21.06
Alexandrine parakeet 3 223.0 ±5.3 5.699 ±0.492 1,096.26 ±89.81 572.74 ±28.11
Goffins cockatoo 2 244.8 ±3.5 8.275 ±0.548 1,160.59 ±101.87 792.22 ±39.88
Gray parrot 2 453.5 ±47.4 8.827 ±0.859 1,565.93 ±128.99 880.59 ±2.06
Sulfur-crested cockatoo 1 430.0 10.131 2,121.93 1,001.81
Kea 1 708.0 13.593 2,148.67 975.57
Blue and yellow macaw 1 1,008.0 20.731 3,135.79 1,800.03
Variation, max./min. 43.3×18.1×13.8×13.3×
Songbirds
Goldcrest 3 4.5 ±0.1 0.357 ±0.022 163.87 ±8.67 44.16 ±6.57
Zebra finch 3 17.4 ±2.1 0.494 ±0.040 135.98 ±6.82 59.75 ±2.03
Blackcap 3 16.6 ±1.3 0.774 ±0.037 156.73 ±18.91 86.28 ±9.18
Great tit 3 17.1 ±0.3 0.940 ±0.066 225.98 ±46.97 115.43 ±23.43
Starling 3 73.1 ±1.9 1.855 ±0.047 482.50 ±88.29 215.64 ±13.49
Blackbird 3 85.0 ±7.5 1.887 ±0.117 379.41 ±43.33 222.57 ±27.48
Azure-winged magpie 2 84.1 ±16.0 3.393 ±0.486 740.59 ±0.35 349.49 ±33.17
Hill mynah 2 262.1 ±30.7 3.670 ±0.362 906.13 ±45.38 380.79 ±3.56
Eurasian jay 3 160.0 ±12.5 4.597 ±0.307 1,085.42 ±159.56 484.42 ±32.87
Magpie 3 178.6 ±11.5 5.425 ±0.617 897.27 ±57.43 535.97 ±15.96
Jackdaw 3 209.7 ±25.1 6.023 ±0.305 967.99 ±106.66 565.92 ±37.87
Rook 3 429.3 ±35.6 8.357 ±0.312 1,508.72 ±38.25 855.55 ±92.10
Raven 3 1,070.7 ±73.2 14.135 ±0.558 2,170.68 ±72.67 1,242.85 ±98.19
Variation, max./min. 237.9×39.6×16×28.1×
Other birds
Rock pigeon 3 322.5 ±22.7 2.095 ±0.123 309.96 ±33.33 262.18 ±18.94
Red junglefowl 3 861.3 ±107.3 2.819 ±0.200 220.84 ±44.50 286.68 ±17.35
Barn owl 3 369.7 ±37.7 5.618 ±0.404 689.54 ±39.64 522.49 ±25.29
Emu 2 32,600.0 ±1,414.2 21.811 ±2.037 1,335.40 ±29.01 1,528.66 ±118.97
Species ordered by increasing brain size. All values are given as mean ±SD; n, number of individuals analyzed.
Olkowicz et al. www.pnas.org/cgi/content/short/1517131113 9of13
Table S2. Cellular scaling rules for brains of parrots and songbirds
Dependent
variable
Independent
variable Power law r
2
Pvalue
(exponent)
95% confidence
interval
Parrots
M
BR
N
BR
M
BR
=2.669 ×10
10
×N
BR1.144
0.979 <0.000 1 1.0201.269
M
TEL
N
TEL
M
TEL
=1.356 ×10
9
×N
TEL1.075
0.967 <0.000 1 0.9281.221
M
DIE
N
DIE
M
DIE
=5.027 ×10
15
×N
DIE2.031
0.827 <0.000 1 1.3952.668
M
TEC
N
TEC
M
TEC
=5.492 ×10
16
×N
TEC1.998
0.945 <0.000 1 1.6452.350
M
CB
N
CB
M
CB
=3.030 ×10
11
×N
CB1.198
0.974 <0.000 1 1.0521.344
M
BS
N
BS
M
BS
=1.752 ×10
20
×N
BS2.968
0.911 <0.000 1 2.3043.633
M
BR
O
BR
M
BR
=3.267 ×10
10
×O
BR1.170
0.989 <0.000 1 1.0761.263
M
TEL
O
TEL
M
TEL
=9.805 ×10
10
×O
TEL1.126
0.991 <0.000 1 1.0431.208
M
DIE
O
DIE
M
DIE
=2.622 ×10
17
×O
DIE1.102
0.989 <0.000 1 1.0151.189
M
TEC
O
TEC
M
TEC
=3.822 ×10
10
×O
TEC1.160
0.938 <0.000 1 0.9431.380
M
CB
O
CB
M
CB
=2.440 ×10
10
×O
CB1.160
0.970 <0.000 1 1.0311.343
M
BS
O
BS
M
BS
=2.567 ×10
10
×O
BS1.186
0.987 <0.000 1 1.0811.288
Songbirds
M
BR
N
BR
M
BR
=4.699 ×10
11
×N
BR1.227
0.962 <0.000 1 1.0681.387
M
TEL
N
TEL
M
TEL
=4.678 ×10
11
×N
TEL1.134
0.940 <0.000 1 0.9491.320
M
DIE
N
DIE
M
DIE
=2.322 ×10
16
×N
DIE2.208
0.952 <0.000 1 1.8822.520
M
TEC
N
TEC
M
TEC
=2.024 ×10
14
×N
TEC1.736
0.934 <0.000 1 1.4402.033
M
CB
N
CB
M
CB
=2.013 ×10
11
×N
CB1.206
0.972 <0.000 1 1.0721.340
M
BS
N
BS
M
BS
=2.776 ×10
17
×N
BS2.445
0.950 <0.000 1 2.0812.810
M
BR
O
BR
M
BR
=1.536 ×10
9
×O
BR1.093
0.998 <0.000 1 1.0601.125
M
TEL
O
TEL
M
TEL
=5.399 ×10
9
×O
TEL1.043
0.997 <0.000 1 1.0071.080
M
DIE
O
DIE
M
DIE
=6.292 ×10
9
×O
DIE0.992
0.996 <0.000 1 0.9531.032
M
TEC
O
TEC
M
TEC
=1.465 ×10
9
×O
TEC0.946
0.994 <0.000 1 0.8970.996
M
CB
O
CB
M
CB
=2.102 ×10
10
×O
CB1.191
0.965 <0.000 1 1.0431.339
M
BS
O
BS
M
BS
=2.907 ×10
9
×O
BS1.040
0.993 <0.000 1 0.9801.100
Power laws were calculated from the average species values listed in Tables S1 and S3S5. BR, brain; BS, brainstem; CB, cerebellum;
DIE, diencephalon; M, mass (in grams); N, number of neurons; O, number of other (nonneuronal) cells; r
2
, coefficient of determination
calculated from the reduced major axis regression of species averages; TEC, tectum; TEL, telencephalon.
Olkowicz et al. www.pnas.org/cgi/content/short/1517131113 10 of 13
Table S3. Mass of the major brain divisions of 28 bird species
Species Telencephalon, g
Subpallium,
% of Tel Diencephalon, g Tectum, g Cerebellum, g Brainstem, g
Parrots
Green-rumped
parrotlet
0.851 ±0.039 16.0 0.067 ±0.003 0.066 ±0.004 0.098 ±0.002 0.064 ±0.003
Budgerigar 0.935 ±0.039 17.1 0.078 ±0.012 0.073 ±0.002 0.144 ±0.006 0.086 ±0.005
Cockatiel 1.617 ±0.098 15.4 0.119 ±0.006 0.136 ±0.010 0.196 ±0.011 0.138 ±0.016
Eastern rosella 2.009 ±0.053 15.2 0.145 ±0.002 0.175 ±0.015 0.230 ±0.017 0.156 ±0.006
Monk parakeet 2.663 ±0.168 15.9 0.162 ±0.008 0.150 ±0.005 0.281 ±0.002 0.165 ±0.013
Alexandrine parakeet 4.390 ±0.412 16.7 0.292 ±0.025 0.240 ±0.009 0.506 ±0.040 0.272 ±0.013
Goffins cockatoo 6.689 ±0.429 17.7 0.399 ±0.045 0.288 ±0.001 0.571 ±0.034 0.328 ±0.038
Gray parrot 6.973 ±0.780 14.8 0.431 ±0.079 0.331 ±0.035 0.638 ±0.007 0.454 ±0.028
Sulfur-crested cockatoo 8.072 16.2 0.496 0.304 0.836 0.423
Kea 11.383 16.7 0.504 0.372 0.825 0.509
Blue and yellow macaw 17.565 18.2 0.783 0.506 1.245 0.632
Variation, max./min. 20.7×11.7×7.7×12.7×9.9×
Songbirds
Goldcrest 0.225 ±0.023 22.0 0.024 ±0.002 0.045 ±0.004 0.040 ±0.002 0.023 ±0.002
Zebra finch 0.327 ±0.026 15.6 0.032 ±0.006 0.043 ±0.003 0.056 ±0.009 0.036 ±0.003
Blackcap 0.516 ±0.036 12.0 0.056 ±0.002 0.071 ±0.003 0.084 ±0.002 0.047 ±0.004
Great tit 0.675 ±0.051 14.0 0.056 ±0.002 0.073 ±0.008 0.090 ±0.007 0.046 ±0.004
Starling 1.287 ±0.018 11.2 0.113 ±0.008 0.155 ±0.015 0.193 ±0.013 0.107 ±0.011
Blackbird 1.272 ±0.120 14.2 0.125 ±0.014 0.171 ±0.007 0.213 ±0.015 0.107 ±0.009
Azure-winged magpie 2.594 ±0.401 14.5 0.171 ±0.014 0.217 ±0.020 0.271 ±0.035 0.140 ±0.018
Hill mynah 2.577 ±0.300 16.2 0.184 ±0.007 0.323 ±0.018 0.367 ±0.015 0.218 ±0.022
Eurasian jay 3.360 ±0.296 15.1 0.251 ±0.009 0.369 ±0.034 0.411 ±0.013 0.205 ±0.004
Magpie 4.193 ±0.520 11.4 0.263 ±0.015 0.318 ±0.035 0.453 ±0.037 0.197 ±0.021
Jackdaw 4.705 ±0.221 11.5 0.280 ±0.022 0.339 ±0.010 0.483 ±0.043 0.216 ±0.012
Rook 6.648 ±0.246 13.2 0.322 ±0.013 0.425 ±0.014 0.657 ±0.036 0.306 ±0.013
Raven 11.307 ±0.450 9.8 0.570 ±0.086 0.623 ±0.073 1.145 ±0.116 0.49 ±0.012
Variation, max./min. 50.3×23.7×14.5×28.6×21.3×
Other birds
Rock pigeon 1.095 ±0.090 16.5 0.190 ±0.006 0.281 ±0.016 0.332 ±0.013 0.196 ±0.014
Red junglefowl 1.567 ±0.162 14.8 0.245 ±0.014 0.345 ±0.022 0.369 ±0.024 0.293 ±0.010
Barn owl 4.141 ±0.328 7.0 0.347 ±0.007 0.192 ±0.006 0.510 ±0.080 0.427 ±0.011
Emu 14.238 ±1.515 8.8 1.218 ±0.072 1.184 ±0.152 3.399 ±0.181 1.773 ±0.116
Species ordered by increasing brain size. All values are given as mean ±SD.
Olkowicz et al. www.pnas.org/cgi/content/short/1517131113 11 of 13
Table S4. Number of neurons in the major brain divisions of 28 bird species
Species Telencephalon, ×10
6
Subpallium,
% of Tel Diencephalon, ×10
6
Tectum, ×10
6
Cerebellum, ×10
6
Brainstem, ×10
6
Parrots
Green-rumped parrotlet 120.84 ±4.68 14.4 3.25 ±0.49 11.44 ±1.04 89.62 ±5.08 2.05 ±0.21
Budgerigar 186.00 ±4.09 19.9 3.34 ±0.16 11.72 ±2.09 118.73 ±8.31 2.03 ±0.46
Cockatiel 301.81 ±40.57 14.4 3.64 ±1.68 16.11 ±1.11 129.15 ±7.77 2.06 ±0.76
Eastern rosella 406.32 ±77.24 18.0 4.30 ±0.77 16.98 ±2.62 211.98 ±25.42 2.31 ±0.49
Monk parakeet 476.50 ±73.49 16.9 4.78 ±1.33 17.49 ±0.76 195.53 ±22.69 2.47 ±0.44
Alexandrine parakeet 703.96 ±94.37 18.4 5.16 ±0.96 22.95 ±4.78 361.36 ±10.29 2.84 ±0.52
Goffins cockatoo 715.08 ±95.62 16.2 6.63 ±0.37 22.71 ±2.02 413.10 ±7.67 3.07 ±0.24
Gray parrot 1,147.54 ±73.25 25.9 6.75 ±2.71 20.90 ±2.53 387.40 ±51.18 3.35 ±0.69
Sulfur-crested cockatoo 1,490.59 23.8 12.05 27.06 588.41 3.82
Kea 1,630.49 21.4 7.13 26.57 481.14 3.35
Blue and yellow macaw 2,459.15 22.0 7.85 31.22 633.52 4.06
Variation, max./min. 20.4×2.4×2.7×7.1×2×
Songbirds
Goldcrest 76.13 ±5.77 15.7 2.73 ±0.17 16.99 ±1.39 66.78 ±3.28 1.26 ±0.10
Zebra finch 64.66 ±2.84 14.6 2.40 ±0.42 10.62 ±0.50 56.61 ±9.84 1.69 ±0.46
Blackcap 59.71 ±6.23 12.6 3.15 ±0.24 15.34 ±2.31 76.94 ±13.02 1.58 ±0.14
Great tit 96.38 ±27.43 13.9 3.47 ±0.67 16.19 ±0.91 108.27 ±20.71 1.67 ±0.52
Starling 257.08 ±66.07 12.0 4.00 ±0.88 25.44 ±0.54 193.74 ±25.00 2.24 ±0.58
Blackbird 157.63 ±13.12 13.6 4.26 ±0.55 28.26 ±3.53 186.73 ±33.99 2.52 ±0.034
Azure-winged magpie 454.25 ±36.08 12.0 4.99 ±1.34 35.35 ±0.60 243.49 ±35.05 2.51 ±0.65
Hill mynah 484.29 ±49.32 15.3 4.90 ±0.22 47.97 ±0.33 365.74 ±4.07 3.22 ±0.24
Eurasian jay 600.49 ±139.10 11.9 6.75 ±0.66 45.95 ±2.22 429.35 ±28.75 2.88 ±0.47
Magpie 497.94 ±26.98 11.0 6.83 ±1.14 40.44 ±2.80 349.25 ±27.82 2.82 ±0.14
Jackdaw 541.37 ±63.55 9.2 6.82 ±1.55 40.94 ±4.55 375.46 ±41.89 3.39 ±0.73
Rook 917.85 ±68.58 10.6 7.56 ±1.48 42.72 ±3.59 536.28 ±29.31 4.30 ±0.45
Raven 1,355.34 ±73.26 11.2 10.15 ±2.99 47.65 ±7.31 753.64 ±27.34 3.90 ±0.18
Variation, max./min. 22.7×4.2×4.5×13.3×3.1×
Other birds
Rock pigeon 83.35 ±20.53 13.8 2.81 ±0.69 23.76 ±2.69 197.72 ±11.37 2.31 ±0.31
Red junglefowl 73.79 ±2.46 17.8 4.02 ±0.76 25.50 ±3.26 114.45 ±39.59 3.08 ±0.57
Barn owl 453.73 ±13.53 3.6 7.54 ±1.19 9.33 ±1.22 214.31 ±32.07 4.63 ±0.68
Emu 471.57 ±3.54 6.8 10.22 ±2.48 33.72 ±3.74 814.61 ±17.66 5.28 ±1.60
Species ordered by increasing brain size. All values are given as mean ±SD.
Olkowicz et al. www.pnas.org/cgi/content/short/1517131113 12 of 13
Table S5. Number of nonneuronal cells in the brain divisions of 28 bird species
Species Telencephalon, ×10
6
Subpallium,
% of Tel Diencephalon, ×10
6
Tectum, ×10
6
Cerebellum, ×10
6
Brainstem, ×10
6
Parrots
Green-rumped parrotlet 83.01 ±2.45 15.0 11.95 ±1.05 12.42 ±2.35 16.18 ±0.21 11.45 ±0.98
Budgerigar 103.22 ±2.23 20.2 15.49 ±3.29 13.92 ±2.91 27.08 ±4.59 16.35 ±1.72
Cockatiel 142.78 ±14.01 19.1 20.83 ±2.14 19.45 ±2.96 29.72 ±4.78 21.88 ±3.05
Eastern rosella 193.56 ±40.48 15.3 27.47 ±4.39 31.22 ±0.85 38.96 ±2.26 26.97 ±4.38
Monk parakeet 259.75 ±20.81 16.1 34.24 ±0.53 24.40 ±1.78 45.40 ±2.50 29.70 ±1.39
Alexandrine parakeet 353.72 ±34.43 16.2 47.91 ±5.07 47.38 ±8.48 79.03 ±7.98 44.70 ±1.78
Goffins cockatoo 544.50 ±7.14 24.6 66.51 ±9.81 48.00 ±0.51 79.60 ±15.59 53.60 ±7.85
Gray parrot 602.93 ±2.07 10.2 72.44 ±5.46 51.88 ±6.34 90.70 ±2.96 62.63 ±4.71
Sulfur-crested cockatoo 688.63 17.7 87.57 47.89 114.59 63.12
Kea 721.18 19.6 73.85 40.18 78.33 62.03
Blue and yellow macaw 1,383.27 16.3 115.1 66.95 152.48 82.23
Variation, max./min. 16.7×9.6×5.4×9.4×7.2×
Songbirds
Goldcrest 20.48 ±5.03 21.6 4.09 ±0.75 6.69 ±0.19 8.85 ±1.62 4.05 ±0.37
Zebra finch 28.89 ±1.23 19.5 5.45 ±1.06 7.13 ±0.99 10.79 ±2.49 7.49 ±0.92
Blackcap 39.8 ±3.12 11.5 9.79 ±0.79 12.08 ±2.93 16.19 ±2.57 8.41 ±0.40
Great tit 60.11 ±18.23 17.1 10.29 ±1.75 11.20 ±0.77 25.94 ±5.47 7.90 ±1.12
Starling 115.05 ±8.09 14.6 21.84 ±1.57 26.31 ±1.42 33.68 ±5.78 18.76 ±1.34
Blackbird 106.02 ±16.65 18.2 23.54 ±0.36 31.18 ±3.71 42.11 ±8.19 19.72 ±0.81
Azure-winged magpie 220.70 ±14.36 15.9 29.20 ±9.40 38.51 ±4.45 37.46 ±0.75 23.62 ±5.70
Hill mynah 199.22 ±8.05 18.6 36.30 ±3.52 58.41 ±1.74 51.85 ±0.90 35.00 ±0.12
Eurasian jay 282.53 ±34.69 16.1 47.84 ±3.71 59.62 ±7.81 59.59 ±12.83 34.83 ±3.65
Magpie 316.64 ±9.68 11.6 50.33 ±5.41 66.29 ±3.52 64.73 ±8.41 37.99 ±1.72
Jackdaw 366.72 ±40.30 13.5 51.35 ±6.03 54.36 ±1.95 57.82 ±3.60 35.67 ±1.47
Rook 562.20 ±79.47 14.6 54.05 ±3.73 79.57 ±10.68 102.14 ±13.90 57.59 ±9.50
Raven 790.64 ±80.16 12.2 94.86 ±13.90 110.13 ±22.37 173.96 ±4.03 73.27 ±3.51
Variation, max./min. 38.6×23.2×16.5×19.7×18.1×
Other birds
Rock pigeon 102.02 ±16.85 17.6 36.85 ±1.58 41.55 ±0.74 47.61 ±3.58 34.16 ±5.04
Red junglefowl 131.98 ±9.48 17.0 38.42 ±4.70 45.39 ±7.09 28.28 ±9.67 42.63 ±0.98
Barn owl 339.01 ±16.56 8.2 49.74 ±4.50 25.39 ±2.60 54.52 ±10.61 53.82 ±7.26
Emu 865.37 ±74.72 11.2 124.24 ±10.16 123.63 ±16.28 231.05 ±2.58 184.36 ±20.40
Species ordered by increasing brain size. All values are given as mean ±SD.
Other Supporting Information Files
Dataset S1 (XLSX)
Dataset S2 (XLSX)
Dataset S3 (DOCX)
Olkowicz et al. www.pnas.org/cgi/content/short/1517131113 13 of 13
... To determine the structural features associated with high intelligence, one can survey the numerous studies of cognition in animals, and search for those taxa that stand out as having exceptional abilities in general features like learning and problem solving -and what structural features they share. Many lines of evidence indicate that larger brains (ideally measured as the number of neurons and synapses; Herculano-Houzel 2009; Olkowicz et al. 2016, but usually measured more crudely by volume or weight) typically have increased learning and problemsolving ability (Deaner et al. 2007;MacLean et al. 2014), although encephalization in combination with large brain size may also contribute to intelligence (Benson-Amram et al. 2016). Broadscale comparisons across the animal taxa on Earth ( Figure 1) indicate that the general features leading to large brain size are: large body size (Isler et al. 2013;Tsuboi et al. 2018;, endothermic homeothermy (Yu et al. 2014), a high quality diet (DeCasien et al. 2017), and given these three traits, inclusion in an outlier clade that for unresolved reasons have higher than average encephalization, e.g. ...
... These taxa are endothermic and have higher than average encephalization, i.e., their brain mass values lie at the upper edge of the cloud of points for endotherms when log brain size is plotted against log body size (colored red in Figure 1; Yu et al. 2014;Herculano-Houzel 2017). The brain sizes of the parrots and corvids are smaller in volume and mass than those of the other high-cognition taxa, but they also have double the density of neurons (and plausibly also higher synapse densities per neuron) and a higher proportion of brain neurons located in the primary part of the brain used in problem solving (pallial telencephalon) compared with other birds, primates, toothed whales and other mammals (Olkowicz et al. 2016). An additional indication of the association between brain size and intelligence is the weak but highly significantly positive correlation between brain size and IQ in humans (r=0.24, ...
Preprint
When animals evolve sufficient intelligence and dexterity to be able to learn to fabricate utility products (UPs) like tools, the UP's they produce become part of an induced-reproduction system that intrinsically shares many life-like traits with biological organisms, including genome-like fabrication and operation information that is physically-encoded in the animal fabricator’s neural networks. When this set of life- like traits includes a sufficient capacity for system-improving cultural evolution (UP-evolvability), the UPs become ‘para-alive’, i.e., nearly alive, or a form of non-biological UP-paralife that is equivalent to the life- status of biological viruses, plasmids, and transposons. In the companion paper I focus on the evolution of UP-paralife in the context of modern, language-capable humans and its predicted evolution going forward in time (Rice 2022). Here I look backward in time and focus on the origin of UP-paralife and its subsequent coevolution with human intelligence. I begin by determining the pathways leading to the evolution of large brains in the rare lineages of biological life that have sufficient intelligence to learn to fabricate tools –a critical first step in the evolution of UP-paralife. The simplest forms of these learning- based UPs, made by species like chimpanzees and New Caledonian crows, represent only proto-UP- paralife because they lack sufficient UP-evolvability. Expanded UP-evolvability required a combination of three attributes that enabled continuous niche-expansion of the animal fabricator via a new and advanced form of UP-mediated teamwork (TW): i) self-domestication that facilitated TW among low-related individuals, ii) learned volitional words (protolanguage) that represent ephemeral UPs that coordinate TW, and iii) learned fabrication of simple flaked-stone tools with cutting and chopping capabilities (a UP to make other structural UPs) that expanded teammate phenotypes and TW capabilities. This specific triad of attributes is synergistic because each one acts as a TW-enhancer that can gradually erode different components of the three major constraints on TW operation and expansion: too much selfishness, insufficient coordination signals, and insufficient physical traits of teammates. The increase in UP- evolvability was transformative and marked the origin of UP-paralife and the initiation of coevolution between UP-paralife (cultural evolution) and the intelligence of its hominin/human symbiont (genetic evolution) that fostered 2.5 million years of: i) continuous brain size increase and niche-expansion within the genus Homo, and ii) parallel advances in the diversity, complexity and uses of UP-paralife. This coevolution also fostered evolutionary expansion of word-based communication, and eventually language, that acted in a catalyst-like manner to facilitate the evolution of increasingly complex forms of imagination, reasoning, mentalizing, and UP-generating technology. I next focus on the evolution of creativity in the human lineage –in the form of divergent thinking and creative imagination. I conclude that the evolution of this advanced cognitive feature required a preadaptation of sufficient intelligence and is the component of human cognition that was the major causal factor generating the greatly expanded diversity and complexity of UP-paralife currently associated with modern humans. Lastly, I apply my findings to the issue of the prevalence of extraterrestrial intelligent life. I conclude that any exoplanets with detected chemical life will very rarely (e.g., probability ~10-5 for a planet closely matching Earth’s characteristics) have evolved intelligence equalling or exceeding that of humans.
... NCL neurons are selectively responsive to numerosity in visual displays 19,28-30 , even in numerically naïve crows 31 and newborn domestic chicks 32 , but also encode the planning and execution of goal-directed movements 33,34 . Therefore, the NCL operates at the apex of the perception-action loop in the avian brain and is considered the avian analog of the mammalian prefrontal cortex (PFC) [35][36][37] . We hypothesized that neurons in the NCL Nature Communications | (2022) 13:6913 1 1234567890():,; 1234567890():,; ...
... NCL neurons are selectively responsive to numerosity in visual displays 19,[28][29][30] , even in numerically naïve crows 31 and newborn domestic chicks 32 , but also encode the planning and execution of goal-directed movements 33,34 . Therefore, the NCL operates at the apex of the perception-action loop in the avian brain and is considered the avian analog of the mammalian prefrontal cortex (PFC) [35][36][37] . We hypothesized that neurons in the NCL represent the numerical sensorimotor transformation process. ...
Article
Full-text available
Translating a perceived number into a matching number of self-generated actions is a hallmark of numerical reasoning in humans and animals alike. To explore this sensorimotor transformation, we trained crows to judge numerical values in displays and to flexibly plan and perform a matching number of pecks. We report number selective sensorimotor neurons in the crow telencephalon that signaled the impending number of self-generated actions. Neuronal population activity during the sensorimotor transformation period predicted whether the crows mistakenly planned fewer or more pecks than instructed. During sensorimotor transformation, both a static neuronal code characterized by persistently number-selective neurons and a dynamic code originating from neurons carrying rapidly changing numerical information emerged. The findings indicate there are distinct functions of abstract neuronal codes supporting the sensorimotor number system.
... Importantly, birds are also the only non-mammalian group to express AQP4 on astrocytic endfeet facing the blood vessels, a necessary cellular component of the glymphatic system 6,15 . Finally, as birds are homeotherms with high metabolic rates and large brains more densely packed with neurons than in mammals 16,17 , they might have a greater need to clear metabolic waste from the brain. ...
Preprint
Full-text available
Mammalian sleep has been implicated in maintaining a healthy extracellular environment in the brain. During wakefulness, neuronal activity leads to the accumulation of toxic proteins implicated in Alzheimer’s disease. The glymphatic system is thought to clear these proteins by flushing cerebral spinal fluid (CSF) through the brain. In mice, this process occurs during non-rapid eye movement (NREM) sleep. And, in humans, increased ventricular CSF flow during NREM sleep, visualized using functional magnetic resonance imaging (fMRI), is also thought to be coupled to flow through the glymphatic system. But does waste clearance occur throughout sleep or is it specific to NREM sleep? Using fMRI of naturally sleeping pigeons, we show that REM sleep, a paradoxical state with wake-like brain activity, during which we experience our most vivid dreams, is accompanied in birds with the activation of brain regions involved in processing visual information, including optic flow during flight. We further demonstrate that, compared to NREM sleep, REM sleep is associated with a sharp drop in ventricular CSF flow. Consequently, functions linked to brain activation during REM sleep might come at the expense of waste clearance during NREM sleep.
... Kuhl et al. (2006) have shown that auditory experience drives a progressive process of integration of language-specific phonemes in auditory memory, which may be indicative of analogous neural circuitry to that observed in songbirds and rodents. Following this work, a parallel to birdsong template theory (Konishi, 1963b) has been put forward and elaborated by Kuhl and colleagues (Kuhl, 1992;Kuhl and Meltzoff, 1996;Kuhl et al., 2006; see also Vihman, 2019). 2 Crucially, recent iterations of Frank Guenther's DIVA model (Guenther and Vladusich,2 In this context it is worth noting that the degree to which the organization of the songbird brain parallels that of humans (and other mammals) is subject to extensive, as of yet unsettled debate (Reiner et al., 2004;Petkov and Jarvis, 2012;Olkowicz et al., 2016;Prather et al., 2017). 2012; Guenther, 2016) present a coherent argument for how such conversion from auditory speech "chunk" component to motor vocal production behavior may take place; that is, two-way prediction of motor and sensory domains facilitates the establishment of a "speech sound map" (Guenther, 2016). ...
Article
Full-text available
Every normally developing human infant solves the difficult problem of mapping their native-language phonology, but the neural mechanisms underpinning this behavior remain poorly understood. Here, motor constellation theory, an integrative neurophonological model, is presented, with the goal of explicating this issue. It is assumed that infants’ motor-auditory phonological mapping takes place through infants’ orosensory “reaching” for phonological elements observed in the language-specific ambient phonology, via reference to kinesthetic feedback from motor systems (e.g., articulators), and auditory feedback from resulting speech and speech-like sounds. Attempts are regulated by basal ganglion–cerebellar speech neural circuitry, and successful attempts at reproduction are enforced through dopaminergic signaling. Early in life, the pace of anatomical development constrains mapping such that complete language-specific phonological mapping is prohibited by infants’ undeveloped supralaryngeal vocal tract and undescended larynx; constraints gradually dissolve with age, enabling adult phonology. Where appropriate, reference is made to findings from animal and clinical models. Some implications for future modeling and simulation efforts, as well as clinical settings, are also discussed.
Article
Full-text available
The allometric scaling of the brain size and neuron number across species has been extensively studied in recent years. With the exception of primates, parrots, and songbirds, larger brains have more neurons but relatively lower neuronal densities than smaller brains. Conversely, when considering within-population variability, it has been shown that mice with larger brains do not necessarily have more neurons but rather more neurons in the brain reflect higher neuronal density. To what extent this intraspecific allometric scaling pattern of the brain applies to individuals from other species remains to be explored. Here, we investigate the allometric relationships among the sizes of the body, brain, telencephalon, cerebellum, and optic tectum, and the numbers of neurons and non-neuronal cells of the telencephalon, cerebellum, and optic tectum across 66 individuals originated from an intercross between wild and domestic chickens. Our intercross of chickens generates a population with high variation in brain size, making it an excellent model to determine the allometric scaling of the brain within population. Our results show that larger chickens have larger brains with moderately more neurons and non-neuronal cells. Yet, absolute number of neurons and non-neuronal cells correlated strongly and positively with the density of neurons and non-neuronal cells, respectively. As previously shown in mice, this scaling pattern is in stark contrast with what has been found across different species. Our findings suggest that neuronal scaling rules across species are not a simple extension of the neuronal scaling rules that apply within a species, with important implications for the evolutionary developmental origins of brain diversity.
Chapter
Brain morphology has become a key element to predict a wide array of cognitive and behavioral, sensory and motor abilities, and to determine evolutionary rates of phenotypic transformation. Our information on early bird brain morphology comes of natural endocasts or studies of the intracranial cavity. Although the first studies of fossil bird brains were published almost two centuries ago, there is still relatively little known about the avian brain and its evolution compared with other groups such as mammals. This is due primarily to the fact that few three-dimensionally preserved skulls of early birds are recognized. The avian brain occupies the entire intracranial cavity, so that it is possible to reconstruct high-quality 3D virtual endocast models that can be used as excellent proxies for both volume and morphology of the brain. This technique has driven advances in avian paleoneurology from 2000 onwards. In this chapter, we provide a holistic view of the main features of the avian brain and senses, its disparity and potential use in paleobiological inferences, and discuss the main changes across the transition from non-avian theropods to derived Neornithes.
Article
Evolution of Learning and Memory Mechanisms is an exploration of laboratory and field research on the many ways that evolution has influenced learning and memory processes, such as associative learning, social learning, and spatial, working, and episodic memory systems. This volume features research by both outstanding early-career scientists as well as familiar luminaries in the field. Learning and memory in a broad range of animals are explored, including numerous species of invertebrates (insects, worms, sea hares), as well as fish, amphibians, birds, rodents, bears, and human and nonhuman primates. Contributors discuss how the behavioral, cognitive, and neural mechanisms underlying learning and memory have been influenced by evolutionary pressures. They also draw connections between learning and memory and the specific selective factors that shaped their evolution. Evolution of Learning and Memory Mechanisms should be a valuable resource for those working in the areas of experimental and comparative psychology, comparative cognition, brain–behavior evolution, and animal behavior.
Article
The evolution of the nervous system progressed through cellular diversification and specialization of functions. Conceptually, the nervous system is composed of electrically excitable neuronal networks connected by chemical synapses and nonexcitable glial cells that provide for homeostasis and defense. The evolution of neuroglia began with the emergence of the centralized nervous system and proceeded through a continuous increase in their complexity. In the primate brain, especially in the brain of humans, the astrocyte lineage is exceedingly complex, with the emergence of new types of astroglial cells possibly involved in interlayer communication and integration. This paper describes the evolution of neuroglia, which began with the emergence of the centralized nervous system and proceeded through a continuous increase in their complexity. In the primate brain, especially in the brain of humans, the astrocyte lineage is exceedingly complex, with the emergence of new types of astroglial cells possibly involved in interlayer communication and integration.
Article
There is a reason this primer is called ‘The avian brain’ and not ‘The birdbrain’ even though the latter term might be more accessible. The term birdbrain was long used as a derogative, yet I hope that this primer will inspire you to see it more as I do — as a compliment. This primer will take you on a short tour. First, I will explore superficial differences between birds and mammals. Then I will briefly mention the behavioral capabilities of birds and explain why the bird brain has the neural makeup to generate these behaviors. This will lead me to briefly describe some key neural circuits and structures and to conclude with the ongoing challenge to identify functional differences in the neural substrates of avian and mammalian cognition. The aim of this short article, therefore, is not to give a full review of the avian brain but rather to highlight important aspects with a focus on the comparison to the mammalian brain and cognition.
Article
The present contribution offers a descriptive account of two recent books concerning shamanism, Homayun Sidky’s The Origins of Shamanism, Spirit Beliefs, and Religiosity: A Cognitive Anthropological Perspective (2017) and Sergio Botta’s Dagli sciamani allo sciamanesimo. Discorsi, credenze, pratiche (2018). The commentary starts by supplying a brief historical contextualization of the subfield of shamanic studies in both Anthropology and the History of Religions, highlighting the main trends and widespread approaches. Sidky’s neurocognitive account and Botta’s poststructural historiographical walk-through are then taken into consideration and reviewed. The conclusions under-score the need for an integration between these two perspectives and urge cognitive historians to collaborate with like-minded anthropologists in order to further the study of shamanism and prevent the subfield from becoming de novo monopolized by paranormal and postmodern anthropology.
Article
Full-text available
Understanding the mechanisms of evolution of brain pathways for complex behaviours is still in its infancy. Making further advances requires a deeper understanding of brain homologies, novelties and analogies. It also requires an understanding of how adaptive genetic modifications lead to restructuring of the brain. Recent advances in genomic and molecular biology techniques applied to brain research have provided exciting insights into how complex behaviours are shaped by selection of novel brain pathways and functions of the nervous system. Here, we review and further develop some insights to a new hypothesis on one mechanism that may contribute to nervous system evolution, in particular by brain pathway duplication. Like gene duplication, we propose that whole brain pathways can duplicate and the duplicated pathway diverge to take on new functions. We suggest that one mechanism of brain pathway duplication could be through gene duplication, although other mechanisms are possible. We focus on brain pathways for vocal learning and spoken language in song-learning birds and humans as example systems. This view presents a new framework for future research in our understanding of brain evolution and novel behavioural traits.
Article
Full-text available
Although reconstruction of the phylogeny of living birds has progressed tremendously in the last decade, the evolutionary history of Neoaves-a clade that encompasses nearly all living bird species-remains the greatest unresolved challenge in dinosaur systematics. Here we investigate avian phylogeny with an unprecedented scale of data: >390,000 bases of genomic sequence data from each of 198 species of living birds, representing all major avian lineages, and two crocodilian outgroups. Sequence data were collected using anchored hybrid enrichment, yielding 259 nuclear loci with an average length of 1,523 bases for a total data set of over 7.8 × 10(7) bases. Bayesian and maximum likelihood analyses yielded highly supported and nearly identical phylogenetic trees for all major avian lineages. Five major clades form successive sister groups to the rest of Neoaves: (1) a clade including nightjars, other caprimulgiforms, swifts, and hummingbirds; (2) a clade uniting cuckoos, bustards, and turacos with pigeons, mesites, and sandgrouse; (3) cranes and their relatives; (4) a comprehensive waterbird clade, including all diving, wading, and shorebirds; and (5) a comprehensive landbird clade with the enigmatic hoatzin (Opisthocomus hoazin) as the sister group to the rest. Neither of the two main, recently proposed Neoavian clades-Columbea and Passerea-were supported as monophyletic. The results of our divergence time analyses are congruent with the palaeontological record, supporting a major radiation of crown birds in the wake of the Cretaceous-Palaeogene (K-Pg) mass extinction.
Article
Full-text available
The ability to imitate complex sounds is rare, and among birds has been found only in parrots, songbirds, and hummingbirds. Parrots exhibit the most advanced vocal mimicry among non-human animals. A few studies have noted differences in connectivity, brain position and shape in the vocal learning systems of parrots relative to songbirds and hummingbirds. However, only one parrot species, the budgerigar, has been examined and no differences in the presence of song system structures were found with other avian vocal learners. Motivated by questions of whether there are important differences in the vocal systems of parrots relative to other vocal learners, we used specialized constitutive gene expression, singing-driven gene expression, and neural connectivity tracing experiments to further characterize the song system of budgerigars and/or other parrots. We found that the parrot brain uniquely contains a song system within a song system. The parrot "core" song system is similar to the song systems of songbirds and hummingbirds, whereas the "shell" song system is unique to parrots. The core with only rudimentary shell regions were found in the New Zealand kea, representing one of the only living species at a basal divergence with all other parrots, implying that parrots evolved vocal learning systems at least 29 million years ago. Relative size differences in the core and shell regions occur among species, which we suggest could be related to species differences in vocal and cognitive abilities.
Article
Full-text available
Significance A six-layered neocortex is a hallmark feature of the mammalian brain. Connections among layers and progressive changes in the neural coding properties of each layer define a microcircuit thought to perform the computations underlying complex behavior. Birds lack a six-layered cortex. Yet, they demonstrate complex cognition and behavior. Recent anatomical studies propose that adjacent regions of the avian pallium are homologs of neocortical layers. Here, we show that the avian auditory pallium exhibits the same information-processing principles that define the mammalian neocortical microcircuit. Results suggest that the cortical microcircuit evolved in a common ancestor of mammals and birds and provide a physiological explanation for neural processes that give rise to complex behavior in the absence of cortical lamination.