Scaling Patterns of Cerebellar Petrosal Lobules in Euarchontoglires: Impacts of Ecology and Phylogeny

Abstract

The petrosal lobules (in whole or part homologous with the paraflocculi) of the cerebellum regulate functions associated with vision including smooth pursuit and velocity control of eye movements, suggesting a possible relationship between the petrosal lobules and behavioral adaptation. Previous studies have produced diverging conclusions regarding the lobules’ ecological signal. The current study examines lobule scaling within an ecologically diverse but phylogenetically constrained sample of extant mammals to determine if ecology influences relative petrosal lobule size. Using the endocasts of 140 Euarchontoglires (Primates, Scandentia, Dermoptera, Lagomorpha, Rodentia), petrosal lobule size was evaluated relative to endocranium and body size, accounting for phylogenetic relationships and ecology (locomotor behavior, diet, activity pattern). Results show a strong positive relationship between lobule size and both endocranial volume and body mass. Phylogeny is a major factor in the scaling of the petrosal lobules, with significant differences in relative size identified between orders and suborders. Concerning ecology, fossorial taxa were found to have significantly smaller petrosal lobules relative to body mass compared to other locomotor groups across Euarchontoglires. The small lobules possessed by this group may reflect an adaptation related to reduced visual reliance. In contrast to previous research, no relationship was identified between relative lobule size and any other ecological variables. While variation in relative lobule size may be adaptively significant in some groups (i.e. fossorial species), it is critical to study the evolution of petrosal lobule size within a narrow phylogenetic scope, with inclusion of fossil material to inform our understanding of evolutionary trajectories.

Full text

FULL LENGTH ARTICLE
Scaling patterns of cerebellar petrosal lobules in
Euarchontoglires: Impacts of ecology and phylogeny
Madlen M. Lang
1
| Ornella C. Bertrand
2
| Gabriela San Martin-Flores
3
|
Chris J. Law
4,5,6
| Jade Abdul-Sater
1
| Shayda Spakowski
1
| Mary T. Silcox
1
1
Department of Anthropology, University
of Toronto Scarborough, Toronto,
Ontario, Canada
2
School of GeoSciences, University of
Edinburgh, Grant Institute, Edinburgh,
Scotland, UK
3
Flinders University, College of Science
and Engineering, Bedford Park, South
Australia, Australia
4
Richard Gilder Graduate School,
Department of Mammalogy, and Division
of Paleontology, American Museum of
Natural History, New York, New
York, USA
5
Department of Biology, University of
Washington, Seattle, Washington, USA
6
Department of Integrative Biology, The
University of Texas at Austin, Austin,
Texas, USA
Correspondence
Madlen M. Lang, Department of
Anthropology, University of Toronto
Scarborough, Toronto, ON M1C 1A4,
Canada.
Email: madlen.lang@mail.utoronto.ca
Funding information
Early Career Provost Fellowship at the
University of Texas; Gerstner Family
Foundation at the American Museum of
Natural History; National Science
Foundation, Grant/Award Number: NSF
postdoctoral fellowship (DBI-1906248);
Natural Sciences and Engineering
Research Council of Canada, Grant/
Award Numbers: CGS-M Award,
Discovery Grant
Abstract
The petrosal lobules (in whole or part homologous with the paraflocculi) of
the cerebellum regulate functions associated with vision including smooth pur-
suit and velocity control of eye movements, suggesting a possible relationship
between the petrosal lobules and behavioral adaptation. Previous studies have
produced diverging conclusions regarding the lobules' ecological signal. The
current study examines lobule scaling within an ecologically diverse but phylo-
genetically constrained sample of extant mammals to determine whether
ecology influences relative petrosal lobule size. Using the endocasts of
140 Euarchontoglires (Primates, Scandentia, Dermoptera, Lagomorpha,
Rodentia), petrosal lobule size was evaluated relative to endocranium and
body size, accounting for phylogenetic relationships and ecology (locomotor
behavior, diet, activity pattern). Results show a strong positive relationship
between lobule size and both endocranial volume and body mass. Phylogeny is
a major factor in the scaling of the petrosal lobules, with significant differences
in relative size identified between orders and suborders. Concerning ecology,
fossorial taxa were found to have significantly smaller petrosal lobules relative
to body mass compared to other locomotor groups across Euarchontoglires.
The small lobules possessed by this group may reflect an adaptation related to
reduced visual reliance. In contrast to previous research, no relationship was
identified between relative lobule size and any other ecological variables.
While variation in relative lobule size may be adaptively significant in some
groups (i.e., fossorial species), it is critical to study the evolution of petrosal
lobule size within a narrow phylogenetic scope, with inclusion of fossil mate-
rial to inform our understanding of evolutionary trajectories.
KEYWORDS
ecology, endocast, Euarchontoglires, neuroanatomy, paraflocculi, petrosal lobule
1|INTRODUCTION
Analyses of the scaling patterns of the mammalian brain
and its components provide valuable information about
how the brain responds to selective pressures and can
help identify broader evolutionary trends. To this end,
numerous studies have drawn connections between eco-
logical specialization and variation in the size and
Received: 27 July 2021 Revised: 2 February 2022 Accepted: 21 February 2022
DOI: 10.1002/ar.24929
Anat Rec. 2022;1–32. wileyonlinelibrary.com/journal/ar © 2022 American Association for Anatomy. 1

morphology of the mammalian brain and its particular
structures (e.g., Barton & Harvey, 2000; de Winter &
Oxnard, 2001; DeCasien & Higham, 2019). These studies
generally operate on the premise that an adaptation
requiring an increase in the information sent to certain
neural tissues will result in an increase in the relative size
of those tissues (i.e., Jerison's, 1973,1979 principle of
proper mass), and therefore variation in the relative sizes
of functionally significant brain tissues reflects adaptive
specialization (Barton & Harvey, 2000).
The evolutionary relevance of interspecific variation
in the brain and its parts in relation to sensory and cogni-
tive adaptations has been extensively investigated in pri-
mates (DeCasien & Higham, 2019). Variation in the size
of the neocortex, the olfactory bulbs, and in functionally
significant cortical regions has been attributed to ecologi-
cal behaviors associated with vision and activity pattern
(Barton, 1998; Barton et al., 1995; Kirk, 2006; Powell
et al., 2017), diet and foraging strategy (Clutton-Brock &
Harvey, 1980; DeCasien et al., 2017; DeCasien &
Higham, 2019; Harvey et al., 1980; Powell et al., 2017),
and sociality (Barton, 2006; Dunbar, 1992,1995;
Dunbar & Shultz, 2007; Kudo & Dunbar, 2001;
Sawaguchi & Kudo, 1990). Similar ecology-related scaling
patterns have also been identified in rodents
(e.g., Bernard & Nurton, 1992; Bertrand et al., 2017,2018,
2019a;2021; Krubitzer et al., 2011; Lemen, 1980; Mace
et al., 1981; Meier, 1983; Pilleri et al., 1984; Roth &
Thorington, 1982) and in other small mammals (Barton
et al., 1995; Bhatnagar & Kallen, 1974; de Winter &
Oxnard, 2001; Eisenberg & Wilson, 1978,1981;
Gittleman, 1986,1991; Harvey et al., 1980; Mace
et al., 1981). These studies indicate that species belonging
to independent lineages can exhibit neurological conver-
gences in functionally and ecologically important brain
structures, causing brain architecture to diverge from
those of phylogenetically closer species (de Winter &
Oxnard, 2001).
The widespread availability of high-resolution x-ray
computed tomography (CT) data has led to the prolifera-
tion of studies evaluating variation in the endocast and
endocranium of extinct and extant groups (Balanoff &
Bever, 2020). Of these studies, however, comparatively
few have focused on the evolution of the mammalian cer-
ebellum and its components specifically, within the con-
text of variation as a function of ecological and
behavioral factors (Barks et al., 2015; Ferreira-Cardoso
et al., 2017; Ridgway & Hanson, 2014). One component
of the cerebellum, the petrosal lobules, is of particular
interest. As part of the vestibulocerebellum (or arch-
icerebellum), the petrosal lobules are among the evolu-
tionarily oldest anatomical components of the cerebellum
(Kheradmand & Zee, 2011). The vestibulocerebellum is
comprised of the nodulus, ventral uvula, and floccular-
parafloccular complex (Kheradmand & Zee, 2011).
Together, these brain structures play a major role in
the stabilization of eye movements and in the vest-
ibuloocular reflex (VOR) (Hiramatsu et al., 2008;
Kheradmand & Zee, 2011; Rambold et al., 2002). Within
the vestibulocerebellum, the petrosal lobules are part of
the floccular–parafloccular complex, sometimes referred to
as the “floccular lobe”(Fukushima et al., 1999;Krauzlis&
Lisberger, 1994), “floccular complex”(Rambold et al.,
2002), or “floccular region”(Belton & McCrea, 2000a,
2000b). The floccular–parafloccular complex includes the
flocculi, paraflocculi, and the petrosal lobules (as a compo-
nent of the paraflocculi). These tissues receive visual and
vestibular information from the visual cortex and the
semicircular canals and output to the ocular muscles to
regulate eye movements (Burne & Woodward, 1983;
Hiramatsu et al., 2008; Voogd & Barmack, 2006). Specifi-
cally, the floccular–parafloccular complex stabilizes visual
images on the retina by generating compensatory eye
movements (Rambold et al., 2002), regulates smooth pur-
suit eye movements to prevent retinal image blur of a
moving object (Ilg & Thier, 2008;Nagao,1992;Rambold
et al., 2002; Shojaku et al., 1990;Zeeetal.,1981), and
velocity control of eye movements (Hiramatsu
et al., 2008). The petrosal lobules themselves extend later-
ally from the cerebellum to fill a distinct cavity in the cra-
nium called the subarcuate fossa in many vertebrate
groups including mammals (Gannon et al., 1988), birds
(Walsh et al., 2013), and pterosaurs (Witmer et al., 2003)
(Figure 1). The fossa is formed by the petrosal bone and
extends through the arch of the anterior semicircular canal
(Gannon et al., 1988;Figure2).
The homology of the cerebellar structures that are
housed in the subarcuate fossa can vary between orders,
which has implications for their nomenclature and how
we understand their functions between taxa. In Glires
(Panezai et al., 2020; Tan et al., 1995) and Scandentia (Ni
et al., 2018) the lobules are formed by the paraflocculi,
which are divided into the ventral and dorsal par-
aflocculus (Tan et al., 1995). In contrast, the petrosal lob-
ules of primates are formed by only a portion of the
paraflocculi (Voogd & Barmack, 2006). While the dorsal
and the ventral paraflocculi are functionally linked to the
petrosal lobules (Voogd et al., 2012), they sit within the
endocranial cavity adjacent to the flocculus and
subarcuate fossa (Xiong et al., 2010). Previous analysis
examining the scaling of the subarcuate fossa and the
petrosal lobules/paraflocculus of mammals, including
primates, rodents, and lagomorphs, found a strong corre-
lation between the size of the fossa and the tissues which
sit inside it (Gannon et al., 1988). Significantly, this
means it is possible to obtain an estimate of the size of
2LANG ET AL.

these lobules using the subarcuate fossa. As a result, we
use the term “petrosal lobule”to refer to the tissue occu-
pying the subarcuate fossa instead of “paraflocculus”for
all taxa in this analysis for two reasons: (a) this sample
includes primates and therefore the paraflocculi are not
being measured per se across all taxa and (2) as we are
using endocasts and not neural tissue, any measurements
obtained represent the potential maximum volume of the
tissue occupying the subarcuate fossa of the petrosal bone
and consequently, the term “petrosal lobule”is more
accurate in describing what can be measured.
As the petrosal lobules are protected inside the
subarcuate fossa, they are difficult to study using micro-
electrode techniques and therefore their specific function
as part of the floccular–parafloccular complex is difficult
to assess (Hiramatsu et al., 2008). Nevertheless, there is
strong evidence that the petrosal lobules play a key role
in smooth pursuit and velocity control of eye movements
(Hiramatsu et al., 2008). Within primates, the petrosal
lobules receive projections from the dorsolateral pontine
nucleus (Glickstein et al., 1994; Nagao et al., 1997;
Xiong & Nagao, 2002; Xiong et al., 2010), which repre-
sents a major relay area for smooth pursuit eye move-
ments to the cerebellum (May et al., 1988). Connections
to the lobules, via the pontine nucleus, have been identi-
fied in the medial temporal/medial superior temporal
FIGURE 1 Images of Indri indri (AMNH 100506) endocast depicting petrosal lobules (red) and the subarcuate fossa. (a) Sagittal cross
section of cranium showing subarcuate fossa, (b) left lateral view of endocast in semi-transparent cranium, (c) right lateral view, (d) ventral
view, and (e) left lateral view of endocast; scale =10 mm
FIGURE 2 Endocast of Perodicticus potto (MCZ 42622)
showing the petrosal lobules (red) protruding through and anterior
semicircular canal (inner ear in yellow). Endocast shown (a) right
lateral view and (b) ventral view of endocast; scale =10 mm
LANG ET AL.3

extrastriate visual areas of the parietal cortex (Glickstein
et al., 1994), also known as important sites for smooth
pursuit eye movements (Newsome et al., 1988). The lob-
ules also project to preoculomotor structures via the lat-
eral interstitial nuclei (Xiong & Nagao, 2002).
Significantly, lesioning of the lobules in macaques
reduced the velocity of smooth pursuit eye movements
(Hiramatsu et al., 2008). There have been no systematic
studies of the functional role of the rodent paraflocculi
(=petrosal lobules). However, the neural connections of
the ventral and dorsal paraflocculi of the rat are consis-
tent with those of the macaque ventral and dorsal par-
aflocculi (Osanai et al., 1999), which play a role in
smooth pursuit eye movements as well as VOR adapta-
tions (Rambold et al., 2002). Like the primate petrosal
lobules, the paraflocculi of rodents appear to be a major
receiving area of the cerebellum for information from the
visual cortices (Burne & Woodward, 1983). Overall, the
petrosal lobules of the cerebellum in euarchontoglirans
seem to play a significant role in an animal's ability to
regulate eye movements to maintain a clear image of a
moving object.
Some research has suggested that the large fossae, and
therefore lobules, possessed by birds and pterosaurs reflect
adaptive specialization to flight (Witmer et al., 2003;but
see Walsh et al., 2013) with connections drawn between
fossa/lobule size and ecology within birds (Ferreira-
Cardoso et al., 2017). Within mammals, however, support
for this connection has varied. Analysis of fossil and extant
rodent endocasts by Bertrand and colleagues identified a
trend toward smaller petrosal lobules in progressively
more fossorial sciuroids (Bertrand et al., 2018,2021)as
well as a marked increase in lobular size in early arboreal
squirrels (Bertrand et al., 2017,2021). Conversely, analysis
by Ferreira-Cardoso et al. (2017) found that, in a diverse
sample of primarily extant mammals, ecological variables
could not explain variation in fossa/lobule size, and varia-
tion was primarily attributed to phylogenetic influence.
The contrasting perspectives offered by these analyses, and
the potential usefulness of the fossa and associated brain
structure if added to our current repertoire of brain fea-
tures that can be measured in fossil endocasts, necessitates
further examination.
The purpose of this study is to perform such an exam-
ination on a large dataset of virtual endocasts from
closely related species (Superorder Euarchontoglires [Pri-
mates, Scandentia, Dermoptera, Rodentia, Lagomorpha])
to determine whether ecology influences petrosal lobule
scaling patterns. Specifically, this paper seeks to evaluate
(a) the scaling relationship between petrosal lobule size
relative to the size of the rest of the endocranium, and to
body size; (b) the impact of phylogeny on relative petro-
sal lobule size; and (c) the influence of ecological
behaviors on relative petrosal lobule size within a phylo-
genetically constrained group of mammals. If the results
indicate that ecology plays a significant role in petrosal
lobule scaling (as measured from the size of the
subarcuate fossa), then these structures may be used to
help interpret endocranial morphology of fossil endocasts
for this group from the perspective of sensory ecology.
2|MATERIALS AND DATA
COLLECTION
Endocasts of 140 extant euarchontoglirans, including
lagomorphs (15), primates (38 =Haplorhini [18],
Strepsirrhini [20]), rodents (71 =Squirrel-related clade
[27], Ctenohystrica [17], Mouse-related clade [27]),
scandentians (14), and dermopterans (2) (Table 1;
Figure 3) were used. All endocasts were constructed
using micro-CT scans of crania. Endocranial and petrosal
lobule volumes were taken from the literature for
Sciuroidea (Bertrand et al., 2017,2018,2019b,2021),
Lagomorpha (L
opez-Torres et al., 2020) and Scandentia
(San Martin-Flores et al., 2018). Thirty-four of the
nonsquirrel-related clade rodents, seven lagomorphs, and
the two dermopterans were scanned using a high-
resolution x-ray micro-CT scanner at the Shared Mate-
rials Instrumentation Facility (SMIF), Duke University
(North Carolina). These specimens were loaned from the
American Museum of Natural History (AMNH). The
micro-CT scans used to create the endocasts of the
remaining species were obtained from Morphosource
(see Table S1 for detailed information about provenance,
scanning locations, scanning parameters for all speci-
mens). Crania were primarily chosen based on neu-
rocranial completeness irrespective of sex. While sex may
be a factor influencing brain size and shape, especially
for anthropoids, the primates included in this sample
exhibit little sexual dimorphism (Aristide et al., 2015;
Fleagle, 2013). Only adults were included in this analysis,
age was determined using dental eruption. The primate
sample used for these analyses was limited to
strepsirrhines and platyrrhines. This is because the
subarcuate fossa does not scale linearly relative to brain
volume in cercopithecoids as it does in strepsirrhines and
platyrrhines and is almost completely absent, expect for a
small depression, in many catarrhines including great
apes and humans (Gannon et al., 1988).
Virtual endocasts were produced using the x-ray CT
scanned images of each cranium. Segmentation of the
endocrania was performed in AVIZO
®
9.1.1 software
(Visualization and Sciences Group, 1995–2020)usinga
WACOM Cintiq 21UX tablet. The virtual endocasts were
manually segmented using the Avizo program by
4LANG ET AL.

TABLE 1 List of species and associated petrosal lobule volumes (PLV), total endocranial volumes (ECV), petrosal lobule percentage of endocranial volume (%PLV), body mass estimates
(BM), and ecological categories for locomotor behavior, activity pattern, and diet
Order Species Catalog # ECV (mm
3
) PLV (mm
3
) % PLV BM (g) Locomotion Activity pattern Diet
Lagomorpha Sylvilagus bachmani MVZ 228957 5,967.82 126.86 2.13 973.33 Saltatorial Crepuscular Folivore
Sylvilagus audubonii MVZ 43914 7,688.30 158.17 2.06 715.40 Saltatorial Crepuscular Generalist herbivore
Sylvilagus floridanus NCSM 8490 8,456.42 199.27 2.36 1,741.87 Saltatorial Crepuscular Generalist herbivore
Oryctolagus cuniculus AMNH 34816 9,156.09 207.03 2.26 1,796.07 Saltatorial Nocturnal Folivore
Lepus townsendii MVZ 20773 13,923.30 303.30 2.18 2,265.02 Saltatorial Nocturnal Generalist herbivore
Lepus arcticus AMNH 42139 15,679.80 270.10 1.72 4,003.10 Saltatorial Crepuscular Generalist herbivore
Lepus alleni MVZ 76201 15,945.70 275.22 1.73 2,183.01 Cursorial Crepuscular Generalist herbivore
Lepus californicus NCSM 8305 14,218.60 325.40 2.29 1955.70 Cursorial Nocturnal Generalist herbivore
Lepus americanus AMNH97648 10,020.40 201.00 2.01 998.68 Saltatorial Nocturnal Generalist herbivore
Brachylagus idahoensis AMNH 92869 4,992.15 153.04 3.07 339.56 Terrestrial Crepuscular Folivore
Poelagus marjorita AMNH 51052 11,597.13 210.83 1.82 2,480.24 Saltatorial Nocturnal Generalist herbivore
Ochotona pallasi AMNH 59712 2,078.92 59.32 2.85 223.85 Terrestrial Cathemeral Generalist herbivore
Ochotona hyperborea MVZ 183687 1,992.21 57.11 2.87 118.87 Terrestrial Diurnal Generalist herbivore
Ochotona princeps AMNH 120698 2,195.74 74.74 3.40 202.92 Terrestrial Diurnal Generalist herbivore
Ochotona collaris MVZ 183686 2,412.48 76.33 3.16 128.78 Terrestrial Diurnal Generalist herbivore
Rodentia Graphiurus platyops AMNH 88170 951.69 8.89 0.93 42.61 Scansorial Nocturnal Omnivore
Dryomys nitedula AMNH 206584 940.03 14.30 1.52 29.99 Arboreal Nocturnal Faunivore
Glis glis AMNH 160904 1,895.41 29.79 1.57 105.92 Arboreal Nocturnal Generalist herbivore
Aplodontia rufa AMNH 42389 7,892.50 64.41 0.82 1,470.15 Fossorial Nocturnal Generalist herbivore
Ratufa affinis USNM 488104 12,313.70 233.00 1.89 1,070.19 Arboreal Diurnal Omnivore
Xerus rutilus AMNH 179092 6,001.71 113.01 1.88 352.65 Terrestrial Diurnal Granivore
Paraxerus cepapi USNM 367956 3,053.30 44.66 1.46 137.67 Scansorial Cathemeral Omnivore
Heliosciurus rufobrachium USNM 378091 6,076.64 113.61 1.87 353.72 Arboreal Diurnal Omnivore
Protoxerus stangeri USNM 435027 6,076.64 113.61 1.87 764.37 Arboreal Diurnal Omnivore
Funisciurus pyrropus USNM 435043 4,554.13 99.54 2.19 300.09 Scansorial Crepuscular Omnivore
Tamias minimus USNM 298500 1,521.38 34.80 2.29 36.94 Scansorial Diurnal Omnivore
Urocitellus richardsonii AMNH 146619 3,157.99 55.33 1.75 245.78 Terrestrial Diurnal Generalist herbivore
Cynomys ludovicianus AMNH114522 7,220.30 70.97 0.98 938.70 Terrestrial Diurnal Generalist herbivore
Marmota marmota AMNH 15062 15,195.80 127.40 0.84 4,546.97 Terrestrial Diurnal Generalist herbivore
Dremomys rufigenis USNM 488602 5,866.09 125.71 2.14 416.93 Scansorial Diurnal Omnivore
(Continues)
LANG ET AL.5

TABLE 1 (Continued)
Order Species Catalog # ECV (mm
3
) PLV (mm
3
) % PLV BM (g) Locomotion Activity pattern Diet
Callosciurus caniceps USNM 294865 7,007.94 123.91 1.77 435.77 Arboreal Diurnal Frugivore
Rhinosciurus laticaudatus USNM 488511 4,383.60 97.36 2.22 505.54 Terrestrial Diurnal Omnivore
Lariscus insignis USNM 488570 4,878.57 113.00 2.32 323.56 Terrestrial Diurnal Omnivore
Sciurus carolinensis AMNH 258346 8,052.59 163.26 2.03 590.38 Arboreal Crepuscular Granivore
Sciurus granatensis USNM 441999 6,323.66 131.63 2.08 335.80 Arboreal Diurnal Granivore
Tamiasciurus hudsonicus USNM 549146 5,146.70 120.02 2.33 255.72 Arboreal Diurnal Granivore
Aeromys tephromelas USNM 481190 11,461.50 166.10 1.45 901.19 Gliding Nocturnal Folivore
Pteromyscus pulverulentus UMNH 481178 3,621.04 49.50 1.37 194.77 Gliding Nocturnal Frugivore
Pteromys volans USNM 172622 2,330.80 37.53 1.61 1,070.19 Gliding Nocturnal Folivore
Petaurista petaurista UMNH 589079 12,317.70 199.90 1.62 1,092.47 Gliding Nocturnal Folivore
Glaucomys volans AMNH 240290 2010.43 33.73 1.68 63.76 Gliding Nocturnal Omnivore
Hylopetes spadiceus UNMH 488639 2,120.74 17.93 0.85 83.95 Gliding Nocturnal Generalist herbivore
Laonastes aenigmamus HL KY213 5,578.76 65.68 1.18 885.09 Terrestrial Nocturnal Generalist herbivore
Lagostomus maximus AMNH 41523 13,326.70 17.60 0.13 4,155.91 Terrestrial Nocturnal Folivore
Chinchilla lanigera AMNH 180038 5,602.34 100.19 1.79 725.02 Terrestrial Nocturnal Folivore
Capromys pilorides UMZC E 3371 13,731.90 6.20 0.05 4,674.27 Terrestrial Diurnal Generalist herbivore
Mesomys hispidus AMNH 80434 2,897.73 29.22 1.01 191.53 Arboreal Nocturnal Omnivore
Pattonomys semivillosus AMNH 96763 5,625.19 40.02 0.71 792.95 Arboreal Nocturnal Frugivore
Myocastor coypus AMNH 93320 18,039.60 70.70 0.39 7,569.19 Semi-aquatic Nocturnal Generalist herbivore
Hoplomys gymnurus AMNH 29548 4,928.85 29.41 0.60 689.41 Terrestrial Nocturnal Frugivore
Ctenomys pearsoni AMNH 206517 2,421.58 4.65 0.19 196.67 Fossorial Diurnal Folivore
Octodon degus AMNH 242477 2,135.45 22.68 1.06 104.07 Terrestrial Diurnal Generalist herbivore
Spalacopus cyanus AMNH 33277 1,665.01 7.85 0.47 99.41 Fossorial Diurnal Generalist herbivore
Coendou prehensilis UF 14899 22,473.00 5.60 0.02 3,113.37 Arboreal Nocturnal Generalist herbivore
Myoprocta acouchy AMNH 94073 12,932.70 73.00 0.56 2,189.32 Terrestrial Diurnal Granivore
Dasyprocta punctata AMNH 14179 21,968.30 168.80 0.77 4,772.96 Terrestrial Diurnal Frugivore
Cavia porcellus HACB CP3 5,729.85 25.91 0.45 1,147.76 Terrestrial Nocturnal Granivore
Cryptomys hottentotus AMNH 219061 1,284.96 1.95 0.15 71.90 Fossorial Nocturnal Generalist herbivore
Bathyergus suillus AMNH 168285 3,826.55 8.91 0.23 776.39 Fossorial Fossorial Generalist herbivore
Dipodomys agilis AMNH 68717 1,239.04 12.24 0.99 68.92 Saltatorial Nocturnal Generalist herbivore
Perognathus flavescens AMNH 104526 340.53 0.49 0.14 9.77 Terrestrial Nocturnal Granivore
6LANG ET AL.

TABLE 1 (Continued)
Order Species Catalog # ECV (mm
3
) PLV (mm
3
) % PLV BM (g) Locomotion Activity pattern Diet
Castor fiber AMNH 244285 44,663.90 390.10 0.87 11,561.39 Semi-aquatic Nocturnal Folivore
Castor canadensis AMNH 258793 42,745.50 427.10 1.00 10,347.60 Semi-aquatic Nocturnal Folivore
Pedetes capensis AMNH 168880 19,544.40 100.60 0.51 2,627.64 Saltatorial Nocturnal Folivore
Idiurus macrotis AMNH 239579 926.66 12.90 1.39 20.61 Gliding Nocturnal Folivore
Anomalurus derbianus AMNH 116551 6,150.99 106.35 1.73 332.45 Gliding Nocturnal Folivore
Allactaga sibirica AMNH 58863 2,040.04 35.56 1.74 86.36 Saltatorial Nocturnal Faunivore
Dipus sagitta UMZC E.3165 2,645.07 29.41 1.11 90.75 Saltatorial Nocturnal Granivore
Jaculus jaculus AMNH 184987 1,803.63 22.43 1.24 54.06 Saltatorial Nocturnal Generalist herbivore
Cannomys badius UMZC E 2850 2,771.32 17.67 0.64 199.46 Fossorial Fossorial Generalist herbivore
Rhizomys pruinosus AMNH 112998 6,862.46 43.20 0.63 841.66 Fossorial Fossorial Generalist herbivore
Rhizomys sumatrensis AMNH 250025 5,284.54 12.45 0.24 343.36 Fossorial Fossorial Generalist herbivore
Tachyoryctes splendens AMNH 269633 2,493.87 12.95 0.52 174.29 Fossorial Nocturnal Generalist herbivore
Ondatra zibethicus AMNH 270062 6,266.19 53.33 0.85 845.26 Semi-aquatic Crepuscular Generalist herbivore
Ellobius talpinus AMNH 97838 984.56 4.28 0.43 32.44 Fossorial Fossorial Generalist herbivore
Cricetus cricetus AMNH 176484 1993.82 22.32 1.12 141.56 Terrestrial Nocturnal Omnivore
Ochrotomys nuttalli AMNH 215371 658.86 8.96 1.36 22.82 Scansorial Crepuscular Granivore
Peromyscus maniculatus AMNH 140807 646.31 10.06 1.56 23.45 Scansorial Nocturnal Omnivore
Tatera indica AMNH 250026 1947.40 11.18 0.57 142.78 Terrestrial Nocturnal Omnivore
Psammomys obesus AMNH 203215 1,430.40 3.53 0.25 130.76 Terrestrial Diurnal Generalist herbivore
Gerbillus watersi UMZC E.1971 493.05 3.25 0.66 16.80 Terrestrial Nocturnal Granivore
Acomys cahirinus AMNH 241326 686.18 8.52 1.24 30.11 Terrestrial Nocturnal Omnivore
Acomys russatus AMNH 238096 841.77 11.88 1.41 42.33 Terrestrial Diurnal Omnivore
Rattus rattus AMNH 275420 1980.85 16.32 0.82 136.35 Scansorial Nocturnal Omnivore
Cricetomys gambianus UMZC E 2261 6,599.58 48.69 0.74 1,308.99 Terrestrial Nocturnal Omnivore
Dendromus insignis AMNH 181028 437.36 4.51 1.03 9.77 Scansorial Cathemeral Granivore
Primates Nycticebus pygmaeus MCZ 6035 7,125.36 29.72 0.42 356.99 Arboreal Nocturnal Gummivore
Loris tardigradus BAA 0006 6,411.77 33.91 0.53 343.92 Slow Arborealist Nocturnal Faunivore
Galago senegalensis MCZ 44134 4,215.87 33.60 0.80 188.72 Arboreal Nocturnal Omnivore
Otolemur crassicaudatus AMNH 80800 13,277.00 69.10 0.52 1,628.34 Arboreal Nocturnal Gummivore
Galagoides demidovii AMNH 50984 2,401.69 21.82 0.91 72.68 Arboreal Nocturnal Faunivore
Euoticus elegantulus MCZ 14658 6,797.01 60.82 0.89 312.53 Arboreal Nocturnal Gummivore
(Continues)
LANG ET AL.7

TABLE 1 (Continued)
Order Species Catalog # ECV (mm
3
) PLV (mm
3
) % PLV BM (g) Locomotion Activity pattern Diet
Perodicticus potto MCZ 42622 15,389.30 101.30 0.66 972.19 Slow Arborealist Nocturnal Omnivore
Arctocebus aureus YPM 014401 8,291.01 36.06 0.43 486.75 Slow Arborealist Nocturnal Faunivore
Daubentonia madagascariensis AMNH 100632 42,309.70 435.90 1.03 2,166.51 Slow Arborealist Nocturnal Omnivore
Cheirogaleus major AMNH 100640 6,342.91 71.28 1.12 507.39 Arboreal Nocturnal Omnivore
Mirza coquereli DPC 0137 6,576.32 50.86 0.77 345.21 Arboreal Nocturnal Omnivore
Microcebus murinus MCZ 45125 1,395.93 18.00 1.29 61.47 Arboreal Nocturnal Omnivore
Lepilemur mustelinus AMNH 170558 7,603.27 88.52 1.16 375.63 Slow Arborealist Nocturnal Folivore
Lemur catta MCZ 44903 24,563.00 205.70 0.84 2,066.28 Scansorial Cathemeral Frugivore
Hapalemur griseus MCZ 44911 13,700.70 146.30 1.07 884.88 Arboreal Cathemeral Folivore
Eulemur fulvus MCZ 44896 25,991.00 193.50 0.74 3,132.42 Arboreal Cathemeral Frugivore
Varecia variegata MCZ 44905 34,771.40 324.10 0.93 5,946.82 Arboreal Diurnal Frugivore
Indri indri AMNH 100506 37,127.60 367.20 0.99 5,039.75 Arboreal Diurnal Folivore
Propithecus diadema MCZ 5016 38,825.10 439.00 1.13 2,869.45 Arboreal Diurnal Frugivore
Avahi laniger MCZ32503 9,308.34 106.56 1.14 341.09 Arboreal Nocturnal Folivore
Tarsius tarsier USNM 200279 2,907.57 25.39 0.87 98.60 Arboreal Nocturnal Faunivore
Tarsius syrichta USNM282761 4,121.58 33.56 0.81 139.31 Arboreal Nocturnal Faunivore
Pithecia pithecia MCZ 31061 29,878.00 221.50 0.74 1,669.48 Arboreal Diurnal Frugivore
Chiropotes satanas MCZ 6028 55,600.80 492.50 0.89 2,326.18 Arboreal Diurnal Frugivore
Cacajao calvus MCZ 1957 68,491.80 394.00 0.58 3,532.10 Arboreal Diurnal Frugivore
Callicebus moloch MCZ 26922 18,810.30 61.10 0.32 802.79 Arboreal Diurnal Frugivore
Cebuella pygmaea USNM 337324 3,679.19 16.86 0.46 47.44 Arboreal Diurnal Omnivore
Mico argentatus MCZ 30579 7,930.20 39.44 0.50 191.33 Arboreal Diurnal Omnivore
Callimico goeldii AMNH 98367 12,866.10 41.90 0.33 285.30 Arboreal Diurnal Omnivore
Leontopithecus rosalia AMNH 235274 11,863.20 30.80 0.26 493.02 Arboreal Diurnal Omnivore
Leontocebus fuscicollis AMNH 74042 9,421.15 41.77 0.44 255.63 Arboreal Diurnal Frugivore
Sapajus apella MCZ 41090 68,811.50 151.70 0.22 3,694.71 Arboreal Diurnal Omnivore
Cebus capucinus MCZ 34326 68,488.50 203.00 0.30 4,366.25 Arboreal Diurnal Omnivore
Saimiri sciureus MCZ 30568 20,353.50 71.00 0.35 581.02 Arboreal Diurnal Omnivore
Aotus trivirgatus MCZ 19802 15,265.70 80.20 0.53 684.15 Slow Arborealist Nocturnal Frugivore
Ateles geoffroyi MCZ BOM-5346 106,133.00 893.00 0.84 7,214.74 Arboreal Diurnal Frugivore
Lagothrix lagotricha USNM 194342 101,254.00 290.00 0.29 5,227.72 Arboreal Diurnal Frugivore
Alouatta palliata MCZ 6001 47,553.30 97.70 0.21 4,641.89 Arboreal Diurnal Folivore
8LANG ET AL.

TABLE 1 (Continued)
Order Species Catalog # ECV (mm
3
) PLV (mm
3
) % PLV BM (g) Locomotion Activity pattern Diet
Dermoptera Galeopterus variegatus AMNH 102703 5,897.58 49.40 0.84 1,100.00 Gliding Nocturnal Folivore
Cynocephalus volans AMNH 16697 5,869.34 18.71 0.32 1,500.00 Gliding Nocturnal Folivore
Scandentia Ptilocercus lowii USNM 481103 1,645.13 16.72 1.02 45.30 Arboreal Nocturnal Faunivore
Dendrogale murina FMNH 46629 1,405.24 20.69 1.47 45.00 Arboreal Diurnal Faunivore
Tupaia belangeri USNM 320655 3,854.83 48.23 1.25 145.40 Terrestrial Diurnal Faunivore
Tupaia glis USNM 311311 3,444.91 40.03 1.16 152.20 Scansorial Diurnal Faunivore
Tupaia gracilis AMNH 103620 2,215.67 36.38 1.64 67.10 Terrestrial Diurnal Faunivore
Tupaia picta UMZC E4063A 3,336.63 43.38 1.30 34.30 Scansorial Diurnal Faunivore
Tupaia montana FMNH 108831 2,810.09 37.26 1.33 126.00 Terrestrial Diurnal Faunivore
Tupaia dorsalis AMNH 1103892 2,465.00 30.21 1.23 26.50 Terrestrial Diurnal Faunivore
Tupaia palawanensis FMNH 62948 2,950.49 33.56 1.14 138.50 Terrestrial Diurnal Faunivore
Tupaia tana AMNH 102829 3,712.24 55.39 1.49 205.30 Terrestrial Diurnal Faunivore
Tupaia minor AMNH 103906 1755.63 21.25 1.21 54.40 Arboreal Diurnal Faunivore
Tupaia javanica AMNH 101672 2,289.09 32.52 1.42 87.80 Scansorial Diurnal Faunivore
Urogale everetti FMNH 61419 4,530.12 53.54 1.18 224.00 Terrestrial Diurnal Omnivore
Dendrogale melanura FMNH 108854 1,480.2 20.08 1.36 49.5 Terrestrial Diurnal Faunivore
Abbreviations: AMNH, American Museum of Natural History; DPC, Duke Primate Center; FMNH, Field Museum of Natural History; MCZ, Museum of Comparative Zoology; MVZ, Museum of Vertebrate Zoology;
NCSM, National Council of Science Museums; UF, Florida Museum of Natural History; UMZC, University Museum of Zoology, Cambridge; USNM, Smithsonian National Museum of Natural History.
LANG ET AL.9

“closing”the endocranial area of a given slice
(i.e., tracing a straight line between two bones to sepa-
rate the endocast from openings for the passage of ves-
sels and/or nerves entering or exiting the endocranium).
When the number of slices in the frontal plane exceeded
1800, one of two methods were used: (a) the interpolate
tool was used to fill in the endocranial area between two
completed slices, with one to four slices separating
them; (b) the file size was reduced using ImageJ
(Rasband, 1997-2018) to create a version of the dataset
that included every other slice (in the frontal plane),
thereby increasing the interslice distance in this dimen-
sion by a factor of two (see Table S1). These methods
were used to decrease segmentation time and loading
times of large data files without sacrificing the quality of
the endocranial reconstruction.
The entire virtual endocast, including the petrosal
lobules, was segmented in the frontal plane. The lob-
ules were then segmented independently in the trans-
verse plane, as it is easier to determine the point at
which they can be most accurately isolated from the
rest of the endocast in this plane. Petrosal lobules
were separated from the rest of the endocast by draw-
ing a straight line across the narrowest point at the
opening of the subarcuate fossa, manually closing off
the fossa from the rest of the endocast (Figure 4). In
some sections, identifying the narrowest point was not
possible. In these cases, the petrosal lobules were iso-
lated by tracing a straight line between the edges of
the anterior semicircular canal, which occurs on
either side of the fossa (Figure 4). This was done to
ensure the most consistent segmentation across taxa,
accounting for variation in the configuration of the
subarcuate fossa. Once isolated, the left and right
petrosal lobules were each assigned a distinct label-
field module. Volumes (mm
3
)werecalculatedin
Avizo using the surface area volume module, which
calculates the volume enclosed in a given surface area
for both the left and right petrosal lobules and for the
entire endocast. The left and right petrosal lobule vol-
umes were then summed to form the total volume of
the lobules. Body masses for each species were derived
from order or suborder specific equations using cra-
nial measurements (Table 2). No equations have been
published to estimate body mass for scandentians or
dermopterans. Consequently, species mean body
masses from Sargis (2002)wereusedforScandentia.
The body masses for Tupaia javanica,andTupaia
belangeri were provided by EJ Sargis (personal com-
munication). Body masses for the dermopterans were
obtained from Stafford and Szalay (2000). All cranial
measurements were taken digitally using the Avizo
3D measurement tool.
2.1 |Ecological categorization
Species were assigned to primary ecological categories
based on their behaviors for activity period, locomotor
behavior, and diet (Table 1; see S2 for sources for ecologi-
cal behaviors). For each primary ecological category, spe-
cies were then placed into high, medium, and low
clustered ecological categories based on the hypothesized
demands their ecological behaviors present to the visual
and vestibular systems. It was necessary to use this
approach to reduce the number of groups being com-
pared to make it possible to run an Ornstein–Uhlenbeck
model (described below), which requires there to be more
datapoints (taxa) within each category than there are cat-
egories (Butler & King, 2004). Regarding activity pattern,
diurnal taxa (n=56) were placed in the high cluster, cre-
puscular (n=10) and cathemeral (n=6) taxa were
placed in the medium cluster, and nocturnal (n=63)
and highly fossorial (n=5) taxa were placed in the low
cluster. These subcategories were created based on the
observation that diurnality is more often associated with
visual specialization both optically (Veilleux &
Kirk, 2014) and neurally (Barton, 1996,1998; Campi &
Krubitzer, 2010; Kirk, 2006) and as a result, these taxa
may require the greater image stabilization afforded by
larger petrosal lobules than taxa which operate in lower
light conditions.
High, medium, and low clusters were also applied to
locomotor modes and dietary primary ecological catego-
ries. For locomotion, arboreal (n=46) and gliding
(n=10) taxa were placed in the high locomotor cluster,
as they navigate through complex three-dimensional,
dynamic, and pliant substrates, which present challenges
to the visual and vestibular systems. Saltatorial (n=13),
scansorial (n=13), semi-aquatic (n=4), and slow arbo-
real (n=6) taxa were sorted into the medium locomotor
cluster as they engage in behaviors (i.e., bounding,
climbing, swimming), which present some challenge to
the visual and vestibular systems, but unlike the arboreal
and gliding species, they engage in frequent terrestrial,
two-dimensional locomotion, or slow deliberate move-
ment which may require fewer visual and vestibular
adjustments (Spoor et al., 2007). Finally, terrestrial
(n=36) and fossorial (n=10), taxa were placed in the
low locomotor cluster, as they engage in locomotor
behaviors, which present less of a stabilization challenge
to the visual and vestibular systems relative to the other
clusters. Although cursorial taxa (n=2) are terrestrial,
due to their rapid movements and sudden changes in
redirection they were placed in the high locomotor clus-
ter with gliding and arboreal taxa.
Lastly, species were sorted into three dietary clusters
based on the potential complexity of information
10 LANG ET AL.

presented to the visual and vestibular systems involved in
foraging or procuring each food item. Faunivorous
(n=20) and omnivorous (n=33) taxa were sorted into
the high dietary cluster, as capturing animal prey is often
associated with visual specializations including higher
visual acuity (Veilleux & Kirk, 2014) and greater orbital
FIGURE 3 Endocasts for primates (purple), rodents (yellow), lagomorphs (green), scandentians (blue), and dermopterans (orange) with
petrosal lobules (red). (a) Saimiri sciureus (MCZ 30568), (b) Euoticus elegantulus (MCZ 14658), (c) Tarsius tarsier (USNM 200279), (d)
Dryomys nitedula (AMNH 206584), (e) Idiurus macrotis (AMNH 239576), (f) Rhizomys sumatrensis (AMNH 250025), (g) Myoprocta acouchy
(AMNH 94073), (h) Lepus arcticus (AMNH 73602), (i) Poelagus marjorita (AMNH 118569), (j) Ochotona hyperborea (MVZ 18367), (k) Tupaia
javanica (AMNH 101672), l) Tupaia gracilis (AMNH103620), (m) Cynocephalus volans (AMNH 16697), (n) Galeopterus variegatus (AMNH
102703); scale =10 mm
LANG ET AL.11

convergence (Barton, 2004; Heesy, 2008; Ross, 1995; Ross
et al., 2007), which are dependent on precise control of
eye movements. Frugivorous (n=16), granivorous
(n=11), and gummivorous (n=3) taxa were placed in
the medium dietary cluster. One of the challenges shared
by these three ecological behaviors is that their food
sources are unevenly distributed over a large area (Mace
et al., 1981; Melin et al., 2014) and are often dispersed
among foliage (Melin et al., 2014). Analysis of the dietary
habits of gummivorous primates indicate that feeding
occurs on select species, which may be patchily distrib-
uted (Nash, 1986). These foraging challenges also apply
to frugivorous foraging, challenges that are frequently
cited as one of the driving forces in the evolution of color
vision among frugivorous primates (e.g., Melin
et al., 2014; Regan et al., 2001). Fewer analyses have been
conducted on the specific foraging challenges associated
with seed predation, but granivory is akin to frugivory as
seed production is not continuous and is also distinct
from folivory as seeds are inconspicuous once dispersed
(Janzen, 1971). These ecological foraging behaviors
would not be expected to be as visually or vestibularly
demanding as omnivory or faunivory, but procurement
of these food sources represents more of a challenge in
terms of visual location of a given food item when com-
pared to folivores (n=22), and generalist herbivores
(n=35), which were placed in the low dietary cluster as
the relevant food sources are more ubiquitous and readily
attainable (Mace et al., 1981).
2.2 |Data analysis
All statistical analyses were performed in R 4.0.2 (R Core
Team, 2020). Quantitative variables, including petrosal
lobule volume (mm
3
), endocranial volume minus petro-
sal lobule volume (mm
3
) (hereafter referred to as
“adjusted endocranial volume”or AEV), petrosal lobule
mass (g), and body mass (g) minus petrosal lobules mass
(g) (hereafter referred to as “adjusted body mass”or
FIGURE 4 Cross-section of the (a) cranium and (b) segmentation of endocast (blue) of Perodicticus potto (MCZ 42622) in the transverse
plane showing the petrosal lobule (red) isolation method. The right petrosal lobule shows isolation at the narrowest point of the entrance to
the subarcuate fossa and the left petrosal lobule shows isolation drawn across edges of the anterior semicircular canal
TABLE 2 Regression equations and measurements used to estimate body masses by taxonomic group with sources
Source Taxon Measurement (mm) Unit Slope y-int pr
2
Silcox et al. (2009) Prosimian Prosthion inion length kg 3.79 6.92 <.0001 .93
Martin (1990) Simian Maximal skull length g 3.89 4.09 —.98
Bertrand et al. (2016) Rodentia Maximal skull length g 3.95 4.23 —.96
Moncunill-Solé et al. (2015) Lagomorpha Maximal occipital condyle width g 4.09 1.526 .000 .96
Note: Log 10 applied to all raw data.
12 LANG ET AL.

ABM), were log transformed prior to statistical analysis.
The independent variables, AEV and ABM, were used to
avoid comparing lobule size to itself in regression ana-
lyses. Petrosal lobule volumes in cubic millimetre were
multiplied by 1.036 to convert them to mass using the tis-
sue conversion from Stephan et al. (1981). Values were
then converted to grams and used in analyses of petrosal
lobule mass and ABM.
Both ordinary least squares (OLS) (package: RRPP,
version: 0.6.1, function: lm.rrpp, Collyer & Adams, 2018)
and phylogenetic generalized least-square (PGLS) simul-
taneously estimated with phylogenetic signal in the resid-
ual error as Pagel's lambda (package: phylolm, function:
phylolm, version: 2.6.2, Ho et al., 2016) analyses were
performed to evaluate the scaling patterns of the petrosal
lobules relative to endocranial volume and body mass
within the sample of euarchontoglirans. Both body mass
and endocranial volume were included as independent
variables. Body mass was included as some taxa are
known to have large brains relative to body mass, which
may affect the size of the lobules relative to endocranial
volume if the lobules are not also correspondingly large
(i.e., if brain expansion occurred in other regions but not
the lobules). For instance, anthropoids have exceptionally
large brains for their body masses (Jerison, 1973), which
has been argued to be primarily related to neocortical
expansion (Kaas, 2012), and so would not necessarily be
reflected in the size of the lobules.
Trees used for phylogenetic analyses were generated
from the online vertebrate phylogeny database, vertlife.
org (2020), using the subsetting tool. Mammalian phylog-
enies from this resource were produced using a “back-
bone-and-patch”Bayesian method from genetic and
fossil data (Upham et al., 2019a,2019b). For this sample
of 140 euarchontoglirans, 10,000 fossil calibrated node-
dated credible trees were generated and downloaded.
From this posterior distribution of phylogenies, a single
tree was obtained using maximum clade credibility func-
tion (maxCladeCred) from the R package phangorn (ver-
sion: 2.5.5; Schliep, 2011).
Ordinal scaling patterns were examined using the
residuals from the OLS regression analysis of petrosal
lobule volume plotted against AEV and petrosal lobule
mass plotted against ABM. An analysis of variance
(ANOVA) using a randomized residual permutation pro-
cedure (rrpp) with 1,000 iterations was performed to
examine significant differences among orders. An addi-
tional ANOVA was performed with primates divided into
the suborders (Haplorhini and Strepsirrhini) to assess dif-
ferences in relative lobule size within Primates, given
that anthropoids tend to have larger brains relative to
body mass than their strepsirrhine relatives. The PGLS
models were run using Pagel's λcorrelation structure
(Freckleton et al., 2002) for Euarchontoglires overall, and
rodents and primates separately (package: phylolm, ver-
sion: 2.6.2, Ho et al., 2016). The phylogenetic signal of
the residuals obtained from the relationship between
petrosal lobule volume and AEV and the relationship
between petrosal lobule mass and ABM was assessed
using Pagel's λ(Pagel, 1999; Freckleton et al., 2002).
To evaluate whether relative lobule size differed
between ecological groups, a phylogenetic ANOVA (pAN-
OVA) was performed on residuals from PGLS analyses of
petrosal lobule volume and AEV, and petrosal lobules
mass and ABM (package: RRPP, version: 0.6.1, functions:
lm.rrpp, ANOVA; Collyer & Adams, 2018) according to
the primary ecological categories for each ecological group
(i.e., locomotor behaviors, activity pattern, diet) using a
Brownian motion covariance matrix. This analysis was
also repeated on rodents and primates separately given the
differences between the two orders in terms of the func-
tional components of the brain which comprise the lobules
and in light of the extreme ecological diversity of rodents,
factors that may produce distinct ecological scaling pat-
terns in each order. For the ANOVAs which identified sig-
nificant differences between ordinal and ecological
categories, a post hoc pairwise comparison was performed
using the function pairwise from the RRPP R package
(Collyer & Adams, 2018) to examine the significance of
differences among specific groups. The hypothesis that
locomotor behavior, activity pattern, and diet influenced
the evolution of relative petrosal lobule size was tested
using generalized evolutionary models (Butler &
King, 2004;Hansen,1997). The first model fit was a
single-rate Brownian motion model (BM1) that models
how the variance of relative petrosal lobule size accumu-
lates proportionally to evolutionary time under a random
walk. The second model fit was a single-peak Ornstein–
Uhlenbeck model (OU1) that constrains relative petrosal
lobule size to evolve toward one optimum. The BM1 and
OU1 models serve as null hypotheses that relative petrosal
lobule size does not differ between ecological groups.
The last three models fit were multipeak OUM that
allowed locomotor behavior (OUMloc), activity pattern
(OUMact), and diet (OUMdiet) to exhibit different adap-
tive optima. All models were fit using the OUwie function
in the R package OUwie (version: 2.6; Beaulieu
et al., 2012) across 1,000 stochastically mapped trees to
take into account uncertainty in phylogenetic topology
and the ancestral character states. The high, medium, and
low clusters (as described above) were used to examine
ecology related changes along evolutionary lineages based
on the evidence that these clusters are associated with
greater challenges to the visual and vestibular system.
Within this modeling analysis, it is expected that
species in the high cluster groups will have the largest
LANG ET AL.13

lobules, and therefore greater capacity for image stabili-
zation, and that species in the low cluster groups will
have the smallest lobules. The evolution of locomotor
behavior, activity pattern, and dietary clusters are
inferred by performing stochastic character mapping
with symmetric transition rates between regimes
(Bollback, 2006; Huelsenbeck et al., 2004; Nielsen, 2002)
using the make.simmap function in the phytools R
package (version: 0.7.47; Revell, 2012). Ten stochastic
character maps across 100 tree topologies randomly
drawn from the posterior distribution of trees were sim-
ulated (Upham et al., 2019a,2019b), resulting in 1,000
character maps for each set of locomotor behavior,
activity pattern, and dietary regimes. Relative support
for each of the five models was assessed through compu-
tation of small sample corrected Akaike weights
(AICcW). All models with ΔAICc < 2 were considered
to be supported by the data (Burnham &
Anderson, 2004). Using the residuals from regression
analyses for lobule volume/mass plotted against AEV
and ABM, ancestral character states for lobule size were
estimated using maximum likelihood with the fastAnc
function from the phytools package (version: 0.7.47;
Revell, 2012) as a heuristic tool to help visualize differ-
ences in relative lobule size across lineages.
Phylogenetic ANOVAs were also run on log
endocranial volume and log body mass to evaluate the
possible relationship between the ecological variables
(locomotor behavior, diet, and activity pattern) and total
endocranial volume and body mass respectively which
may influence the results of petrosal lobule analysis
(package: RRPP, version: 0.6.1, functions: lm.rrpp, anova;
Collyer & Adams, 2018).
3|RESULTS
Volumetric analysis indicates that on average petrosal lob-
ules occupy 1.15% of the total endocranial volume. Within
this sample, Coendou prehensilis, (a nocturnal, arboreal
caviomorph rodent) has the relatively smallest petrosal
lobules, at 0.02% endocranial volume and Ochotona
princeps (a diurnal, terrestrial lagomorph) has the rela-
tively largest at 3.4% endocranial volume. Both OLS and
PGLS regression of volumetric data (petrosal lobule vol-
ume plotted against AEV) indicate a significant positive
linear relationship between the two variables, with AEV
accounting for 61.5% (p=.001) and 60% (p=<.000) of
the variation in petrosal lobule volume, respectively
(Table 3). Similar results were obtained from the mass
OLS and PGLS regression analyses (petrosal lobule mass
plotted against ABM) in which there is a significant posi-
tive linear relationship between the two variables
(Table 3). However, ABM accounted for less of the varia-
tion in petrosal lobule mass at 52.4% (p=<.000) for OLS
regression and 48.2% (p=<.000) for PGLS regression.
ANOVA of residuals from the OLS regression of
petrosal lobule volume and AEV identified significant dif-
ferences among ordinal groups (Table 4). Specifically, the
post hoc pairwise tests indicate that lagomorphs have rel-
atively larger petrosal lobule volumes compared to all
other orders (Table 5; Figure 5a,b), ranging between
TABLE 3 Regression statistics from petrosal lobule analyses of log
10
petrosal lobule volume (mm
3
) plotted against log
10
adjusted
endocranial volume (mm
3
) (PLV–AEV) and log
10
petrosal lobule mass (g) plotted against log
10
adjusted body mass (g) (PLM–ABM) using
PGLS (Pagle's λ) and ordinary least squares (OLS) linear modeling for Euarchontoglires (n=140), Rodentia (n=71), and Primates (n=38)
separately
Data Models Slope Intercept b r
2
sig. (p)λAIC LogLik
Euarchontoglires PLV–AEV Pagel's λ0.845 1.494 .600 .000 0.966 3.137 2.432
OLS 0.842 1.450 .615 .001 NA NA NA
PLM–ABM Pagel's λ0.466 2.546 .482 .000 0.949 40.760 16.380
OLS 0.575 2.776 .524 .001 NA NA NA
Rodentia PLV–AEV Pagel's λ0.835 1.524 .524 .000 0.777 47.060 19.530
OLS 0.907 1.734 .545 .001 NA NA NA
PLM–ABM Pagel's λ0.458 2.704 .424 .000 0.861 59.450 25.720
OLS 0.497 2.741 .400 .001 NA NA NA
Primates PLV–AEV Pagel's λ0.920 1.843 .829 .000 0.761 20.820 14.410
OLS 0.855 1.603 .781 .001 NA NA NA
PLM–ABM Pagel's λ0.661 2.956 .792 .000 0.15 5.619 6.809
OLS 0.673 2.994 .801 .001 NA NA NA
14 LANG ET AL.

1.75% and 3.5% of total endocranial volume. ANOVA of
residuals from OLS regression of petrosal lobule mass
and ABM (Table 5) identified significant differences
between lagomorphs and both rodents and
dermopterans, with lagomorphs having relatively larger
lobules. Rodents were significantly different from all
other ordinal groups except for dermopterans. However,
there was substantial variation in lobule size relative to
body mass in rodents (Figure 5c,d). Dermopterans were
significantly different from all other ordinal groups
(except Rodentia), possessing comparatively smaller lob-
ules relative to ABM. Significant differences were also
identified between scandentians and dermopterans, and
scandentians and rodents, with the former having com-
paratively larger lobules relative to ABM in both cases
(Figure 5c,d).
Within primates, haplorhines have significantly
smaller petrosal lobules than strepsirrhines relative to
AEV (Tables 6and 7; Figure 5a,b) but not relative to
ABM (Tables 6and 7; Figure 5c,d), although the ranges
of relative petrosal lobule volume and mass between the
two groups largely overlap. Overall, Order accounts for
22.1% (p=.001; Table 4) of the variation in the residuals
of petrosal lobule volume and AEV and 18.2% (p=.001;
Table 4) of the variation in the residuals of petrosal lob-
ule mass and ABM. As would be expected from this
result, both volumetric and mass PGLS analyses identi-
fied a strong phylogenetic signal, with Pagel's λ=0.97
(p< .001) for lobule volume relative to AEV analysis and
Pagel's λ=0.95 (p< .001) for lobule mass relative to
ABM analysis (Table 3). The ancestral state reconstruc-
tion documents these lobule scaling patterns across
the tree (Figure 6a,b). Relative to both AEV (Figure 6a)
and ABM (Figure 6b), lagomorphs, strepsirrhines,
scandentians, and sciurid rodents are reconstructed as
having evolved proportionately larger petrosal lobules.
However, the difference between the haplorhines and
strepsirrhines is less marked when considered relative
to ABM.
Phylogenetic ANOVAs of relative petrosal lobule vol-
ume did not identify significant differences among any of
the primary ecological groups in Euarchontoglires
(Table 8; Figure 7), rodents (Table 10), or primates
(Table 11). However, phylogenetic ANOVAs of relative
petrosal lobule mass identified significant differences
among locomotor categories (Table 9; Figure 7b) within
Euarchontoglires. The post hoc pairwise tests indicate
that fossorial taxa had significantly smaller relative lobule
masses compared to arboreal, semi-aquatic, slow arbo-
realist, and terrestrial locomotor groups. When pAN-
OVAs of relative lobule mass was performed on rodents
separately, no significant differences were detected
TABLE 4 Results from ANOVA of residuals from OLS regression of log
10
petrosal lobule volume (mm
3
) plotted against log
10
adjusted
endocranial volume (mm
3
) (PLV–AEV) and log
10
petrosal lobule mass (g) plotted against log
10
adjusted body mass (g) (PLM–ABM) by Order
Data Df SS Residual SS R
2
FZSig. (p)
PLV–AEV 4 3.616 12.765 .221 9.559 3.050 .001
PLM–ABM 4 3.681 16.534 .182 7.515 2.738 .002
TABLE 5 Results of post hoc tests based on ANOVA of residuals from OLS regression of log
10
petrosal lobule volume (mm
3
) plotted
against log
10
adjusted endocranial volume (mm
3
) (PLV–AEV) and log
10
petrosal lobule mass (g) plotted against log
10
adjusted body mass (g)
(PLM–ABM) by Order
Data Order Dermoptera Lagomorpha Primates Rodentia Scandentia
PLV–AEV Dermoptera —0.021 0.557 0.433 0.184
Lagomorpha 0.021 —0.001 0.001 0.014
Primates 0.557 0.001 —0.482 0.059
Rodentia 0.433 0.001 0.482 —0.133
Scandentia 0.184 0.014 0.059 0.133 —
PLM–ABM Dermoptera —0.010 0.042 0.128 0.039
Lagomorpha 0.010 —0.064 0.001 0.202
Primates 0.042 0.064 —0.007 0.825
Rodentia 0.128 0.001 0.007 —0.031
Scandentia 0.039 0.202 0.825 0.031 —
Note: Significant p-values shown in bold.
LANG ET AL.15

among any of the primary ecological groups for activity
pattern, diet, and locomotor behavior. However, the
alpha level for locomotor behavior was very close to our
cut off (.05) at p=.051 (Table 10). Post hoc tests for rela-
tive lobule mass and locomotor behavior in rodents iden-
tify significant differences between the fossorial group
and arboreal, semi-aquatic, and terrestrial groups
(Table S4). Within primates, phylogenetic ANOVAs of
relative petrosal lobule volume and mass identified no
significant differences among any of the ecological
groups (Table 11).
Across Euarchontoglires, the single peak OU1 model
(ΔAICc =0.00, W
A
=0.48) and the OUMloc model
(ΔAICc =1.94, W
A
=0.18) had the best fit for relative
petrosal lobule volume (Table 12). While the other
ecological models (i.e., OUMact, OUMdiet) also had com-
parable fits, they are just outside of the ΔAICc cut
off (ΔAICc =2.01 and 2.05, respectively; Table 12).
Concerning relative petrosal lobule mass for
Euarchontoglires, only the single peak OU1 model was
the best-fitted model (W
A
=0.95; Table 13). In rodents,
all models (except for OUMdiet) had comparable fits for
relative petrosal lobule volume (ΔAICc =0.00–1.66;
Table 12).OnlythesinglepeakOU1modelhadthebest
fit for relative petrosal lobule volume mass (W
A
=0.50;
Table 13). Although the single peak OU1 models has the
lowest ΔAICc values and is therefore best supported
among the other models for relative petrosal lobule vol-
ume in Euarchontoglires and Rodentia, the fact that all
models (except OUMdiet for rodents) had comparable fits
FIGURE 5 (a) Scatter plot of log
10
petrosal lobule volume (mm
3
) plotted against log
10
adjusted endocranial volume (mm
3
) for
140 Euarchontoglires categorized by order and suborder (Primates). (b) Boxplot of residuals from OLS regression of log
10
petrosal lobule
volume (mm3) plotted against log
10
adjusted endocranial volume (mm
3
) categorized by order and Primate suborder. (c) Scatter plot of log
10
petrosal lobule mass (g) plotted against log
10
adjusted body mass (g) for 140 Euarchontoglires categorized by order and Primate suborder. (d)
Boxplot of residuals from OLS regression of log
10
petrosal lobule mass (g) plotted against log
10
adjusted body mass (g) categorized by order
and suborder (Primates)
16 LANG ET AL.

(ΔAICc < 2) suggests that there is no single factor among
these ecological groups that can account for variation in
relative petrosal lobule volume within Euarchontoglires
overall or within Rodentia. In primates, only the single
peak OU1 model had the best fit for relative petrosal lob-
ule volume (Table 12), whereas the OUMloc and OU1
models had comparable fits for relative petrosal lobule
mass (Table 13). However, parametric bootstrapping of the
OUMloc model revealed that the 95% confidence intervals
of the theta values overlapped between high, medium, and
low clustered locomotor regimes. This indicates that
despite the support for the OUMloc model, variation
among these locomotor groups largely overlaps and it is
therefore difficult to identify an adaptive pattern associ-
ated with locomotor behaviors.
Significant differences in total endocranial volume
and body mass were identified between some of the
groups for locomotor behavior and diet (Table S3). How-
ever, because the differences in total endocranial volume
and body mass between locomotor groups are either
(a) the result of the small sample size of large bodied and
large brained semi-aquatic species or (b) involved a sig-
nificant degree of overlap between the tested ecological
groups, we do not believe that the patterns observed in
petrosal lobules size relative to adjusted endocranial vol-
ume or adjusted body mass are the result of scaling pat-
terns occurring at the endocranial volume or body mass
level.
4|DISCUSSION
OLS regression analyses indicates that there is a strong
positive correlation between petrosal lobule size, as deter-
mined from the size of the subarcuate fossa, relative to
both endocranial volume and body mass in
Euarchontoglires. Furthermore, a significant portion of
the variation in petrosal lobule size is attributed to varia-
tion in endocranial volume and body mass. Analysis of
the scaling patterns of the subarcuate fossa in primates
and other mammals including Carnivora, Rodentia,
Lagomorpha, Marsupialia, and Cetacae presented by
Gannon et al. (1988) identified a strong positive correla-
tion between subarcuate fossa volume and endocranial
volume. Although the statistical methods used by Gan-
non et al. (1988) have been challenged in more recent
TABLE 6 Results from ANOVA of residuals from OLS regression of log
10
petrosal lobule volume (mm
3
) plotted against log
10
adjusted
endocranial volume (mm
3
) (PLV–AEV) and log
10
petrosal lobule mass (g) plotted against log
10
adjusted body mass (g) (PLM–ABM) by order
and suborder for primates (Haplorhini and Strepsirrhini)
Data Df SS Residual SS R
2
FZsig. (p)
PLV–AEV 5 4.096 12.284 .250 8.937 3.296 .001
PLM–ABM 5 3.688 16.528 .182 5.980 2.703 .002
TABLE 7 Results of post hoc tests based on ANOVA of residuals from OLS regression of log
10
petrosal lobule volume (mm
3
) plotted
against log
10
adjusted endocranial volume (mm
3
) (PLV–AEV) and log
10
petrosal lobule mass (g) plotted against log
10
adjusted body mass (g)
(PLM–ABM) by Order and Suborder for Primates (Haplorhini and Strepsirrhini)
Data Order Dermoptera Haplorhini Lagomorpha Rodentia Scandentia Strepsirrhini
PLV–AEV Dermoptera —0.927 0.021 0.433 0.184 0.313
Haplorhini 0.927 —0.001 0.060 0.009 0.043
Lagomorpha 0.021 0.001 —0.001 0.014 0.001
Rodentia 0.433 0.060 0.001 —0.133 0.504
Scandentia 0.184 0.009 0.014 0.133 —0.429
Strepsirrhini 0.313 0.043 0.001 0.504 0.429 —
PLM–ABM Dermoptera —0.047 0.010 0.128 0.039 0.039
Haplorhini 0.047 —0.088 0.036 0.765 0.850
Lagomorpha 0.010 0.088 —0.001 0.202 0.119
Rodentia 0.128 0.036 0.001 —0.031 0.016
Scandentia 0.039 0.765 0.202 0.031 —0.916
Strepsirrhini 0.039 0.850 0.119 0.016 0.916 -
Note: Significant p-values shown in bold.
LANG ET AL.17

studies (i.e., S
anchez-Villagra, 2002), their results are
largely consistent with those presented here using more
rigorous statistical methods and phylogenetic controls.
The result of these analyses differ somewhat from
recent analyses performed by Ferreira-Cardoso et al.
(2017) on petrosal lobule volume from a relatively large
sample of mammals (48 species from 13 orders). In this
study, however, they assessed lobule size relative to total
endocranial volume minus lobule volume, and only con-
sidered body mass in relation to the residuals from that
FIGURE 6 Ancestral character state reconstruction of lobule size (as determined from subarcuate fossa size) using residuals from
ordinary least squares (OLS) regression analyses for lobule volume/mass plotted against (a) AEV and (b) ABM. Scale reflects residual values.
Fossorial taxa highlighted in gray. (Phylogenetic tree produced using maximum clade credibility for this sample of 140 euarchontoglirans,
based on 10,000 fossil calibrated node-dated credible trees derived from Upham et al., 2019a, 2019b)
18 LANG ET AL.

relationship. Here, lobule size was evaluated relative to
both endocranial volume and body mass separately to
help parse out differences in lobule scaling for highly
encephalized groups (i.e., the platyrrhines). Evaluating
lobule scaling relative to these variables independently
helped to identify subordinal scaling patterns between
strepsirrhines and haplorhines.
4.1 |Phylogenetic signals and ordinal
patterns
In analyses of relative petrosal lobule volume as deter-
mined from subarcuate fossa volume, lagomorphs had
significantly larger lobules compared to all other orders
(Figure 5a,b). However, relative to body mass, the petro-
sal lobules of lagomorphs largely fall within the range of
all other orders only significantly differing from
Dermoptera and Rodentia, which possess comparatively
smaller relative lobule masses (Figure 5c,d). This suggests
two possible scenarios for the evolution of the petrosal
lobules within Lagomorpha, which are not necessarily
mutually exclusive: (a) that there was an increase in the
size of the lobules relative to brain/endocranial size
within the lagomorph lineage occurring soon after their
separation from Rodentia or (b) that there was an
increase in the size of the rest of the brain/endocranium
(excluding the lobules/subarcuate fossa) in other
euarchontogliran groups independently.
The ancestral state reconstruction (ASR) analysis for
both relative lobule volume and mass suggests that the
lobule size increased within the lagomorph lineage since
their common ancestor with Rodentia, which
may suggest that the first option is better supported.
However, the ASR analysis was based only on
modern data since relevant information from fossil
euarchontoglirans is only patchily available. In the
absence of data from fossils, these analyses really only
constitute a heuristic device that produces hypotheses to
be tested with fossil data. Nevertheless, the endocasts of
Mesozoic mammals and cynodonts are described as hav-
ing large petrosal lobules, which indicates that this char-
acteristic may be ancestral for Mammalia (Kielan-
Jaworowska, 1986). However, quantitative data have
been obtained for only a handful of Mesozoic mammals
(Csiki-Sava et al., 2018; Macrini, 2006; Macrini, De
Muizon et al., 2007; Macrini, Rougier, et al., 2007) reveal-
ing a wide range of values (0.23%–5.98% of endocranial
TABLE 8 Results of pANOVA of residuals from PGLS regression of log
10
petrosal lobule volume (mm
3
) plotted against log
10
adjusted
endocranial volume (mm
3
) (PLV–AEV) and log
10
petrosal lobule mass (g) plotted against log
10
adjusted body mass (g) (PLM–ABM) by
primary ecological categories for locomotor behavior, activity pattern, and diet in Euarchontoglires
Data Df SS Residual SS R
2
FZ sig. (p)
PLV–AEV Locomotor behavior Categories 8 0.030 0.004 .101 1.843 1.221 .117
Residuals 131 0.267 0.002 .899
Total 139 0.297
Activity pattern Categories 4 0.004 0.001 .012 0.411 0.653 .772
Residuals 135 0.293 0.002 .988
Total 139 0.297
Diet Categories 6 0.021 0.004 .071 1.701 1.101 .120
Residuals 133 0.276 0.002 .929
Total 139 0.297
PLM–ABM Locomotor behavior Categories 8 0.088 0.011 .200 4.106 2.539 .002
Residuals 131 0.349 0.003 .800
Total 139 0.437
Activity pattern Categories 4 0.0047 0.001 .011 0.367 0.810 .806
Residuals 135 0.432 0.003 .989
Total 139 0.437
Diet Categories 6 0.031 0.005 .070 1.677 1.074 0.146
Residuals 130 0.406 0.003 .930
Total 136 0.437
Note: Significant p-values shown in bold.
LANG ET AL.19

volume) suggesting that more data are necessary to deter-
mine the ancestral condition for placental mammals (but
see Bertrand et al., 2022). Given that lagomorphs are
nested within Euarchontoglires, it seems more
parsimonious to suggest that this condition is a product
of evolutionary changes within that clade. But to consider
this question more rigorously, quantitative data for out-
groups to Euarchontoglires need to be added.
FIGURE 7 Boxplot of residuals from PGLS regression of log
10
petrosal lobule volume (mm
3
) plotted against log
10
adjusted endocranial
volume (mm
3
) and log
10
petrosal lobule mass (g) plotted against log
10
adjusted body mass (g) for 140 Euarchontoglires categorized by
locomotor behavior (a,b), diet (c,d), and activity pattern (e,f)
20 LANG ET AL.

The only fossil lagomorph for which petrosal lobule
data are known is the early Oligocene stem lagomorph,
Megalagus turgidus, which also possessed large petrosal
lobules relative to endocranial volume, similar in size to
those of extant leporids, but smaller than extant
ochotonids (L
opez-Torres et al., 2020). Significantly, the
size of the petrosal lobules relative to endocranial volume
in M. turgidus is larger than observed in early rodents
and stem primates, which does suggest some shift in the
relative size (or relative importance) of this part of the
brain in lagomorph evolution. However, extant lago-
morphs are reconstructed as having had lower
encephalization quotients (EQ) compared to extant
rodents, and the EQ was also found to be lower for M.
turgidus compared to early rodents (L
opez-Torres
et al., 2020). These various lines of evidence suggest that
TABLE 9 Results of post hoc tests based on pANOVA of residuals from PGLS regression of log
10
petrosal lobule mass (g) plotted against
log
10
adjusted body mass (g) for locomotor behavior in Euarchontoglires
Locomotor behavior Ar Cu Fo Gl Sl Sc Sa SAr Tr
Arboreal —0.762 0.007 0.38 0.973 0.051 0.712 0.681 0.719
Cursorial 0.762 —0.074 0.422 0.607 0.301 0.982 0.965 0.85
Fossorial 0.007 0.074 —0.269 0.069 0.097 0.032 0.034 0.004
Gliding 0.38 0.422 0.269 —0.526 0.977 0.387 0.366 0.307
Saltatorial 0.973 0.607 0.069 0.526 —0.367 0.741 0.759 0.828
Scansorial 0.051 0.301 0.097 0.977 0.367 —0.215 0.184 0.001
Semiaquatic 0.712 0.982 0.032 0.387 0.741 0.215 —0.975 0.815
Slow arboreal 0.681 0.965 0.034 0.366 0.759 0.184 0.975 —0.857
Terrestrial 0.719 0.85 0.004 0.307 0.828 0.001 0.815 0.857 —
Abbreviations: Ar, Arboreal; Cu, cursorial; Fo., fossorial; Gl, gliding; Sl, saltatorial; Sc, scansorial; Sa, semiaquatic; SAr, slow arboreal; Tr, terrestrial.
Note: Significant p-values shown in bold.
TABLE 10 Results of pANOVA of residuals from PGLS regression of log
10
petrosal lobule volume (mm
3
) plotted against log
10
adjusted
endocranial volume (mm
3
) (PLV AEV) and log
10
petrosal lobule mass (g) plotted against log
10
adjusted body mass (g) (PLM ABM) by
primary ecological categories for locomotor behavior, activity pattern, and diet in Rodentia
Data Df SS Residual SS R
2
FZ sig. (p)
PLV–AEV Locomotor behavior Categories 6 0.028 0.005 .124 1.511 0.937 .160
Residuals 64 0.196 0.003 .876
Total 70 0.224
Activity pattern Categories 4 0.004 0.001 .017 0.292 1.009 .855
Residuals 66 0.220 0.003 .983
Total 70 0.224
Diet Categories 4 0.024 0.005 .106 1.544 0.896 .176
Residuals 33 0.200 0.003 .894
Total 37 0.224
PLM–ABM Locomotor behavior Categories 6 0.043 0.007 .168 2.161 1.486 .051
Residuals 64 0.214 0.003 .832
Total 70 0.258
Activity pattern Categories 4 0.003 0.001 .014 0.227 1.396 .927
Residuals 66 0.254 0.004 .986
Total 70 0.258
Diet Categories 4 0.029 0.006 .111 1.631 0.973 .152
Residuals 33 0.229 0.004 .889
Total 37 0.258
Note: Significant p-values shown in bold.
LANG ET AL.21

both postulated explanations are likely at work. It
appears that the large size of the petrosal lobules relative
to endocranial volume in lagomorphs is being driven in
part by their otherwise small brains. But as the brain
evolved in lagomorph evolution, the petrosal lobules may
have been prioritized when other regions were not, or at
least were not to the same degree as in other
euarchontogliran clades (e.g., the neocortex, which
expanded in lagomorph evolution, but not to the same
degree as in Primates or some rodent lineages; Bertrand
et al., 2017;2019b; Long et al., 2015;L

opez-Torres
et al., 2020). As such, there may be some ecological basis
for this contrast, perhaps related to the saltatory behavior
of leporids or need for predator detection in all lago-
morphs as prey animals. However, this suggested rela-
tionship between lobule size and ecological behaviors
within Lagomorpha requires additional testing.
Distinct scaling relationships for relative petrosal
lobule volume and mass were also identified within Pri-
mates. Specifically, haplorhines had significantly
smaller petrosal lobules compared to strepsirrhines rela-
tive to endocranial volume (Figure 5a,b). Based on the
ancestral state reconstruction, it appears that the small
relative lobule volumes identified in the strepsirrhines
are being driven primarily by the lemuriforms and not
the lorisiforms, or at least not to the same extent
(though again this observation would benefit from test-
ing with fossils; Figure 6a). However, there is little dif-
ference in the size of the lobules relative to body mass
between the two suborders (Figures 5c,d and 6b),
although there are particular lineages that show evi-
dence of proportional decrease in both relative lobule
volume and mass (e.g., Daubentonia and Ateles). As
such, the pattern observed in relation to endocranial
volume is likely strongly influenced by the significant
expansion of the rest of the brain, primarily in the neo-
cortex, in anthropoids (Barton, 1996;Jerison,1973;
Kaas, 2012).
The scaling patterns of the scandentians generally
mirror those of the rodents relative to both volume and
mass. This is not surprising given the ecological analogies
which have been drawn between them, especially
between scandentians and squirrels, and the neural con-
vergences in other visual structures between the two
groups (Kaas, 2002). However, relative lobule volumes
and masses of the squirrel-related clade are reconstructed
as being larger than those of the scandentians
(as visualized in the ASR; Figure 6a,b), which would be
consistent with evidence from the fossil record that
expansion of this part of the brain occurred in sciuroid
evolution (Bertrand et al., 2017,2018,2021). In contrast,
dermopterans have small lobules relative to both
TABLE 11 Results of pANOVA of residuals from PGLS regression of log
10
petrosal lobule volume (mm
3
) plotted against log
10
adjusted
endocranial volume (mm
3
) (PLV–AEV) and log
10
petrosal lobule mass (g) plotted against log
10
adjusted body mass (g) (PLM–ABM) by
primary ecological categories for locomotor behavior, activity pattern, and diet in Primates
Data Df SS Residual SS R
2
FZ sig. (p)
PLV–AEV Locomotor behavior Categories 2 0.001 0.000 .012 0.213 0.672 .784
Residuals 35 0.192 0.003 .880
Total 37 0.218
Activity pattern Categories 2 0.000 0.000 .009 0.164 0.950 .849
Residuals 35 0.043 0.001 .991
Total 37 0.044
Diet Categories 4 0.005 0.001 .105 0.965 0.261 .451
Residuals 33 0.039 0.001 .895
Total 37 0.044
PLM–ABM Locomotor behavior Categories 2 0.000 0.000 .006 0.109 1.192 .896
Residuals 35 0.069 0.002 .994
Total 37 0.069
Activity pattern Categories 2 0.001 0.000 .009 0.151 0.977 .853
Residuals 35 0.068 0.002 .991
Total 37 0.069
Diet Categories 4 0.008 0.002 .119 1.113 0.441 .349
Residuals 33 0.061 0.002 .881
Total 37 0.069
22 LANG ET AL.

endocranial volume and body mass. Although this might
be unexpected in arboreal, gliding animals, the small lob-
ules of the dermopterans are consistent with the highly
variable and often small lobules reported for gliding
rodents (Bertrand et al., 2017,2021).
The evidence that phylogeny significantly influences
petrosal lobule scaling patterns is supported by the strong
phylogenetic signal identified in petrosal lobule size rela-
tive to both endocranial volume (Pagle's λ=0.966) and
body mass (Pagle's λ=0.949). Ferreira-Cardoso et al.
(2017) also identified a strong phylogenetic signal in the
scaling relationship between petrosal lobule volume and
endocranial volume in a smaller but more phylogenetically
diverse sample of mammals (Pagle's λ=0.93). Based on
these results, it appears that relative lobule size is strongly
influenced by phylogeny across Mammalia (Ferreira-
Cardoso et al., 2017) and within Euarchontoglires.
4.2 |Ecological patterns
It was hypothesized that the subarcuate fossa and corre-
spondingly the petrosal lobules would be larger in taxa
which engage in ecological behaviors that place greater
demand on the visual and the vestibular systems and
smaller in taxa that do not engage in such demanding
behaviors. Specifically, the behaviors hypothesized to be
more demanding, therefore requiring the greater image
stabilization afforded by larger petrosal lobules, include
diurnality, arboreality, and faunivory; and the behaviors
hypothesized to be the least demanding included
nocturnality, fossoriality, terrestriality, and folivory. Con-
trary to these predictions, the results presented here sug-
gest that general patterns in ecology, as encapsulated in
our scoring system, do not strongly influence variation in
the size of the petrosal lobules across Euarchontoglires.
While there is stronger support for the ecological models
in the Ornstein–Uhlenbeck analysis of clustered groups
(i.e., high, medium, low categories) than for the Brownian
motion model across Euarchontoglires for both relative
lobule volume and mass, the single peak OU model (OU1)
was best supported. This suggests that while ecological fac-
tors may influence the evolution of relative lobule size, the
ecological clusters used in this study may not effectively
capture this variation and/or that other factors not
accounted for by these ecological clusters are influencing
petrosal lobule evolution in Euarchontoglires. Correspond-
ingly, in nearly all of the pANOVAs, no significant differ-
ences were identified between ecological categories for
TABLE 12 Comparisons of evolutionary model fit for petrosal
lobule volume based on residuals of OLS regression of log
10
petrosal lobule volume (mm
3
) plotted against log
10
adjusted
endocranial volume (mm
3
)
Model Fit ΔAICc W
A
Euarchontoglires
BM1 174.70 78.07 0.00
OU1 96.63 0.00 0.48
OUMloc 98.56 1.94 0.18
OUMact 98.67 2.05 0.17
OUMdiet 98.64 2.01 0.17
Rodentia
BM1 20.20 1.58 0.17
OU1 21.78 0.00 0.36
OUMloc 20.51 1.26 0.19
OUMact 20.11 1.66 0.16
OUMdiet 19.56 2.21 0.12
Primates
BM1 54.57 5.58 0.03
OU1 48.99 0.00 0.54
OUMloc 51.95 2.96 0.12
OUMact 52.37 3.38 0.10
OUMdiet 51.01 2.02 0.20
Note: Significant ΔAICc values shown in bold.
TABLE 13 Comparisons of evolutionary model fit for petrosal
lobule mass based on residuals of OLS regression of log
10
petrosal
lobule mass (g) plotted against log
10
adjusted body mass (g)
Model Fit ΔAICc W
A
Euarchontoglires
BM1 270.88 128.84 0.00
OU1 142.04 0.00 0.95
OUMloc 150.14 8.10 0.02
OUMact 150.12 8.08 0.02
OUMdiet 150.21 8.16 0.02
Rodentia
BM1 63.24 2.84 0.12
OU1 60.40 0.00 0.50
OUMloc 62.77 2.37 0.15
OUMact 63.68 3.28 0.10
OUMdiet 62.99 2.60 0.14
Primates
BM1 3.99 5.85 0.03
OU1 9.32 0.51 0.38
OUMloc 9.83 0.00 0.49
OUMact 4.45 5.39 0.03
OUMdiet 5.72 4.11 0.06
Note: Significant ΔAICc values shown in bold.
LANG ET AL.23

locomotor behavior, diet, and activity pattern. Importantly,
however, an ecological pattern was identified in the pAN-
OVA with respect to fossorial rodents, which have smaller
relative lobule masses compared to several of the other
locomotor categories across Euarchontoglires, including
the arboreal, semi-aquatic, slow arborealist, and terrestrial
groups.
With the exception of the fossorial rodents, these
results support comparable analyses performed by
Ferreira-Cardoso et al. (2017), which found that ecology
did not play a role in relative lobule volume in a diverse
sample of extinct and extant taxa from across Mammalia.
There are several factors that can help to explain the dif-
ferent results obtained here for the fossorial rodents.
First, the sample in this study is more phylogenetically
constrained and therefore the anatomical structure which
forms the lobules is more likely to be homologous
between taxonomic groups. Second, in this study there
are a greater number of species in each of the ecological
categories, including the fossorial group, allowing for
greater resolution of lobular variation for each category.
And third, lobule size was evaluated against both
endocranial volume and body mass, whereas Ferreira-
Cardoso et al. (2017) only assessed lobule size against
endocranial volume in ecological analyses. Nevertheless,
evaluating lobule scaling patterns relative to both
endocranial volume and body mass is valuable given the
differences in relative lobule volume and mass identified
in certain groups, especially those which are highly
encephalized (i.e., anthropoids).
The relationship between relative lobule mass and
fossoriality supports recent analyses of petrosal lobule
scaling in rodents (Bertrand et al., 2017,2018,2021).
Analysis using the endocasts of extinct and extant
sciuroid rodents identified a marked downgrade in the
size of the petrosal lobules relative to both endocranial
volume and body mass in progressively more fossorial
aplodontids (Bertrand et al., 2018). These changes in rela-
tive lobule size were attributed to a decrease in visual
ability associated with burrowing behavior in the
aplodontid lineage (Bertrand et al., 2018). Additionally,
analysis of locomotor behavior and brain evolution noted
significant differences in the percentage of petrosal lobule
volume to endocranial volume between fossorial rodents
and their arboreal relatives (Bertrand et al., 2021), which
is consistent with the pattern presented here. Analyses of
the inner ear in caviomorph rodents indicate that the
subarcuate fossa is larger in taxa which engage in rapid
complex movements, as in the semi-aquatic Myocastor,
and smaller in burrowing taxa, like the fossorial
Ctenomys (Arnaudo et al., 2020). Outside of neural anat-
omy, the transition to fossoriality is associated with many
cranial (Agrawal, 1967), postcranial (Samuels & Van
Valkenburgh, 2008; Wölfer et al., 2019), and soft tissue
adaptions (i.e., vibrissae, ear, and tail length; Verde
Arregoitia et al., 2017) within rodents appropriate to this
unique ecological niche. Likewise, the small relative lob-
ule masses of the fossorial rodents may therefore reflect a
relaxation of adaptive pressure on these visual structures
or active selection given the energetic costs of
maintaining brain tissue (Williams & Herrup, 1988).
Reduced visual reliance has been shown to have pro-
found effects on the brain including the size of the neo-
cortex. Specifically, the visual cortex is reduced in shrews
(Catania et al., 1999), echolocating bats (Krubitzer, 1995)
and some moles (Catania & Kaas, 1995). In the case of
the fossorial rodents, these slow-moving species, which
spend a significant part of their lives in low-light condi-
tions where olfaction and vibrissal sensing are better
suited (Stein, 2000), have a reduced need for precise con-
trol of eye movements afforded by larger lobules.
The small relative lobule masses of the fossorial
rodents identified in this analysis may ultimately relate
to overall brain size, as some fossorial rodents
(i.e., Cryptomys hottentotus) are known to have smaller
brains relative to body mass (Bernard & Nurton, 1992).
Fossorial rodents are reported to have smaller lobule vol-
umes relative to endocranial volume compared to other
locomotor groups in previous studies (Bertrand
et al., 2018,2021) and appear to have smaller lobule vol-
umes in this analysis as well (Figure 7a). However, no
significant differences were identified in relative lobule
volume between the fossorial taxa and other locomotor
groups in Euarchontoglires more generally, or between
relative lobule volume or mass and any ecological vari-
able in the rodents separately. It is likely that the identifi-
cation of ecological patterns in relative lobule size by
Bertrand et al. (2018,2021) was made possible by the
inclusion of fossil material and analysis of lobule scaling
patterns through time within specific lineages. Addition-
ally, the lack of clear ecological patterns identified in the
Ornstein–Uhlenbeck model may relate to the clustering
of ecological groups necessary to run the analysis, which
may obfuscate the kind of within lineage changes that
Bertrand et al. (2018,2021) documented.
The absence of significant differences in relative lob-
ule size between the fossorial rodents and other groups
when analyzed in the context of endocranial volume in
this analysis may relate to the evolution of digging and
fossoriality within Rodentia. Fossoriality has evolved
independently several times in different rodent groups
(here represented by Bathyergidae, Spalacidae,
Aplodontidae) and has produced disparate adaptations in
skeletal bone structure (Amson & Bibi, 2021) and skull
morphology (Fournier et al., 2021). This disparity is the
product of a variety of factors including evolutionary
24 LANG ET AL.

history, soil type, time spent above ground, and signifi-
cantly, digging behavior (Stein, 2000). For example,
Ctenomys engages in “scratch-digging”behavior to exca-
vate burrows in which the forelimbs are primarily used
to break apart the soil, while its close relative,
Spalacopus, engages in “chisel-tooth”digging in which
large procumbent incisors are used to break up the soil
(Stein, 2000). In some species, as in Ellobius, the head
itself is used in a shoveling motion to excavate tunnels.
The diversity of digging behaviors and disparity in fosso-
rial adaptations (Amson & Bibi, 2021; Fournier
et al., 2021) may have influenced subarcuate fossa size
and therefore lobule size, especially if modifications have
been made to the braincase. While it is most likely that
the small lobules of these fossorial species relate to their
reduced reliance on vision, a feature that characterizes
fossorial groups (Stein, 2000), this result, together with
previous analyses (Bertrand et al., 2018,2021) highlights
the need for clade specific, fossil informed analyses of fos-
sorial adaptations.
Similar analyses of fossil and extant squirrels identi-
fied an increase in the relative size of the lobules between
early fossil rodents and the later occurring early arboreal
squirrel, Cedromus wilsoni (Bertrand et al., 2017,2021),
changes that were attributed to improved vision associ-
ated with the transition to arboreality (Bertrand
et al., 2017,2018) or in the very least, which helped to
facilitate the transition to arboreality in the squirrel line-
age (Bertrand et al., 2021). These analyses document
clear ecological scaling relationships in lobule size when
patterns are informed by fossils. Although there may
have been changes in the relative size of the petrosal lob-
ules associated with early ecological transitions within
Rodentia, specifically with respect to members of the
squirrel-related clade (Bertrand et al., 2017,2018,2021),
no such pattern is identified within the current analysis
focused on extant rodents despite the significant variabil-
ity in relative lobule size and the diversity of ecological
behaviors. The analyses by Bertrand et al. (2017) on brain
variation in fossil and extant squirrels found that they did
not possess large lobules relative to endocranial volume
and the size of the lobules relative to body mass for
C. wilsoni were within the range of extant squirrels.
Ancestral state reconstruction of lobule size by Bertrand
et al. (2021) suggests that squirrels had larger lobules
prior to their transition to arboreality, which was then
followed by an increase in overall brain size as they
became more arboreal. As such, this raises the possibility
that there may be shifts in lobule size within particular
lineages that are being masked at the scale of the current
analysis, or by subsequent events in brain evolution.
The absence of an ecological signal in petrosal lobule
size is unexpected for primates. The primate visual
system is highly specialized for the accurate perception of
distance and detail (Dominy et al., 2001), which depend
on precise control of eye movements. For instance, com-
pared to other mammals, primates have a high degree of
orbital convergence (Heesy, 2004,2008; Ross, 1995; Ross
et al., 2007). The effect of this convergence is that the
visual fields seen by each eye overlap significantly
(Heesy, 2004; Ross, 1995). The extensive overlap of the
visual fields in primates creates a large zone of stereo-
scopic depth perception (Howard & Rogers, 1995), using
the differences between the two images seen by the eyes
viewing an object at slightly different angles
(Heesy, 2008). Importantly, accurate depth perception
and the high degree of binocular field overlap require
precise coordination of the eyes to fuse the two large
monocular fields into one image and maintain fixation
on an object of interest (Wallace et al., 2013). Addition-
ally, orbital convergence in primates is associated with
ecological behaviors, in that faunivorous species tend to
have more convergent orbits compared to non-
faunivorous species (Heesy, 2008,2009). Orbital conver-
gence is also positively correlated with the size of the
visual structures of the brain, including the lateral genic-
ulate nucleus and the primary visual cortex
(Barton, 2004). Furthermore, morphological studies of
the primate eye and orbit demonstrate that haplorhine
primates have exceptionally high visual acuity (i.e., the
ability to make finely detailed differentiations between
closely spaced objects) compared to other mammals
(Veilleux & Kirk, 2009). Haplorhines also possess a reti-
nal fovea, a pit at the back of the eye with a high density
of photoreceptors, where visual acuity is greatest (Kay
et al., 1997). Precise control of eye movements is essential
for taxa with high visual acuity, especially haplorhines
(Kirk & Kay, 2004), where a viewed object must be
brought to the fovea to be seen in greatest detail. Analysis
of the relationship between visual acuity and the dimen-
sions of the semicircular canals, which provide proprio-
ceptive information also used to regulate eye movements,
suggest a strong correlation between the two, such that
as visual acuity increases so does the curvature of the
canals allowing for more precise perception of the direc-
tion and velocity of head and body movements (Kemp &
Kirk, 2014). Given (a) that the visual adaptations pos-
sessed by primates, including orbital convergence and
high visual acuity are predicated on precise control of eye
movements; (b) the evidence that visual specializations
are known to influence the size of the brain, neocortex,
and specific visual structures; and (c) the connection
between visual acuity and the semicircular canals, it is
surprising that the lobules, which regulate smooth pur-
suit and velocity of eye, are not correspondingly large
within primates, even relative to body size. Instead,
LANG ET AL.25

lagomorphs, with low orbital convergence (Heesy, 2004)
and low expected acuity as herbivores (Veilleux &
Kirk, 2009), possess the largest lobules, at least relative to
endocranial volume.
Although these results are unexpected, there are some
possible explanations as to why no ecological signal was
identified in the size of the petrosal lobules within Pri-
mates, especially considering the uncertainty surround-
ing the specific function of the tissues which fill the
subarcuate fossa in these groups. Despite the evidence
that the lobules play a significant role in the control of
eye movements (Hiramatsu et al., 2008) they are not the
only part of the brain which performs this function. As
mentioned, the petrosal lobules are part of a functional
unit, the floccular–parafloccular complex of the
vestibulocerebellum, which regulates VOR, stabilizes the
eyes, and controls the velocity of eye movements
(Hiramatsu et al., 2008; Ilg & Thier, 2008; Nagao, 1992;
Rambold et al., 2002; Shojaku et al., 1990; Zee
et al., 1981). Although the petrosal lobules of primates
are specifically involved with the control of smooth pur-
suit eye movements, this function is also performed by
the ventral paraflocculus (Nagao, 1992; Nagao
et al., 1997), which sits outside the fossa. While some
parts of the complex may perform more specific func-
tions, the anatomical divisions within the primate
floccular–parafloccular complex do not reflect functional
separations (Noda & Mikami, 1986).
Furthermore, it is worth noting that most of the ana-
lyses that assess the function of the petrosal lobules are
based on macaques. Catarrhines are known to have
smaller lobules that do not scale linearly with body mass,
as they do in other mammals and primates (Gannon
et al., 1988). Considering the systematic reduction in lob-
ular size within this lineage it is perhaps not surprising
there is little functional compartmentalization in the cat-
arrhine petrosal lobule. With hominids lacking the lob-
ules and the fossa entirely (Gannon et al., 1988), other
parts of the cerebellum have presumably taken on the
functional role of this structure long before its disappear-
ance. As a result, functional conclusions drawn from the
catarrhine brain may not be applicable to the petrosal
lobules of strepsirrhines and platyrrhines, in which the
lobules are comparatively large (Gannon et al., 1988).
What is more, it is also somewhat problematic to infer
the specifics of petrosal lobule function in rodents using
the morphologically derived macaques. In haplorhine
primates, concerted eye movements are essential for bin-
ocular fusion and centering an object of interest on the
fovea of each eye (Wallace et al., 2013). Analysis of eye
movements in the rat, however, indicate that the left and
right eye do not move in a concerted fashion, precluding
the possibility of primate-like binocular fusion (Wallace
et al., 2013). Instead, a field of binocular overlap is
maintained over the animal's head despite variation in
the alignment of the eyes and rapid movements of the
head and body (Wallace et al., 2013). In small mammals
like the rat, the most important function of the visual sys-
tem is the detection of predators at a distance
(De Franceschi et al., 2016; Land, 2013; Wallace
et al., 2013), as location of food and navigation of sub-
strates can be achieved through olfaction and vibrissae
(Hollander et al., 2012; Kleinfeld et al., 2006). If concerted
control of eye movements, which the petrosal lobules
play a role in regulating, is not an integral component of
rodent visual behavior, and the regulation of eye move-
ments is primarily related to predator detection, then
there may not be a relationship between the size of the
lobules and locomotor behaviors, diet, or activity pattern.
There may instead be a relationship between anti-
predator visual behavior and the size of neural structures
regulating eye movements. For example, ochotonids are
noted for having large lobules relative to brain size, even
larger than in the saltatorial leporids (L
opez-Torres
et al., 2020). This group of small terrestrial mammals
often lives in rocky open environments, which exposes
them to both aerial and terrestrial predators (Ivins &
Smith, 1983; Holmes, 1991). Precise and rapid eye move-
ment for predator detection may be imperative to identify
and alert colony members of potential threats (Ivins &
Smith, 1983; Volodin et al., 2018) necessitating larger lob-
ules in this group.
Some research has also identified a neural connection
between the auditory cortex and the floccular–
parafloccular complex of rodents (Azizi et al., 1985;
Azizi & Woodward, 1990; Du et al., 2017). In electrical
stimulation experiments on the rat brain, approximately
33% of the neurons in the paraflocculi were responsive to
stimulus of the auditory cortex (Azizi et al., 1985) and
71.4% of neurons in the contralateral auditory cortex
were responsive to stimulus in the paraflocculus (Du
et al., 2017). The functional significance of this connec-
tion is unknown, but a feedback loop between the two
structures is suggested to play a role in tinnitus in
humans (Du et al., 2017; Mennink et al., 2020). At pre-
sent, no studies have attempted to examine both visual
and auditory roles of the paraflocculus, and doing so is
likely to be exceedingly difficult as categorizing animals
based on auditory requirements would require even more
speculation than doing so based on visual requirements.
But that does not mean that such patterns do not exist.
The absence of a relationship between petrosal lobule
size and ecology (with the exception of the fossorial
rodents) could also relate to constraints on fossa expansion
within the cranium. The semicircular canals surround the
subarcuate fossa and scale according to body mass and
26 LANG ET AL.

locomotor agility (Jeffery et al., 2008; Spoor et al., 2007)—
the petrosal lobules may be constrained to some degree by
the size of these canals. However, the limited relationship
between locomotor behavior and lobule/fossa size in this
analysis negates the suggestion that the fossa scales solely
as a product of changes in semicircular canal size. From
an ontogenetic perspective, the subarcuate fossa is not
formed via ossification of tissue surrounding the petrosal
lobules. Instead, the formation of the subarcuate fossa is
connected to the growth and development of the semicir-
cular canals, and the petrosal lobules secondarily occupy
the fossa (McClure & Daron, 1971; Jeffery & Spoor, 2006).
Although the fossa and the canals are developmentally
connected (Jeffery & Spoor, 2006), the size of the fossa
does not appear to be entirely dependent on the size of the
canals. Other factors related to cranial scaling may have a
significant impact on fossa and lobular scaling. For exam-
ple, locomotor behavior is suggested to play a role in cra-
nial (Lu et al., 2014) and endocranial shape (Bertrand
et al., 2019a) in squirrels. This may relate to the small lob-
ules of gliding squirrels as they possess a large auditory
bulla, which may limit the space available for the fossa
(Bertrand et al., 2019a). Despite the prospective impor-
tance of visual tracking in arboreal and volant locomotion,
some gliding rodent species lack the lobules entirely
(i.e., Petinomys setosus; Bertrand et al., 2017). Furthermore,
species with globular crania (i.e., hominoids) also do not
possess a subarcuate fossa, which is presumably related to
factors other than ecology.
5|CONCLUSION
In this article, the scaling patterns of the petrosal lobules,
as determined from endocranial reconstructions of
subarcuate fossa size, were examined in 140 extant
euarchontoglirans to evaluate scaling relationships with
the rest of the endocranium and with body mass to iden-
tify phylogenetic patterns and to determine if ecological
factors play a role in the relative sizes of these structures
using phylogenetically controlled analyses. These ana-
lyses indicate that the size of the petrosal lobules is posi-
tively correlated with both endocranial volume and body
mass, which is largely consistent with previous research
(Ferreira-Cardoso et al., 2017; Gannon et al., 1988). Over-
all, phylogeny appears to be a major factor in the scaling
of the petrosal lobules, with significant differences in rel-
ative petrosal lobule size identified between orders and
suborders. Specifically, lagomorphs had significantly
larger petrosal lobules compared to all other orders rela-
tive to endocranial volume and haplorhines had signifi-
cantly smaller petrosal lobules compared to
strepsirrhines relative to endocranial volume. The ordinal
scaling patterns identified in petrosal lobule size relative
to endocranial volume differed to some degree from those
identified in petrosal lobule mass relative to body mass.
The relative lobule mass of the lagomorphs was only sig-
nificantly different from the rodents and the
dermopterans, while no difference was identified
between haplorhine and strepsirrhine relative lobule
masses. These contrasts highlight the importance of
doing both types of comparisons, since they offer differ-
ent perspectives on the evolutionary process. Rodents
were also found to be significantly different from all
other groups, except for dermopterans, in relative lobule
mass, but this pattern is difficult to characterize given the
range of variation for relative lobule mass in this group.
These contrasting results imply a complex interplay
between the evolution of the size of the petrosal lobules
with the evolution of body mass and relative brain size.
While there is evidence that the unique scaling pat-
terns identified in several of these phylogenetic groups
may have been ecologically driven early in their respec-
tive lineages (Bertrand et al., 2017,2018,2019a,2021), no
connection was identified between most of the ecological
factors tested here and relative lobule size within this
sample of extant taxa. However, significant differences in
lobule size were identified for fossorial taxa in relative
lobule mass across Euarchontoglires, as the fossorial taxa,
comprised exclusively of rodents, had small lobules com-
pared to arboreal, scansorial, semi-aquatic, slow arbo-
realist, and terrestrial groups. Small lobules and
subarcuate fossae have been documented in other studies
of fossorial aplodontiids and ischyromyids (Bertrand
et al., 2018,2021) and caviomorphs (Arnaudo
et al., 2020). Together these results indicate that the small
lobules in fossorial rodents may reflect an adaptation to
burrowing or in the least a relaxation of selective pres-
sure, where the need for larger lobules and increased
coordination of eye movements is decreased relative to
other locomotor behaviors.
The lack of a relationship identified between relative
lobule size and other ecological factors may be influenced
by differences in the functional anatomical organization
among groups of the lobules (i.e., between Primates and
Glires), which necessitates a more phylogenetically con-
strained analysis than has been presented here. Further-
more, it is possible that adaptive changes to the petrosal
lobules occurred early in these lineages, but that subse-
quent changes may have obscured those patterns in the
extant sample, as suggested by the contrast between the
results here and those found in the study of fossil
sciuroids and ischyromyids (Bertrand et al., 2017,2018,
2021). The underlying goal of this study was to better
understand the scaling relations of the petrosal lobules in
Euarchontoglires, with the hope that the results may be
LANG ET AL.27

applied in analyses of fossil endocasts to better under-
stand the sensory repertoire of extinct species. As ecology
related scaling patterns appear only in fossorial taxa
within this extant sample it is difficult to draw clear-cut
conclusions about the significance of lobule size in fossil
specimens outside of this ecological category. As a result,
the extent to which variation in these lobules and the
subarcuate fossa can be used to interpret the morphology
and sensory ecology of fossil taxa is limited and needs to
be tested within specific clades and in the context of con-
sidering evolutionary trajectories (informed by fossil
material) for both overall brain size and body mass.
ACKNOWLEDGMENTS
The authors would like to acknowledge Dr. P. Cox,
Dr. A. Harrington, Dr. R. Asher, and Dr. M. Lowe for
access to CT scans through the Morphosource database.
The authors thank Dr. E. Sargis for his communications
on scandentian body masses. The authors would also like
to thank the work/study students, Koda MacLellan,
Pamela Santos, Tejnarine Persaud, and James Graves for
their efforts in creating endocasts for this project.
AUTHOR CONTRIBUTIONS
Madlen M. Lang: Conceptualization (equal); data
curation (lead); formal analysis (equal); methodology
(equal); visualization (lead); writing –original draft (lead).
Ornella C. Bertrand: Formal analysis (supporting);
investigation (supporting); methodology (supporting);
writing –review and editing (supporting). Gabriela San
Martin-Flores: Investigation (supporting); writing –
review and editing (supporting). Chris J. Law: Formal
analysis (supporting); methodology (supporting); visualiza-
tion (supporting); writing –review and editing
(supporting). Jade Abdul-Sater: Investigation
(supporting); writing –review and editing (supporting).
Shayda Spakowski: Investigation (supporting); writing –
review and editing (supporting). Mary T. Silcox: Concep-
tualization (supporting); formal analysis (supporting);
funding acquisition (lead); methodology (supporting);
supervision (lead); writing –original draft (supporting);
writing –review and editing (supporting).
DATA ACCESSIBILITY
All endocasts used for this analysis will be made available
on Morphosource in the following project: https://www.
morphosource.org/projects/000424317?locale=en.
ORCID
Madlen M. Lang https://orcid.org/0000-0003-2604-4733
Ornella C. Bertrand https://orcid.org/0000-0003-3461-
3908
Mary T. Silcox https://orcid.org/0000-0002-4174-9435
REFERENCES
Agrawal, V. C. (1967). Skull adaptations in fossorial rodents.
Mammalia,31, 300–312.
Amson, E., & Bibi, F. (2021). Differing effects of size and lifestyle
on bone structure in mammals. BMC Biology,19(1), 1–18.
Aristide, L., Dos Reis, S. F., Machado, A. C., Lima, I.,
Lopes, R. T., & Perez, S. I. (2015). Encephalization and diversifi-
cation of the cranial base in platyrrhine primates. Journal of
Human Evolution,81,29–40.
Arnaudo, M. E., Arnal, M., & Ekdale, E. G. (2020). The auditory
region of a caviomorph rodent (Hystricognathi) from the early
Miocene of Patagonia (South America) and evolutionary con-
siderations. Journal of Vertebrate Paleontology,40(2), e1777557.
Azizi, S. A., Burne, R. A., & Woodward, D. J. (1985). The auditory
corticopontocerebellar projection in the rat: Inputs to the par-
aflocculus and midvermis. An anatomical and physiological
study. Experimental Brain Research,59(1), 36–49.
Azizi, S. A., & Woodward, D. J. (1990). Interactions of visual and
auditory mossy fiber inputs in the paraflocculus of the rat: A
gating action of multimodal inputs. Brain Research,533(2),
255–262.
Balanoff, A. M., & Bever, G. S. (2020). The role of endocasts in the
study of brain evolution. In J. H. Kaas (Ed.), Evolutionary neu-
roscience (pp. 29–49). Academic Press.
Barks,S.K.,Calhoun,M.E.,Hopkins,W.D.,Cranfield,M.R.,
Mudakikwa,A.,Stoinski,T.S.,Patterson,F.G.,
Erwin,J.M.,Hecht,E.E.,Hof,P.R.,&Sherwood,C.C.
(2015). Brain organization of gorillas reflects species differ-
ences in ecology. AmericanJournalofPhysicalAnthropol-
ogy,156(2), 252–262.
Barton, R. A. (1996). Neocortex size and behavioral ecology in pri-
mates. Proceedings of the Royal Society of London. Series B: Bio-
logical Sciences,263(1367), 173–177.
Barton, R. A. (1998). Visual specialization and brain evolution in
primates. Proceedings of the Royal Society of London. Series B:
Biological Sciences,265(1409), 1933–1937.
Barton, R. A. (2004). Binocularity and brain evolution in primates.
Proceedings of the National Academy of Sciences,101(27),
10113–10115.
Barton, R. A. (2006). Primate brain evolution: Integrating compara-
tive, neurophysiological, and ethological data. Evolutionary
Anthropology: Issues, News, and Reviews: Issues, News, and
Reviews,15(6), 224–236.
Barton, R. A., & Harvey, P. H. (2000). Mosaic evolution of brain
structure in mammals. Nature,405(6790), 1055–1058.
Barton, R. A., Purvis, A., & Harvey, P. H. (1995). Evolutionary radi-
ation of visual and olfactory brain systems in primates, bats
and insectivores. Philosophical Transactions of the Royal Society
of London. Series B: Biological Sciences,348(1326), 381–392.
Beaulieu, J. M., Jhwueng, D. C., Boettiger, C., & O'Meara, B. C.
(2012). Modeling stabilizing selection: expanding the Ornstein–
Uhlenbeck model of adaptive evolution. Evolution: Interna-
tional Journal of Organic Evolution,66(8), 2369–2383.
Belton, T., & McCrea, R. A. (2000a). Role of the cerebellar flocculus
region in cancellation of the VOR during passive whole body
rotation. Journal of Neurophysiology,84(3), 1599–1613.
Belton, T., & McCrea, R. A. (2000b). Role of the cerebellar floccu-
lus region in the coordination of eye and head movements
during gaze pursuit. Journal of Neurophysiology,84(3), 1614–
1626.
28 LANG ET AL.

Bernard, R. T. F., & Nurton, J. (1992). Ecological correlates of rela-
tive brain size in some South African rodents. South African
Journal of Zoology,28(2), 95–98.
Bertrand, O. C., Amador-Mughal, F., Lang, M. M., & Silcox, M. T.
(2018). Virtual endocasts of fossil Sciuroidea: Brain size reduc-
tion in the evolution of fossoriality. Palaeontology,61(6),
919–948.
Bertrand, O. C., Amador-Mughal, F., Lang, M. M., & Silcox, M. T.
(2019b). New virtual endocasts of Eocene Ischyromyidae and
their relevance in evaluating neurological changes occurring
through time in Rodentia. Journal of Mammalian Evolution,
26(3), 345–371.
Bertrand, O. C., Amador-Mughal, F., & Silcox, M. T. (2017). Virtual
endocast of the early Oligocene Cedromus wilsoni
(Cedromurinae) and brain evolution in squirrels. Journal of
Anatomy,230(1), 128–151.
Bertrand, O. C., Püschel, H. P., Schwab, J. A., Silcox, M. T., &
Brusatte, S. L. (2021). The impact of locomotion on the brain
evolution of squirrels and close relatives. Communications biol-
ogy,4(1), 1–15.
Bertrand, O. C., Schillaci, M. A., & Silcox, M. T. (2016). Cranial
dimensions as estimators of body mass and locomotor habits in
extant and fossil rodents. Journal of Vertebrate Paleontology,36
(1), e1014905. https://doi.org/10.1098/rspb.2020.0665
Bertrand, O. C., San Martin-Flores, G., & Silcox, M. T. (2019a).
Endocranial shape variation in the squirrel-related clade and
their fossil relatives using 3D geometric morphometrics: Contri-
butions of locomotion and phylogeny to brain shape. Journal of
Zoology,308(3), 197–211.
Bertrand, O. C., Shelley, S. L., Williamson, T. E., Wible, J. R.,
Chester, S. G., Flynn, J. J., Holbrook, L. T., Lyson, T. R.,
Meng, J., Miller, I. M., Püschel, H. P., Smith, T., Spaulding, M.,
Tseng, Z. J., & Brusatte, S. L. (2022). Brawn before brains in
placental mammals after the end-Cretaceous extinction. Sci-
ence,376(6588), 80–85.
Bhatnagar, K. P., & Kallen, F. C. (1974). Cribriform plate of eth-
moid, olfactory bulb and olfactory acuity in forty species of
bats. Journal of Morphology,142(1), 71–89.
Bollback, J. P. (2006). SIMMAP: Stochastic character mapping of
discrete traits on phylogenies. BMC Bioinformatics,7(1), 1–7.
Burne, R. A., & Woodward, D. J. (1983). Visual cortical projections
to the paraflocculus in the rat. Experimental Brain Research,
49(1), 55–67.
Burnham, K. P., & Anderson, D. R. (2004). Multimodel inference:
Understanding AIC and BIC in model selection. Sociological
Methods & Research,33(2), 261–304.
Butler, M. A., & King, A. A. (2004). Phylogenetic comparative anal-
ysis: A modeling approach for adaptive evolution. The Ameri-
can Naturalist,164(6), 683–695.
Campi, K. L., & Krubitzer, L. (2010). Comparative studies of diurnal
and nocturnal rodents: Differences in lifestyle result in alter-
ations in cortical field size and number. Journal of Comparative
Neurology,518(22), 4491–4512.
Catania, K. C., & Kaas, J. H. (1995). Organization of the somatosen-
sory cortex of the star-nosed mole. Journal of Comparative Neu-
rology,351(4), 549–567.
Catania, K. C., Lyon, D. C., Mock, O. B., & Kaas, J. H. (1999). Corti-
cal organization in shrews: Evidence from five species. Journal
of Comparative Neurology,410(1), 55–72.
Clutton-Brock, T. H., & Harvey, P. H. (1980). Primates, brains and
ecology. Journal of Zoology,190(3), 309–323.
Collyer, M. L., & Adams, D. C. (2018). RRPP: An r package for
fitting linear models to high-dimensional data using residual
randomization. Methods in Ecology and Evolution,9(7), 1772–
1779.
Csiki-Sava, Z., Vremir, M., Meng, J., Brusatte, S. L., & Norell, M. A.
(2018). Dome-headed, small-brained Island mammal from the
late cretaceous of Romania. Proceedings of the National Acad-
emy of Sciences,115(19), 4857–4862.
De Franceschi, G., Vivattanasarn, T., Saleem, A. B., &
Solomon, S. G. (2016). Vision guides selection of freeze or
flight defense strategies in mice. Current Biology,26(16), 2150–
2154.
de Winter, W., & Oxnard, C. E. (2001). Evolutionary radiations and
convergences in the structural organization of mammalian
brains. Nature,409(6821), 710–714.
DeCasien, A. R., & Higham, J. P. (2019). Primate mosaic brain evo-
lution reflects selection on sensory and cognitive specialization.
Nature Ecology & Evolution,3(10), 1483–1493.
DeCasien, A. R., Williams, S. A., & Higham, J. P. (2017). Primate
brain size is predicted by diet but not sociality. Nature Ecology &
Evolution,1(5), 1–7.
Dominy, N. J., Lucas, P. W., Osorio, D., & Yamashita, N. (2001).
The sensory ecology of primate food perception. Evolutionary
Anthropology: Issues, News, and Reviews: Issues, News, and
Reviews,10(5), 171–186.
Du, Y., Liu, J., Jiang, Q., Duan, Q., Mao, L., & Ma, F. (2017). Par-
aflocculus plays a role in salicylate-induced tinnitus. Hearing
Research,353, 176–184.
Dunbar, R. I. M. (1992). Neocortex size as a constraint on group size
in primates. Journal of Human Evolution,22(6), 469–493.
Dunbar, R. I. M. (1995). Neocortex size and group size in primates:
A test of the hypothesis. Journal of Human Evolution,28(3),
287–296.
Dunbar, R. I. M., & Shultz, S. (2007). Understanding primate brain
evolution. Philosophical Transactions of the Royal Society B: Bio-
logical Sciences,362(1480), 649–658.
Eisenberg, J. F., & Wilson, D. E. (1978). Relative brain size and
feeding strategies in the Chiroptera. Evolution,32, 740–751.
Eisenberg, J. F., & Wilson, D. E. (1981). Relative brain size and
demographic strategies in didelphid marsupials. The American
Naturalist,118(1), 1–15.
Ferreira-Cardoso, S., Araújo, R., Martins, N. E., Martins, G. G.,
Walsh, S., Martins, R. M. S., Kardjilov, N., Manke, I.,
Hilger, A., & Castanhinha, R. (2017). Floccular fossa size is not
a reliable proxy of ecology and behavior in vertebrates. Scien-
tific Reports,7(1), 1–11.
Fleagle, J. G. (2013). Primate adaptation and evolution (3rd ed.).
Academic press.
Fournier, M., Hautier, L., & Rodrigues, H. G. (2021). Evolution
towards fossoriality and morphological convergence in the
skull of Spalacidae and Bathyergidae (Rodentia). Journal of
Mammalian Evolution,28,1–15.
Freckleton, R. P., Harvey, P. H., & Pagel, M. (2002). Phylogenetic
analysis and comparative data: A test and review of evidence.
The American Naturalist,160(6), 712–726.
Fukushima, K., Fukushima, J., Kaneko, C. R., & Fuchs, A. F.
(1999). Vertical Purkinje cells of the monkey floccular lobe:
LANG ET AL.29

Simple-spike activity during pursuit and passive whole body
rotation. Journal of Neurophysiology,82(2), 787–803.
Gannon, P. J., Eden, A. R., & Laitman, J. T. (1988). The subarcuate
fossa and cerebellum of extant primates: Comparative study of
a skull-brain interface. American Journal of Physical Anthropol-
ogy,77(2), 143–164.
Gittleman, J. L. (1986). Carnivore brain size, behavioral ecology,
and phylogeny. Journal of Mammalogy,67(1), 23–36.
Gittleman, J. L. (1991). Carnivore olfactory bulb size: Allometry,
phylogeny and ecology. Journal of Zoology,225(2), 253–272.
Glickstein, M., Gerrits, N., Kralj-Hans, I., Mercier, B., Stein, J., &
Voogd, J. (1994). Visual pontocerebellar projections in the
macaque. Journal of Comparative Neurology,349(1), 51–72.
Hansen, T. F. (1997). Stabilizing selection and the comparative
analysis of adaptation. Evolution,51(5), 1341–1351.
Harvey, P. H., Clutton-Brock, T. H., & Mace, G. M. (1980).
Brain size and ecology in small mammals and primates.
Proceedings of the National Academy of Sciences,77(7), 4387–
4389.
Heesy, C. P. (2004). On the relationship between orbit orientation
and binocular visual field overlap in mammals. The Anatomical
Record Part A: Discoveries in Molecular, Cellular, and Evolution-
ary Biology: An Official Publication of the American Association
of Anatomists,281(1), 1104–1110.
Heesy, C. P. (2008). Ecomorphology of orbit orientation and the
adaptive significance of binocular vision in primates and other
mammals. Brain, Behavior and Evolution,71(1), 54–67.
Heesy, C. P. (2009). Seeing in stereo: The ecology and evolution of
primate binocular vision and stereopsis. Evolutionary Anthro-
pology,18(1), 21–35.
Hiramatsu, T., Ohki, M., Kitazawa, H., Xiong, G., Kitamura, T.,
Yamada, J., & Nagao, S. (2008). Role of primate cerebellar
lobulus petrosus of paraflocculus in smooth pursuit eye move-
ment control revealed by chemical lesion. Neuroscience
Research,60(3), 250–258.
Ho, L.S.T., Ane, C., Lachlan, R., Tarpinian, K., Feldman, R., Yu, Q.,
van der Bijl, W., Maspons, J., Vos, R., & Ho, M.L.S.T., (2016).
Package ‘phylolm’.http://cran.r-project.org/web/packages/
phylolm/index.html
Hollander, J. L., Vander Wall, S. B., & Longland, W. S. (2012).
Olfactory detection of caches containing wildland versus culti-
vated seeds by granivorous rodents. Western North American
Naturalist,72(3), 339–347.
Holmes, W. G. (1991). Predator risk affects foraging behavior of
pikas: Observational and experimental evidence. Animal Behav-
ior,42(1), 111–119.
Howard, I. P., & Rogers, B. J. (1995). Binocular vision and stereopsis.
Oxford University Press.
Huelsenbeck, J. P., Larget, B., & Alfaro, M. E. (2004). Bayesian phy-
logenetic model selection using reversible jump Markov chain
Monte Carlo. Molecular Biology and Evolution,21(6), 1123–
1133.
Ilg, U. J., & Thier, P. (2008). The neural basis of smooth pursuit eye
movements in the rhesus monkey brain. Brain and Cognition,
68(3), 229–240.
Ivins, B. L., & Smith, A. T. (1983). Responses of pikas (Ochotona
princeps, Lagomorpha) to naturally occurring terrestrial preda-
tors. Behavioral Ecology and Sociobiology,13(4), 277–285.
Janzen, D. H. (1971). Seed predation by animals. Annual Review of
Ecology and Systematics,2(1), 465–492.
Jeffery, N., Ryan, T. M., & Spoor, F. (2008). The primate subarcuate
fossa and its relationship to the semicircular canals part II:
Adult interspecific variation. Journal of Human Evolution,
55(2), 326–339.
Jeffery, N., & Spoor, F. (2006). The primate subarcuate fossa and its
relationship to the semicircular canals part I: Prenatal growth.
Journal of Human Evolution,51(5), 537–549.
Jerison, H. J. (1973). Evolution of the brain and intelligence. Aca-
demic Press.
Jerison, H. J. (1979). Brain, body and encephalization in early pri-
mates. Journal of Human Evolution,8(6), 615–635.
Kaas, J. (2002). Convergences in the modular and areal organiza-
tion of the forebrain of mammals: Implications for the recon-
struction of forebrain evolution. Brain, Behavior and Evolution,
59(5–6), 262–272.
Kaas, J. H. (2012). The evolution of neocortex in primates. Progress
in Brain Research,195,91–102.
Kay, R. F., Ross, C., & Williams, B. A. (1997). Anthropoid origins.
Science,275(5301), 797–804.
Kemp, A. D., & Christopher Kirk, E. (2014). Eye size and visual
acuity influence vestibular anatomy in mammals. The Anatomi-
cal Record,297(4), 781–790.
Kheradmand, A., & Zee, D. S. (2011). Cerebellum and ocular motor
control. Frontiers in Neurology,2, 53.
Kielan-Jaworowska, Z. (1986). Brain evolution in Mesozoic mam-
mals. Rocky Mountain Geology,24(3), 21–34.
Kirk, E. C. (2006). Visual influences on primate encephalization.
Journal of Human Evolution,51(1), 76–90.
Kirk, E. C., & Kay, R. F. (2004). The evolution of high visual acuity
in the Anthropoidea. In J. G. Fleagle & R. F. Kay (Eds.),
Anthropoid origins (pp. 539–602). Springer.
Kleinfeld, D., Ahissar, E., & Diamond, M. E. (2006). Active sensa-
tion: Insights from the rodent vibrissa sensorimotor system.
Current Opinion in Neurobiology,16(4), 435–444.
Krauzlis, R. J., & Lisberger, S. G. (1994). Simple spike responses of
gaze velocity Purkinje cells in the floccular lobe of the monkey
during the onset and offset of pursuit eye movements. Journal
of Neurophysiology,72(4), 2045–2050.
Krubitzer, L. (1995). The organization of neocortex in mammals:
Are species differences really so different? Trends in Neurosci-
ences,18(9), 408–417.
Krubitzer, L., Campi, K. L., & Cooke, D. F. (2011). All rodents are
not the same: A modern synthesis of cortical organization.
Brain, Behavior and Evolution,78(1), 51–93.
Kudo, H., & Dunbar, R. I. (2001). Neocortex size and social network
size in primates. Animal Behavior,62(4), 711–722.
Land, M. F. (2013). Animal vision: Rats watch the sky. Current Biol-
ogy,23(14), R611–R613.
Lemen, C. (1980). Relationship between relative brain size and
climbing ability in Peromyscus.Journal of Mammalogy,61(2),
360–364.
Long, A., Bloch, J. I., & Silcox, M. T. (2015). Quantification of neo-
cortical ratios in stem primates. American Journal of Physical
Anthropology,157(3), 363–373.
L
opez-Torres, S., Bertrand, O. C., Lang, M. M., Silcox, M. T., &
Fostowicz-Frelik, Ł. (2020). Cranial endocast of the stem
30 LANG ET AL.

lagomorph Megalagus and brain structure of basal
Euarchontoglires. Proceedings of the Royal Society B,287(1929),
20200665.
Lu, X., Ge, D., Xia, L., Huang, C., & Yang, Q. (2014). Geometric
morphometric study of the skull shape diversification in
Sciuridae (Mammalia, Rodentia). Integrative zoology,9(3),
231–245.
Mace, G. M., Harvey, P. H., & Clutton-Brock, T. H. (1981). Brain
size and ecology in small mammals. Journal of Zoology,193(3),
333–354.
Macrini, T. E. (2006). The evolution of endocranial space in mam-
mals and non-mammalian cynodonts. [PhD dissertation]. Uni-
versity of Texas.
Macrini, T. E., De Muizon, C., Cifelli, R. L., & Rowe, T. (2007). Digi-
tal cranial endocast of Pucadelphys andinus, a Paleocene
metatherian. Journal of Vertebrate Paleontology,27(1), 99–107.
Macrini, T. E., Rougier, G. W., & Rowe, T. (2007). Description of a
cranial endocast from the fossil mammal Vincelestes
neuquenianus (Theriiformes) and its relevance to the evolution
of endocranial characters in therians. The Anatomical Record:
Advances in Integrative Anatomy and Evolutionary Biology,
290(7), 875–892.
Martin, R. D. (1990). Primate origins and evolution: A phylogenetic
reconstruction. Chapman and Hall.
May, J. G., Keller, E. L., & Suzuki, D. A. (1988). Smooth-pursuit eye
movement deficits with chemical lesions in the dorsolateral
pontine nucleus of the monkey. Journal of Neurophysiology,
59(3), 952–977.
McClure, T. D., & Daron, G. H. (1971). The relationship of the
developing inner ear, subarcuate fossa and paraflocculus in the
rat. American Journal of Anatomy,130(2), 235–249.
Meier, P. T. (1983). Relative brain size within the North American
Sciuridae. Journal of Mammalogy,64(4), 642–647.
Melin, A. D., Young, H. C., Mosdossy, K. N., & Fedigan, L. M.
(2014). Seasonality, extractive foraging and the evolution of pri-
mate sensorimotor intelligence. Journal of Human Evolution,
71,77–86.
Mennink, L. M., van Dijk, J. M. C., & van Dijk, P. (2020). The cere-
bellar (para) flocculus: A review on its auditory function and a
possible role in tinnitus. Hearing Research,398, 108081.
Moncunill-Solé, B., Quintana, J., Jordana, X., Engelbrektsson, P., &
Köhler, M. (2015). The weight of fossil leporids and ochotonids:
Body mass estimation models for the order L agomorpha. Jour-
nal of Zoology,295(4), 269–278.
Nagao, S. (1992). Different roles of flocculus and ventral par-
aflocculus for oculomotor control in the primate. Neuroreport,
3(1), 13–16.
Nagao, S., Kitamura, T., Nakamura, N., Hiramatsu, T., &
Yamada, J. (1997). Differences of the primate flocculus and
ventral paraflocculus in the mossy and climbing fiber input
organization. Journal of Comparative Neurology,382(4),
480–498.
Nash, L. T. (1986). Dietary, behavioral, and morphological aspects
of gummivory in primates. American Journal of Physical
Anthropology,29(S7), 113–137.
Newsome, W. T., Wurtz, R. H., & Komatsu, H. (1988). Relation of
cortical areas MT and MST to pursuit eye movements.
II. Differentiation of retinal from extraretinal inputs. Journal of
Neurophysiology,60(2), 604–620.
Ni, R. J., Huang, Z. H., Luo, P. H., Ma, X. H., Li, T., & Zhou, J. N.
(2018). The tree shrew cerebellum atlas: Systematic nomencla-
ture, neurochemical characterization, and afferent projections.
Journal of Comparative Neurology,526(17), 2744–2775.
Nielsen, R. (2002). Mapping mutations on phylogenies. Systematic
Biology,51(5), 729–739.
Noda, H., & Mikami, A. (1986). Discharges of neurons in the dorsal
paraflocculus of monkeys during eye movements and visual
stimulation. Journal of Neurophysiology,56(4), 1129–1146.
Osanai, R., Nagao, S., Kitamura, T., Kawabata, I., & Yamada, J.
(1999). Differences in mossy and climbing afferent sources
between flocculus and ventral and dorsal paraflocculus in the
rat. Experimental Brain Research,124(2), 248–264.
Pagel, M. (1999). The maximum likelihood approach to
reconstructing ancestral character states of discrete characters
on phylogenies. Systematic Biology,48(3), 612–622.
Panezai,S.K.,Luo,Y.,Vibulyaseck,S.,Sarpong,G.A.,
Nguyen-Minh, V. T., Nedelescu, H., Hirano, S., &
Sugihara, I. (2020). Reorganization of longitudinal com-
partments in the laterally protruding paraflocculus of the
postnatal mouse cerebellum. Journal of Comparative Neu-
rology,528(10), 1725–1741.
Pilleri, G., Gihr, M., & Kraus, C. (1984). Cephalization in rodents
with particular reference to the Canadian beaver (Castor can-
adensis). In Investigations on Beavers. Brain Anatomy
Institute.
Powell, L. E., Isler, K., & Barton, R. A. (2017). Re-evaluating the
link between brain size and behavioral ecology in primates.
Proceedings of the Royal Society B: Biological Sciences,
284(1865), 20171765.
R Core Team. (2020). R: A language and environment for statistical
computing. R Foundation for Statistical Computing. http://
www.R-project.org/
Rambold, H., Churchland, A., Selig, Y., Jasmin, L., &
Lisberger, S. G. (2002). Partial ablations of the flocculus and
ventral paraflocculus in monkeys cause linked deficits in
smooth pursuit eye movements and adaptive modification of
the VOR. Journal of Neurophysiology,87(2), 912–924.
Rasband, W.S. (1997-2018). ImageJ, U. S. National Institutes of
Health, Bethesda, Maryland, USA. https://imagej.nih.gov/ij/
Regan, B. C., Julliot, C., Simmen, B., Viénot, F., Charles-
Dominique, P., & Mollon, J. D. (2001). Fruits, foliage and the
evolution of primate colour vision. Philosophical Transactions
of the Royal Society of London. Series B: Biological Sciences,
356(1407), 229–283.
Revell, L. J. (2012). phytools: An R package for phylogenetic com-
parative biology (and other things). Methods in Ecology and
Evolution,3(2), 217–223.
Ridgway, S. H., & Hanson, A. C. (2014). Sperm whales and killer
whales with the largest brains of all toothed whales show
extreme differences in cerebellum. Brain, Behavior and Evolu-
tion,83(4), 266–274.
Ross, C. F. (1995). Allometric and functional influences on primate
orbit orientation and the origins of the Anthropoidea. Jo