Vocal development is typically thought of as the adaptive coordination of the vocal apparatus (i.e., the lungs, larynx and the mouth) and the associated muscles and neural systems that influence its activity. In adult animals, however, vocal behavior does not occur in isolation of the operations of other motor systems. Studies investigating the real-time coordination between vocal and locomotor outputs show that some vocalizations can be produced concurrently with locomotor activity and/or postural changes, while others cannot (Suthers et al., 1972; Blumberg, 1992; Fusani et al., 1996; Williams, 2001; Wong and Waters, 2001; Holderied and von Helversen, 2003; Branchi et al., 2004; Cooper and Goller, 2004; Berg et al., 2013; Dalziell et al., 2013; Hoepfner and Goller, 2013; Ota et al., 2015; Alves et al., 2016; Laplagne and Elías Costa, 2016; Ullrich et al., 2016). For example, the ‘A’, ‘B’ and ‘D’ song types of the lyrebird (Menura novaehollandiae) co-occur with courtship dance wing flaps, while the ‘C’ type occurs when wings are still (Dalziell et al., 2013). In rats (Rattus norvegicus), 50 kHz ultrasonic vocalizations are produced during locomotor activity, while 20 kHz vocalizations occur when rats are immobile (Laplagne and Elías Costa, 2016). Similarly, bats in flight coordinate the production of echolocation sounds with particular phases of their wing beats (Suthers et al., 1972; Wong and Waters, 2001; Holderied and von Helversen, 2003). Thus, the adaptive coordination achieved during vocal development must also include other developing motor systems as important factors, notably those related to posture and locomotion. How this ‘vocal-locomotor’ coordination is accomplished over the course of development is not well understood.

In humans, current theoretical frameworks posit that the different motor systems are interactive over the course of development; one system can influence another in ways that change as a function of time (Thelen, 1991; Adolph, 2008; Iverson, 2010; Adolph and Robinson, 2015; Libertus and Hauf, 2017). In support of this, there is evidence that locomotor skills at one time point predict speech ability months or even years later (LeBarton and Iverson, 2013; Walle and Campos, 2014; Wang et al., 2014; He et al., 2015; LeBarton and Iverson, 2016; Libertus and Violi, 2016; Walle, 2016; Garrido et al., 2017; Libertus and Hauf, 2017; Salavati et al., 2017; West et al., 2019). However, only a handful of empirical studies investigated human infant vocal-locomotor coordination (Ejiri and Masataka, 2001; Fagan and Iverson, 2007; Abney et al., 2014; Berger et al., 2017). These studies showed that infant production of pre-linguistic vocalization is highly sensitive to body movement. For example, infants who are beginning to transverse their environment (i.e., crawling) are less likely to vocalize during locomotion than while they are sitting (Ejiri and Masataka, 2001; Fagan and Iverson, 2007; Abney et al., 2014; Berger et al., 2017). What is missing is an understanding of how vocal and locomotor outputs are coordinated in real-time in infants and how this coordination may change over the course of development.

Key to understanding these developmental dynamics is to also identify physiological conditions that may promote (or potentially inhibit) coordination between different motor outputs. One candidate is the state of arousal, a product of the autonomic nervous system and relevant for a range of behaviors (Pfaff, 2006). An animal would be said to exhibit a high arousal state if it is more alert to sensory stimuli, more motorically active and more reactive (Pfaff, 2006). The role of arousal is essentially to allocate metabolic energy (i.e., to prepare the body for action). As it relates to vocal production, arousal modulates respiration, which in turn provides the power for vocal output. Humans, for example, exhibit an increase in arousal--as measured by heart rate--prior to speaking (Lynch et al., 1980; Linden, 1987). In developing individuals, variable and spontaneous behaviors are ubiquitous, providing the scaffolding for more complex and organized behaviors later in life (Blumberg et al., 2013). These early behaviors, including vocal output and other bodily movements, primarily reflect the interplay between the infants’ arousal states, sensorimotor coordination and biomechanical conditions (Robinson et al., 2000). Thus, investigating the relationship between arousal fluctuations and the development of vocal and locomotor behaviors may prove to be illuminating.

Using marmoset monkeys as a model, here we investigate the relationship between vocal and locomotor systems and arousal levels during postnatal development. In the vocal domain, infant marmosets spontaneously produce sequences of immature and mature vocalizations, and these are linked to real-time changes in arousal levels (Zhang and Ghazanfar, 2016). Over the course of approximately two months, infants exhibit changes in the acoustic properties of their vocalizations that reflect a transition from producing mostly immature-sounding contact calls (e.g., cries) to mature-sounding contact calls (e.g., phees) (Takahashi et al., 2015; Zhang and Ghazanfar, 2016; Teramoto et al., 2017; Zhang and Ghazanfar, 2018). As in humans (Goldstein and Ja, 2008), this transition is facilitated by, and dependent upon, social reinforcement from parents (Takahashi et al., 2015; Gultekin and Hage, 2017; Takahashi et al., 2017; Gultekin and Hage, 2018). Moreover, these parallels with human prelinguistic development occur in the same life history stage (early infancy) (Ghazanfar and Liao, 2018). In the postural and locomotor domains, marmoset monkey development transitions from immature to mature forms in a pattern that is also similar to human development (e.g., righting reflex before sitting, and crawling before walking) (Wang et al., 2014; Braun et al., 2015; Schultz-Darken et al., 2016).

We address three fundamental questions: (1) Does one motor system – vocal or postural-locomotor – mature first or do they follow an overlapping trajectory? (2) How are vocal-postural-locomotor systems coordinated and do these coordination dynamics shift across development? (3) How do real-time fluctuations in arousal relate to vocal production, locomotion and their coordination?

During their first two months of postnatal life, we measured infant marmoset behavior in a controlled context for 10 minutes approximately every 2 days. In each session, individuals were placed in a testing box in an experiment room that was outside visual and auditory range of their family groups. This brief ‘isolation’ context is a standard testing paradigm used to elicit vocalizations (Takahashi et al., 2015; Zhang and Ghazanfar, 2016) and to study the postures and locomotion of marmoset infants (Wang et al., 2014; Braun et al., 2015). The subjects were seven marmosets (three females) from three different parental pairs (two sets of twins, one set of triplets). We recorded the behaviors of each subject across ~30 sessions (28–33 per subject for a total of 220 sessions).

For vocal behaviors, we focused on two types of contact calls – cries and phees (Figure 1A). As described previously (Takahashi et al., 2015), cries are immature contact calls that have a short duration and noisy spectral properties (i.e., high Wiener entropy); phees are mature-sounding contact calls that have a longer duration and tonal spectral properties (i.e., low Wiener entropy). Cries transform into phees over the course of development (Takahashi et al., 2015; Zhang and Ghazanfar, 2016; Takahashi et al., 2017).

Figure 1

Download asset Open asset

We measured five types of postural behaviors – righting reflex, head raising, forelimb support, hindlimb support, and hanging (Wang et al., 2014; Braun et al., 2015) (Figure 1B). The righting reflex is when infants re-establish their body orientation so that their hands and feet are on the ground; head raising is when infants lift their head off the ground and look forward or up; forelimb support is when infants sit on the ground with their hands touching the ground; hindlimb support is when infants sit on the ground with their hands off the ground; hanging is when infants grasp elements in their environment (e.g., bars of the testing box) so that their hands and feet do not touch the ground.

We measured five types of locomotor behaviors – crawling, digging, jumping, climbing, and walking (Wang et al., 2014; Braun et al., 2015) (Figure 1C). Crawling is when infants move forward on the ground with their stomach touching the ground; digging is when infants move their hands back and forth across the ground; jumping is when infants push themselves off the ground or cage to move from one location to another; climbing is when infants traverse across the cage; walking is when infants traverse across the ground in a standing orientation.

Finally, for all seven infants, we concurrently measured arousal levels by acquiring heart rates during the sessions using non-invasive surface electrocardiography (Borjon et al., 2016; Zhang and Ghazanfar, 2016).

The vocal system matures before postural and locomotor systems

By measuring both vocal and postural-locomotor behaviors longitudinally in developing marmosets, we determined how these motor systems changed relative to one another. We first classified each behavior as immature or mature by measuring how their use shifted across development. In the vocal domain, the proportion of time producing cries decreased across development, while the proportion of time producing phees increased (Figure 2A; Table 1; Appendix 1.1) (Takahashi et al., 2015; Zhang and Ghazanfar, 2016). As indicated by physiology and biomechanics (Takahashi et al., 2015; Zhang and Ghazanfar, 2016; Zhang and Ghazanfar, 2018), cries were categorized as an immature contact call, and phees were categorized as a mature version of the contact call. The proportion of postural time engaged in righting reflexes decreased across development, while the proportion of postural time engaged in hindlimb support increased (Figure 2A; Table 1; Appendix 1.2). Thus, righting reflexes were categorized as an immature posture, and hindlimb support was categorized as a mature posture. The proportion of locomotor time engaged in crawling decreased across development, while the proportion of locomotor time engaged in walking increased (Figure 2A; Table 1; Appendix 1.3). Thus, crawling was categorized as an immature locomotor behavior, while walking was categorized as mature. These postural-locomotor classifications for developing marmosets are consistent with previous findings (Wang et al., 2014).

Figure 2

Download asset Open asset

Table 1

Behavior	β (SE)	T value	Adjusted P	Classification
Vocal behaviors (n = 192 observation days)
Cry	−0.0128 (0.0019)	6.88	0.0011	immature
Phee	0.0129 (0.0017)	7.43	0.0004	mature
Postural behaviors (n = 201 observation days)
Forelimb support	−0.0010 (0.0020)	0.51	>0.05	NA
Hanging	−0.0018 (0.0011)	1.67	>0.05	NA
Hindlimb support	0.0130 (0.0034)	3.78	0.0373	mature
Raising head	−0.0021 (0.0009)	2.32	>0.05	NA
Righting reflex	−0.0081 (0.0015)	5.54	0.0020	immature
Locomotor behaviors (n = 191 observation days)
Climbing	−0.0023 (0.0011)	2.10	>0.05	NA
Crawling	−0.0160 (0.0019)	8.35	0.0019	immature
Digging	0.0033 (0.0011)	3.03	>0.05	NA
Jumping	0.0005 (0.0002)	2.37	>0.05	NA
Walking	0.0147 (0.0012)	11.95	<0.0001	mature

We used the immature-mature classifications to summarize developmental changes using a ‘maturity index’ (see Materials and methods)--a value that represents the proportion of mature behaviors observed relative to all immature and mature behaviors across postnatal days. Values below 0.5 indicate that immature behavior is more common, and values above 0.5 indicate that mature behavior is more common. The population-level developmental trajectories of all three motor systems (vocal = 192 observation days, postural = 201 observation days, locomotor = 191 observation days) followed s-shape patterns that ended in maturity indices around one, which represents adult-like motor outputs (Figure 2B).

We tested whether the developmental time courses of the vocal system and postural-locomotor systems overlap (Figure 2C). The null hypothesis is that vocal and postural-locomotor systems transition from immature to mature forms around the same time. This would suggest that the development of the different motor systems reflect a global process of neural or physiological maturation (Gesell, 1929; McGraw, 1943). The alternative hypothesis is that the postural-locomotor system develops either before or after the vocal system. If postural-locomotor behaviors were to mature first, it would suggest that a developed post-cranial body is a prerequisite for producing mature contact calls.

Supporting the ‘vocal behavior develops first’ hypothesis, contact calls transitioned to a more mature form around 10 postnatal days, while postural and locomotor behavior transitioned around postnatal days 19 and 21, respectively (Figure 2D). A generalized linear mixed model (GLMM) showed that the relationship between maturity indices and postnatal day fits a logistic regression, with maturity indices increasing with age (n = 519 observation days, β ±SE = 0.21±0.03, z = 6.33, p<0.0001; Appendix 1.4). The same model showed that the vocal maturity indices were larger than the maturity indices of postural (β ±SE = −2.03±0.58, z = 3.48, p=0.0005; Appendix 1.4) and locomotor behavior (β ±SE = −2.86±0.98, z = 2.93, p=0.0034; Appendix 1.4) (Figure 2E). In other words, vocal development occurred earlier than postural and locomotor development. However, simply because vocal-postural-locomotor systems follow different trajectories does not mean that they do not interact in real-time. The question of whether vocal-locomotor coordination changes over the course of development is addressed next.

Mature contact call production and locomotor activity become increasingly coordinated during development

Infant marmosets require considerable muscular effort to produce mature contact calls (phees) (Takahashi et al., 2015; Zhang and Ghazanfar, 2016; Zhang and Ghazanfar, 2018), and adult marmosets tend to produce mature contact calls during periods of reduced locomotor activity (Borjon et al., 2016). Therefore, our overarching hypothesis was that locomotor activity inhibits mature contact call production. Using linear mixed effect models (LMMs), we found initial support for this hypothesis when examining the relationships between vocal acoustic parameters, locomotor activity and postnatal day (Figure 3A,B). Call duration was negatively associated with locomotor activity (n = 9609 calls, β ±SE = −0.56±0.13, t = 4.25, Bonferroni-Holm adjusted p=0.0001; Appendix 1.5) while Wiener entropy (higher entropy means noisier) was positively associated with locomotor activity (n = 9606 calls, β ±SE = 0.93±0.12, t = 7.68, Bonferroni-Holm adjusted p<0.0001; Appendix 1.6). In other words, on average, infant marmosets produced short, noisier calls when locomotor activity was high, and longer, more tonal adult-like calls when locomotor activity was reduced.

Figure 3 with 1 supplement see all

Download asset Open asset

Next, we tested three hypotheses about the developmental dynamics of vocal-locomotor coordination. We know that by 1–2 months of age, marmosets only produce mature sounding contact calls (Takahashi et al., 2015; Zhang and Ghazanfar, 2016; Zhang and Ghazanfar, 2018). One hypothesis is that they simply learn to stop moving when they need to produce a contact call. Doing so would eschew any potential physiological constraints on the production of mature sounding contact calls. In such a scenario, we would predict the number of contact calls produced during movement would decrease, while those produced during periods of immobility would increase. An alternative hypothesis is that as marmosets grow bigger, they become more capable of producing mature sounding contact calls while moving. This outcome would suggest that infants overcome any potential physiological constraints on vocal-locomotor coordination. In such a scenario, we would predict that the number of contact calls produced during movement would increase, while those produced during periods of immobility would decrease.

To test the above hypotheses, we first needed to estimate locomotor activity on a continuous scale. These estimates were extracted from frame-by-frame pixel differences in the video footage of the sessions (Figure 3—figure supplement 1). We first summarized the temporal real-time dynamics of locomotor activity surrounding vocal production events (Figure 3C). We found that locomotor activity started to increase above the 95% threshold of the bootstrap significance test ~5 s before the immature call onsets and then decrease back to inside the 95% threshold ~10 s after call offsets. In contrast, locomotor activity started to decrease below the 95% threshold of the bootstrap significance test ~8 s before mature contact call onsets and then increase back to inside the 95% threshold ~10 s after call offsets. A LMM confirmed that the average locomotor activity was higher during immature contact calls as compared to mature contact calls (n = 9609 calls, β ±SE = −0.07±0.01, t = 5.19, p<0.0001; Appendix 1.7). When these temporal dynamics were mapped across postnatal days, we found that locomotor activity during immature contact call production remained around or above the randomized expected levels of locomotor activity. In contrast, early in postnatal life, mature contact call production occurred when locomotor activity was below the 95% threshold of the bootstrap significance test; however, their production gradually became coordinated with increased levels of locomotor activity (Figure 3D). Locomotor activity during mature contact calls was higher during late development (PND 52–61) as compared to early development (PND 1–10) (Early = 914 calls; Late = 932 calls; β ±SE = 0.04±0.01, t = 3.05, p=0.0056; Appendix 1.8).

These data therefore support the alternative hypothesis, that the potential constraints of producing mature contact calls during movements are mitigated as the infants get older. Thus, even though postural-locomotor maturity does not appear to be a prerequisite for vocal development (Figure 2), vocal-locomotor coordination is still an important component of marmoset monkey motor development. This finding then raises the question of how real-time fluctuations in physiological condition may predict call production and locomotor activity in developing infants. Presumably, infants in elevated states of arousal are more likely to engage in these motor behaviors. The question of whether temporal associations between motor output and arousal levels change over the course of development is addressed next.

Mature call production and locomotion occurs during elevated arousal levels

We first tested hypotheses about the developmental dynamics of arousal state during marmoset contact call production. One very basic hypothesis is that the production of mature contact calls requires elevated arousal levels more than does the production of immature contact calls (Teramoto et al., 2017). Being in an elevated arousal state also means that individuals may have more respiratory power needed to generate mature sounding calls (Borjon et al., 2016; Zhang and Ghazanfar, 2016). Another hypothesis (that doesn’t preclude the first one) is that mature contact calls are more likely to occur during elevated arousal levels earlier in development. This would be consistent with computational models that suggest that younger infants are unable to effectively coordinate the elements of their vocal apparatus to produce mature contact calls and may require enhanced respiratory power to do so (Takahashi et al., 2015; Teramoto et al., 2017); indeed, there is now empirical support for the link between respiratory power and mature contact call production (Zhang and Ghazanfar, 2018). Conversely, a third hypothesis is that mature contact calls are more likely to occur during elevated arousal states later in development than earlier. This could occur if, as the infants get older, their overall motivation changes (i.e., less stressed by isolation) and so higher arousal levels are needed to motivate vocal production. An elevated arousal state may also enable respiratory power and laryngeal tension (Zhang and Ghazanfar, 2016); this could lead to better coordination between vocal production and locomotion (Figure 3C,D).

We found that mature contact calls occurred during elevated levels of arousal, a pattern that become more pronounced as development proceeded. A summary of the real-time dynamics of heart rate fluctuations in the 10 s before to 10 s after vocal production revealed that marmosets produced immature contact calls when heart rate percentiles dropped below the 95% threshold of the bootstrap significance test (Figure 4A). Production of mature contact calls, on the other hand, occurred when heart rate percentiles went above the 95% threshold of the bootstrap significance test. In both cases, changes in arousal levels were ‘global’, meaning that the change in arousal occurred well before (at least 10 s) the start of the call. A LMM confirmed that heart rate percentiles were higher during mature contact calls than during immature contact calls (n = 6215 calls, β ±SE = 3.38±1.45, t = 2.32, p=0.0276; Appendix 1.9). The developmental dynamics of arousal levels during vocal production further supported this main effect (Figure 4B). Early in postnatal life, heart rate percentiles during immature contact calls were inside the 95% threshold of the bootstrap significance test, but at around two weeks, heart rates during immature contact calls started to decrease below the 95% threshold. In contrast, mature contact calls were produced when heart rate percentiles were at, or above, the 95% threshold of the bootstrap significance test during the first two months of postnatal life. There was a marginal increase in heart rate percentiles during mature contact calls during late development (PND 52–61) as compared to early development (PND 1–10) (Early = 490 calls; Late = 780 calls; β ±SE = 6.28±3.48, t = 1.81, p=0.0795; Appendix 1.10).

Figure 4 with 1 supplement see all

Download asset Open asset

We next tested hypotheses about the developmental dynamics of arousal during locomotor activity. One basic hypothesis is that locomotion occurs during elevated arousal states, which is a pattern well supported by studies of heart rate during physically demanding tasks (Rotstein et al., 2004; Baker et al., 2008). Another hypothesis (that is not mutually exclusive of the first one) is that locomotor activity is more likely to occur during elevated arousal levels earlier in development. One interpretation of this result is that, as with vocal production, younger infants are less able to coordinate their body movements and require enhanced respiratory power to do so. Conversely, a third hypothesis is that locomotor activity is more likely to occur during elevated arousal levels later in development than earlier. As with vocal production, this could occur if, as the infants get older, their overall motivation changes (i.e., less stressed by isolation) and so higher arousal levels are needed to motivate movement. An elevated arousal state may also enable respiratory power and laryngeal tension to enable better coordination between vocal production and locomotion (Figure 3C,D).

Like vocal production, we found that locomotor activity occurred during elevated arousal levels, a pattern that appeared to become more pronounced later in development. A summary of the real-time dynamics of heart rate fluctuations in the 10 s before to 10 s after locomotion events revealed that these events occurred when heart rate percentiles were elevated above the 95% threshold of the bootstrap significance test (Figure 4C). As with vocal production, this change in arousal was ‘global’, meaning that the change in arousal happened well before (at least 10 s) the start of movement. The developmental dynamics of arousal levels during locomotor activity further supported this main effect result (Figure 4D). Early in postnatal life, heart rate percentiles during locomotor activity were within the 95% threshold of the bootstrap significance test, but continued to increase throughout the first two months of postnatal life. There was a marginal increase in heart rate percentiles during locomotor events during late development (PND 52–61) as compared to early development (PND 1–10) (Early = 752 calls; Late = 566 calls; β ±SE = 8.12±4.04, t = 2.01, p=0.0580; Appendix 1.11).

Our data suggest that mature contact call production and locomotor activity are both associated with elevated levels of arousal, an association that becomes more pronounced with age. As such, arousal state may be an important predictor of whether infant marmosets coordinate these different motor outputs. We tested this idea next.

Coordination of mature contact call production with locomotor activity occurs during elevated levels of arousal

Over the course of infant marmoset development, both locomotor activity and arousal levels predicted whether an infant produces a mature contact call over an immature call. Low levels of locomotor activity support mature contact call production early in development, and elevated arousal levels support mature contact call production later in development. The missing piece of the puzzle is whether there is a connection between vocal-locomotor coordination and arousal state. Here, we test, and find support for, the hypothesis that arousal levels during mature contact call production are elevated when marmosets are moving around (Figure 5A). Locomotor activity during mature contact call production was positively associated with heart rate percentiles (n = 3966 calls, β ±SE = 9.66±4.05, t = 2.39, p=0.0208; Appendix 1.12; Figure 5B). In other words, this result suggests that individuals in an elevated arousal state are better able to coordinate mature vocal production with locomotion. Specifically, this positive association characterized infants that were one month old (1–30 days; n = 1705 calls, β ±SE = 16.32±6.60, t = 2.47, Bonferroni-Holm adjusted p=0.0382; Appendix 1.12), but not two months old (31–61 days; n = 2261 calls, β ±SE = 2.63±5.24, t = 0.50, Bonferroni-Holm adjusted p=0.6220; Appendix 1.12). This means that a positive association between locomotor activity and heart rate percentiles was not simply a by-product of age. Instead, there may be a particularly robust relationship between locomotor activity and heart rate during early infanthood when individuals are transitioning from producing immature to mature contact calls.

Figure 5

Download asset Open asset

Vocal development is a dynamic process that involves the interaction of multiple systems, from the biomechanics and muscles of the vocal apparatus to the nervous system and the social environment (Thelen, 1991; Teramoto et al., 2017). Using developing marmoset monkeys, we sought to understand how these processes related to vocal output are influenced by other systems of motor behavior, specifically posture and locomotion. First, we examined when the transition from immature to mature calls occurs relative to the transition from immature to mature body postures and locomotion. Second, we examined the putative temporal coordination of vocal production with locomotor activity, and whether such coordination changes over the course of development. Finally, we investigated whether fluctuating arousal levels (estimated from heart rate) predicts vocal and locomotor output. We found that marmoset monkey vocalizations develop sooner than postural-locomotor skills, that locomotor activity gradually becomes coordinated with the production of mature-sounding contact calls, and that this vocal-locomotor coordination occurs during elevated levels of arousal.

As part of this full submission, we have uploaded our data to Dryad (https://doi.org/10.5061/dryad.cp75158).

1. 1. Gustison ML
   2. Borjon JI
   3. Takahashi DY
   4. Ghazanfar AA

   (2019) Dryad Digital Repository

   Data from:Vocal and locomotor coordination develops in association with arousal state.

   https://doi.org/10.5061/dryad.cp75158

1. 1. Abney DH
   2. Warlaumont AS
   3. Haussman A
   4. Ross JM
   5. Wallot S

   (2014) Using nonlinear methods to quantify changes in infant limb movements and vocalizations

   Frontiers in Psychology 5:771.

   https://doi.org/10.3389/fpsyg.2014.00771

   • PubMed
   • Google Scholar
2. 1. Adolph KE

   (2008)

   Encyclopedia of Infant and Early Childhood Development

   359–373, Motor and physical development: locomotion, Encyclopedia of Infant and Early Childhood Development, San Diego, CA, Academic Press, 10.1016/B978-012370877-9.00104-3.

   • Google Scholar
3. Book

   1. Adolph KE
   2. Robinson SR

   (2015)

   Motor development. In: Handbook of child psychology and developmental science

   In: Lerner RM, Liben L, Muller U, editors. Cognitive Processes, 2 (7). New York: Wiley. pp. 114–157.

   • Google Scholar
4. 1. Alves JA
   2. Boerner BC
   3. Laplagne DA

   (2016) Flexible coupling of respiration and vocalizations with locomotion and head movements in the freely behaving rat

   Neural Plasticity 2016:1–16.

   https://doi.org/10.1155/2016/4065073

   • Google Scholar
5. 1. Baker SE
   2. Hipp J
   3. Alessio H

   (2008) Ventilation and speech characteristics during submaximal aerobic exercise

   Journal of Speech, Language, and Hearing Research 51:1203–1214.

   https://doi.org/10.1044/1092-4388(2008/06-0223)

   • Google Scholar
6. 1. Bates D
   2. Mächler M
   3. Bolker B
   4. Walker S

   (2015) Fitting linear mixed-effects models using lme4

   Journal of Statistical Software 67:.

   https://doi.org/10.18637/jss.v067.i01

   • Google Scholar
7. 1. Berg KS
   2. Beissinger SR
   3. Bradbury JW

   (2013) Factors shaping the ontogeny of vocal signals in a wild parrot

   The Journal of Experimental Biology 216:338–345.

   https://doi.org/10.1242/jeb.073502

   • PubMed
   • Google Scholar
8. 1. Berger SE
   2. Cunsolo M
   3. Ali M
   4. Iverson JM

   (2017) The trajectory of concurrent motor and vocal behaviors over the transition to crawling in infancy

   Infancy 22:681–694.

   https://doi.org/10.1111/infa.12179

   • Google Scholar
9. 1. Blumberg MS

   (1992) Rodent ultrasonic short calls: locomotion, biomechanics, and communication

   Journal of Comparative Psychology 106:360–365.

   https://doi.org/10.1037/0735-7036.106.4.360

   • PubMed
   • Google Scholar
10. Book

    1. Blumberg MS

    (2001) The developmental context of thermal homeostasis

    In: Blass E. M, editors. Developmental Psychobiology. Handbook of Behavioral Neurobiology, 13. Boston, MA: Springer. pp. 199–228.

    https://doi.org/10.1007/978-1-4615-1209-7_6

    • Google Scholar
11. 1. Blumberg MS
    2. Marques HG
    3. Iida F

    (2013) Twitching in sensorimotor development from sleeping rats to robots

    Current Biology 23:R532–R537.

    https://doi.org/10.1016/j.cub.2013.04.075

    • PubMed
    • Google Scholar
12. 1. Blumberg MS
    2. Sokoloff G

    (1998) Thermoregulatory competence and behavioral expression in the young of altricial species--revisited

    Developmental Psychobiology 33:107–123.

    https://doi.org/10.1002/(SICI)1098-2302(199809)33:2<107::AID-DEV2>3.0.CO;2-N

    • PubMed
    • Google Scholar
13. 1. Borjon JI
    2. Takahashi DY
    3. Cervantes DC
    4. Ghazanfar AA

    (2016) Arousal dynamics drive vocal production in marmoset monkeys

    Journal of Neurophysiology 116:753–764.

    https://doi.org/10.1152/jn.00136.2016

    • PubMed
    • Google Scholar
14. 1. Bramble DM
    2. Carrier DR

    (1983) Running and breathing in mammals

    Science 219:251–256.

    https://doi.org/10.1126/science.6849136

    • PubMed
    • Google Scholar
15. 1. Branchi I
    2. Santucci D
    3. Puopolo M
    4. Alleva E

    (2004) Neonatal behaviors associated with ultrasonic vocalizations in mice (Mus musculus): A slow-motion analysis

    Developmental Psychobiology 44:37–44.

    https://doi.org/10.1002/dev.10150

    • PubMed
    • Google Scholar
16. 1. Braun K
    2. Schultz-Darken N
    3. Schneider M
    4. Moore CF
    5. Emborg ME

    (2015) Development of a novel postnatal neurobehavioral scale for evaluation of common marmoset monkeys

    American Journal of Primatology 77:401–417.

    https://doi.org/10.1002/ajp.22356

    • PubMed
    • Google Scholar
17. 1. Burkart JM
    2. Hrdy SB
    3. Van Schaik CP

    (2009) Cooperative breeding and human cognitive evolution

    Evolutionary Anthropology: Issues, News, and Reviews 18:175–186.

    https://doi.org/10.1002/evan.20222

    • Google Scholar
18. 1. Case TJ

    (1978) On the evolution and adaptive significance of postnatal growth rates in the terrestrial vertebrates

    The Quarterly Review of Biology 53:243–282.

    https://doi.org/10.1086/410622

    • PubMed
    • Google Scholar
19. 1. Cooper BG
    2. Goller F

    (2004) Multimodal signals: enhancement and constraint of song motor patterns by visual display

    Science 303:544–546.

    https://doi.org/10.1126/science.1091099

    • PubMed
    • Google Scholar
20. 1. Dalziell AH
    2. Peters RA
    3. Cockburn A
    4. Dorland AD
    5. Maisey AC
    6. Magrath RD

    (2013) Dance choreography is coordinated with song repertoire in a complex avian display

    Current Biology 23:1132–1135.

    https://doi.org/10.1016/j.cub.2013.05.018

    • PubMed
    • Google Scholar
21. 1. Derrickson EM

    (1992) Comparative reproductive strategies of altricial and precocial eutherian mammals

    Functional Ecology 6:57–65.

    https://doi.org/10.2307/2389771

    • Google Scholar
22. 1. Ejiri K
    2. Masataka N

    (2001) Co‐occurences of preverbal vocal behavior and motor action in early infancy

    Developmental Science 4:40–48.

    https://doi.org/10.1111/1467-7687.00147

    • Google Scholar
23. 1. Fagan MK
    2. Iverson JM

    (2007) The influence of mouthing on infant vocalization

    Infancy 11:191–202.

    https://doi.org/10.1111/j.1532-7078.2007.tb00222.x

    • PubMed
    • Google Scholar
24. 1. Friard O
    2. Gamba M

    (2016) BORIS: a free, versatile open-source event-logging software for video/audio coding and live observations

    Methods in Ecology and Evolution 7:1325–1330.

    https://doi.org/10.1111/2041-210X.12584

    • Google Scholar
25. 1. Fusani L
    2. Hutchinson RE
    3. Hutchinson JB

    (1996) Vocal-postural co-ordination of a sexually dimorphic display in monomorphic species: the barbary dove

    Behaviour 134:321–335.

    https://doi.org/10.1163/156853997X00566

    • Google Scholar
26. 1. Garrido D
    2. Petrova D
    3. Watson LR
    4. Garcia-Retamero R
    5. Carballo G

    (2017) Language and motor skills in siblings of children with autism spectrum disorder: a meta-analytic review

    Autism Research 10:1737–1750.

    https://doi.org/10.1002/aur.1829

    • PubMed
    • Google Scholar
27. Book

    1. Gesell A

    (1929)

    Infancy and Human Growth

    New York: The Macmillan Company.

    • Google Scholar
28. 1. Ghazanfar AA
    2. Liao DA

    (2018) Constraints and flexibility during vocal development: insights from marmoset monkeys

    Current Opinion in Behavioral Sciences 21:27–32.

    https://doi.org/10.1016/j.cobeha.2017.11.015

    • PubMed
    • Google Scholar
29. Book

    1. Ghazanfar AA
    2. Takahashi DY

    (2017)

    The evo-devo of vocal communication: Insights from marmoset monkeys

    In: Kaas JH, editors. The Evolution of Nervous Systems (2nd). Elsevier Press. pp. 317–324.

    • Google Scholar
30. 1. Goldstein MH
    2. Ja S

    (2008) Social feedback to babbling facilitates vocal learning

    Psychological Science 19:515–523.

    https://doi.org/10.1111/j.1467-9280.2008.02117.x

    • Google Scholar
31. 1. Gros-Louis J
    2. West MJ
    3. Goldstein MH
    4. King AP

    (2006) Mothers provide differential feedback to infants' prelinguistic sounds

    International Journal of Behavioral Development 30:509–516.

    https://doi.org/10.1177/0165025406071914

    • Google Scholar
32. 1. Gultekin YB
    2. Hage SR

    (2017) Limiting parental feedback disrupts vocal development in marmoset monkeys

    Nature Communications 8:14046.

    https://doi.org/10.1038/ncomms14046

    • PubMed
    • Google Scholar
33. 1. Gultekin YB
    2. Hage SR

    (2018) Limiting parental interaction during vocal development affects acoustic call structure in marmoset monkeys

    Science Advances 4:eaar4012.

    https://doi.org/10.1126/sciadv.aar4012

    • PubMed
    • Google Scholar
34. 1. He M
    2. Walle EA
    3. Campos JJ

    (2015) A Cross-National investigation of the relationship between infant walking and language development

    Infancy 20:283–305.

    https://doi.org/10.1111/infa.12071

    • Google Scholar
35. 1. Hoepfner AR
    2. Goller F

    (2013) Atypical song reveals spontaneously developing coordination between multi-modal signals in brown-headed cowbirds (Molothrus ater)

    PLOS ONE 8:e65525.

    https://doi.org/10.1371/journal.pone.0065525

    • PubMed
    • Google Scholar
36. 1. Holderied MW
    2. von Helversen O

    (2003) Echolocation range and wingbeat period match in aerial-hawking bats

    Proceedings of the Royal Society of London. Series B: Biological Sciences 270:2293–2299.

    https://doi.org/10.1098/rspb.2003.2487

    • PubMed
    • Google Scholar
37. Book

    1. Hrdy SB

    (2007) Evolutionary context of human development: The cooperative breeding model

    In: Salmon C. A, Shackelford T. K, editors. Family Relationships: An Evolutionary Perspective. Oxford: Oxford University Press. pp. 9–32.

    https://doi.org/10.1093/acprof:oso/9780195320510.003.0003

    • Google Scholar
38. 1. Iverson JM

    (2010) Developing language in a developing body: the relationship between motor development and language development

    Journal of Child Language 37:229–261.

    https://doi.org/10.1017/S0305000909990432

    • PubMed
    • Google Scholar
39. 1. Kuznetsova A
    2. Brockhoff PB
    3. Christensen RHB

    (2017) lmerTest Package: Tests in Linear Mixed Effects Models

    Journal of Statistical Software 82:.

    https://doi.org/10.18637/jss.v082.i13

    • Google Scholar
40. 1. Laplagne DA
    2. Elías Costa M

    (2016) Rats synchronize locomotion with ultrasonic vocalizations at the subsecond time scale

    Frontiers in Behavioral Neuroscience 10:184.

    https://doi.org/10.3389/fnbeh.2016.00184

    • PubMed
    • Google Scholar
41. 1. LeBarton ES
    2. Iverson JM

    (2013) Fine motor skill predicts expressive language in infant siblings of children with autism

    Developmental Science 16:815–827.

    https://doi.org/10.1111/desc.12069

    • PubMed
    • Google Scholar
42. 1. LeBarton ES
    2. Iverson JM

    (2016) Associations between gross motor and communicative development in at-risk infants

    Infant Behavior and Development 44:59–67.

    https://doi.org/10.1016/j.infbeh.2016.05.003

    • PubMed
    • Google Scholar
43. 1. Libertus K
    2. Hauf P

    (2017) Editorial: motor skills and their foundational role for perceptual, social, and cognitive development

    Frontiers in Psychology 8:301.

    https://doi.org/10.3389/fpsyg.2017.00301

    • PubMed
    • Google Scholar
44. 1. Libertus K
    2. Violi DA

    (2016) Sit to talk: relation between motor skills and language development in infancy

    Frontiers in Psychology 7:475.

    https://doi.org/10.3389/fpsyg.2016.00475

    • PubMed
    • Google Scholar
45. 1. Linden W

    (1987) A microanalysis of autonomic activity during human speech

    Psychosomatic Medicine 49:562–578.

    https://doi.org/10.1097/00006842-198711000-00002

    • PubMed
    • Google Scholar
46. 1. Locke JL

    (2006) Parental selection of vocal behavior : crying, cooing, babbling, and the evolution of language

    Human Nature 17:155–168.

    https://doi.org/10.1007/s12110-006-1015-x

    • PubMed
    • Google Scholar
47. 1. Lukas D
    2. Clutton-Brock T

    (2012) Cooperative breeding and monogamy in mammalian societies

    Proceedings of the Royal Society B: Biological Sciences 279:2151–2156.

    https://doi.org/10.1098/rspb.2011.2468

    • Google Scholar
48. 1. Lynch JJ
    2. Thomas SA
    3. Long JM
    4. Malinow KL
    5. Chickadonz G
    6. Katcher AH

    (1980) Human speech and blood pressure

    The Journal of Nervous and Mental Disease 168:526–534.

    https://doi.org/10.1097/00005053-198009000-00002

    • PubMed
    • Google Scholar
49. 1. MacLarnon AM
    2. Hewitt GP

    (1999) The evolution of human speech: the role of enhanced breathing control

    American Journal of Physical Anthropology 109:341–363.

    https://doi.org/10.1002/(SICI)1096-8644(199907)109:3<341::AID-AJPA5>3.0.CO;2-2

    • PubMed
    • Google Scholar
50. 1. Maclarnon A
    2. Hewitt G

    (2004) Increased breathing control: another factor in the evolution of human language

    Evolutionary Anthropology: Issues, News, and Reviews 13:181–197.

    https://doi.org/10.1002/evan.20032

    • Google Scholar
51. Book

    1. McGraw M

    (1943)

    The Neuromuscular Maturation of the Human Infant

    New York: Columbia University Press.

    • Google Scholar
52. 1. Muir GD

    (2000) Early ontogeny of locomotor behaviour: a comparison between altricial and precocial animals

    Brain Research Bulletin 53:719–726.

    https://doi.org/10.1016/S0361-9230(00)00404-4

    • PubMed
    • Google Scholar
53. 1. Orlikoff RF

    (2008) Voice production during a weightlifting and support task

    Folia Phoniatrica Et Logopaedica 60:188–194.

    https://doi.org/10.1159/000128277

    • PubMed
    • Google Scholar
54. 1. Ota N
    2. Gahr M
    3. Soma M

    (2015) Tap dancing birds: the multimodal mutual courtship display of males and females in a socially monogamous songbird

    Scientific Reports 5:6–11.

    https://doi.org/10.1038/srep16614

    • Google Scholar
55. Book

    1. Pfaff D

    (2006)

    Brain Arousal and Information Theory

    Cambridge, MA: Harvard University Press.

    • Google Scholar
56. 1. Porges SW
    2. Furman SA

    (2011) The early development of the autonomic nervous system provides a neural platform for social behavior: a polyvagal perspective

    Infant and Child Development 20:106–118.

    https://doi.org/10.1002/icd.688

    • PubMed
    • Google Scholar
57. 1. Price KJ
    2. Lin M
    3. Feng J
    4. Goldman R
    5. Sears A
    6. Jacko JA

    (2006) Motion does matter: an examination of speech-based text entry on the move

    Universal Access in the Information Society 4:246–257.

    https://doi.org/10.1007/s10209-005-0006-8

    • Google Scholar
58. 1. Price ER
    2. Dzialowski EM

    (2018) Development of endothermy in birds: patterns and mechanisms

    Journal of Comparative Physiology B 188:373–391.

    https://doi.org/10.1007/s00360-017-1135-0

    • Google Scholar
59. 1. Robinson SR
    2. Blumberg MS
    3. Lane MS
    4. Kreber LA

    (2000) Spontaneous motor activity in fetal and infant rats is organized into discrete multilimb bouts

    Behavioral Neuroscience 114:328–336.

    https://doi.org/10.1037/0735-7044.114.2.328

    • PubMed
    • Google Scholar
60. Book

    1. Rosenblatt JS

    (1976)

    Stages in the early behavioural development of altricial young of selected species of non-primate mammals

    In: Bateson PPG, Hinde RA, editors. Growing Points Ethology. Cambridge: Cambridge University Press. pp. 345–381.

    • Google Scholar
61. 1. Rotstein A
    2. Meckel Y
    3. Inbar O

    (2004) Perceived speech difficulty during exercise and its relation to exercise intensity and physiological responses

    European Journal of Applied Physiology 92:431–436.

    https://doi.org/10.1007/s00421-004-1160-z

    • PubMed
    • Google Scholar
62. 1. Salavati S
    2. Einspieler C
    3. Vagelli G
    4. Zhang D
    5. Pansy J
    6. Burgerhof JGM
    7. Marschik PB
    8. Bos AF

    (2017) The association between the early motor repertoire and language development in term children born after normal pregnancy

    Early Human Development 111:30–35.

    https://doi.org/10.1016/j.earlhumdev.2017.05.006

    • PubMed
    • Google Scholar
63. 1. Schilling N

    (2005) Ontogenetic development of locomotion in small mammals--a kinematic study

    Journal of Experimental Biology 208:4013–4034.

    https://doi.org/10.1242/jeb.01875

    • PubMed
    • Google Scholar
64. 1. Schultz-Darken N
    2. Braun KM
    3. Emborg ME

    (2016) Neurobehavioral development of common marmoset monkeys

    Developmental Psychobiology 58:141–158.

    https://doi.org/10.1002/dev.21360

    • PubMed
    • Google Scholar
65. 1. Snowdon CT

    (1996) Infant care in cooperatively breeding species

    Advances in the Study of Behavior 25:643–688.

    https://doi.org/10.1016/S0065-3454(08)60345-9

    • Google Scholar
66. Book

    1. Solomon NG
    2. French JA

    (2007)

    Cooperative Breeding in Mammals

    Cambridge: Cambridge University Press.

    • Google Scholar
67. Book

    1. Starck JM
    2. Ricklefs RE

    (1998)

    Patterns of development: The altricial-precocial spectrum

    In: Starck JM, Ricklefs RE, editors. Avian Growth and Development. Evolution Within the Altricial Precocial Spectrum. New York: Oxford University Press. pp. 3–30.

    • Google Scholar
68. 1. Sundberg J
    2. Leanderson R
    3. von Euler C
    4. Knutsson E

    (1991) Influence of body posture and lung volume on subglottal pressure control during singing

    Journal of Voice 5:283–291.

    https://doi.org/10.1016/S0892-1997(05)80057-8

    • Google Scholar
69. 1. Suthers RA
    2. Thomas SP
    3. Suthers BJ

    (1972)

    Respiration, wing-beat and ultrasonic pulse emission in an echo-locating bat

    The Journal of Experimental Biology 56:37–48.

    • Google Scholar
70. 1. Takahashi DY
    2. Fenley AR
    3. Teramoto Y
    4. Narayanan DZ
    5. Borjon JI
    6. Holmes P
    7. Ghazanfar AA

    (2015) The developmental dynamics of marmoset monkey vocal production

    Science 349:734–738.

    https://doi.org/10.1126/science.aab1058

    • Google Scholar
71. 1. Takahashi DY
    2. Fenley AR
    3. Ghazanfar AA

    (2016) Early development of turn-taking with parents shapes vocal acoustics in infant marmoset monkeys

    Philosophical Transactions of the Royal Society B: Biological Sciences 371:20150370.

    https://doi.org/10.1098/rstb.2015.0370

    • Google Scholar
72. 1. Takahashi DY
    2. Liao DA
    3. Ghazanfar AA

    (2017) Vocal learning via social reinforcement by infant marmoset monkeys

    Current Biology 27:1844–1852.

    https://doi.org/10.1016/j.cub.2017.05.004

    • PubMed
    • Google Scholar
73. 1. Tchernichovski O
    2. Nottebohm F
    3. Ho CE
    4. Pesaran B
    5. Mitra PP

    (2000) A procedure for an automated measurement of song similarity

    Animal Behaviour 59:1167–1176.

    https://doi.org/10.1006/anbe.1999.1416

    • PubMed
    • Google Scholar
74. 1. Teramoto Y
    2. Takahashi DY
    3. Holmes P
    4. Ghazanfar AA

    (2017) Vocal development in a Waddington landscape

    eLife 6:e20782.

    https://doi.org/10.7554/eLife.20782

    • PubMed
    • Google Scholar
75. Book

    1. Thelen E

    (1991)

    Motor aspects of emergent speech: A dynamic approach

    In: Krasnegor NA, Rumbaugh DM, Schiefelbusch RL, Studdert-Kennedy M, editors. Biological and Behavioral Determinants of Language Development. Hillsdale, NJ: Lawrence Erlbaum Associates. pp. 339–362.

    • Google Scholar
76. 1. Tipps ST
    2. Mira MP
    3. Cairns GF

    (1981)

    Concurrent tracking of infant motor and speech development

    Genetic Psychology Monographs 104:303–324.

    • PubMed
    • Google Scholar
77. 1. Ullrich R
    2. Norton P
    3. Scharff C

    (2016) Waltzing Taeniopygia : integration of courtship song and dance in the domesticated australian zebra finch

    Animal Behaviour 112:285–300.

    https://doi.org/10.1016/j.anbehav.2015.11.012

    • Google Scholar
78. 1. Vihman MM

    (2014)

    The First Two Years

    Phonological development, The First Two Years, 2nd Edition, Chichester, UK, John Wiley.

    • Google Scholar
79. 1. Walle EA

    (2016) Infant social development across the transition from crawling to walking

    Frontiers in Psychology 7:960.

    https://doi.org/10.3389/fpsyg.2016.00960

    • PubMed
    • Google Scholar
80. 1. Walle EA
    2. Campos JJ

    (2014) Infant language development is related to the acquisition of walking

    Developmental Psychology 50:336–348.

    https://doi.org/10.1037/a0033238

    • PubMed
    • Google Scholar
81. 1. Wang Y
    2. Fang Q
    3. Gong N

    (2014) Motor assessment of developing common marmosets

    Neuroscience bulletin 30:387–393.

    https://doi.org/10.1007/s12264-013-1395-y

    • PubMed
    • Google Scholar
82. 1. Wasserman K
    2. Whipp BJ
    3. Koyl SN
    4. Beaver WL

    (1973) Anaerobic threshold and respiratory gas exchange during exercise

    Journal of Applied Physiology 35:236–243.

    https://doi.org/10.1152/jappl.1973.35.2.236

    • PubMed
    • Google Scholar
83. 1. West KL
    2. Leezenbaum NB
    3. Northrup JB
    4. Iverson JM

    (2019) The relation between walking and language in infant siblings of children with autism spectrum disorder

    Child Development 90:.

    https://doi.org/10.1111/cdev.12980

    • PubMed
    • Google Scholar
84. 1. Williams H

    (2001)

    Choreography of song, dance and beak movements in the zebra finch (Taeniopygia guttata)

    The Journal of Experimental Biology 204:3497–3506.

    • PubMed
    • Google Scholar
85. 1. Wong JG
    2. Waters DA

    (2001)

    The synchronisation of signal emission with wingbeat during the approach phase in soprano pipistrelles (Pipistrellus pygmaeus)

    The Journal of Experimental Biology 204:575–583.

    • PubMed
    • Google Scholar
86. Book

    1. Yingling JM

    (1981)

    Temporal Features of Infant Speech: A Description of Babbling Patterns Circumscribed by Postural Achievement

    University of Denver.

    • Google Scholar
87. 1. Zhang YS
    2. Ghazanfar AA

    (2016) Perinatally influenced autonomic system fluctuations drive infant vocal sequences

    Current Biology 26:1249–1260.

    https://doi.org/10.1016/j.cub.2016.03.023

    • PubMed
    • Google Scholar
88. 1. Zhang YS
    2. Ghazanfar AA

    (2018) Vocal development through morphological computation

    PLOS Biology 16:e2003933.

    https://doi.org/10.1371/journal.pbio.2003933

    • PubMed
    • Google Scholar
89. Book

    1. Zuberbühler K

    (2012)

    Cooperative breeding and the evolution of vocal flexibility

    In: Tallerman M, Gibson K. R, editors. The Oxford Handbook of Language Evolution. Oxford: Oxford University Press. pp. 71–81.

    • Google Scholar

Author details

1. Morgan L Gustison

   Princeton Neuroscience Institute, Princeton University, Princeton, United States

   Present address

   Department of Integrative Biology, University of Texas at Austin, Austin, United States

   Contribution

   Data curation, Formal analysis, Investigation, Writing—original draft

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-1162-8966

2. Jeremy I Borjon

   1. Princeton Neuroscience Institute, Princeton University, Princeton, United States
   2. Department of Psychology, Princeton University, Princeton, United States

   Contribution

   Data curation, Investigation, Writing—review and editing

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0001-9114-3362

3. Daniel Y Takahashi

   1. Princeton Neuroscience Institute, Princeton University, Princeton, United States
   2. Department of Psychology, Princeton University, Princeton, United States

   Contribution

   Conceptualization, Data curation, Validation, Investigation, Methodology, Writing—review and editing

   For correspondence

   takahashiyd@gmail.com

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0003-4972-001X

4. Asif A Ghazanfar

   1. Princeton Neuroscience Institute, Princeton University, Princeton, United States
   2. Department of Psychology, Princeton University, Princeton, United States
   3. Department of Ecology and Evolutionary Biology, Princeton University, Princeton, United States

   Contribution

   Conceptualization, Supervision, Funding acquisition, Validation, Writing—original draft, Writing—review and editing

   For correspondence

   asifg@princeton.edu

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0003-1960-7470

• 1,103

  views

• 179

  downloads

• 10

  citations

Views, downloads and citations are aggregated across all versions of this paper published by eLife.