Available via license: CC BY-NC-ND 4.0
Content may be subject to copyright.
Infancy
-
RESEARCH ARTICLE
OPEN ACCESS
Forms and Functions of Gestures in Preverbal 12‐ to 15‐
Months Old Infants
Shreejata Gupta
1,2
| Eulalie Pequay
1
| Clément François
2
| Isabelle Dautriche
1
1
Centre de Recherche en Psychologie et Neurosciences (CRPN), CNRS, Aix‐Marseille Université, Marseille, France |
2
Laboratoire Parole et Langage (LPL),
CNRS, Aix‐Marseille Université, Aix‐en‐Provence, France
Correspondence: Shreejata Gupta (shreejata@gmail.com)
Received: 12 March 2024 | Revised: 21 October 2024 | Accepted: 27 November 2024
Funding: This work, carried out within the Institute of Convergence ILCB (ANR‐16‐CONV‐0002), has beneted from support from the French government
(France 2030), managed by the French National Agency for Research (ANR) and the Excellence Initiative of Aix‐Marseille University (A*MIDEX). This
research has been supported by a grant from the Fondation de France (Grant 00147512/WB‐2023‐51888).
Keywords: co‐speech gestures | comparative methods | gesture development | language acquisition | ontogeny
ABSTRACT
Speech and co‐speech gestures always go hand in hand. Whether we nd the precursors of these co‐speech gestures in infants
before they master their native language still remains an open question. Except for deictic gestures, there is little agreement on
the existence of iconic, non‐referential and conventional gestures before children start producing their rst words. Here, we
bridge this knowledge gap by leveraging an ethological method already established for describing speech independent gestures
in nonhuman primates, to analyze the spontaneous gestures produced by infants when interacting with their caregivers. We
manually annotated video recordings of infant‐caregiver interactions (26 h) from the CHILDES platform, to describe the gesture
forms, types and functions in six infants from 12 to 15 months of age. We describe 62 gesture forms in the preverbal repertoire.
These were categorized into deictic, iconic, non‐referential and conventional gesture types, similar to co‐speech gesture types.
We also nd that the type‐function relation of preverbal gestures map similarly to type‐meaning relation of co‐speech gestures.
Taken together, our results illustrate linguistic properties of infant gestures in the absence of speech, suggesting them to be
precursors of co‐speech gestures.
1
|
Introduction
Across cultures and age groups, we all gesture while speaking
(Hewes 1974). These co‐speech gestures play a crucial role in
enhancing communication, either by complementing the spoken
message or conveying information that is not expressed through
speech alone (Goldin‐Meadow 1999). They also facilitate speech
comprehension (Kelly et al. 2015) and shape social interactions
(Holler and Levinson 2019; Clark 1996). Co‐speech gestures can
be broadly categorized into four types (Kendon 2004;
McNeill 1992; Cartmill and Goldin‐Meadow 2016): deictic ges-
tures, such as pointing, indicate the position of a referent in the
conversational space; iconic gestures, such as pretending to sip
from a glass when uttering “do you want water”, transparently
reect certain aspects of the meaning that is being conveyed;
conventional or emblematic gestures, such as a “thumb up” to
mean “okay”, have conventionalized meanings
1
that are not
iconically related to the form of the gesture. Finally, non‐
referential gestures (also known as beat gestures), are rhythmic
movements that are aligned with the speech prosody. While
deictic, iconic, and conventional gestures make semantic contri-
butions to speech, non‐referential gestures do not. Rather, the
latter have pragmatic functions used to punctuate the structure of
the discourse (Ekman and Friesen 1969; Bavelas et al. 1992). They
This is an open access article under the terms of the Creative Commons Attribution‐NonCommercial‐NoDerivs License, which permits use and distribution in any medium, provided the original work
is properly cited, the use is non‐commercial and no modications or adaptations are made.
© 2024 The Author(s). Infancy published by Wiley Periodicals LLC on behalf of International Congress of Infant Studies.
Infancy, 2025; 30:e12645 1 of 18
https://doi.org/10.1111/infa.12645
act as markers of speech‐acts, indicating the illocutionary act that
a speaker is engaged in (Kendon 1995,2004). Although extensive
research has examined the meanings of co‐speech gestures (for a
review, see Wagner, Malisz, and Kopp 2014; Kendon 2004;
McNeill 1992; Bavelas et al. 1992), their developmental trajectory
in relation to language acquisition remains poorly explored. This
is largely because understanding the meaning of these gestures
depends not only on interpreting the gestures themselves but also
on understanding the accompanying speech and the relationship
between the two. A key open question is thus whether early
precursors of these co‐speech gestures emerge during the pre-
verbal stage, before infants fully master their native language. The
aims of this study are to identify the forms, types and functions of
infant gestures, and thereafter, to test whether the different
gesture types found in infants map on to similar semantic and
pragmatic functions as those found in adult co‐speech gestures.
Infants use intentional gestures to communicate before they
start speaking; deictic gestures, especially pointing, develop in
infants typically between 9 and 12 months of age (Bates,
Benigni, and Bretherton 1979; Bates 1974). Infants point to
request objects in their vicinity (imperative pointing) (Rohlng,
Grimminger, and Lüke 2017) and even to refer to absent ob-
jects encountered in the past (Bohn et al. 2018). Pointing as-
sumes more complex functions, such as attracting others'
attention toward an object or event (declarative pointing)
(Liszkowski et al. 2004,2007) or informing or interrogating
others and learning new labels (interrogative pointing) (Lucca
and Wilbourn 2019; Kovács et al. 2014; Begus and South-
gate 2012; Tomasello, Carpenter, and Liszkowski 2007; Lisz-
kowski et al. 2006). Besides pointing, infants produce other
deictic gestures (e.g., giving, and showing) that successfully
elicit responses in adult social partners thus suggesting
clear imperative and declarative communicative functions (Orr
and Kashy Rosenbaum 2023; Karasik, Tamis‐LeMonda, and
Adolph 2014; Beuker et al. 2013; Goldin‐Meadow and Ali-
bali 2013; Kirk et al. 2013; Tomasello, Carpenter, and Lisz-
kowski 2007; Tamis‐LeMonda, Bornstein, and Baumwell 2001;
Carpenter et al. 1998).
In contrast to deictic gestures, there is no consensus on the
onset of iconic gestures. Some studies suggest that iconic ges-
tures are produced as early as 11‐ to 12‐months (L. P. Acredolo
and Goodwyn 1985; L. Acredolo and Goodwyn 1988), while
others suggest that iconic gestures only emerge from between 22
and 27 months of age (Behne, Carpenter, and Tomasello 2014;
Özçalışkan and Goldin‐Meadow 2011). For example, L. P.
Acredolo and Goodwyn (1985) found that 12‐months‐olds raise
their arm to mean “big” while Özçalışkan and Goldin‐
Meadow (2011) reported a similar gesture (holding the hand
above head to indicate “tall”), only emerging around 26 months.
The resulting discrepancy may be due to methodological dif-
ferences: while the rst study (L. P. Acredolo and Good-
wyn 1985) primarily reported on gesture production by 11‐
month‐olds from individual case studies and interviews with
caregivers, the second study (Özçalışkan and Goldin‐
Meadow 2011) analyzed direct video recordings of infants aged
14–34 months.
In this paper, we maintain a clear operational distinction be-
tween iconic and conventional gestures—we categorize gestures
as iconic when their forms resemble an action or an object
(McNeill 1992); we categorize gestures as conventional only
when there is an arbitrary relationship between their forms and
meanings and have shared conventional meanings (Ekman and
Friesen 1969; Kendon 2004; McNeill 1992; Cartmill and Goldin‐
Meadow 2016; Cartmill, Beilock, and Goldin‐Meadow 2012).
The denition of conventional gestures has been operational-
ized differently across studies using the terms “iconic” and
“symbolic” interchangeably (Goodwyn, Acredolo, and
Brown 2000; L. Acredolo and Goodwyn 1988). Caselli
et al. (2012), on the other hand, describe a list of conventional
gestures produced in conjunction with words at 18‐ to 24‐
months of age that includes both arbitrary conventional ges-
tures, such as waving the hand for bye‐bye, and iconic gestures
that resemble actions, such as giving a slap or sleeping by
leaning the head on a hand or pillow and closing the eyes. Such
interchangeable use of the names may have arisen from the fact
that some iconic gesture forms evolve into arbitrary conven-
tional gestures over time (Werner and Kaplan 1963). This leads
to a lack of clarity about the onset of conventional gestures in
infants (Caselli et al. 2012; Bates and Dick 2002). Moreover, we
do not consider nominal gestures (Mishra et al. 2021) which are
similar to “adaptors” (Ekman and Friesen 1969) as these are not
produced with any communicative intent, or at least the
communicative intent is not always evident (see Cartmill and
Goldin‐Meadow 2016).
Finally, non‐referential gestures, may emerge around 2–3 years
of age, when language is already in place, and are mostly
associated with pragmatic properties of communication (Vila‐
Gimenez and Prieto 2021; Levy and McNeill 2013; Nicoladis,
Mayberry, and Genesee 1999; Bavelas et al. 1992). In 5‐ to 6‐
year‐olds, these non‐referential gestures carry distinct roles of
information structuring (brief biphasic palm‐turns to emphasize
words in a narrative; Rohrer et al. 2022). Some non‐referential
gestures have been argued to appear earlier: from 9 months of
age, infants produce rhythmic hand movements synchronized
with their vocalizations (Iverson and Fagan 2004; Iverson and
Thelen 1999). However, it is unclear at this stage what functions
these movements serve in communication and whether they can
be compared to adult gestures. Throughout the manuscript, we
follow the broad category “non‐referential gestures”
(McNeill 1992; Prieto et al. 2018) for those which do not have a
direct referent. We deliberately avoid calling them “beat ges-
tures” as it is dened top‐down in relation to speech (Prieto
et al. 2018), and we do not want to presuppose that “beat ges-
tures” occur in preverbal infants. Furthermore, even in adults,
beats are now understood to be only a subset of a larger category
of other non‐referential co‐speech gestures (Prieto et al. 2018).
In summary, with the exception of deictic gestures, there is a
large variation in the reported onset of iconic, conventional and
non‐referential gestures. Although these categories of gestures
are sometimes reported to appear preverbally (Green
et al. 2023), it remains unclear whether these are precursors to
the co‐speech gesture types found in adults. This is because the
type and the function of gestures are often determined by the
accompanying speech or by the knowledge of cultural conven-
tions (in the case of conventional gestures). Even apparently
transparent gestures such as deictic gestures rely heavily on the
shared linguistic and social context to be understood
2 of 18 Infancy, 2025
(Clark 1996). For instance, when a child points at an apple
without producing any word, it could either be interpreted as
that she wants the apple (imperative function) or she wants to
show that there is an apple (declarative functions); in their one‐
word phase, however, if a child points to an apple and says
“give” (Butcher, Goldin‐Meadow, and McNeill 2000), the
imperative function of the pointing is clear. Such a top‐down
approach makes it difcult to identify the precursor of co‐
speech gestures in the absence of speech in early infancy.
In contexts where there is no linguistic information, such as in
non‐human primate communication, the functions of gestures
can be nevertheless inferred from the situational and social
contexts by observing the behavior of both the signaler and the
receiver. In these studies, the function of a gesture in the pre-
vailing context can be deduced based on whether the receiver's
reaction satises the signaler. The termination of further
gesturing from the signaler is considered to have satised the
signaler and is termed an apparently satisfactory outcome (ASO,
Hobaiter and Byrne 2014). This method has revealed that pri-
mate gestures are produced exibly and intentionally (Call and
Tomasello 2020), two key properties of human language (Cor-
ballis 2002). However, deictic, iconic, or non‐referential gestures
do not naturally occur in primate communication and thus
seem to be unique to humans (but see Wilke et al. 2022; Krause
et al. 2018; Douglas and Moscovice 2015; Gupta and Sinha 2016;
Pika and Mitani 2009; Gómez 2005; Leavens and Hopkins 1999).
Therefore, it is still unknown whether the different types of co‐
speech gestures only emerge after mastering spoken language,
or if they appear earlier in infancy.
In the present study, we adapt the ethological method originally
developed for primates to analyze the spontaneous gestures
produced by 12‐ to 15‐month‐old infants during interactions
with their caregivers (Cartmill and Byrne 2010; Genty
et al. 2009; Graham et al. 2018; Grund et al. 2024; Gupta and
Sinha 2019; Molesti, Meguerditchian, and Bourjade 2020). Un-
like previous research that primarily relied on top‐down ap-
proaches to identify gestures—such as parental reports (e.g., L.
Acredolo and Goodwyn 1988) or on gestures accompanied by
speech (e.g., Özçalışkan and Goldin‐Meadow 2011)—our
approach offers a systematic and comprehensive documentation
of gesture inventories and their functions, independent of
speech‐related or functional classications. Recent studies have
used this ethological method to compare gesture forms in apes
and human infants (Kersken et al. 2019; Rodrigues, Maroco, and
Frota 2021). However, these studies did not analyze infants'
preverbal gesture functions (but see Kersken et al. 2019, who
analyzed gesture exibility in terms of the number of social
goals). Here, we aim to ll this gap. Specically, we (1) describe
longitudinally the gesture forms produced by infants, and
compare them with the repertoire of gesture forms reported in
previous studies (Rodrigues, Maroco, and Frota 2021; Kersken
et al. 2019); (2) Dene the type of these gestures (deictic, iconic,
non‐referential, conventional) and (3) dene their semantic and
pragmatic functions.
Our objective is to test whether the different types of gestures
found in infants are mapped to similar semantic and pragmatic
functions as those found in adult co‐speech gestures. In adults,
the functions of co‐speech gestures are strictly tied to their
types, which are directly derived from speech: referential
(deictic, iconic and conventional) gestures are semantic, while
non‐referential ones are pragmatic. However, this top‐down
approach cannot be applied to classify preverbal infant ges-
tures for two reasons. First, infant gestures may lack spoken
reference, making it impossible to use the meaning of accom-
panying speech to derive the type and function of the gesture.
And second, except for deictic gestures, there is limited under-
standing of the other types and functions of gestures in the in-
fant's repertoire to use this in a top‐down approach. Therefore,
we assume that all preverbal gesture forms can be categorized
into multiple gesture types, and simultaneously have semantic
and pragmatic functions. This allows for a ground‐up emer-
gence of the relation between infant gesture forms, types and
their functions, without any presuppositions. Once established,
this relation can be compared to co‐speech gesture type‐function
relationships found in adults. In particular, we predict that if the
preverbal gestures contain the precursors of linguistic features
of adult co‐speech gestures, we will nd that (i) referential
gestures in the preverbal stage have a wider span of semantic
functions than non‐referential gestures, as referential co‐speech
gestures are found to bring semantic diversity to speech (Ken-
don 2004); (ii) non‐referential gestures in infants have more
declarative pragmatic functions than referential gestures, thus
acting as precursors to narrative performance of beat gestures in
children and adults (Vila‐Gimenez and Prieto 2021; Levy and
McNeill 2013; Nicoladis, Mayberry, and Genesee 1999).
2
|
Method
2.1
|
Data
Video recordings of spontaneous infant‐caregiver interactions
were taken from two open‐source corpora from CHILDES ‐the
PhonBank French‐Paris and the PhonBank French‐Lyon
corpora (videos and transcriptions can be downloaded at
https://phon.talkbank.org/access/French/Paris.html;https://
phon.talkbank.org/access/French/Lyon.html). For our study,
we used longitudinally collected recordings from 12‐ to 15‐
months of six infants—Antoine, Madeleine, and Theophile
(French‐Paris corpus); Anaïs, Marie, and Nathan (French‐Lyon
corpus). This resulted in 25.74 h of videos. We chose these
corpora for two reasons: rst, selecting a French corpus facili-
tated working with French‐speaking students and second,
several of our colleagues have already studied speech and non‐
verbal signal development in these two corpora, providing us
with an opportunity for future comprehensive cross‐study
analyses.
2.1.1
|
French‐Paris Corpus
This corpus contains video and audio recordings. These re-
cordings have been investigated primarily from the perspective
of language development, although deictic gestures were coded
(Morgenstern and Parisse 2012). The infants were lmed at
home about once a month for an hour in contexts of playing,
taking a bath and having dinner. The three infants that we
studied from this corpus had consistent video recordings from
3 of 18
12‐ to 15‐months of age. Moreover, their linguistic competence
(measured as the number of word types per hour of recording &
number of words per hour of recording), became comparable by
the age of 3‐years, although with individual differences at a
younger age (Morgenstern and Parisse 2012).
2.1.2
|
French‐Lyon Corpus
This corpus consists of video and audio recordings previously
analyzed from the perspective of speech and grammar develop-
ment (Demuth and Tremblay 2008). The children and their par-
ents were recorded at home for approximately one hour every two
weeks in similar contexts as those in the French Paris corpus. The
three infants that we studied from this corpus had the best quality
video recordings (Jourdain 2022). Their linguistic competence
(measured as the number of utterances) was comparable between
2‐ to 3‐years of age, although Anaïs had an exceptionally high rate
of utterance production (Jourdain 2020,2022).
2.1.3
|
Ethical Statement
We used pre‐existing open‐source video corpora, and we abided
by all the ethical requirements set by the data sharing platform.
2.2
|
Coding Scheme
We used the ELAN video coding software (Version 6.7, Max
Planck Institute for Psycholinguistics 2023) to code the form,
type, and semantic‐pragmatic functions of gestures occurring
during spontaneous infants‐caregiver interactions. A schematic
representation of the consecutive steps followed for coding the
videos is provided in Figure 1. The coding process can be
described in four steps:
Step 1. Identify Gestures
To identify gestures, we observed the infant's body pos-
tures and/or movements of the hands, legs and head that met
the following criteria (for nonhuman primate gestures, Call
and Tomasello 2020 and for human infants, Bates, Camaioni,
and Volterra 1975): (i) the movements are targeted toward
specic receivers, (ii) the movements are voluntarily produced
by the infant, that is they are not based on imitative process
(see exclusion criteria) and are voluntarily controlled
depending on the attention state of the receiver (iii) the in-
fant waits for the receiver's response after producing the
movement, (iv) in the absence of a response from the
receiver, the infant persistently produces the same or different
movements until a response is elicited, (v) on elicitation of an
ASO from the receiver, the infant stops producing any
movement targeted to the receiver (Hobaiter and Byrne 2014).
We coded infants' actions as potential gestures only if they
met all of these criteria. These ve criteria not only establish
that the infants' gestures reect a clear intent to communicate
but also that the function of these gestures (see Step 4) is
derived from their interactive communication with a
caregiver.
Step 2. Dene Gesture Forms
The gestures identied in Step 1were then visually described
in terms of the body part used to execute them and their
movement patterns. To ensure coding consistency, we checked
whether the fundamental aspects of gesture forms were present,
even if there were variations in execution among individuals
and contexts. For instance, we coded all raised arm gestures
above the shoulders as “arm/raise.” A complete list of the
gesture forms, along with the criteria used for their categori-
zation is given in Table 1.
Step 3. Dene Gesture Types
Once the gesture forms were dened, we classied them into
the following four types corresponding to the types of adults' co‐
speech gestures (McNeill 1992):
FIGURE 1
|Schematic representation of the various steps to code forms, types, and meaning functions of gestures dened in 12‐ to 15‐months old
infants.
4 of 18 Infancy, 2025
TABLE 1 |Name and descriptions of gesture forms in the repertoire of 12‐ to 15‐months‐olds with corresponding gesture type classication.
Gesture form Description Gesture type
1 Arm/hand ing Moving arm/s or hand/s rapidly and repetitively in the direction of the
receiver
NR
2 Arm jerk Rapid, horizontal movement of one or both arm/s or hand/s, keeping it bent at
the elbow
I, NR
3 Arm jerk with object Like “Arm jerk”, while holding an object D, NR
4 Arm raise Raising arm/s above shoulder and holding it vertically in the air D, I, NR
5 Arm swing Swinging motion of one or both arm/s, sometimes bent at the elbow or falling
straight by either side of the body
NR
6 Arm throw Rapid vertical movements of one or both arm/s, raised above the shoulders I, NR
7 Arm wave Large repeated back and forth motion of the arm raised above the shoulder D, NR
8 Arm wave with object Like Arm wave, while holding an object D
9 Beckon Rapid closing and opening of the st/s while holding out arm/s away from the
body
D
10 Bite Holding receiver's body part between the teeth or lips of the signaler NR
11 Bounce Rapid vertical movements of the upper body, while sitting or standing D, I, NR
12 Bow Bringing head toward the ground, while bending at the waist I
13 “Chin‐chin” Bringing drinking glass in contact with that of the receiver and making a
raising movement as the glasses touch
CO
14 Clap Bringing repeatedly both palms together in audible contact NR
15 Eating Pretending to mix imaginary food in a bowl (real or imaginary), pretending to
scoop out imaginary food with a spoon (real or imaginary), pretending to
bring the spoon (real or imaginary) to the signaler's mouth, pretending to eat
I
16 Embrace Wrapping one or both arm/s around the receiver and maintaining physical
contact
NR
17 Facepalm Bringing two hands over the face and keeping the face covered with palms;
not repeated, like in the context of playing hide and seek or peekaboo
NR
18 Finger‐icking Rapid movements of the ngers (of one or both hand/s) while touching the
signaler's mouth
NR
19 “Flying kiss” Covering the signaler's mount with a hand followed extending the palm
toward a receiver, pretending to send them a kiss
CO
20 “Good‐bye” Hand wave to mean goodbye; arm usually bent at the shoulder, or the arm
held out straight ahead away from the body
CO
21 Grab and hold Firmly closing one or both hand/s over part of the receiver's body and
maintain the hold for >1 s
NR
22 Hand in mouth Bringing one or both hands to the signaler's own mouth and putting multiple
or one nger inside the mouth, sometimes giving it a quick bite
I, NR
23 Hand jerk Rapid vertical or horizontal movements of one or both arms and/or hand,
keeping it bent at the elbow
NR
24 Hand to mouth Bringing one or both hand/s toward or inside the receiver's mouth I, NR
25 Head bob Rapid vertical movements of the head; either repeated movements, or
produced just once to say “yes”
D, I, NR, CO
26 Head jerk Sudden short movement/s of the head, sideways or slightly backwards NR
27 Head shake Shaking head from side to side; either repeated movements, or produced just
once to say “no”
NR, CO
28 Head tilt Tilting the head to one side and keeping it tilted, while looking at the receiver NR
29 “Hello” Opening and closing of the rst to mean “hello”, while bending the arm at the
elbow
CO
30 Hold head Lifting both arms above the head, placing hands on the head and holding it D, I, NR
(Continues)
5 of 18
TABLE 1 |(Continued)
Gesture form Description Gesture type
31 Kiss like mouth movements Making kiss‐like action toward the receiver, without actually kissing NR
32 Leg ing Rapid movements of one or both leg/s away from the body, bending at the
knees
NR
33 Mouth to body touch Bringing mouth, with lips slightly apart, in contact to receiver's body and
maintaining the touch for >1 s
NR
34 Music conducting Holding out two arms away from the body and waving them simultaneously
in a horizontal motion, bending them slightly at the elbows
I
35 Offer hand Reaching out one or both hand/s toward the receiver with index nger
pointed out or open palm turned upwards or palm turned downwards
D, NR
36 Offer object As “Offer hand”, while holding an object, and a physical transfer of the object
was effectuated at the end
D
37 Open mouth Opening the mouth slightly, making rounded shape with the lips I, NR
38 Open palm turned upwards Extending out one or both arm/s toward the receiver, turning the palm/s
upwards toward the receiver
D
39 Pat other Bringing the palm of one or both hand/s in contact with the receiver's body,
followed by repeated rapid movements, gently touching the receiver's body
each time
NR
40 Pat self Bringing one or both hands in contact with signaler's own body followed by
repeated rapid movements to gently touch own body
D, NR
41 Peekaboo Alternatingly covering and uncovering the face with one or both hand/s,
while looking at the receiver during the face‐uncovered phase
NR
42 Point Extending whole arm of the signaler with open palm or with closed palm or
index nger pointing out toward an object/another individual/self
D
43 Point and touch As “Point”, but elaborating it with touching the object/another individual/self
with the extended arm or pointed nger
D
44 Puckered lips Narrowing the lips, bringing them together, usually associated with crinkling
of the nose
I
45 Push Hand/s, placed over object/receiver, and a force exerted to move the object/
receiver away from the signaler, yet not physically displacing the object/
receiver
NR
46 Raspberry bubble Making motorboat sound uttering the lips together, making a cluster of spit‐
bubbles
D, I, NR
47 Reach Extending one or both arm/s, upper body leaning forward, toward the receiver D
48 Reach for object Extending one or both arm/s, open palms turned downwards, upper body
leaning forward, toward an object
D
49 Shake object Holding an object in one or both hand/s, shaking the object, either with arms
held up above the head, or extended out in front, or arms held close to the
body of the signaler
D, NR
50 Show object As “Offer object” but not letting the receiver take the object D
51 Slap object One or repeated forceful contact of one or both hand/s with the surface of an
object
D, NR
52 Slap surface One or repeated forceful contact of one or both hand/s with the surface of a
furniture or the ground
D, NR
53 Stick out tongue Sticking out tongue from the mouth and holding it out for >1 s NR
54 Stomp feet Rapid, forceful vertical movements of one or both leg/s, lifting up and
bringing in contact to the ground or the surface on which the signaler is
standing
NR
55 Sway backwards and forwards Movement of the torso backwards and forwards, while standing or sitting NR
56 Sway side to side Movement of the torso from side to side while standing or sitting I, NR
(Continues)
6 of 18 Infancy, 2025
�Deictic: when the gesture directly indicated a referent
(object, person) present in the immediate environment
�Iconic: when the gesture form describes the attributes of an
object and/or actions
�Non‐referential: when the gesture form has no direct
referent in the environment
�Conventional: when a gesture form bears resemblance to a
socially conventionalized gesture with precise meanings
and can be replaced by a single word (e.g., Good‐bye or
Hello)
Step 4. Dene Gesture Functions
We assumed that all gesture forms could simultaneously
have semantic and pragmatic functions. To determine the se-
mantic function of a gesture, we asked “what did the signaler
achieve by using a gesture form?” For example, if an infant
persistently points to an object until the caregiver retrieves it,
the infant achieved the goal of obtaining a desired object; the
semantic function of this pointing here is coded as “acquire
object.” For deducing the pragmatic function of a gesture, we
asked “what did the gesture achieve in the context of commu-
nication?” For example, when the infant persistently points to
an object until the caregiver retrieves it for him/her, the gesture
fullls the pragmatic function of placing a “request” to the
caregiver. We achieved this using the “Apparently Satisfactory
Outcome” (ASO), which is based on the response of the re-
ceivers (often the caregivers) that appears to satisfy the signaler
(the infants). Satisfaction of the signaler was indicated by
termination of further gesturing upon elicitation of a response
(Hobaiter and Byrne 2014).
2.2.1
|
Exclusion Criteria
We did not analyze gesture forms, types and functions when
the receiver of the gesture and the subsequent response were
not clearly visible in the recordings. This excluded 784 insta-
nces out of the 1734 identied gestures across all of the corpora
(45%). Although this exclusion criterion may result in an
underestimation of the full range of preverbal gestures, it ensures
that the gestures we analyze are intentionally communicative and
that their functions can be accurately identied.
Additionally, during interactions with adults, if infants were
observed to repeat gestures made by adults (i.e., imitation), we
did not code these as gestures as they lacked communicative
intent. However, in cases where gestures initially learned
through imitation were later used by infants with communica-
tive intent—specically to initiate interactions or solicit actions
—we coded these as gestures.
For example: Antoine, at 15 months, observed his aunt using a
was observed to learn the “Music conducting” gesture from his
aunt while listening to music. His aunt later held his hands and
trained him on the arm movements as a music conductor would
do. In this case, we did not code “Music conducting” movement
as a gesture. In another situation from the same recording,
Antoine looked at his aunt and voluntarily made the same arm
movements as the music replayed, and his aunt followed his
suit. In this second case, we coded “Music conducting” as an
iconic gesture produced by Antoine to “solicit action” of music
conducting from his aunt.
2.2.2
|
Reliability
Coder1 (SG) coded all the videos for gesture forms, types and
functions. Inter‐rater reliability was assessed for 20% (5 h) of
all the videos, which corresponded to 106 identied gesture
tokens coded by Coder2 (EP). Agreement between Coder1
and Coder2 was 93% for gesture forms, 91% for gesture types,
and 82% for gesture functions. Cohen's Kappa statistics for
these coding decisions were 0.91, 0.88, and 0.78, respectively,
rendering our agreement score as “Substantial” to “Perfect”
(Landis and Koch 1977). SG has extensive video coding
experience for gestures and other social behaviors in ma-
caques, apes, and human adults. EP is a naïve coder with
beginners' level experience with ELAN software and a theo-
retical knowledge about the goals and principles of the pre-
sent study and was provided with the coding scheme in this
TABLE 1 |(Continued)
Gesture form Description Gesture type
57 Sweep surface Rapid horizontal movements of one or both hand/s along the surface of a
furniture or the ground
NR
58 Telephone Making a hand‐shape, as if holding a telephone, bringing it to an ear and
pretending to speak to someone; same gesture can be produced using an
object, as if holding a “telephone”, while it is not an actual telephone
I
59 Turn head Rapid movement to turn the head away from the receiver; head could be
turned on either right or left side, but not both
NR
60 Wide open mouth Widely opening the mouth making a rounded shape with the lips, so much so
that the cheeks are extended behind
D, I, NR
61 “Wow” Opening the mouth widely, in a rounded shape and placing both hands one
on each cheek
CO
62 Wrist rotate Rotating both wrists repeatedly in circular motion, usually holding arms close
to the body
D
Abbreviations: CO =conventional, D =deictic, I =iconic, NR =non‐referential.
7 of 18
study as a reference for reliability assessment. The coding
instructions used for the reliability coding have been provided
in Supporting Information S3.
2.2.3
|
Analysis
All statistical analyses were performed using the software R
(version 4.3.1, 2023‐06‐16). Figures 2and 3were generated us-
ing David Sjoberg's “ggsankey” package (Sjoberg 2021; R Core
Team 2021).
3
|
Results
Our analyses were conducted on the 950 gesture tokens (mean
number of gestures produced per infant =93.6, SEM =32.2)
which complied with the denition criteria of communicative
intent and allowed us to code their semantic and pragmatic
functions.
3.1
|
Gesture Forms
Gesture forms are described based on their physical forms and
movement patterns. We identied and described 62 unique
gesture forms in the repertoire of 12‐ to 15‐month‐old infants
(Table 1, Column 1, 2). Each infant produced an average of
13.12 gesture forms (SEM =1.2; range: 8–28). We conducted a
one‐way ANOVA to compare the mean number of gesture forms
across ages, and since there was no signicant main effect of age
(F(3, 20) =0.29, p=0.81), we removed this factor as a predictor
in all subsequent descriptions and analyses. The number of
gesture forms for each individual, from 12 to 15 months of age,
is provided in Table S1.
Eight gesture forms were used by all the six infants, nine forms
were shared by ve infants, and seven forms by four infants.
Thus, 24 (39%) gesture forms were shared among half of the
study group. These shared gestures taken together were pro-
duced most frequently among all instances of gesturing
observed across individuals of all ages (83% of all the gesture
tokens; Table S2, Column 3). Pointing alone made up 23% of all
the gesture tokens performed, which is not surprising given that
pointing gestures are the earliest to emerge in infants, serving a
wide range of functions (Bates 1974).
Eighteen gesture forms solely occurred in the repertoire of a
single infant (4.08% of all the gesture tokens; Table S2, Col-
umn 3), and 14 gestures were found in the repertoire of two
infants (8.38% of all the gesture tokens; Table S2, Column 3).
These gestures may not be idiosyncratic, as several of them
were found in previous reports of infant gestures from
different corpora (Table S2, Column 4 and 5). Note that we
found only 36% overlap between our gesture forms repertoire
with the one described by Kersken et al. (2019) for 1‐ to 2‐
year‐olds; and only 25% overlap with the one described
by Rodrigues, Maroco, and Frota. (2021) for 7‐ to 12‐months‐
olds. This is a rather low overlap, given that 90% of
gesture forms overlap between apes and human infants
(Kersken et al. 2019), and meanings of some of these
forms are also shared between apes and adult humans
(Henderson et al. 2024). We come back to this in the general
discussion.
FIGURE 2
|Classication of the 24 gesture forms shared among half of the study group (83% of all the gesture tokens) into the four possible
gesture types. The number of ows connecting a gesture form node on the left to the gesture type node on the right corresponds to the number
of types each form was assigned to. Gesture forms “Point” and “Wide open mouth” are highlighted as examples that have been illustrated in the text.
8 of 18 Infancy, 2025
3.2
|
Gesture Types
On average a gesture form could be classied into more than
one (1.44) gesture type (SEM =0.1) (range: 1–4) (Table 1, Col-
umn 3). 70.1% of the gesture forms (44 out of 62) were classied
as non‐referential, 38.71% (24/62) as deictic, 27.42% (17/62) as
iconic and 12.9% (8/62) as conventional. Figure 2shows a
schematic representation of the classication for the subset of 24
gestures forms that were shared among half of the study group,
into the four possible gesture types (83% of all the gesture to-
kens; Table S2, Column 3). For example, the gesture “Point” is
always classied as deictic, since it is used for referring to an
object or person in the immediate environment (see Figure 2,
highlighted in red). The gesture “Wide open mouth” can be
categorized as iconic, when it is used to describe the action of
eating and categorized as non‐referential when it is used to
express surprise during social interaction (see Figure 2, high-
lighted in red). There were only a few gesture forms that were
classied as conventional gestures, and most of them were
produced by a single individual, except for Good‐bye, which was
found in four out of the six infants (Table S2, Column 1 and 2).
We modeled the percentage of gesture type (Perc) using a mixed
model specied as: Perc ~ GestureType þAge þ(1 ∣infant). A
near‐zero variance of the random intercept (σ2=2.497e−19,
SD =4.997e−10) suggested negligible variability across infants.
There was no main effect of Age (β=1.293e−10, t=0, p=1)
and a main effect of gesture type (χ
2
(3) =111.73, p<0.001):
there were more non‐referential gestures (46.7%; SEM =3.24)
than deictic (37.7%; SEM =3.16) (β= −9.01, SEM =3.54,
t= −2.54, p=0.01), iconic (9.19%; SEM =1.24) (β= −37.56,
SEM =3.54, t= −10.58, p<0.001) and conventional gestures
(6.33%; SEM =1.54) (β= −40.42, SEM =3.548, t= −11.392,
p<0.001) (Figure 4). There were more deictic gestures than
iconic (β= −28.54, SEM =3.54, t= −8.13, p<0.001) and
conventional gestures (β= −31.40, SEM =3.50, t= −8.95,
p<0.001). A detailed account of the age‐wise variation of
proportions of gesture types in infants from 12 to 15 months can
be found in Table S3.
3.3
|
Functions of Gestures
We identied the following 11 semantic functions of the ges-
tures produced by infants from 12‐ to 15‐months: “acquire ob-
ject/food”, “describe action”, “express emotion”, “play”, “reject
object”, “response to other”, “seek attention”, “seek informa-
tion”, “share attention”, “share object”, “solicit action/play”.
These gestures fulll 5 unique pragmatic functions as derived
from their effects on the receivers during the interaction:
“request”, “declare”, “initiate interaction”, “refuse interaction”,
and “turn‐taking” (Table 2).
A gesture form can be assigned to more than one semantic or
pragmatic function depending on its contextual use. Figure 3
below shows a schematic representation of the classication for
the subset of 24 gesture forms that are shared among half of the
study group into their corresponding semantic and pragmatic
functions (83% of all the gesture tokens; Table S2, Column 3).
For instance, “Point” (highlighted in red, Figure 3) is a gesture
form that can have the semantic meaning “acquire object”,
pragmatically functioning as a “request” to the receiver; “Point”
could also be used to “share attention” on an object of interest,
FIGURE 3
|Classication of the 24 gesture forms shared among half of the study group (83% of all the gesture tokens) into the eleven semantic
functions identied, and their corresponding ve pragmatic functions. The number of ows connecting a gesture form node on the left to the
semantic and pragmatic function nodes in the middle and on the right correspond to the number of functions each form was assigned to.
Gesture forms “Point” is highlighted as an example that has been illustrated in the text.
9 of 18
thus, pragmatically function to “declare”. We observed an
interesting elaboration of pointing‐ “Point and touch” gesture‐
(2% of all the gesture tokens, shared among two infants, hence
not visible in Figure 3; see Table S2, Column 3) which solely
have the semantic function of seeking information about novel
objects of interest.
3.3.1
|
Semantic Functions
Overall, we found that deictic gestures were used for 9 out of the
11 semantic functions, and iconic gestures were used for 5 out of
11. The predominant ve semantic functions of referential
gestures (deictic, iconic) were “acquire food/object” (23.61%,
SEM =5.58), “play” (11.18%, SEM =6.2), “share attention”
(25.83%, SEM =1.92), “share object” (13.26%, SEM =0.75) and
“solicit action/play” (25.46%, SEM =7.16). Additionally, in our
study, infants produced all iconic gestures without uttering any
words. All iconic gestures were used for the semantic function
“describe actions” (e.g., “Raspberry bubbling” to request to blow
on hot food, Supporting Information S2, Video S2), except one
gesture (“Sway side to side”) which was used to describe
movement attributes of a toy.
Non‐referential gestures (e.g., “Head shake” to express disin-
terest in food, Supporting Information S2, Video S2), although
used for 10 out of the 11 identied semantic functions, were
mostly used to “express emotion” (87.32%, SEM =1.7)
(Figure 5a).
The semantic functions of conventional gestures (e.g., “Good‐
Bye” and “Flying kiss” to end an interaction, Supporting
Information S2, Video S3) were more restricted: they were solely
used to fulll two functions: “response to others” (53.16%,
SEM =7.79) and “share attention” (40.42%, SEM =10.94)
(Figure 5a). Since these conventional gestures already have
precise semantic meanings, we excluded these from our ana-
lyses of the diversity index of semantic functions.
We evaluated the diversity of semantic functions by quantifying
their richness (i.e., how many different semantic functions) and
evenness (i.e., how many gesture forms belong to each semantic
functional category) for referential (deictic, iconic) and non‐
referential gestures, using the Shannon‐Weiner diversity index
(Shannon 1949), and compared them with the Hutcheson t‐test
(Hutcheson 1970). The Shannon‐Weiner diversity index is a
measure used in ecological studies to calculate species diversity
in a community as a function of the proportion of each species
relative to the total number of species present in the commu-
nity. Here, we draw analogies by replacing an ecological
“community” by infant “gesture type” (referential & non‐
referential), while replacing the “total number of species in a
community” by “semantic functions” of infant gestures. The
diversity index for semantic functions of infant referential ges-
tures (deictic and iconic) (Shannon diversity index =1.84) is
signicantly higher (t(311) =12.1, p<0.001) than that of non‐
referential gestures (Shannon diversity index =0.61). This result
conrms that infants' referential gestures cover a wider range of
semantic functions than do non‐referential gestures.
3.3.2
|
Pragmatic Functions
As can be seen in Figure 5b, deictic and iconic gestures pre-
dominantly achieve a “request” (mean_deictic =51.46%,
SEM =2.72; mean_iconic =57.87%, SEM =5.47) and a decla-
ration (“declare”) (mean_deictic =27.6%, SEM =2.43; mean_-
iconic =27.17%, SEM =4.6). These gestures also facilitate “turn‐
taking” (mean_deictic =9.38%, SEM =1.9; mean_-
iconic =29.86%, SEM =7.64). It appears that the “describe
FIGURE 4
|Mean percentage of non‐referential, deictic, iconic, and conventional gestures in the infant gesture repertoire (n=6). The error bars
represent standard errors of the mean. Each shape corresponds to one infant.
10 of 18 Infancy, 2025
action” semantic function of the iconic gestures, also facilitate
turn‐taking, as if in conversation‐like structures.
Non‐referential gestures were predominantly used for the
pragmatic function “declare” (83.82%, SEM =0.9; Figure 5b).
The percentage of declarative functions of non‐referential ges-
tures was signicantly higher than that of the other pragmatic
functions combined (independent sample t‐test: t(31) =6.38,
p<0.001). A one‐way ANOVA revealed that non‐referential
gestures were signicantly more likely to have a declarative
function than referential gestures (F(3,8) =118.9, p<0.001).
Post‐hoc Tukey tests showed that the percentage of declarative
function for non‐referential gestures (83.81%, SEM =0.9) was
higher than the percentage of declarative function for deictic
(7.53%, SEM =2.43; p<0.001), iconic (27.17%, SEM =4.6;
p<0.001) and conventional gestures (12.5%; p<0.001). This
result conrms our expectations of infant non‐referential ges-
tures to have more declarative functions than the referential
gestures, thus building upon their role in narrative perfor-
mances and discourse structuring in children (Vila‐Gimenez
and Prieto 2021; Levy and McNeill 2013; Nicoladis, Mayberry,
and Genesee. 1999).
TABLE 2 |Descriptions of semantic and pragmatic functions of gesture forms found in the repertoire of 12‐ to 15‐months old infants.
Semantic function (“what did the signaler achieve by
using a gesture?”) Description
Acquire object/food Intention of the signaler is to obtain an object/food of desire and
is measured by ASO of receiving the desired object/food
Describe action Intention of the signaler is to represent an action or attributes of
an object using gestures
Express emotion Intention of the signaler is to express states of happiness,
sadness, anger or disinterest, etc. toward an ongoing situation
Play Intention of the signaler is to begin or continue a game
Reject object Intention of the signaler is to directly reject any object offered to
them; different from expression of disinterest, as this function
directly refers to a physical object
Response to other Intention of the signaler is to gesturally respond to the receiver's
question or request
Seek attention Intention of the signaler is to attract the receiver's attention
either to begin an interaction or continue an ongoing one
Seek information Intention of the signaler is to inquire about something in the
environment
Share attention Intention of the signaler is to attract receiver's attention on an
external object
Share object Intention of the signaler is to share an object with the receiver
Solicit action/play Intention of the signaler is to elicit a particular action in the
receiver
Pragmatic function (“what did the gesture achieve in the
context of communication ?”) Description
Request The gesture was performed to make a request to the receiver;
usually connected to the semantic functions “acquire object/
food”, “describe action”, “seek information”, “solicit, action/
play”
Declare The gesture was performed to declare an emotion state or a fact;
usually connected to semantic functions “express emotion”,
“seek attention”, “share attention”, “share object”
Initiate interaction The gesture was performed to start a new interaction or revive
an interaction that had been paused; usually connected to
semantic functions “play”, “seek attention”, “share object”
Refuse interaction The gesture was performed to terminate an interaction; usually
connected to semantic functions “express emotion”, “reject
object”
Turn‐taking The gesture was performed to facilitate taking turns alternating
with the receiver, during an ongoing interaction; usually
connected to all the semantic functions described above
11 of 18
Conventional gestures, on the other hand, were almost exclu-
sively used to “turn‐taking” in social interactions (93.75%,
SEM =6.25; Figure 5b), illustrating pragmatic roles of these
preverbal gestures in coordinating social exchanges, like “Good‐
bye” followed by “Flying kiss” with the receiver to say good‐bye
(Supporting Information S2, Video S3).
A detailed account of the age‐wise variation in semantic and
pragmatic functions of infant gestures can be found in Tables S4
and S5. Example videos of gesture forms and their subsequent
classication into gesture types and functions are provided in
Supporting Information S2, Videos S1,S2, and S3.
4
|
Discussion
The objective of this paper was to examine precursors of adult
co‐speech gestures in infants (12‐ to 15‐months) prior to
mastering spoken language. Previous research has focused pri-
marily on pointing gestures and their signicance in language
acquisition. Here, we used a descriptive observational method,
predominantly applied to primates, to identify the form, type
and function of gestures used by preverbal infants. Our results
show that infants employed a repertoire of 62 gesture forms,
which can be classied into the four gesture types that char-
acterize adult co‐speech gestures (deictic, iconic, non‐referential
and conventional) (Kendon 2004; McNeill 1992; Ekman and
Friesen 1969). In addition, we nd that, in infants, referential
gestures (deictic and iconic) cover a wider range of semantic
functions than non‐referential gestures, while non‐referential
gestures have more declarative functions than referential ges-
tures. This is indicative of an early emergence of linguistic
features similar to co‐speech gestures, but in the absence of
speech.
4.1
|
Gesture Forms and Gesture Types
We nd a strikingly low overlap (25% and 36%) between the
repertoire of gestures forms identied in the present study and
those found in previous studies (Rodrigues, Maroco, and
Frota 2021; Kersken et al. 2019). The overlap of repertoires
described by Kerksen and colleagues with those by Rodrigues
and colleagues is however very high (81%). Recently, a consid-
erable overlap of semantic functions between chimpanzee and
human gestures has been reported (Henderson et al. 2024),
suggesting an evolutionary continuity of form‐meaning re-
lations. The discrepancy between our report and other infant
gesture inventories may be due to the fact that these studies
used a pre‐dened repertoire of ape gestures (Hobaiter and
Byrne 2011) as a reference for conducting a phylogenetic com-
parison of gesture forms. Regardless, it is important to note that
all of these studies, including our own, relied on small sample
sizes (n=6 for 25 h of recordings in the present study; n=13 for
15 h of recordings in Kersken et al. 2019;n=10 for 50 h of
recordings in Rodrigues, Maroco, and Frota 2021). We also
acknowledge the fact that the 62 gestures that we identied
here, since we analyzed only the gestures for which a clear
response from the receivers was visible. This only comprised
about half of the total gesture tokens coded in the dataset.
Therefore, it is important to consider that no single study can
denitively dene the exhaustive repertoire of gestures found in
infancy, however, a collective effort can give rise to a clearer
picture of the infant gestural repertoire, in terms of the diversity
of gesture forms, communicative functions and variations as a
response to socio‐cultural environments.
In contrast to previous studies focusing on gesture forms, we
further classied them into gesture types and found that infant
gestures, without speech, also classify as deictic, iconic, non‐
FIGURE 5
|Percentage of (a) semantic functions of (b) pragmatic functions of deictic, iconic, non‐referential and conventional gesture types in 12‐
to 15‐month‐old infants.
12 of 18 Infancy, 2025
referential and conventional gestures, similar to adults' co‐
speech gestures.
Deictic gestures in infants included pointing, reaching,
showing object, offering object and reaching for object. These
gestures have been extensively documented in earlier work
and can emerge, as early as in 9 months old (Rohlng,
Grimminger, and Lüke 2017; Liszkowski et al. 2004,2007;
Bates, Benigni, and Bretherton 1979; Bates 1974), Additionally,
we observed that infants sometimes elaborate their pointing
gesture followed by “touch”‐ this “point and touch” gesture
serves the semantic function of seeking information about
novel objects. Pointing is already known to be used for
learning new labels, paving the way for asking questions later
in life (Lucca and Wilbourn 2019; Kovács et al. 2014; Begus
and Southgate 2012; Tomasello, Carpenter, and Liszkow-
ski 2007; Liszkowski et al. 2006). The “point and touch”
gesture further indicates the infants' intention to gather in-
formation about specic objects of interest. The caregiver's
responses (providing the labels for the objects) is not a random
outcome of the interaction, but rather a response to the in-
fants' intention. Future studies should determine whether this
specialized form of deictic gesturing (“point and touch”),
rather than pointing in general (Lucca and Wilbourn 2018,
2019; Begus, Gliga, and Southgate 2014), serves as a precursor
to children's ability to ask questions (Chouinard, Harris, and
Maratsos 2007).
We observed iconic gestures, without accompanying speech in
12‐ to 15‐months‐olds conrming ndings from previous studies
(L. P. Acredolo and Goodwyn 1985; L. Acredolo and Good-
wyn 1988; Green et al. 2023). However, other corpus studies
report the use of iconic gestures much later around 22–26‐
months (Green et al. 2023; Behne, Carpenter, and Toma-
sello 2014; Özçalışkan and Goldin‐Meadow 2011). In adults, the
functions of iconic co‐speech gestures are tightly linked to
speech (McNeill 1992)‐ their “meaning” is dened by the
spoken label. Surely, this top‐down approach makes it difcult
to identify iconicity in preverbal gestures in the absence of
words. As a result, the identication of iconic gestures often
only occurs after the production of the rst words. In contrast,
our bottom‐up, speech‐independent method enabled us to
identify iconic gestures by matching them with preceding and/
or following actions, referential contexts and reactions from the
receivers. This method could reveal the existence of iconic
gestures before procient word‐use.
The majority of the gestural repertoire of 12‐ to 15‐month‐old
infants consists of non‐referential gestures. These gestures are
produced intentionally and directed to a target receiver, simi-
larly to other gesture types. However, unlike deictic, iconic or
conventional gestures, they do not have an external referent.
Interestingly, 22 of the 42 non‐referential gestures are rhythmic
in nature (e.g., Arm jerk, Arm throw, Bounce, Leg ing, Swaying
side to side, among others), perhaps indicating a continuum
with the rhythmic arm movements synchronized with babbling
reported from younger infants (Iverson and Fagan 2004; Iverson
and Thelen 1999; Thelen 1979). This could also be the potential
“proto‐versions of beat gestures” (Pouw and Fuchs 2022) in
adults, which are included in a larger category of non‐referential
co‐speech gestures (Prieto et al. 2018).
Conventional gestures represented the smallest proportion
of gesture types (7 out of 62; e.g., “Chin‐Chin”, “Flying kiss”,
“Good‐bye”, “Head shake”, “Head bob”, “Hello” and “Wow”)
and were limited to certain individuals (Table S2). Individual
variation could be attributed to levels of exposure to these ges-
tures, or even varying degrees of time spent with adults
(L. Acredolo and Goodwyn 1988).
4.2
|
Gesture Functions
We examined the relationship between the different types of
infant gestures and their functions, thus, addressing a previ-
ously neglected area of research. To do so we adopted the
Apparently Satisfactory Outcome (ASO) method from primate
studies (Hobaiter and Byrne 2014) to indirectly derive gesture
functions as a consequence of particular contexts, relying only
on the receiver's responses. Our results indicate that infants use
deictic and iconic gestures for a large array of semantic func-
tions. These gestures enable them to acquire their desired ob-
ject, share attention with others on objects of desire, and
describe actions to solicit actions from others. Their pragmatic
functions are, therefore, mostly imperative (to request) and
declarative. Co‐speech referential gestures are also known to
add semantic diversity to speech (Kendon 2004), suggesting an
ontogenetic continuity here.
We found that iconic gestures in preverbal infants predomi-
nantly describe actions, rather than attributes of objects. Similar
observations had been made by Özçalışkan and Goldin‐
Meadow (2011) and L. Acredolo and Goodwyn (1988) in older 2‐
year‐old children. This early emergence of iconic action repre-
sentation suggests a capacity to associate forms with meanings,
even in the absence of speech. From an embodied cognition
perspective, it is suggested that infants develop representations
of bodily actions as their motor skills mature; this then leads to
iconic representation of actions through gestures, which may
serve as the rst step toward symbolic communication and
language acquisition (Fuchs, Etzelmüller, and Tewes 2016;
Werner and Kaplan 1963), found ubiquitously across languages
(Perniss, Thompson, and Vigliocco 2010).
Our results show that infant non‐referential gestures largely serve
declarative purposes. The main semantic function of infant non‐
referential gestures was to express happiness, discomfort, dislike,
or anger toward others or about an event. Gesturing stopped when
an appropriate reaction was elicited from the audience, either by
continuing the interaction (in case of positive emotion), or
termination of the interactions (in case of negative emotions).
These declarative gestures in infants are thus interactional, as
reported for adult non‐referential co‐speech gestures (beat ges-
tures) (Bavelas et al. 1992; Kendon 2004).
Conventional gestures such as “yes”, “no”, hand wave for “goo-
dbye” have already been reported in 12 to 34‐month‐olds (Fusaro
and Vallotton 2011; Özçalışkan and Goldin‐Meadow 2011). We
13 of 18
found that the main pragmatic function of these gestures was
turn‐taking, suggesting that conventional gestures may enable
conversation‐like structures between the infant and their care-
givers before procient spoken language develops.
4.3
|
Gesture Flexibility
Infant gesture forms are not inexibly tied to their corresponding
gesture types, nor their functions; they are rather used depending
on the context, exhibiting a form‐to‐function dissociation. In the
infant repertoire described here, we found that a single gesture,
such as pointing at an object, can carry two distinct semantic
functions—“give me the object” or “look at the object”. These
serve two different pragmatic functions: the rst is a “request”,
and the second is a “declaration.” We also observe that multiple
distinct gestures can have a similar function. For example, to
express emotions we describe a wide range of non‐referential
gestures in these infants (e.g., gestures Hand ing,Leg ing,
Head shake were all used to express the emotion of not being
content with the situation). This genre of form‐to‐function
dissociation is a key cognitive marker of involuntary, inten-
tional communication, also described for gesture production in
non‐human primates (Call and Tomasello 2020).
It is worth noting that co‐speech gestures in adults follow a
more rigid classication: referential gestures are primarily se-
mantic and non‐referential gestures are predominantly prag-
matic in nature. However, this strict distinction between
referential and non‐referential gesture functions may result
from the top‐down inuence of speech. It might be valuable to
consider a bottom‐up, ethological approach to classify adult
gestures, similar to the method we used in this study. This could
lead to a more nuanced understanding of co‐speech gestures
and broaden their expanding their meaning‐scape, as has been
recently proposed (Rohrer, Delais‐Roussarie, and Prieto 2020;
Prieto et al. 2018). Our open‐ended approach, which assumes
that all co‐speech gestures can have both semantic and prag-
matic functions depending on the context, could lead to a
multidimensional, multimodal classication of speech‐gesture
integration in humans, as envisioned by Rohrer et al. (2022).
4.4
|
Challenges, Limitations and Future
Directions
One important challenge of the current ASO method is that
gesture functions are deduced indirectly from the infant's
behavior in response to the receiver's reaction to their gesture.
In essence this indirect, contextual interpretation is similar to
approaches used in past studies that inferred function by
looking at the context in which gestures occur (Patel‐
Grosz 2023; Tomasello and Call 2019; Tomasello et al. 1994,
1997; Goldin‐Meadow, Alibali, and Church 1993; Goldin‐
Meadow 1999), but with a key distinction. Inferring a gesture's
function cannot be done solely by observing what happens
before and after the gesture. Rather, our method identies the
infant's intended communicative goal more precisely by deter-
mining which specic response from the recipient satises the
infant, as indicated by the cessation of gesturing. To illustrate,
consider an example: when an infant points to a doll and the
adult brings the doll to her, satisfying the infant, we infer that
the infant's communicative goal was to request the doll. In
another scenario, if the adult responds by saying, “Yes, that's
your doll”, and this satises the infant, we infer that her goal
was to declaratively point at the doll, sharing her interest and
attention with the adult. Without these responses and the in-
fant's subsequent reactions, it would be impossible to determine
the semantic functions (requesting the object vs. sharing
attention) or the pragmatic functions (a request vs. a declara-
tion) of the original pointing gesture. Thus, the gesture alone,
even in a familiar context, can have an ambiguous meaning
without considering the interaction that follows. The ASO
method, by focusing on the infant's response to the recipient's
reaction, provides a deeper understanding of the infant's
communicative intent, going beyond merely analyzing the
surrounding context.
Perhaps then, one of the most important methodological limi-
tations of the ASO approach is an almost exclusive focus on
gestures that require the receiver to do something in response.
This criterion led us to discard 45% of all gesture instances.
Thus, while the current approach reveals a complex and rich
repertoire of gestures that could not be reported with previous
methods, the inventory of gestures reported here is likely to be
an underestimation of the actual gesture repertoire of infants.
The corpora of infant‐caregiver interactions that we used here
were originally created with the aim to examine factors driving
spoken language development (MacWhinney 2000; Demuth and
Tremblay 2008; but also see Morgenstern and Parisse 2012).
Thus, often in these videos, we could not see the child, perhaps
leading to an oversight of the gestures produced in these mo-
ments. This, in turn, could have resulted in an underestimation
of some gesture forms and their functions in the infant reper-
toire. One possibility is that the individual variation that we
observe in our data is a result of such a sampling bias rather
than being by actual variations in gesture production at an in-
dividual level. Due to this unresolvable distinction between a
potential developmental process and sampling bias, we
refrained from performing any longitudinal analyses on the
current data. Future studies on infant gesture development
should ideally be conducted on video corpora recorded with
gesture production as the main goal (e.g., Iverson and Goldin‐
Meadow 2005; Özçalışkan and Goldin‐Meadow 2011). Our pri-
mary goal was to study the early precursors of co‐speech ges-
tures, but in this study, we focus on non‐speech gestures alone.
To bridge this gap, future research should explore how these
non‐speech gestures relate to infant vocalizations, eventually
leading to the integration of gestures and speech. While we have
not examined this integration in the current paper, we plan to
investigate it in future studies by documenting the longitudinal
development of gesture‐vocalization combinations and their
functional roles in early communication.
4.5
|
Conclusions
In sum, our results demonstrate that preverbal infants possess a
wide range of gesture forms that align with the four co‐speech
14 of 18 Infancy, 2025
gesture types typically observed in adults. Our ndings mark an
important rst step in understanding the development of co‐
speech gestures independently of spoken language. This prog-
ress was made possible by using an ethological approach
(Hobaiter and Byrne 2014), which allows for the identication
of gestures and their functions based solely on behavioral sig-
nals. This approach, which does not rely on speech, offers new
insights into the early development of gestural communication
in infancy.
Author Contributions
Shreejata Gupta: conceptualization, data curation, formal analysis,
funding acquisition, project administration, writing–original draft,
writing–review & editing. Eulalie Pequay: methodology. Clément
François: project administration, resources, supervision, validation,
writing–original draft, writing–review & editing. Isabelle Dautriche:
project administration, resources, supervision, validation, writing–
original draft, writing–review & editing.
Acknowledgments
This work, carried out within the Institute of Convergence ILCB (ANR‐
16‐CONV‐0002), has beneted from support from the French govern-
ment (France 2030), managed by the French National Agency for
Research (ANR) and the Excellence Initiative of Aix‐Marseille Univer-
sity (A*MIDEX). This research has been supported by a grant from the
Fondation de France (Grant 00147512/WB‐2023‐51888).
Conicts of Interest
The authors declare no conicts of interest.
Data Availability Statement
The data that support the ndings of this study are available from the
corresponding author upon reasonable request.
Endnotes
1
Conventional gestures are traditionally considered as “co‐speech”
gestures although they can sometimes replace words (“pro‐speech”
gestures; Schlenker 2020; Ekman and Friesen 1969).
References
Acredolo, L., and S. Goodwyn. 1988. “Symbolic Gesturing in Normal
Infants.” Child Development 59, no. 2: 450–466. https://doi.org/10.2307/
1130324.
Acredolo, L. P., and S. W. Goodwyn. 1985. “Symbolic Gesturing in
Language Development: A Case Study.” Human Development 28, no. 1:
40–49. https://doi.org/10.1159/000272934.
Bates, E. 1974. “Language and Context: Studies in the Acquisition of
Pragmatics.” Doctoral diss., The University of Chicago.
Bates, E., L. Benigni, and I. Bretherton. 1979. Cognition and Commu-
nication in Infancy. New York: Academic.
Bates, E., L. Camaioni, and V. Volterra. 1975. “The Acquisition of Per-
formatives Prior to Speech.” Merrill‐Palmer Quarterly of Behavior and
Development 21, no. 3: 205–226.
Bates, E., and F. Dick. 2002. “Language, Gesture, and the Developing
Brain.” Developmental Psychobiology: The Journal of the International
Society for Developmental Psychobiology 40, no. 3: 293–310. https://doi.
org/10.1002/dev.10034.
Bavelas, J. B., N. Chovil, D. A. Lawrie, and A. Wade. 1992. “Interactive
Gestures.” Discourse Processes 15, no. 4: 469–489. https://doi.org/10.
1080/01638539209544823.
Begus, K., T. Gliga, and V. Southgate. 2014. “Infants Learn What They
Want to Learn: Responding to Infant Pointing Leads to Superior
Learning.” PLoS One 9, no. 10: e108817. https://doi.org/10.1371/journal.
pone.0108817.
Begus, K., and V. Southgate. 2012. “Infant Pointing Serves an Inter-
rogative Function.” Developmental Science 15, no. 5: 611–617. https://
doi.org/10.1111/j.1467‐7687.2012.01160.x.
Behne, T., M. Carpenter, and M. Tomasello. 2014. “Young Children
Create Iconic Gestures to Inform Others.” Developmental Psychology 50,
no. 8: 2049–2060. https://doi.org/10.1037/a0037224.
Beuker, K. T., N. N. Rommelse, R. Donders, and J. K. Buitelaar. 2013.
“Development of Early Communication Skills in the First Two Years of
Life.” Infant Behavior and Development 36, no. 1: 71–83. https://doi.org/
10.1016/j.infbeh.2012.11.001.
Bohn, M., L. Zimmermann, J. Call, and M. Tomasello. 2018. “The Social‐
Cognitive Basis of Infants' Reference to Absent Entities.” Cognition 177:
41–48. https://doi.org/10.1016/j.cognition.2018.03.024.
Butcher, C., S. Goldin‐Meadow, and D. McNeill. 2000. “Language and
Gesture.” In Gesture and the Transition From One‐To Two‐Word Speech:
When Hand and Mouth Come Together, 235–257. Cambridge, USA:
Cambridge University Press.
Call, J., and M. Tomasello. 2020. The Gestural Communication of Apes
and Monkeys. New York, USA: Psychology Press.
Carpenter, M., K. Nagell, M. Tomasello, G. Butterworth, and C. Moore.
1998. “Social Cognition, Joint Attention, and Communicative Compe-
tence From 9 to 15 Months of Age.” Monographs of the Society for
Research in Child Development 63, no. 4: i–174. https://doi.org/10.2307/
1166214.
Cartmill, E. A., S. Beilock, and S. Goldin‐Meadow. 2012. “A Word in the
Hand: Action, Gesture and Mental Representation in Humans and Non‐
Human Primates.” Philosophical Transactions of the Royal Society B:
Biological Sciences 367, no. 1585: 129–143. https://doi.org/10.1098/rstb.
2011.0162.
Cartmill, E. A., and R. W. Byrne. 2010. “Semantics of Primate Gestures:
Intentional Meanings of Orangutan Gestures.” Animal Cognition 13, no.
6: 793–804. https://doi.org/10.1007/s10071‐010‐0328‐7.
Cartmill, E. A., and S. Goldin‐Meadow. 2016. “Gesture.” In APA
Handbook of Nonverbal Communication, edited by D. Matsumoto, H. C.
Hwang, and M. G. Frank, 307–333. Washingtom DC, USA: American
Psychological Association.
Caselli, M. C., P. Rinaldi, S. Stefanini, and V. Volterra. 2012. “Early
Action and Gesture ‘Vocabulary’ and its Relation With Word Compre-
hension and Production.” Child Development 83, no. 2: 526–542. https://
doi.org/10.1111/j.1467‐8624.2011.01727.x.
Chouinard, M. M., P. L. Harris, and M. P. Maratsos. 2007. “Children's
Questions: A Mechanism for Cognitive Development.” Monographs of
the Society for Research in Child Development 72: i–129.
Clark, H. H. 1996. Using Language. Cambridge, UK: Cambridge Uni-
versity Press.
Corballis, M. C. 2002. “From Hand to Mouth: The Origins of Language.”
New Jersey, USA: Princeton University Press.
Demuth, K., and A. Tremblay. 2008. “Prosodically‐Conditioned Vari-
ability in Children's Production of French Determiners.” Journal of
Child Language 35, no. 1: 99–127. https://doi.org/10.1017/s03050009
07008276.
Douglas, P. H., and L. R. Moscovice. 2015. “Pointing and Pantomime in
Wild Apes? Female Bonobos Use Referential and Iconic Gestures to
15 of 18
Request Genito‐Genital Rubbing.” Scientic Reports 5, no. 1: 13999.
https://doi.org/10.1038/srep13999.
Ekman, P., and W. V. Friesen. 1969. “The Repertoire of Nonverbal
Behavior: Categories, Origins, Usage, and Coding.” Semiotica 1, no. 1:
49–98. https://doi.org/10.1515/semi.1969.1.1.49.
Fuchs, T., G. Etzelmüller, and C. Tewes. 2016. “The Embodied Devel-
opment of Language.” In Embodiment in Evolution and Culture,
107–128.
Fusaro, M., and C. D. Vallotton. 2011. “Giving a Nod to Social Cogni-
tion: Developmental Constraints on the Emergence.” In Integrating
Gestures: The Interdisciplinary Nature of Gesture, 121–135.
Genty, E., T. Breuer, C. Hobaiter, and R. W. Byrne. 2009. “Gestural
Communication of the Gorilla (Gorilla gorilla): Repertoire, Intention-
ality and Possible Origins.” Animal Cognition 12, no. 3: 527–546. https://
doi.org/10.1007/s10071‐009‐0213‐4.
Goldin‐Meadow, S. 1999. “The Role of Gesture in Communication and
Thinking.” Trends in Cognitive Sciences 3, no. 11: 419–429. https://doi.
org/10.1016/s1364‐6613(99)01397‐2.
Goldin‐Meadow, S., and M. W. Alibali. 2013. “Gesture's Role in
Learning and Development.” In Oxford Library of Psychology. The Ox-
ford Handbook of Developmental Psychology.Body and Mind, P. D.
Zelazo, ed. Vol. 1, 953–973. New York: Oxford University Press. https://
doi.org/10.1093/oxfordhb/9780199958450.013.0033.
Goldin‐Meadow, S., M. W. Alibali, and R. B. Church. 1993. “Transitions in
Concept Acquisition: Using the Hand to Read the Mind.” Psychological
Review 100, no. 2: 279–297. https://doi.org/10.1037//0033‐295x.100.2.279.
Gómez, J.‐C. 2005. “Requesting Gestures in Captive Monkeys and Apes:
Conditioned Responses or Referential Behaviours?” Gesture 5, no. 1–2:
91–105. https://doi.org/10.1075/gest.5.1‐2.08gom.
Goodwyn, S. W., L. P. Acredolo, and C. A. Brown. 2000. “Impact of
Symbolic Gesturing on Early Language Development.” Journal of
Nonverbal Behavior 24, no. 2: 81–103. https://doi.org/10.1023/a:
1006653828895.
Graham, K. E., C. Hobaiter, J. Ounsley, T. Furuichi, and R. W. Byrne.
2018. “Bonobo and Chimpanzee Gestures Overlap Extensively in
Meaning.” PLoS Biology 16, no. 2: e2004825. https://doi.org/10.1371/
journal.pbio.2004825.
Green, K., C. Osei‐Cobbina, M. Perlman, and S. Kita. 2023. “Infants Can
Create Different Types of Iconic Gestures, With and Without Parental
Scaffolding.” In Gesture and Speech in Interaction Conference (GeSpIn
2023).
Grund, C., G. Badihi, K. E. Graham, A. Safryghin, and C. Hobaiter. 2024.
“GesturalOrigins: A Bottom‐Up Framework for Establishing Systematic
Gesture Data Across Ape Species.” Behavior Research Methods 56, no. 2:
986–1001.
Gupta, S., and A. Sinha. 2016. “Not Here, There! Possible Referential
Gesturing During Allogrooming by Wild Bonnet Macaques, Macaca
radiata.” Animal Cognition 19, no. 6: 1243–1248. https://doi.org/10.
1007/s10071‐016‐1012‐3.
Gupta, S., and A. Sinha. 2019. “Gestural Communication of Wild Bonnet
Macaques in the Bandipur National Park, Southern India.” Behavioural
Processes 168: 103956. https://doi.org/10.1016/j.beproc.2019.103956.
Henderson, M., P. G. Grosz, K. E. Graham, C. Hobaiter, and P. Patel‐
Grosz. 2024. “Shared Semantics: Exploring the Interface Between Hu-
man and Chimpanzee Gestural Communication.” Mind and Language
39, no. 4: 454–471. https://doi.org/10.1111/mila.12500.
Hewes, G. W. 1974. “Gesture Language in Culture Contact.” Sign
Language Studies 4, no. 1: 1–34. https://doi.org/10.1353/sls.1974.0010.
Hobaiter, C., and R. W. Byrne. 2011. “The Gestural Repertoire of the
Wild Chimpanzee.” Animal Cognition 14, no. 5: 745–767. https://doi.
org/10.1007/s10071‐011‐0409‐2.
Hobaiter, C., and R. W. Byrne. 2014. “The Meanings of Chimpanzee
Gestures.” Current Biology 24, no. 14: 1596–1600. https://doi.org/10.
1016/j.cub.2014.05.066.
Holler, J., and S. C. Levinson. 2019. “Multimodal Language Processing
in Human Communication.” Trends in Cognitive Sciences 23, no. 8: 639–
652. https://doi.org/10.1016/j.tics.2019.05.006.
Hutcheson, K. 1970. “A Test for Comparing Diversities Based on
Shannon Formula.” Journal of Theoretical Biology 29, no. 1: 151–154.
https://doi.org/10.1016/0022‐5193(70)90124‐4.
Iverson, J. M., and M. K. Fagan. 2004. “Infant Vocal–Motor Coordina-
tion: Precursor to the Gesture–Speech System?” Child Development 75,
no. 4: 1053–1066. https://doi.org/10.1111/j.1467‐8624.2004.00725.x.
Iverson, J. M., and S. Goldin‐Meadow. 2005. “Gesture Paves the Way for
Language Development.” Psychological Science 16, no. 5: 367–371.
https://doi.org/10.1111/j.0956‐7976.2005.01542.x.
Iverson, J. M., and E. Thelen. 1999. “Hand, Mouth and Brain. The
Dynamic Emergence of Speech and Gesture.” Journal of Consciousness
Studies 6, no. 11–12: 19–40.
Jourdain, M. 2020. “A Quantitative and Qualitative Corpus Study of the
Acquisition of Topic Constructions in Child French.” Doctoral diss.,
PhD diss., Université de Lille and KU Leuven.
Jourdain, M. 2022. “The Emergence of Information Structure in Child
Speech: The Acquisition of c’est‐clefts in French.” Cognitive Linguistics
33, no. 1: 121–154. https://doi.org/10.1515/cog‐2021‐0022.
Karasik, L. B., C. S. Tamis‐LeMonda, and K. E. Adolph. 2014. “Crawling
and Walking Infants Elicit Different Verbal Responses From Mothers.”
Developmental Science 17, no. 3: 388–395. https://doi.org/10.1111/desc.
12129.
Kelly, S., M. Healey, A. Özyürek, and J. Holler. 2015. “The Processing of
Speech, Gesture, and Action During Language Comprehension.” Psy-
chonomic Bulletin and Review 22, no. 2: 517–523. https://doi.org/10.
3758/s13423‐014‐0681‐7.
Kendon, A. 1995. “Gestures as Illocutionary and Discourse Structure
Markers in Southern Italian Conversation.” Journal of Pragmatics 23,
no. 3: 247–279. https://doi.org/10.1016/0378‐2166(94)00037‐f.
Kendon, A. 2004. Gesture: Visible Action as Utterance. Cambridge, UK:
Cambridge University Press.
Kersken, V., J.‐C. Gómez, U. Liszkowski, A. Soldati, and C. Hobaiter.
2019. “A Gestural Repertoire of 1‐to 2‐Year‐Old Human Children: In
Search of the Ape Gestures.” Animal Cognition 22, no. 4: 577–595.
https://doi.org/10.1007/s10071‐018‐1213‐z.
Kirk, E., N. Howlett, K. J. Pine, and B. Fletcher. 2013. “To Sign or Not to
Sign? The Impact of Encouraging Infants to Gesture on Infant Language
and Maternal Mind‐Mindedness.” Child Development 84, no. 2: 574–590.
https://doi.org/10.1111/j.1467‐8624.2012.01874.x.
Kovács, A. M., T. Tauzin, E. Téglás, G. Gergely, and G. Csibra. 2014.
“Pointing as Epistemic Request: 12‐Month‐Olds Point to Receive New
Information.” Infancy 19, no. 6: 543–557. https://doi.org/10.1111/infa.
12060.
Krause, M. A., M. A. Udell, D. A. Leavens, and L. Skopos. 2018. “Animal
Pointing: Changing Trends and Findings From 30 Years of Research.”
Journal of Comparative Psychology 132, no. 3: 326–345. https://doi.org/
10.1037/com0000125.
Landis, J. R., and G. G. Koch. 1977. “An Application of Hierarchical
Kappa‐Type Statistics in the Assessment of Majority Agreement Among
Multiple Observers.” Biometrics 33, no. 2: 363–374. https://doi.org/10.
2307/2529786.
Leavens, D. A., and W. D. Hopkins. 1999. “The Whole‐Hand Point: The
Structure and Function of Pointing From a Comparative Perspective.”
Journal of Comparative Psychology 113, no. 4: 417–425. https://doi.org/
10.1037//0735‐7036.113.4.417.
16 of 18 Infancy, 2025
Levy, E. T., and D. McNeill. 2013. “Narrative Development as Symbol
Formation: Gestures, Imagery and the Emergence of Cohesion.” Cul-
ture & Psychology 19, no. 4: 548–569. https://doi.org/10.1177/
1354067x13500328.
Liszkowski, U., M. Carpenter, A. Henning, T. Striano, and M. Toma-
sello. 2004. “Twelve‐Month‐Olds Point to Share Attention and Interest.”
Developmental Science 7, no. 3: 297–307. https://doi.org/10.1111/j.1467‐
7687.2004.00349.x.
Liszkowski, U., M. Carpenter, T. Striano, and M. Tomasello. 2006. “12‐
and 18‐Month‐Olds Point to Provide Information for Others.” Journal of
Cognition and Development 7, no. 2: 173–187. https://doi.org/10.1207/
s15327647jcd0702_2.
Liszkowski, U., M. Carpenter, and M. Tomasello. 2007. “Pointing Out
New News, Old News, and Absent Referents at 12 Months of Age.”
Developmental Science 10, no. 2: F1–F7. https://doi.org/10.1111/j.1467‐
7687.2006.00552.x.
Lucca, K., and M. P. Wilbourn. 2018. “Communicating to Learn: Infants’
Pointing Gestures Result in Optimal Learning.” Child Development 89,
no. 3: 941–960. https://doi.org/10.1111/cdev.12707.
Lucca, K., and M. P. Wilbourn. 2019. “The What and the How:
Information‐Seeking Pointing Gestures Facilitate Learning Labels and
Functions.” Journal of Experimental Child Psychology 178: 417–436.
https://doi.org/10.1016/j.jecp.2018.08.003.
MacWhinney, B. 2000. The CHILDES Project: The Database, Vol. 2. New
Jersey, USA: Lawrence Erlbaum Associates Inc. Publishers.
Max Planck Institute for Psycholinguistics. 2023. “ELAN (Version 6.7).”
[Computer software], Language Archive. Nijmegen: Max Planck Insti-
tute for Psycholinguistics. https://archive.mpi.nl/tla/elan.
McNeill, D. 1992. Hand and Mind: What Gestures Reveal About Thought.
Chicago, USA: University of Chicago Press.
Mishra, A., V. Ceballos, K. Himmelwright, S. McCabe, and L. Scott.
2021. “Gesture Production in Toddlers With Autism Spectrum Disor-
der.” Journal of Autism and Developmental Disorders 51, no. 5: 1658–
1667. https://doi.org/10.1007/s10803‐020‐04647‐5.
Molesti, S., A. Meguerditchian, and M. Bourjade. 2020. “Gestural
Communication in Olive Baboons (Papio anubis): Repertoire and
Intentionality.” Animal Cognition 23, no. 1: 19–40. https://doi.org/10.
1007/s10071‐019‐01312‐y.
Morgenstern, A., and C. Parisse. 2012. “The Paris Corpus.” Journal of
French Language Studies 22, no. 1: 7–12. https://doi.org/10.1017/
s095926951100055x.
Nicoladis, E., R. I. Mayberry, and F. Genesee. 1999. “Gesture and Early
Bilingual Development.” Developmental Psychology 35, no. 2: 514–526.
https://doi.org/10.1037//0012‐1649.35.2.514.
Orr, E., and G. Kashy Rosenbaum. 2023. “The Context of Infants' Giving
Gestures in Mother‐Infant Dyads: Typical Giving Gestures and Those
Contingent on Exploration and Play.” Infancy 28, no. 3: 619–633.
https://doi.org/10.1111/infa.12531.
Özçalışkan, Ş., and S. Goldin‐Meadow. 2011. “Is There an Iconic
Gesture Spurt at 26 Months.” In Integrating Gestures: The Interdisci-
plinary Nature of Gesture. Amsterdam, NL: John Benjamins.
Patel‐Grosz, P. 2023. “The Search for Universal Primate Gestural Mean-
ings.” In Proceedings of Sinn und Bedeutung 27, 500–518 (November).
Perniss, P., R. L. Thompson, and G. Vigliocco. 2010. “Iconicity as a
General Property of Language: Evidence From Spoken and Signed
Languages.” Frontiers in Psychology 1: 227. https://doi.org/10.3389/
fpsyg.2010.00227.
Pika, S., and J. C. Mitani. 2009. “The Directed Scratch: Evidence for a
Referential Gesture in Chimpanzees.” Prehistory of Language 1:
166–181.
Pouw, W., and S. Fuchs. 2022. “Origins of Vocal‐Entangled Gesture.”
Neuroscience and Biobehavioral Reviews 141: 104836. https://doi.org/10.
1016/j.neubiorev.2022.104836.
Prieto Vives, P., A. Cravotta, O. Kushch, P. L. Rohrer, and I. Vilà‐
Giménez. 2018. “Deconstructing Beat Gestures: A Labelling Proposal.”
In Proceedings of the 9th International Conference on Speech Prosody,
2018 June 13–16; Poznań, Poland, edited by K. Klessa, J. Bachan, A.
Wagner, M. Karpiński, and D. Śledziński, 201–205. Lous Tourils: ISCA.
https://doi.org/10.21437/SpeechProsody.2018‐41.
R Core Team. 2021. R: A Language and Environment for Statistical
Computing. Vienna, Austria: R Foundation for Statistical Computing.
https://www.R‐project.org/.
Rodrigues, E. D., J. Maroco, and S. Frota. 2021. “Communicative Ges-
tures in 7–12‐Month Infants: A Phylogenetic Comparative Approach.”
Infant and Child Development 30, no. 5: e2262. https://doi.org/10.1002/
icd.2262.
Rohlng, K. J., A. Grimminger, and C. Lüke. 2017. “An Interactive View
on the Development of Deictic Pointing in Infancy.” Frontiers in Psy-
chology 8: 1319. https://doi.org/10.3389/fpsyg.2017.01319.
Rohrer, P. L., E. Delais‐Roussarie, and P. Prieto. 2020. “Beat Gestures for
Comprehension and Recall: Differential Effects of Language Learners
and Native Listeners.” Frontiers in Psychology 11: 575929. https://doi.
org/10.3389/fpsyg.2020.575929.
Rohrer, P. L., J. Florit‐Pons, I. Vilà‐Giménez, and P. Prieto. 2022.
“Children Use Non‐Referential Gestures in Narrative Speech to Mark
Discourse Elements Which Update Common Ground.” Frontiers in
Psychology 12: 661339. https://doi.org/10.3389/fpsyg.2021.661339.
Schlenker, P. 2020. “Gestural Grammar.” Natural Language and Lin-
guistic Theory 38, no. 3: 887–936. https://doi.org/10.1007/s11049‐019‐
09460‐z.
Shannon, C. E. 1949. The Mathematical Theory of Communication, by
CE Shannon (And Recent Contributions to the Mathematical Theory of
Communication), W. Weaver. Champaign, IL, USA: University of Illinois
Press.
Sjoberg, D. 2021. “Ggsankey: Sankey, Alluvial and Sankey Bump Plots.”
R Package Version 0.0, 99999.
Tamis‐LeMonda, C. S., M. H. Bornstein, and L. Baumwell. 2001.
“Maternal Responsiveness and Children's Achievement of Language
Milestones.” Child Development 72, no. 3: 748–767. https://doi.org/10.
1111/1467‐8624.00313.
Thelen, E. 1979. “Rhythmical Stereotypies in Normal Human Infants.”
Animal Behaviour 27: 699–715. https://doi.org/10.1016/0003‐3472(79)
90006‐x.
Tomasello, M., and J. Call. 2019. “Thirty Years of Great Ape Gestures.”
Animal Cognition 22, no. 4: 461–469. https://doi.org/10.1007/s10071‐
018‐1167‐1.
Tomasello, M., J. Call, K. Nagell, R. Olguin, and M. Carpenter. 1994.
“The Learning and Use of Gestural Signals by Young Chimpanzees: A
Trans‐Generational Study.” Primates 35, no. 2: 137–154. https://doi.org/
10.1007/bf02382050.
Tomasello, M., J. Call, J. Warren, G. T. Frost, M. Carpenter, and K.
Nagell. 1997. “The Ontogeny of Chimpanzee Gestural Signals: A Com-
parison Across Groups and Generations.” Evolution of Communication
1, no. 2: 223–259. https://doi.org/10.1075/eoc.1.2.04tom.
Tomasello, M., M. Carpenter, and U. Liszkowski. 2007. “A New Look at
Infant Pointing.” Child Development 78, no. 3: 705–722. https://doi.org/
10.1111/j.1467‐8624.2007.01025.x.
Vila‐Gimenez, I., and P. Prieto. 2021. “The Value of Non‐Referential
Gestures: A Systematic Review of Their Cognitive and Linguistic Ef-
fects in Children’s Language Development.” Children 8, no. 2: 148.
https://doi.org/10.3390/children8020148.
17 of 18
Wagner, P., Z. Malisz, and S. Kopp. 2014. “Gesture and Speech in
Interaction: An Overview.” Speech Communication 57: 209–232. https://
doi.org/10.1016/j.specom.2013.09.008.
Werner, H., and B. Kaplan. 1963. Symbol Formation. New York, USA:
Psychology Press.
Wilke, C., Ns. J. Lahiff, K. H. Sabbi, D. P. Watts, S. W. Townsend, and
K. E. Slocombe. 2022. “Declarative Referential Gesturing in a Wild
Chimpanzee (Pan troglodytes).” Proceedings of the National Academy of
Sciences 119, no. 47: e2206486119. https://doi.org/10.1073/pnas.22064
86119.
Supporting Information
Additional supporting information can be found online in the Sup-
porting Information section.
18 of 18 Infancy, 2025
