ArticlePDF Available

Prosody in Infant-Directed Speech Is Similar Across Western and Traditional Cultures


Abstract and Figures

When speaking to infants, adults typically alter the acoustic properties of their speech in a variety of ways compared with how they speak to other adults; for example, they use higher pitch, increased pitch range, more pitch variability, and slower speech rate. Research shows that these vocal changes happen similarly across industrialized populations, but no studies have carefully examined basic acoustic properties of infant-directed (ID) speech in traditional societies. Moreover, some scholars have suggested that ID speech is culturally specific and does not exist in some small-scale societies. We examined fundamental frequency (F0) production and speech rate in mothers speaking to both infants and adults in three cultures: Fijians, Kenyans, and North Americans. In all three cultures, speakers used higher F0 when speaking to infants relative to when speaking to other adults, and they also used significantly greater F0 variation and fewer syllables per second. Previous research has found that American mothers tend to use higher pitch than do mothers from other cultures, but when maternal education was controlled in the current study, we did not find a significant difference in average pitch across our three populations. This is the first research systematically comparing spontaneous ID and adult-directed speech prosody between Western and traditional societies, and it is consistent with a large body of evidence showing similar acoustic patterns in ID speech across industrialized populations.
Content may be subject to copyright.
This article was downloaded by: [Simon Fraser University]
On: 16 January 2015, At: 20:06
Publisher: Routledge
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered
office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK
Click for updates
Journal of Cognition and Development
Publication details, including instructions for authors and
subscription information:
Prosody in Infant-Directed Speech Is
Similar Across Western and Traditional
Tanya L. Broesch a b & Gregory A. Bryant c
a Simon Fraser University , Canada
b Emory University
c University of California , Los Angeles
Accepted author version posted online: 18 Oct 2013.Published
online: 16 May 2014.
To cite this article: Tanya L. Broesch & Gregory A. Bryant (2013): Prosody in Infant-Directed Speech
Is Similar Across Western and Traditional Cultures, Journal of Cognition and Development, DOI:
To link to this article:
Taylor & Francis makes every effort to ensure the accuracy of all the information (the
“Content”) contained in the publications on our platform. However, Taylor & Francis,
our agents, and our licensors make no representations or warranties whatsoever as to
the accuracy, completeness, or suitability for any purpose of the Content. Any opinions
and views expressed in this publication are the opinions and views of the authors,
and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content
should not be relied upon and should be independently verified with primary sources
of information. Taylor and Francis shall not be liable for any losses, actions, claims,
proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or
howsoever caused arising directly or indirectly in connection with, in relation to or arising
out of the use of the Content.
This article may be used for research, teaching, and private study purposes. Any
substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing,
systematic supply, or distribution in any form to anyone is expressly forbidden. Terms &
Conditions of access and use can be found at
Downloaded by [Simon Fraser University] at 20:06 16 January 2015
Prosody in Infant-Directed Speech Is Similar
Across Western and Traditional Cultures
Tanya L. Broesch
Simon Fraser University, Canada and Emory University
Gregory A. Bryant
University of California, Los Angeles
When speaking to infants, adults typically alter the acoustic properties of their speech in a variety of
ways compared with how they speak to other adults; for example, they use higher pitch, increased
pitch range, more pitch variability, and slower speech rate. Research shows that these vocal changes
happen similarly across industrialized populations, but no studies have carefully examined basic
acoustic properties of infant-directed (ID) speech in traditional societies. Moreover, some scholars
have suggested that ID speech is culturally specific and does not exist in some small-scale societies.
We examined fundamental frequency (F
) production and speech rate in mothers speaking to both
infants and adults in three cultures: Fijians, Kenyans, and North Americans. In all three cultures,
speakers used higher F
when speaking to infants relative to when speaking to other adults, and they
also used significantly greater F
variation and fewer syllables per second. Previous research has
found that American mothers tend to use higher pitch than do mothers from other cultures, but when
maternal education was controlled in the current study, we did not find a significant difference in
average pitch across our three populations. This is the first research systematically comparing
spontaneous ID and adult-directed speech prosody between Western and traditional societies, and
it is consistent with a large body of evidence showing similar acoustic patterns in ID speech across
industrialized populations.
Communicating with infants presents adults with a pragmatic challenge. Because babies have
limited communicative competence, adults must adjust many of their behaviors to accommodate
babies’ needs. There are a variety of documented behaviors that caregivers perform to accomplish
this. One common strategy is to modify the voice in a way that not only gets a baby’s attention,
but also helps direct behavior in a desired way. Compared with how adults speak to other adults,
parents often alter their speech patterns, specifically by adjusting prosodic characteristics such as
vocal pitch (making their voice go up and down in tone), amplitude (voice loudness), and speech
rate. Adults also use a variety of other techniques, such as simplifying their speech, using diminu-
tives, and singing (Ferguson, 1977; Fernald, 1992).
Tanya L. Broesch is currently at Simon Fraser University but was a student at Emory University during data
Correspondence should be sent to Tanya L. Broesch, Ph.D., Department of Psychology, Simon Fraser University,
8888 University Drive, Burnaby, BC, Canada V5A 1S6. E-mail:
Copyright #2014 Taylor & Francis Group, LLC
ISSN: 1524-8372 print=1532-7647 online
DOI: 10.1080/15248372.2013.833923
Downloaded by [Simon Fraser University] at 20:06 16 January 2015
Cross-cultural and cross-linguistic observations with mothers reveal that similar vocal
modifications transcend cultural and linguistic boundaries, suggesting that the tendency to
modify infant-directed (ID) speech is a species-specific adaptation designed to facilitate effective
mother–infant communication. However, although simplified speech patterns have been
observed in a number of cultures around the world (Ferguson, 1964), the acoustic properties
of ID speech have not been studied systematically in small-scale, traditional societies.
Documenting the existence of particular speech modifications across diverse cultures is impor-
tant in understanding whether this behavior represents a more general human communication
phenomenon or if it is culturally specific. In the present study, we recorded and acoustically
analyzed ID speech and adult-directed (AD) speech in three disparate cultures—rural Fijians,
the Bukusu of Kenya, and middle-class Americans—and confirmed our expectation that ID
speech manifests similarly across quite different cultural groups.
ID speech can be broken down into two basic elements: simplified speech register and
acoustic modification of speech. Simplified speech register has been previously documented
in both industrialized and nonindustrialized cultures, but the acoustic properties of speech have
been reported in only large urban and industrialized societies (Ferguson, 1964, 1977). The goal
of this study was to determine if mothers in traditional societies alter their speech acoustically
when speaking to an infant compared with when speaking to an adult. Additionally, it has been
proposed that acoustic modification is more typical when addressing prelinguistic infants,
whereas simplified speech is more common when addressing older infants and young children
(Fernald, 1992). Therefore, it is important to measure acoustic features of speech when address-
ing infants to determine if vocal changes are similar across diverse societies.
Simplified speech registers to infants and young children have been observed in many
European, Asian, and African cultures and are often assumed to be a human universal (Ferguson,
1964, 1977; Fernald, 1992). Evidence for ID speech exists in urban centers worldwide, such as
Australia, Japan, Thailand, Syria, the Middle East, India, Russia, France, Bangkok, Germany,
Mexico, the United States, Canada, and Great Britain, among others (Ferguson, 1964, 1977;
Fernald, 1992; Fernald et al., 1989; Kitamura, Thanavishuth, Burnham, & Luksaneeyanawin,
2002). Ferguson (1977) summarized evidence for adult vocal modification in 15 different
languages and 23 different societies and reported a difference in ‘‘speech register’’ by adults
when addressing children, which can refer to a change in any linguistic feature of speech, such
as vocabulary or syntax. In addition, work by Watson-Gegeo and Gegeo (1986) suggested that
Kwara’ae (Melanesian group of the Solomon Islands) caregivers modified their speech register
in similar ways to ID speech found in other societies. Although these researchers reported
linguistic variation (simplification of speech) when talking to young children, they did not pro-
vide any systematic analysis of the speech in that particular culture. Their evidence consisted of
self-reports, natural observations, and interviews about ID speech.
Some of the most detailed ethnographic observations and linguistic analyses of caregiver–
infant interactions in small-scale indigenous societies have come from work in Western Samoa
(Ochs, 1986; Schieffelin & Ochs, 1986). These researchers reported many differences from
American children in the way toddlers produced language, as well as differences in the ways par-
ents spoke to their children. Interestingly, Western Samoan caregivers were reported to not ‘‘com-
municate’’ with their infants in the typical way that is observed and described in the West (Ochs,
1982). In particular, observations of Americans indicate that mothers speak directly to infants
(even a few days old) and engage in face-to-face interaction, often interpreting movements and
Downloaded by [Simon Fraser University] at 20:06 16 January 2015
behaviors as intentional. This is in sharp contrast to what is observed in Western Samoan
parent–child interactions where infants are not spoken to for social engagement and their beha-
viors are not treated as social acts. Infants are sung to and cooed over for distraction rather than
as a conversational partner (Ochs, 1982). However, no acoustic analysis of speech has been
reported from their work in Western Samoa, leaving open the question of whether these mothers
modify their speech in ways similar to mothers in Western cultures.
Recent research examining the perception of ID speech in traditional societies suggests that
Western ID speech is highly recognizable by speakers regardless of the language they speak
(Bryant & Barrett, 2007; Bryant, Lie´nard, & Barrett, 2012). Bryant and Barrett (2007) recorded
American mothers producing ID and AD speech in four intention categories (prohibition, approval,
comfort, and attention), and the recorded tokens had all the typical acoustic features found
in ID speech across many languages. They then presented these recordings to Shuar
hunter-horticulturalists in Amazonian Ecuador and found that they were able to distinguish ID from
AD speech as well as reliably identify the intention categories of both kinds of speech, with an
advantage in ID speech. This work showed that even when presented with speech in a language
they do not understand, adult listeners can recognize important aspects of speakers’ informative
intentions, and ID speech is particularly well suited for the job. Bryant and Barrett (2007) attributed
this cross-cultural recognition success to the form–function relationship between physical
properties of signals and senders’ communicative intentions. This relationship underlies the
structure of signaling systems not only across human cultures, but across mammalian species as
well (E. S. Morton, 1977; Owren & Rendall, 2001). Because ID speech is often used with preverbal
babies, it is the sound that matters most, and so we should expect principles of animal signaling to
be operating that apply independent of language and culture. Not only should prosodically marked
ID speech be a common strategy across most if not all cultures, but the nature of the marking should
be quite similar as a function of the common communicative goals of speakers to infants.
To date, almost no research has examined closely the acoustic features of ID speech in
traditional societies. One study looking at three Quiche mothers found no evidence of systemati-
cally higher pitch when these women were speaking to their infants, but the results are difficult to
interpret given the extremely small sample size and post-hoc nature of the speech extraction
(Pye, 1986; Ratner & Pye, 1984). There is, however, strong evidence for acoustic correlates of
different affective communicative contexts across several languages in industrialized populations.
For example, Fernald and colleagues (1989) recorded mothers speaking to their infants in
German, French, Italian, and Mandarin Chinese and acoustically analyzed the prosodic contours
of the speech. They reported that different communicative contexts elicited different prosodic
forms. For example, prohibitive utterances tended to have abrupt amplitude onsets and lower
overall pitch, but approval utterances often had higher average pitch, and greater pitch variation,
including a prominent fundamental frequency (F
) rise–fall contour. The acoustic properties of a
mother comforting, prohibiting, approving, or getting the attention of a young child were distinct
across these intention categories and were similar across languages (Fernald, 1989).
To better understand the cultural similarities and variations of ID speech, we explored maternal
speech to infants and adults in three culturally distinct populations. Given the vast amount of eth-
nographic literature indicating that mothers modify their speech register when addressing infants,
as well as the highly similar mappings between acoustic structure and communicative function,
we expected the acoustic properties of ID speech to be similar across all societies, with mothers
increasing their average F
variation, and F
range when speaking to infants compared with
Downloaded by [Simon Fraser University] at 20:06 16 January 2015
when speaking to adults, as well as mothers speaking more slowly to infants than they speak to
other adults.
Participants and Location
Forty-three mothers were recorded producing both ID and AD speech. These paired samples
included 12 mothers from Fiji, 3 from Kenya, and 28 from the United States. The pitch profiles
of the Kenyan and Fijian mothers were highly similar, and due to the small Kenyan sample (3),
they were combined to comprise a non-Western group. The mean age of the mothers was 31;3
(SD ¼73 months >range ¼21;0–45;0) and there was no significant difference between Western
¼32;8; SD ¼59 months; range ¼24;0–42;0) and non-Western mothers (M
SD ¼86 months; range ¼21;0–45;0), t(24) ¼1.4, p¼.174. We collected maternal age infor-
mation for only 26 participants (11 non-Western, 15 Western) out of 43, due to experimenter error
(Western) and mothers not knowing their age (non-Western). During ID speech, mothers spoke to
their own infants, and the average age of these children was 8.3 months (SD ¼3.6 months;
range ¼2–16 months). There was no significant age difference between the non-Western infants
¼8.6 months; SD ¼4.5 months; range ¼2–16 months) and Western infants (M
months; SD ¼2.9 months; range ¼2–13 months), t(38) ¼0.43, p¼.67. Infant age information
was missing for 3 infants in the Western group. The amount of formal education attained
by all mothers was 13;11 (SD ¼50 months; range ¼7;0–22;0), and there was a significant
difference between non-Western mothers (9;6; SD ¼24 months; range ¼7;0–12;0) and Western
mothers (16;10; SD ¼2;1; range ¼14;0–22;0), t(23) ¼8.82, p<.001. We collected education
information for 25 participants (10 non-Western, 15 Western).
An additional 30 participants (26 non-Western, 4 Western) were analyzed for ID speech only,
as we did not have recordings of AD speech for them. The mean age of these mothers was 24;6
(SD ¼77 months; range ¼17;0–41;0). The mean age of the infants was 8.5 months (SD ¼3.9
months; range ¼2–15 months). We analyzed these 30 mothers together with the 43 who were
recorded producing both ID and AD speech, comprising a total sample of 73 for a separate
analysis of ID speech acoustic features.
Fiji. The study was conducted in two distinct locations in Fiji: the Yasawa Island Group,
located in the Northwestern group of the Fiji Islands, and the Lau Island Group, located in the
Southeastern group of the Fiji Islands. These two locations are similar in that they both rely on
subsistence agriculture and marine foraging and fishing for their livelihood (Henrich, 2004;
Sahlins, 1962). Each village has a population of less than 150, and there is no secondary school
on either of the island groups. From the Fijian mainland of Viti Levu, the villages on Yasawa
Island are 1 to 2 days of travel by boat, with air access. The villages of the Lau Island group
are 5 days by boat, with no other access (no air strip). These Fijian villages were largely devoid
of Western media influences, as neither island had access to television or newspapers at the time
of this study, with the exception of the medical station in the village of Tovu, Lau, which
occasionally allowed public viewings of sports games on specific days.
Early childhood practices are similar in the two regions as children are cared for primarily
by mothers in the first few years of life with help from other female villagers, but without
Downloaded by [Simon Fraser University] at 20:06 16 January 2015
significant alloparenting (H. Morton, 1996). Typically, mothers receive help in household
domestic chores while tending to infants in the first 3 to 6 months of life. In addition, it is
not uncommon for older siblings or cousins to help out with childcare, but they rarely do in
the 1st year of life. All mothers were the self-declared primary caregivers for all of our
infant–mother dyads in this study. All participants were recruited by word of mouth after the
village elders granted consent to conduct the study in these regions.
Kenya. The study was conducted in the Eastern region of Kenya near the rift valley in the
Bungoma district. Mothers were recruited and tested in Chemwa Village, a small village with
a population of 1,220 that is composed of 198 households and is located near a large town center
(Bungoma town). Access to electricity is limited so households experience little influence of
Western media (no television or magazines). Media exposure comes in the form of national
(Kenyan) newspapers and local radio. Television and Western media exist in Bungoma town,
which is about 1 hr away by bicycle taxi. The daily life of mothers consists of childrearing and
income generation activities, as well as participation in community organizations. Most house-
holds in this region rely on small-income stores and labor on sugarcane and maize crops for a
daily wage. The socialization goals and parenting practices of mothers in this region have been
identified as distinct from Western styles of parenting in the 1st year of life, with an emphasis
on nurturing and protecting rather than psychosocial development (LeVine, 1994). All mothers
in this sample were self-declared primary caregivers.
United States. The study was conducted in a psychology laboratory in Atlanta, GA. All
mothers were contacted through a database composed of families solicited through a variety of
recruitment methods such as mailings, local day-care centers, and birthing centers. All mothers
were self-declared primary caregivers.
Recording Procedure
Mothers were video-recorded digitally (Sony DCR-SR45, Dolby Digital AC-3 audio
compression, 448 kbps, 20 Hz–20 kHz frequency response) interacting with their infants as part
of a study examining mother–infant interactions and the early social environment across cultures.
Prior to recording, informed consent was obtained by a native speaker in each of the locations.
After consent was obtained, mothers and their infants were brought to the testing location and were
seated in a quiet corner of the room or outdoor area. They were asked to interact naturally with
their infant, with the goal of keeping the infant content for 10 min (see the Appendix for text of
the instructions). Both the mother and infant were seated on the floor, with the infant facing the
mother and within arm’s reach. The mothers were asked not to pick up the infant, but touching
was allowed at the mothers’ discretion. They were also told that if the infant cried, we would stop
the camera, but if the infant fussed, they could signal to us to stop and we would only use the first
few minutes of video, which would be sufficient for the study. Mothers were not explicitly
instructed to talk or play with their infants, but instead were told to ‘‘do whatever they preferred’’
as long as the infant stayed within view of the camera. To obtain AD speech, mothers were asked
questions at the beginning and end of their infant interactions. They were asked general descriptive
questions such as the age and sex of the infant, as well as questions about their thoughts of the
Downloaded by [Simon Fraser University] at 20:06 16 January 2015
Audio Extraction
The first 10 s of uninterrupted vocalization directed toward the infant (and toward the adult for
AD speech) was extracted for acoustic analysis. We defined a vocalization as any utterance or
sound coming from the adult while engaging with the infant (or adult), but we did not include
sounds derived from only lip movements (e.g., speech raspberries). These vocalizations were
extracted (i-movie software) and exported as uncompressed wav audio files (44.1 kHz, 16 bit,
mono). The recording context was originally designed to capture episodes of free interaction
between mothers and infants, so some mothers had only brief episodes of AD speech. In cases
where AD speech was less than 10 s in duration, we captured any instance of AD speech
(occasionally occurring during multiple vocalizations). Background and infant noise was not
edited out at this stage, but if the infant vocalized consistently for more than 3 s, it was
considered an interruption and the next 10-s instance was captured instead.
Acoustic Analysis
Due to the spontaneous nature of the recordings, many sounds were present in addition to the
mothers vocalizing and had to be edited out. All parts of the recordings that included nontarget
vocalizations (e.g., other people talking, crying babies, overlapping speech, animals, etc.) were
removed prior to acoustic analysis, and some recorded clips were not able to be analyzed (n¼5)
due to excessive nontarget vocalizations=sounds. In the ID speech clips, 75%of the original
recordings were retained on average. For AD speech recordings, clips were shorter and more
variable in length; however, 76%of the recordings were retained in the analyses, comparable
to the ID speech samples.
Edited audio clips were analyzed using Praat Version 5.2.21 (Boersma & Weenink, 2011).
We measured F
variability (F
SD), and minimum and maximum F
values using an
autocorrelation method. Octave errors and other analytical errors were removed by hand or were
fixed through pitch-setting adjustment. Default pitch settings suggested by Praat were used for
women (120–600 Hz) but changes to these settings were done on a case-by-case basis after
visual inspection of the F
values and never exceeded 20 Hz adjustment in the lower limit
and 60 Hz in the higher limit. F
SD and F
range were used as measures of pitch variation.
range was measured by subtracting the minimum F
from the maximum F
value for each
speaker. All F
values were converted to semitones (relative to 50 Hz) for statistical analyses.
Speech rate analysis was conducted on the edited clips of ID and AD recordings using a Praat
script that detects syllable nuclei automatically (De Jong & Wempe, 2009) and generates
syllables-per-second output. All files were low-pass filtered (4 kHz) to remove extraneous noise
that could affect intensity and voicing measurements used in the script algorithm.
We measured vocal pitch (F
) in mothers interacting with infants in a free interaction context and
speaking with another adult, and we expected that mothers would use higher pitch and greater
pitch range in ID speech compared with AD speech. Moreover, we expected that this difference
in speech type would be similar across cultural groups. A repeated-measures multivariate analysis
Downloaded by [Simon Fraser University] at 20:06 16 January 2015
of variance (MANOVA) was conducted, with speech type (ID or AD) as a within-subjects factor,
culture (Western and non-Western) as a between-subjects factor, and semitone conversions of
mean F
SD, and F
range as dependent variables. There was a main effect of speech type,
F(3, 39) ¼4.46, p<.01, partial g
¼.26, for mean F
and F
SD but not F
range, and there
was no Culture Speech Type interaction F(3, 39) ¼2.28, p>.05. Both Western and
non-Western mothers used higher F
when speaking to infants than when speaking to adults,
F(1, 41) ¼11.28, p<.01, partial g
¼.22. Both groups also had greater F
variation (F
F(1, 41) ¼5.00, p<.05, partial g
¼.12. There was no significant increase in F
(max F
– min F
), F(1, 41) ¼2.74, p¼.11, partial g
¼.06. Western and non-Western mothers
did not differ significantly on any of the pitch dimensions, other than a marginal difference
in overall F
in ID speech, with Western mothers using higher overall F
,F(1, 41) ¼3.43,
p¼.07, partial g
¼.08. Recorded clips varied in the percentage of usable sound for acoustic
analysis, and this was included in the model as a covariate but was not significant (F<1). See
Table 1 for means and standard deviations.
Means and Standard Deviations of Three Pitch Measurements and Speech Rate in Infant-Directed and
Adult-Directed Speech Across Non-Western and Western Cultural Groups (N¼43)
Speech type
Non-Western (n ¼15) Western (n ¼28)
Mean F
Range Syl=s Mean F
Range Syl=s
Infant-directed 252 (35.2) 55 (22.4) 243 (83.2) 3.09 (0.75) 287 (57.4) 68 (24.0) 302 (91.7) 3.09 (0.94)
Adult-directed 231 (39.7) 32 (22.9) 121 (100.0) 4.08 (0.65) 234 (37.5) 50 (21.5) 224 (107.0) 3.96 (0.74)
Note. All F
values are in Hertz. Standard deviations are in parentheses. Syl=s¼syllables per second.
FIGURE 1 Mean fundamental frequency (F
) by speech type (infant-directed speech [IDS]=adult-directed speech
[ADS]) and culture. (Color figure available online.)
Downloaded by [Simon Fraser University] at 20:06 16 January 2015
We also measured speech rate with the expectation that mothers would speak more slowly
when addressing infants than when speaking to another adult. A repeated-measures MANOVA
was conducted, with speech type (ID or AD) as a within-subjects factor, culture (Western and
non-Western) as a between-subjects factor, and speech rate (number of syllables per second)
FIGURE 2 Standard deviation of fundamental frequency (F
) by speech type (infant-directed speech [IDS]=adult-
directed speech [ADS]) and culture. (Color figure available online.)
FIGURE 3 Mean speech rate (syllables=second) by speech type (infant-directed speech [IDS]=adult-directed speech
[ADS]) and culture. (Color figure available online.)
Downloaded by [Simon Fraser University] at 20:06 16 January 2015
as the dependent variable. There was a significant effect of speech type, F(1, 41) ¼27.07,
p<.001, partial g
¼.40, with both Western and non-Western mothers using a slower speech
rate when speaking to infants (M¼3.09, SD ¼0.87) than when speaking to adults (M¼4.00,
SD ¼0.71), but there was not a significant effect for culture, F(1, 41) ¼0.11, p¼.74, partial
¼.03, nor was there a significant interaction of culture and speech rate, F(1, 41) ¼0.12,
p¼.73, partial g
¼.003. See Figures 1 through 3 for graphs of the results.
Average pitch was marginally different across cultures, so to further explore this, we exam-
ined 30 additional mothers (26 Western, 4 non-Western) who were recorded producing only ID
speech (i.e., no AD speech data available). We conducted a separate analysis of variance includ-
ing all 73 mothers with culture (Western or non-Western) as a between-subjects factor and mean
as the dependent measure. In this model, Western mothers produced significantly higher
mean F
than did non-Western mothers, F(1, 71) ¼9.38, p¼.003, d¼0.80. We had education
data for a subset of this sample (n¼39), and in this subset, the mean F
difference remained
significant, F(1, 37) ¼6.06, p¼.02, d¼0.92, but when mothers’ education was controlled as
a covariate, this effect was eliminated, F(1, 36) ¼0.866, p¼.36.
Communicating effectively with infants is a challenge shared by caregivers worldwide. Although
there are documented differences across cultures in the ways people interact with preverbal
babies, we should expect certain patterns to emerge independent of language and culture.
Research on the prosodic features of ID speech has shown that particular features of vocalizations
arise consistently across quite different languages and cultures. But most studies of ID speech
have been done using industrialized populations with common exposure to Western media.
We measured a perceptually salient acoustic dimension of ID speech (F
perceived as vocal pitch)
in mothers from two non-Western, traditional societies, and found that pitch changes in ID speech
relative to AD speech were very similar in these groups, as well as other languages in which
acoustic data have been reported (Fernald et al., 1989). Mothers in all three cultures increased
their vocal pitch when talking to babies and increased their pitch variability relative to when they
spoke to another adult. Additionally, we found that mothers in all cultures slowed their speaking
rate to infants relative to speaking to another adult, which is also consistent with previous work
examining ID speech across different languages. This research constitutes the first systematic,
basic acoustic analysis of ID speech from any traditional society and supports the hypothesis that
ID speech prosody manifests similarly across disparate cultures.
Because of close relationships between acoustic variables in vocal signals and their communi-
cative functions, we should expect a great deal of cross-cultural similarity, and even cross-species
similarity, in how form follows function (Bryant & Barrett, 2007; Cosmides, 1983; Fernald, 1992;
E. S. Morton, 1977; Owren & Rendall, 2001). ID speech in humans is a rather unique communi-
cative behavior among primates, but at least one other species, squirrel monkeys (Saimiri
boliuiensis peruuiensis), also alter their vocalizations in a context-dependent manner when
signaling to immature offspring (Biben, Symmes, & Bernhards, 1989). In our closest living
relatives, the chimpanzees, some modified gestures have been observed in mothers with their
infants, but no vocalizations resembling human ID speech have been documented (Falk,
2004). Falk (2004) argued that an increased juvenile dependency period led to selection pressure
Downloaded by [Simon Fraser University] at 20:06 16 January 2015
on mothers to be particularly skilled at maintaining close proximity to their infants through
gestures and vocalizations.
ID speech is a reliably developing trait in humans that helps solve adaptive problems asso-
ciated with effective communication with preverbal infants. We should expect cultural evol-
utionary processes to result in some differences across groups in the manifestation of any
communicative behavior. But studies of vocal emotion across cultures have revealed a great deal
of structural similarities independent of language, making many emotional vocalizations univer-
sally recognizable (e.g., Bryant & Barrett, 2007, 2008; Pell, Monetta, Paulmann, & Kotz, 2009;
Sauter, Eisner, Ekman, & Scott, 2010; Scherer, Banse, & Wallbott, 2001). The same regularities
that occur in vocal emotions across cultures also likely occur in ID speech given the close ties
between these phenomena (Fernald, 1992; Trainor, Austin, & Desjardins, 2000). Increases in
pitch and pitch variability in ID speech likely function to get infants’ attention and communicate
emotional intentions in a maximally effective way (Fernald, 1992; Werker & McLeod, 1989),
though some later developmental linguistic functions might be served as well, possibly with
trade-offs in effectiveness as the signals change over time (Fernald, 1992). A variety of factors
is at work in the shaping of maternal vocal strategies with infants, including developmental
issues and the local ecology. Although the current research does not speak to the specific role
of ID speech in cognitive development, it does support the notion that certain aspects of ID
speech function similarly across cultures given the accumulating evidence that the relative
changes from AD speech manifest quite similarly in all cultures studied to date. The current
study provides some evidence that mothers’ education can play a role in how mothers talk to
babies. We found that in a subsample of mothers for whom we had education information, cul-
tural differences in average pitch in ID speech were eliminated once the level of education was
controlled. These data suggest that mothers with higher education might use higher overall pitch
in ID speech, but not AD speech. The issue of how education, age, and culture interact with ID
speech certainly warrants further examination.
In cultures where caretakers rarely speak directly to infants, we should expect other nonvocal
strategies to fulfill common communicative goals such as getting an infant’s attention or com-
municating emotional intentions. But when caretakers are speaking to infants, similar forms will
likely emerge as a function of similar communicative goals across cultures. There still remains
the possibility for trade-offs between communicative strategies in particular contexts (e.g., Reilly
& Bellugi, 1996) or across languages. There is also evidence for functional trade-offs in prosody
between affective and linguistic forms that could have communicative importance in ID speech
(McRoberts, Studdert-Kennedy, & Shankweiler, 1995). For example, when affective prosody
is in organizational conflict with prosodic indicators of syntactic structure, speakers’ prosodic
strategies will vary depending on the age of the target child and the relationship between the
speaker and child (Kempe, Schaeffler, & Thoresen, 2010).
For decades, linguists and anthropologists have reported the richness of ID speech in great
detail—from alterations in the prosody to adjusting the content and vocabulary in an effort to
accommodate infants. Here we only explore two dimensions of a quite complex behavior. The nat-
ure of the physical recordings did not permit additional analyses of other potentially differentiating
acoustic variables such as breathiness, harmonics-to-noise ratio, or cepstral peak prominence that
correlate with perceptible voice quality changes. Future work should explore the different
communicative contexts in which ID speech is used and how context changes the various
physical properties of the signal. Studies have shown across languages that similarities across
Downloaded by [Simon Fraser University] at 20:06 16 January 2015
communicative contexts do exist (e.g., Fernald, 1992), but no work has examined specific infant
behavioral responses. A continuing effort of analyzing ID speech in indigenous cultures will help
further illustrate the apparent universality of the structural features of this way of speaking to
infants, and these data, combined with more nuanced work examining the communicative func-
tions over developmental time, will help us understand this important aspect of mother–infant
We thank all of the mothers and their infants for their participation, as well as the research
assistants in Fiji, Kenya, and the Emory Infant and Child Lab. We also thank Nivja de Jong
and Jennifer Widerhorn for assistance with the acoustic analysis. Portions of this work were
presented at the 23rd Annual Meeting of the Human Behavior and Evolution Society in
Montpellier, France.
Biben, M., Symmes, D., & Bernhards, D. (1989). Contour variables in vocal communication between squirrel monkey
mothers and infants. Developmental Psychobiology,22, 617–631.
Boersma, P., & Weenink, D. (2011). Praat: Doing phonetics by computer (Version 5.2.21) [Computer software].
Retrieved from
Bryant, G. A., & Barrett, H. C. (2007). Recognizing intentions in infant-directed speech: Evidence for universals.
Psychological Science,18, 746–751.
Bryant, G. A., & Barrett, H. C. (2008). Vocal emotion recognition across disparate cultures. Journal of Cognition and
Culture,8, 135–148.
Bryant, G. A., Lie´nard, P., & Barrett, H. C. (2012). Recognizing infant-directed speech across distant cultures: Evidence
from Africa. Journal of Evolutionary Psychology,10, 147–159.
Cosmides, L. (1983). Invariances in the acoustic expression of emotion during speech. Journal of Experimental
Psychology: Human Perception & Performance,9, 864–881.
De Jong, N. H., & Wempe, T. (2009). Praat script to detect syllable nuclei and measure speech rate automatically.
Behavior Research Methods,41, 385–390.
Falk, D. (2004). Prelinguistic evolution in early hominins: Whence motherese? Behavioral and Brain Sciences,27,
Ferguson, C. (1964). Baby talk in six languages. American Anthropologist,66, 103–114.
Ferguson, C. (1977). Baby talk as a simplified register. In C. Snow & C. Ferguson (Eds.), Talking to children
(pp. 219–236). Cambridge, UK: Cambridge University Press.
Fernald, A. (1989). Intonation and communicative intent in mothers’ speech to infants: Is the melody the message? Child
Development,60, 1497–1510.
Fernald, A. (1992). Human maternal vocalizations to infants as biologically relevant signals: An evolutionary
perspective. In J. H. Barkow, L. Cosmides, & J. Tooby (Eds.), The adapted mind: Evolutionary psychology and
the generation of culture (pp. 391–449). New York, NY: Oxford University Press.
Fernald, A., Taeschner, T., Dunn, J., Papousek, M., de Boysson-Bardies, B., & Fukui, I. (1989). A cross-language
study of prosodic modifications in mothers’ and fathers’ speech to preverbal infants. Journal of Child Language,
16, 447–501.
Henrich, J. (2004). Foundations of human sociality: Economic experiments and ethnographic evidence from fifteen
small-scale societies. New York, NY: Oxford University Press.
Kempe, V., Schaeffler, S., & Thoresen, J. C. (2010). Prosodic disambiguation in child-directed speech. Journal of
Memory and Language,62, 204–225.
Downloaded by [Simon Fraser University] at 20:06 16 January 2015
Kitamura, C., Thanavishuth, C., Burnham, D., & Luksaneeyanawin, S. (2002). Universality and specificity in infant-
directed speech: Pitch modifications as a function of infant age and sex in a tonal and non-tonal language. Infant
Behavior and Development,24, 372–392.
LeVine, R. (1994). Communication and social learning during infancy. In R. LeVine, S. Dixon, S. LeVine, A. Richman,
P. H. Leiderman, C. H. Keefer, & T. B. Brazelton (Eds.), Childcare and culture (pp. 196–223). Cambridge, UK:
Cambridge University Press.
McRoberts, G., Studdert-Kennedy, M., & Shankweiler, D. (1995). The role of fundamental frequency in
signaling linguistic stress and affect: Evidence for a dissociation. Perception & Psychophysics,57,
Morton, E. S. (1977). On the occurrence and significance of motivation-structural rules in some bird and mammal
sounds. The American Naturalist,111, 855–869.
Morton, H. (1996). Becoming Tongan: An ethnography of childhood. Honolulu, HI: University of Hawaii Press.
Ochs, E. (1982). Talking to children in Western Samoa. Journal of Language Socialization,2, 77–104.
Ochs, E. (1986). From feelings to grammar: A Samoan case study. In B. Schieffelin & E. Ochs (Eds.), Language socia-
lization across cultures: Vol. 3 Studies in the social and cultural foundations of language (pp. 251–270). Cambridge,
UK: Cambridge University Press.
Owren, M. J., & Rendall, D. (2001). Sound on the rebound: Bringing form and function back to the forefront in
understanding nonhuman primate vocal signaling. Evolutionary Anthropology,10, 58–71.
Pell, M. D., Monetta, L., Paulmann, S., & Kotz, S. A. (2009). Recognizing emotions in a foreign language. Journal of
Nonverbal Behavior,33, 107–120.
Pye, C. (1986). Quiche Mayan speech to children. Journal of Child Language,13, 85–100.
Ratner, N. B., & Pye, C. (1984). Higher pitch in BT is not universal: Acoustic evidence from Quiche Mayan. Journal of
Child Language,11, 515–522.
Reilly, J. S., & Bellugi, U. (1996). Competition on the face: Affect and language in ASL motherese. Journal of Child
Language,23, 219–239.
Sahlins, M. D. (1962). Culture and nature on a Fijian island. Ann Arbor, MI: University of Michigan Press.
Sauter, D., Eisner, F., Ekman, P., & Scott, S. K. (2010). Cross-cultural recognition of basic emotions through nonverbal
emotional vocalizations. Proceedings of the National Academy of Sciences,107, 2408–2412.
Scherer, K. R., Banse, R., & Wallbott, H. G. (2001). Emotion inferences from vocal expression correlate across languages
and cultures. Journal of Cross Cultural Psychology,32, 76–92.
Schieffelin, B., & Ochs, E. (1986). Language socialization. Annual Review of Anthropology,15, 163–191.
Trainor, L. J., Austin, C. M., & Desjardins, R. N. (2000). Is infant-directed speech prosody a result of the vocal
expression of emotion? Psychological Science,11, 188–195.
Watson-Gegeo, K., & Gegeo, D. (1986). Calling-out and repeating routines in Kwara’ae children’s language socialization.
In B. Schieffelin & E. Ochs (Eds.), Language socialization across cultures: Vol. 3 Studies in the social and cultural
foundations of language (pp. 17–50). Cambridge, UK: Cambridge University Press.
Werker, J. F., & McLeod, P. J. (1989). Infant preference for both male and female infant-directed talk: A
developmental study of attentional and affective responsiveness. Canadian Journal of Psychology,43,
Script of Instructions for Mothers Provided Prior to Video Recording
We will be videotaping you interacting with your infant. Please just do whatever you
normally do with your infant to keep him or her engaged. We will be videotaping and ask that
you continue to interact with your child=infant for 10 min. Please do not lift up your infant
at anytime as we have the camera on the infant. Please do not use any props. Thank you
very much.
Downloaded by [Simon Fraser University] at 20:06 16 January 2015
Tutakuwa tukichukua video yako ukihusiana na mtoto wako. Tafadhali fanya jinsi wewe
hufanya kwa kawaida ili kumfanya angalau awe anafanya jambo, yani, asisubae. Tutakuwa
tukichukua video na tunakuhimiza kuendelea kucheza na mtotowako kwa dakika 10. Tafadhali
usimuinue mwanako kwasababu tumeangalisha kamera kwake. Tafahdhali usitumievihimili
vyovyote. Asante sana.
O keitou na tabaka tiko na nomudrau veimaliwai kei na levumu. Keitou na qai kerea kevaka o
rawa ni o cakava ga na ka o dau cakava e na veisiga me dau vakawelei koya kina. Keitou na
tabaka me 10 na miniti na nomudrau tiko vata. Keitou kerea talega me o kakua ni o keveta
na luvemu ka ni na tabaki tiko. E kerei talega me o kakua ni vakyagataka e dua na vakawele.
Vinaka vakalevu.
Downloaded by [Simon Fraser University] at 20:06 16 January 2015
... Nonetheless, various studies revealed the presence of CDC features in non-WEIRD cultures (e.g., higher pitch [112]; slower speaking rate [113]; and repetitions, diminutives, and simpler syntax [114]). Overall, the results suggest that both similarities (e.g., in pitch [113]) and differences [115] between WEIRD and non-WEIRD cultures do exist. ...
... Nonetheless, various studies revealed the presence of CDC features in non-WEIRD cultures (e.g., higher pitch [112]; slower speaking rate [113]; and repetitions, diminutives, and simpler syntax [114]). Overall, the results suggest that both similarities (e.g., in pitch [113]) and differences [115] between WEIRD and non-WEIRD cultures do exist. However, not all CDC features can be found in every culture. ...
Full-text available
Humans communicate with small children in unusual and highly conspicuous ways (child-directed communication (CDC)), which enhance social bonding and facilitate language acquisition. CDC-like inputs are also reported for some vocally learning animals, suggesting similar functions in facilitating communicative competence. However, adult great apes, our closest living relatives, rarely signal to their infants, implicating communication surrounding the infant as the main input for infant great apes and early humans. Given cross-cultural variation in the amount and structure of CDC, we suggest that child-surrounding communication (CSC) provides essential compensatory input when CDC is less prevalent—a paramount topic for future studies.
... There are a variety of choices for a caregiver when attempting to verbally encourage some behaviour in a young child (i.e. approval utterances). Local social conventions could shape caregivers' behaviour to slow down precipitously, for example, or perhaps to increase the pitch, or even do both [62]. In some societies, speaking directly to infants is relatively less common (e.g. ...
Like most human non-verbal vocalizations, laughter is produced by speakers of all languages, across all known societies. But despite this obvious fact (or perhaps because of it), there is little comparative research examining the structural and functional similarity of laughter across speakers from different cultures. Here, we describe existing research examining (i) the perception of laughter across disparate cultures, (ii) conversation analysis examining how laughter manifests itself during discourse across different languages, and (iii) computational methods developed for automatically detecting laughter in spoken language databases. Together, these three areas of investigation provide clues regarding universals and cultural variations in laughter production and perception, and offer methodological tools that can be useful for future large-scale cross-cultural studies. We conclude by providing suggestions for areas of research and predictions of what we should expect to discover. Overall, we highlight how important questions regarding human vocal communication across cultures can be addressed through the examination of spontaneous and volitional laughter. This article is part of the theme issue ‘Cracking the laugh code: laughter through the lens of biology, psychology and neuroscience’.
... CDS has characteristics that are shared across various cultures (Broesch & Bryant, 2015;Ferguson, 1977;Fernald, 1989), but the extent to which CDS and adultdirected speech (ADS) differ from one another varies across cultures (Broesch & Bryant, 2018;Farran et al., 2016;Fernald et al., 1989). Despite this variability, listeners can readily distinguish whether the speech they hear addresses a child or an adult and extract intentions from CDS even in unfamiliar languages (Bryant & Barrett, 2007;Fernald, 1989;Golinkoff & Alioto, 1995). ...
This study investigated adult listeners’ ability to detect age-related cues in child-directed speech (CDS). Participants ( N = 186) listened to two speech recordings directed at children between the ages of 6 to 44 months and guessed which had addressed a younger or an older child. The recordings came from North American English-speaking mothers and listeners were native speakers of Turkish with varying degrees of English knowledge. Participants were randomly assigned to listen either to the original recordings or to the low-pass filtered versions. Accuracy was above chance level across all groups. Participants’ English level, age and the age difference between the addressees significantly predicted accuracy. After controlling for these variables, we found a significant effect of condition. Participants’ accuracy tended to be better in the unfiltered condition with the exception of male participants without children. These results suggest that age-related variations in child-directed speech are perceptually available to adult listeners. Further, even though sensitivity to the age-related cues is facilitated by the availability of content-related cues in speech, it does not seem to be solely dependent on these cues, providing further support for the form-function relations in CDS.
... The majority of studies on IDS have been conducted with American English populations (for the overall prevalence of English in child language studies, see Kidd & Garcia, 2021) which have been described as having more extreme IDS properties than other languages , suggesting that generalisability might be limited. While higher and more variable pitch might be the two robust characteristics of IDS present across cultures and languages (Broesch & Bryant, 2015;Farran et al., 2016;McClay et al., 2021;Narayan & McDermott, 2016; but see also Han et al., 2020Han et al., , 2021, vowel space expansion, on the other hand, has not been reported consistently across languages. For instance, increased vowel space expansion in IDS vs. ADS has not been found in Dutch (Benders, 2013), German (Audibert & Falk, 2018), Cantonese (Xu Rattanasone et al., 2013), Lenakel and Southwest Tanna (McClay et al., 2021), and inconsistently reported in Norwegian (Englund & Behne, 2006;Kartushina, Robbestad, et al., 2021). ...
Full-text available
The present study examines the acoustic properties of infant-directed speech (IDS) as compared to adult-directed speech (ADS) in Norwegian parents of 18-month-old toddlers, and whether these properties relate to toddlers’ expressive vocabulary size. Twenty-one parent- toddler dyads from Tromsø, Northern Norway participated in the study. Parents (16 mothers, 5 fathers), speaking a Northern Norwegian dialect, were recorded in the lab reading a storybook to their toddler (IDS register), and to an experimenter (ADS register). The storybook was designed for the purpose of the study, ensuring identical linguistic contexts across speakers and registers, and multiple representations of each of the nine Norwegian long vowels. We examined both traditionally reported measures of IDS: pitch, pitch range, vowel duration and vowel space expansion, but also novel measures: vowel category compactness and vowel category distinctiveness. Our results showed that Norwegian IDS, as compared to ADS, had similar characteristics as in other languages: higher pitch, wider pitch range, longer vowel duration, and expanded vowel space area; in addition, it had less compact vowel categories. Further, parents’ hyper-pitch, that is, the within-parent increase in pitch in IDS as compared to ADS, and vowel category compactness in IDS itself, were positively related to toddlers' vocabulary. Our results point towards potentially facilitating roles of parents’ increase in pitch when talking to their toddler and of consistency in vowel production in early word learning.
... Evidence suggests not only that WEIRD parents produce ID speech more often, but that some do it in more extreme ways than others. Broesch & Bryant [82] found that US and non-Western (Fijian and Kenyan) mothers both produced ID speech using higher f o and f o variation (s.d.) than when producing AD speech, but US mothers increased f o to a greater extent than their non-Western counterparts. Interestingly, when mothers' education was controlled in the analysis, the cultural difference disappeared. ...
The study of human vocal communication has been conducted primarily in Western, educated, industrialized, rich, democratic (WEIRD) societies. Recently, cross-cultural investigations in several domains of voice research have been expanding into more diverse populations. Theoretically, it is important to understand how universals and cultural variations interact in vocal production and perception, but cross-cultural voice research presents many methodological challenges. Experimental methods typically used in WEIRD societies are often not possible to implement in many populations such as rural, small-scale societies. Moreover, theoretical and methodological issues are often unnecessarily intertwined. Here, I focus on three areas of cross-cultural voice modulation research: (i) vocal signalling of formidability and dominance, (ii) vocal emotions, and (iii) production and perception of infant-directed speech. Research in these specific areas illustrates challenges that apply more generally across the human behavioural sciences but also reveals promise as we develop our understanding of the evolution of human communication. This article is part of the theme issue ‘Voice modulation: from origin and mechanism to social impact (Part II)’.
Full-text available
When interacting with infants, humans often alter their speech and song in ways thought to support communication. Theories of human child-rearing, informed by data on vocal signalling across species, predict that such alterations should appear globally. Here, we show acoustic differences between infant-directed and adult-directed vocalizations across cultures. We collected 1,615 recordings of infant- and adult-directed speech and song produced by 410 people in 21 urban, rural and small-scale societies. Infant-directedness was reliably classified from acoustic features only, with acoustic profiles of infant-directedness differing across language and music but in consistent fashions. We then studied listener sensitivity to these acoustic features. We played the recordings to 51,065 people from 187 countries, recruited via an English-language website, who guessed whether each vocalization was infant-directed. Their intuitions were more accurate than chance, predictable in part by common sets of acoustic features and robust to the effects of linguistic relatedness between vocalizer and listener. These findings inform hypotheses of the psychological functions and evolution of human communication.
Having recognized the need for diversity spotlighted by Kidd and Garcia – but given that sampling all the world’s languages is infeasible – we focus on which dimensions of variability researchers should prioritize. We consider three major approaches to the study of child language learning, namely, language as a (1) cognitive puzzle, (2) clinical/educational object, and (3) window onto socialization. We discuss how what is important about ‘diversity’ from each of these perspectives dictates the sociolinguistic communities from which researchers should sample.
Thanks to the French-German Agenda 2020, many new French-German day-care centres have opened in the region of the Upper Rhine, guaranteeing an early bilingual and bicultural upbringing.The present thesis elaborates an inventory of the contents, strategies and language practices offered in the respective structures and, even more important, of the linguistic and cultural interactional competencies that can actually be acquired by the children visiting these day-care centres.For this purpose, the present thesis analyses different corpora such as the legal framework for early childhood care in France and Germany, the websites of the French-German day-care centres of the Upper Rhine, the interviews made with the centres’ employees and directors, and, last not least, the field notes taken in 3 centres that have been visited as well as 28 hours of audiorecorded caretaker-child interactions. In so doing, it uses an ethnolinguistic research design in order to identify fields of tension as well as the synergies to be exploited.
Full-text available
Infant-directed speech (IDS), a speaking style distinguished by its higher pitch, slower tempo, and exaggerated intonation, has been documented in speech directed towards infants across many cultures and languages. Previous research shows that IDS in the context of parent-infant interactions is associated with advances in children's language learning. While we have long known that fathers, like mothers, produce IDS, most research on IDS to this day has focussed exclusively on female (maternal) speech. In light of the recent societal changes which have increased fathers' time spent in hands-on activities with children, I argue that the research on IDS needs to move away from using the maternal template. I first describe why IDS plays a central role in language development. I then outline known similarities and differences between paternal and maternal IDS, and point to potential biases and challenges within the research that has been conducted so far. Finally, I outline why investigators should strive to include fathers in research on IDS, and conclude with recommendations for follow-up work that can help advance our understanding of fathers' contributions to children's linguistic development.
Full-text available
Whereas the perception of emotion from facial expression has been extensively studied cross-culturally, little is known about judges’ ability to infer emotion from vocal cues. This article reports the results from a study conducted in nine countries in Europe, the United States, and Asia on vocal emotion portrayals of anger, sadness, fear, joy, and neutral voice as produced by professional German actors. Data show an overall accuracy of 66% across all emotions and countries. Although accuracy was substantially better than chance, there were sizable differences ranging from 74% in Germany to 52% in Indonesia. However, patterns of confusion were very similar across all countries. These data suggest the existence of similar inference rules from vocal expression across cultures. Generally, accuracy decreased with increasing language dissimilarity from German in spite of the use of language-free speech samples. It is concluded that culture- and language-specific paralinguistic patterns may influence the decoding process.
When communicating with infants, caregivers often modulate their speech in an effort to make their communicative and informative intentions more clear. Infant-directed (ID) speech differs acoustically from adult-directed (AD) speech, and systematically varies according to different kinds of intentions. This way of speaking to infants is thought to be a species-specific adaptation, as research has documented highly similar patterns in ID speech across a variety of cultures. A recent study has also shown that people from an indigenous non-Western culture (Shuar of Amazonian Ecuador) can reliably discriminate ID speech from AD speech in a language they do not speak, and distinguish between four different intention categories (prohibition, attention, comfort, and approval). The current research attempted to replicate this finding in a traditional African population, the Turkana of northwestern Kenya. In three experiments, we found that Turkana adults were able to discriminate between ID and AD speech produced in English by American mothers, and they could also distinguish between several intention categories in both ID and AD speech. Signal detection analysis revealed that ID speech was marginally more discriminable than AD speech, but overall rate of intention recognition was similar across speech types. These results partially support the hypothesis that ID speech is universally recognizable due to the formfunction relationship between acoustic signals and their communicative purpose, but there were differences in performance between Turkana and Shuar that merit further investigation.