Cognitive Science 38 (2014) 1675–1686

Copyright © 2014 Cognitive Science Society, Inc. All rights reserved.

ISSN: 0364-0213 print / 1551-6709 online
DOI: 10.1111/cogs.12128

Foundational Tuning: How Infants’ Attention to Speech

Predicts Language Development
Athena Vouloumanos,a Suzanne Curtinb
a
Department of Psychology, New York University
b
Department of Psychology, University of Calgary
Received 6 October 2012; received in revised form 13 September 2013; accepted 20 September 2013

Abstract
Orienting biases for speech may provide a foundation for language development. Although
human infants show a bias for listening to speech from birth, the relation of a speech bias to later
language development has not been established. Here, we examine whether infants’ attention to
speech directly predicts expressive vocabulary. Infants listened to speech or non-speech in a pref-
erential listening procedure. Results show that infants’ attention to speech at 12 months signifi-
cantly predicted expressive vocabulary at 18 months, while indices of general development did
not. No predictive relationships were found for infants’ attention to non-speech, or overall atten-
tion to sounds, suggesting that the relationship between speech and expressive vocabulary was not
a function of infants’ general attentiveness. Potentially ancient evolutionary perceptual capacities
such as biases for conspecific vocalizations may provide a foundation for proficiency in formal
systems such language, much like the approximate number sense may provide a foundation for
formal mathematics.

Keywords: Cognitive development; Language acquisition; Speech perception; Longitudinal;

Predictor

1. Introduction

To learn language, children must be able to select meaningful signals among the range
of stimuli in their environment. This requires infants learning a spoken language to distin-
guish and attend to speech—linguistic sounds generally produced by the human vocal
tract. Many species show a bias for their own species’ (i.e., conspecific) vocalizations
over other species’ vocalizations. For example, swamp sparrows preferentially learn con-
specific song compared with heterospecific sparrow song (e.g., Marler & Peters, 1977).

Correspondence should be sent to Athena Vouloumanos, Department of Psychology, New York Univer-
sity, 6 Washington Place, New York, NY 10003. E-mail: athena.vouloumanos@nyu.edu
1676 A. Vouloumanos, S. Curtin / Cognitive Science 38 (2014)

Human infants have similar biases for the communicative signals of their species, privi-
leging speech over filtered speech (Spence & DeCasper, 1987), warbled tones (Samples
& Franklin, 1978), white noise (Butterfield & Siperstein, 1970; Colombo & Bundy,
1981), and synthetic sine-wave analogs of speech (Vouloumanos & Werker, 2004, 2007).
Recently, Shultz and Vouloumanos (2010) found that 3-month-old infants preferred lis-
tening to speech compared with other naturally occurring sounds in their environment,
even other human communicative sounds such as laughter. A bias for speech, apparent
even in newborns (Vouloumanos & Werker, 2007), may constitute an orienting mecha-
nism toward conspecific linguistic vocalizations that provides a foundation for subsequent
language development. In this paper, we examine whether an attentional bias for speech
predicts later language development. Extracting species-specific linguistic signals from
their auditory environments could allow infants to analyze speech signals more thor-
oughly, providing a mechanism for early language acquisition.
Language development is influenced by competence in underlying sensory systems;
specifically, spoken language perception relies on basic auditory processing. For example,
infants’ rapid auditory processing abilities at 7.5 months predict emerging language abili-
ties at 24 months (Benasich & Tallal, 2002). Findings with children yield similar results;
identification of speech sounds in noise predicts language impairment in children with
Specific Language Impairment, and in typically developing participants (Ziegler, Pech-
Georgel, George, Alario, & Lorenzi, 2005). Language development is thus affected by
competence in the basic sensory systems that support its functions: Individuals with better
auditory skills achieve better spoken language proficiency.
More specifically, early speech perception abilities predict some aspects of later lan-
guage skills. For example, infants’ discrimination of two acoustically distinct vowels
(/u/ and /y/) at 6 months significantly correlates with language abilities at 13–
24 months of age (Tsao, Liu, & Kuhl, 2004). Similarly, performance on a word–object
associative learning task using minimally differing words, such as bih and dih (Stager
& Werker, 1997) at 17 and 20 months is related to performance on children’s stan-
dardized tests of language comprehension and production up to two and a half years
later (Bernhardt, Kemp, & Werker, 2007). This suggests that infants’ ability to associ-
ate minimally distinct words to objects, which requires sensitivity to phonological dif-
ferences, might be related to later word learning abilities. Other speech processing
abilities such as speech segmentation at 12 months also correlate with expressive
vocabulary at 24 months and language skills at 4–6 years, though not with IQ (New-
man, Ratner, Jusczyk, Jusczyk, & Dow, 2006). At the same time, studies of infants’
lexical processing show that speed and accuracy in word recognition at 25 months are
correlated with earlier lexical and grammatical development from 12 to 25 months and
with expressive language, IQ and working memory at 8 years; but speed of word rec-
ognition at 15 and 18 months does not predict measures of lexical and grammatical
development at 25 months, leaving open the specific relationships between the con-
structs (Fernald, Perfors, & Marchman, 2006; Marchman & Fernald, 2008). However,
as these authors acknowledge, the relationship between speech perception abilities and
 A. Vouloumanos, S. Curtin / Cognitive Science 38 (2014) 1677

language development could be due to non-linguistic factors such as general cognitive

and/or social skills, rather than being specific to language learning.
Indeed, general cognitive factors influence speech perception abilities in infancy. For
example, infants between 8 and 11 months who succeeded on A-not-B and means-end
tasks showed reduced discrimination of non-native speech contrasts, while their native
speech discrimination remained intact (Conboy, Sommerville, & Kuhl, 2008; Lalonde &
Werker, 1995). The negative relationship between discrimination of non-native speech
contrasts and cognitive control may result from infants’ developing ability to attend to
relevant rather than irrelevant information (e.g., Werker & Curtin, 2005). Both the cogni-
tive tasks and the speech perception tasks required infants to inhibit attention to irrelevant
information and tapped into the same underlying ability: inhibitory control (Conboy et al.,
2008; Lalonde & Werker, 1995). Moreover, one of these studies suggested a specific rela-
tionship between 11-month-old infants’ attention to relevant information (reflected in bet-
ter discrimination of native speech contrasts) and their vocabulary size assessed at the
same age (Conboy et al., 2008). These studies did not, however, examine whether atten-
tion to relevant information at an earlier age may predict aspects of language develop-
ment such as expressive vocabulary at a later age.
Preliminary evidence that individual infants’ attention to speech sounds more generally
could have an impact on their language development comes from studies of atypical lan-
guage development. Kuhl, Coffey-Corina, Padden, and Dawson (2005) tested auditory
preferences for infant-directed speech samples pitted against non-speech analogs in 2.5–
4-year-old children with autism spectrum disorder (ASD). They found that the children
with ASD who preferred listening to non-speech over speech were more likely to exhibit
deficits in expressive language ability, suggesting that attention to speech is related to
language production within individual children. Further, children with language impair-
ments also show atypical preferences for particular types of speech stimuli relative to typ-
ically developing children. Five-year olds with autism do not orient reliably to their name
(Dawson, Meltzoff, Osterling, Rinaldi, & Brown, 1998) and 3–4-year-olds with autism
prefer listening to superimposed voices over their mother’s voice (Klin, 1992). This con-
trasts with typically developing children, who attend to their name by 4.5 months of age
(Mandel, Jusczyk, & Pisoni, 1995) and prefer listening to their mother’s voice from birth
(DeCasper & Fifer, 1980). Early processing differences for speech may also predict read-
ing ability in school-age children, as neonatal electrophysiological responses to speech
and non-speech predicted children who were dyslexic or below average readers at 8 years
of age (Molfese, 2000). Together, this work suggests that the absence of typical responses
to speech may be associated with delays or impairments in language development. How-
ever, the relation between a speech bias and language development in the general popula-
tion, and in particular the degree to which attention to speech predicts later expressive
vocabulary development, has not been examined. Nor is it known whether early speech
processing is related specifically to later language abilities or whether it reflects more
general cognitive abilities.
In this study, we examine a cohort of infants prospectively and test whether attention
to speech at 12 months predicts expressive vocabulary at 18 months. We also examine
1678 A. Vouloumanos, S. Curtin / Cognitive Science 38 (2014)

the degree to which infants’ general cognitive and motor development correlates with
attention to speech and language outcomes at 18 months. At the same time, we consider
two alternative possibilities: that language development is related to increased general
attentiveness to sounds, or to more advanced general development, two possibilities
which would ground language development in earlier general cognitive competencies
(e.g., Lalonde & Werker, 1995) rather than more specific attention to speech.

2. Method

2.1. Participants

Thirty infants (Mage = 12.5 months, SD = 0.26) participated in this study. Of these 30
infants, 27 completed the experimental task. We collected observational data (see below)
from 29 infants at 12 months and 26 at 18 months, and parental reports of language
development (see below) from 26 infants at 12 months and from 23 at 18 months.

2.2. Stimuli

Auditory stimuli were of two types (used in Vouloumanos & Werker, 2004, 2007): a
speech set composed of 12 monosyllabic nonsense tokens of “lif” and “neem” spoken by a
female native English speaker, and a non-speech set composed of 12 complex non-speech
analogs matched to the 12 speech tokens. Each complex non-speech token consisted of
four time-varying sinusoidal waves that track the center of the fundamental frequency
(preserving pitch contour) and portions of the first three formants (F1, F2, and F3) of its
speech counterpart while retaining the amplitude envelope, relative formant amplitude,
and relative intensity of speech (Remez, Rubin, Berns, Pardo, & Lang, 1994; Remez,
Rubin, Pisoni, & Carrell, 1981). Specifically, for the “lif” non-speech analogs, sine waves
tracked the first formant of the initial consonant segment (“l”), and the first three formants
of the vocalic segment (“i”). The analog for the fricative “f” was created using a filtered
white noise generator. For the “neem” non-speech analogs, sine waves tracked the first for-
mant of the initial (“n”) and final (“m”) consonants, and the first three formants of the
vocalic segment (“ee”). Adults did not spontaneously perceive the non-speech analogs as
speech (Vouloumanos, Kiehl, Werker, & Liddle, 2001). Each set of 12 tokens varied in
intonational contour (average minimum and maximum pitch: 197 and 350 Hz, respec-
tively) and duration (525–1,155 ms). Experimental trials consisted of tokens separated by
300–500 ms of silence presented at 65  5 dB.

2.3. Experimental procedure

Infants were tested using a modified version of the sequential looking preference pro-
cedure (Cooper & Aslin, 1990, 1994; Vouloumanos & Werker, 2004). Infants were seated
on a parent’s lap in front of a single video monitor. A movie of a flashing light was used
 A. Vouloumanos, S. Curtin / Cognitive Science 38 (2014) 1679

to attract the infant’s attention to the screen. Once the infant fixated on the screen, the
trial was initiated. A black-and-white checkerboard was displayed in tandem with one set
of sounds. Five speech trials and five non-speech trials were presented in random order.
Stimulus presentation was of a fixed length: A trial consisted of tokens separated by 300–
450 ms of silence, for a fixed trial length of 40 s. The presentation of auditory and visual
stimuli was controlled by Habit X (Cohen, Atkinson, & Chaput, 2004). Infants were vid-
eotaped during the task to allow for offline coding. The main dependent measure was
looking time to the checkerboard for each sound type.

2.4. Measures of linguistic development and general development

We assessed language development using the MacArthur-Bates Communicative Devel-

opment Inventory (MB-CDI; Feldman et al., 2000; Fenson et al., 1994). The MB-CDI is
a widely used parental report measure that has been validated in both typically develop-
ing and in atypical populations. Parents report children’s word comprehension and
production. We converted the raw MB-CDI scores to percentiles (using published age-
appropriate norms; Fenson et al., 2007) and used these in all analyses to test the
relationship between attention to speech and expressive vocabulary.
Infants’ general development was assessed using the Mullen Scales of Early Learning
(Mullen, 1995), a laboratory-based observational measure. The Mullen is a comprehen-
sive assessment of language, motor, and perceptual abilities for children of all ability lev-
els. The Mullen consists of five scales: visual reception (e.g., turns cup right-side up),
receptive language (e.g., recognizes own name), expressive language (e.g., produces three
consonant sounds), fine motor (e.g., uses pincer grasp while seated), and gross motor abil-
ities (e.g., pulls self to stand). The Mullen Early Learning Composite score is designed to
measure cognitive ability and is based on the first four scales. This is standardized for
children aged 0–69 months. The gross motor scale covers children from 0 to 29 months.

3. Results

3.1. Overall preference for speech

At 12 months, infants listened to speech (M = 12.0 s, SD = 5.2) significantly longer

overall than to non-speech (M = 9.6 s, SD = 3.9), t(26) = 2.29, p = .03, r = .41. A 2
(sound: speech, non-speech) by 5 (experimental block: 1, 2, 3, 4, 5) repeated-measures
analysis of variance (ANOVA) revealed a main effect of sound, F(1, 20) = 6.02, p = .023,
gp2 = .23, confirming longer listening to speech than non-speech, and a main effect of
block, F(4, 80) = 16.81, p < .001, gp2 = .46, consistent with infants’ looking time
decreasing over time. There was no reliable sound by block interaction, F(4, 80) = 1.77,
p = .14, gp2 = .08, suggesting that the difference between speech and non-speech listen-
ing times was stable across blocks. Twenty of 27 infants listened longer to speech than
non-speech (significant by binomial test, p = .02). This establishes that the speech prefer-
1680 A. Vouloumanos, S. Curtin / Cognitive Science 38 (2014)

ence observed in prior studies in younger infants, from newborns to 7-month-olds (Vou-
loumanos & Werker, 2004, 2007), persists until the end of the first year of life.

3.2. Variability and reliability of parental report and observational measures

MB-CDI scores varied considerably between individuals. MB-CDI expressive language

scores at 12 months ranged from 0 to 22 spoken words, with a mean of 7.8 words
(SD = 7.3), similar to data reported in prior studies (Fenson et al., 1994). These corre-
spond to MB-CDI expressive language percentile scores ranging from 1 to 87. MB-CDI
expressive language scores at 18 months ranged from 0 to 320 spoken words, with a
mean of 77.4 words (SD = 73.0). These correspond to MB-CDI expressive language per-
centile scores ranging from 1 to 97. Mullen Composite scores at 12 months varied
between 84 and 126 (M = 103.7, SD = 9.6), and at 18 months varied between 66 and
127 (M = 100.7, SD = 15.9). MB-CDI expressive language percentile scores and Mullen
Composite scores of general development were relatively stable across the two ages
(MB-CDI: r = .57, N = 20, p = .009; Mullen: r = .34, N = 25, p = .093), although lan-
guage measures seemed more stable than general development scores across this time
period. Finally, parental reports of expressive vocabulary on the MB-CDI correlated with
observational laboratory measures of expressive language using the Mullen at both ages
(12 months: r = .76, N = 26, p < .001; 18 months: r = .77, N = 22, p < .001).

3.3. Specific relationship between orienting to speech and language development

We found a specific predictive relationship between orienting to speech and expressive

language. We used multiple regression to test models with the following predictors of
MB-CDI scores at 18 months: (Step 1) orienting time to speech at 12 months; (Step 2)
Mullen Composite scores at 12 months; and (Step 3) MB-CDI scores at 12 months. Ori-
enting time to speech at 12 months was the only significant predictor of MB-CDI scores
at 18 months (see Table 1). When MB-CDI scores at 12 months were included as a pre-
dictor in the full model in Step 3, the effect size for orienting time to speech was mod-
estly attenuated. In addition, the significance level dropped to marginal (p < .10), which
might suggest that MB-CDI at 12 months was driving the effect. However, this seems
unlikely as MB-CDI at 12 months was itself not a significant contributor and its effect
size was much smaller. Orienting time to speech was consistently the largest relative pre-
dictor for MB-CDI scores at 18 months, and the only one that emerged significant at any
step. When we included orienting time to non-speech and overall looking time as covari-
ates in the model, the pattern of results remained the same with speech the only signifi-
cant predictor of MB-CDI at 18 months (these results are not presented). Neither
orienting time to non-speech nor overall looking time was a significant predictor.
We found no relationship between orienting to speech and general development at
18 months as measured by the Mullen Composite score. To test this relationship, we
regressed the same predictors [(Step 1) orienting time to speech at 12 months; (Step 2)
Mullen Composite scores at 12 months; and (Step 3) MB-CDI scores at 12 months] on
 A. Vouloumanos, S. Curtin / Cognitive Science 38 (2014) 1681

Table 1
Linear fixed effects model of predictor variables at 12 months on MB-CDI expressive language outcome at
18 months
B SE B b
Step 1
Constant 10.28 16.59
Orientation to speech 3.75 1.28 .59*
Step 2
Constant 111.26 64.38
Orientation to speech 3.25 1.26 .51*
Mullen composite 1.04 .64 .32
Step 3
Constant 97.26 64.53
Orientation to speech 2.56 1.37 .40†
Mullen composite .81 .66 .25
MB-CDI .35 .29 .27
Notes. R2 = .35 for Step 1, ΔR2 = .10 for Step 2, ΔR2 = .05 for Step 3.
*p < .05. †p < .10.

Mullen Composite scores at 18 months. The only significant predictor of Mullen Com-
posite scores at 18 months was the Mullen Composite score at 12 months. Orienting time
to speech was not a significant predictor (see Table 2). When we included orienting time
to non-speech and overall looking time as covariates in the model results were similar,
and neither orienting time to non-speech nor overall looking time was a significant pre-
dictor of Mullen Composite score at 18 months.

Table 2
Linear fixed effects model of predictor variables at 12 months on Mullen general development scores at
18 months
B SE B b
Step 1
Constant 85.23 10.50
Orientation to speech 1.34 .84 .35
Step 2
Constant 7.43 44.62
Orientation to speech .87 .83 .23
Mullen composite .82 .46 .39†
Step 3
Constant 14.13 45.24
Orientation to speech 1.41 .88 .37
Mullen composite 1.13 .49 .54*
MB-CDI .32 .21 .39
Notes. R2 = .12 for Step 1, ΔR2 = .14 for Step 2, ΔR2 = .09 for Step 3.
*p < .05, †p < .10.
1682 A. Vouloumanos, S. Curtin / Cognitive Science 38 (2014)

4. Discussion

Attention to speech predicts later language development. Infants who listened longer to
speech at 12 months scored higher on expressive vocabulary measures at 18 months than
infants with shorter listening times to speech. Notably, expressive vocabulary was not
reliably predicted by overall looking time, looking time to non-speech, or general indices
of development, speaking against the possibility that proficiency in expressive vocabulary
was due to infants being generally more attentive or being more advanced in their general
development. Instead, this finding is consistent with the proposal that an orienting mecha-
nism toward conspecific linguistic vocalizations provides a foundation for subsequent
language development.
Prior studies had shown a general predictive relationship between general auditory pro-
cessing and language abilities (e.g., Benasich & Tallal, 2002). General relationships also
exist between visual attention (habituation curves and subsequent recovery) in early
infancy and higher standard intelligence scores in childhood (e.g., Bornstein & Sigman,
1986; Colombo, 2001; Lewis & Brooks-Gunn, 1981; Slater, 1997). Here, we show a spe-
cific relationship between increased attention to a particular type of auditory stimulus,
speech, and later expressive vocabulary. These relatively specific effects are consistent
with other domain-specific individual differences, for example, within the domain of
number. Individual differences in the approximate number sense of 14-year-olds correlate
with their past scores of mathematical achievement at the age of 6 (Halberda, Mazzocco,
& Feigenson, 2008). This is consistent with specific relationships between potentially
ancient evolutionary abilities—such as conspecific vocalization biases, and the approxi-
mate number sense—and proficiency in formal systems such language and mathematics.
The current results complement and extend existing findings showing that specific
speech perception abilities (e.g., vowel discrimination, speech segmentation, use of pho-
netic detail in word learning) are related to language abilities (e.g., Bernhardt et al.,
2007; Newman et al., 2006; Tsao et al., 2004). Prior studies showed that specific phonetic
abilities (e.g., discriminating between vowels) correlated with aspects of language perfor-
mance (e.g., vocabulary size; Tsao et al., 2004), demonstrating that different language
processing tasks that tap into a common underlying linguistic ability are related across
development. This is consistent with findings suggesting that 17- to 20-month-olds’ abil-
ity to associate minimally distinct words to objects, which requires sensitivity to phono-
logical differences, might influence later word learning abilities (Bernhardt et al., 2007).
Here, we find evidence for a more general relationship between infants’ attention to a
class of relevant stimuli and their expressive language abilities. Our analyses show that
this relationship holds even when we include aspects of general development as measured
by the Mullen scales as a predictor. It is possible that the lack of correlation between
attention to speech and general development stems from lower variability in the general
development scores. However, the proposed specificity of the relationship between atten-
tion to speech and later language is consistent with previous work showing that speech
processing at 12 months correlates with expressive vocabulary at 24 months and language
skills at 4–6 years of age but not with cognitive skills (Newman et al., 2006). This
 A. Vouloumanos, S. Curtin / Cognitive Science 38 (2014) 1683

previous study suggests that speech processing does not predict later cognitive skills. Our
results suggest the converse may also be true: Earlier measures of general development
may not be strong predictors of later expressive language. Together these studies show
that language skills and general development have separable components.
Although orienting time to speech was the only significant predictor of expressive
vocabulary at 18 months, the effect only trended toward significance when we included
expressive vocabulary at 12 months as a predictor, because vocabulary measures were
highly correlated across ages. This final model is tightly controlled because by including
12-month expressive vocabulary as a predictor any remaining effect of orienting to
speech is an effect over and above any variable that influenced expressive vocabulary at
12 months. Thus, although the orienting to speech effect is no longer significant at
p < .05, the fact that it remains a large effect (b = .40) is very promising. It will be
important to replicate this effect in future research with a larger sample.
One open question concerns which properties of the speech stimulus draw infants’
attention. Speech is a complex, biological, and familiar sound for infants. Any one of
these properties might suffice to draw infants’ attention, conferring an advantage in later
language development. However, infants are selective in their listening preferences from
an early age, with 3-month-olds attending to speech more than to other complex, biologi-
cal, and familiar sounds such as human communicative vocalizations including laughter
(as well as monkey vocalizations or environmental sounds that share some subsets of
these properties; Shultz & Vouloumanos, 2010; Vouloumanos, Hauser, Werker, & Martin,
2010).
A second important unanswered question concerns the specific role that a bias for
speech plays in language acquisition. A speech bias may facilitate language development;
alternatively, a speech bias may be essential for language acquisition. Research with atyp-
ical populations may provide some insight into a possible facilitatory or foundational role
for a speech bias (Marcus & Rabagliati, 2006). Individuals with ASD who experience
language difficulties do not pay selective attention to speech sounds (Kuhl et al., 2005;
Paul, Chawarska, Fowler, Cicchetti, & Volkmar, 2007). In addition to displaying general
language delays (Baron-Cohen, 1995), word learning also appears to be different in ASD
individuals. Although children diagnosed with ASD apply exclusion constraints like
mutual exclusivity (de Marchena, Eigsti, Worek, Ono, & Snedeker, 2011), they do not
take advantage of eye gaze as a cue for disambiguating which object is being labeled,
and “weigh associative cues” more heavily (Baron-Cohen, Baldwin, & Crowson, 1997).
The absence of a behavioral bias for speech in individuals with ASD is correlated with
delays or impairments in language development (e.g., Kuhl et al., 2005), providing at
least some evidence for a fundamental role for speech biases in language development.
A predictive relationship between attending to speech at 12 months and vocabulary at
18 months does not entail a causal relationship. Speech and later vocabulary may be
linked through a range of possible relations. For example, infants who have learned that
speech is relevant may listen longer to speech and produce more words at 12 months,
and these infants then maintain their relatively high vocabulary 6 months later. Note that
under this hypothesis, orienting to speech would cease to be a significant predictor of
1684 A. Vouloumanos, S. Curtin / Cognitive Science 38 (2014)

expressive language at 18 months when expressive language at 12 months is a predictor

in the model; our finding that the predictive effect remains strong though not significant
leaves this possibility open, though we find it unlikely. At the same time, infants encoun-
ter speech in social communication contexts. Thus, another possibility is that infants’
ability to use the social context in which speech occurs influences later language develop-
ment. Future work needs to clarify the nature of the relationship between listening to
speech and later language development.
The demonstrated relationship between attending to speech and expressive language
abilities has important implications for our understanding of mechanisms of language
development and for assessing long-term language and social-communicative outcomes
(Curtin & Vouloumanos, 2013). Deviations in speech preference may be early indicators
of language delays and subsequent disorders.

Acknowledgments

This research was supported by NYU Research Funds (A. V.) and the Alberta Center
for Child, Family and Community Research (ACCFCR) (S. C.). We would like to thank
Mijke Rhemtulla for her expert statistical advice.

References

Baron-Cohen, S. (1995). Mindblindness: An essay on autism and theory of mind. Cambridge, MA: MIT
press.
Baron-Cohen, S., Baldwin, D. A., & Crowson, M. (1997). Do children with autism use the speaker’s
direction of gaze strategy to crack the code of language? Child Development, 68(1), 48–57.
Benasich, A. A., & Tallal, P. (2002). Infant discrimination of rapid auditory cues predicts later language
impairment. Behavioural Brain Research, 136, 31–49.
Bernhardt, B. M., Kemp, N., & Werker, J. F. (2007). Early word–object associations and later language
development. First Language, 27(4), 315–328.
Bornstein, M. H., & Sigman, M. D. (1986). Continuity in mental development from infancy. Child
Development, 57, 251–274.
Butterfield, E. C., & Siperstein, G. N. (1970). Influence of contingent auditory stimulation upon non-
nutritional suckle. In J. F. Bosma (Ed.), Third symposium on oral sensation and perception: The mouth of
the infant (pp. 313–334). Springfield, IL: Charles C. Thomas.
Cohen, L.B., Atkinson, D.J., & Chaput, H.H. (2004). Habit X: A new program for obtaining and organizing
data in infant perception and cognition studies (Version 1.0). Austin: University of Texas.
Colombo, J. (2001). The development of visual attention in infancy. Annual Review of Psychology, 52, 337–
367.
Colombo, J. A., & Bundy, R. S. (1981). A method for the measurement of infant auditory selectivity. Infant
Behavior & Development, 4(2), 219–223.
Conboy, B. T., Sommerville, J. A., & Kuhl, P. K. (2008). Cognitive control factors in speech perception at
11 months. Developmental Psychology, 44(5), 1505–1512.
Cooper, R. P., & Aslin, R. N. (1990). Preference for infant-directed speech in the first month after birth.
Child Development, 61, 1584–1595.
 A. Vouloumanos, S. Curtin / Cognitive Science 38 (2014) 1685

Cooper, R. P., & Aslin, R. N. (1994). Developmental differences in infant attention to the spectral properties
of infant-directed speech. Child Development, 65(6), 1663–1677.
Curtin, S., & Vouloumanos, A. (2013). Speech preference predicts autistic-like behavior in 18- months-olds
at risk for Autism Spectrum Disorder. Journal of Autism and Developmental Disorders, 43, 2114–2120.
Dawson, G., Meltzoff, A. N., Osterling, J., Rinaldi, J., & Brown, E. (1998). Children with autism fail to
orient to naturally occurring social stimuli. Journal of Autism and Developmental Disorders, 28(6), 479–
485.
DeCasper, A. J., & Fifer, W. P. (1980). Of human bonding: Newborns prefer their mothers’ voices. Science,
208(4448), 1174–1176.
Feldman, H. M., Dollaghan, C. A., Campbell, T. F., Kurs-Lasky, M., Janosky, J. E., & Paradise, J. L. (2000).
Measurement properties of the MacArthur communicative development inventories at ages one and two
years. Child Development, 71(2), 310–322.
Fenson, L., Dale, P. S., Reznick, J. S., Bates, E., Thal, D. J., & Pethick, S. J. (1994). Variability in early
communicative development. Monographs of the Society for Research in Child Development, 59(5), 1–
173; discussion 174–185.
Fenson, L., Marchman, V. A., Thal, D. J., Dale, P. S., Reznick, J. S., & Bates, E. (2007). MacArthur-Bates
Communicative Development Inventories: User’s guide and technical manual (2nd ed.). Baltimore, MD:
Paul H. Brookes Publishing Company.
Fernald, A., Perfors, A., & Marchman, V. A. (2006). Picking up speed in understanding: Speech processing
efficiency and vocabulary growth across the 2nd year. Developmental Psychology, 42(1), 98–116.
Halberda, J., Mazzocco, M. M., & Feigenson, L. (2008). Individual differences in non-verbal number acuity
correlate with math achievement. Nature, 455(7213), 665–668.
Klin, A. (1992). Listening preferences in regard to speech in four children with developmental disabilities.
Journal of Child Psychology and Psychiatry, 33(4), 763–769.
Kuhl, P. K., Coffey-Corina, S., Padden, D., & Dawson, G. (2005). Links between social and linguistic
processing of speech in preschool children with autism: Behavioral and electrophysiological measures.
Developmental Science, 8(1), F1–F12.
Lalonde, C. E., & Werker, J. F. (1995). Cognitive influences on cross-language speech perception in infancy.
Infant Behavior & Development, 18(4), 459–475.
Lewis, M., & Brooks-Gunn, J. (1981). Visual attention at three months as a predictor of cognitive
functioning at two years of age. Intelligence, 5(2), 131–140.
Mandel, D. R., Jusczyk, P. W., & Pisoni, D. B. (1995). Infants’ recognition of the sound patterns of their
own names. Psychological Science, 6, 315–318.
de Marchena, A., Eigsti, I. M., Worek, A., Ono, K. E., & Snedeker, J. (2011). Mutual exclusivity in autism
spectrum disorders: Testing the pragmatic hypothesis. Cognition, 119(1), 96–113.
Marchman, V. A., & Fernald, A. (2008). Speed of word recognition and vocabulary knowledge in infancy
predict cognitive and language outcomes in later childhood. Developmental Science, 11(3), F9–F16.
Marcus, G., & Rabagliati, H. (2006). What developmental disorders can tell us about the nature and origins
of language. Nature Neuroscience, 9(10), 1226–1229.
Marler, P., & Peters, S. (1977). Selective vocal learning in a sparrow. Science, 198(4316), 519–521.
Molfese, D. L. (2000). Predicting dyslexia at 8 years of age using neonatal brain responses. Brain and
Language, 72(3), 238–245.
Mullen, E. (1995). Mullen scales of early learning. Circle Pines, MN: American Guidance.
Newman, R., Ratner, N. B., Jusczyk, A. M., Jusczyk, P. W., & Dow, K. A. (2006). Infants’ early ability to
segment the conversational speech signal predicts later language development: A retrospective analysis.
Developmental Psychology, 42(4), 643–655.
Paul, R., Chawarska, K., Fowler, C., Cicchetti, D., & Volkmar, F. (2007). “Listen my children and you shall
hear”: Auditory preferences in toddlers with autism spectrum disorders. Journal of Speech, Language, and
Hearing Research, 50(5), 1350–1364.
1686 A. Vouloumanos, S. Curtin / Cognitive Science 38 (2014)

Remez, R. E., Rubin, P. E., Berns, S. M., Pardo, J. S., & Lang, J. M. (1994). On the perceptual organization
of speech. Psychological Review, 101(1), 129–156.
Remez, R. E., Rubin, P. E., Pisoni, D. B., & Carrell, T. D. (1981). Speech perception without traditional
speech cues. Science, 212(4497), 947–949.
Samples, J. M., & Franklin, B. (1978). Behavioral responses in 7 to 9 month old infants to speech and non-
speech stimuli. Journal of Auditory Research, 18(2), 115–123.
Shultz, S., & Vouloumanos, A. (2010). Three-month-olds prefer speech to other naturally occurring signals.
Language Learning and Development, 6(4), 241–257.
Slater, A. (1997). Can measures of infant habituation predict later intellectual ability? Archives of Disease in
Childhood, 77(6), 474–476.
Spence, M. J., & DeCasper, A. J. (1987). Prenatal experience with low-frequency maternal-voice sounds
influence neonatal perception of maternal voice samples. Infant Behavior & Development, 10(2), 133–142.
Stager, C.L., & Werker, J.F. (1997). Infants listen for more phonetic detail in speech perception than in word
learning tasks. Nature, 388, 381–382.
Tsao, F.-M., Liu, H.-M., & Kuhl, P. K. (2004). Speech perception in infancy predicts language development
in the second year of life: A longitudinal study. Child Development, 75(4), 1067–1084.
Vouloumanos, A., Hauser, M. D., Werker, J. F., & Martin, A. (2010). The tuning of human neonates’
preference for speech. Child Development, 81, 517–527.
Vouloumanos, A., Kiehl, K. A., Werker, J. F., & Liddle, P. F. (2001). Detection of sounds in the auditory
stream: Event-related fMRI evidence for differential activation to speech and nonspeech. Journal of
Cognitive Neuroscience, 13(7), 994–1005.
Vouloumanos, A., & Werker, J. F. (2004). Tuned to the signal: The privileged status of speech for young
infants. Developmental Science, 7(3), 270–276.
Vouloumanos, A., & Werker, J. F. (2007). Listening to language at birth: Evidence for a bias for speech in
neonates. Developmental Science, 10(2), 159–164.
Werker, J. F., & Curtin, S. (2005). PRIMIR: A developmental framework of infant speech processing.
Language Learning and Development, 1(2), 197–234.
Ziegler, J. C., Pech-Georgel, C., George, F., Alario, F. X., & Lorenzi, C. (2005). Deficits in speech
perception predict language learning impairment. Proceedings of the National Academy of Sciences USA,
102(39), 14110–14115.
Copyright of Cognitive Science is the property of Wiley-Blackwell and its content may not be
copied or emailed to multiple sites or posted to a listserv without the copyright holder's
express written permission. However, users may print, download, or email articles for
individual use.