ArticlePDF Available

What studies of communication with infants ask us about psychology: Baby-talk and other speech registers.


Reviews the research concerning the possible functional roles of the prosodic aspects of infant-directed speech ("baby-talk"), and shows that this style of speaking could have both attentional and affective functions (in addition to linguistic ones). Studies of other speech registers could profit from taking similar approaches and even using similar techniques to those used in infant-directed speech research to tease apart the linguistic, attentional, and affective components of these other speaking styles. Nursing-home register is used to exemplify the importance of examining the social, emotional, and linguistic ramifications of a communication style separately. (PsycINFO Database Record (c) 2012 APA, all rights reserved)
What Studies of Communication with Infants Ask Us
About Psychology: Baby-talk and Other Speech Registers
Acadia University
This article begins wilh a brief review of
research concerning ihe possible functional
roles of the prosodic aspects of infant-directed
speech ("baby-talk"), showing that this style of
speaking could have both attentional and
affective functions (in addition to linguistic
Tt is then argued that studies of other
speech registers could profit from taking
similar approaches and even using similar
techniques to those used in infant-directed
speech research to tease apart the linguistic,
attentional and affective components of these
other speaking styles. Nursing-home register is
used to exemplify the importance of examin-
ing the social, emotional and linguistic ramifi-
cations of a communication style separately.
Studies of the special speech register we use
with infants both inform us about psychology
in general and raise questions about other
communicative behaviours that might not
have been asked if it were not for these
developmental studies. Since knowing what
questions to ask is often the keystone to
important research, this is an important
contribution developmental psychology can
make to psychology. The methods used in
infant-directed speech studies also suggest
how questions about other styles of verbal
communication might best be answered.
Baby Register
One of the more interesting features of inter-
actions with babies is the particular way in
which people modify their speech when
Canadian Psychology/Psychologie canadienne, 34:3
addressing infants. This type of speech,
which is usually referred to as "baby-talk" or
"motherese", differs from normal adult-
directed talk in many ways (in the order of
100 documented characteristics, Brown,
1986) including having simplified syntax,
shorter utterances, more questions, more
repetition, and special prosodic features.
Since not only mothers but all adults, and
even preschool children have been shown to
use this specific style of speech when address-
ing infants, it is more appropriately referred
to as "infant-directed speech" or "infant-
directed talk" (Werker & Mclx-od, 1989).
The characteristic prosodic features of
infant-directed talk include higher pitch,
exaggerated pitch modulation, elongated
vowels, longer pauses and increased
rhythmicity: What Darwin (1877) called the
"sweet music of the species". These character-
istic modifications arc found in a wide range
of diverse languages such as Japanese,
French, Italian, German, Mandarin and
English (e.g., Fernald et
1989); a fact that
has led people to speculate that infant-
directed talk prosody may be an important,
functional aspect of the infant's social envi-
Three categories of functions have been
suggested for infant-directed talk prosody. 1)
Linguistic: These modifications simplify and
highlight relevant linguistic components of
speech. 2) Attentional: The modifications
may be effective in gaining and maintaining
infants' attention. 3) Affective: The modifica-
tions may contribute to positive affective
interactions between parents and infants.
Note that these are not mutually exclusive
I Fernald (1032a) has recently gone further and
developed arguments about (he evolution of ihis
speech register based on the functions it lias been
shown to serve.
Speech Registers 283
To examine the evidence for the latter two
possible functions for infant-directed talk,
Janet Werker and I did several experiments
suggest ways in which speech registers
more generally might be studied. Our lirst
simple experiment was designed to deter-
mine if infants prefer to attend to infant-
directed over adult-directed talk when the
speaker is a male. Prior to our work, four-
month old in fan
had been shown to prefer-
entially attend to infant-directed over adult-
directed talk when spoken by a female (e.g.,
Fernald, 1985), however there were no simi-
lar studies published using male speakers.
The stimuli used in this first study were
audio-video recordings of one actor and one
actress reciting an identical script to both a
six-month old (in infant-directed prosody),
and to an adult (in adult-directed prosody).
By having an identical script, we controlled
for many linguistic differences between
conditions, and by using video tapes we
ensured that subjects don't influence the
speaker and that all subjects within a treat-
ment condition get (lie same stimuli.
The method we used lo ask infants which
stimuli they prefer involved showing them
the recorded stimuli in a sequeniial preferen-
tial looking procedure while they were held
facing a monitor by a parent. Infants in (his
paradigm were presented with alternating
stimuli for fixed periods of time and the
amount of time they attended to each was
measured by an observer looking through a
one-way mirror. This measure of looking
expressed as the percentage of tinie
infants looked at the stimuli, was our oper-
ational definition of attcntional effectiveness.
We found that infant-directed talk was
watched for significantly longer periods of
time than was adult-directed talk. There was
no significant interaction between talk-type
and speaker, demonstrating quite clearly that
infants will attend to infant-directed lalk
more than adult-directed talk for both male
and female speakers.
In our second experiment we began to
address the question of whether infants
exhibit differential affective responsiveness
to infant-directed and adult-directed talk and
assessed age related changes in responsive-
ness to infant-directed talk. To do this,
infants of 4 - 5.5 and 7.5 - 9 months were
compared on both their attentional and
affective responsiveness to male and female
infant- and adult-directed talk using the same
sequential preferential looking paradigm.
As shown in Figure 1, the auentiorial
responsiveness data indicated that younger
infants looked longer overall than did older
infants and both age groups looked longer
at infant-directed talk than at adult-directed
talk. Finding such an attentional bias has
been interpreted as evidence for affective
involvement as
however, selective atten-
tion does not necessarily indicate an affective
preference. For example, infants may choose-
to attend to a stimulus because of its inten-
sity, complexity, novelty, or a number of
factors theoretically independent of positive
valence (sec Pcgg, Wcrkcr
McLeod, 1992).
Infants may even stare at potentially frighten-
ing (i.e., looming) stimuli under some condi-
tions (see Yonas, 1981).
To assess the possible affective role of
infant-directed speech more direcdy, we had
trained coders—unaware of the recording
conditions—rate the video recordings of
each infant made during the experiment as
the infants were exposed to the different
speech styles. Subjective ratings were made
of each infant in each condition on the basis
of judgement of how much the infant was
trying lo interact, how interested the infant
appeared lo be, and the infant's emotional
These highly correlated ratings were
combined lo obtain a single score oi affective
responsiveness for each infant in each condi-
tion such that infants judged to be highly
interactive, interested, and experiencing
positive emotions received high
percent looking time is likely to contribute
to these ratings, these scores more direcdy
measure emotional responsiveness than does
looking time alone.
Our results showed that infants in the
younger age group were judged more affect-
284 McLeod
Infant-Directed Speech
Adult-Directed Speech
4-5 Months7-9 Months
Fig. I Attentional responsiveness, upcralionalizcd as percent looking lime, to iiiiam-
dire<:tcd and adult-directed talk per age group lo both tin: male and female speakers
(from Wcrkcr & McLcod, 19H9).
ively responsive than were infants in the
older age group. Infants were also judged
more responsive to the actress than to the
actor. Most importantly, infants showed more
affective response to the infant-directed than
to the adult-directed stimuli. As suggested in
Figure 2, there was also a two-way interaction
between talker (the actor or the actress) and
talk-type with the adult-directed versus infant-
directed speech difference being significant
only for the female stimuli. Together these
data show that the prosodic characteristics of
infant-directed speech register can have both
aiientional and affective roles and these
effects may be somewhat independent.
As social "interactions" and communication
can only be understood by examining all
those involved, Janet Wcrker and
also asked
what reciprocal roles infants' differential
responsiveness to infant-directed talk might
have on the adults with whom they are inter-
acting. The results just summarized demon-
strated that trained adults could discriminate
between infants who were exposed to infant-
directed talk or adult-directed talk on the
basis of the infant's behaviour, but these data
did not address the much more intriguing
possibility that naive adults might show a
preference for infants' responses to infant-
directed versus adult-directed talk.
To test this, we had undergraduate sub-
jects watch one-minute video segments of a
sample of different infants from
age group recorded during the previous ex-
Speech Registers 285
Infant-Directed Speech
Adult-Directed Speech
4-5 Months7-9 Months
Fig. 2 Mean affective responsiveness, operationalized
average rating score, to iniant-
directcd and aduU-dircclcd talk per age group to both tin: male and female speakers
(from Wcrker & McLcod, 1989).
while they were watching the male
or the female reciting the script in adult-
direcied and infant-directed prosody. After
watching each video segment subjects filled
out a rating form, developed by Kathleen
Bloom (1985), asking their impressions of
the infant they had just seen on the five-
dimensions of: pleasant, friendly, fun, like-
and cuddly.
The pattern of results was very .similar to
those from the previous experiment. Subjects
showed strong preferences for babies
recorded while viewing the actress in com-
parison to recordings of the same babies
made while viewing the actor, and for babies
recorded viewing infant-directed talk over
recordings of the same babies made while
viewing adult-directed talk. Again there was
also an interaction between talker and talk-
type with greater differences in the judge-
ments of infants viewing the female's infant-
versus adult-directed talk than between
judgements of infants watching the male's
two speech styles. These results indicate that
naive adults judge babies who are exposed
to some infant-directed talk (e.g., that of our
actress) as friendlier, more cuddly, etc.
Infants are therefore not only more respon-
sive to infant-directed talk but may also be
more likely to display behaviours which
adults find appealing when they are spoken
to in this way. Infant-directed talk could
thereby initiate a sequence of events between
adults and infants wherein each becomes
more attractive to the partner. Such positive
feedback could conceivably help maintain
both participants' attention and interest in
the interaction, as well as facilitate attach-
ment through the more positive emotions
thai both experience.
It is important to realize that infant-
directed talk is normally accompanied by
exaggerated facial expressions as well as
exaggerated prosody. Both our actor and
actress displayed more smiling, eyebrow
arching and rhythmical head movements
during their infant-directed speech than in
the adult-directed condition. It was import-
286 Mcl-eod
ant, therefore, for us to assess the relative
contribution of the facial and vocal features
to infants' differential responsiveness to these
stimuli. We did this by testing 4 - 5.5 month
old infants over a two day period with both
the actress's infant-directed and adult-
directed facial patterns paired with identical
"musak" tracks, and her infant-directed and
adult-directed vocal tracks paired wiih an
identical visual segment of her nodding with
a neutral facial expression.
There were no attentional differences in
the resulting looking time data from this
experiment. The results of affective ratings,
however, showed greater overall responsive-
ness to the infant-directed stimuli than to the
adult-directed stimuli. When examined separ-
ately, we found a significant preference for
the vocal features of the infant-directed over
die adult-directed stimuli, but not for die
facial features of die infant-directed stimulus
relative to die adult-directed stimulus, sug-
gesting diat the vocaJ features may play a
more important role than the facial features
in the preference infants exhibit for die
infant-directed talk of our actress.
To summarize these studies, Werker and
McLeod (1989) demonstrated diat infants of
- 5.5 and 7.5 - 9 months of age are more
responsive to infant-directed talk than to
adult-directed talk by both attentional and
affective measures. We also demonstrated
that younger infants are more responsive
than older infants. Furthermore, tliis "prefer-
held true for bodi male and female
speakers and the vocal features alone were
sufficient to get these findings with our
female stimuli. Perhaps most interestingly,
infants exposed to some infant-directed talk
were judged to be more "attractive" by naive
In what has turned out to be an excellent
example of the "homotypic" versus
"heterotypic" continuity problem laced by
many developmental psychologists, a third
group of infants, under two-months of age,
not only did not show a preference for
infant-directed talk but showed no indication
that dicy even discriminated between the two
types of speech (Pegg ct al., 1992). If one
believes that neonales arc predisposed to
prefer the prosodic characteristics of infant-
directed talk, these data are problem-
atic—especially since Cooper and Aslin
(1990) found infants less than one-month of
age to prefer infant-directed speech. If prior
to one-mondi and after four-mondis infants
prefer this speech register, why didn't, our
A clue to die answer came from the fact that
our original procedure used a fixed duration
sequential preferential looking task that
while appropriate for four to ninc-month-
did not appear to be sensitive to the
abilities of three to eight-wcck-old subjects.
We had to reformulate the question in a way
that the younger infants could answer.
Group-specific operational definitions are
often necessary in psychology, especially
when groups differ greatly in abilities (as is
typically the case in developmental studies).
Since I wanl to make the case that the
research I am summarizing here could profit-
ably be transferred to other social situations,
1 will now briefly discuss the paradigm we
used with younger subjects.
Judith Pegg, Janet Werker and I redid this
experiment with seven to nine-week-olds
using an infant controlled habituation-
dishiibi I nation procedure in which the
infants' looking lime determines the length
of each trial (Pegg et al., 1992). In this more
sensitive procedure, when the infants fixated
on a light above the monitor, they were then
presented with infant-directed or adult-
directed talk stimuli paired with a
checkerboard display for as long as they
looked at the monitor. Repeated presenta-
tions were given until the infants habituated,
at which time two dishabituation trials widi
the oiher stimulus were presented.
Briefly, we found differential recovery
between the control group and the experi-
mental groups, demonstrating that infants of
seven weeks of age are able to discriminate
between infant-directed and adult-directed
Speech Registers 287
speech styles when delivered by the same
speaker using this paradigm.
Since we were also interested in whether
or not there was a preference for infant-
directed over adult-directed speech (not just
discrimination), we also analyzed whether the
amount of recovery in looking time was
greater in some conditions than in others
and whether infant-directed speech was
looked at longer than adult-directed speech
when presented during the first habituation
With the female speaker groups, these
analyses revealed that when infants were
presented with female adult-directed speech
during the habituation phase of the experi-
ment and female infant-directed speech
during dishabiluation, dishabituation was
markedly greater than when the female-
speaker's stimuli were presented in the
reverse order.2 When the initial habituation
trials were examined, infants looked longer
when presented with infant-directed than
with adult-directed speech for both male and
female speakers.
Before relating these findings to other
areas of psychology, I will summarize another
set of studies on infant-directed speech that
lakes a different approach to distinguishing
between the effects of the "universal"
prosodic features of this speech register and
the specific linguistic aspects of an infant's
environment, in determining additional and
affective responsiveness.
Cross-language studies of baby-register The
simple idea behind these studies was that if
infants' preference for infant-directed talk is
due primarily to universal prosodic modifica-
young infants should prefer infant-
directed talk even in an unfamiliar language.
To examine this, we studied 4.5 and 9 month
old English- and Cantonese-learning infani.s
tested with stimuli from both English and
Cantonese mothers (recorded while they
2 The results with our malt; speaker's stimuli were
different in that the mean increase in looking time
during dishabidiation was not significant in either
of the male speaker experimental conditions.
were addressing their own infants at the
same ages as the subjects).
One reason why
used Cantonese as our
second language was that it is a tone-lan-
guage wherein the same sound can have
different meanings depending on the
pitch-contour of the voice3. The modifica-
tions made when addressing infants can
sometimes involve sacrificing semantic infor-
mation for the sake of maintaining infant-
directed talk prosody; just
semantic aspects
of American Sign language are sometimes
sacrificed for the sake of positive affect when
signing to infants (Rcilly, 1990). By selecting
a non-European language it
also felt that
this would provide a more rigorous test for
cross-language preferences for infant-directed
While preliminary analyses of the attcn-
tionai preference results of these experi-
ments included order effects and dilfcrcnccs
that may be related to differences in the
acoustic characteristics of the two mothers'
voices, they nevertheless provide good evi-
dence that English-learning infants attended
to infant-directed speech significantly longer
than to adult-directed speech when delivered
in Cantonese. Cantonese infants also pre-
ferred Cantonese infant-directed speech,
however, neither they nor the English infants
looked longer to the English speaker's infant-
directed speech than to her adult-directed
In other cross-language studies using
seven-week-old infant subjects with different
stimuli we have also found attentional prefer-
ences for infant-directed over adull-dircctcd
speech in the unfamiliar language (Pegg el
1992). In related research, Fernald and
Kuhl (1987) have shown the attentional
preference for infant-directed speech to be
present even when the speech samples were
low-pass filtered, thereby removing linguistic
information. Together these studies show
quite conclusively that the universal prosodic
aspects of this speech register function to
3 Cantonese can have as many as nine different
tonal categories.
288 McLeod
maintain infants' attention and this effect
appears independent of any linguistic func-
tion infant-directed speech serves.4
When affective responsiveness was the
dependent variable of interest with older
infants, English subjects showed a preference
for infant-directed stimuli when presented in
either English or Cantonese subjects, and the
Cantonese-learning infants showed an
affective preference for infant-directed over
adult-directed stimuli in Cantonese but not
in English (although again order effects
complicated the picture somewhat).
Although beyond the scope of the present
paper, it is possible that this measure may be
more sensitive to the effects of experience
than attentional preference even though the
prosodic aspects of speech associated with
communicating emotion are also similar
across cultures and languages (Frick, 1985).
Regardless, the different results for
alientional versus affective measures high-
lights the importance of separating these
Age effects in these cross-language experi-
ments also suggest the interesting possibility
that infant-directed speech registers of vari-
ous languages may be more similar
prosodically than adult-directed registers. Do
our language environments diverge develop-
mentally in ways that are not just correlates
of increasing linguistic complexity? More
generally, arc cultural differences in adult
social behaviour greater than cross-cultural
differences in adult-infant behavior?
Although these issues are beyond the focus
of the present discussion, these are questions
for social scientists and linguists that arise
directly from research in developmental
While support for the hypothesis that
infant-directed speech would be preferred
even in an unfamiliar langauge was equivo-
'I Kcrnald (19921)) briefly mentions related findings
that suggest scolding and praising in infant-
dirt'trtcd but not adult-dirciMt'd prosody results m
appropriate affective responses with some (hut not
all) unfamiliar languages.
the main point to be made here from all
these studies of "baby-register" is that
although vocal signals are very com-
plex—containing not only linguistic informa-
tion but also many other messages conveyed
in the manner in which the speech is
possible to isolate the linguis-
tic aspects of speaking style from these other
messages. Furthermore, it is possible to
isolate the possible functional roles that these
embedded messages may play.
Other Speech Registers
Infant-directed speech was the first, and is
the most thoroughly, studied speech register.
Among the others recognized by socio-lin-
guisl.s are: foreigner-talk, "doggerel", nursing-
hoine-talk, sick-room-talk (hospital-talk), and
sporls-announcer-talk. On some dimensions
there are similarities among registers. For
example, it has been suggested that the
special registers used with babies and
foreigners are almost identical in terms of
linguistic modifications (restricted vocabu-
lary, frequent pauses, etc.), although
prosodic differences do exist, (e.g., foreigner-
talk is often loud, baby-talk is often whis-
pered) (Freed, 1981). It has also been
hypothesized that the basis for the similar
modifications is that both babies and
foreigners are "linguistically insufficient".
(Infants, however, are also cognitively limited
and of different social status, whereas
foreigners presumably are neither.) The
similar linguistic modifications observed in
baby-talk and "doggerel" arc also quite
remarkable. While there may be some
communicatively relevant similarities between
speaking to very young children and dogs,
the existence of similarities between doggerel
and baby-talk has been used to question any
"linguistic teaching" hypothesis for the func-
tion of these registers (Hirsh-Pasek &
Treiman, 1981).
Natural communicative signals are typically
multidimensional. Some dimensions differ
in terms of their controllability and there-
fore what they are best suited to convey.
The components of infant-directed speech
Speech Registers 289
verbal - linguistic
facial expressions
body, hands, legs, feet
[speech prosody?]
vocal quality
Fig. 3 Ekman's and Fressen's (1969) controllability-leakage hierarchy of com-
municative channels (adopted from Brown, 1986) with a suggested location for
speech prosody.
(and the other registers identified) that have
heen examined fall at different points along
the controllability-leakage hierarchy of com-
municative channels proposed by Ekman
and Friescn (1969). As shown in Figure 3,
linguistic information is most controllable
and vocal quality information least, with
non-vocal signals intermediate. Given this,
social register use involving both linguistic
and vocal quality (or at least prosody) alter-
ations, might be expected to have both
linguistic and emotional or personal effects.
If the prosodic modifications made when
addressing an infant are intermediate in
their controllability (as I have suggested in
Figure 3) this would fit with the Werker and
McLeod (1989) findings that the vocal fea-
tures of talk alone were sufficient to get
greater affective responsiveness to infant-
directed talk over adult-directed talk while,
facial features along were not sufficient
in-so-far-as infants were emotionally more
responsive to the channel that was more
likely to convey accurate emotional
The linguistic characteristics of infant-
directed talk have been well studied, and
considerable research suggests that the modi-
fications made in talk to infants relative to
talk to adults, could function to facilitate
language acquisition (although there is
considerable controversy in this area, for
example see Gleitman, Newport, & Gleitman,
Several studies have shown that the
segmentation of speech sounds into sen-
tences, clauses and phrases is facilitated by
infant-directed speech (Hirsh-Pasek el al.,
Dexter, 1988), perhaps in pan
due to the melodic structure of infant-
directed speech (as suggested by Trehuh &
Trainor, 1990). Six-month-olds can also more
easily discriminate medial position syllables
in trisyllabic utterances when infant-directed
speech stress patterns are used (Karzon,
In light of the data now available on
infant-directed talk it
counterproductive to
ask if infant-directed talk serves a linguistic
or affective or alien tional role, as if these are
mutually exclusive. Studies of the effects of
other speech registers, however, have been
290 McLeod
more narrowly focussed
their linguistic
The factors identified
that influence communicative register have
been categorized
in a
simplest being:
mode (written
versus verbal versus drums, etc.);
2) ihe field,
which refers
to the
subject matter
situational context
of the
3) the
social relation between
communicants. Tenor
can be
ized along dimensions
of: 1)
3) formality (intimacy vs.
prosodic aspects
directed talk
have studied fall primarily
under this third category
The infant-directed talk studies that I have
summarized demonstrate that
it is
to examine
different speech styles,
their linguistic ones.
is it
important that
all the
ible effects
the alterations
speech made
in social registers (although
methods by which we
diiswill undoubted-
ly vary among
populations studied).
The final suggestion
future research
make comes from
finding that
infants being spoken
to in
ways that make them more
It is
therefore important
only examine
communication styles
on the
these signals
also examine
the reciprocal effects these have
on the
of the
message. Just
Goldberg (this volume) reminds
attachment is
reciprocal process, so
are the
various speech
communicative styles
find that people
to use
special registers
reflect social roles
only perceive
their speech
particular ways
make assumptions about
intimacy, power,
etc. but
they may also elicit,
in the recipients of the speech, changes that are
consistent with these expectations. We want to
know: 1) What is in the mind of the
2) How does using that style of
communication influence the thinking and
behaviour of the recipient?; and 3) How do
these influences, in turn, alter the percep-
tions of both the recipient and sender of that
style of communication?
A classic study showing such self-fulfilling
prophecies, done by Rosenthal andJacohson
examined the effects of instilling false
expectations in teachers about the future
performance of a randomly selected group
of elementary school students. Teachers were
told that approximately one-third of their
students were expected to show spurts in
performance during the school year.
Although students were randomly designated
as "spurters", this group showed significant
increases in their IQ scores, relative to their
control classmates over the school year.
A more dramatic example of how differen-
tial treatment of someone might influence
others not involved in the interaction comes
from the clever experiment by Snyder, Tanke
and Berscheid (19V7) in which male univer-
sity students conversed with an unfamiliar
female by phone. The male subjects were
previously given photographs they were told
were of the person with whom they were
going to converse. Subjects who thought they
were talking to an attractive female (based
on independent ratings of the photograph)
rated her more favourable alter their conver-
sation than when these male subjects thought
she was unattractive. Most, importantly, inde-
pendent subjects listening only to the female half
of these conversations and without photographs
also judged the woman whose conversational
partner thought was attractive more favourably.
The male subjects
thought they were
to an
attractive woman spoke
a way that
to the
sound more attractive
One area where adoption
of a
register leading
self-fulfilling prophesies
be a
with nursing-home:
In one
study, Caporael
Culbertson (1986) found that
Speech Registers 291
of ihe speech directed to more depend-
ent nursing-home residents was classified by
naive raters as being in baby-talk prosody,
whereas more independent residents were
not spoken to in this way—suggesting that
dependency is an important component of
nursing-home register usage. The concern is
that this communicative style could instill
dependency in die elderly to whom it is
directed, however, to my knowledge tfiis has
yet lo be tested directly. To do so will require
careful control over linguistic (semantic)
aspects of communication (perhaps by using
video-taped segments of identical script
delivered in different registers) and will
require multiple dependent measures to
evaluate die various suggested effects.
Infants are by definition without language
(the very term coming from die Latin infant
meaning incapable of speech). To study
these "animals" researchers must cidier use
elhological approaches or come up with
indirect methodologies (such as those I have
mentioned above) to "ask" the questions that
interest us. With verbal subjects in nursing-
homes this may not seem necessary. Certain-
ly, questionnaires administered to residents
could provide excellent measures of the
various effects nursing-home register may
serve, however, more elhological measures,
such as independent, ratings of aiienlivcncss
to the speaker (e.g., eye contact) and facial
affect could provide additional insight. Simi-
larly, video-tapes of various messages (per-
haps instructions on taking medication)
delivered by staff could he used to compare
the instructional and affective effectiveness
(assessed in pan by naive coders observing
the participants) of nursing-home and nor-
mal adult registers.
I am not suggesting the adoption of this
style of communication is a "bad" thing. In
a study by Cohen and Faulkner (1986) for
example, it was shown that the stress pattern
of "elderspeak" facilitates comprehension.
But we can and should separate the various
specific effects of the components of these
registers. The medical register cases are
particularly interesting in this regard because
linguistically the speaker has to avoid pro-
fessional jargon or risk not being understood
while also avoiding talking down to the
Although in theory adopting a counter-
register could have positive effects, the rela-
tive involuntary nature of the cues that trans-
mil the affective messages of speech may
make this impossible to realize without alter-
ing the thoughts and feelings of the social
register user. Such effects are possible (e.g.,
in the Rosenlhal & Jacobson study discussed
above) but the mechanisms by which these
effects are manifest are not well understood.
Communicative style is undoubtedly an
important component.
I would like lo thank Tara for inviting me to
participate in thai symposium and
would like
lo lhank Kathleen Bloom for her helpful com-
ments. I would also like to acknowledge the
crucial role of my collaborators in the baby-talk
studies, Janel Werkcr and Judith Pc-gg. I thank
them for this and the valuable comments they
made on an earlier draft of this manuscript
Preparation of this manuscript was supported
by an
grant Reprints arc available from
Dr. Peter McLeod, Infant and Family Research
Centre, Department of Psychology, Acadia
University, Wolfville, Nova Scotia, BOPlXO.
Bloom, K. (1985). Ruling scale developed lo
assess parent ratings of
manuscript (Copyright).
Brown, R. (1986). Social Psychology. The Sec-
ond Edition. New York: The Free Press.
Caporacl, L.R. (1981). The paralanguage of
care-giving: Baby talk to the institutionally
aged. Journal of Personality and Socitil Psy-
chology, 40(5), 87fi-884.
Caporael, L.R., & Culbertson, Gil. (1986).
Verbal response modes of baby talk and
other speech at institutions for the aged.
Language and Communication, 6, 99-112.
Cohen, G., & Faulkner, 1). (1986). Does
292 Mcl-eod
"cldcrspeak" work? The effect of intonation
and stress on comprehension and recall of
spoken discourse in old age.
6, 91-98.
Cooper, R.P.,
Aslin, R.N. (1990). Preference
for infant-directed speech in the first
month after birth.
Darwin, C.R- (1877). A biographical sketch of
an infant.
2, 286-294.
Ekman, P., & Friesen, W.B. (1969). The reper-
toire of nonverbal behavior: Categories, ori-
usage, and coding.
I, 49-98.
Femald, A. (1985). Four-month-old infants
prefer to listen to "molherese".
8, 181-195.
Fernald, A. (1992a). Hviman maternal vocal-
izations to infants as biologically relevant
signals: An evolutionary perspective. In, J.H.
Rarkow, L. Cosmides,
Tooby (Eds.) The
Evolutionary psychology and
(pp. 391-428).
Oxford: Oxford University Press.
Fernald, A. (1992b). Meaningful melodies in
mothers' speech to infants. In, H.
Papousek, U. Jurgens,
Comparative and
(pp. 262-282). Cam-
bridge: Cambridge University Press.
Fernald, A., & Kuhl, PK. (1987). Acoustic
determinants of infant preference for
mothcrcse speech.
Infant Behavior and
Fernald, A., Taeschner, T., Dunn, J.,
Papousek, M., Boysson-Bardies, R., & Fukui,
I. (1989). A cross-language study of
prosodic modifications in mothers' and
fathers' speech to preverbal infants.
Freed, B.F. (1981). Foreigner bilk, baby talk,
native talk.
International Journal
28, 19-39.
Frick, R.W. (1985). Communicating emotions:
The role of prosodic features.
97(3), 412-429.
Gleitman, L.R., Newport, E.I.., & Gleitman, fI.
The current status of the mothcrese
II, 43-79.
Hirsh-Pasek, K.,
Trciman, R. (1981). Dog-
gerel: Mothcrese in a new context.
9, 229-237.
Hirsh-Pasck, K., Nelson,
Jusczyk, P.W.,
Cassidy, K., Dmss, B., & Kennedy, L.
Clauses are perceptual units for
young infants.
26, 268-286.
Karzon, G. (1985). Discrimination of polysyl-
labic sequences by one-to-four month old
Experimental Child
39, 326-342.
Miller, J., & Dexter, E.R. (1988). Effect of
speaking rate and lexical status on phonetic
Human Perception
J.E., Wcrker,
& McLeod, P.J.
Preference for infant-directed over
adult-directed speech: Evidence from
7-week-oId infants.
Infant Behavior and
(1990). Child-directed language in
Taking expressions at face value.
Behaviour and
51 (Abstract).
Rosenlhal, R-, & Jacobson, I,. (1968).
York: Holt, Reinhart
Snyder, M., Tankc, ED, & Berscheid, E.
Social perception and interpersonal
behavior: On the self-fulfilling nature of
social stereotypes.
Personality and
35, 656-666.
Trehub, S.E-, & Trainor, I. J. (1990). Rules for
listening in infancy. In J.T. Enns (Ed.), The
87-119). North-Holland: Elsevier
Science Publishers.
Worker, J.F., &MtI.eod, P.J. (1989). Infant
erence for male and female infant-directed
developmental study of altentional
and affective responsiveness.
43(2), 230-246.
Yonas, A. (1981). Infants' responses to optical
information for collision. In R.N. Aslin, J.R.
Alberts, & M.R. Patterson (Eds.),
v2. The
visual system
(pp. 313-332).
New York: Academic Press.
... As hypothesized, IDS may function developmentally to communicate affect, regulate infants' arousal and attention, and facilitate speech perception and language comprehension [16,82]. ...
Full-text available
Various aspects of motherese also known as infant-directed speech (IDS) have been studied for many years. As it is a widespread phenomenon, it is suspected to play some important roles in infant development. Therefore, our purpose was to provide an update of the evidence accumulated by reviewing all of the empirical or experimental studies that have been published since 1966 on IDS driving factors and impacts. Two databases were screened and 144 relevant studies were retained. General linguistic and prosodic characteristics of IDS were found in a variety of languages, and IDS was not restricted to mothers. IDS varied with factors associated with the caregiver (e.g., cultural, psychological and physiological) and the infant (e.g., reactivity and interactive feedback). IDS promoted infants' affect, attention and language learning. Cognitive aspects of IDS have been widely studied whereas affective ones still need to be developed. However, during interactions, the following two observations were notable: (1) IDS prosody reflects emotional charges and meets infants' preferences, and (2) mother-infant contingency and synchrony are crucial for IDS production and prolongation. Thus, IDS is part of an interactive loop that may play an important role in infants' cognitive and social development.
... Secondary baby talk is interesting to compare to primary baby talk because the former clearly does not have any tutorial function (Caporael et al., 1983;Ferguson, 1977;Hirsh-Pasek & Treiman, 1982;McLeod, 1993). Thus, any similarities between secondary and primary baby talk suggest a function not solely tutorial as an explanation. ...
Full-text available
Common knowledge tells us that people talk "baby talk" to dogs, but how similar talk to dogs and infants is remains unexplored. In this study, people's talk to dogs during 46 play interactions was coded and compared with previous studies about talk to infants. Talk to dogs and infants shared numerous features, including a high-pitch register, a low mean length of utterance, high frequencies of grammatically acceptable utterances, present-tense verbs, repetitiveness, and attention-getting devices. Differences were also present. Talk to dogs contained shorter sentences and more imperatives and exact repetitions; talk to infants had more questions, declaratives, and deictic utterances. Reasons for the similarities and differences are elaborated on. Both forms involve communicating with a limited and inattentive addressee, controlling the addressee's attention and behavior by focusing on an object or activity, and expressing friendliness and affection. They differ in that talk to infants is more likely to treat the other as a conversant and, especially, to tutor the other in naming.
... Several theorists have emphasized the adaptive significance of BT (Fernald, 1992; Fisher & Tokura, 1996; McLeod, 1993; Papoušek et al., 1990; Sachs, 1977). ...
The most robust finding on infants' listening preferences has been widely characterized as a preference for baby talk (BT) over adult-directed speech (ADS). Although prosodic modifications characteristic of BT also convey positive affect, differences in affect across BT and ADS speech registers have not been controlled in previous studies. This set of experiments sought to elucidate the basis for 6-month-olds' listening preference by independently manipulating affect and speech register. When affect was held constant, no preference for any speech register was observed. Moreover, when ADS stimuli presented more positive affect than BT stimuli, infants' preferences followed the positive affect. Higher and more variable pitch was neither necessary nor sufficient for determining infants' preferences, although pitch characteristics may modulate affect-based preferences. The BT preference is thus attributable to a more general preference for speech that imparts relatively positive affect, a preference perhaps ascribable to a preexisting general-purpose mechanism opportunistically exploited by language.
... They typically do so by elevating pitch, exaggerating pitch changes, slowing tempo, simplifying vocabulary and syntax, increasing the number of repetitions, and lengthening pauses between words (Fernald et al., 1989;Fernald & Simon, 1984;Grieser & Kuhl, 1988). This speaking style, commonly referred to as infant-directed (ID) speech, has been observed with infants from birth to 2 years of age (McLeod, 1993), and several studies have demonstrated that it elicits stronger visual and affective responding from infants than does adult-directed (AD) speech (Cooper & Aslin, 1990;Fernald, 1985;Fernald & Kuhl, 1987;Pegg, Werker, & McLeod, 1992;Werker & McLeod, 1989). It is possible that infants in the present study were more attentive to the emotion displays portrayed by their own mothers because the facial and vocal expressions were more infant directed. ...
This research investigated the role of person familiarity in the ability of 3.5-month-old infants to recognize emotional expressions. Infants (N = 72) were presented simultaneously with two filmed facial expressions, happy and sad, accompanied by a single vocal expression that was concordant with one of the two facial expressions. Infants' looking preferences and facial expressions were coded. Results indicated that when the emotional expressions were portrayed by each infant's own mother, infants looked significantly longer toward the facial expressions that were accompanied by affectively matching vocal expressions. Infants who were presented with emotional expressions of an unfamiliar woman did not. Even when a brief delay was inserted between the presentation of facial and vocal expressions, infants who were presented with emotional expressions of their own mothers looked longer at the facial expression that was sound specified, indicating that some factor other than temporal synchrony guided their looking preferences. When infants viewed the films of their own mothers, they were more interactive and expressed more positive and less negative affect. Moreover, infants produced a greater number of full and bright smiles when the sound-specified emotion was "happy," and particularly when they viewed the happy expressions of their own mothers. The average duration of negative affect was significantly longer for infants who observed the unfamiliar woman than for those who observed their own mothers. These results show that when more contextual information-that is, person familiarity-was available, infants as young as 3.5 months of age recognized happy and sad expressions. These findings suggest that in the early stages of development, infants are sensitive to contextual information that potentially facilitates some of the meaning of others' emotional expressions.
Full-text available
This thesis studies early functioning of autistic infants, still unclear. Our developmental study of family home movies shows that they lack interest in others since the first year of life, but begin to actively avoid others only during the second year of life. Thus there could be a dynamic leading to this final relational deficit (maybe through a intersubjective deficit). To shed light on this dynamic, we took parents-infant synchrony into account, using computational methods on family home movies: infants show a very early deficit in orienting toward others and then in intersubjectivity; parents do respond normally to their infant but they are more insistent in soilliciting it as early as the first semester. However, may the infant relational deficit have affective consequences on its parents in the long run ? Motherese or infant-directed speech (IDS) reveals emotion in parent-infant interaction; belonging to an interactive loop, it supports both affective interaction and cognitive development: attention, language... May it sustain autistic infant's social engagement ? And may it decrease over time because of the interactive dysfunction (lack of social interest and lack of intersubjectivity) ? An algorithm detecting IDS was created and applied to a pilot study before to use it on all our films. First résults, to be confirm, suggest that IDS may be efficient in sustaining autistic infant's reactivity, but maternal, not paternal IDS efficiency could drop over time. In conclusion, we have laid a groundwork for studying dynamic interactions of autistic infants with their parents in order to approach the processes leading to the final autistic syndrom.
We examined mothers' speech to 2-year-old children in both quiet and moderately noisy conditions. Mothers were recorded while teaching their children two words, one of which occurred in the context of other people speaking. Parents used characteristics of infant-directed speech (IDS) to these older children. Even more interesting was that many of the prosodic changes typical in IDS were accentuated in noise. In a context in which multiple people were speaking simultaneously, mothers spoke to their children with both increased pitch and increased word duration relative to when they spoke in a quiet condition. However, these changes were quite small, lending only mixed support to proposals that me of the advantages of IDS is that it is easier to separate from background noise.
This study investigated pitch and communicative intent in mothers' infant-directed speech spoken to their infants at birth, 3, 6, 9, and 12 months. Audio recordings of mothers (6 with female, and 6 with male infants) talking to another adult and to their infant at 5 ages were low-pass filtered and rated by 60 adults on 5 scales (Positive or Negative Affect, Express Affection, Encourage Attention, Comfort or Soothe, and Direct Behavior). Mean fundamental frequency (F0) and pitch range of utterances were also measured. Utterances associated with positive affect tend to peak at 6 and 12 months, whereas more directive utterances peaked at 9 months. Mean F0 followed the age trend for affective utterances, and pitch range followed the trend for directive utterances. The results suggest mother speech patterns reflect, complement, and perhaps facilitate infant development.
This research investigated the role of person familiarity in the ability of 3.5-month-old infants to recognize emotional expressions. Infants (N= 72) were presented simultaneously with two filmed facial expressions, happy and sad, accompanied by a single vocal expression that was concordant with one of the two facial expressions. Infants' looking preferences and facial expressions were coded. Results indicated that when the emotional expressions were portrayed by each infant's own mother, infants looked significantly longer toward the facial expressions that were accompanied by affectively matching vocal expressions. Infants who were presented with emotional expressions of an unfamiliar woman did not. Even when a brief delay was inserted between the presentation of facial and vocal expressions, infants who were presented with emotional expressions of their own mothers looked longer at the facial expression that was sound specified, indicating that some factor other than temporal synchrony guided their looking preferences. When infants viewed the films of their own mothers, they were more interactive and expressed more positive and less negative affect. Moreover, infants produced a greater number of full and bright smiles when the sound-specified emotion was “happy,” and particularly when they viewed the happy expressions of their own mothers. The average duration of negative affect was significantly longer for infants who observed the unfamiliar woman than for those who observed their own mothers. These results show that when more contextual information—that is, person familiarity —was available, infants as young as 3.5 months of age recognized happy and sad expressions. These findings suggest that in the early stages of development, infants are sensitive to contextual information that potentially facilitates some of the meaning of others' emotional expressions.
This study examined the interaction of two sets of speaker practices which are useful to listeners-the acoustic reduction of given as opposed to new words in speech and the acoustic and grammatical modifications typical of speech to infants. English-speaking mothers described a puppet show to their 14- month-old infants and to an adult. This task elicited the repeated use of a common set of target words in spontaneous utterances. First- and second-mentioned target words differed in similar ways in infant- and adult-directed speech: Repeated words were shorter, quieter, lower-pitched, and less variable in pitch than first-mentioned words. Repeated words were also placed in less prominent positions relative to other words in the same utterance: That is, they were less likely than new words to be utterance-final, to receive sentence stress, and to occur on an utterance pitch peak. Thus, the attenuation of given information in speech appears insensitive to large differences in the speaker′s view of the listener′s sophistication. We suggest that some of the benefits of the distinction between given and new words might be available to the infant listener: The acoustic difference between given and new words might serve to automatically attract infants′ attention to new words at the expense of background information.
Virtually all the research to date on infant preference for infant-directed talk has been conducted with English-learning infants. This study was designed to test whether the preference for native infant-directed (ID) communication extends to a language group other than English, and whether infants' preference for ID communication extends to a nonnative language. English- and Cantonese-learning infants 4.5 and 9.0 months of age were tested on their preference for filmed displays of a Cantonese-speaking female addressing either her own infant (ID) or an adult (AD). Both groups of infants showed a robust attentional and affective preference for ID over AD in Cantonese. These results strengthen the claim that the special properties evident in ID communication may have universal attentional and affective significance.
Full-text available
outline a set of principles governing infants' deployment of attention to auditory events / infants initially look in the direction of sounding objects, later reaching for the objects / they find some qualities highly salient, such as female voices in general and the mother's voice in particular / infants selectively attend to the pitch contours and rhythms of animated speech and musical sequences / they encode finer details of some musical sequences, notably those typical of their culture / simplicity or familiarity of the sequences and greater maturity of the infant lead to more comprehensive auditory processing identify a number of parallels between infants' processing of speech and music, and propose directions for future research (PsycINFO Database Record (c) 2012 APA, all rights reserved)
Full-text available
argue that the characteristic vocal melodies of human mothers' speech to infants are biologically relevant signals that have been shaped by natural selection discuss the current debate about the status of human language as an evolved mechanism, as a way of examining criteria for when it is appropriate to invoke natural selection as a causal explanation for the evolution of human behavior / examine the antiadaptationist argument that language is an "exaptation" and therefore not a product of selection, as well as the argument that adaptive features must be designed for optimal efficiency the characteristics of mothers' speech to infants are described in some detail, along with research on the communicative functions of intonation in infant-directed speech / examine infant-directed speech in the context of ethological research on vocal communication and maternal behavior in nonhuman primates, in order to identify selection pressure relevant to understanding the adaptive functions of human maternal speech (PsycINFO Database Record (c) 2012 APA, all rights reserved)
The involuntary confinement of persons with mental illnesses involves balancing the rights of the individual hospitalized with that person's well-being and the safety of society. This chapter reviews the historical development of civil commitment and other types of involuntary hospitalization, and highlights the forensic evaluation issues that arise in these contexts. First, the background and legal developments that have shaped civil commitment are discussed, highlighting the increased constitutional protections provided individuals in the latter third of the 20th century. Then, the legal foundations for involuntary hospitalization in the criminal justice system are discussed, focusing on persons adjudicated not guilty by reason of insanity, prison inmates and parolees, and defendants adjudicated incompetent to stand trial. After reviewing the legal developments in civil commitment and involuntary hospitalization, this chapter reviews the evaluative issues most common to the mental health professionals who practice in these contexts.Keywords:civil commitment;involuntary hospitalization;civil rights;mental illness;dangerousness
Studied "elderspeak"—the particular speech patterns often used in communicating with the elderly—to evaluate its efficacy. 15 male and 15 female older Ss (aged 68–80 yrs) of good health were compared with 15 male and 15 female younger Ss (aged 19–33 yrs). Texts selected from newspapers were recorded in a focal stress pattern designed to assist linguistic processing, and in nonfocal stress and no stress versions. Comprehension of older and younger Ss was determined through written answers to content questions. For older Ss, focal stress resulted in better comprehension than no stress, which, in turn, was better than the nonfocal stress. For younger Ss, the differences were not statistically significant. For older Ss, comprehension and recall of spoken language were improved by stress placed at locations optimal for linguistic processing. (PsycINFO Database Record (c) 2012 APA, all rights reserved)
Reviews research on the expression of emotion through the nonverbal (prosodic) features of speech. Findings show that emotions can be expressed prosodically, apparently through a variety of prosodic features. This communication appears to be largely the same for different individuals and cultures, suggesting that the prosodic expression of emotion is not conventional. Some correlations between dimensions of emotions (e.g., anxiety, aggression) and prosodic features are discussed; activity or arousal seems to be signaled by increased pitch height, pitch range, loudness, and rate. The possibility that prosodic contours (patterns of pitch and loudness over time) are used to communicate specific emotions is explored. A number of authors suggest that anger is communicated by an even contour with occasional sharp increases in pitch and loudness. Methodological difficulties with the acoustical manipulation of relevant auditory and articulatory features are noted. It is suggested that a major step in investigating the prosodic expression of emotion will be learning how to synthesize various articulatory and auditory features. (3 p ref) (PsycINFO Database Record (c) 2012 APA, all rights reserved)
Studied patterns of caregivers' speech to the elderly in 2 institutional settings: in 1 facility about 45 residents required intensive nursing care; and the other was a health-related facility for people needing only moderate care. It was expected that residents needing only a little care would be spoken to as the caregivers—20 nurses aides—spoke to each other, but those requiring more care would receive speech accommodated to them—baby talk (BT). Speech recordings were made at lunch as the caregivers interacted with residents. With residents requiring less care, BT was very rare. About 24% of the talk addressed to the more incapacitated residents was BT. It appeared that the contextual significance of BT was in dependency relationships. Institutionalization per se did not account for BT. Further research is recommended to determine whether the elderly respond well to BT. (PsycINFO Database Record (c) 2012 APA, all rights reserved)
Examined the self-fulfilling influences of social stereotypes on dyadic social interaction. Conceptual analysis suggests that a perceiver's actions based upon stereotype-generated attributions about a specific target individual may cause the behavior of that individual to confirm the perceiver's initially erroneous attributions. A paradigmatic investigation of the behavioral confirmation of stereotypes involving physical attractiveness (e.g., "beautiful people are good people") is presented. 51 male "perceivers" interacted with 51 female "targets" (all undergraduates) whom they believed to be physically attractive or physically unattractive. Tape recordings of each participant's conversational behavior were analyzed by naive observer judges for evidence of behavioral confirmation. Results reveal that targets who were perceived (unknown to them) to be physically attractive came to behave in a friendly, likeable, and sociable manner in comparison with targets whose perceivers regarded them as unattractive. (42 ref) (PsycINFO Database Record (c) 2012 APA, all rights reserved)
2 experiments examined behavioral preferences for infant-directed (ID) speech over adult-directed (AD) speech in young infants. Using a modification of the visual-fixation-based auditory-preference procedure, Experiments 1 and 2 examined whether 12 1-month-old and 16 2-day-old infants looked longer at a visual stimulus when looking produced ID as opposed to AD speech. The results showed that both 1-month-olds and newborns preferrred ID over AD speech. Although the absolute magnitude of the ID speech preference was significantly greater, with the older infants showing longer looking durations than the younger infants, subsequent analyses showed no significant difference in the relative magnitude of this effect. Differences in overall looking times between the 2 groups apparently reflect task variables rather than differences in speech processing. These results suggest that infants' preference for the exaggerated prosodic features of ID speech is present from birth and may not depend on any specific postnatal experience. However, the possible role of prenatal auditory experience with speech is considered.