Speech Perception in Infants

Peter D. Eimas; Einar R. Siqueland; Peter Jusczyk; James Vigorito

Science, New Series, Vol. 171, No. 3968. (Jan. 22, 1971), pp. 303-306.

Stable URL:
http://links.jstor.org/sici?sici=0036-8075%2819710122%293%3A171%3A3968%3C303%3ASPII%3E2.0.CO%3B2-V

Science is currently published by American Association for the Advancement of Science.

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at
http://www.jstor.org/about/terms.html. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained
prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in
the JSTOR archive only for your personal, non-commercial use.

Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at
http://www.jstor.org/journals/aaas.html.

Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed
page of such transmission.

The JSTOR Archive is a trusted digital repository providing for long-term preservation and access to leading academic
journals and scholarly literature from around the world. The Archive is supported by libraries, scholarly societies, publishers,
and foundations. It is an initiative of JSTOR, a not-for-profit organization with a mission to help the scholarly community take
advantage of advances in technology. For more information regarding JSTOR, please contact support@jstor.org.

http://www.jstor.org
Mon Feb 18 09:29:12 2008
excreted by ten hyperkinetic children References and Wotes 9. G . Brooker, L. J. Thomas, M. hf. Appleman,
Biochernisfrj~7. 4157 (1968).
off medication was similar to that ex- 1. W. E. Bunney, Jr., G. F. Borge, D. L. 10. E. W. Sutherlznd. G. A. Robison, R. W.
Murphy, F . K. Goodwin, paper presented at Butcher, Circ~rlatiorl 37. 179 (1968).
creted by an age-matched group of the annual meeting of the American Psychi- 11. R. C . Haynes, S. B. Koritz, F. G. Pcron,
normal controls. In addition, the mean atric Assoc., Bal Harbour, Fla., 1969; W. E. J . Hiol. Chenr. 234, 1421 (1959).
Bunncy, Jr., D. L. Murphy, F. K. Goodwin, 12. R. V. Fnrese, L. G. Linarelli, W. H. Glins-
values for both the hyperactive patients G. F. Borgc, Lancet 1970-1, 1022 (1970). mann, B. R. Ditzion, M. I. Paul, G. L. Pauk,
and control children were in the same 2. H. Crarner and W. Kuhlo. Acta Nerrrol. Endocriizolo~y 85. 867 (1969).
Psychiaf. Belg. 67, 658 (1967). 13. P. S. Schonhofer, I. F. Skidmore, M. I. Paul,
range as our adult normal controls. 3. M. I. Paul, B. R. Ditzion, G. L. Pauk, D. S. B. R. Ditzion, G. L. Pauk. G. Kiishna, B. B.
After prolonged physical activity (foot- Janowsky, Amer. J. Ps)'clziat. 126, 1493 Brodic, unpublished data.
(1970); M. I. Paul. B. R. Ditzion, D. S. 14. R. Strom-Olsen and H. W. Weil-Malherbe,
ball), there was no significant differ Janonsky, Lancet 1970-1, 88 (1970). J . Ment. Sci. 104, 696 (1958); A. Bergsman.
ence between the pre- and postexercise 4. Y. fI. Abdullah and K. Hamadah, I.arrcet Acta Psycl?irrt. Nerrrol. Scanrl. Srdppl. 33,
1970-1, 378 (1970). 5133 (1959); N. Shinfuku, 0. Michio, K.
levels of urinary cyclic AMP in seven 5. M. I. Paul, H. Cramer, F. K. Goodwin, ibid., Masao, Yotlngo Acta filerl. 5, 109 (1961);
p. 996; Arch. Gen. Psj>chiaf.,in press. R. B. Sloane, W. Hughes, hf. L. Haust, Can.
normal subjects. In contrast to our Psychiat. Ass. J . 11, 6 (1966).
6. M. I. Paul, B. R. Ditzion, G. L. Pauk.
findings, one study suggested that ex- Phar~ttacology 3, 148 (1970). 15. J. J . Schildkraut, E. K. Gordon, J . Durell,
J . Psyclziat. Re.r. 3, 213 (1965).
ercise may elevate urinary cyclic AMP 7. G. L. Gessa, J. Forn, A. Taglian~onte, G. 16. W. E. Bunney, Jr., and J . M. Davis, Arch.
Krishna, in Role of Cyclic A M P in Ser[ronnl
levels ( 17). Robison et al. also recently Function, E. Costa and P. Greengard, Eds.
Cen. Psycltiat. 13, 483 (1965).
(Raven, New York, in press). 17. D. Eccleston, R. Loose, I. A. P~illar,R. F.
reported normal levels of cerebral spinal Sugden, Lancet 1970-11, 612 (1970).
8. \V. E. Bunney and D. A. Hamburg. Arch. 18. G. R. Robison et al., ibid., p. 1028.
fluid cyclic AMP in manic patients; Gen. Psyclziat. 9, 280 (1963); A. Beige], D.
however, they did not study patients at Murphy, W. E. Bunncy, unpublished data. 23 July 1970; revised 14 October 1970 m
the time of the switch into mania and
thtrs would have missed a transient
marked peak at that time (IS). In our
studies, cyclic AMP excretion is inde- Speech Perception in Infants
pendent of age and sex ( 3 ).We have re-
viewed the known factors influencing Abstract. Discrimination o j synthetic speech .so~~nds was st~idiecl in 1- and 4-
cyclic AMP excretion ( 3 , 5), but the month-old infants. T h e speech so~trzdsvaried along alr acou~ticdiinension pre-
relative proportion of cyclic AMP com- viously sholtvz to cue phonemic distinctions arnong tlze voiced and voiceless stop
ing from extrarenal sources remains to consonants in adults. Dircrimii~ahility was measured by an itzcrecise itz cotzdi-
be determined. Thus, at this time, one tiorzed response rate to a second speech ~ o u n dalter habituation to the first speech
cannot state whether the cyclic AMP ,sound. Recovery from habituation was greater for a given acoustic difference
response is mediated centrally or is a ~ztherz the tn-o stimuli wete from different adult phonemic categories than when
reflection of peripheral metabolism. De- they were from the same category. T h e discontinuity in discrimination at tlze
creases in the excretion of urinary cy- region o/ the adult phonemic bourzdary was taken as evidence for cafegorical
clic AMP have been demonstrated in perception.
patients with pseudohypoparathyroidism.
Thus, changes in calcium metaboliqm In this study of speech perception, investigation of this nature (2) revealed
may be associated with alterations in cy- it was found that 1- and 4-month-old that the perception of this cue was very
clic AMP metabolism ( 3 ) .Although the infants were able to discriminate the nearly categorical in the sense that
changes in urinary cyclic AMP may acoustic cue urderlying the adult pho- listeners could discriminate continuous
be secondary to catecholamine or cal- nemic distinction between the voi ed variations in the relative onset of the
cium changes, our evidence documents and voiceless stop consonants / b / and first formant very little better than they
an alteration in an important process / p / . Moreover, and morc important, could identify the sound patterns abso-
that acconlpanies and, in at least one there was a tendency in these subjects lutely. That is, listeners could readily
instance, preceded gross behavioral toward categorical perception: discrim- discriminate between the voiced and
changes. This further suggests the im- ination of the same physical difference voiceless stop consonants, just as they
portance of biochemical changes in the was reliably better across the adult would differentially label them, but they
manic-depressive illness. It is of inter- phonemic boundary than within the were virtually unable to hear intra-
est to consider the possibility that the adult phonemic category. phonemic ditt'erences, despite the f a 3
concept of cyclic AMP as a trigger Earlier research using synthetic that the acoustic variation was the same
mechanism for metabolic processes may speech sounds with adult subjects un- in both conditions. The most measur-
be relevant to the switch process from covered a sufficient cue for the per- able indication of this catcgorical per-
depression to mania. ceived distinction in English between ception was the occurrence of a high
MICHAELI. PAUL the voiced and voiceless forms of the peak of discriminability at the boundary
Neuropsychiatric Institute, Centt7r stop consonants, /b-p/, /d-t/, and /g- between the voiced and voiceless stops,
for the Health Sciences, University k / , occurring in absolute initial position and a nearly chance level of discrim-
o f California, Los Angeles 90024 ( 1 ) . The cue, which is illustrated in inability among stimuli that represented
HINKICH C R A M E H the spectrograms displayed in Fig. 1, is acoustic variations of the same pho-
Laboratory o f Chernical Phart~~ncology, the onset of the first formant relative to neme. Such categorical perception is
National Heart and Lung Institute, the second and third formants. It is pos- not found with nonspeech sounds that
Betl~esda,hfaryland 20014 sible to construct a series of stimuli that vary continuously along physical con-
WILLIAME. BUNNEY, JR. vary continuously in the relative onset tinua such as frequency or intensity.
Laboratory of Clinical Science, time of the first formant, and to investi- Typically, listeners are able to discrim-
National Institute of Mental Realtlz, gate listeners' ability to identify and inate many more stimuli than they are
Betkesda, Maryland 20014 discriminate these sound patterns. An able to identify absolutely, and the dis-
22 JANUARY 1971 303
 the special processing to which sounds Not all languages studied make use
1 of speech are subjected and thus t o be of the three modal positions. English,
for example, uses only two locations, a
short lag in voicing and a relatively long
lag in voicing. Prevoicing o r long voic-
ing lead, found in Thai, for example,
11 nearly so, it may be thought t o be is omitted. Of interest, however, is the
t 10rnsec
reasonably close to the biological- basis fact that all languages use the middle

- '1
-,,- 7-
>
-O
d

F-3
'" of speech and hence of special interest location, short voicing lag, which, given
to students of language development. certain other necessary articulatory
Though the distinctions made along the events, corresponds to the English
voicing dimension are not phonetically voiced stop / b / , and one o r both of
the same in all languages, it has becn the remaining modal values. The acous-
found in the cross-language research of tic consequences for two modes of pro-

i100
. rnsec
Fig. 1. Spectrograms of synthetic speech
showing two conditions of voice onset time
(VOT): slight voicing lag in the upper
figure and long voicing lag in the lower
figure. The symbols F-l* F-2$ and F-3 rep-
resent the first three formants, that is, the
relatively intense bands of energy in the
speech spectrum, [courtesy of L. ~
and A. S. Abramson]
criminability functions d o not normally
show the same high peaks and low
troughs found in the case of the voic-
ing distinction ( 3 ) . The strong and un-
usual tendency for the stop consonants
to be perceived in a categorical manner
has been assumed to be the result of
75 - 20 D
Lisker and Abramson (5) that the duction are shown in Fig. 1; these cor-
usages are not arbitrary, but rather respond to short and long voicing lags,
very much constrained. I n studies of / b / and / p / , respectively.
the production of the voicing distinc- Given the strong evidence for uni-
tion in 11 diverse languages, these in- versal-and presumably biologically de-
vestigators found that, with only minor termined-modes of production for the
exceptions, the various tokens fell at voicing distinction, we should suppose
three values along a single continuum. that there might exist complementary
i The
~ continuum,
k ~ ~called voice onset time processes of perception ( 6 ) . Hence, if
( V O T ) , is defined as the time between we are to find evidence marking the
the release burst and the onset of beginnings of speech perception in a
laryngeal pulsing o r voicing. H a d the linguistic mode, it would appear rea-
location of the phonetic distinctions sonable to initiate our search with in-
been arbitrary, then different languages vestigations of speech sounds differing
might well have divided the V O T con- along the voicing continuum. What was
tinuum in many different ways, con- done experimentally, in essence, was to
strained only by the necessity to space compare the discriminability of two
the different modal values of V O T suffi- synthetic spcech sounds separated by a
ciently far apart as to avoid confusion. fixed difference in V O T under two con-
ditions: in the first condition the two
stimuli to be discriminated lay on op-
I
- 20 S L
I
-
0
I
I

I
posite sides of the adult phonemic
boundary, whereas in the second con-
I I I
I

I dition the two stimuli were from the

I I
I

t I I
same phonemic category.
I
The experimental methodology was a
60
-
8

ur I
- I
I

- I
I modification of the reinforcement pro-
2
e

I I

I I

I I
cedure developed by Siqueland ( 7 ) .

liL
9
".
l
I I
I After obtaining a baseline rate of high-
I
F
t
I
amplitude, nonnutritive sucking f o r
.Y-
- I - I
- I
I
I
I
I each infant, the presentation and inten-
U I
2l
"
I I
I
sity of an auditory stimulus was made
Y-
g 4 5
O
L

2 30- I
I
-
!\-
I
I
I
I
I
I
contingent
high-amplitude
uponsucking.
the i nfant's
The nipple
rate on
which the child sucked was connected
of

!\
$e

e I
I
0 I
I B to a positive pressure transducer that
e
I I
provided polygraphic recordings of all
w
m
I I I
I responses and a digital record of cri-
= 15 - I
I
-
I

I
I
-
terional high-amplitude sucking re-
I
I
I
I
II sponses. Criterional responses activated
I
I I a power supply that increased the in-
I I
tensity of the auditory feedback. A
:
I I
I I I I I I I I I I - 1 1 1 1 1 , I l l I I I I I I
I
I I I I I

B 5 4 3 2 1 1 2 3 59 3 2 1 1 2 3 8 5 4 3 2 1 1 2 1 4 - sucking rate per set-

ond maintained the stimulus at maxi-
Tirile ( m ~ n )
munl intensity, about 75 db (13 db
Fig. 2. Mean number of sucking responses for the 4-month-old infants, as a function over the background intensity o f
of time and experimental condition. The dashed line indicates the occurrence of the 62 d b ) .
stimulus shift, or in the case of the control group the time at which the shift would
have occurred. The letter B stands for the baseline rate. Time is measured with The presentation of an auditory stim-
reference to the moment of stimulus shift and indicates the 5 minutes prior to and lllus in this manner results in
the 4 minutes after shift. an increase in the rate of sucking com-
304 SCIENCE, VOL. 171
pared with the baseline rate. With con-
tinued presentation of the initial stim-
ulus, a decrement in the response rate
occurs, presumably as a consequence of
the lessening of the reinforcing prop-
erties of the initial stimulus. When it
was apparent that attenuation of the
reinforcing properties of the initial stim-
ulus had occurred, as indicated by a
decrement in the conditioned sucking
rate of at least 20 percent for two con-
secutive minutes compared with the im-
mediately preceding minute, a second
auditory stimulus was presented with-
out interruption and again contingent
upon sucking. The second stimulus was
maintained for 4 minutes after which
the experiment was terminated. Control
subjects were treated in a similar man-
ner, except that after the initial de-
crease in response rate, that is, after
habituation, no change was made in
the auditory stimulus. Either an increase
in response rate associated with a
change in stimulation or a decrease of
smaller magnitude than that shown by
the control subjects is taken as infer-
ential evidence that the infants per-
ceived the two stimuli as different.
The stimuli were synthetic speech
sounds prepared by means of a paraliel
resonance synthesizer at the Haskins
Laboratories by Lisker and Abramson.
There were three variations of the bi-
labial voiced stop / b / and three varia-
tions of its voiceless counterpart / p / .
The variations between all stimuli were
in VOT, which for the English stops
/ b / and / p / can be realized acoustically
by varying the onset of the first formant
relative to the second and third form-
ants and by having the second and
third formants excited by a noise source
during the interval when the first form-
ant is not present. Identification func-
tions from adult listeners ( 8 ) have in-
dicated that when the onset of the first
formant leads or follows the onset of
the second and third formants by less
than 25 msec perception is almost in-
variably / b / . When voicing follows the
release burst by more than 25 msec the
perception is / p / . Actually the sounds
are perceived as /ba/ or /pa/, since the
patterns contain three steady-state form-
ants appropriate for a vowel of the type
/ a / . The six stimuli had VOT values of
-20, 0, +20, +40, +60, and +SO
msec. The negative sign indicates that
voicing occurs before the release burst.
The subjects were 1- and 4-month-old
infants, and within each age level half
of the subjects were males and half
were females.
22 JANUARY 1971
Fig. 3. The mean change in response rate
as a function of experimental treatments,
shown separately for the 1- and 4-month-
old infants. (See text for details.)
The main experiment was begun
after several preliminary studies estab-
lished that both age groups were re-
sponsive to synthetic speech sounds as
measured by la reliable increase in the
rate of sucking with the response-con-
tingent presentation of the first stimulus
( P < .01) . Furthermore, these studies
showed that stimuli separated by dif-
ferences in VOT of 100, 60, and 20
msec were discriminable when the stim-
uli were from different adult phonemic
categories; that is, there was reliable
recovery of the rate of sucking with a
change in stimulation after habituation
(P < .05). The finding that a VOT dif-
ference of 20 msec was discriminable
permitted within-phonemic-category dis-
criminations of VOT with relatively
realistic variations of both phonemes.
In the main experiment, there were
three variations in VOT differences at
each of two age levels. In the first con-
dition, 20D, the difference in VOT be-
tween the two stimuli to be discrim-
inated was 20 msec and the two stimuli
were from different adult phonemic
categories. The two stimuli used in con-
dition 20D had VOT values of +20 and
+40 msec. I n the second condition,
20S, the VOT difference was again 20
msec, but now the two stimuli were
from the same phonemic category. In
this condition the stimuli had VOT
values of -20 and 0 msec or $-60 and
+80 msec. The third condition, 0, was
a control condition in which each sub-
ject was randomly assigned one of the
six stimuli and treated in the same man-
ner as the experimental subjects, except
that after habituation no change in
stimulation was made. The control
group served to counter any argument
that the increment in response rate as-
sociated with a change in stimulation
was artifactual in that the infants
tended to respond in a cyclical manner.
Eight infants from each age level were
randomly assigned to conditions 20D
and 20S, and ten infants from each age
level were assigned to the control con-
dition.
Figure 2 shows the minute-by-rnin-
ute response rates for the 4-month-old
subjects for each of the training condi-
tions separately. The results for the
younger infants show very nearly the
identical overall pattern of results seen
with the older infants. In all conditions
at both age levels, there were reliable
conditioning effects: the response rate
in the third minute prior to shift was
significantly greater than the baseline
rate of responding ( P < .01). As was
expected from the nature of the pro-
cedure, there were also reliable habitu-
ation effects for all subjects. The mean
response rate for the final 2 minutes
prior to shift was significantly lower
than the response rate for the third min-
ute before shift ( P < .01). As is ap-
parent from inspection of Fig. 1, the
recovery data for the 4-month-old in-
fants were differentiated by the 'nature
of the slhift. When the mean response
rate during the 2 minutes after shift
was compared with the response rate
for the 2 minutes prior to shift, condi-
tion 20D showed a significant incre-
ment ( P < .05), whereas condition 20s
showed a nonsignificant decrement in
responding ( P > .05). In the control
condition, there was a fairly substantial
decrement in responding during the first
2 minutes of what corresponded to the
shift period in the experimental condi-
tions. However, the effect failed to
reach the .05 level of significance, but
there was a reliable decrement when the
mean response rate for the entire 4
minutes after shift was compared with
the initial 2 minutes of habituation
( P < .02). The shift data for the
younger infants were quite similar. The
only appreciable difference was that in
condition 20s #there was a nonsignifi-
cant increment in the response rate
during the first 2 minutes of shift.
In Fig. 3 the recovery data are sum-
marized for both age groups. The mean
change in response rate (that is, the
mean response rate for the initial 2
minutes of shift minus the mean re-
sponse rate during the final 2 minutes
before shift) is displayed as a function
 of experimental treatments and age.
Analyses of these data revealed that the
magnitude of recovery for the 20D con-
dition was reliably greater than that for
the 20s condition ( P < .01). In addi-
tion, the 20D condition showed a great-
er rate of responding than did the con-
trol condition ( P < -01), while the dif-
ference between the 20s and control
conditions failed to attain the .05 level
of significance.
In summary, the results strongly indi-
cate that infants as young as 1 month
of age are not only responsive to speech
sounds and able to make fine discrim-
inations but are also perceiving speech
sounds along the voicing continuum in
a manner approximating categorical
perception, the manner in which adults
perceive these same sounds. Another
way of stating this effect is that infants
are able to sort acoustic variations of
adult phonemes into categories with
relatively limited exposure to speech,
as well as with virtually no experience
in producing these same sounds and
certainly with little, if any, differential
reinforcement for this form of behav-
ior. The implication of these findings is
that the rneans by which the catego-
8. L. Lisker and A. S. Abramson, Proc. Int. facilities of the Haskins Laboratories. We also
Congr. Phone!. Sci. 6th (1970), p. 563. thank Drs. A. M. Liberman, I. G. Mattingly,
9. ,Supported by grants H D 03386 and H D 04146 A. S. Abramson, and L. Lisker for their critical
from the National Institute of Child Health comments. Portions of this study were p r e
and Human Development. P.J. and J.V. were sented before the Eastern Psychological As-
supported by the NSF Undergraduate Partici- sociation, Atlantic City (April 1970).
pation Program (GY 5872). We thank Dr. F.
S. Cooper for generously making available the 14 September 1970 a
Randomization and the Draft Lottery
Abstract. Fifty "random permutations" were prepared for use by the Selecti~e
Service System as a basis for a two-stage randomization that preceded the lottery
drawing on I July 1970. This report identifies the perm~ttations used. It also
gives the orders in which calendar dates and numbers were put into and drawn
from t w o drums and the correlations between them.
The Selective Service System asked dars" and 25 "random permutations"
the National Bureau of Standards of the numbers 1 to 365 for use in
(NBS) to prepare 25 "random calen- connection with preparations for the
draft lottery conducted on 1 July 1970,
which determined the order in which
'Fable 1. Coefficients of correlation f r ) and men boll3 in 1951 would be called for
their significance levels (P) for pairs of induction into military service (. I .) .
permutations used in and produced by the
draft lottery procedures. I, Calendar dates
In order that all details of the prep-
and ranks in the order loaded into capsules; arations be reproducible, the permuta-
permutations 53 and 43, respectively. 11, tions were drawn from published tables
Calendar dates (ranks) in the order of entry
into drum. comouted from ~ermutation 51 Ltables and in (2), pp. 152-2291.
(43) and permuiation 46. 111,~~alendar dates Table S contains 38 permutations of
(ranks) in the order drawn from the drum; 1 to 500; by omitting
permutation shown in Fig. 2 (Fig. 3 ) .
numbers greater than 365. 38 uermu-
u

rical perception of speech, that is, per-
ception i n a linguistic mode, is ac-
complished may well be part of the
biological makeup of the organism and,
moreover, that these means must be
operative at an unexpectedly early age.
PETERD. EIMAS
Calendar dates Ranks tations of the numbers 1 to 365 were
Pair
r PY r P" formed. Table 9 contains 20 permuta-
tions of the numbers 1 to 1000; by
considering numbers from 1 to 365 and
those from 501 to 865, 40 permutations
* Probability that correlation between two ran- of the numbers 1 to 365 were formed.
dom pertnutations exceeds (in magnitude) the
observed value. An assortment of tests for randomness

PETERJUSCZYK
JAMES VIGORITO 1. NOVEYRER 3 n 30 rlnVEMNER 11/30 334
Departmerzt o f Psychology, 2. AUG'IST 1 9 IR AIJGUST 8/18 230
Browrz University, 3. APRIL R 8 4PYIL 4/R 98
Providence, Rhode Island 02912 4, J'JhlF 1'1 I Y .liJhiE 6/!q 165
- .-
5 OCTORER 2 2 2 2 OCTOIJER 1@/22 295
References and Notes he JAtliJARY 9 R JANUAHY 118 8
7. DECEMPER 1 2 1 2 OECEMRER 12/ 12 3Y6
1. A. hl. Libcrman. P. C. Delattre, F. S. Cooper, Au sT 27 2 7 PIJSIJST 8/27 239
.
I
Language and Speech 1, 153 (1958); A. M.
Liberrnan, F. Ingernann, L. Lisker, P. C . Delat- 9 EPT E 3ER 2 1 2n s E P T E ! ~ ~ E ? 9/28 263
tre. F. S. Cooper, I. Acorcst. Soc. Amer. 31, I9 , S E p T E M R E II 8 9 SEPTEI.'RE[I 9/9 251
1490 (1959). It should be emphasized that the 31 r)ECEMOFR 12/3 1 365
cues underlying the voicing distinction as dis- 1 '
cussed in the present report apply only to 12 , OC T 0 R E R 1 3 13 !>cTo~~FIR l0/13 2 86
sound segments in absolute initial position. 13. FE~RUARY 2 2 FERRlJARY 2/ 2 33
2. A. M. Liberman, K. S . Harris, H. S. Hoffman,
H. Lane, I . E x p . Psychol. 61, 370 (1961). 14. MAY 15 14 M A Y 5/15 135
3. P. D. Eimas, Language and Speech 6 , 206 15 . O C T rli R E R 2 (1 20 ~cToHFR 1f l / 2 f l 29 3
(1963); G. A. Miller, Psychol. Rev. 63, 81
(1956); R. S . Woodworth and H. Schlosberg,
Erperimenfol Psychology (Holt, New York, 17
'P 1 6
A
OH
9 1.. 1
4 ncToRFR
16 A P R I L
10/4
4/16
277
106
1954). 1 , 3
hll3VFMRER 3 ~IOVEMRER 11/3 3 I7 7
4. A. hf. Liberman, F. S. Cooper, D. P. Shank-
ueiler. 1%. Studdert-Kennedy, Psychol. Rev. 74,
431 (1967); M. Studdert-Kennedy, A. M. Liber-
]
7 11 ,
* ' '' ' 3
A
')
1)
€ ?.
G I)5 T 2
*?n
23
J ~ N E
AUGUST
6/20
R/?3
171
2 35
man, K. S. Harris, F. S. Cooper, ibid. 77,
234 (1970); M. Studdert-Kennedy and D.
Shankweiler, I . Acoust. Soc. Amer., in press. 22r
' MAY 3
0 CTn E R 9
3
9
YAY
oCTOREQ
5/3
1019
123
282
5. L. Lisker and A. S. Abramson, Word 20, 23 , rlA Y 2 2 NAY 5/ 2 122
6.
384 (1964).
P. Lieberman, Li81guislic Inquiry 9, 307 (1970).
' J'.JNE 1 1 JUNE b/ 1 152
7. E. R. Sicjueland, address presented before the Fig. 1. P a r t of a random calendar ("calendar number 53") prepared by NBS, that
29th Inte~national Congress of Psychology,
London, England (August 1969); - and was used f o r loading dates into capsules, showing the redundant f o r m a t used f o r
printing. [Derived f r o m pages 204-205 of Moses and Oakford ( 2 ) l
C. A. DeLucia, Science 165, 1144 (1969).
306 SCIENCE, VOL. 171
http://www.jstor.org

LINKED CITATIONS
- Page 1 of 1 -

You have printed the following article:

Speech Perception in Infants
Peter D. Eimas; Einar R. Siqueland; Peter Jusczyk; James Vigorito
Science, New Series, Vol. 171, No. 3968. (Jan. 22, 1971), pp. 303-306.
Stable URL:
http://links.jstor.org/sici?sici=0036-8075%2819710122%293%3A171%3A3968%3C303%3ASPII%3E2.0.CO%3B2-V

This article references the following linked citations. If you are trying to access articles from an
off-campus location, you may be required to first logon via your library web site to access JSTOR. Please
visit your library's website or contact a librarian to learn about options for remote access to JSTOR.

References and Notes

7
Visual Reinforcement of Nonnutritive Sucking in Human Infants
Einar R. Siqueland; Clement A. DeLucia
Science, New Series, Vol. 165, No. 3898. (Sep. 12, 1969), pp. 1144-1146.
Stable URL:
http://links.jstor.org/sici?sici=0036-8075%2819690912%293%3A165%3A3898%3C1144%3AVRONSI%3E2.0.CO%3B2-L

NOTE: The reference numbering from the original has been maintained in this citation list.