Imitation of Facial and Manual Gestures by Human Neonates Andrew N. Meltzoff, M. Keith Moore Science, New Series, Volume 198, Issue 4312 (Oct. 7, 1977), 75-78. Stable URL: bttp//links jstor.org/sici?sic!=0036-8075% 281977 1007%293%3A 198%3A4312%3C75%3AIOFAMG%3E2.0,CO%3B2-H ‘Your use of the ISTOR archive indicates your acceptance of JSTOR’s Terms and Conditions of Use, available at hhup:/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. Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the sereen or printed page of such transmission. Science is published by American Association for the Advancement of Science, Please contact the publisher for further permissions regarding the use of this work. Publisher contact information may be obtained at hup:/www.jstor.org/journals/aaas.html, Science ©1977 American Association for the Advancement of Science ISTOR and the ISTOR logo are trademarks of ISTOR, and are Registered in the U.S. Patent and Trademark Office For more information on ISTOR contact jstor-info@umich.edv, ©2002 JSTOR bupulwww jstor.org/ Tue Oct 22 14:40:40 2002References and Notes 1. Butta ga 1, Siverman, Porcher Mee ae ei, Sie e tach fat R Hen Pore Mo Sas 3 WP cham an A Reerhaun, ache atte 00 eh Scac an M Ete N08 on Buta Mey a MISimer, 6; Borge. Simon My Boch tum ivennans tC. Wye, Cin Res 1, TSyac ahaa Siege Sab Seg dig J Pre 44,8 soskis and C. Shagass, Povehophysiolgy Ht. Footie A 2a, Marth, Se Pn FRAC Hal Ni. Roprepor, H, K. Hopkins, fl, Sivnan Sec i, dir Bel. Seal Q358)'S Dumont and P De. Hrewstadblae Ch Newrphyaa 12505 ‘isa Slade tt Rev. Neri 2.87 97, Sil we ceca ce tt eso ‘mera ne fling eae”) t@pdb $250) brat inne hes OD FEcagesiah Reeremere orem ene {stato fe nN fring iin a mechan, a epitancin neta eas ‘he rslael behaloal seins om the ‘ya lle 1 Sgt tee Seta BLM ae Sena ‘The use of Faxed presents ewo inteelated rts the inal he concn at ft eeu [NM Seeroeenhalge Ce Seto 1 35) Thetet Fae rodice paras and was daconnued when. ‘ver the final tale to desynchroniz ona: id ‘imo Peay tony rere weed soon he sal was tated 30 8 pied CO, and body tonperatore were contin: fous mantred ord mane wis rea 10, Conn aimlaton cletrodes were formed from an led we ence 0 tan fe in es Fah {heir niacin, Rear condiates wee a {eer ca and enznal= to oleae Wnglsing EEC dexynctonteio [Re Bonralet sd "Newman Tuior Elec apie: Cn wep Bh 405 ire daaiencrimed coor T80 poles per secon Each in termite 33 eecpey es meseniation of he shcrages eat coming of ve EPs fr Boh ‘Gani OR suoulaion were collected uring Ser tly SARE wafason son Say vet mete ‘ed in served as boeing data HE GeefiSamic eaten tee 16, Spe ator anya variance or see tent tiers er gout a Fit ee me eae Coase Me Sinaiton satay ers amt re Pape amon ra oen 15. T Opis, Science 19, 3431963): D. Sais Hiccroencophaloge. Cin Newophio 3. (ise he stan ad BK Crapper Hen if 1 9S) WE. Footer el. Bap 1s, NoBagi, Fave, € Loeb M, Mantes 1 Neurophiel 28 45 99). 1 Wath a Cop, New 1, (ea Tal Bal id 247909, v7 The rn ied Sia #10 = 3.4, ara he prope by RF tinlaon I fer 8, 9) e58, P 1 MBemeiess, M. Demetresc, Tos, le troencephlogr, Cl, Newrophyil 1,1 948 Wana’ Me Kamat’, b. Crete igs Btn Rest 373 09 VG. Sree bain er 4,51 (1908) D: M. Peony an IM Orem, ap, Neurol 8,10 1951). 19, Mi Stare, Bran es, 919968, 2a Sap ite Stece 185301 (981: J. Orem Md bs Me Feeney fraln Rew 9, 2009708 ME Feeney and M Orem, Ps. ohn 9 21, TLE Grubb, paper prevented at Army Science Confrence, ioe 16 22, Previn er nisi a negative correlation feween EP augmenting sod heal). We eine withdrawal as the depres of movement ‘Teay tom the pons stinull"Assmenter re ‘tect moe tothe stoi tn da he resuers Se par of ther intl resetion wn dete Seutdnwal. We have the impression th he r= Tice oreo Be fack a te re The dactopancytheteloe, may bei the ‘etntion of wih 2, Seppo ty NEF arm BNSTEOI6S 01. ‘man Engineering Laboratory to LHL 31 January 1977 revived 23 May 1977 Imitation of Facial and Manual Gestures by Human Neonates Abstract, Infants berween [2 and 21 days of age can imitate both facial and manu- al gestures; this behavior cannot be explained in terms af either conditioning or innate releasing mechanisms Such imitation implies that human neonates can ‘equate their own unseen behaviors with gestures they see others perform. Piaget and other students of devel ‘opmental psychology consider the imita- tion of facial gestures to be a landmark ‘achievement in infant development. In fants are thought to pass this milestone at approximately 8 to 12 months of age. Infants younger than this have been pos- tulated to lack the perceptual-cognitive sophistication necessary to match a ges- lure they see with a gesture of their own Which they cannot see (J). The experi- ments we report show that the infant's imitative competence has been under- estimated. We find that 12- to 21-day-old Fig. 1. Sample photographs from videotape recordings of 2- to 3-week-old infants imi infants can imitate both facial and manu- al gestures (Fig. 1) This result has impli- cations for our conception of innate hu- ‘man abilities and for theories of social and cognitive development, ‘An experimental evaluation of the ne- ‘onate’s imitative competence raises sev- eral methodological difficulties. One con- sists of distinguishing true imitation from a global arousal response. For example, fone can conclude nothing about imita- tion if an infant produces more tongue protrusions in response to a tongue pro- trusion demonstration than he does 10 ing) tongue protrusion, (b) mouth opening, and (clip protrusion demonstrated by an adult experthe presentation of a neutral facial ex pression. It would be more parsimonious simply to conclude that a moving, human face is arousing forthe infant and that in- creased oral activity is part of the in- fant’s arousal response. A second issue involves. controlling interactions be- tween adult and infant that might shape the imitative response. We found that if parents were informed of the imitative tasks we planned to examine, they prac- ticed these gestures with their infants be- fore coming into the laboratory so that their baby “would do well on the test.” In reviewing films of preliminary work, wwe also noticed that the examiner tended to alter the rhythm of his tongue protru- sion as a function of the response of the infant. These kinds of interactions would expose findings of imitation t0 a variety of explanations, including the possibility ‘that the infants were merely being condi tioned to imitate tongue protrusion. A third issue concerns the scoring of the in: fant’s responses. The movements tested were not generally produced in a dis- crete, unambiguous fashion, and not sur- prisingly, there were gross differences in the scoring function of whether or not the observer knew which gesture had been demonstrated to the infant. In the experiments we now report, these three issues are addressed as fol lows. (i) Bach infant's response to one ‘gesture is compared to his response to nother similar gesture demonstrated by the same adult, at the same distance from the infant, and at the same rate of movement, For instance, we test wheth- er infants produce more tongue protru- sions after an adult demonstrates tongue protrusion than after the same adult demonstrates mouth opening, and vice versa. If differential imitation occurs, it ‘cannot be attributed to a mere arousal of ‘oral activity by a dynamic, human face. (ii) Parents were not told that we were ‘examining imitation until after the stud ies were completed: moreover, the ex periments were designed to preclude the possibility that the experimenter might alter the rhythm of his demonstration as 2 function of the infant's response. (il) ‘The infant’s reactions were videotaped and then scored by observers who were uninformed of the gesture shown to the infant they were scoring (2). In experiment 1, the subjects were six \N \N Gesture thown to stant Fig. 2. Distribution of “yes” judgments asa funetion ofthe gesture show to the infant during experiment Six judges) Shaded bars indicate the ii ‘The maximum possible numberof judgment ive reaction. (a) Number of judgments that infants foreach bar was 96 (sx infants and Fesponded wit ip protrusion (LP) to each ofthe four gestures shown them. (b) mouth-opening (MO) judgments, (e)tongue-proteusion (TP) judgments, and (J) sequential-finger-movement infants ranging in age from 12 to 17 days (Y= 143 days). Three were male and three female. Testing began with a 90- second period in which the experimenter presented an unreactive, “passive face™ (lips closed, neutral facial expression) to the infant. Each infant was then shown the following four gestures in a different random order: lip. protrusion, mouth ‘opening, tongue protrusion, and sequen- tial finger movement (opening and clos- ing the hand by serially moving the fin- gers). Each gesture was demonstrated four times in a 1$-second stimulus-pre- sentation period. This period was imme- diately followed by a 20-second response period for which the experimenter stopped performing the gesture and re- sumed a passive face. In order to allow for the possibility that the infants might not watch the fist stimulus presentation, the procedure allowed a maximum of three stimulus presentations and corre- sponding response periods for any one gesture. Half the cases required only one stimulus presentation. In those cases ne- ccessitating more than one stimulus pre- sentation, the 20-second response period used in assessing imitation was the one following the final presentation of the gesture, A 7-second passive-face period separated the presentation of each new type of gesture from preceding ones, The videotape recordings of the re- sponse periods were scored in a random order by undergraduate volunteers. Two groups of six coders were used. One group scored the infant's facial behavior; the other scored the manual responses, ‘The face coders were informed that the infant in each videotaped segment was shown one of the following four ges- tures: lip protrusion, mouth opening, tongue protrusion, or passive face. They were instructed to order the four ges- tures by ranks from the one they thought it most likely the infant in each segment \was imitating to the one they thought was least likely. No other training was given. The hand coders were treated identically, except that they were in- formed that the infant in each segment (SFM) judgments. was presented with one of the following Condition | Baseline | Baseline period [tsoeimertall Response perioa 1 [eerimental] Response period 2 exposure seconds) exposure 1] (150 seconds) _[exposure 2] (150 seconds) experimenter | Passive Passive tace | Gesture 1] Passive face | Gesture 2| Passive face Infant Paci No pacitier Pacitier No pacitier Pacitier No pacitier Fig 3. Schematic illustration of the pacifier technique for assessing facial ation in neonates in experiment 2. Half ofthe infants were exposed to the gestures in the order tongle protrusion, mouth opening: the other half were exposed to the gestures in the reverse order SCIENCE, VOL. 8hand gestures: sequential finger move- ‘ment, finger protrusion, hand opening, (or passive hand. For the purposes of analysis, the two highest ranks and the two lowest ranks ‘were collapsed. This procedure yields di cchotomous judgments of whether it was likely or unlikely (hereafter referred to as, yes" or no") that the infants were im= itating a particular gesture. The distribu- tion of “yes” judgments for each infant gesture peaked when the corresponding sesture was demonstrated by the experi= menter (Fig. 2). In all four instances, Cochran Q tests (@) reveal that the judged behavior of the infants varies sig nificantly as a function of the gestures they are shown [lip protrusion, P < .01 (Fig. 2); mouth opening, P< .02 (Fig. 2b); tongue protrusion, P< 05 (Fig. 2c); and sequential finger movement, P< 001 (Fig. 2d). That this variation is, attributable 10 imitation is supported by the fact that none of these effects is sig- nificant when the judgments correspond- ing to the imitative reaction (shaded col- lumns in Fig. 2) are excluded from the analyses. Experiment 1 avoided a prolonged stimulus-presentation period during Which the experimenter might alter the timing of his gesturing as a function of the infant's responses. However, in adopting a fixed stimulus-presentation period as brief as 15 seconds, it was sometimes necessary to repeat the pre- sentation to ensure that the infants ac tually saw the gesture they were to imi tate. This procedure then opened the possiblity that the experimenter might unwittingly have been prefitering the ddata by readministering the stimulus pre~ sentations until the random behavior of the infant coincided with the behavior demonstrated. A second study was therefore designed which is not open to this potential objection. ‘The subjects in experiment 2 were 12 infants ranging in age from 16 to 21 days (K = 193). Six were male and six fe- male. They were shown both mouth- ‘opening and a tongue-protrusion gesture in a repeated-measures design, counter- balanced for order of presentation. The experimental procedure is illustrated in Fig. 3. Testing began with the insertion of a pacifier into the infant's mouth. In- fants were allowed to suck on it for 30 seconds while the experimenter present- ed a passive face. The pacifier was then removed, and a 150-second baseline pe- riod was'timed. After the baseline peri- cd, the pacifier was reinserted into the infant's mouth, and the first gesture was demonstrated "until the experimenter “a “lb im 4 il 1 a | | H ' : 1 ict Fig. 4. Total frequency of (a) tongue-proteu- sion and (b) mouth-opening responses for three conditions in experiment 2. Abbrev ations: baseline period; TP, tongue-proru Sion response period: and MO, moutivopen: ing response period Judged that the infant had watched it for 15 seconds. The experimenter then stopped gesturing, resumed a passive face, and only then removed the pacifier. ‘A 150.second response period, during which the experimenter maintained his, passive face, was clocked. Immediately thereafter the pacificer was reinserted, and the second gesture was presented in an identical manner (4). Infants did not tend to open their ‘mouths and let the pacifier drop out du ing the mouth-opening demonstration; nor did they push out the pacifier with their tongues during the tongue-protru- sion demonstration. On the contrary, they sucked actively with the pacifier re- maining firmly within their mouths dur- ing the stimulus-presentation period. ‘Thus, the pacifier technique () safe- guards against the experimenter’s alter- ing his gesturing as a function of the imi tative responses ofthe infant and (i) per- mits the experimenter to demonstrate the gesture until the infant has seen it, while ensuring that the experimenter's assessment of this point is unco: taminated by any knowledge of the i fant’s imitative response. ‘The 36 videotaped segments (12 in- fants for 3 periods each) were scored in a random order by an undergraduate assi tant who was uninformed of the struc- ture of the experiment. The frequencies of tongue protrusions and mouth open- ings were tallied for each videotaped seg- ment (5). The results demonstrate that ‘neonates imitate both tongue protrusion and mouth opening (Fig. 4). As assessed by Wilcoxon matched-pairs signed-ranks tests (), significantly more tongue-pro- trusion responses occurred after that gesture had been presented than during the baseline period (P < .005) or after the mouth-opening gesture (P< .005) Similarly, there were significantly more mouth-opening responses after that ges- ture had been demonstrated than during the baseline period (P <_.08) or after the tongue-protrusion gesture (P < .05), tis noteworthy that under the present exper- imental conditions, the infants had to delay their imitation until after the ges- ture to be imitated had vanished from the perceptual field ‘At least three different mechanisms could potentially underlie the imitation we report 1) It could be argued thatthe imitation is based on reinforcement administered by either the experimenter or the par- ‘ents. In order to prevent the experiment- €e from shaping the infants imitative re- sponding, the procedure directed that he maintain an unreactive, neutral face dur- ing the response period. The experiment- er's face was videotaped throughout both experiments in order to evaluate whether this procedure was followed. ‘The videotaped segments were shown to observers whose task it was to score any reinforcements that the experimenter ad~ ministered. No smiles or vocalizations were noted in any trial. Indeed, the only changes from the passive face occurred in three trials in experiment 1, when the experimenter was judged to “blink ex- tremely rapidly.” Considering only ex- periment 2, then, the experimental pro- ‘cedure does not appear to have been vi lated, and therefore, differential shaping ‘of the mouth-opening and tongue-protru- sion responses during the successive 130-second response periods is an un- likely source of the effects obtained. Since none of the parents were informed about the nature of the study, special practice on imitative tasks at home in preparation for the experiment was avoided. Further, informal questioning revealed that no’ parent was aware of ever having seen babies imitating in the first 21 days of life; indeed, most were astonished at the idea. Thus, a history of Parental reinforcement seems an improb- able basis for imitation at this very early ee. 2) This early imitation might be based fon an innate releasing mechanism such as that described by Lorenz and Tin- bergen (6). This view would hold that tongue protrusion, mouth opening, lip protrusion, and sequential finger move. ‘ment are each fixed-action patterns and that each is released by the correspond- ing adult gesture (sign stimulus). The overall organization of the infant's imi- tative response, particularly its lack of stereotypy, does not favor this inter- pretation. In addition, the fact that in- fants imitate not one, but four different gestures, renders this. approach un wieldy. ”3) The hypothesis we favor is that this imitation is based on the neonate’s ca pacity to represent visually and proprio- ceptively perceived information in a form common to both modalities. The in fant could thus compare the sensory in formation from his own unseen motor behavior to a “'supramodal” representa. tion ofthe visually perceived gesture and construct the match required (7). In brief, we hypothesize that the imitative responses observed are not innately or- ‘ganized and “released,” but are accom- plished through an active matching pro- ccess and mediated by an abstract repre sentational system. Our recent obser- vations of facial imitation in six new: borns—one only 60 minutes old—sus- ‘gest (0 us that the ability to use inter- ‘modal equivalences isan innate ability of ‘humans. If this isso, we must revise our current conceptions of infancy, which hold that such a capacity is the product ‘of many months of postnatal devel. ‘opment. The ability to act on the basis of fan abstract representation of per cceptually absent stimulus becomes the starting point for psychological devel- ‘opment in infancy and not its culmina tion, ANDREW N. MELTZOFF © Department of Experimental Psychology, Oxford University, Oxford, England 0X1 3UD M. Kerri Moore Child Development and Mental Retardation Center, University of Washington, Seattle 98195 Reterences and Notes 1 trex Page Py Deamsendinie Asrieemicareane kee yokes ieee leg a Nene ae Hn ste in et i en eee ee? Se eA Beye Ne aredant de ee ace ad ete Sa sels cern i a, om See aeaneta aS ata gee frigalfataana Bo eat ERR SES Metta Sos ere ie am etapa se Pte sc an ae ee aie etn kat tne i Ease yee eee Seri has Tau Bad ed antec tae et a one ac te Saves sgegnrrouae otc Balik chen eee te Rae pts nt an sce ete ac a as rig beatae uae seine Pedi tee eer tami cts ctw hee ws no sacun erence P > be {cen the durason afte preentation of the Tongue potion Uo ef secongs) and fpowte opening Ot 48" seconts) peste Felina war reveal that ifr cons ted to make tucking movements or abou! ee ‘Sedsafers pace was ened Therefore: poser as tonived ant elo the beplng ‘Bhi Stsecood taslne or response period The nan’ orl actviy ding hile Was ot inled inte anal +. Kone pt way red ol when he ioogae ede at omy eyood the Wes ‘moat opening was talied ony when the nat {aly opened us mouthIntobverver ages nen (nerf precment ded bythe tl fmber af sgrcemeris pus ssateemeats) wat ip fr ba tongue prurion perc tad pout opening (percent 6. Reevens and A Eeberaen, 2: Terpochol 2. Tah Tahcgen. i of Fin Ox foul Univ: Pree, New York 31, 7. SSipramodal ie wed, Toigwing T. Bower Detlopment Irfan (Freeman, San Fan fSsco, 87), to dente that the representation «nba oe sey may ‘te prevented athe cna Meeting of the So- ‘Seo for Research in Chis Detelopment Der See alos 0013 Ape 197 Prins of ts Saccusiy Ce vee a Ee ating ee Ce ent Retrditon Center of the University of Pete ithe eer Riker beg Suspestins, We ae epecally SASS Siler Se fey Gopnik’ V Hanson. R. Hart. Me MeCary. . Seep Sie rete alten Child Bevel nt Menta Ketarlaiton Comer (WH0, University of SE SH ‘Transplantable Pancreatic Carcinoma of the Rat Abstract. Panereatie carcinoma, which developed in a male Fischer 344 rat fed 0.1 percent nafenopin for 20 months, is being successfully transplanted into weanling rats. The tumor cells contain variable numbers of zymogen granules, and the endo- plasmic reticulum and the Golgi apparatus appear prominent. This transplantable tumor, which displays substantial amylase and lipase activity. should serve as a useful model system for immuno- and chemotherapeutic experiments, as well as for the study of synthesis, storage, and release of zymogen proteins in neoplastic cells. Epidemiological studies indicate an ‘unequivocal increase in the incidence of pancreatic carcinoma in several coun- tries during the past three decades (I, 2). In the United States, pancreatic carci ‘noma ranks as the fourth most common ccause of death by cancer, exceeded only by cancer of the lung, large bowel, and breast (2). Difficulty in early diagnosis, as well as lack of adequate knowledge of its biological behavior, appear to be ma- jor factors contributing to the poor prog noxis of panereatie carcinoma in humans G). Since several studies suggest that panereatie cancer in man may be etio- logically related to exogenous chemicals and thus preventable (4), attempts are Fig. 1. (A) Histological appearance of the original pancreatic carcinoma from a male Fischer 344 rat treated with nafenopin for 20 ‘months. Acinar dilferentation i evident, and humerous mitoses ate present. (Hematoxstin and eosin; 80.) (B) Subcutaneous transplant ‘of pancreatic carcinoma (second generation) fixed in 2.8 peroent glutaraldehyde m0. 1M eae fodylate bulfer, pH 7.4, for 30 minutes and then in I percent OsO,. This section (0.8 jm thick) of plastic-embedded tissue shows ‘merous seeretory granules in the eytoplasm of tumor cells. (Toluidine blue: *480,) being made to develop suitable animal models of this cancer (5) which could serve as an effective system for various experimental manipulations aimed at preventing or altering the natural pro- gression of the disease, Here we de- scribe a transplantable pancreatic car noma of the rat which is capable of pro- ducing amylase and lipase. ‘The primary tumor developed in the panereas of a male Fischer 344 rat that was fed nafenopin Q-methyl-21p-(1. 2. 3, 4tetrahydro-I-naphthyl)phenoxyjpro- Pionie acid; Su-13437), at a dietary con- centration of 0.1 percent for 20 months, Nafenopin is a potent hepatic peroxisome proliferator (6) and, as reported elsewhere (7).the majority of rats fed this compound develop liver tumors. The primary pancre- atic tumor was highly vascular, measured {6m in diameter. and contained several cystic spaces filled with straw-colored fluid. Metastases were present in the lv ct. Histologically, the tumor was a well- to-poorly differentiated pancreatic acinar carcinoma originating from exocrine tis- sue (Fig. 1A). On electron microscopic examination, the primary pancreatic car- cinoma cells revealed large nuclei with prominent nucleoli: the cytoplasm dis- played abundant rough endoplasmic re- ticulum and prominent Golgi apparatus. Numerous zymogen granules were also seen in the tumor cells. Portions of this primary tumor were minced and diluted in sterile normal saline for inoculation in- to the peritoneal cavity at laparotomy, SCIENCE, VOL. 198