Mode of presentation affects infants' preferential attention to - - PDF document

See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/279037116

Mode of presentation affects infants' preferential attention to singing and speech

Article in Music Perception · December 2014

DOI: 10.1525/MP.2014.32.2.160

CITATIONS

13

READS

132

1 author: Eugenia Costa-Giomi The Ohio State University

31 PUBLICATIONS 832 CITATIONS

SEE PROFILE

All content following this page was uploaded by Eugenia Costa-Giomi on 12 January 2017.

The user has requested enhancement of the downloaded file.

Mode of Presentation Affects Infants’ Preferential Attention to Singing and Speech Author(s): Eugenia Costa-Giomi Source: Music Perception: An Interdisciplinary Journal, Vol. 32, No. 2 (December 2014), pp. 160-169 Published by: University of California Press Stable URL: http://www.jstor.org/stable/10.1525/mp.2014.32.2.160 . Accessed: 02/03/2015 18:24

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp

.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms

• f scholarship. For more information about JSTOR, please contact support@jstor.org.

.

University of California Press is collaborating with JSTOR to digitize, preserve and extend access to Music Perception: An Interdisciplinary Journal. http://www.jstor.org

This content downloaded from 128.83.63.180 on Mon, 2 Mar 2015 18:24:20 PM All use subject to JSTOR Terms and Conditions

MODE OF PRESENTATION AFFECTS INFANTS’ PREFERENTIAL ATTENTION

TO SINGING AND SPEECH EUGENIA COSTA-GIOMI University of Texas – Austin

ALMOST FROM BIRTH, INFANTS PREFER TO ATTEND

to human vocalizations associated with speech over many other sounds. However, studies that have focused

• n infants’ differential attention to speech and singing

have failed to show a speech listening bias. The purpose

• f the study was to investigate infants’ preferential

attention to singing and speech presented in audiovisual and auditory mode. Using an infant-controlled prefer- ence procedure, 11-month-olds were presented with audiovisual stimuli depicting a woman singing or recit- ing a song (Experiment 1, audiovisual condition). The results showed that infants attended significantly longer to singing than to speech. In Experiment 2 (visual con- dition), infants watched the same videos presented with no sound and in Experiment 3 (auditory condition), they listened to the singing and speech stimuli in English and a foreign language. No differences in length

• f attention to singing and speech were found in either
• experiment. The results of the study reconcile the seem-

ingly contradicting findings of previous investigations and show that mode of presentation affects infants’ preferential attention to speech and singing. The facili- tating effects of facial cues on infants’ processing of speech and singing are discussed. Received: July 19, 2013, accepted March 8, 2014. Key words: auditory development, music, singing, speech, multimodal perception

I

NFANTS ARE IMMERSED IN SOUND EVEN BEFORE

they are born. They are exposed to music, language, and noises of all types almost every second of their

• lives. The sounds infants hear are produced in a variety
• f ways. They hear music played by audio-only devices

such as home stereos and portable music players and also by dynamic audiovisual sources such as interactive music toys, computers, and TVs. Sometimes they see mom singing directly to them and sometimes they simply hear her singing in the next room. In order for infants to understand the content and structure of music, they must learn to integrate and disassociate the multiple audiovisual cues of such diverse music experiences. Research on infants’ music perception has focused almost exclusively on their ability to detect differences, patterns, and similarities in music presented to them in auditory mode and in the absence of visual cues. Although many of the procedures used in infant research accompany the presentation of sound with an image, the image has no other value than directing infants’ gaze and facilitating the measurement of atten- tion to the auditory stimuli. Such approach has increased our understanding of infants’ sensitivity to a wide variety of music parameters but has provided limited information as to how they integrate the visual and auditory information in which they are immersed in daily life. Visual information can distort the perception of audi- tory stimuli. Studies on the McGurk effect illustrate this phenomenon in a language context (e.g., McGurk & MacDonald, 1976) and investigations completed with music conductors (e.g., Morrison, Price, Geiger, & Cor- nacchio, 2009) and performers (e.g., Juchniewicz, 2008) illustrate it in a music context. The findings of these and

• ther studies show that when presented with the same

auditory information matched with different visual information, subjects often detect nonexistent differ- ences in sound. For example, Morrison et al. (2009) found that identical ensemble performances were rated as more or less expressive depending on whether they were matched with videos of conductors displaying more or less expressive gestures. Infants do not seem immune to the influence of visual information on their processing of auditory stimuli. The McGurk effect has been found in infants as young as 5 months old (Rosemblum, Schmucker, & Johnson, 1997). Although visual information can create surprising auditory illusions, it can also aid in the understanding

• f auditory stimuli. This is true for adults trying to follow

a conversation in a noisy environment (e.g., Ross, Saint- Amour, Leavitt, Javitt, & Foxe, 2007), and for infants trying to decode the structure, content, and meaning of

• speech. The facial movements that accompany speech

Music Perception, VOLUME 32, ISSUE 2, PP. 160–169, ISSN 0730-7829, ELECTRONIC ISSN 1533-8312.  2014 BY THE REGENTS OF THE UNIVERSITY OF CALIFORNIA ALL

RIGHTS RESERVED. PLEASE DIRECT ALL REQUESTS FOR PERMISSION TO PHOTOCOPY OR REPRODUCE ARTICLE CONTENT THROUGH THE UNIVERSITY OF CALIFORNIA PRESS’S RIGHTS AND PERMISSIONS WEBSITE, HTTP://WWW.UCPRESSJOURNALS.COM/REPRINTINFO.ASP. DOI: 10.1525/MP.2014.32.2.160

160

Eugenia Costa-Giomi This content downloaded from 128.83.63.180 on Mon, 2 Mar 2015 18:24:20 PM All use subject to JSTOR Terms and Conditions

provide infants with dynamic and correlated audiovisual cues that facilitate language learning (Blossom & Morgan, 2006; Hollich, Newman, & Jusczyk, 2005; Teinonen, Aslin, Alku, & Csibra, 2008). Visual infor- mation also aids in the understanding of speech embedded in music. For example, it is easier to under- stand song lyrics when watching the singers than when just listening to them (Hidalgo-Barnes & Massaro, 2007; Jesse & Massaro, 2010). Additionally, the visual information displayed by a singer affects the percep- tion of music parameters such as the size of the inter- vals sung (Thompson & Russo, 2007; Thompson, Russo, & Livingstone, 2010). The visual cues associated with speaking and singing are particularly important during infancy. They enhance the communication of affect and emotion, regulate infant behavior, and foster socialization in humans as well as other species (Falk, 2004). Infant-directed (ID) communication is indeed characterized by exaggerated facial expressions, gestures, and movements such as raised eyebrows, smiling, and head nods (Dissanayake, 2000) which are perceived as friendly and inviting (Falk, 2004; Huron, Dahl, & Johnson, 2009). The gestures that accompany ID communication affect the acoustical characteristics of the sounds being produced (Huron & Shanahan, 2013; Ohala, 1984; Tartter, 1980). Smiling and moving the brows convey positive affect to vocal sounds by raising pitch height, increasing resonance, and producing a brighter timbre (Huron & Shanahan, 2013; Ohala, 1984; Tartter, 1980). Not surprisingly, infants are more attracted to ID speech and singing than non-ID communication (Fernald, 1985; Trainor, 1996). The characteristics of ID singing, such as exaggerated dynamics, articulation, and facial expressions make it particularly appealing to infants. In fact, mothers talk to their infants in a way that resembles singing: Infant- directed speech (i.e., ID speech) portrays the melodious contours, exaggerated stress patterns, and elongated vowels (Fernald, 1985) typical of infant music (Trehub, Unyk, & Trainor, 1993). It is clear that ID speech and ID singing share many common characteristics and are both effective in capturing and sustaining infants’ inter-

• est. But ID singing may be actually better than ID

speech at eliciting and maintaining their attention (Nakata & Trehub, 2004). Nakata and Trehub (2004) compared the behaviors of two groups of 6-month-old infants towards videos of their own mothers talking or singing and found that infants in the singing group attended significantly longer and spent more time look- ing at the video without moving than did infants in the speech group. The authors suggested that infants were more attentive to singing than speech possibly because maternal singing is particularly effective in inducing emotional communication. Additionally, they proposed that mothers’ small repertoire and stereotyped singing performance style might be more comforting and plea- surable to their infants than the highly variable speaking style characteristic of ID speech. The idea that maternal singing plays a privileged role in infants’ emotional development may explain the con- tradicting results of Nakata and Trehub’s (2004) study showing a listening bias for singing over speech, and two recent studies that failed to show such bias (Corbeil, Trehub, & Peretz; 2013; Costa-Giomi & Ilari, 2014). In these studies the music stimulus was sung or hummed by a woman unfamiliar to the infants instead of their

• mothers. Perhaps it is only maternal singing that elicits

sustained attention in infants. Corbeil et al. (2013) pro- posed an alternative explanation for the seemingly con- flicting findings regarding infants’ differential attention to song and speech. They suggest that the main contrib- utor to infants’ preferential attention is the affect con- veyed by the stimuli rather than the nature of the spoken and sung stimuli itself. In their study, infants listened longer to happy stimuli regardless of whether it was sung or spoken. However, considering that the happy stimuli portrayed ID speech or song and the neutral stimuli, adult-directed speech or hummed mel-

• dies, the results of the study may simply reflect infants’

known preference for ID stimuli. There was another difference between the study showing a singing bias (Nakata & Trehub, 2004) and the ones that failed to do so (Corbeil et al, 2013; Costa-Giomi & Ilari, 2014). Whereas in Nakata and Trehub (2004) infants watched videotapes of speech and song, in Corbeil et al. (2013) and Costa-Giomi and Ilari (2014), infants listened to audio recordings of spoken and sung melodies. It is possible that the facial expres- sions and movements of the speaker and singer affect infants’ preferential attention to the videos. In fact, there’s evidence that mothers smile and move more when singing to their infants than when speaking to them (Plantinga, Trehub, & Russo, 2011, as cited in Corbeil et al. 2013). The more animated visual cues that accompany maternal singing may be responsible for infants’ increased attention towards such stimuli (Nakata & Trehub, 2004). Corbeil et al. (2013) and Costa-Giomi and Ilari (2014) did not find a listening bias towards singing when using audio recordings, but they did not find a listening bias for speech either. This is surprising considering the results of extensive research showing infants’ preferential attention to speech over nonspeech sounds (Butterfield & Siperstein, 1970; Colombo & Bundy, 1981; Samples &

Infant Preferential Attention 161 This content downloaded from 128.83.63.180 on Mon, 2 Mar 2015 18:24:20 PM All use subject to JSTOR Terms and Conditions

Franklin, 1978; Schultz & Vouloumanos, 2010; Spence & DeCasper, 1987; Vouloumanos & Werker, 2004, 2007; Vouloumanos et al., 2010). A wide variety of nonspeech sounds have been investigated including filtered speech, white noise, water sounds, sine waves, monkey calls, and non-referential human sounds such as laughing (But- terfield & Siperstein, 1970; Colombo & Bundy, 1981; Samples & Franklin, 1978; Schultz & Vouloumanos, 2010; Spence & DeCasper, 1987; Vouloumanos & Wer- ker, 2004, 2007; Vouloumanos et al., 2010). It is remarkable that given the consistency of the results regarding infants’ preferential attention to speech, the few studies that included singing failed to show a speech listening bias (Nakata & Trehub, 2004; Corbeil et al., 2013; Costa-Giomi & Ilari, 2014). The purpose of the present study was to investigate the effect of mode of presentation on infants’ differential attention to singing and speech. Infants’ attention to speech and singing was measured under three conditions: audiovisual mode, visual-only mode, and auditory-

• nly mode. The design and stimuli of the present

study differed from those of Nakata and Trehub’s (2004) in critical ways. Nakata and Trehub used real interactions between mothers and infants when recording the singing and spoken stimuli. This allowed the authors to compare infants’ responses to natural and ecologically valid stimuli. It is worth noting, how- ever, that the results of the study were based on infants’ behaviors when watching the videotapes and not during the live singing and speech episodes and that there was no direct comparison between infants’ attention to singing and speech; infants were presented with either singing or speech. Because Nakata and Trehub recorded the stimuli in the presence of infants, the videos needed to be edited to delete or minimize the vocalizations produced by the infants during the recording session. The editing process consisted of muting the sound when infants’ vocalizations occurred in the absence of mothers’ vocalizations, as well as reducing the sound volume of the video to 10% when infants’ and mothers’ vocalizations occurred simulta-

• neously. It is unknown how the gaps in sound created

by the editing process affected infants’ responses to the stimuli. Furthermore, it is difficult to establish which component of the videos elicited infants’ atten- tion to singing and speech, as both the auditory and visual information differed substantially between the two conditions. The music and language content selected by the mothers as well as the gestures, facial expressions, and movements they used when singing and talking to their infants were different between the two types of stimuli. In summary, the conclusion that infants’ attend more intently to singing than to speech is questionable on the basis of the differences between the two types of stimuli and between the two groups

• f children. To minimize differences between conditions,

in the present study infants were exposed to both sing- ing and speech, and the stimuli consisted of the same woman singing and reciting the same text under con- trolled conditions. The mode of presentation varied in the three experiments. In Experiment 1, infants were presented with audiovisual stimuli of the woman sing- ing and talking, in Experiment 2, infants watched the same videos presented with no sound (i.e., visual only), and in Experiment 3, infants heard but didn’t see the videos (i.e., auditory only). Experiment 1 The purpose of the experiment was to study infants’ differential attention to singing and speech presented in audiovisual mode. The music and speech videos depicted a woman unfamiliar to the infants singing and reciting the text of a folk song. The verbal content of the two stimuli was thus the same. The use of a preferential looking procedure allowed for a within-subjects design and a direct comparison of infants’ responses to singing and speech. This procedure has been used extensively to determine comparative attention to two stimuli (Cooper & Aslin, 1994; Vouloumanos & Werker, 2004).

METHOD

• Sample. Sixteen 11-month-old infants born full term

and with no known disabilities participated in the experiment (M ¼ 344 days). They were recruited from the registry of infants born in an urban center in the Southwest following IRB procedures. The data from four infants were not included in the analyses because

• f computer failure (n ¼ 1) or interruptions during the

testing (fire alarm, n ¼ 1; infants or caregivers sneezing and coughing, n ¼ 2). The final sample included 12 infants, half of whom were males. Music stimuli. The music stimuli consisted of two verses and the refrain of a Canadian folk song. The French version of the song had been used in a previous inves- tigation on infants’ attention to speech and music pre- sented in auditory mode (Costa-Giomi & Ilari, 2014). In the present study, the English version was used and presented to infants raised in English-speaking families. The song was in 6/8 and D major. According to the parents, the song was unfamiliar to the infants. A young woman was videotaped singing and reciting the song. She was asked to sing and talk expressively 162

Eugenia Costa-Giomi This content downloaded from 128.83.63.180 on Mon, 2 Mar 2015 18:24:20 PM All use subject to JSTOR Terms and Conditions

and to look straight at the camera. She sang without vibrato in Bb. The two videos (i.e., song and speech) were edited to equalize the loudness and length of the videos using the software Adobe After Effect. The edited videos lasted 30 s each.

• Procedure. Infants sat on their caregivers’ lap in a quiet,

dimly lit room facing a 42’’ monitor (Panasonic TH-42PHD5) equipped with side speakers. A hidden video camera (Panasonic Super Dynamic II WV-CP460) positioned below the monitor captured infants’ behaviors and transmitted them to an adjacent room where the researcher observed, analyzed, and recorded all responses. Theresearcher,whocouldnothearorseethestimulibeing presented in the testing room, registered the duration of infants’ gaze to the monitor during each trial using the software Habit X (Cohen, Atkinson, & Chaput, 2004). Infants’ differential attention to the stimuli was mea- sured using an infant controlled preference procedure (Cooper & Aslin, 1994; Vouloumanos & Werker, 2004). This procedure allowed for the comparison of infants’ attention to the two videos by determining the time they spent looking at each. Each of the videos was presented 10 times in alternating order with half of the infants watching the speech video first and the other half watching the singing video first. The videos were pre- sented in an infant-controlled setting: the video played for as long as the infant exhibited focused attention. The procedure began with the display of a flashing circle on the monitor to attract the infant’s attention towards the screen. As soon as the infant looked for- ward towards the screen, one of the videos started play-

• ing. The video played as long as the infant looked at the

screen and up to 30 s (i.e., the duration of the video). If the infant looked away from the monitor for 2 s or more during a trial, the video stopped. The flashing circle reappeared on the monitor to redirect the infant’s atten- tion to the screen. Once the infant looked forward, the next video started playing from the beginning of the

• song. The same protocol was repeated for the 20 trials
• f the experiment, 10 singing and 10 speech.

RESULTS AND DISCUSSION

Infants’ cumulative looking time to the singing and speech videos were calculated and compared through a paired t-test. The results revealed a significant differ- ence in attention to the two stimuli, t(11) ¼ 3.05, p < .01. Infants attended significantly longer to singing (M ¼ 74.13, SD ¼ 23.18) than speech (M ¼ 55.06, SD ¼ 25.85) (Figure 1). The analysis of infants’ individual responses supported this finding: 10 of the 12 attended longer to the singing video than the speech video. This finding is in agreement with the results of Nakata and Trehub (2004), who found longer sustained atten- tion to singing than speech with 6-month-old infants. Although in both studies infants were presented with audiovisual stimuli, there were many differences between the two investigations. Whereas Nakata and Trehub used infant-directed speech and singing, the videos of the present study were recorded in the absence

• f infants. Infants were exposed to videos of their own

mothers in Nakata and Trehub’s study but a strange woman in the present study. The singing and speech videos in Nakata & Trehub’s were very diverse as the mothers chose what and how to talk and sing to their

• infants. On the other hand, the videos of the present

study consisted of a woman singing and reciting the same song facing the camera, with controls over the duration, loudness, and delivery of the text. Finally, the studies were different in their design and method, as in one of the investigations infants watched either the singing or the speech video, whereas in the present one, all infants watched both videos. Despite all these differ- ences, the conclusions of the two studies are the same: infants show preferential attention to singing over speech. The stimuli of the present experiment were prepared in such a way as to reduce the differences between the speech and singing stimuli. The text, duration, loudness, and actress were the same in the two videos, yet there were slight visual differences between them. For example,

FIGURE 1. Infants’ cumulative looking time to speech and song presented in audiovisual (Experiment 1), visual-only (Experiment 2), and audio-only mode (Experiment 3). Infant Preferential Attention 163

This content downloaded from 128.83.63.180 on Mon, 2 Mar 2015 18:24:20 PM All use subject to JSTOR Terms and Conditions

gestures associated with breathing and sustained vowels were more noticeable in the sung rendition than in the recited rendition of the song. Although these differences were subtle, they may have driven infants’ attention to the singing video. A second experiment was completed to determine whether this was case. Experiment 2 The purpose of Experiment 2 was to determine whether infants’ differential attention to the audiovisual stimuli

• f Experiment 1 was triggered by the visual cues of the
• videos. Infants were exposed to the same stimuli used in

Experiment 1 but presented in silence. Preferential attention to the visual-only singing stimulus would pro- vide evidence that visual cues drove infants’ attention to the singing videos of Experiment 1.

METHOD

• Sample. Twelve 11-month-old infants (six girls) born

full term and with no known disabilities participated in the experiment (M ¼ 339 days). They were recruited from the registry of infants born in an urban center in the Southwest, following IRB procedures. Music stimuli & procedure. The music stimuli were identical to those of Experiment 1 except that infants watched the 10 speech and 10 singing videos played with no sound. The same infant-controlled preference procedure of Experiment 1 was used in Experiment 2.

RESULTS AND DISCUSSION

Infants’ cumulative looking time to the singing and speech visual stimuli were calculated and compared through a paired t-test. The results showed no signifi- cant differences in attention to the two stimuli, t(11) ¼ 0.04, p ¼ .97. In fact, the cumulative looking time averages to the singing and speech videos were remark- ably similar (M ¼ 78.57, SD ¼ 41.75 and M ¼ 78.33, SD ¼ 37.20 respectively) (Figure 1). Infants’ sustained attention to the videos was higher under the visual-only mode of presentation of Experi- ment 2 than under the audiovisual setting of Experi- ment 1 (see Figure 1). This may be explained by the novelty of the video-only stimuli as it is unlikely that infants had much exposure to mute faces talking and

• singing. It is possible that infants’ differential attention

to the speech and singing videos of Experiment 2 was masked by their high attention to the novel mode of

• presentation. To explore this idea, infants’ cumulative

looking time to the first 10 videos (i.e., five speech and five singing) and last 10 videos were compared. A drop in attention between the two sets of videos would be indicative of the typical process of familiarization that

• ccurs with repeated exposure, and rule out the possi-

bility that a generalized novelty effect obscured differ- ences in attention to speech and singing. A 2 x 2 (Type

• f Simulus [speech, singing] x Trial Set [first half, last

half]) analysis of variance (ANOVA) showed a signifi- cant main effect for set, F(1, 11) ¼ 11.32, p < .01, but no main effect for type of stimulus, F(1, 11) < 0.01, p ¼ .97,

• r interaction effects, F(1, 11) ¼ 1.05, p ¼ .33. Looking

time decreased significantly between the first and the second half of the experiment (M ¼ 47.85 SD ¼ 20.91 and M ¼ 30.6, SD ¼ 22.49 respectively) but did not differ between the speech and singing stimuli. These results indicate that infants’ lack of preferential atten- tion to singing and speech presented in visual-only mode was not caused by the novelty of the visual-only mode of presentation. Additionally, to directly compare infants’ responses to the audiovisual stimuli of Experiment 1 and the visual-

• nly stimuli of Experiment 2, a 2 x 2 ANOVA (Type of

Stimulus [speech, singing] x Mode of Presentation [audiovisual, visual]) was completed. The results showed a significant main effect for type of stimulus, F(1, 22) ¼ 5.39, p ¼ .03, likely driven by infants’ preferential atten- tion to singing in Experiment 1, and no significant main effect for mode of presentation, F(1, 22) ¼ 1.13, p ¼ .30. The interaction between type of stimulus and mode of presentation was significant, F(1, 22) ¼ 5.14, p ¼ .03. Infants attended longer to singing than to speech in the audiovisual condition, protected alpha level ¼ .025, t(11) ¼ 3.05, p < .01, but not in the visual condition, t(11) ¼ 0.04, p ¼ .97. Attention to speech didn’t differ between conditions, protected alpha level ¼ .025, t(22) ¼ 1.78, p ¼ .09, nor did attention to singing, t(22) ¼ 0.26, p ¼ .80. In summary, although infants’ total cumulative looking time was comparable between the audiovisual and auditory conditions, the focus of their attention was not: They attended longer to singing than to speech only in the audiovisual condition. Experiment 2 was completed to determine whether the results of Experiment 1 showing infants’ preferential attention to singing over speech was triggered by differ- ences in the visual cues of the videos. The analyses indicate that infants’ preferential attention for singing presented in audiovisual mode could not be attributed exclusively to the visual information portrayed in the

• videos. To study the effects of the auditory information

embedded in the videos on infants’ preferential atten- tion to singing and speech, a third experiment was com-

• pleted. In Experiment 3, infants were exposed to speech

and singing presented in auditory mode. 164

Eugenia Costa-Giomi This content downloaded from 128.83.63.180 on Mon, 2 Mar 2015 18:24:20 PM All use subject to JSTOR Terms and Conditions

Experiment 3 Previous studies on infants’ attention to speech and sing- ing presented in auditory mode have shown no prefer- ential attention to either stimulus (Corbeil et al., 2013; Costa-Giomi & Ilari, 2014). In those studies, however, the stimuli portrayed a language foreign to the infants. From the results of Experiment 1 based on audiovisual stimuli in infants’ native language and the previous studies based

• n auditory stimuli in a foreign language, it is difficult to

determine whether infants’ attention to singing and speech is affected by mode of presentation (i.e., audiovi- sual vs auditory) or language (native vs foreign). The present experiment addressed this question.

METHOD

• Sample. Twenty-seven 11-month-old infants born full

term and with no known disabilities participated in the experiment (M ¼ 347 days). They were recruited from the registry of infants born in an urban center in the Southwest, following IRB procedures. The data from three infants were not included in the analyses because they had been exposed to French (n ¼ 1), or because technical problems interrupted the testing procedure (n ¼ 2). The final sample included 24 infants, half of whom were males.

• Stimuli. The music stimuli consisted of the sung and

recited renditions of the folk song used in Experiments 1 and 2 recorded in English and French. Half of the infants heard the stimuli in English and half in French. Infants in the English condition were exposed to the audio track of the videos used in Experiment 1. Infants in the French condition were exposed to audio record- ings of the same woman portrayed in the English videos reciting and singing the tune in French. All the stimuli were of the same duration (i.e., 30 s) and comparable amplitude levels. The visual stimulus of Experiment 3 was different from those of the previous two experiments. Whereas in Experiments 1 and 2 the monitor displayed the videos of a woman singing and talking, in Experiment 3, the monitor displayed a static image of a sunflower. This procedure served the purpose of focusing infants’ gaze on the monitor to measure the length of their atten-

• tion. It is important to note that infants viewed the same

image while listening to the various auditory stimuli; using an invariant image permitted the measurement of infants’ differential attention to the auditory stimuli.

• Procedure. The procedures were identical to those of

Experiments 1 and 2 except that in Experiment 3, a trial consisted of the presentation of the static image of a sunflower on the monitor while one of the recordings played through the speakers.

RESULTS AND DISCUSSION

Infants’ cumulative looking times to the singing and speech stimuli were calculated and compared. A 2 x 2 (Type of Stimulus [singing, speech] x Language [native, foreign]) ANOVA showed no significant main or inter- action effects (type of stimulus F(1, 47) ¼ 1.19 p ¼ .29; language F(1, 47) ¼ 0.18, p ¼ .68; interaction F(1, 47) ¼ 0.53, p ¼ .47). The results showed no differences in infants’ sustained attention to singing and speech regardless of the language of the stimuli. Although infants’ differential attention to the French recordings (speech: M ¼ 62.61, SD ¼ 17.41; song: M ¼ 73.10, SD ¼ 24.10) was more pronounced than to the English recordings (speech: M ¼ 70.93, SD ¼ 41.10; song: M ¼ 73.04, SD ¼ 22.85), the interaction between lan- guage and type of stimulus was not significant (Figure 1). These findings support the results of previous research showing infants’ lack of preferential attention to singing or speech presented in auditory mode (Cor- beil et al., 2013; Costa-Giomi & Ilari, 2014). Overall, the results of the present experiment and those of Experi- ment 1 suggest that mode of presentation affects infant preferential attention to speech and singing: Infants attended longer to song than to speech when presented with audiovisual stimuli but showed no preferential attention to either speech or singing when presented with auditory stimuli. Further analyses were completed to compare the results of Experiments 1 and 3 directly. Because in both experiments infants were exposed to the spoken and sung rendition of the English song, a 2 x 2 ANOVA (Mode of Presentation [audiovisual, auditory] x Type

• f Stimulus [speech, singing]) on infants’ attention to

the recited and sung English song was performed. Of particular interest were the results regarding the inter- action between the two variables because they would reveal whether mode of presentation affects infants’ preferential attention to speech and singing. The results indeed showed a significant interaction, F(1, 22) ¼ 4.49, p ¼ .05: Infants attended longer to singing than to speech presented in audiovisual mode, protected alpha ¼ .025, t(11) ¼ 3.05, p < .01, but showed no preferential attention to either speech or singing pre- sented in auditory mode, t(11) ¼ 0.42, p ¼ .68. Atten- tion to speech didn’t differ between experiments, t(22) ¼ 1.64, p ¼ .11, nor did attention to singing, t(22) ¼ 0.21, p ¼ .83. The results of the analysis also showed a significant main effect for type of stimulus, F(1, 22) ¼ 7.02, p ¼ .01 which was likely driven by

Infant Preferential Attention 165 This content downloaded from 128.83.63.180 on Mon, 2 Mar 2015 18:24:20 PM All use subject to JSTOR Terms and Conditions

infants’ differential attention to singing and speech in Experiment 1. Finally, the analysis revealed no signifi- cant main effects for mode of presentation, F(1, 22) ¼ 0.73, p ¼ .40. Although infants’ overall attention was comparable between the audiovisual and auditory con- ditions, the focus of their attention was not: They attended longer to singing than to speech only in the audiovisual condition. The results of the three experiments combined indi- cate that infants attended longer to singing than to speech only when presented with audiovisual stimuli, and that this difference in attentional behavior could not be explained by the visual cues embedded in the videos or by the auditory content of the stimuli alone. These results were corroborated when directly compar- ing infants’ responses to the English stimuli in the three

• experiments. A 2 x 3 (Type of Stimulus [speech, singing]

x Mode of Presentation [Audiovisual, Visual, Auditory]) ANOVA showed significant main effects for type of stimulus, F(1, 33) ¼ 4.95, p ¼ .03 and for the interaction between type of stimulus and mode of presentation, F(2, 33) ¼ 5.14, p ¼ .03, but no significant main effects for mode of presentation, F(2, 33) ¼ 1.13, p ¼ .30. General Discussion The combined results of the three experiments indicate that mode of presentation affects infants’ preferential attention to singing and speech. When exposed to audiovisual stimuli of a woman singing or reciting a song, 11-month-olds attended significantly longer to the singing stimulus. However, when simply listening to the same stimuli, infants showed no preferential atten- tion for either singing or speech. According to the results of Experiment 2, the difference in infants’ responses to the audiovisual stimuli couldn’t be attrib- uted exclusively to the visual information displayed in the videos: Infants exhibited no preferential attention to either singing or speech when they watched the videos in silent mode. These findings show that it was the combination of auditory and visual cues rather than solely the auditory or solely the visual information that triggered longer attention to singing than to speech in the audiovisual condition. Infants’ integration of visual and auditory cues has been studied extensively in the context of language but not in the context of music. For example, it is well known that the facial cues provided by speakers help infants process the information embedded in speech. The facilitating effect of visual cues on infants’ encoding

• f language may explain infants’ shorter attention to

speech than to singing in the present study. Research

• n language development has shown that the synchrony

between facial movements and speech facilitates infants’ language segmentation (e.g., Hollich et al., 2005), and that the association between mouth shape and sound facilitates their perception of vowels (e.g., Kuhl & Meltz-

• ff, 1982). As infants become older and more proficient

in their understanding of language, their attention to facial audiovisual cues changes in terms of duration and focus (Romberg, 2011). As a result, young infants ben- efit to a larger extent from attention to the correlated audiovisual cues of a speaker’s face than do 12-month-

• lds (Romberg, 2001). Considering that infants in the

present study were 11-months old, their shorter atten- tion to audiovisual speech than to audiovisual singing may be indicative of their more efficient encoding of audiovisual cues and understanding of language; they may not have attended very long to the spoken stimulus simply because they didn’t need to. Because infants’ integration of visual and auditory cues has received little attention in regards to music perception, we do not know if or how the facial cues provided by singers help infants process the informa- tion embedded in song. However, studies completed with adults have shown that the length of attention to singers’ facial cues is related to the complexity of the auditory stimuli. Russo, Sandstrom, and Maksimowski (2011) found that adults looked longer at the mouth of a singer when sung intervals were presented in combi- nation with other auditory information than in isolation. That an increase in the amount of auditory information resulted in longer attention to the singer’s facial cues may help explain infants’ longer attention to singing than to speech in the present study. In comparison to the speech stimulus, the singing video was more complex as it con- tained not only musical information but also the lan- guage information portrayed by the lyrics. It is likely that the increased cognitive load created by the combi- nation of music and language challenged infants’ atten- tional resources resulting in longer looks to the singing than to the speech videos. In other words, infants may have attended longer to singing than to speech because of the amount of information portrayed in song. Although there’s no clear evidence that the complex- ity resulting from the combination of music and lan- guage increases infants’ attention to song, we know that such a combination affects the processing of informa- tion embedded in short melodic sequences. Lebedeva and Kuhl (2010) found that 11-month-olds encoded changes in melodic information when the melody was sung with a repeating syllable but failed to encode the changes when the melody was sung with changing syl-

• lables. They suggested that the amount of information

166

Eugenia Costa-Giomi This content downloaded from 128.83.63.180 on Mon, 2 Mar 2015 18:24:20 PM All use subject to JSTOR Terms and Conditions

• f the stimulus with variegated syllables contributed to

infants’ decrease in melodic discrimination. Would lon- ger attention to the more complex stimulus have resulted in better encoding of the music information? The results

• f Lebedeva and Kuhl’s study make it impossible to

answer the question as the length of time that infants could attend to the stimuli had been pre-established by the researchers. However, in the present study, infants chose to attend to the more complex stimulus for a high amount of time thus supporting the idea that the amount of information embedded in song imposes higher cognitive demands resulting in longer sustained

• attention. The two proposed explanations for infants’

shorter attention to speech than to singing in audiovisual mode, namely the facilitating effects of visual cues on the processing of speech and the cognitive demands of pro- cessing song, complement each other. The facial cues provided by the speaker/singer may have facilitated infants’ processing of the stimulus containing less infor- mation (speech) resulting in shorter attention, but may have been insufficient for shortening the processing of the stimulus containing more information (singing). We know that preferential attention to relevant audi- tory stimuli develops very early in life both in humans (Vouloumanos et al., 2010) and other species (e.g., Marler & Peters, 1977). Infants prefer listening to speech over a wide variety of sounds (Butterfield & Siperstein, 1970; Colombo & Bundy, 1981; Samples & Franklin, 1978; Schultz & Vouloumanos, 2010; Spence & DeCasper, 1987; Vouloumanos & Werker, 2004, 2007; Vouloumanos et al., 2010), even shortly after birth (Vouloumanos & Werker, 2007). That in none of the three experiments presented here infants showed the expected speech bias may seem surprising if one doesn’t consider that both the speech and singing stimuli were language-based. Learn- ing language is obviously a compelling reason for infants to attend to human vocalizations whether embedded in speech or music. In fact, in a study in which the music stimulus did not include words, infants showed the expected speech bias (Corbeil et al., 2013). It is also possible that infants’ consistent high atten- tion to the singing stimuli of the present study reflects their natural attraction to music in general. Although infants are discriminative listeners capable of perceiving subtle changes in many music parameters (e.g., Trehub, 2001), they often show no preferential attention to novel music of contrasting styles and timbres, or varying dynamics and tempo. For example, 12-month-olds lis- ten attentively to classical music played with the original instrumentation intended by the composer or a modi- fied arrangement identified as ‘‘baby’’ music showing no preferential attention to either (Merkow & Costa- Giomi, 2014). They also display undifferentiated atten- tion to instrumental music played with and without expressive nuances (i.e., changes in tempo, dynamics, and timbre) (Trainor, Marie, Gerry, Whiskin, & Unrau, 2012). Similarly, research with young children has shown that three- to six-years-olds show no preferential attention to novel lullabies and children’s songs played in various vocal or instrumental renditions (Sims & Cassidy, 1997) and listen attentively to recorded music in laboratory and natural settings for remarkably long periods of time (Sims & Cassidy, 1997; Sims, 2005). Furthermore, the length of young children’s engagement in listening to novel music seems unaffected by the con- text of the listening activity (e.g., forced or free choice of music; structured or free listening activity) (Sims & Cassidy, 1997; Sims, 2005). Considering the privileged perceptual status of speech and the seemingly natural appeal of music, one would expect infants to be attracted to the combination of music and language found in

• song. Indeed, the results of the present study showed that

11-month-olds listened to singing for a similar or longer length of time than to speech, a stimulus that is known to be captivating to infants. That this was true for stimuli portraying infants’ native language as well as a foreign language and in a style that was not infant-directed fur- ther supports the attractiveness of song to infants. Parents and caregivers from all over the world use singing to capture, redirect, and engage children’s atten- tion (e.g., Barrett, 2009; Young, 2008). They sing lullabies, children’s songs, popular songs and even made up songs to children as part of their daily routines (Custodero & Johnson-Green, 2003; Custodero, Rebello Britto, & Brooks-Gunn, 2003; Dissanayake, 2000, Field, 1999; Ilari, 2005; Rock, Trainor, & Addison, 1999; Trainor, Clark, Huntley, & Adams, 1997; Trehub et al., 1993; Unyk, Tre- hub, Trainor, & Schellenberg, 1992). The results of the present study provide support for this universal practice that engages infants’ attention and enables communica- tion (Dissanayake, 2000), modulates infant arousal (Shenfield, Trehub, & Nakata, 2003), creates vital emo- tional bonds (Trehub & Nakata, 2001-2002), and facil- itates musical development (Costa-Giomi, 2013). Author Note This research was supported by the Butler Professional Development Fund. Correspondence concerning this article should be addressed to Eugenia Costa-Giomi, Butler School of Music, University of Texas-Austin, 2406 Robert Ded- man Dr. Stop E3100, Austin, TX 78712-1555. E-mail: costagiomi@austin.utexas.edu.

Infant Preferential Attention 167 This content downloaded from 128.83.63.180 on Mon, 2 Mar 2015 18:24:20 PM All use subject to JSTOR Terms and Conditions

References

BARRETT, M. S. (2009). Sounding lives in and through music: A narrative inquiry of the ‘‘everyday’’ musical engagement of a young child. Journal of Early Childhood Research, 7, 115-134. doi: 10.1177/1476718X09102645 BLOSSOM, M., & MORGAN, J. (2006). Does the mouth say what the face says? A study of infants’ sensitivity to visual prosody. In D. Bamman, T. Magnitskaia, & C. Zaller (Eds.), Proceedings

• f the 30th Annual Boston University Conference on Language

Development (pp. 24-35) Somerville, MA: Cascadilla Press. BUTTERFIELD, E. C., & SIPERSTEIN, G. N. (1970). Influence of contingent auditory stimulation upon non-nutritional suckle. In J. F. Bosma (Ed.), Third symposium on oral sensation and perception: The mouth of the infant (pp. 313-334). Springfield, IL: Charles C. Thomas. COHEN L. B., ATKINSON D. J., & CHAPUT H. H. (2004). Habit X: A new program for obtaining and organizing data in infant perception and cognition studies (Version 1.0) [Computer software] Austin, TX: University of Texas. COLOMBO, J. A., & BUNDY, R. S. (1981). A method for the measurement of infant auditory selectivity. Infant Behavior and Development, 4, 219-223. doi: 10.1016/S0163-6383(81) 80025-2 COOPER, R. P., & ASLIN, R. N. (1994). Developmental differ- ences in infant attention to the spectral properties of infant directed speech. Child Development, 65, 1663-1677. doi 10.2307/1131286 CORBEIL, M., TREHUB S. E., & PERETZ, I. (2013) Speech vs. singing: Infants choose happier sounds. Frontiers in Psychology, 4, 372. doi: 10.3389/fpsyg.2013.00372 COSTA-GIOMI, E. (2013). Infant’s perception of timbre in music. In J.-L. Leroy (Ed), Actualite ´s des Universaux en Musique / Topics in Musical Universals (pp. 183-197). Paris: Editions des Archives Contemporaines. COSTA-GIOMI, E., & ILARI, B. J. (2014). Infants’ preferential attention to sung and spoken stimuli. Journal of Research in Music Education, 62, 188-194. doi: 10.1177/002242941 4530564 CUSTODERO, L. A., & JOHNSON-GREEN, E. A. (2003). Passing the cultural torch: Musical experience and musical parenting of

• infants. Journal of Research in Music Education, 51, 102-114.

doi: 10.2307/3653454 CUSTODERO, L. A., REBELLO BRITTO, P., & BROOKS-GUNN, J. (2003). Musical lives: A collective portrait of American parents and their young children. Journal of Applied Developmental Psychology, 24, 553-572. doi:10.1016/j.appdev.2003.08.005 DISSANAYAKE, E. (2000) Antecedents of the temporal arts in early mother-infant interaction. In N. L. Wallin, B. Merker, &

• S. Brown (Eds.), The origins of music (pp. 389-410).

Cambridge, MA: MIT Press. FALK, D. (2004). Prelinguistic evolution in early hominins: Whence motherese? Behavioral and Brain Sciences, 27, 491-541. doi: 10.1017/S0140525X04000111 FERNALD, A. (1985). Four-month-old infants prefer to listen to

• motherese. Infant Behavior and. Development, 8, 181–195.

doi: 10.1016/S0163-638380005-9 FIELD, T. (1999). Music enhances sleep in pre-school children. Early Child Development and Care 150, 65-80. 10.1080/ 0300443991500106 HIDALGO-BARNES, M., & MASSARO, D. (2007). Read my lips: An animated face helps communicate musical lyrics. Psychomusicology, 19, 3-12. doi: 10.1037/h0094037 HOLLICH, G., NEWMAN, R. S., & JUSCZYK, P. W. (2005). Infants’ use of synchronized visual information to separate streams of

• speech. Child Development, 76, 598-613. doi:10.1111/j.1467-

8624.2005.00866.x HURON, D., DAHL, S., & JOHNSON, R. (2009). Facial expression and vocal pitch height: Evidence of an intermodal association. Empirical Musicology Review, 4, 93-100. HURON, D., & SHANAHAN, D. (2013). Eyebrow movements and vocal pitch height: Evidence consistent with an ethological

• signal. Journal of the Acoustical Society of America, 133,

2947-2952. doi:10.1121/1.4798801 ILARI, B. (2005). On musical parenting of young children: Musical beliefs and behaviours of mothers and infants. Early Child Development and Care, 175, 647-660. doi:10.1080/ 0300443042000302573 JESSE, A., & MASSARO, D. W. (2010). Seeing a singer helps comprehension of the song’s lyrics. Psychonomic Bulletin and Review, 17, 323-328. doi: 10.3758/PBR.17.3.323 JUCHNIEWICZ, J. (2008). The influence of physical movement on the perception of musical performance. Psychology of Music, 36, 417-427. doi: 10.1177/0305735607086046 KUHL, P. K., & MELTZOFF, A. N. (1982). The bimodal perception

• f speech in infancy. Science, 218, 1138-1141. doi: 10.1126/

science.7146899 LEBEDEVA, G. C., & KUHL, P. K. (2010). Sing that tune: Infants’ perception of melody and lyrics and the facilitation of phonetic recognition in songs. Infant Behavior and Development, 33, 419-430. doi: 10.1016/j.infbeh.2010.04.006 MARLER, P., & PETERS, S. (1977). Selective vocal learning in a sparrow. Science, 198, 519-521. doi: 10.1126/ science.198.4316.519 MCGURK, H., & MACDONALD, J. (1976). Hearing lips and seeing voices: A new illusion. Nature, 264, 746-748. MERKOW, C. H., & COSTA-GIOMI, E. (2014). Infants’ attention to synthesized baby music and original acoustic music. Early Child Development and Care, 184, 73-83. doi:10.1080/ 03004430.2013.772993

168

Eugenia Costa-Giomi This content downloaded from 128.83.63.180 on Mon, 2 Mar 2015 18:24:20 PM All use subject to JSTOR Terms and Conditions

MORRISON, S. J., PRICE, H. E., GEIGER, C. G., & CORNACCHIO,

• R. A. (2009). The effect of conductor expressivity on ensemble

performance evaluation. Journal of Research In Music Education, 57, 37-49. doi: 10.1177/0022429409332679 NAKATA, T., & TREHUB, S. E. (2004). Infants’ responsiveness to maternal speech and singing. Infant Behavior and Development, 27, 455-464. doi: 10.1016/j.infbeh.2004.03.002 OHALA, J. (1984). An ethological perspective on common cross- language utilization of F0 in voice, Phonetica, 41, 1-16. ROCK, A. M., TRAINOR, L. J., & ADDISON, T. L. (1999). Distinctive messages in infant directed lullabies and playsongs. Developmental Psychology, 35, 527-534. ROMBERG, A. R. (2011). Infant language learning from multi- modal speech cues: Developmental change and the role of visual attention (Unpublished doctoral dissertation). University of Wisconsin-Madison. ROSENBLUM, L. D., SCHMUCKLER, M. A., & JOHNSON, J. A. (1997). The McGurk effect in infants. Perception and Psychophysics, 59, 347-357. doi:10.3758/BF03211902 ROSS, L. A., SAINT-AMOUR, D., LEAVITT, V. M., JAVITT, D. C., & FOXE, J. J. (2007). Do you see what I am saying? Exploring visual enhancement of speech comprehension in noisy envir-

• nments. Cerebral Cortex, 17, 1147-1153. doi:10.1093/cercor/

bhl024 RUSSO, F. A., SANDSTROM, G. M., & MAKSIMOWSKI, M. (2011). Mouth versus eyes: Gaze fixation during perception of sung interval size. Psychomusicology: Music, Mind and Brain, 21, 98-107. doi:10.1037/h0094007 SAMPLES, J. M., & FRANKLIN, B. (1978). Behavioral responses in 7 to 9 month old infants to speech and non-speech stimuli. Journal of Auditory Research, 18, 115-123. SHENFIELD, T., TREHUB, S. E., & NAKATA, T. (2003). Maternal singing modulates infant arousal. Psychology of Music, 31, 365-375. doi: 10.1177/03057356030314002 SHULTZ, S., & VOULOUMANOS, A. (2010). Three-month-olds prefer speech to other naturally occurring signals. Language Learning and Development, 6, 241-257. doi: 10.1080/ 15475440903507830 SIMS, W. L., & CASSIDY, J. W. (1997). Verbal and operant responses of young children to vocal versus instrumental song performances. Journal of Research in Music Education, 45, 234-244. SIMS, W. L. (2005). Effects of free versus directed listening on duration of individual music listening by prekindergarten

• children. Journal of Research in Music Education, 53, 78-86.

SPENCE, M. J., & DECASPER, A. J. (1987). Prenatal experience with low-frequency maternal-voice sounds influence neonatal perception of maternal voice samples. Infant Behavior and Development, 10, 133-142. doi: 10.1016/0163-6383(87)90028-2 TARTTER, V. C. (1980). Happy talk: Perceptual and acoustic effects

• f smiling on speech. Perception and Psychophysics 27, 24-27.

TEINONEN, T., ASLIN, R. N., ALKU, P., & CSIBRA, G. (2008). Visual speech contributes to phonetic learning in 6-month-old

• infants. Cognition, 108, 850-855. doi:10.1016/j.cognition.

2008.05.009 THOMPSON, W. F., & RUSSO, F. A. (2007). Facing the music. Psychological Science, 18, 756-757. THOMPSON, W. F, RUSSO, F. A., & LIVINGSTONE, S. (2010). Facial expressions of pitch structure in music performance. Psychonomic Bulletin and Review, 17, 317-322 TRAINOR, L. J. (1996). Infant preferences for infant-directed versus noninfant-directed playsongs and lullabies. Infant Behavior and Development, 19, 83-92. doi: 10.1016/S0163- 638390046-6 TRAINOR, L. J., CLARK, E. D., HUNTLEY, A., & ADAMS, B. (1997). The acoustic basis of preferences for infant directed singing. Infant Behavior and Development, 20, 383-396. TRAINOR, L. J., MARIE, C., GERRY, D., WHISKIN, E., & UNRAU,

• A. (2012). Becoming musically enculturated: Effects of music

classes for infants on brain and behavior. Annals of the New York Academy of Sciences, 1252, 129-138. TREHUB, S. E. (2001). Musical predispositions in infancy. In R. J. Zatorre & I. Peretz (Eds.), The biological foundations of music (pp. 1-16). New York: New York Academy of Sciences. TREHUB, S. E., & NAKATA, T. (2001-2002). Emotion and music in

• infancy. Musicae Scientiae, Special Issue, 37-61.

TREHUB, S. E., UNYK, A.M., KAMENETSKY, S. B., HILL, D. S., TRAINOR, L. J., HENDERSON, J. L. & SARAZA, M. (1997). Mothers’ and fathers’ singing to infants. Developmental Psychology, 33, 500-507. doi: 10.1037//0012-1649.33.3.500 TREHUB, S. E., UNYK, A. M., & TRAINOR, L. J. (1993). Maternal singing in cross-cultural perspective. Infant Behavior and Development, 16, 285-295. UNYK, A. M., TREHUB, S. E., TRAINOR, L. J., & SCHELLENBERG,

• E. G. (1992). Lullabies and simplicity: A cross-cultural per-
• spective. Psychology of Music, 20, 15-28.

VOULOUMANOS, A., HAUSER, M. D., WERKER, J. F., & MARTIN,

• A. (2010). The tuning of human neonates preference for
• speech. Child Development, 81, 517-527. doi:10.1111/j.1467-

8624.2009.01412.x VOULOUMANOS, A., & WERKER, J. F. (2004). Tuned to the signal: The privileged status of speech for young infants. Developmental Science, 7, 270-276. doi:10.1111/j.1467- 8624.2009.01412.x VOULOUMANOS, A., & WERKER, J. F. (2007). Listening to lan- guage at birth: Evidence for a bias for speech in neonates. Developmental Science, 10, 159-171. doi: 10.1111/j.1467- 7687.2007.00549 YOUNG, S. (2008) Lullaby light shows: Everyday musical experience among under-two year-olds. International Journal of Music Education, 26, 33-46. doi: 10.1177/ 0255761407085648

Infant Preferential Attention 169 This content downloaded from 128.83.63.180 on Mon, 2 Mar 2015 18:24:20 PM All use subject to JSTOR Terms and Conditions

View publication stats View publication stats