Canadian and Japanese preschoolers' creation of happy and sad

69

© 2012 by Psychomusicology DOI: 10.1037/h0094005

Psychomusicology: Music, Mind & Brain 2011, Vol. 21, No. 1 & No. 2

a bstr act—Canadian and Japanese 4- and 5-year-old children were asked to create a happy song or a sad song. Approximately 60% of Canadian preschoolers pro-duced a song under these circumstances in contrast to roughly 40% of Japanese preschoolers. Of the Japanese children who produced a song, most simply reproduced a familiar song. By contrast, Canadian children typi-cally included novel words, melodies, or both. The lyrics of Canadian children’s happy songs focused on candy, friends, family, and vigorous physical activity. Their sad lyrics focused on thwarted goals and the absence of family and friends. Their happy melodies showed influences of the major mode, along with dotted or syn-copated rhythms, whereas their sad songs showed sup-pressed melodic range and contour. Japanese children made much less use of emotion words or emotion-evok-ing events in their happy or sad songs. We discuss so-cialization and early music education differences across cultures that may contribute to the observed differences. k ey wor ds—invented song, emotion, preschool chil-dren, cross-cultural

Although there has been considerable investigation of preschool children’s singing of invented and stan-dard songs (e.g., Campbell, 1998; Davidson, 1985; Davidson, McKernon, & Gardner, 1981; Davidson & Scripp, 1994; Dowling, 1984; Kelley & Sutton-Smith, 1987; McKernon, 1979; Moog, 1968/1976; Moorhead & Pond, 1941, 1942; Omi, 1994), rela-tively little is known about the communication of expressive intentions in their singing. In musically oriented homes, mothers often convey information to toddlers in the context of improvised songs, and toddlers respond by joining the “singing conver-sation” (Kelley & Sutton-Smith, 1987). These sung exchanges are likely to continue beyond

toddlerhood, with mother and child taking turns at improvising song variations (Umezawa, 1990, cited in Adachi, 1994). Young children generate idiosyncratic songs for communicative purposes such as stopping a baby brother from touching food (Omi, 1994). These spontaneous songs, or “musical utterances” (Campbell, 1998, p. 67), often trigger improvised musical exchanges among friends, who repeat, modify, and extend the melodic motif while transmitting ideas and messages to each other (Campbell, 1998, pp. 28-30; Whiteman, 2001). Although preschoolers successfully communicate verbal instructions in their invented songs, it is unclear whether they can create songs to convey distinct emotional messages. Much research on the communication of emotion in music has focused on musicians’ contrastive performances (e.g., happy, sad) of the same piece of music or on listeners’ perception of the intended emotion (Gabrielsson & Juslin, 1996; Juslin, 1997; Juslin & Sloboda, 2001; Ohgushi & Hattori, 1996; Senju & Ohgushi, 1987). By contrast, relatively few studies have explored how musically untrained adults (Yamasaki, 2002) and children communicate emotion in their musical performances (Adachi, 2000; Adachi & Trehub, 1998, 1999a; Umemoto, 1994; Yamasaki, 2006). Such research has indi-cated that untrained Japanese adults and young children (5- to 6-year-olds) use variations in loud-ness and tempo to express happiness, sadness, and anger in their improvised percussion performances

Canadian and Japanese preschoolers’ creation of happy and sad songs

M ay u M i a daCh i Hokkaido University

Sa ndr a E . T r Eh u b University of Toronto & Université de Montréal

Mayumi Adachi, Department of Psychology, Hokkaido University. Sandra E. Trehub, University of Toronto & Université de Montréal. Correspondence concerning this article should be addressed to: Mayumi Adachi, Department of Psychology, Hokkaido University, N10 W7, Kita-ku, Sapporo, 060-0810, Japan. E-mail: [email protected]

Mayumi Adachi & Sandra E. Trehub

70

(Yamasaki, 2002, 2006). In addition to variations in loudness and tempo, Canadian children 4-12 years of age vary pitch level and tone of voice to distin-guish happy from sad versions of a familiar song (Adachi & Trehub, 1998). In their invented songs, Canadian 6- to 12-year-olds express happiness by means of dotted or syncopated rhythms, and they express sadness by means of isochronous rhythms, legato articulation, and affectively negative lyrics (Adachi & Trehub, 1999a, 1999b). Japanese 8- to 10-year-olds vary tempo, loudness, rhythm, and interval size to gener-ate distinctive melodies to go with prescribed happy and sad lyrics, but the melodies generated by 5- and 6-year-olds are unrelated to the emotions expressed in the lyrics (Umemoto, 1999, pp. 56-57). In the present investigation we sought to ascer-tain how preschool children use lyrics and melodies to convey happiness and sadness through their invented songs. Although preschoolers seem to be capable of creating songs (Campbell, 1998; Kelley & Sutton-Smith, 1987; McKernon, 1979; Moog, 1968/1976; Moorhead & Pond, 1941, 1942; Omi, 1994), it is unclear whether they can express their emotional intentions consistently or conventionally in such songs. Although Japanese 5- and 6-year-olds failed to use distinctive melodic features for happy and sad lyrics (Umemoto, 1994), the provision of set lyrics may have interfered with their performance. Children in the present study had complete flex-ibility with the lyrics and melodies of their invented songs. We explored the expression of emotion in invented songs by preschoolers from Canada (Experiment 1) and Japan (Experiment 2). The focus on North American and Japanese children had the potential to reveal differences arising from contrasting cultural values such as individualism versus collectivism (Stephan, Stephan, Saito, & Barnett, 1998), different display rules and attitudes about self-expression (Pittam, Gallois, Iwawaki, & Kroonenberg, 1995; Safdar et al., 2009), different socialization practices (Abe & Izard, 1999; Hayashi, Karasawa, & Tobin, 2009), and different systems of music education (Adachi & Chino, 2004). In contrast to North American children, who are socialized to

value the expression of emotion in public, Japanese children are encouraged to develop robust bound-aries between their private and public persona (Barnlund, 1975; Doi, 1986; La Voy et al., 2001). That may be why Japanese children’s figure drawings are less likely to include smiles than those of American children (La Voy et al., 2001). One might also expect more muted musical expressiveness from Japanese preschoolers than from Canadian preschoolers.

exper i ment 1

In previous research, 6- to 12-year-old Canadian chil-dren created happy and sad songs by manipulating the lyrics and expressive performing style (Adachi & Trehub, 1999b). Their sad songs exhibited more legato articulation than their happy songs regard-less of children’s age, sex, or singing ability. Older girls (11- to 12-year-olds) used dotted or syncopated rhythms, and the more proficient singers capitalized on modality (i.e., major for happiness and minor for sadness). To some extent, then, school-aged Canadian children made use of emotional expres-sive devices that are conventional in Western art music (e.g., Juslin, 1997, 2001). Our goal in the pres-ent experiment was to examine younger Canadian children’s free expression of happiness and sadness in lyrics and melodies.

Method

Participants

The participants in the present experiment were part of a larger investigation of the development of expressive singing in Canadian children 4-12 years of age (for details, see Adachi & Trehub, 1998). Although the original sample included 83 Canadian preschoolers 4-5 years of age from middle-class families, technical difficulties reduced the available sample to 66 preschoolers. Of those 66 children, 40 of them, 14 boys (M = 4.50 years, SD = .52) and 26 girls (M = 4.42 years old, SD = .50), provided mate-rial for detailed auditory analysis in the present study.

Canadian and Japanese preschoolers’ happy and sad songs

71

Apparatus and Procedure

Testing was conducted in a sound-attenuating room with only the child and a female tester pres-ent. To make the room friendly to young children, some color drawings of animal characters were pasted on the wall. VHS recordings were made with a Panasonic VW-1500 camera, and separate digital audio-recordings were made with a Sony TCD-D7 recorder and a Sony ECM-95A microphone. Of the 66 preschoolers, 35 were instructed to make up a happy song and 31 to make up a sad song. To assist them with this task, children were presented with a picture of a rabbit displaying the target emotion, and were asked to describe how the rabbit was feel-ing and why the rabbit was feeling that way. After children identified the target emotion (i.e., feeling happy or sad), they were asked to make up a happy (or sad) song that would sound just like the way the rabbit was feeling. Substantial encouragement and prompting (but no provision of specific ideas) were given to children who exhibited difficulty with the task. On average, children who completed the task took 2 min or less to do so. By contrast, those who showed initial difficulty generally failed to complete the task despite the encouragement offered.

Content analysis of lyrics

A native speaker of Canadian English who was familiar with children’s songs transcribed the lyrics of each song. These transcriptions were confirmed independently by a second native speaker of English. The following aspects of the lyrics were coded categorically. 1. Type of lyrics. The lyrics were coded as “known” if they reproduced those of an existing song, as “variation” if they consisted of modifications of the lyrics of an existing song, as “original/personal” if the lyrics were original and depicted the child’s everyday experience, and as “original/others” if the original lyrics included make-believe stories or incantations. 2. Positive emotion words. This category was coded “present” if the lyrics contained one or more positive emotion words such as happy, glad, fun, good, laugh,

smile, love, like, favorite, delicious, clean, and thank you. Otherwise, it was coded “absent.” 3. Negative emotion words. This category was coded “present” if the lyrics contained one or more negative emotion words such as sad, scary, lonely, alone, bad, angry, hate, disappointed, irritated, cry, sorry, and separated. Otherwise, it was coded “absent.” 4. Positive emotion-evoking events. This category was coded “present” if the lyrics contained one or more descriptions of a positive event (e.g., “Everyone wanted to play with him.” “A bunny was eating an ice cream.”). Otherwise, it was coded “absent.” 5. Negative emotion-evoking events. This category was coded “present” if the target lyrics contained one or more descriptions of a negative event (e.g., “He was bullied.” “He could not play with his friends.”). Otherwise, it was coded “absent.” Inter-rater reliability for each coding category was evaluated by calculating Cohen’s kappa (Siegel & Castellan, Jr., 1988) between the first author and an independent rater who coded five happy and five sad songs (i.e., 25% of songs) that were selected randomly. High reliabilities were confirmed for all coding categories (κs = .75-1.0). Coding from the first author was used for statistical analyses.

Content analysis of melodies

Digital recordings of children’s songs were converted to acoustical files by means of Sound Designer II Software (Digidesign). Two music teachers who had experience with young children listened to five happy and five sad songs (i.e., 25% of all songs), confirming that the following categories could be coded reliably (κs = .69-1.0). Coding from the first author was used for statistical analyses. 1. Melody type. Each melody was coded as “known” if it was a nearly exact copy of another song that the child knew, “variation” if it was a variation of an existing song, “original rhythm” if its novelty was primarily rhythmic, and “original melody” if it featured original pitch and rhythmic patterns. 2. Modality. Most melodies whose modality was apparent were closer to the major than to the minor mode. A melody with minor influences was coded as “minor.” We also coded melodies as “minor” if they had a flattened or reduced pitch range


72

or compressed contour. Melodies were coded as “unclear” if the modality was not readily identifiable or if both modalities were evident. 3. Rhythm. A melody was coded as “dotted” if its rhythm sounded more dotted than isochronous, as “isochronous” if it sounded more isochronous than dotted, and as “unclear” if the rhythm was not read-ily identifiable. 4. Beat. Each melody was coded as “regular” if the singer held a relatively steady beat throughout, as “partly regular” if the singer held a relatively steady beat for a portion of the song, and as “irregu-lar” if there was no evidence of a steady beat.

Results and Discussion

As noted, 40 of the 66 children (60.6%) produced songs in response to the request for happy or sad songs. Specifically, 20 of 35 children (57.1%) produced happy songs, as requested, with the remaining 15 children producing no song. Moreover, 20 of 31 children (64.5%) produced a sad song in response to that request, with the remaining 11 chil-dren producing no song. All analyses were based on the 20 happy and 20 sad songs produced by these children.

Lyrics

Regardless of their intended emotion (happy or sad), 36 Canadian preschoolers, or 90% of the sample, produced original lyrics (n = 36). For the songs with original lyrics, 72.5% (n = 29) featured animal characters such as a bunny (suggested by the picture provided) or dinosaur (original/others) and 17.5% (n = 7) featured personal experiences (origi-nal/personal). Only 10% (n = 4) of children used the lyrics of familiar songs, the remaining children creating variations of familiar lyrics or creating entirely novel lyrics. Thus, subsequent analyses of the lyrics included novel lyrics as well as those borrowed from familiar songs. Positive emotion words (e.g., happy, love, hug, kiss) were used more often in happy songs (n = 8) than in sad songs (n = 2), χ2(1, N = 40) = 4.80, p = .03. Positive emotion-evoking events were also used more frequently in happy (n = 6) than in sad (n = 1)

songs, χ2(1, N = 40) = 4.33, p = .04. Negative emotion words (e.g., sad, crying, lonely) appeared exclusively in children’s sad songs (n = 14), χ2(1, N = 40) = 21.54, p < .0001, but negative emotion-evoking events occurred more frequency in sad (n = 15) than in happy songs (n = 3), χ2(1, N = 40) = 14.55, p < .0001. In Adachi and Trehub (1999a), Canadian 6- to 12-year-olds used positive emotion words or posi-tive emotion-evoking events regardless of the intended emotion. For preschoolers in the present study, however, such lyrical devices featured prom-inently in happy songs. Table 1 depicts examples of emotion-laden themes in the lyrics of preschool-ers’ songs. Positive lyrical themes involving candy, cake, or ice cream were significantly more frequent in happy than in sad songs, χ2(1, N = 40) = 11.61, p = .001. Similarly, actions like hopping, jump-ing, running, or playing were significantly more frequent in happy than in sad songs, χ2(1, N = 40) = 5.63, p = .018. In previous research, death was the prototypi-cal sad theme for older children (Adachi & Trehub, 1999a), but only a single preschooler mentioned death in the present study (Figure 1). Instead, the most typical verbal theme in preschoolers’ sad songs was the blocking of desired actions (i.e., not getting what you wanted), which appeared in half of the sad songs (n = 10) but in none of the happy songs, χ2(1, N = 40) = 13.33, p < .0001. Although friends, family, and related words (Mom, Dad) occurred in both happy and sad songs, their usage differed substantially. For example, happy songs featured family and friends together with the main character (Figure 2), χ2(1, N = 40) = 3.75, p = .05, in contrast to sad songs, which included no family or no friends, χ2(1, N = 40) = 2.06, p = .15 (see Table 1). Just as older children sometimes convey happi-ness or sadness through repetition of emotion words in their invented songs (Adachi & Trehub, 1999a), preschoolers did likewise. For example, one 5-year-old girl sang the following: I’m happy, I’m happy. My friends are happy, too. One 5-year-old boy began his sad song with similar word repetition and ended as follows: Sad words, sad words, sad words, sad. Sad words, sad words, sad words, word. The word was crying and the other one was laughing. The other one stole his ball and he started crying. And then he ran.


73

Some preschool children dramatized the emotional context by means of onomatopoeia and character dialogue: (1) I was sad from all the day. I cry from day to do today. Wah, wah, wah from all the way home… Wah, wah, wah, wah, wah, wah, wah, … (5-year-old boy); (2)… the dinosaurus Rex came and took that toy away. And then the elephant said, ‘No way ballface!’ And then the door came and slammed… (4-year-old boy). These story-telling strategies, which were used by 18% of the children to express happiness or sadness, were not evident in the invented songs of 6- to 12-year-olds (Adachi & Trehub, 1999a), which suggests that boundaries between songs and stories are not as clear-cut for younger children as for older children.

Melodies

The majority of children (75%, n = 30) gener-ated original melodies, with only 25% using familiar melodies, either keeping the entire song (e.g., Barney) intact (10%, n = 4), or borrowing tonal motives (e.g., ABC, Mary had a Little Lamb) for use

with original lyrics (15%, n = 6). For children who generated original melodies, most (55%, n = 22) combined original melodic and rhythmic elements, in contrast to a minority of children (20%, n = 8) whose creative emphasis was on rhythm. Happy melodies exhibited greater influence of the major mode while sad melodies exhibited greater influence of the minor mode, including reduced pitch range and compressed contour, χ2(2, N = 40) = 16.67, p < .0001 (Figure 3). Happy melo-dies tended to include more dotted rhythms than isochronous rhythms, with the reverse for sad melo-dies, χ2(2, N = 40) = 3.89, p = .14 (Figure 4). This tendency was confirmed statistically when rhythmi-cally unclear songs were removed from the analysis, χ2(1, N = 38) = 3.89, p = .049. The intended emotion was unrelated to beat regularity: 27.5% of songs (n = 11) were primarily regular, 37.5% (n = 15) were partially regular, and 35% (n = 14) were irregular. In previous research with older Canadian chil-dren (Adachi & Trehub, 1999a, 1999b), happy songs showed more influence of the major mode than did sad ones, but sad songs exhibited no such

Table 1Emotion themes in Canadian preschool children’s lyrics

Theme Example

Happy lyrics

candy, desserts

hopping, jumping, running, playing

friends, family

Sad lyrics

no friends, no family

blocking of desired actions/hope

*… One day, he gave me a chocolate bar too… *… it saw ice cream shop…

*… Happy bunny running in the forest…*… He hopped all the way to school…*Bunny rabbit, jump jump…*The dinosaur went out to play.

*… Bunny rabbit, bunny ba.. maked a new friend…*… ‘N I know that we are a family… I love you, mom, …

*… I have no friends, that’s why I’m lonely…*… She can’t find her mom or dad or brother or sister...

*… She was so sad because she didn’t get what she wants…


74

Figure 1. A sad song invented by a 5-year-old girl. The score, transcribed by the first author, was checked with a few musicians.

Figure 2. A happy song invented by a 4-year-old girl. The score, transcribed by the first author, was checked with a few musicians.


75

differentiation. In fact, only good 11- and 12-year-old singers clearly differentiated modality across happy and sad songs. Canadian preschoolers in the present study reduced the size of intervals or compressed the melodic contour for their sad songs, but there was little evidence that the minor mode was used consistently to convey sadness (see Figure 1). Twelve (30%) of the preschoolers’ songs were characterized by unclear mode. In four instances, the songs (two happy, two sad) sounded more like spoken narrations than like singing, resulting in unclear rhythm. A number of other instances sounded like chant (Moorhead & Pond, 1941) or rap, lacking a sense of melody but possess-ing a clear sense of rhythm.

exper i ment 2

The results of Experiment 1 revealed that many preschoolers could use their singing to communi-cate happiness and sadness. As was the case for older Canadian children (Adachi & Trehub, 1999a), the intended emotion of Canadian preschoolers’ songs was reflected in their lyrics. However, the predomi-nant association of sad feelings with thwarted goals (e.g., “having no sweets”) and isolation from family or friends was in marked contrast to the predom-inant death theme of older children’s sad songs (Adachi & Trehub, 1999a). Canadian preschoolers incorporated emotion-ally appropriate words and events in their happy

and sad lyrics. In addition, the use of rhythmic and melodic features differentiated their happy from their sad intentions. The purpose of the present experiment was to examine whether young Japanese preschoolers would respond in a similar manner.

Method

Participants

The participants in the present experiment were part of large cross-cultural investigations of children’s singing (e.g., Adachi, 2000). The pres-ent sample included 34 middle-class Japanese preschoolers 4 to 6 years of age. Only 15 of these 34 children, or 44.1%, produced materials for auditory analysis—7 boys (M = 4.43 years old, SD = .54) and 8 girls (M = 4.63 years old, SD = .74).

Apparatus and Procedure

Recordings were done in a quiet, unfamiliar room at the children’s preschool by means of a Sony DCR-PC1 digital video camera, a Sony TCD-D8 digital audio recorder, and a Sony ECM-MS957 microphone. The room was decorated with color drawings similar to those used in Experiment 1. Half of 34 children in the sample were asked to make up a happy song and half were asked to make up a sad song. All of the instructions, pictorial props, encouragement, and prompts were similar to those used in Experiment 1.

Figure 3. Number of happy and sad songs identified as major, minor, and unclear. Melodies with reduced range or compressed contour (flat sound) were classified as minor.

Figure 4. Number of happy and sad songs with dotted, isochronous, and unclear rhythms.


76

Content analysis of lyrics

The lyrics of each song were transcribed by a native Japanese speaker and confirmed by means of independent transcriptions of another native Japanese speaker. The coding categories for the lyrics from Experiment 1 were used in the present experiment. Inter-rater reliability for each coding category was evaluated by calculating Cohen’s kappa (Siegel & Castellan, Jr., 1988) between the first author and the independent rater who coded each category for three happy songs and two sad songs (i.e., 33% of all the lyrics) selected randomly. High reliabilities were confirmed for all coding cate-gories (κs = .90–1.0), and the first author’s coding was used for statistical analyses.

Content analysis of melodies

Digital recordings of children’s songs were converted to acoustical files by means of Protools Software (Digidesign). Two musicians who were familiar with young children’s singing listened to three happy and two sad tunes (i.e., 33% of all tunes) and determined that the categories from Experiment 1 could be coded reliably (κs = .65-1.0). The first author’s coding was used for statistical analyses.

Results and Discussion

Overall, only 44.1% (15 of 34) of Japanese preschool-ers produced any kind of song in response to the request for a made-up song that expressed happi-ness or sadness, with compliance being considerably higher for happy songs (11 of 17, or 64.5%) than for sad songs (4 of 17, or 23.5%). The proportion of sad songs produced by Japanese children (23.5%) differed significantly from the proportion produced by Canadian children (20 of 31, 64.5%), χ2(1, N = 48) = 7.38, p = .007, but the proportion of happy songs for Japanese (64.5%) and Canadian children (20 of 35, or 57.1%) did not differ, χ2(1, N = 52) = 0.60, p = .272. Of those Japanese children who produced happy or sad songs, only 20% (n = 3) created original lyrics (original/others), which consisted of story-like

phrases (e.g., portraying a sparrow or bunny) or simple word repetition (e.g., “Micky”). The remain-ing children (80%, n = 12) used the lyrics of familiar songs, with a small modification in one case. Japanese children’s tendency to use the lyrics of familiar songs rather than creating new lyrics approached conven-tional levels of significance, χ2(1, N = 15) = 3.07, p = .08. The distribution of original lyrics differed across emotion categories, with 2 of the 4 sad songs and only 1 of the 11 happy songs having original lyrics. This outcome contrasts markedly with the 90% incidence of original lyrics produced by Canadian children for their happy and sad songs. Unlike Canadian preschoolers’ songs, neither positive emotion words (e.g., daisuki [love], omoshi-roso [seemingly having fun], kirei [pretty], oishii [delicious]) nor positive emotion-evoking events (e.g., yume o kanaetekureru [a dream comes true], okashi o mitsuketa [found candy]) were reliably asso-ciated with happy intentions. However, none of the happy songs had negative emotion words, χ2(1, N = 11) = 11.00, p = .0009, or negative events, χ2(1, N = 11) = 11.00, p = .0009. By contrast, negative emotion words (e.g., kanashii [sad], naiteita [crying]) or events (e.g., “Suzume-chan ga tobenakunatte imashita [A spar-row could not fly]”) were observed in the sad songs. The differential use of negative lyrics was statisti-cally significant for words, χ2(1, N = 15) = 6.35, p = .012, and approached conventional levels of signifi-cance for events, χ2(1, N = 15) = 2.95, p = .086. Thus, Japanese preschoolers’ use of expressive verbal cues was apparent for sad songs, but not for happy songs, unlike Canadian preschoolers in Experiment 1, who used positive verbal cues for happy songs and nega-tive verbal cues for sad songs. Despite the use of similar visual prompts in the Canadian and Japanese contexts, the words “candy” and “playing” appeared consistently in Canadian preschoolers’ happy songs but in only 2 of the 11 Japanese preschoolers’ happy songs, where they were simply adopted from existing songs. Canadian preschoolers’ typical negative themes, such as the blocking of desired actions or the absence of family or friends, had no counterpart in Japanese preschoolers’ sad songs, which may be due to the small number of original lyrics produced by the Japanese sample.


77

Three children in the present experiment selected “Suzume no Kyodai [Sparrow Brothers],” as their happy song despite its lyrical theme of blocked hope, which pervaded the sad songs of Canadian preschoolers. In the song, the sparrow brothers converse: “What shall we become when we grow up?’ ‘I will become a hawk!’ ‘I will become a penguin!’ Chirp, chirp, chirp, ... But, even when they grow up [after all], sparrows are sparrows. Chirp, chirp.” This song, which was popular across age groups in Japan at the time of data collection (1999), had musical features associated with happiness such as the major mode and occasional staccato articulation. The song can be viewed as having conflicting cues to emotion because of blocked verbal desire (implying sadness) and happy melodic features. It is unclear whether the lyrics were noticeable and comprehensible to young listeners. Although there are suggestions that young children focus on verbal rather than melodic features when evaluating a speaker’s or singer’s feelings (Morton & Trehub, 2001, 2007), that may not always be the case. For example, few children or adults express concern about the plight of the three blind mice in the well-known children’s song (e.g., “She cut off their tails with a carving knife. Did you ever see such a sight in your life?”) It is diffi-cult to know whether singers of the sparrow song in the present experiment chose it because of melodic features typically associated with positive affect, pleasant memories associated with the song, or the song’s popularity and consequent accessibility. As they did with the lyrics, 80% (n = 12) of Japanese preschoolers used melodies from familiar songs. Only 20% (n = 3) generated original melo-dies, in contrast to 75% of the Canadian children. Comparing across emotion categories, 2 out of the 4 sad melodies produced by Japanese children were original, in contrast to only 1 of the 11 happy melodies. None of the Japanese children modi-fied the melodic features of familiar songs, unlike their Canadian counterparts. Japanese children’s tendency to use familiar melodies rather than generating entirely new melodies or engaging in improvisation with familiar melodies approached significance, χ2(1, N = 15) = 3.07, p = .08. The Japanese children’s happy melodies exhib-ited influences of the major mode (n = 8) or unclear

modality (n = 3), χ2(2, N = 11) = 8.91, p = .012, but no modal tendency was evident for their sad melodies. Unlike Canadian preschoolers’ songs in Experiment 1, Japanese preschoolers showed no differential use of rhythm across happy and sad melodies. Rather, the rhythms of Japanese preschoolers’ happy melo-dies sounded more isochronous than dotted or unclear, χ2(2, N = 11) = 12.18, p = .002, quite the oppo-site of that observed with Canadian preschoolers.

gener al discussion

The present study revealed Canadian and Japanese children’s ability to express feelings of happiness and sadness in their singing. However, their approach to the challenge of inventing an emotionally expressive song was starkly different. In response to requests to create a happy or sad song, over 60% of Canadian preschoolers produced some kind of song in contrast to roughly 44% of Japanese children. Although Canadian and Japanese children were similar in the proportion of happy songs produced, Japanese children produced a substantially smaller proportion of sad songs. Most Canadian preschool-ers expressed their emotional intentions by means of invented songs featuring novel lyrics, melodies, or both, but Japanese preschoolers did so largely by means of familiar melodies and lyrics. What accounts for Japanese children’s reluc-tance to produce sad songs and their limited creativity in the context of expressive singing? Although Japanese preschoolers are encouraged to be sensitive to the feelings and needs of others, there is less encouragement for expressing their feelings, especially negative ones, than is the case for American children (Hayashi et al., 2009; Yogo & Onoue, 1998; Zahn-Waxler, Friedman, Cole, Mizuta, & Hiruma, 1996). Japanese schoolchildren, for example, show less emotional expressiveness in their drawings of human figures than do American schoolchildren (La Voy et al., 2001). Japanese pref-erence for restraint in the expression of strong emotions, even for positive emotions, remains evident in adulthood (Safdar et al., 2009). Moreover, although Japanese children are more compliant than American children in the early


78

years of elementary school, tolerance for noncom-pliance at the preschool level seems to be greater in Japan than in North American (Abe & Izard, 1999). Finally, Japanese children’s inclination to sing familiar songs coupled with the prevalence of happy songs in the children’s repertoire may have made noncompliance seem like a suitable option. In short, a variety of factors could have contributed, separately or in combination, to the difference in complying with the request to create happy or sad songs. The frequent use of invented songs by Canadian preschoolers and their infrequent use by Japanese preschoolers raise the possibility of differences in musical creativity favoring young Canadian chil-dren. When Japanese first graders in a previous study were asked to make up a happy or sad song, 70% of them sang familiar songs (Adachi, 2001). Japanese preschoolers’ and first graders’ reliance on songs in their repertoire may be influenced by preschool curricula, which feature daily singing of familiar songs in the same key with piano accom-paniment (Trehub, Schellenberg, & Nakata, 2008) and weekly learning of new songs but no creative music-making (Adachi, 2001, in press). These expe-riences may carry the implicit message that music is something to be learned rather than created, interfering, perhaps, with the imaginative or diver-gent thinking that underlies creative music-making (Adachi & Chino, 2004). By contrast, musical utter-ances shared among playmates in North America (Campbell, 1998) and Australia (Whiteman, 2001) as well as rhymes and chants observed in play-grounds throughout the world (Opie & Opie, 1973) have their roots in free play and freely expressed exuberance. Such unbounded activity may enhance creative music-making both on and off the playground. Even though Japanese preschoolers also engage in exuberant playground singing, almost all of the songs produced are either standard songs with Western tonal influence or traditional songs with Japanese tonal influences (warabe-uta) that were invented by previous generations of Japanese chil-dren, with some improvisation of word rhythms (Fujita, 1997, 2000, 2001, 2002; Fujita & Nakakura, 1996; Ogawa & Imagawa, 2008). This is in sharp

contrast to the spontaneous songs observed in an Australian daycare center, the majority of which are invented (Whiteman, 2001). When creative music-making is introduced to the elementary-school curriculum in Japan (Adachi & Chino, 2004), it unleashes the creative musical potential of Japanese children. By the fourth grade, 75% of those Japanese children provide original songs when asked to do so (Adachi, 2000, 2001). Nevertheless, there are suggestions that the creative potential of Asian individuals is inhibited by prevailing cultural values of submissive individu-ality and conformance to group ideals (Ng, 2001). The idealization of conformance is evident in the well-known Japanese proverb, “the nail that sticks up gets hammered down.” Children are exposed to these cultural values in a variety of ways, includ-ing the lyrics of popular songs such as “Sparrow Brothers.” Being correct or doing the right thing in the eyes of adults may also be implicated (Ng, 2001). In that respect, choosing a familiar song would be less risky than creating one that might be considered inappropriate or substandard. Cultural pressures for conformity can be coun-tered, to some extent, by settings that nurture individuality or free expression. For example, Japanese first-graders who attend daycare programs that emphasize individual expression invent happy or sad songs twice as frequently as their peers in standard daycare settings (Adachi, 2004). In any case, the net result of Canadian preschoolers’ predominant use of invented songs and Japanese preschoolers’ predominant use of familiar songs was that Canadian children’s productions provided greater insight into their developing expressive abilities. On the surface, the happy songs of Canadian preschoolers showed conventional musical influ-ences such as the major mode and dotted or syncopated rhythms. While it is tempting to attribute implicit understanding of modality to preschool-ers because of influences of the major mode on happy songs and minor mode in the sad songs, it is premature to do so. For one thing, the catego-ries of major and minor were defined very broadly for use with a sample of preschoolers. Moreover, the sad melodies were characterized primarily by


79

reduced interval size or compressed contour rather than specific features of the minor mode. Adults also use compressed contour to convey sadness in their speech (Scherer, 1981), and Canadian chil-dren do so in their sad renditions of familiar songs (Adachi & Trehub, 1998). Finally, children do not associate major and minor modes with happy and sad emotions until about 6 years of age (Dalla Bella, Peretz, Rousseau, & Gosselin, 2001). With respect to timing, children’s differential use of dotted or syncopated rhythms for happy contexts may reflect associations between specific actions (e.g., skip-ping, jumping, dancing) and positive mood. The style of Canadian preschoolers’ invented songs ranged from storytelling-like to highly musi-cal. For example, the happy song of one 4-year-old (Figure 2) featured the major mode along with dotted or syncopated rhythms and affectively posi-tive lyrics. The sad song of one 5-year-old (Figure 1) manipulated modality from major to minor as the final theme of death was introduced. For the most part, however, sad songs featured compressed contours and affectively negative lyrics. In general, the melodies of happy and sad songs were inte-grated with the lyrics and exhibited some expressive devices characteristic of Western art music. The lyrics of Canadian preschoolers’ invented songs featured typical everyday experiences, much like those that appear in their stories (Saarni, 1999). In contrast to death as the primary sad theme for 6- to 12-year-old Canadian singers (Adachi & Trehub, 1999a), the primary sad theme of Canadian preschool singers was absent friends or family. Interestingly, happy themes were more consistent in preschoolers’ lyrics than in those of older chil-dren (Adachi & Trehub, 1999a), which may reflect older children’s increasing experience and verbal fluency. Japanese preschoolers’ overwhelming produc-tion of familiar songs (80%) rather than original songs made it difficult to assess their developing expressiveness. The features observed in their lyrics and melodies largely reflected the practices of composers and popular performers rather than the children’s own expressive abilities. Perhaps it is not surprising, then, that child-like emotional themes were largely absent from the lyrics.

Performances in a storytelling (i.e., speech-like) manner were uncommon both for Canadian and Japanese preschoolers, in contrast to the higher incidence reported by Umemoto (1999). Perhaps the necessity of generating melodies for prescribed texts in Umemoto’s study increased the prominence of the lyrics, promoting spoken narration. In sum, the present study revealed that preschool-ers are capable of expressing happy and sad intentions by means of singing, either by inventing songs, as Canadian children did, or by performing familiar songs, as Japanese children did. It remains to be determined whether the emotional intentions of such invented songs are interpretable to child and adult audiences. It would also be of interest to determine whether preschool children who fail to generate novel songs exhibit comparable inhibition in other creative realms such as drawing, fantasy play, and dancing.

r efer ences

Abe, J. S. A., & Izard, C. E. (1999). Compliance, noncompliance strategies, and the correlates of compliance in 5-year-old Japanese and American children. Social Development, 8, 1-20.

Adachi, M. (1994). The role of the adult in the child’s early musical socialization: A Vygotskian perspective. The Quarterly: Journal of Music Teaching & Learning, 5(3), 26-35.

Adachi, M. (2000). Nihonjin-jido no jodo-hyogen: Monogatari to kashi no naiyo kara [How do Japanese children express emotion in the stories and lyrics?]. Proceedings of the Fall Meeting of the Japanese Society for Music Perception and Cognition (pp. 65-68). Kyoto, Japan: JSMPC.

Adachi, M. (2001). Why can’t Japanese first grade children make up songs? In Y. Minami & M. Shinzanoh (Eds.), Proceedings of the Third Asia-Pacific Symposium on Music Education Research, Vol. II (pp. 49-50). Nagoya, Japan: Aichi University of Education.

Adachi, M. (2004). Ongaku de jiko-hyogen suru “me” o motte umareru kodomo-tachi [Children born with expressive potential in music]. Shoto-kyoiku Shiryo, 779, 56-59.

Adachi, M. (in press). Musically rich but creatively poor: A cautionary tale of music nurturing in Japanese preschools. In P. S. Campbell & T. Wiggins (Eds.), The Oxford handbook of children’s musical cultures. New York: Oxford University Press.


80

Adachi, M., & Chino, Y. (2004). Inspiring creativity through music. In S. Lau, A. N. N. Hui, & G. Y. C. Ng (Eds.), Creativity: When east meets west (pp. 305-340). Singapore: World Scientific.

Adachi, M., & Trehub, S. E. (1998). Children’s expression of emotion in song. Psychology of Music, 26, 133-153.

Adachi, M., & Trehub, S. E. (1999a). Children’s communication of emotion in song. In S. W. Yi (Ed.), Music, mind and science (pp. 454-465). Seoul: Seoul National University Press.

Adachi, M., & Trehub, S. E. (1999b). Musical emotion in the songs of Canadian children. Proceedings of the Fall Meeting of the Japanese Society for Music Perception and Cognition (pp. 69-74).

Adachi, M., & Trehub, S. E. (2003). Yoji no jisaku-uta ni mirareru jodo hyogen. [Preschoolers’ expression of emotion through invented songs]. Annual of Music Psychology and Therapy, 32, 3-10.

Barnlund, D. C. (1975). Public and private self in Japan and the United States. Tokyo: Simul Press.

Campbell, P. S. (1998). Songs in their heads: Music and its meaning in children’s lives. New York: Oxford University Press.

Davidson, L. (1985). Tonal structures of children’s early songs. Music Perception, 2, 361-374.

Davidson, L., McKernon, P., & Gardner, H. (1981). The acquisition of song: A developmental approach. In Documentary report of the Ann Arbor Symposium: Applications of psychology to the teaching and learning of music (pp. 301-315). Renton, VA: Music Educators National Conference.

Davidson, L., & Scripp, L. (1994). Conditions of giftedness: Musical development in the preschool and early elementary years. In R. F. Subotnik & K. D. Arnold (Eds.), Beyond Terman: Contemporary longitudinal studies of giftedness and talent (pp. 155-185). Norwood, NJ: Ablex.

Della Bella, S., Peretz, I., Rousseau, L., & Gosselin, N. (2001). A developmental study of the affective value of tempo and mode in music. Cognition, 80, B1-B10.

Doi, T. (1986). The anatomy of the self: The individual versus society. Tokyo: Kodansha.

Dowling, W. J. (1984). Development of musical schemata in children’s spontaneous singing. In W. R. Crozier & A. J. Chapman (Eds.), Cognitive processes in the perception of art (pp. 145-163). Amsterdam: North-Holland.

Fujita, F. (1997). Youji wa seikatsu no naka de ongakuteki na mono o dou mochiiteiruka [How do young children use something musical in their own lives?]. Proceedings of the 1997 Meeting of Japan Society of Research on Early Childhood Care and Education (pp. 224-225). Tokyo: Japan Society of Research on Early Childhood Care and Education.

Fujita, F. (2000). Kotoba o dousa o ongakuteki ni matomeru: Hoikuen 1-saiji-kurasu no kodomotachi no ongakukoudou no kansatsu-kiroku kara [Generating words and actions in musical ways: From field notes of musical behaviours in a 1-year-old classroom at a nursery school]. Proceedings of the 2000 Meeting of Japan Society of Research on Early Childhood Care and Education (pp. 272-273). Tokyo: Japan Society of Research on Early Childhood Care and Education.

Fujita, F. (2001). Iki o awasete rizumikaru na outou o tanoshimu: Hoikuen 2-saiji-kurasu no kodomotachi no ongakukoudou no kansatsu-kiroku kara [Enjoying rhythmical responses in unison: From field notes of musical behaviours in a 2-year-old classroom at a nursery school]. Proceedings of the 2001 Meeting of Japan Society of Research on Early Childhood Care and Education (pp. 530-531). Tokyo: Japan Society of Research on Early Childhood Care and Education.

Fujita, F. (2002). Rizumikaru na kotoba no sousaku o tanoshimu: Hoikuen 3-saiji kurasu no kodomotachi no ongakukoudou no kansatsu-kiroku kara [Enjoying making up rhythmical words: From field notes of musical behaviours in a 3-year-old classroom at a nursery school]. Proceedings of the 2002 Meeting of Japan Society of Research on Early Childhood Care and Education (pp. 76-77). Tokyo: Japan Society of Research on Early Childhood Care and Education.

Fujita, F., & Nakakura, M. (1996). Kodomo ga utaihajimeru toki: Hoikuen 1-saiji-kurasu no ongaku-koudou no kansatsu kara [When children begin singing: From observations of musical behaviours in a 1-year-old classroom at a nursery school]. Proceedings of the 1996 Meeting of Japan Society of Research on Early Childhood Care and Education (pp. 74-75). Tokyo: Japan Society of Research on Early Childhood Care and Education.

Gabrielsson, A., & Juslin, P. N. (1996). Emotion expression in music performance: Between the performer’s intention and the listener’s experience. Psychology of Music, 24, 68-91.

Hayashi, A., Karasawa, M., & Tobin, J. (2009). The Japanese preschool’s pedagogy of feeling: Cultural strategies for supporting young children’s emotional development. Ethos, 37, 32-49.

Juslin, P. N. (1997). Emotional communication in music performance: A functionalist perspective and some data. Music Perception, 14, 383-418.

Juslin, P. N. (2001). Communicating emotion in music performance: A review and a theoretical framework. In P. N. Juslin & J. A. Sloboda (Eds.), Music and emotion (pp. 309-337). Oxford, UK: Oxford University Press.

Juslin, P. N., & Sloboda, J. A. (Eds.). (2001). Music and emotion. Oxford, UK: Oxford University Press.


81

Kelley, L., & Sutton-Smith, B. (1987). A study of infant musical productivity. In J. C. Peery, I. W. Peery, & T. W. Draper (Eds.), Music and child development (pp. 35-53). New York: Springer-Verlag.

McKernon, P. E. (1979). The development of first songs in young children. In D. Wolf (Ed.), Early symbolization (pp. 43-58). San Francisco, CA: Jossey-Bass.

Moog, H. (1976). The musical experience of the pre-school child (C. Clarke, Trans.). London: Schott. (Original work published in German in 1968)

La Voy, S. K., Pedersen, W. C., Reitz, J. M., Brauch, A. A., Luxenberg, T. M., & Nofsinger, C. C. (2001). Children’s drawings: A cross-cultural analysis from Japan and the United States. School Psychology International, 22, 53-63.

Moorhead, G. E., & Pond, E. (1941). Music for young children: I. Chant. (Pillsbury Foundation Studies). Santa Barbara, CA: Pillsbury Foundation for Advancement of Music Education.

Moorhead, G. E., & Pond, E. (1942). Music for young children: II. General observations. (Pillsbury Foundation Studies). Santa Barbara, CA: Pillsbury Foundation for Advancement of Music Education.

Morton, B., & Trehub, S. E. (2001). Children’s understanding of emotion in speech. Child Development, 72, 834-843.

Morton, B., & Trehub, S. E. (2007). Children’s judgments of emotion in song. Psychology of Music, 35, 629-639.

Ng, A. K. (2001). Why Asians are less creative than Westerners. Singapore: Prentice Hall.

Ogawa, Y., & Imagawa, K. (2008). Ongaku-suru kodomo o tsukamaetai: Jikken-kenkyusha to fieldworker no taiwa [How to capture musicking children through the eyes of experimentalists and fieldworkers]. Okayama, Japan: Fukurou Shuppan.

Ohgushi, K., & Hattori, M. (1996). Emotional communication in performance of vocal music. In B. Pennycook, & E. Costa-Giomi (Eds.), Proceedings of the Fourth International Conference on Music Perception and Cognition (pp. 269-274). Montreal: McGill University.

Omi, A. (1994). Children’s spontaneous singing: Four song types and their musical devices. Journal of Kawamura Gakuen Woman’s University, 5(2), 61-76.

Opie, I, & Opie, P. (1973). The lore and language of school-children. Oxford, UK: Oxford University Press.

Pittam, J., Gallois, C., Iwawaki, S., & Kroonenberg, P. (1995). Australian and Japanese concepts of expressive behavior. Journal of Cross-Cultural Psychology, 26, 451-473.

Saarni, C. (1999). The development of emotional competence. New York, NY: Guilford Press.

Safdar, S., Friedlmeier, W., Matsumoto, D., Yoo, S. H., Kwantes, C. T., Kakai, H., & Shigemasu, E. (2009). Variations of emotional display rules within and across cultures: A comparison between Canada, USA,

and Japan. Canadian Journal of Behavioural Science, 41, 1-10.

Scherer, K. R. (1981). Speech and emotional states. In J. K. Darby, Jr. (Ed.), Speech evaluation in psychiatry (pp. 189-220). New York: Grune & Stratton.

Senju, M., & Ohgushi, K. (1987). How are the player’s ideas conveyed to the audience? Music Perception, 4, 311-324.

Siegel, S., & Castellan, Jr., N. J. (1988). Nonparametric statistics for the behavioral sciences (2nd ed.). New York: McGraw-Hill.

Stephan, C. W., Stephan, W. G., Saito, I., & Barnett, S. M. (1998). Emotional expression in Japan and the United States: The nonmonolithic nature of individualism and collectivism. Journal of Cross-Cultural Psychology, 29, 728-748.

Trehub, S. E., Schellenberg, E. G., & Nakata, T. (2008). Cross-cultural perspectives on pitch memory. Journal of Experimental Child Psychology, 100, 40-52.

Umemoto, T. (1994). Children’s knowledge about song. In I. Deliege (Ed.), Proceedings of the 3rd International Conference on Music Perception and Cognition (pp. 121-122). Liège, Belgium: ESCOM.

Umemoto, T. (1999). Kodomo to ongaku [The child and music]. Tokyo: University of Tokyo Press.

Umezawa, Y. (1990). Ongakuteki-taiwa to shite no “kyoen”: Hahaoya to kodomo no “kyoen” o tegakarini [“Co-performance” as a musical communication: A case of co-performance between a mother and her child]. Aichi University of Education Technical Report, 39, 1-15.

Whiteman, P. J. (2001). How the bananas got their pyjamas: A study of the metamorphosis of preschoolers’ spontaneous singing as viewed through Vygotsky’s Zone of Proximal Development (Unpublished doctoral dissertation). University of New South Wales, Australia.

Yamasaki, T. (2002). Emotional communication in improvised performance by musically untrained players. In Kato (Ed.), Proceedings of the 17th Congress of the International Association of Empirical Aesthetics (pp. 521-524). Osaka, Japan: IAEA.

Yamasaki, T. (2006). Youji ni yoru ongaku-ensou o tsuujita kanjouteki-imi no dentatsu [Communication of emotional intention through music performance by young children]. Journal of Music Perception and Cognition, 12, 1-14.

Yogo, M., & Onoue, K. (1998). Social sharing of emotion in a Japanese sample. In A. Fischer (Ed.), ISRE98’: Proceedings of the Conference of the International Society for Research on Emotions (pp. 335-340). Amsterdam: The International Society for Research on Emotions.

Zahn-Waxler, C., Friedman, R. J., Cole, P. M., Mizuta, I., & Hiruma, N. (1996). Japanese and United States preschool children’s responses to conflict and distress. Child Development, 67, 2462-2477.


82

Mayumi adachi worked as a piano teacher after receiving her B.A. in music education from Niigata University. She studied piano pedagogy at Teachers College at Columbia University, where she completed her M.A. and Ed.M. in music and music educa-tion while studying

psychology. She received her Ph.D. in psychomu-sicology from the University of Washington, and worked as a postdoctoral fellow at the University of Toronto at Mississauga. Her interdisciplinary background gave her flexibility in her career, teach-ing music education (Yamanashi University) and psychology (current), and conducting quantitative and qualitative research on a variety of phenom-ena surrounding the development and learning of music. She chaired ICMPC10 in 2008, archived at http://www.jsmpc.info/archives/icmpc10archive/index.html.

Sandra Trehub, who obtained her doctor-ate from McGill University in 1973, is Professor Emeritus of Psychology at the University of Toronto and Visiting Professor at the International Laboratory for Brain, Music and Sound Research (BRAMS) at the Université de Montréal. She has published exten-sively on infants’ perception of music and parents’ songs to infants. Although most of her research is conducted in her laboratory, she sometimes travels to remote regions of the world to observe mothers’ musical interactions with infants.

Mayumi Adachi Sandra E. Trehub

au t hor note s

biogr aph ies

Funding for data collection was provided by a grant from Social Sciences and Humanities Research Council (SSHRC) of Canada to the second author. The authors also acknowledge the assistance and encouragement of AIRS, which is also funded by SSHRC. Portions of Experiment 1 were reported in a Japanese-language jour-nal with limited circulation (Adachi & Trehub,

2003). We thank Jane Campbell (Experiment 1) and Miho Tanaka (Experiment 2) for assistance with data collection. We also thank Loretta Falco and Lori Jiang (Experiment 1) as well as Japanese students (Experiment 2) for transcription of the lyrics and coding for reliability. Finally, we thank Gary Vasseur for his helpful comments on the manuscript.

Canadian and Japanese preschoolers' creation of happy and sad

Documents

Transcript of Canadian and Japanese preschoolers' creation of happy and sad