Exploring the Transition Phenomenon of “Song and Musical Play” in Infants

ABSTRACT There is a lack of research regarding the use of the still-face procedure to assess infants’ comfort and reengagement in stressful situations. While previous studies have already investigated the transitional phenomenon based on Winnicott’s theory, this study uses “stuffed toy” and “song and musical play” with the still-face procedure. This study aimed to examine the transition phenomenon of infants in “song and musical play” and provide a new perspective regarding psychological support for infants. Infant responses and maternal availabilities were also analyzed. The study participants were 32 infants (M age = 17.97 months, SD = 1.43) and their mothers (M age = 32.28 years, SD = 2.96). The results showed that infants were receptive to the “stuffed toy” and displayed positive responses. Additionally, the appearance ratios of “song and musical play” imitations increased. This suggests that the infants used “song and musical play” to manage and control their affect in stressful situations. Furthermore, when the mothers exhibited maternal availability as an emotional rhythm, the infants’ responses became positive. Therefore, the infants demonstrated the transitional phenomenon to symbolize inner objects for social participation using “song and musical play” as the transitional object.

In the still-face experiment (Tronick et al., 1978, p. 3), there are episodes in which the infant and the parent interact, the parent becomes still-faced, and then the two interact again. Still-face is performed during the still-face episode of this experiment, wherein the infant and parent lack responsive or tuned affective interaction. In the still-face episode, infants try to initiate social dialogue with the parent by staring and raising their voices in an attempt to change their parent's still-face into an expression of affective interaction. However, they give up when they are not able to interact with their parents. The effect of the still-face continues in the form of negative interactions in the episodes that occur after the still-face (carry-over effect; Tronick et al., 1978, p. 11). This occurs because the infant perceives a lack of parental availability in the context of affective interaction, which violates the expectations of mutual emotion sharing (Legerstee, 2005, p. 92, p. 140). It has been suggested that the negative effects of the still-face condition in a parent-infant relationship may cause the infant to have an ambivalent response or form a false self (Winnicott, 1958, Chapter 15;Stern, 1995, Chapter 6).
For such parents and infants, Stern (1995, Chapter 6) provides a clinical intervention through affective attunement, which makes the emotional experience of parents and infants an "affective shape form" that attunes with the flow of time, much like experiencing music. According to Eckerdal and Merker (2009, pp. 254-255), in daily life, "song and musical play" allows parents to intuitively interact with their infants' sensitivity to music and calm them when they are anxious. In addition, according to Winnicott (1971a, Chapter 5), musical playing by infants is a transitional phenomenon (Winnicott, 1958, Chapter 16;Gronlnick, 1990, Chapter 5) essential for their emotional development. Winnicott (1971a, Chapter 3) refers to the state in which infants use transitional objects as the "transitional phenomena." Playing is a transitional phenomenon that involves sharing with others, and thus, develops into a cultural experience. A transitional object is a soft object, such as a blanket, which symbolizes some part-object, such as the mother's breast for infants (Winnicott, 1958, Chapter 16). According to Winnicott (1958, Chapter 16), infants have the illusion that their mother is a part of them. An infant uses transitional objects to relieve stress caused by the disillusionment process, which leads to accepting the reality that the mother is not a part of oneself and then comforting, and reassuring oneself (Winnicott, 1958, Chapter 16). As the transitional phenomenon evolves into playing, the infant plays with the mother, alternating between subjective and external reality, depending on maternal availability. In the process, the infant accepts the mother as an external reality, but the mother's representation (object) inside them becomes disjointed. Further, when the mother involves herself in the infant's play, the infant may or may not accept participation. Infants have their own rhythms and timbres in their facial expressions and movements, so when the parent sympathizes with an acceptable tone, the infant forms a true self while accepting the external mother (Winnicott, 1958, Chapter 17;Gronlnick, 1990, Chapter 5).
However, parents do not always communicate musically with their infant's emotions (Stern, 1995, Chapter 6). Moreover, clinical intervention may be required if the infant and parent continue to lack interaction, as in the still-face experiment. "Song and musical play" is part of a support program for infants and their parents who require psychological therapy (Ogane, 2009, p. 165). Additionally, musical communication between infants and their parents has been studied for developmental effects (Malloch & Trevarthen, 2009, Chapter 1). However, empirical studies are required to determine the effects of "Song and musical play" (Eckerdal & Merker, 2009, Chapter 11).
For these reasons, it is expected that song and musical play will not have the carry-over effect that occurs after still-face. It is thought that song and musical play in the stressful negative situation of an infant frowning reduces negative emotions. This suggests that song and musical play have become transitional phenomena for infants. Thus, it is hypothesized that infants can regulate their own negative emotions in stressful situations through the transitional phenomenon of song and musical play. To prove this hypothesis, this study introduces "song and musical play" into the still-face procedure and quantitatively analyzes the facial expressions and movements of infants.
According to the studies dealing with the "emergence of musical expressiveness" (Eckerdal & Merker, 2009, p. 248), "intention imitation" (Meltzoff, 1995, p. 847), "expecting to enjoy the interaction" (Legerstee, 2005, p. 152), "understanding of intentions in others" (Legerstee, 2005, pp. 180-181), "schema of being with another" (Stern, 1985, Chapter 5), and "joint attention" (Tomasello, 1995, p. 108), infants around the age of 18 months can cope with anxiety caused by stressful situations while imitating "song and musical play" and show expectations and intentions for affective interaction with their parents. It is also assumed that 18-month-old infants can relate to the "song and musical play" of objects that are not their parents in stressful situations and use them to play. Infants can also positively perceive, relate to, and own stuffed toys or other soft objects presented. In both cases, the infant can exercise controls using the outer objects that are not part of them. According to Winnicott (1971a, Chapter 3), such infant interaction, that is, playing, is a transient phenomenon that leads to cultural experiences.
Therefore, in this study, we provided a stuffed toy to the infants during the still-face procedure that introduced the "song and musical play." If the infant receives the stuffed toy at the beginning of the still-face experiment, the toy, rather than the parents, may be used to control emotions. It is predicted that at that time, there will be no carry-over effect for infants because the positive expressions of infants will increase and the gaze focused on the experimenter who plays the "song and musical play" will intensify. In addition, it is assumed that infants who do not receive the stuffed toy may feel anxious about their relationship with the objects and cry for help from their parents (Winnicott, 1964, Chapter 9). At that time, it is expected that the infant will depict a negative expression when gazing at the nearby objects and the parent, and a carry-over effect will be observed.
In the still-face procedure of Egami et al. (2008, pp. 154-157), infants sat on the parent's knee and interacted with the experimenter; the results were, similar to those of Tronick et al. (1978, p. 11). It seems that the influence of the difference in the method of "song and musical play" by the parent can be controlled by the experimenter.' However, placing the infant on a parent‛s lap, should confirm the effect of the parents.
In this study, the stuffed toy and "song and musical play" were included as part of the still-face procedure. The experimenter introduced "song and musical play," and the responses of the infants and the parents were quantitatively analyzed. Thus, the transition phenomenon of infants in "song and musical play" was analyzed and a new perspective on psychological support for infants is offered.

Participants
We distributed information on research collaboration at four childcare support centers in city A; 41 mothers offered to participate. The contents and method of this study were explained to the participants, and verbal and written consent for cooperation was obtained. Except for nine mothers and their infants who were excluded from participation due to excessive crying, 32 infants (experimental month of age; M = 17.97, SD = 1.43) and their mothers (experimental year of age; M = 32.28, SD = 2.96) were included in the study. The infant participants comprised 19 boys and 13 girls. The mothers confirmed that the infants were not diagnosed with any developmental disorders, and that they were growing steadily.

Procedures
The experiment was conducted in August 2019 in the counseling room of A City's Child Care Support Center (Aichi Pref., Japan). The experimenter was in their early 60s, with more than 20 years of experience in clinical psychological support for infants and their parents in nursery schools as well as the appropriate skills and abilities necessary for this study. The mother and infant entered the counseling room and sat, about 50 cm away from and, facing the experimenter. Initially, after entering, the infant was seated on their mother's knee. Then, the experimenter looked at the infant, called the infant's name, showed them the stuffed toy, encouraged them to pick it up and play, and talked about how to play with them.
First, the experimenter gazed at the infant and engaged in "song and musical play" for 30 seconds ("song and musical play" episode, hereinafter called "SAMP"), then the experimenter gazed at the infant with a neutral expression for 30 seconds (still-face episode, hereinafter called "SF"). Finally, the experimenter gazed at the infant and a different "song and musical play" was performed for 30 seconds (re-song and musical play episode, hereinafter called "re-SAMP"). Different "song and musical plays" were performed in the SAMP and the re-SAMP to control the influence of habituation. For the "song and musical play," popular songs were selected from infant nursing sites based on the survey by Wakaya, (2018). ("song and musical play" of SAMP, see Figure S1; re-SAMP, see Figure S2).
In all episodes, one webcam was installed between the infant and the experimenter and, one was installed on each side of the infant and the mother; the Otacle Motion Analysis OTL8 system synchronized the three webcams, recorded, and saved them as video data in a period of two milli-seconds. The video data for the three cameras were played back synchronously, and the counter value was displayed. In all episodes to be analyzed, the counter values at the beginning and the end of the behaviors of the infant and mother were recorded, and the appearance rate (%) of each behavior was calculated. To examine the reliability of the data to be analyzed, 12 data sets, representing 19.35% of the 64 data sets including the infants and their mothers, were randomly selected and coded by two coders (a certified public psychologist and a clinical psychologist). When the results of coding by one of the authors and the two coders were used to calculate the intraclass correlation coefficient for each behavior in each of the three episodes, a high agreement rate was confirmed between intraclass correlations (ICC) of .96 and .98.

Ethical treatment of study participants
The study was conducted in accordance with the ethical standards of the American Psychological Association. This study received the approval of the Institutional Review Board on "Studies on Humans" at [BLINDED FOR REVIEW]. Verbal and written consent from the 41 mothers, who offered to participate, was obtained.

Response to infants' stuffed toy
We classified infants as either receptive (pick up, etc.) or non-receptive (does not see, pick up, or throw, and so on) based on their response to the stuffed toy.

Infant's expression
The expression of each infant during the three episodes (SAMP, SF, and re-SAMP) were classified into four types: positive expressions (smiling, opening, laughing, and the like), negative expressions (angry, crying, sad, dissatisfaction, and others), neutral expressions (expressionless), and indistinguishable states. The ratio of expressions in each episode was calculated.

Infant's gaze
Each infant's gaze during the three episodes (SAMP, SF, and re-SAMP) was classified into four types: experimenter, mother, near objects, and indistinguishable state. The ratio of the gaze in each episode was calculated.

Imitation of infants' "song and musical play"
For the three episodes (SAMP, SF, and re-SAMP), the percentage of each episode in which the infants imitated the "song and musical play" was calculated.

Maternal availability
Regarding the maternal availability in the three episodes (SAMP, SF, and re-SAMP), the ratio of the appearance of engaging in "song and musical play," calling, smiling, lulling, and holding in each episode was calculated.

Infants' responses to the stuffed toy
In total, 22 (68.75%) infants, including 15 boys and 7 girls had a receptive response to the stuffed toy, with an average age of 17.82 months (SD = 1.33). Altogether, 10 infants (31.25%), including four boys and six girls, had a non-receptive response, with an average age of 18.22 months (SD = 1.72).

Infants' expression
Descriptive statistics for the appearance rate of each expression in the receptive and non-receptive responses to the stuffed toy in the three episodes are shown in Table 1. The appearance rate of neutral expressions for the stuffed toy was high for both receptive responses (M = 63.45, SD = 37.67) and nonreceptive responses (M = 56.94, SD = 35.53). The Kruskal-Wallis tests used to compare the expressions in each episode showed that the results were not significant. The Brunner-Munzel test was used to compare the appearance rates of each expression in infants who responded receptively and nonreceptively to the stuffed toy, as well as those who did not respond receptively to the stuffed toy. In the re-SAMP episode, infants with receptive responses to the stuffed toy displayed more positive expressions than those with non-receptive responses (W = 1.88, df = 20.19, p < .10) (Figure 1). Infants with receptive responses to the stuffed toy displayed fewer negative expressions in the re-SAMP episode (M = 11.84, SD = 29.99) than in the SAMP episode (M = 19.41, SD = 36.77), with no carryover effect. Infants with non-receptive responses to the stuffed toy displayed more negative expressions in the re-SAMP episode (M = 25.09, SD = 33.45) than in the SAMP episode (M = 24.19, SD = 23.38), showing a carryover effect. However, the Kruskal-Wallis tests for these two results were not significant.

Infant's gaze
The descriptive statistics of the appearance rate of each gaze in the receptive and non-receptive responses to the stuffed toy in the three episodes are shown in Table 1. The infants' gaze was often directed at the experimenter, for both receptive (M = 50.36, SD = 36.80) and non-receptive (M = 48.51, SD = 30.05) responses to the stuffed toy. As a result of the Kruskal-Wallis test, a main effect was observed in the infants' gaze toward the experimenter in the episodes of the receptive response to the stuffed toy (χ2 (2) = 17.89, p < .001). According to the Steel-Dwass test for multiple comparison, the receptive response to the stuffed toy showed that infants' gaze toward the experimenter was more notable in the re-SAMP and SAMP episodes than in the SF episode ( Figure 2). Furthermore, a main effect was observed in the infants' gaze toward the experimenter in the non-receptive response to the stuffed toy (χ2(2) = 12.24, p < .01). The Steel Dwass test for multiple comparisons revealed that the infants' gaze, in this case, was more notable for the re-SAMP episode than the SF episode. A main effect was observed in the infants' gaze toward near objects in the episodes of the receptive response (χ2(2) = 22.48, p < .001) and the non-receptive response (χ2(2) = 5.85, p < .10) to the stuffed toy. The Steel-Dwass test for multiple comparisons revealed that the intensity of the infants' gaze toward the near objects in the receptive and non-receptive response was greater in the SF episode than in the re-SAMP and SAMP episodes (Figure 3). In the re-SAMP episode, the infants' gaze toward the experimenter in both receptive and non-receptive responses to the stuffed toy increased, but the Kruskal-Wallis test results were not significant. Regardless of the receptive and non-receptive responses to the stuffed toy, the infants' gaze toward the experimenter engaging in "song and musical play" in the re-SAMP episode was more notable than in the other episodes.

Imitation of infant's "song and musical play"
Descriptive statistics on the proportion of "song and musical play" imitations that appeared in the three episodes of receptive and non-receptive infants' (regarding their responses to the stuffed toy) are shown in Table 2. As a result of the Kruskal-Wallis test, a main effect was observed in the imitation of "song and musical play" in the episodes of receptive response to the stuffed toy (χ2(2) = 9.10, p < .05) (Figure 4). A Steel-Dwass multiple comparisons test revealed that, infants with receptive responses to stuffed toys had a higher imitation rate of "song and musical play" in the re-SAMP and SAMP episodes Figure 1. Appearance rate of infants' positive expressions in receptive/non-receptive response to the stuffed toy. Note. SAMP, "song and musical play" episode; SF, still-face episode; re-SAMP, re-"song and musical play" episode; +p < .10.
than in the SF episode. In the re-SAMP episode, the Brunner-Munzel test showed that the appearance rate of the "song and musical play" imitation of the response to the stuffed toy was higher for the receptive response to the stuffed toy than for the non-receptive response (W = 1.81, df = 76.57, p < .10). Infants who received stuffed toys had an increased imitation rate of "song and musical play" in the re-SAMP episode; thus, a carry-over effect did not occur. Table 3 shows the descriptive statistics for maternal availability to the infants regarding the average appearance rate of the receptive and non-receptive responses to the stuffed toy in each episode. The Figure 3. Appearance rate of infants' gazing at nearby objects in receptive/non-receptive response to the stuffed toy. Note. SAMP, "song and musical play" episode; SF, still-face episode; re-SAMP, re-"song and musical play" episode; *p < .05; ***p < .001. SAMP, song and musical play episode; SF, still-face episode; re-SAMP, re-song and musical play episode; M, mean; SD, standard deviation.
Kruskal-Wallis test showed that "song and musical play" had a main effect in the episodes of the infant's receptive response to the stuffed toy (χ2(2) = 12.24, p < .001) and non-receptive response (χ2(2) = 4.47, p < .10). The Steel-Dwass multiple comparison test showed that, when the infant had a receptive response to the stuffed toy, the appearance rate of "song and musical play" was higher in the re-SAMP and SAMP episodes than in the SF episode. Furthermore, in the case of non-receptive responses to the stuffed toy, the appearance rate of "song and musical play" tended to be higher in the re-SAMP episode than in the SF episode. The Kruskal-Wallis test showed a main effect on the appearance rate of "lulling the infant" in the episode of the receptive response to the stuffed toy (χ2(2) = 6.60, p < .05). The Steel-Dwass multiple comparisons test showed that the receptive response to the stuffed toy tended to be higher in the SF episode than in the SAMP episode.

Infants with a receptive response to stuffed toys
Infants predominantly had a neutral expression in every episode, and their gaze was often directed at the experimenter in the SAMP and re-SAMP episodes, indicating that the infants were observing the experimenter. In the re-SAMP episode, the experimenter was gazed at with a more positive expression than in the other episodes, and the appearance rate of "song and musical play" imitation was high, suggesting that the "song and musical play" was perceived as positive. While participating in the stressful situations of the still-face experiment, infants who held the stuffed toy showed interest in and positive impressions toward the stuffed toy and "song and musical play," indicating that they were coping with their stress. This is related to an infant's strategy of dealing with stress by using an object that can generate positive emotions in stressful situations. However, the infants' gaze in the SF episode was often directed at near objects, indicating that they were coping with their stress. Infants were only partially positive and needed their mothers as well as stuffed toy to cope with stress. These results suggest that the infants may have used the stuffed toy as a transition object. In the re-SAMP episodes, the positive expressions and gaze toward the experimenter were more notable than in the other episodes, and the increase in "song and musical play" imitation indicated that the negative emotional carry-over effect after the SF episode did not occur. This shows that infants control their emotions in stressful situations. The experimenter's first introduction to "song and musical play" is unfamiliar to the infant and was offered to infants as something that facilitated development. It was to reflect the mother's expectations and intentions for her infant to behave that way. Infants satisfy the expectations and intentions of interacting with their mothers by referring to their mother's expectations and intentions and imitating "song and musical play" in a tripartite relationship with the experimenter. Rather than depending on their mother to cope with stress, infants accepted "songs and musical play" SAMP, song and musical play episode; SF, still-face episode; re-SAMP, re-song and musical play episode; M, mean; SD, standard deviation; *p < .05; **p < .01.
that contained their mother's expectations and intentions. This suggests that "song and musical play" is a transition phenomenon for infants.

Infants with a non-receptive response to stuffed toys
Infants with a non-receptive response to the stuffed toys gazed at the experimenter during the re-SAMP and SAMP episodes; however, positive expressions and "song and musical play" imitations in the re-SAMP episode were less notable than receptive responses. It was suggested that infants inferred the intention of social behavior to participate in the "song and musical play" by socially referencing the state of the experimenter, but did not use the "song and musical play" to transform the negative emotions caused by their stress into positive ones. As the negative expressions did not decrease and the positive ones were lesser in the infants' receptive responses, it can be inferred that the infants could not control their emotions by themselves. It is probable that the infant's gaze was more focused on the objects during the SAMP and re-SAMP episodes, and more focused on the mother during the SF episode. The infants were coping with stress by gazing at the mother to get their attention or engaging in distracting behavior. This suggests that infants who showed non-receptive responses to the stuffed toy need their mothers to cope with stress.

Being with others and the transition phenomenon
This study's results support that the infants inferred the intention from "song and musical play" through social referencing. When the infants were on their mothers' knees, they were able to monitor their mother's condition without looking at them. Therefore, the behavior accompanying their mother's expectation and intention to be positive for the infant was transmitted to the infant's body as an emotional rhythm and affected their emotions. The infant then gazed at the experimenter playing "song and musical play" while monitoring the mother, who was with them during the stressful situation, which indicated their mother's expectations and intentions for the "song and musical play." In addition, the increase in the positive expressions in the re-SAMP episode after the SF episode indicates that the "song and musical play" directed the infants toward positive emotions. Furthermore, the infants imitated "song and musical play" when they perceived it as positive, even if their expectations of sharing emotions with their mothers could not be fulfilled. In such cases, infants will be able to hold a cohesive perception of their disjointed presence in stressful situations and be able to be with others. Therefore, if the mother is unable to tune into the tone of the infant, "song and musical play" is thought to have a similar clinical intervention effect as a transition phenomenon. Furthermore, infants' emotional development is promoted with the process of cultural experience through "song and musical play."

Conclusion
The pool of existing literature on "song and musical play" does not examine the efficacy of interventions using psychological experiments. This study confirms that "song and musical play," as a clinical intervention, tunes the emotional bond between infants and their parents, as well as promotes their emotional development, and provides a cultural experience to the infants. Therefore, the infants displayed the transitional phenomenon to symbolize inner objects for social participation using "song and musical play" as the transitional object. They acted to reflect their mother's expectations and intentions and used "song and musical play" to deal with stressful situations between subjective and external realities, as well as to feel their selves true. A transition phenomenon was observed in which the infants controlled their emotions by playing with "song and musical play" and demonstrated positive expressions. According to the results of this experiment, the infant's abilities, such as cognition and sociality, and the transition process of internal representations mutually influence each other. In addition, interpreting psychoanalytic theory about infants from empirical experiments motivates scientific evidence-based clinical interventions, expands the understanding of the infant mind, and contributes to psychoanalytic support. Further, for infants around 18 months of age, the experience of playing with others through communicative musicality as "song and musical play" becomes a transient phenomenon, encouraging them to participate in cultural activities. However, basic research on "song and musical play" is required, such as what factors affect "song and musical play" in infants. Accordingly, it is necessary to conduct developmental research on the characteristics of age-specific responses. Moreover, how "song and musical play" experiences the transition phenomenon by infants of each age depends largely on interpretation; therefore, the interpretation method must be explored.

Author contributions
All authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by Hironobu Ogane and Kaname Ogane. The first draft of the manuscript was written by Hironobu Ogane and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.