Preterm mother-infant interaction: speaking and humming during kangaroo care

1 Universidade do Minho Escola de Psicologia Fernanda Lisboa Guimarães Preterm Mother-Infant Interaction: Speaking and Humming during Kangaroo Care Preterm Mother-Infant Interaction: Speaking and Humming during Kangaroo Care Fernanda Lisboa Guimarães Uminho | 2023 Outubro de 2023 2 ii Fernanda Lisboa Guimarães Preterm Mother-Infant Interaction: Speaking and Humming during Kangaroo Care Dissertação de Mestrado Mestrado em Clínica da Infância e Adolescência Trabalho realizado sob a orientação da Professora Doutora Bárbara Figueiredo Coorientação da Professora Doutora Eduarda Carvalho Universidade do Minho Escola de Psicologia Outubro de 2023 ii DIREITOS DE AUTOR E CONDIÇÕES DE UTILIZAÇÃO DO TRABALHO POR TERCEIROS Este é um trabalho académico que pode ser utilizado por terceiros desde que respeitadas as regras e boas práticas internacionalmente aceites, no que concerne aos direitos de autor e direitos conexos. Assim, o presente trabalho pode ser utilizado nos termos previstos na licença abaixo indicada. Caso o utilizador necessite de permissão para poder fazer um uso do trabalho em condições não previstas no licenciamento indicado, deverá contactar o autor, através do RepositóriUM da Universidade do Minho. Licença concedida aos utilizadores deste trabalho Atribuição-NãoComercial-SemDerivações CC BY-NC-ND https://creativecommons.org/licenses/by-nc-nd/4.0/ iii Agradecimentos À Professora Doutora Bárbara Figueiredo, que acompanhou este trabalho do início ao fim, por ter contribuído para o meu desenvolvimento pessoal e profissional, tornando-me uma pessoa mais resiliente. Obrigada por possibilitar a minha participação no 18º Congresso Mundial da Saúde Mental Infantil e por me ter aberto portas para tantas outras oportunidades. À Professora Doutora Eduarda Carvalho por toda a disponibilidade e conhecimento transmitido, por todas as palavras de incentivo e experiências partilhadas. Agradeço com todo coração a minha família por tornarem este percurso possível e por partilharem ele comigo. Por serem sempre o meu porto seguro, por acreditarem em mim, por arranjarem forma de estarem perto e por me ensinarem tanto. Em especial, à minha mãe e ao meu pai por me amarem incondicionalmente, por fazerem o impossível para me ver feliz e por me ensinarem todos os dias o significado do amor. Obrigada por serem a melhor parte de mim. À minha avó por estar sempre comigo, por me transmitir sabedoria, força e vontade de viver, por estar sempre pronta para me ajudar e por fazer com que tudo pareça possível. Ao meu avô, por me fazer sentir a melhor neta do mundo. À dinda e à Sandra, por estarem sempre a distância de uma ligação e por todo o apoio sempre, de todas as formas. A todos aqueles que de alguma forma fizeram-se presentes e me ajudaram a chegar aqui. À Odete por ser incansável, genuína, presente, por me lembrar sempre do essencial, por me mimar e por me dar sempre toda a força que preciso para continuar. À Rita por ser casa, conforto e segurança. Por me ensinar tanto e por crescer comigo. À Maria por ser colo e verdade. Por me fazer crescer e por ser criança comigo ao mesmo tempo. À Sofia por ser tão pura, por recarregar as minhas energias e por ser tão especial. À Beatriz por ser leveza e companhia. Obrigada por dividires neurónios comigo. À Carol por sempre me ouvir e trazer o melhor de mim, por ser um exemplo a seguir. À Ana por todos os momentos de risadas e por dividir as mesmas incertezas comigo. À Eliana por toda ajuda e partilha de experiências. A todos os meus amigos que fizeram parte desta jornada e ao pessoal do CAB e do CAUM por tornarem genuinamente o meu caminho, um caminho muito mais feliz. iv STATEMENT OF INTEGRITY I hereby declare having conducted this academic work with integrity. I confirm that I have not used plagiarism or any form of undue use of information or falsification of results along the process leading to its elaboration. I further declare that I have fully acknowledged the Code of Ethical Conduct of the University of Minho. Braga, October 16, 2023 v Preterm Mother-Infant Interaction: Speaking and Humming during Kangaroo Care Abstract The quality of mother-infant interaction is crucial for the child’s development. These interactions with preterm infants are even more challenging. Maternal vocalizations in the hospital setting seem to have positive effects for both mother and infant. This study aimed to analyze the differences in the quality of preterm mother-infant interaction during maternal speaking and maternal humming while the preterm infant is held in supported diagonal flexion (SDF) position of kangaroo care, and to evaluate these differences according to the infant's gestational age at birth, intrauterine and extrauterine age and sex. Forty mother-infant dyads (mother’s age: 21-48 years) and their preterm infants (gestational age: 25-34 weeks), recruited at the Neonatal Intermediate Care Unit of a Lisbon Hospital, were evaluated using the Portuguese version of the Interaction Rating Scales. Maternal speaking and humming reveal different characteristics and purposes in mother-infant interaction and mother and infant can adjust depending on whether the interaction is spoken or hummed. The greater infant's gestational and intra and extrauterine age, as well as male sex, appear to be favorable factors for interaction. The results show the importance of disseminating this knowledge to guide neonatal care practices, leading to better infant development. Keywords : preterm infants; mother-infant interaction; maternal speaking; maternal humming vi Interação Mãe-Bebé Pré-termos: Fala e Humming Materno durante o Contacto Pele-a-Pele Resumo A qualidade da interação mãe-bebé é crucial para o desenvolvimento da criança. Com bebés pré-termos, essa interação é mais desafiadora. Vocalizações maternas no contexto hospitalar têm revelado efeitos positivos para a mãe e para o bebé. Este estudo teve como objetivo analisar as diferenças na qualidade da interação mãe-bebé pré-termos, durante a fala e o humming materno, com o bebé em posicionamento de flexão diagonal com suporte em canguru, e analisar essas diferenças de acordo com a idade gestacional ao nascimento, a idade cronológica e o sexo do bebé. Foram avaliadas quarenta díades de mães (21 – 48 anos) e seus bebés pré-termos (25 – 34 semanas de idade gestacional), recrutados na Unidade de Cuidados Intermediários Neonatais de um Hospital de Lisboa, através da versão portuguesa das Interaction Rating Scales. A fala e o humming materno revelam diferentes características e propósitos na interação mãe-bebé e a mãe e o bebé ajustam-se conforme a interação é falada ou cantada. As maiores idades gestacional e intra e extrauterina, assim como o sexo masculino, parecem ser fatores favoráveis para a interação. Os resultados mostram a importância da divulgação desse conhecimento para a orientação de práticas de cuidados neonatais, promovendo um melhor desenvolvimento do bebé. Palavras-Chave : bebés pré-termos; interação mãe-bebé; fala materna; canto materno vii Index Preterm Mother-infant Interaction: Speaking and Humming during Kangaroo Care ............................... 9 Prematurity .................................................................................................................................... 9 Preterm mother-infant interaction ................................................................................................... 9 Maternal speaking and humming .................................................................................................. 10 Objectives .................................................................................................................................... 11 Methodology ..................................................................................................................................... 11 Procedure .................................................................................................................................... 11 Participants .................................................................................................................................. 12 Instruments .................................................................................................................................. 14 Interaction Rating Scales ........................................................................................................... 14 Data analyses ............................................................................................................................... 14 Results ............................................................................................................................................. 15 Differences between maternal speaking and humming in mother-infant interactions ...................... 15 Differences in mother-infant interactions according to the infant’s gestational age .......................... 16 Differences in mother-infant interactions according to the infant’s intra and extrauterine age .......... 18 Differences in mother-infant interactions according to the infant’s sex ............................................ 19 Discussion ........................................................................................................................................ 21 Differences between speaking and humming ................................................................................ 21 Differences in mother-infant interaction according to the infant’s gestational age ........................... 22 Differences in mother-infant interaction according to the infant’s intra and extrauterine age ........... 23 Differences in mother-infant interaction according to the infant’s sex ............................................. 24 Limitations and future implications ............................................................................................... 24 Final Remarks .............................................................................................................................. 25 References ....................................................................................................................................... 26 14 Instruments Interaction Rating Scales (IRS; Field, 1980; Portuguese version by Figueiredo, 2023) This study used the face-to-face Interaction Rating Scale (IRSff), a mother-infant interaction rating scale made up of 20 items, 10 evaluating the mothers’ behaviors and 10, the infants' behavior. Videos of the interactions were used for the assessments. The dimensions assessed for the infants during the interaction were: status (from sleepy to alert), meaningful physical activity, atypical physical activity, head orientation, eye contact, gaze aversion, facial expressions, positive vocalizations, negative vocalizations and play behavior. The dimensions analyzed for the mothers were: status (from anxious/depressed to attentive/alert), meaningful physical activity, head orientation, eye contact, facial expressions, meaningful vocalizations, silence, contingent response, infantilized behaviors and play behaviors. Each item was scored from 1 to 3 according to the behavior displayed by the mother and the infant. The final result was the sum of the scores for each item divided by the number of items on the scale. Higher total scores indicated a better quality of mother-infant interaction (Figueiredo & Dias, 2013). For the evaluation of eye contact in the infant scale, gaze maintenance was not taken into account, and an eye contact was counted every time the infant looked at the mother’s eyes. Play behavior was not included because the infants were too small to make this assessment. Therefore, this item was removed from the study and from the statistical analyses, both in the mother's and the infant's assessment. The Portuguese version has high internal consistency, with a Cronbach's alpha of 0.85 for infants, 0.91 for mothers, test-retest reliability, and concurrent and predictive validity (Figueiredo & Dias, 2013). The weighted kappa coefficient was used to check inter-rater agreement. According to Altman (1991), kappa values equal to or above 0.61 indicate good agreement. The percentage of agreement in mothers ranged from 55% to 100% and in infants from 70% to 100%, with overall good agreement between the three raters. Only 5 items scored lower: the mothers' status and eye contact in speaking, the mothers' status and facial expression in humming and the infants' negative vocalizations in speaking. Kappa statistics for some items (n=5) could not be calculated due to at least one of the researchers having constant values. The items not calculated were the mothers' silence in maternal speaking and humming, the infants' gaze aversion in speaking and the infants' head orientation and gaze aversion in humming. Data analyses Data analysis was carried out using IBM® SPSS® Statistics (version 28.0). Data were described using absolute and relative frequencies for categorical variables and means and standard deviations for continuous variables. 15 The Bonferroni-corrected generalized estimating equations (GEE) model for the adjustment of multiple comparisons was used to analyze the differences in the quality of mother-infant interaction during speaking and humming (objective 1) and to analyze the differences in the quality of mother-infant interaction during maternal speaking and maternal humming according to the infant's gestational age at birth, intrauterine and extrauterine age and sex (objective 2). This analysis compared groups during both speaking and humming simultaneously to control for type I errors. Significance was set at p < 0.05 for all analyses. Results Differences between maternal speaking and humming in mother-infant interactions There were no significant differences in the mothers' interactional behaviors ( p = .676) when speaking or humming. However, the mothers were significantly ( p = .005) more attentive and alert when speaking ( M = 1.90; SD = .74) than when humming ( M = 1.64; SD = .55); they showed significantly ( p = .035) more meaningful vocalizations while humming ( M = 2. 54, SD = .51) than when speaking ( M = 2.33, SD = .60); and significantly ( p = .014) more infantilized behaviors while speaking ( M = 2.17, SD = .56) than when humming ( M = 1.93, SD = .39). In contrast, infants had significantly ( p = .025) more interactional behaviors during maternal speaking ( M = 2.51, SD = .21) than humming ( M = 2.43, SD = .18). They also showed significantly ( p = .001) more alert behaviors ( M = 2.37, SD = .55), significantly ( p = .007) less atypical physical activity ( M = 2.82, SD =. 43) and significantly ( p = .022) more eye contact ( M = 2.20, SD = .52) during maternal speaking than humming ( M = 2.11, SD = .50; M = 2.59, SD = .61; and M = 2.02, SD = .55. However, they showed significantly ( p = .048) less gaze aversion during maternal humming ( M = 2.82, SD = .42) than speaking ( M = 2.69, SD = .50). Table 2 Differences between maternal speaking and humming in mother-infant interactions Speaking Humming 95% CI p M SD M SD LL UL Mothers Interactional behavior 2.26 .29 2.24 .22 -.05 .08 .676 Status 1.90 .74 1.64 .55 .08 .42 .005* Meaningful physical activity 1.87 .60 1.78 .51 .08 .27 .309 16 Head orientation 2.89 .40 2.99 .05 -.22 .02 .116 Eye contact 1.74 .75 1.71 .82 -.20 .27 .779 Facial expression 2.02 .26 2.07 .30 -.15 .05 .315 Meaningful vocalizations 2.33 .60 2.54 .51 -.40 -.01 .035* Silence 3.00 .00 2.99 .05 -.01 .02 .311 Contingent response 2.41 .57 2.53 .57 -.31 .06 .189 Infantilized behavior 2.17 .56 1.93 .39 .05 .42 .014* Infants Interactional behavior 2.51 .21 2.43 .18 .01 .13 .025* Status 2.37 .55 2.11 .50 .10 .40 .001* Meaningful physical activity 2.38 .69 2.44 .49 -.25 .13 .542 Atypical physical activity 2.82 .43 2.59 .61 .06 .40 .007* Head orientation 2.96 .22 2.88 .40 -.05 .21 .210 Eye contact 2.20 .52 2.02 .55 .03 .34 .022* Gaze aversion 2.69 .50 2.82 .42 -.26 -.00 .048* Facial expression 3.00 .00 2.99 .05 -.01 .02 .311 Positive vocalization 1.31 .47 1.22 .46 -.05 .23 .205 Negative vocalization 2.85 .38 2.87 .38 -.12 .09 .746 Note. CI = Confidence interval: LL = Lower limit; UL = Upper limit; * p < .05 Differences in mother-infant interactions according to the infant’s gestational age The comparisons of groups according to gestational age did not reveal any significant differences (p = .433) in mothers' interactional behavior when they were speaking or humming to their infants. Mothers whose infants were older than 214 days gestational age at birth showed a more alert and attentive posture in relation to their infants while speaking (M = 2.42, SD = .58) than humming (M = 2.08, SD = .49). Infants over 214 days gestational age were also significantly (p = .042) more alert during maternal speaking (M = 1.97, SD = .76) than humming (M = 1.70, SD = .52). Infants at a gestational age of 214 days or less showed a significantly (p = .043) higher mean value of atypical physical activity during humming (M = 2.85, SD = .46) than speaking (M = 2.60, SD = .58). 17 Table 3 Differences in mother-infant interactions according to the infant´s gestational age at birth Speaking Humming p ≤ 214 days >214 days ≤ 214 days >214 days M ± SD M ± SD M ± SD M ± SD Mothers Interactional behavior 2.51 ± .31 2.52 ± .31 2.41 ± .29 2.49 ± .21 .433 Status 2.32 ± .54ab 2.42 ± .58b 2.15 ± .51ab 2.08 ± .49a .008* Meaningful physical activity 2.45 ± .68 2.32 ± .71 2.32 ± .51 2.57 ± .43 .090 Head orientation 2.85 ± .33 2.80 ± .52 2.63 ± .58 2.55 ± .66 .052 Eye contact 2.98 ± .07 2.93 ± .30 2.78 ± .55 2.97 ± .10 .289 Facial expression 2.11 ± .52 2.28 ± .52 1.95 ± .54 2.08 ± .57 .090 Meaningful vocalizations 2.25 ± .65 2.42 ± .55 2.55 ± .50 2.53 ± .53 .170 Silence 3.00 ± .00 3.00 ± .00 2.98 ± .07 3.00 ± .00 .305 Contingent response 2.47 ± .63 2.35 ± .50 2.47 ± .57 2.60 ± .58 .452 Infantilized behavior 2.17 ± .64 2.17 ± .50 1.88 ± .52 1.98 ± .20 .104 Infants Interactional behavior 2.17 ± .21 2.34 ± .25 2.19 ± .20 2.27 ± .22 .138 Status 1.82 ± .73ab 1.97 ± .76b 1.58 ± .59ab 1.70 ± .52a .042* Meaningful physical activity 1.72 ± .39 2.03 ± .72 1.68 ± .58 1.88 ± .44 .327 Atypical physical activity 2.60 ± .58a 2.78 ± .41ab 2.85 ± .46b 2.80± .38ab .043* Head orientation 2.85 ± .50 2.93 ± .30 3.00 ± .00 2.98 ± .07 .241 Eye contact 1.50 ± .67 1.98 ± .77 1.62 ± .78 1.80 ± .88 .171 Gaze aversion 3.00 ± .00 3.00 ± .00 3.00 ± .00 2.98 ± .07 .305 Facial expression 1.98 ± .30 2.07 ± .23 2.02 ± .28 2.13 ± .31 .433 Positive vocalization 1.30 ± .43 1.32 ± .51 1.17 ± .37 1.27 ± .55 .552 Negative vocalization 2.77 ± .45 2.93 ± .30 2.82 ± .49 2.92 ± .24 .441 Note. a,b Equal letters indicate no significant differences (Bonferroni test, p < .05); * p < .05 18 Differences in mother-infant interactions according to the infant’s intra and extrauterine age The analysis of the interactions between maternal speaking and humming according to the intra and extrauterine age of the infants revealed that mothers of infants at an intra and extrauterine age of 239 days or less showed significantly ( M = 1.88, SD = .44) less active and alert behaviors when humming ( p < .001) than in the other intra and extrauterine age groups ( M = 2.25, SD = .61; M = 2.48, SD = .48; and M = 2.35, SD = .45). Additionally, mothers of infants in the same group (intra and extrauterine age: ≤ 239 days) showed significantly ( p = .013) fewer deviations of head orientation while speaking ( M = 2.85, SD = .33) than when humming ( M = 2.55, SD = .54). Mothers of infants over 239 days of intra and extrauterine age showed significantly ( p = .002) more facial expressions when speaking ( M = 2.40, SD = .44) than mothers of infants at an intra and extrauterine age of 239 days or less while speaking ( M = 2.00, SD = .53) and humming ( M = 1.83, SD = .55). Table 4 shows that infants at an intra and extrauterine age greater than 239 days were significantly ( p = .016) more alert during maternal speaking ( M = 2.03, SD = .75) than infants of both ages during maternal humming ( M = 1.57, SD = .47) and speaking ( M = 1.72, SD = .62). Moreover, infants over 239 days of intra and extrauterine age displayed significantly more physical activity during maternal speaking ( M = 2.12, SD = .58) than infants aged 239 days or less during both maternal speaking ( M = 1.63, SD = .53) and humming ( M = 1.65, SD = .48), with a statistically significant difference ( p = .022). Table 4 Differences in mother-infant interactions according to the infant’s intra and extrauterine age Speaking Humming p ≤ 239 days >239 days ≤ 239 days >239 days Mean ± SD Mean ± SD Mean ± SD Mean ± SD Mothers Interactional behavior 2.50 ± .27 2.53 ± .34 2.42 ± .20 2.48 ± .30 .297 Status 2.25 ± .61b 2.48 ± .48b 1.88 ± .44a 2.35 ± .45b <.001* Meaningful physical activity 2.40 ± .70 2.37 ± .70 2.50 ± .44 2.38 ± .53 .750 Head orientation 2.85 ± .33b 2.80 ± .52ab 2.55 ± .54a 2.63 ± .69ab .013* Eye contact 2.98 ± .07 2.93 ± .30 2.87 ± .37 2.88 ± .45 .217 Facial expression 2.00 ± .53a 2.40 ± .44b 1.83 ± .55a 2.20 ± .51ab .002* Meaningful vocalizations 2.40 ± .55 2.27 ± .65 2.52 ± .54 2.57 ± .50 .106 19 Silence 3.00 ± .00 3.00 ± .00 2.98 ± .07 3.00 ± .00 .305 Contingent response 2.43 ± .53 2.38 ± .61 2.62 ± .49 2.45 ± .64 .449 Infantilized behavior 2.20 ± .52 2.13 ± .62 2.03 ± .15 1.83 ± .52 .083 Infants Interactional behavior 2.23 ± .21 2.27 ± .27 2.21 ± .17 2.26 ± .25 .776 Status 1.75 ± .72ab 2.03 ± .75b 1.57 ± .47a 1.72 ± .62a .016* Meaningful physical activity 1.63 ± .53a 2.12 ± .58b 1.65 ± .48 a 1.92 ± .53ab .022* Atypical physical activity 2.75 ± .42 2.63 ± .58 2.85 ± .33 2.80 ± .50 .226 Head orientation 3.00 ± .00 2.78 ± .55 2.98 ± .07 3.00 ± .00 .118 Eye contact 1.68 ± .75 1.80 ± .78 1.67 ± .83 1.75 ± .84 .960 Gaze aversion 3.00 ± .00 3.00 ± .00 2.98 ± .07 3.00 ± .00 .305 Facial expression 2.08 ± .28 1.97 ± .24 2.10 ± .24 2.05 ± .35 .134 Positive vocalization 1.25 ± .43 1.37 ± .51 1.13 ± .31 1.30 ± .57 .275 Negative vocalization 2.95 ± .16 2.75 ± .51 2.95 ± .12 2.78 ± .52 .346 Note. a,b Equal letters indicate no significant differences (Bonferroni test, p < .05); * p < .05 Differences in mother-infant interactions according to the infant’s sex The analysis of infants' sex during maternal speaking and humming revealed that mothers of male infants had significantly ( p = .018) more interactional behaviours during maternal speaking ( M = 2.59, SD = .25) than mothers of female infants during humming ( M = 2.38, SD = .20). The results of maternal status revealed that mothers of male infants had significantly more alert behaviours ( p = .001) during speaking ( M = 2.52, SD = .41) than mothers of both female ( M = 2.06, SD = .60) and male ( M = 2.17, SD = .41) infants while humming. In addition, mothers of male infants had significantly fewer deviations of head orientation ( p = .018) during maternal speaking ( M = 2.89, SD = .30) than mothers of both female ( M = 2.52, SD = .70) and male ( M = 2.65, SD = .55) infants during humming. Finally, mothers of male infants had significantly ( p = .001) more meaningful vocalizations ( M = 2.73, SD = .38) during humming than mothers of female infants during both speaking ( M = 2.17, SD = .61) and humming ( M = 2.31, SD = .57). The results of infant status revealed that, regardless of the infants' sex, mean values during speaking ( M = 1.86, SD = .66; M = 1.93, SD = .84) were significantly ( p = .042) higher than the means for infants during maternal humming ( M = 1.62, SD = .58; M = 1.67, SD = .52). Male infants had significantly ( p = .002) less atypical physical activity during maternal speaking ( M = 2.89, SD = .30) and humming ( M = 2.95, SD = .16) than female infants during maternal speaking ( M = 2.44, SD = .59). 20 Finally, male infants had significantly ( p = .046) fewer negative vocalizations during maternal humming (M = 3.00, SD = .00) than female infants during speaking ( M = 2.78, SD = .46) and humming ( M = 2.70, SD = .53). Note. a,b Equal letters indicate no significant differences (Bonferroni test, p < .05); * p < .05 Table 5 Differences in mother-infant interactions according to the infant’s sex Speaking Humming Male Female Male Female p M ± SD M ± SD M ± SD M ± SD Mothers Interactional behavior 2.59 ± .25b 2.42 ± .35ab 2.51 ±.28ab 2.38 ± .20a .018* Status 2.52 ± .41b 2.19 ± .66ab 2.17 ± .41a 2.06 ± .60a .001* Meaningful physical activity 2.52 ± .65 2.22 ± .72 2.53 ± .51 2.33 ± .44 .439 Head orientation 2.89 ± .30b 2.74 ± .55ab 2.65 ± .55a 2.52 ± .70a .018* Eye contact 3.00 ± .00 2.91 ± .32 2.85 ± .50 2.91 ± .25 .167 Facial expression 2.27 ± .48 2.11 ± .57 2.02 ± .50 2.02 ± .63 .083 Meaningful vocalizations 2.47 ±.57ab 2.17 ± .61a 2.73 ± .38b 2.31 ± .57a .001* Silence 3.00 ± .00 3.00 ± .00 2.98 ± .07 3.00 ± .00 .306 Contingent response 2.50 ± .53 2.30 ± .60 2.65 ± .56 2.39 ± .56 .216 Infantilized behavior 2.15 ± .59 2.19 ± .55 2.00 ± .41 1.85 ± .37 .083 Infants Interactional behavior 2.29 ± .22 2.20 ± .26 2.26 ± .22 2.20 ± .20 .538 Status 1.86 ± .66b 1.93 ± .84b 1.62 ± .58a 1.67 ± .52a .042* Meaningful physical activity 1.85 ± .58 1.91 ± .63 1.74 ± .46 1.83 ± .59 .716 Atypical physical activity 2.89 ± .30b 2.44 ± .59a 2.95 ± .16b 2.67 ± .57ab .002* Head orientation 2.91 ± .43 2.87 ± .38 3.00 ± .00 2.98 ± .08 .204 Eye contact 1.83 ± .70 1.63 ± .82 1.76 ± .90 1.65 ± .74 .823 Gaze aversion 3.00 ± .00 3.00 ± .00 3.00 ± .00 2.98 ± .08 .303 Facial expression 2.02 ± .13 2.04 ± .38 2.09 ± .23 2.06 ± .37 .597 Positive vocalization 1.38 ± .52 1.22 ± .40 1.17 ± .47 1.28 ± .46 .150 Negative vocalization 2.91 ± .31ab 2.78 ± .46a 3.00 ± .00b 2.70 ± .53a .046* 21 Discussion The present study sought to analyze the characteristics of mother-infant interaction during maternal speaking and maternal humming, as well as the effects of the infant's gestational age at birth, intra and extrauterine age and sex, during maternal speaking and humming, on mother-infant interaction. the results show that speaking and humming have different effects on mother-infant interaction, and speaking is associated with a more alert behavior for both partners in the interaction. the infant’s higher gestational age and the infant’s higher intra and extrauterine age seem to optimize mother infant interaction during maternal speaking. It is also concluded that male sex seems to be a facilitating factor for mother-infant interaction. Differences between speaking and humming We first conclude that speaking and humming produce different effects on mother-infant interaction. Compared to maternal humming, maternal speaking was associated with a more active and alertness state for both partners during interactions. This finding is in line with those reported by some previous studies, which found that maternal speaking was associated with a state of quiet alertness and more motor activity, whereas humming kept the infant in a state of active sleep (Filippa et al., 2013; Arias & Peña, 2016). Likewise, mothers had more infantilized behaviors - more simplified movements and imitations – while interacting with the infant, and infants had more eye responses to their mothers - both more eye contacts and aversions to the mother's gaze – during maternal speaking. This may be assigned to a greater activation of the infant during maternal speaking (Filippa et al., 2013). Also, both mothers and fathers change their vocal qualities depending on the infant's state of wakefulness or sleep (Saliba et al., 2020), and the vocal interaction is reciprocal and complementary between the dyad. Therefore, when the infant is in a more alert state, the mother has more infantilized behaviors and interactions with the infant, and the infant, in turn, makes more eye contact with the mother. Another study that included infants between 29 and 37 weeks of gestational age who were exposed to their mother’s speaking and humming also found an increase in eye opening, with higher values during maternal speaking (Filippa et al., 2020). The higher prevalence of an active sleep state during humming may explain infants’ lower gaze aversion responses to their mothers’ eye contact (Filippa et al., 2013; Saliba et al., 2020). However, mothers produce a repetitive lullaby humming, and future studies should see whether these same behaviors are observed when the mothers sing in different styles, either (lullabies or play songs (songs sung in a playful, upbeat style). 22 Mothers produced fewer meaningful vocalizations during speaking than humming. Meaningful vocalizations are those appropriate to respond to an infant's vocalizations; they occur during turn-taking and give the infant a greater opportunity to alternate with their mother's vocalizations. During humming, however, vocal flow is less interrupted, and phrase duration and melodic contours are more repetitive, which constructs anticipation of the end of the phrases, prompting infants to produce overlapping vocalizations (Carvalho et al., 2019). Infants, in turn, had a greater interactional behavior with the mother in speaking compared to humming, which might indicate that speaking is a more favorable condition for mother-infant interaction. However, this result could be explained by the fact that during speaking, as the infant is in a more alert state, there is a greater possibility of the infant having more interactive behaviors with the mother. Atypical physical activities are strange motor movements, without clear interactive purposes, or which result from high malaise/self-regulation difficulties. In infants, they were observed less frequently during maternal speaking than humming, which was against expectations and a study with term infants that found that infant-directed humming was more effective than speaking in delaying the onset of infant crying and distress (Corbeil et al., 2015). The differences found in dyad interaction during maternal speaking and humming may also be explained by the fact that mothers have more experience in speaking to infants than humming to them. The limited privacy and intimacy that mothers may have in the NICU may inhibit them more while humming than speaking. Differences in mother-infant interaction according to the infant’s gestational age Less preterm dyads were more alert during the interaction, suggesting that higher gestational age can optimize mother-infant interaction during speaking. The analysis of the effect of gestational age at birth revealed that mothers of less preterm infants (> 214 days gestational age) were more alert during maternal speaking than humming, in agreement with two studies (Filippa et al., 2013; Shellaas et al., 2019). The latter included preterm infants at ≥35 weeks gestational age and reported that maternal speaking was associated with a decrease in sleep and an increase in wakefulness. This might have been an effect of the age of preterm infants, as mothers seem to adapt to the expectations that they have of their infants. Mothers seem to expect more responsiveness and involvement from older infants and are more active than the mothers of younger infants, maybe because they see younger infants as more vulnerable and less competent for interactions. Less preterm infants were also more alert during maternal speaking than humming. 23 However, it was also found that lower gestational age at birth seems to contribute to lower atypical physical activity during humming. More preterm infants, born at 214 days gestational age or less, had less atypical physical activity during maternal humming than speaking, suggesting that humming, more than speaking, may favor self-regulation (Haslbeck, 2013) in more preterm infants, or that infants may be calmer because of the state of drowsiness that humming seems to induce (Filippa et al., 2013; Saliba et al., 2020). Differences in mother-infant interaction according to the infant’s intra and extrauterine age Older dyads had greater alert behavior, infants show more meaningful physical activity, and mothers more facial expressions, which suggests that the infants' higher intra and extrauterine age seems to optimize the development of mother-infant interaction during maternal speaking. Mothers of younger infants, whose intra and extrauterine age was ≤ 239 days, were more attentive and alert and had fewer deviations of head orientation during maternal speaking than humming. this may be associated with the fact that infants are at a greater state of alertness during maternal speaking (filippa 2013, shellaas et al., 2019). however, a study with full-term infants found that humming fosters greater infant involvement, as they their attention increases and they maintain eye contact and reduce their movements, which contributes to greater emotional coordination with their mothers (nakata & trehub, 2004). Mothers of older infants (intra and extrauterine age > 239 days) were more attentive and alert during both activities than mothers of younger infants during humming, which makes sense given that older infants are able to interact more with their mothers. Studies should be conducted to compare the age differences of preterm infants, but also to analyze how differences in maternal speaking and humming affect the mothers themselves. Mothers of older infants also showed more facial expressions while speaking than mothers of younger infants when speaking or humming. Again, this was expected, given that the older the infants, the greater their capacity to communicate and interact actively with their mothers. The immaturity of preterm infants may affect their perception and discrimination of auditory stimuli, which makes their responses to their mother's voice less robust than those of full-term infants (de Regnier et al., 2002). The use of more facial expressions during maternal speaking than humming should be expected, as humming seems to indicate a less active state than speaking (Filippa et al., 2013). However, humming may be more favorable for preterm infants with less intra and extrauterine age, since the results indicate that humming is associated with a less alert state when compared to speaking. 30 Shellhaas, R. A., Burns, J. W., Barks, J. D. E., Hassan, F., & Chervin, R. D. (2019). Maternal Voice and Infant Sleep in the Neonatal Intensive Care Unit. Pediatrics , 144 (3), e20190288. https://doi.org/10.1542/peds.2019-0288 Shoemark, H. (2019). Returning musicality to parents: Constructing a strategy using parental voice. International Journal for Birth and Parent Education, 6(2), 29–32. Shoemark, H., Dahlstrøm, M., Bedford, O., & Stewart, L. (2021). The Effect of a Voice-Centered PsychoEducational Program on Maternal Self-Efficacy: A Feasibility Study. International Journal of Environmental Research and Public Health , 18 (5), 2537. https://doi.org/10.3390/ijerph18052537 Souza, M. S., Bandeira, L. D., Soares, M. das G. S., Soares, M. A. S., Dias Filho, J. C., Sousa, A. M. C., Silva, É. B. da, Nascimento, G. P. de A., Silva, A. S. E. S., Cândido, G. da S., Nascimento, I. G. do, Silva Filho, E. A. da, Silva, A. R. C., Sousa, Y. A. S., & Andrade, F. S. F. de. (2022). Método Canguru na UTI neonatal: benefícios para a saúde e vínculo materno-infantil. Research, Society and Development , 11 (13), e160111335072. https://doi.org/10.33448/rsd-v11i13.35072 Trehub, S. E., Plantinga, J., & Russo, F. A. (2015). Maternal Vocal Interactions with Infants: Reciprocal Visual Influences. Social Development , 25 (3), 665–683. https://doi.org/10.1111/sode.12164 Trevarthen, C. (1993). The self born in intersubjectivity: The psychology of an infant communicating. In U. Neisser (Ed.), The perceived self: Ecological and interpersonal sources of self-knowledge (pp. 121–173). Cambridge University Press. Wachman, E. M., & Lahav, A. (2010). The effects of noise on preterm infants in the NICU. Archives of Disease in Childhood - Fetal and Neonatal Edition , 96 (4), F305–F309. https://doi.org/10.1136/adc.2009.182014 Winnicott, D. W. (1999). Os bebês e suas mães. São Paulo: Martins Fontes. World Health Organization. (2022, November 15). WHO advises immediate skin to skin care for survival of small and preterm babies . https://www.who.int/news/item/15-11-2022-who-advisesimmediate-skin-to-skin-care-for-survival-of-small-and-preterm-babies World Health Organization. (2023, May 10). Preterm birth . https://www.who.int/news-room/factsheets/detail/preterm-birth