- Email: [email protected]
Inspiratory Phonation in Baby Voice
ARTICLE IN PRESS Inspiratory Phonation in Baby Voice *Kathleen Wermke, *Asin Ahmad Haschemi, †,‡Volker Hesse, and §Michael P. Robb, *Würzburg, †Berlin-Lichtenberg, and ‡Berlin, Germany, and §Christchurch, New Zealand
Summary: Objective. This study aimed to evaluate the developmental occurrence of inspiratory phonations (IPs) in the spontaneous cries of healthy infants across the first 10 weeks of life. Study Design. This is a populational retrospective study. Participants. The spontaneous crying of 17 healthy infants (10 were male) was retrospectively investigated. Materials and Methods. Sound files of spontaneously uttered cries that were repeatedly recorded once per week for across the first 10 weeks of life were retrospectively analyzed. Frequency spectra and waveforms were used to identify the occurrence of IPs and to measure the duration and fundamental frequency (fo) of each instance of IP. Results. A consistent number of IPs were identified across the 10-week period. All infants were observed to produce IPs in their spontaneous cries, although the frequency of occurrence was not consistent across infants. A marked sex difference was observed with female infants producing a higher number of IPs compared to males. The duration and fo of IPs did not differ significantly across the 10 weeks or between sexes. Conclusions. The production of IPs is a regularly occurring phenomenon in healthy, normally developing infants’ spontaneous crying. The proportional difference in the production of IPs between female and male infants, observed for the first time here, is postulated to be linked to sex-based differences (including steroidal hormones) in respiratory anatomy and physiology. Key Words: Inspiratory phonation–Cry–Fundamental frequency–Infant–Sex difference.
INTRODUCTION A baby’s voice is characterized on the one hand by high melodic capacity and on the other hand by aperiodic, as well as inspiratory phonation (IP). IP is the audible phonation that occurs during the inspiratory phase of the breath cycle. Anecdotal reports of this striking acoustic feature date back to the classic cry research of Bosma et al.1 In adults, this type of phonation can be either an involuntary or voluntary act. As an involuntary act, IP occurs naturally during situations such as laughing or sighing. Involuntary IP has also been found in pathologic voice conditions involving a prolapsed airway,2 a subglottal obstruction,3 or abnormal laryngeal muscle patterns.4 IP can occur voluntarily as a form of speech production in various languages (eg, Scandinavian and New Zealand English) to achieve specific paralinguistic functions.5 The deliberate production of IP has been used as a therapeutic tool to facilitate normal laryngeal valving in some vocal pathologies (eg, ventricular phonation).6,7 To produce deliberate IP, supraglottic pressure needs to be sufficiently high to create a Bernoulli effect, thereby closing the glottis during ingressive airflow. Vanhecke et al8 and Orlikoff et al9 have examined the acoustic and physiological characteristics of IP. Vanhecke et al’s and Orlikoff et al’s collective results Accepted for publication April 12, 2017. The study was partially supported by the German Research Foundation (DFG, WE-1724/4-1) and the Max Planck Institute for Human Cognitive and Brain Sciences, Leipzig as part of Research Group 381—Early Language Development and Specific Language Disorders (FR-519/18-1). From the *Center for Pre-Speech Development and Developmental Disorders, Department of Orthodontics, University of Würzburg, Würzburg, Germany; †German Center for Growth, Development and Health Encouragement during Childhood and Youth, Children’s Hospital Lindenhof, Berlin-Lichtenberg, Germany; ‡Institute for Experimental Paediatric Endocrinology, Charité—University Medicine, Berlin, Germany; and the §School of Health Sciences, University of Canterbury, Christchurch, New Zealand. Address correspondence and reprint requests to Kathleen Wermke, Center for Pre-Speech Development and Developmental Disorders, University Clinics, Julius Maximilian University of Würzburg, Würzburg 97070, Germany. E-mail: [email protected] Journal of Voice, Vol. ■■, No. ■■, pp. ■■-■■ 0892-1997 © 2017 The Voice Foundation. Published by Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jvoice.2017.04.005
indicate that IP is an acoustically harmonic phenomenon associated with distention of the laryngeal ventricles, lengthening of the quadrangular membrane of the vocal folds, reduced vocal fold contact, and increased transglottal airflow. An example of involuntary IP can be found in infant crying. Although IP has not escaped recognition over the past 50 years of infant cry research, this feature has received less attention than expiratory features of crying. There remains only one dedicated report on the IP of infant cry.10 Over 20 years ago, these researchers performed a detailed analysis of the pain cries produced by 20 healthy, full-term infants at 2 days of age. Their interest in IP was motivated by the assumption that IP reflects a form of laryngeal constriction of ingressive airflow. As such, this particular feature may hold diagnostic relevance for medical conditions thought to be linked to upper-airway obstruction (eg, obstructive sleep apnea and sudden unexpected death syndrome). The researchers found that most but not all infants produced IPs, and these particular sounds were significantly shorter in duration and higher in fo compared with expiratory cries. Two explanations were provided for the occurrence of IPs. First, IPs may reflect a disruption in the expected sequence of onset of the posterior cricoarytenoid muscle (ie, a vocal fold abductor) and diaphragmatic muscle activity, thereby disposing the infant to phonate upon inspiration. Alternatively, IP may relate to the configuration of the infant vocal tract, whereby the confined structures within the vocal tract obstruct the upper airway. The single report by Grau et al10 suggests that IP is a common feature of neonates’ pain cries. However, the study was limited to the newborn period. In addition, cries were elicited via a painful stimulus rather than spontaneously, and hence were more excitatory in nature. Newman11 has proposed differential neural circuitry for pain and spontaneous cries. Pain cries are a form of involuntary expression and appear to be controlled by the limbic system, whereas spontaneous cries reflect a form of voluntary expression and may be under cortical control.
ARTICLE IN PRESS 2
Journal of Voice, Vol. ■■, No. ■■, 2017
Based on the past results obtained by Grau et al,10 which were paired with the acoustic and physiological characteristics of IP reported for adults,8 we wished to further investigate the occurrence and the acoustic features of this unique aspect of baby voice in the context of naturally occurring spontaneous cries. The aim of the research presented here was to investigate (1) whether IP occurs regularly in healthy infants’ spontaneous crying and (2) whether the phenomenon exists beyond the newborn period. Consequently, retrospective analysis of IP over the first 10 weeks was performed in a group of infants who showed no signs of developmental, psychological, or language disorder over the first 5 years of life. MATERIALS AND METHODS Participants Seventeen healthy, full-term infants (10 were males) were considered in the present retrospective study. Participants were recruited, recorded, and medically examined at the Children’s Hospital Lindenhof in Berlin, Germany. These infants were a subset of a broader longitudinal study examining genetic and environmental factors influencing language development during the first 5 years of age. Institutional ethics approval was granted and parents provided written consent for their infant’s participation. The hearing capabilities of the infants across the observation period were assessed to be normal based on a combination of otoacoustic emission and brainstem evoked response audiometry. Relevant somatic measures were documented at birth, four and eight weeks (Table 1). None of these measures demonstrated a significant sex difference in this sample. All infants demonstrated normal development throughout the data collection period and as guaranteed by regular medical and developmental assessments across the first 2 years of life. Data collection Cry samples were collected from each infant once per week across the first 10 weeks of life. Cry signals were recorded using a portable digital recorder (Sony TCD-D100) and microphone (Sony ECM-950/957). Sampling frequency was 48 kHz and the dynamic range was 16 bits. All cry samples were obtained in quiet areas in the hospital (first week) or home environment (thereafter). The
infants were recorded lying in a supine position in their mother’s presence. Cry recording began when an infant started to fuss or at a time when the mother would normally feed the child. The microphone was handheld at a distance of approximately 10–15 cm from the child’s mouth. Only spontaneous cries were recorded (ie, when the infants were hungry or thirsty), and no cries were induced through inflicting pain. Recordings had an average duration of 2 minutes per session. Data analysis The expiratory and inspiratory cries produced by each infant were analyzed using a commercially available hardware system (CSL-4500; Kay-PENTAX, Montvale, NJ). Each cry sample was displayed as an amplitude-by-time waveform with the corresponding narrow-band (45 Hz) spectrogram. Instances of IP were identified by visual inspection of waveforms and frequency spectra paired with auditory-perceptual cues. Vertical cursors were superimposed on the dual displays and manually positioned to identify onset and offset points within the cry. A typical display of expiratory crying and IP is illustrated in Figure 1. Each expiratory event and the corresponding inspiratory signal phonated within the complete recording session per infant was analyzed for IP occurrence (number of events = 100%). This was also the approach used by Grau et al10 in their initial exploration of IP. Thus, for reason of comparability, we applied Grau et al’s method here. However, a more elaborate analysis, including normalized temporal features, is in preparation to re-evaluate the observed sex difference. The following features were measured: IP frequency of occurrence: defined as the total number of IPs occurring for each infant’s session of crying. The proportional occurrence of IP was calculated for each infant based on the total number of expiratory cries produced during each recording session. IP duration: defined as the total duration from the onset to the offset of IP activity. Onset was defined as the initial increase in amplitude from the baseline signal. Offset was indicated by a decrease in amplitude and a return to baseline.
TABLE 1. Somatic Measures Sex Female
Male
n
Birth
4 wk
8 wk
Weight (g) Length (cm) Head circumference (cm) Breast circumference (cm)
3332 (227) 50.1 (1.0) 34.5 (0.8) 32.7 (1.2)
4195 (479) 54.6 (1.8) 37.0 (1.0) 36.3 (1.3)
5094 (458) 58.0 (1.9) 38.9 (0.8) 38.4 (1.7)
Weight (g) Length (cm) Head circumference (cm) Breast circumference (cm)
3326 (301) 50.3 (1.8) 34.9 (1.4) 31.3 (1.4)
4295 (277) 54.4 (1.3) 37.6 (1.1) 36.2 (1.1)
5328 (383) 58.3 (1.2) 39.3 (1.2) 39.4 (0.9)
7
10
Data reported are mean values and corresponding standard deviations.
ARTICLE IN PRESS Kathleen Wermke, et al
Inspiratory Phonation in Baby Voice
3
FIGURE 1. Amplitude-by-time waveform and corresponding narrow-band spectrogram of infant spontaneous crying. An alternating sequence of expiratory crying and inspiratory phonation (arrows) is displayed. IP fo: defined as the average frequency component of the IP. For each instance of inspiratory cry, the minimum and maximum fo values were identified. The geometric mean was derived from these minimum and maximum values. Intrajudge measurement reliability for IP duration and fo feature was based on the remeasurement of IP instances across five weekly recording sessions of five randomly selected infants. The original measurements were compared to the remeasured samples, resulting in average remeasurement errors of 5 ms and 30 Hz, respectively. Intrajudge reliability for identification of instances of IP was based on a randomly selected subsample of 400 single events (breath cycles). These 400 events were reanalyzed without prior information about instances of IP. The intrajudge reliability between the original and reanalyzed events was assessed according to the Cohen index kappa, which resulted in an index of 0.98 (P < 0.001), indicating a high degree of reliability in identifying IP phenomena. Statistical methods Descriptive statistics and one- and two-factorial repeatedmeasures analyses of variance (ANOVAs) were performed using SPSS Statistics v23 (IBM, USA). Boxplots diagrams were performed, expressing the 25th–75th percentiles (box) and the median as a horizontal line. RESULTS The results are presented in two sections. The first section reports the results related to the overall and proportional occurrence of IPs. The second section contains the results for the acoustic analysis of IP duration and fo.
IP frequency of occurrence The results of the developmental analysis of the frequency of occurrence of inspiratory cries are listed in Table 2. Across all infants, the proportional occurrence of IPs in a crying session ranged from 31% to 46%. Results of a repeated-measures ANOVA found no significant difference in the occurrence of IPs across the 10-week period (Greenhouse-Geisser: F(9, 90) = 1.61, P = 0.2). The weekly sex-specific distribution of IP over the observation period is displayed in Figure 2. A more frequent IP occurrence in female infants compared to male infants is evident in all but 2 weeks. Averaging the individual data over the whole observation time also found that female infants had higher values than male infants (mean [standard deviation]/median: female infants 47 [11]%/46%; male infants 31 [20]%/28%). This difference was statistically significant (Mann-Whitney U test, P = 0.04). IP duration and fo The weekly results of the acoustic analysis of IP are listed in Table 1. For the entire group of infants, the duration and fo of IP showed a minimal change across the 10-week period. Results of the repeated-measures two-way ANOVAs (week × sex) were not significant for duration (Greenhouse-Geiser: week F(9, 81) = 1.1, P > 0.05; week × sex F(9, 81) = 0.45, P > 0.05) and for fo (Greenhouse-Geiser: week F(9, 81) = 3.0, P > 0.05; week × sex F(9, 81) = 0.54, P > 0.05). Multivariate analysis of variance was used to examine sex differences in fo and duration with data averaged over the whole observation time. The multivariate analysis of variance revealed no significant multivariate effect for sex (Wilks λ = 0.89, F(2, 14) = 0.89, P > 0.05, η2 = 0.11). However, the median values in female infants were
ARTICLE IN PRESS 4 44 (23) 157 (56) 632
43 (23) 153 (59) 583 (41)
40 (25) 139 (40) 642 34 (23) 135 (45) 633 35 (35) 129 (47) 635 33 (28) 154 (63) 678 32 (28) 124 (35) 702 31 (27) 120 (23) 678 45 (32) 129 (40) 727 I (%) Duration (ms) Mean fo (Hz) 6868 Total
FIGURE 2. Boxplot demonstrating the developmental course and interindividual variation of the inspiratory phonatory activity (horizontal line marking indicates overall medians).
Data reported are mean values and corresponding standard deviations. n is the total number of analyzed cries.
38 (28) 148 (48) 608 (69) 27 (22) 148 (56) 614 (79) 22 (28) 133 (74) 576 (43) 41 (36) 135 (59) 708 (103) 3831 Male
I (%) Duration (ms) Mean fo (Hz) Female
I (%) Duration (ms) Mean fo (Hz)
52 (26) 139 (32) 738 (188)
31 (33) 123 (26) 656 (89)
32 (19) 124 (26) 690 (181)
22 (21) 145 (37) 638 (105)
46 (31) 134 (34) 738 (251)
18 (24) 140 (29) 652 (70)
44 (20) 147 (47) 662 (195) 44 (22) 147 (45) 644 (163) 53 (38) 137 (35) 668 (174) 55 (23) 183 (67) 692 (121)
46 (30) 151 (48) 654
40 (30) 145 (31) 653 (118)
36 (26) 152 (37) 610
29 (26) 145 (17) 585 (59)
46 (24) 150 (35) 625 (143) 56 (30) 174 (60) 655 (177)
9 8 7 6 5 4 3 2 1 n
3037
Gender
TABLE 2. Developmental Course of the Frequency of Occurrence of Inspiratory (I) Cries and Their Duration and Mean fo Over the First 10 Weeks of Life
10
46 (26) 167 (67) 661 (192)
Journal of Voice, Vol. ■■, No. ■■, 2017
higher compared to male infants (duration: 154 ms vs 128 ms, fo: 658 Hz vs 595 Hz). DISCUSSION Despite interindividual differences, the study found that all infants showed repeated IP occurrence across the whole observation period. In total, the occurrence varied between 31% and 46% (Table 2) with female infants slightly outnumbering male infants. A higher occurrence of IP in the crying of female infants was observable in the boxplot diagrams in all but 1 week (excluding the second week, Figure 2). Consequently, the results of the present study found the production of IP to be a regularly occurring phenomenon of spontaneous crying in healthy infants between birth and 10 weeks of age without obvious disease or at-risk status. Although statistically not significant, the weekly fo data collected for the group of infants indicated consistently higher fo values for the female infants (Table 2). The high fo for female infants is suggestive of greater transglottal airflow compared with the male infants, as shown in a previous study examining the relationship between glottal airflow and highpitched voices.12 IP seems to be related to respiratory development paired with vocal control. Merkus et al13 reported that a distinct feature of infant respiratory anatomy is a highly compliant chest wall. As a result of this condition, the lungs recoil to a much lower volume in relation to total lung capacity. A particular difficulty associated with compliant chest wall is a tendency for small airway closure. In the case of infant crying, which is characterized by robust phonatory activity, it is possible that the adaptive changes in the respiratory control system, combined with a predisposition for small airway closure, could serve to facilitate the occurrence of IPs.10 In this case, IP should relate to the configuration of the infant vocal tract, whereby the confined structures
ARTICLE IN PRESS Kathleen Wermke, et al
Inspiratory Phonation in Baby Voice
within the newborn vocal tract obstruct the upper airway. Accordingly, a developmental decrease of IP should occur with the onset of the descending larynx then. However, there is a report of minimal descent of the larynx during the period of 2–4 months of age.14 So, it is perhaps not surprising that a developmental pattern in IP was not found across the first 10 weeks. However, earlier findings by the Scandinavian pioneers of cry research, Bosma et al,1 in infants under 10 days of age support the hypothesis of laryngeal IP genesis and questioned the latter pharyngeal constriction hypothesis. These researchers found that the typical sinusoidal pharyngeal motions during expiratory crying and inspiration did not cause any observable phonatory effect in the investigated infants; rather constrictive action of the larynx was suggested, which “under certain conditions of stress” is carried over into the subsequent inspiration.1(p70) Also, Grau et al,10 who systematically defined and quantified IP in newborns for the first time, favored a laryngeal hypothesis: During moments of crying, there is a mis-sequencing of the posterior cricoarytenoid (a vocal fold abductor) and diaphragmatic muscle activity that contributes to closing of the glottis during inspiration. This would define IP as a neurophysiological control phenomenon. A mainly laryngeal genesis of IP, for example, by highly tensed vocal folds, is supported by the higher fo of IP in contrast to the mean fo of expiratory crying. The only previous report of acoustic features of IP was based on cries collected in 2-dayold infants.10 In this report, the average fo of IP was 687 Hz. The data collected for the present group of infants at 1 week of age indicated an fo of 727 Hz. These results align nicely with those of Grau et al10 and validate the observations reported in both studies. In contrast, the mean fo in the expiratory crying of newborns is about 407 Hz.15,16 Over the first 3 months, the fo in expiratory spontaneous crying was found to vary between about 360 and 480 Hz,17 which is considerably lower compared to the fo in IP. Although expiratory crying is controlled by welldeveloped and fine-tuned neurophysiological mechanisms,18,19 the production of IP is mainly governed by physical entities that are less well controlled. Independent of the fact that IP occurrence in young infants could represent immaturity of respiratory-laryngeal control, breathing control is already speechlike. Similar to adult speech, the inspiratory phase of the crying respiratory cycle is much shorter than the expiratory phase. Here, we found the mean IP duration to be between 120 and 157 ms across the first 10 weeks. The first-week average of 139 ms corresponds to the 146-ms duration measurements by Grau et al for 2-day-olds.10 In comparison, the mean duration of expiratory cry phonation during this same time period is approximately 970 ms.15 The speechlike extension of the expiratory phase during crying is a distinctive trait of human infants crying to nonhuman primate calls.20 For the first time, sex differences were observed in the present study with a higher occurrence of IPs in females compared with males. The pattern observed for female and male infants provides support for the recently postulated hormonal effects on phonation at this early age.21 For many years, it has been known that sex hormones influence breathing control and lung maturation. Behan and Wenninger22 summarized possible mechanisms
5
whereby sex hormones could modulate breathing, including direct and indirect effects on gene expression in respiratory neurons. Behan and Wenninger also suggested that the impact of steroid hormones on the developing respiratory control system may be sex specific. This is in agreement with recent findings on the effect of sex hormones on infant crying and babbling21,23 during the first postnatal months. These months are characterized by an elevated sex hormone surge comparable with puberty and hence called “mini-puberty.”24 Sex hormones exert regulatory effects on lung development during the neonatal period with female lungs tending to be smaller and weighing less than those of males.25 There is also indirect evidence to suggest that estrogens are associated with fetal lung maturation occurring sooner among female infants.26 Even though the lungs of female infants are smaller than those of male infants, they exhibit higher forced expiratory flow rates27; that is, the emptying rate of the lungs is higher in female infants than in male infants. These features, combined with possible differences in the rate of ingressive airflow, could lead to the occurrence of more IPs among female infants. Indeed, in the postnatal weeks known for the highest estradiol excess (fourth to eighth weeks of life) during mini-puberty,21,24 were found to be the same weeks in which the present group of female infants exhibited the highest occurrence rate of IP. There is a further argument for the “hormone hypothesis” that is connected with the respiratory development of periodic breathing. Periodic breathing is a normally occurring phenomenon where several pauses (cessations) in breathing occur close together, followed by a series of rapid, shallow breaths.28,29 Wilkinson et al30 found that the periodic breathing cycle duration in infants decreases across the first 4 months of life. Edwards et al31 further defined periodic breathing as reaching a peak at 2–4 weeks’ postnatal age (coinciding with hormone excess peaks during minipuberty) and disappearing by 6 months (fading of mini-puberty). The possible association between sex hormone excess and its importance for early neurophysiological development is intriguing and in need of further study. Finally, the question of whether IPs that persist during the second half-year could carry a bioindicative valence needs further investigation. This could be particularly interesting in infants at risk of developmental disorders with sex-specific incidence. For example, sudden infant death syndrome (SIDS), which is twice as frequent in boys,32,33 seems to be associated with abnormalities in the portion of an infant’s brain that controls breathing.34 SIDS emerges synchronously to the sex hormone excess during mini-puberty between 4 and 16 weeks of life but may occur until 12 months.33 Postmortem analyzed blood samples of SIDSdiagnosed infants showed significantly higher testosterone levels (as they are typical for mini-pubertal levels in boys) compared to a control group.35 There is also a report which indicates that persistence of IP into the babbling period reflects poor control of the respiratory rhythm. Marschik et al36 found instances of inspiratory vocalizations in individuals with Rett syndrome, which is a neurologic disorder affecting predominantly female infants with regression or loss of speech and purposeful hand use, after a few months of seemingly normal development.37 The IPs were not characteristic of comfort vocalizations of typically developing
ARTICLE IN PRESS 6 infants, such as cooing or babbling. Consequently, the observation of IPs in infants beyond 6 months of age may be a feature of pathologic development. The precise age at which IP occurrence should no longer occur cannot be answered with the current data. Our observation period was too short to determine the disappearance of IP. The relatively high occurrence of IP in young infants might reflect traces of phylogenetic paths and thus maturation of brain mechanisms underlying vocal development, because nonhuman primates vocalize on both inhalation and exhalation.20 CONCLUSIONS In conclusion, IP is a perceptually striking feature of the crying behavior exhibited by young infants. The present study sought to evaluate the occurrence and acoustic characteristics of this apparently unusual phonatory behavior. All of the infants were found to have normal physical and language development at 5 years of age, so it would seem that the presence (or absence) of IPs during the first 10 weeks of life may hold little or no predictive or diagnostic relevance. However, many questions remain, and there is a need for continued research in this exciting new direction in infant phonation. Acknowledgments The authors thank all the parents and infants who participated in the study, as well as all colleagues and coworkers of Research Group 381, particularly the project team of Prof. Dr. Manfred Gross at the Charité, University Medical Center Berlin, Department of Audiology and Phoniatrics, who performed the audiological assessments of the infants. The authors also thank Peter Wermke, IT Center of the University of Würzburg, for his support in the data analysis. REFERENCES 1. Bosma JF, Truby HM, Lind J. Cry motions of the newborn infant. Acta Paediatr. 1965;54:60–92. 2. Aronson AE, Bless DM. Clinical Voice Disorders. 4th ed. Thieme Publishers Series; 2009. 3. Gavriely N, Palti Y, Alroy G, et al. Measurement and theory of wheezing breath sounds. J Appl Physiol. 1984;57:481–492. 4. Hogikyan ND, Wodchis WP, Spak C, et al. Longitudinal effects of botulinum toxin injections on voice-related quality of life (V-RQOL) for patients with adductory spasmodic dysphonia. J Voice. 2001;15:576–586. 5. Eklund R. Languages with pulmonic ingressive speech: updating and adding to the list, Proceedings from Fonetik, Lund University Sweden, 2015 31–34. Available at: http://konferens.ht.lu.se/fileadmin/user_upload/conference/ fonetik2015/docs/WP55_Eklund_2_Languages.pdf. Accessed July 2, 2017. 6. Finger LS, Cielo CA. Reverse phonation—physiologic and clinical aspects of this speech voice therapy modality. Braz J Otorhinolaryngol. 2007; 73:271–277. 7. Colton RH, Casper JK, Leonhard R. Understanding Voice Problems: a Physiological Perspective for Diagnosis and Treatment. 4th ed. Lippincott Williams & Wilkins; 2011. 8. Vanhecke F, Lebacq J, Moerman M, et al. Physiology and acoustics of inspiratory phonation. J Voice. 2016;30:769.e9–769.e18. 9. Orlikoff RF, Baken RJ, Kraus DH. Acoustic and physiologic characteristics of inspiratory phonation. J Acoust Soc Am. 1997;102:1838–1845. 10. Grau SM, Robb MP, Cacace AT. Acoustic correlates of inspiratory phonation during infant cry. J Speech Hear Res. 1995;38:373–381. 11. Newman JD. Neural circuits underlying crying and cry responding in mammals. Behav Brain Res. 2007;182:155–165.
Journal of Voice, Vol. ■■, No. ■■, 2017 12. Holmberg EB, Hillman RE, Perkell JS. Glottal airflow and transglottal air pressure measurements for male and female speakers in soft, normal, and loud voice. J Acoust Soc Am. 1988;84:511–529. 13. Merkus FM, de Jongste JC, Stocks J. Respiratory function measurements in infants and children. In: Gosselink R, ed. Lung Function Testing, European Respiratory Society Monographs. 2005:166–194. 14. Stephens RE, Bancroft A, Glaros AG, et al. Anatomic changes related to laryngeal descent from birth to 1 year of age: do they play a role in SIDS? Ear Nose Throat J. 2010;89:313–317. 15. Wermke K, Teiser J, Yovsi E, et al. Fundamental frequency variation within neonatal crying: does ambient language matter? Speech Lang Hear. 2016;doi:10.1080/2050571X.2016.1187903. 16. Wermke K, Ruan Y, Feng Y, et al. Fundamental frequency variation in crying of Mandarin and German neonates. J Voice. 2016;31:255.e25–255.e30. doi:10.1016/j.jvoice.2016.06.009. pii: S0892-1997(16)30124-2, Epub ahead of print. 17. Lind K, Wermke K. Development of the vocal fundamental frequency of spontaneous cries during the first 3 months. Int J Pediatr Otorhinolaryngol. 2002;64:97–104. 18. Wermke K. Neonatal crying behaviors. In: Wright JD, ed. International Encyclopedia of the Social & Behavioral Sciences. 2nd ed. Oxford: Elsevier; 2015:475–480. 19. Wermke K, Mende W, Kempf A, et al., Interaction patterns between melodies and resonance frequencies in infants’ pre-speech utterances, Proceedings of the Fourth International Workshop on Models and Analysis of Vocal Emissions for Biomedical Applications, University of Firenze, Italy, 2005 187–190. 20. MacLarnon AM, Hewitt GP. The evolution of human speech: the role of enhanced breathing control. Am J Phys Anthropol. 1999;109:341–363. 21. Borysiak A, Hesse V, Wermke P, et al. Fundamental frequency of crying in two-month-old boys and girls: do sex hormones during mini-puberty mediate differences? J Voice. 2017;31:128.e21–128.e28. doi:10.1016/ j.jvoice.2015.12.006. Epub 2016 Jan 14. 22. Behan M, Wenninger JM. Sex steroidal hormones and respiratory control. Respir Physiol Neurobiol. 2008;164:213–221. 23. Quast A, Hesse V, Hain J, et al. Baby babbling at five months linked to sex hormone levels in early infancy. Infant Behav Dev. 2016;44:1–10. doi:10.1016/j.infbeh.2016.04.002. Epub 2016 May 18. 24. Becker M, Oehler K, Partsch C, et al. Hormonal “minipuberty” influences the somatic development of boys but not of girls up to the age of 6 years. Clin Endocrinol (Oxf). 2015;83:694–701. doi:10.1111/cen.12827. Epub 2015 Jul 1. 25. Carey MA, Card JW, Voltz JW, et al. It’s all about sex: gender, lung development and lung disease. Trends Endocrinol Metab. 2007;18:308–313. Epub 2007 Aug 30. 26. Torday JS, Nielsen HC, Fencl Mde M, et al. Sex differences in fetal lung maturation. Am Rev Respir Dis. 1981;123:205–208. 27. Langston C, Kida K, Reed M, et al. Human lung growth in late gestation and in the neonate. Am Rev Respir Dis. 1984;129:607–613. 28. Patel M, Mohr M, Lake D, et al. Clinical associations with immature breathing in preterm infants: part 2-periodic breathing. Pediatr Res. 2016;80:28–34. doi:10.1038/pr.2016.58. Epub 2016 Mar 22. 29. Kelly DH, Stellwagen LM, Kaitz E, et al. Apnea and periodic breathing in normal full-term infants during the first twelve months. Pediatr Pulmonol. 1985;1:215–219. 30. Wilkinson MH, Skuza EM, Rennie GC, et al. Postnatal development of periodic breathing cycle duration in term and preterm infants. Pediatr Res. 2007;62:331–336. 31. Edwards BA, Sands SA, Berger PJ. Postnatal maturation of breathing stability and loop gain: the role of carotid chemoreceptor development. Respir Physiol Neurobiol. 2013;185:144–155. doi:10.1016/j.resp.2012.06.003. Epub 2012 Jun 13. 32. Kinney HC, Filiano JJ, Harper RM. The neuropathology of the sudden infant death syndrome. A review. J Neuropathol Exp Neurol. 1992;51:115– 126. 33. Krous HF, Beckwith JB, Byard RW, et al. Sudden infant death syndrome and unclassified sudden infant deaths: a definitional and diagnostic approach. Pediatrics. 2004;114:234–238.
ARTICLE IN PRESS Kathleen Wermke, et al
Inspiratory Phonation in Baby Voice
34. Kinney HC, Richerson GB, Dymecki SM, et al. The brainstem and serotonin in the sudden infant death syndrome. Annu Rev Pathol. 2009;4:517– 550. 35. Emery MJ, Krous HF, Nadeau-Manning JM, et al. Serum testosterone and estradiol in sudden infant death. J Pediatr. 2005;147:586–591.
7
36. Marschik PB, Kaufmann WE, Sigafoos J, et al. Changing the perspective on early development of Rett syndrome. Res Dev Disabil. 2013;34:1236– 1239. doi:10.1016/j.ridd.2013.01.014. Epub 2013 Feb 9. 37. de Bona C, Zappella M, Hayek G, et al. Preserved speech variant is allelic of classic Rett syndrome. Eur J Hum Genet. 2000;8:325–330.
Summary: Objective. This study aimed to evaluate the developmental occurrence of inspiratory phonations (IPs) in the spontaneous cries of healthy infants across the first 10 weeks of life. Study Design. This is a populational retrospective study. Participants. The spontaneous crying of 17 healthy infants (10 were male) was retrospectively investigated. Materials and Methods. Sound files of spontaneously uttered cries that were repeatedly recorded once per week for across the first 10 weeks of life were retrospectively analyzed. Frequency spectra and waveforms were used to identify the occurrence of IPs and to measure the duration and fundamental frequency (fo) of each instance of IP. Results. A consistent number of IPs were identified across the 10-week period. All infants were observed to produce IPs in their spontaneous cries, although the frequency of occurrence was not consistent across infants. A marked sex difference was observed with female infants producing a higher number of IPs compared to males. The duration and fo of IPs did not differ significantly across the 10 weeks or between sexes. Conclusions. The production of IPs is a regularly occurring phenomenon in healthy, normally developing infants’ spontaneous crying. The proportional difference in the production of IPs between female and male infants, observed for the first time here, is postulated to be linked to sex-based differences (including steroidal hormones) in respiratory anatomy and physiology. Key Words: Inspiratory phonation–Cry–Fundamental frequency–Infant–Sex difference.
INTRODUCTION A baby’s voice is characterized on the one hand by high melodic capacity and on the other hand by aperiodic, as well as inspiratory phonation (IP). IP is the audible phonation that occurs during the inspiratory phase of the breath cycle. Anecdotal reports of this striking acoustic feature date back to the classic cry research of Bosma et al.1 In adults, this type of phonation can be either an involuntary or voluntary act. As an involuntary act, IP occurs naturally during situations such as laughing or sighing. Involuntary IP has also been found in pathologic voice conditions involving a prolapsed airway,2 a subglottal obstruction,3 or abnormal laryngeal muscle patterns.4 IP can occur voluntarily as a form of speech production in various languages (eg, Scandinavian and New Zealand English) to achieve specific paralinguistic functions.5 The deliberate production of IP has been used as a therapeutic tool to facilitate normal laryngeal valving in some vocal pathologies (eg, ventricular phonation).6,7 To produce deliberate IP, supraglottic pressure needs to be sufficiently high to create a Bernoulli effect, thereby closing the glottis during ingressive airflow. Vanhecke et al8 and Orlikoff et al9 have examined the acoustic and physiological characteristics of IP. Vanhecke et al’s and Orlikoff et al’s collective results Accepted for publication April 12, 2017. The study was partially supported by the German Research Foundation (DFG, WE-1724/4-1) and the Max Planck Institute for Human Cognitive and Brain Sciences, Leipzig as part of Research Group 381—Early Language Development and Specific Language Disorders (FR-519/18-1). From the *Center for Pre-Speech Development and Developmental Disorders, Department of Orthodontics, University of Würzburg, Würzburg, Germany; †German Center for Growth, Development and Health Encouragement during Childhood and Youth, Children’s Hospital Lindenhof, Berlin-Lichtenberg, Germany; ‡Institute for Experimental Paediatric Endocrinology, Charité—University Medicine, Berlin, Germany; and the §School of Health Sciences, University of Canterbury, Christchurch, New Zealand. Address correspondence and reprint requests to Kathleen Wermke, Center for Pre-Speech Development and Developmental Disorders, University Clinics, Julius Maximilian University of Würzburg, Würzburg 97070, Germany. E-mail: [email protected] Journal of Voice, Vol. ■■, No. ■■, pp. ■■-■■ 0892-1997 © 2017 The Voice Foundation. Published by Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jvoice.2017.04.005
indicate that IP is an acoustically harmonic phenomenon associated with distention of the laryngeal ventricles, lengthening of the quadrangular membrane of the vocal folds, reduced vocal fold contact, and increased transglottal airflow. An example of involuntary IP can be found in infant crying. Although IP has not escaped recognition over the past 50 years of infant cry research, this feature has received less attention than expiratory features of crying. There remains only one dedicated report on the IP of infant cry.10 Over 20 years ago, these researchers performed a detailed analysis of the pain cries produced by 20 healthy, full-term infants at 2 days of age. Their interest in IP was motivated by the assumption that IP reflects a form of laryngeal constriction of ingressive airflow. As such, this particular feature may hold diagnostic relevance for medical conditions thought to be linked to upper-airway obstruction (eg, obstructive sleep apnea and sudden unexpected death syndrome). The researchers found that most but not all infants produced IPs, and these particular sounds were significantly shorter in duration and higher in fo compared with expiratory cries. Two explanations were provided for the occurrence of IPs. First, IPs may reflect a disruption in the expected sequence of onset of the posterior cricoarytenoid muscle (ie, a vocal fold abductor) and diaphragmatic muscle activity, thereby disposing the infant to phonate upon inspiration. Alternatively, IP may relate to the configuration of the infant vocal tract, whereby the confined structures within the vocal tract obstruct the upper airway. The single report by Grau et al10 suggests that IP is a common feature of neonates’ pain cries. However, the study was limited to the newborn period. In addition, cries were elicited via a painful stimulus rather than spontaneously, and hence were more excitatory in nature. Newman11 has proposed differential neural circuitry for pain and spontaneous cries. Pain cries are a form of involuntary expression and appear to be controlled by the limbic system, whereas spontaneous cries reflect a form of voluntary expression and may be under cortical control.
ARTICLE IN PRESS 2
Journal of Voice, Vol. ■■, No. ■■, 2017
Based on the past results obtained by Grau et al,10 which were paired with the acoustic and physiological characteristics of IP reported for adults,8 we wished to further investigate the occurrence and the acoustic features of this unique aspect of baby voice in the context of naturally occurring spontaneous cries. The aim of the research presented here was to investigate (1) whether IP occurs regularly in healthy infants’ spontaneous crying and (2) whether the phenomenon exists beyond the newborn period. Consequently, retrospective analysis of IP over the first 10 weeks was performed in a group of infants who showed no signs of developmental, psychological, or language disorder over the first 5 years of life. MATERIALS AND METHODS Participants Seventeen healthy, full-term infants (10 were males) were considered in the present retrospective study. Participants were recruited, recorded, and medically examined at the Children’s Hospital Lindenhof in Berlin, Germany. These infants were a subset of a broader longitudinal study examining genetic and environmental factors influencing language development during the first 5 years of age. Institutional ethics approval was granted and parents provided written consent for their infant’s participation. The hearing capabilities of the infants across the observation period were assessed to be normal based on a combination of otoacoustic emission and brainstem evoked response audiometry. Relevant somatic measures were documented at birth, four and eight weeks (Table 1). None of these measures demonstrated a significant sex difference in this sample. All infants demonstrated normal development throughout the data collection period and as guaranteed by regular medical and developmental assessments across the first 2 years of life. Data collection Cry samples were collected from each infant once per week across the first 10 weeks of life. Cry signals were recorded using a portable digital recorder (Sony TCD-D100) and microphone (Sony ECM-950/957). Sampling frequency was 48 kHz and the dynamic range was 16 bits. All cry samples were obtained in quiet areas in the hospital (first week) or home environment (thereafter). The
infants were recorded lying in a supine position in their mother’s presence. Cry recording began when an infant started to fuss or at a time when the mother would normally feed the child. The microphone was handheld at a distance of approximately 10–15 cm from the child’s mouth. Only spontaneous cries were recorded (ie, when the infants were hungry or thirsty), and no cries were induced through inflicting pain. Recordings had an average duration of 2 minutes per session. Data analysis The expiratory and inspiratory cries produced by each infant were analyzed using a commercially available hardware system (CSL-4500; Kay-PENTAX, Montvale, NJ). Each cry sample was displayed as an amplitude-by-time waveform with the corresponding narrow-band (45 Hz) spectrogram. Instances of IP were identified by visual inspection of waveforms and frequency spectra paired with auditory-perceptual cues. Vertical cursors were superimposed on the dual displays and manually positioned to identify onset and offset points within the cry. A typical display of expiratory crying and IP is illustrated in Figure 1. Each expiratory event and the corresponding inspiratory signal phonated within the complete recording session per infant was analyzed for IP occurrence (number of events = 100%). This was also the approach used by Grau et al10 in their initial exploration of IP. Thus, for reason of comparability, we applied Grau et al’s method here. However, a more elaborate analysis, including normalized temporal features, is in preparation to re-evaluate the observed sex difference. The following features were measured: IP frequency of occurrence: defined as the total number of IPs occurring for each infant’s session of crying. The proportional occurrence of IP was calculated for each infant based on the total number of expiratory cries produced during each recording session. IP duration: defined as the total duration from the onset to the offset of IP activity. Onset was defined as the initial increase in amplitude from the baseline signal. Offset was indicated by a decrease in amplitude and a return to baseline.
TABLE 1. Somatic Measures Sex Female
Male
n
Birth
4 wk
8 wk
Weight (g) Length (cm) Head circumference (cm) Breast circumference (cm)
3332 (227) 50.1 (1.0) 34.5 (0.8) 32.7 (1.2)
4195 (479) 54.6 (1.8) 37.0 (1.0) 36.3 (1.3)
5094 (458) 58.0 (1.9) 38.9 (0.8) 38.4 (1.7)
Weight (g) Length (cm) Head circumference (cm) Breast circumference (cm)
3326 (301) 50.3 (1.8) 34.9 (1.4) 31.3 (1.4)
4295 (277) 54.4 (1.3) 37.6 (1.1) 36.2 (1.1)
5328 (383) 58.3 (1.2) 39.3 (1.2) 39.4 (0.9)
7
10
Data reported are mean values and corresponding standard deviations.
ARTICLE IN PRESS Kathleen Wermke, et al
Inspiratory Phonation in Baby Voice
3
FIGURE 1. Amplitude-by-time waveform and corresponding narrow-band spectrogram of infant spontaneous crying. An alternating sequence of expiratory crying and inspiratory phonation (arrows) is displayed. IP fo: defined as the average frequency component of the IP. For each instance of inspiratory cry, the minimum and maximum fo values were identified. The geometric mean was derived from these minimum and maximum values. Intrajudge measurement reliability for IP duration and fo feature was based on the remeasurement of IP instances across five weekly recording sessions of five randomly selected infants. The original measurements were compared to the remeasured samples, resulting in average remeasurement errors of 5 ms and 30 Hz, respectively. Intrajudge reliability for identification of instances of IP was based on a randomly selected subsample of 400 single events (breath cycles). These 400 events were reanalyzed without prior information about instances of IP. The intrajudge reliability between the original and reanalyzed events was assessed according to the Cohen index kappa, which resulted in an index of 0.98 (P < 0.001), indicating a high degree of reliability in identifying IP phenomena. Statistical methods Descriptive statistics and one- and two-factorial repeatedmeasures analyses of variance (ANOVAs) were performed using SPSS Statistics v23 (IBM, USA). Boxplots diagrams were performed, expressing the 25th–75th percentiles (box) and the median as a horizontal line. RESULTS The results are presented in two sections. The first section reports the results related to the overall and proportional occurrence of IPs. The second section contains the results for the acoustic analysis of IP duration and fo.
IP frequency of occurrence The results of the developmental analysis of the frequency of occurrence of inspiratory cries are listed in Table 2. Across all infants, the proportional occurrence of IPs in a crying session ranged from 31% to 46%. Results of a repeated-measures ANOVA found no significant difference in the occurrence of IPs across the 10-week period (Greenhouse-Geisser: F(9, 90) = 1.61, P = 0.2). The weekly sex-specific distribution of IP over the observation period is displayed in Figure 2. A more frequent IP occurrence in female infants compared to male infants is evident in all but 2 weeks. Averaging the individual data over the whole observation time also found that female infants had higher values than male infants (mean [standard deviation]/median: female infants 47 [11]%/46%; male infants 31 [20]%/28%). This difference was statistically significant (Mann-Whitney U test, P = 0.04). IP duration and fo The weekly results of the acoustic analysis of IP are listed in Table 1. For the entire group of infants, the duration and fo of IP showed a minimal change across the 10-week period. Results of the repeated-measures two-way ANOVAs (week × sex) were not significant for duration (Greenhouse-Geiser: week F(9, 81) = 1.1, P > 0.05; week × sex F(9, 81) = 0.45, P > 0.05) and for fo (Greenhouse-Geiser: week F(9, 81) = 3.0, P > 0.05; week × sex F(9, 81) = 0.54, P > 0.05). Multivariate analysis of variance was used to examine sex differences in fo and duration with data averaged over the whole observation time. The multivariate analysis of variance revealed no significant multivariate effect for sex (Wilks λ = 0.89, F(2, 14) = 0.89, P > 0.05, η2 = 0.11). However, the median values in female infants were
ARTICLE IN PRESS 4 44 (23) 157 (56) 632
43 (23) 153 (59) 583 (41)
40 (25) 139 (40) 642 34 (23) 135 (45) 633 35 (35) 129 (47) 635 33 (28) 154 (63) 678 32 (28) 124 (35) 702 31 (27) 120 (23) 678 45 (32) 129 (40) 727 I (%) Duration (ms) Mean fo (Hz) 6868 Total
FIGURE 2. Boxplot demonstrating the developmental course and interindividual variation of the inspiratory phonatory activity (horizontal line marking indicates overall medians).
Data reported are mean values and corresponding standard deviations. n is the total number of analyzed cries.
38 (28) 148 (48) 608 (69) 27 (22) 148 (56) 614 (79) 22 (28) 133 (74) 576 (43) 41 (36) 135 (59) 708 (103) 3831 Male
I (%) Duration (ms) Mean fo (Hz) Female
I (%) Duration (ms) Mean fo (Hz)
52 (26) 139 (32) 738 (188)
31 (33) 123 (26) 656 (89)
32 (19) 124 (26) 690 (181)
22 (21) 145 (37) 638 (105)
46 (31) 134 (34) 738 (251)
18 (24) 140 (29) 652 (70)
44 (20) 147 (47) 662 (195) 44 (22) 147 (45) 644 (163) 53 (38) 137 (35) 668 (174) 55 (23) 183 (67) 692 (121)
46 (30) 151 (48) 654
40 (30) 145 (31) 653 (118)
36 (26) 152 (37) 610
29 (26) 145 (17) 585 (59)
46 (24) 150 (35) 625 (143) 56 (30) 174 (60) 655 (177)
9 8 7 6 5 4 3 2 1 n
3037
Gender
TABLE 2. Developmental Course of the Frequency of Occurrence of Inspiratory (I) Cries and Their Duration and Mean fo Over the First 10 Weeks of Life
10
46 (26) 167 (67) 661 (192)
Journal of Voice, Vol. ■■, No. ■■, 2017
higher compared to male infants (duration: 154 ms vs 128 ms, fo: 658 Hz vs 595 Hz). DISCUSSION Despite interindividual differences, the study found that all infants showed repeated IP occurrence across the whole observation period. In total, the occurrence varied between 31% and 46% (Table 2) with female infants slightly outnumbering male infants. A higher occurrence of IP in the crying of female infants was observable in the boxplot diagrams in all but 1 week (excluding the second week, Figure 2). Consequently, the results of the present study found the production of IP to be a regularly occurring phenomenon of spontaneous crying in healthy infants between birth and 10 weeks of age without obvious disease or at-risk status. Although statistically not significant, the weekly fo data collected for the group of infants indicated consistently higher fo values for the female infants (Table 2). The high fo for female infants is suggestive of greater transglottal airflow compared with the male infants, as shown in a previous study examining the relationship between glottal airflow and highpitched voices.12 IP seems to be related to respiratory development paired with vocal control. Merkus et al13 reported that a distinct feature of infant respiratory anatomy is a highly compliant chest wall. As a result of this condition, the lungs recoil to a much lower volume in relation to total lung capacity. A particular difficulty associated with compliant chest wall is a tendency for small airway closure. In the case of infant crying, which is characterized by robust phonatory activity, it is possible that the adaptive changes in the respiratory control system, combined with a predisposition for small airway closure, could serve to facilitate the occurrence of IPs.10 In this case, IP should relate to the configuration of the infant vocal tract, whereby the confined structures
ARTICLE IN PRESS Kathleen Wermke, et al
Inspiratory Phonation in Baby Voice
within the newborn vocal tract obstruct the upper airway. Accordingly, a developmental decrease of IP should occur with the onset of the descending larynx then. However, there is a report of minimal descent of the larynx during the period of 2–4 months of age.14 So, it is perhaps not surprising that a developmental pattern in IP was not found across the first 10 weeks. However, earlier findings by the Scandinavian pioneers of cry research, Bosma et al,1 in infants under 10 days of age support the hypothesis of laryngeal IP genesis and questioned the latter pharyngeal constriction hypothesis. These researchers found that the typical sinusoidal pharyngeal motions during expiratory crying and inspiration did not cause any observable phonatory effect in the investigated infants; rather constrictive action of the larynx was suggested, which “under certain conditions of stress” is carried over into the subsequent inspiration.1(p70) Also, Grau et al,10 who systematically defined and quantified IP in newborns for the first time, favored a laryngeal hypothesis: During moments of crying, there is a mis-sequencing of the posterior cricoarytenoid (a vocal fold abductor) and diaphragmatic muscle activity that contributes to closing of the glottis during inspiration. This would define IP as a neurophysiological control phenomenon. A mainly laryngeal genesis of IP, for example, by highly tensed vocal folds, is supported by the higher fo of IP in contrast to the mean fo of expiratory crying. The only previous report of acoustic features of IP was based on cries collected in 2-dayold infants.10 In this report, the average fo of IP was 687 Hz. The data collected for the present group of infants at 1 week of age indicated an fo of 727 Hz. These results align nicely with those of Grau et al10 and validate the observations reported in both studies. In contrast, the mean fo in the expiratory crying of newborns is about 407 Hz.15,16 Over the first 3 months, the fo in expiratory spontaneous crying was found to vary between about 360 and 480 Hz,17 which is considerably lower compared to the fo in IP. Although expiratory crying is controlled by welldeveloped and fine-tuned neurophysiological mechanisms,18,19 the production of IP is mainly governed by physical entities that are less well controlled. Independent of the fact that IP occurrence in young infants could represent immaturity of respiratory-laryngeal control, breathing control is already speechlike. Similar to adult speech, the inspiratory phase of the crying respiratory cycle is much shorter than the expiratory phase. Here, we found the mean IP duration to be between 120 and 157 ms across the first 10 weeks. The first-week average of 139 ms corresponds to the 146-ms duration measurements by Grau et al for 2-day-olds.10 In comparison, the mean duration of expiratory cry phonation during this same time period is approximately 970 ms.15 The speechlike extension of the expiratory phase during crying is a distinctive trait of human infants crying to nonhuman primate calls.20 For the first time, sex differences were observed in the present study with a higher occurrence of IPs in females compared with males. The pattern observed for female and male infants provides support for the recently postulated hormonal effects on phonation at this early age.21 For many years, it has been known that sex hormones influence breathing control and lung maturation. Behan and Wenninger22 summarized possible mechanisms
5
whereby sex hormones could modulate breathing, including direct and indirect effects on gene expression in respiratory neurons. Behan and Wenninger also suggested that the impact of steroid hormones on the developing respiratory control system may be sex specific. This is in agreement with recent findings on the effect of sex hormones on infant crying and babbling21,23 during the first postnatal months. These months are characterized by an elevated sex hormone surge comparable with puberty and hence called “mini-puberty.”24 Sex hormones exert regulatory effects on lung development during the neonatal period with female lungs tending to be smaller and weighing less than those of males.25 There is also indirect evidence to suggest that estrogens are associated with fetal lung maturation occurring sooner among female infants.26 Even though the lungs of female infants are smaller than those of male infants, they exhibit higher forced expiratory flow rates27; that is, the emptying rate of the lungs is higher in female infants than in male infants. These features, combined with possible differences in the rate of ingressive airflow, could lead to the occurrence of more IPs among female infants. Indeed, in the postnatal weeks known for the highest estradiol excess (fourth to eighth weeks of life) during mini-puberty,21,24 were found to be the same weeks in which the present group of female infants exhibited the highest occurrence rate of IP. There is a further argument for the “hormone hypothesis” that is connected with the respiratory development of periodic breathing. Periodic breathing is a normally occurring phenomenon where several pauses (cessations) in breathing occur close together, followed by a series of rapid, shallow breaths.28,29 Wilkinson et al30 found that the periodic breathing cycle duration in infants decreases across the first 4 months of life. Edwards et al31 further defined periodic breathing as reaching a peak at 2–4 weeks’ postnatal age (coinciding with hormone excess peaks during minipuberty) and disappearing by 6 months (fading of mini-puberty). The possible association between sex hormone excess and its importance for early neurophysiological development is intriguing and in need of further study. Finally, the question of whether IPs that persist during the second half-year could carry a bioindicative valence needs further investigation. This could be particularly interesting in infants at risk of developmental disorders with sex-specific incidence. For example, sudden infant death syndrome (SIDS), which is twice as frequent in boys,32,33 seems to be associated with abnormalities in the portion of an infant’s brain that controls breathing.34 SIDS emerges synchronously to the sex hormone excess during mini-puberty between 4 and 16 weeks of life but may occur until 12 months.33 Postmortem analyzed blood samples of SIDSdiagnosed infants showed significantly higher testosterone levels (as they are typical for mini-pubertal levels in boys) compared to a control group.35 There is also a report which indicates that persistence of IP into the babbling period reflects poor control of the respiratory rhythm. Marschik et al36 found instances of inspiratory vocalizations in individuals with Rett syndrome, which is a neurologic disorder affecting predominantly female infants with regression or loss of speech and purposeful hand use, after a few months of seemingly normal development.37 The IPs were not characteristic of comfort vocalizations of typically developing
ARTICLE IN PRESS 6 infants, such as cooing or babbling. Consequently, the observation of IPs in infants beyond 6 months of age may be a feature of pathologic development. The precise age at which IP occurrence should no longer occur cannot be answered with the current data. Our observation period was too short to determine the disappearance of IP. The relatively high occurrence of IP in young infants might reflect traces of phylogenetic paths and thus maturation of brain mechanisms underlying vocal development, because nonhuman primates vocalize on both inhalation and exhalation.20 CONCLUSIONS In conclusion, IP is a perceptually striking feature of the crying behavior exhibited by young infants. The present study sought to evaluate the occurrence and acoustic characteristics of this apparently unusual phonatory behavior. All of the infants were found to have normal physical and language development at 5 years of age, so it would seem that the presence (or absence) of IPs during the first 10 weeks of life may hold little or no predictive or diagnostic relevance. However, many questions remain, and there is a need for continued research in this exciting new direction in infant phonation. Acknowledgments The authors thank all the parents and infants who participated in the study, as well as all colleagues and coworkers of Research Group 381, particularly the project team of Prof. Dr. Manfred Gross at the Charité, University Medical Center Berlin, Department of Audiology and Phoniatrics, who performed the audiological assessments of the infants. The authors also thank Peter Wermke, IT Center of the University of Würzburg, for his support in the data analysis. REFERENCES 1. Bosma JF, Truby HM, Lind J. Cry motions of the newborn infant. Acta Paediatr. 1965;54:60–92. 2. Aronson AE, Bless DM. Clinical Voice Disorders. 4th ed. Thieme Publishers Series; 2009. 3. Gavriely N, Palti Y, Alroy G, et al. Measurement and theory of wheezing breath sounds. J Appl Physiol. 1984;57:481–492. 4. Hogikyan ND, Wodchis WP, Spak C, et al. Longitudinal effects of botulinum toxin injections on voice-related quality of life (V-RQOL) for patients with adductory spasmodic dysphonia. J Voice. 2001;15:576–586. 5. Eklund R. Languages with pulmonic ingressive speech: updating and adding to the list, Proceedings from Fonetik, Lund University Sweden, 2015 31–34. Available at: http://konferens.ht.lu.se/fileadmin/user_upload/conference/ fonetik2015/docs/WP55_Eklund_2_Languages.pdf. Accessed July 2, 2017. 6. Finger LS, Cielo CA. Reverse phonation—physiologic and clinical aspects of this speech voice therapy modality. Braz J Otorhinolaryngol. 2007; 73:271–277. 7. Colton RH, Casper JK, Leonhard R. Understanding Voice Problems: a Physiological Perspective for Diagnosis and Treatment. 4th ed. Lippincott Williams & Wilkins; 2011. 8. Vanhecke F, Lebacq J, Moerman M, et al. Physiology and acoustics of inspiratory phonation. J Voice. 2016;30:769.e9–769.e18. 9. Orlikoff RF, Baken RJ, Kraus DH. Acoustic and physiologic characteristics of inspiratory phonation. J Acoust Soc Am. 1997;102:1838–1845. 10. Grau SM, Robb MP, Cacace AT. Acoustic correlates of inspiratory phonation during infant cry. J Speech Hear Res. 1995;38:373–381. 11. Newman JD. Neural circuits underlying crying and cry responding in mammals. Behav Brain Res. 2007;182:155–165.
Journal of Voice, Vol. ■■, No. ■■, 2017 12. Holmberg EB, Hillman RE, Perkell JS. Glottal airflow and transglottal air pressure measurements for male and female speakers in soft, normal, and loud voice. J Acoust Soc Am. 1988;84:511–529. 13. Merkus FM, de Jongste JC, Stocks J. Respiratory function measurements in infants and children. In: Gosselink R, ed. Lung Function Testing, European Respiratory Society Monographs. 2005:166–194. 14. Stephens RE, Bancroft A, Glaros AG, et al. Anatomic changes related to laryngeal descent from birth to 1 year of age: do they play a role in SIDS? Ear Nose Throat J. 2010;89:313–317. 15. Wermke K, Teiser J, Yovsi E, et al. Fundamental frequency variation within neonatal crying: does ambient language matter? Speech Lang Hear. 2016;doi:10.1080/2050571X.2016.1187903. 16. Wermke K, Ruan Y, Feng Y, et al. Fundamental frequency variation in crying of Mandarin and German neonates. J Voice. 2016;31:255.e25–255.e30. doi:10.1016/j.jvoice.2016.06.009. pii: S0892-1997(16)30124-2, Epub ahead of print. 17. Lind K, Wermke K. Development of the vocal fundamental frequency of spontaneous cries during the first 3 months. Int J Pediatr Otorhinolaryngol. 2002;64:97–104. 18. Wermke K. Neonatal crying behaviors. In: Wright JD, ed. International Encyclopedia of the Social & Behavioral Sciences. 2nd ed. Oxford: Elsevier; 2015:475–480. 19. Wermke K, Mende W, Kempf A, et al., Interaction patterns between melodies and resonance frequencies in infants’ pre-speech utterances, Proceedings of the Fourth International Workshop on Models and Analysis of Vocal Emissions for Biomedical Applications, University of Firenze, Italy, 2005 187–190. 20. MacLarnon AM, Hewitt GP. The evolution of human speech: the role of enhanced breathing control. Am J Phys Anthropol. 1999;109:341–363. 21. Borysiak A, Hesse V, Wermke P, et al. Fundamental frequency of crying in two-month-old boys and girls: do sex hormones during mini-puberty mediate differences? J Voice. 2017;31:128.e21–128.e28. doi:10.1016/ j.jvoice.2015.12.006. Epub 2016 Jan 14. 22. Behan M, Wenninger JM. Sex steroidal hormones and respiratory control. Respir Physiol Neurobiol. 2008;164:213–221. 23. Quast A, Hesse V, Hain J, et al. Baby babbling at five months linked to sex hormone levels in early infancy. Infant Behav Dev. 2016;44:1–10. doi:10.1016/j.infbeh.2016.04.002. Epub 2016 May 18. 24. Becker M, Oehler K, Partsch C, et al. Hormonal “minipuberty” influences the somatic development of boys but not of girls up to the age of 6 years. Clin Endocrinol (Oxf). 2015;83:694–701. doi:10.1111/cen.12827. Epub 2015 Jul 1. 25. Carey MA, Card JW, Voltz JW, et al. It’s all about sex: gender, lung development and lung disease. Trends Endocrinol Metab. 2007;18:308–313. Epub 2007 Aug 30. 26. Torday JS, Nielsen HC, Fencl Mde M, et al. Sex differences in fetal lung maturation. Am Rev Respir Dis. 1981;123:205–208. 27. Langston C, Kida K, Reed M, et al. Human lung growth in late gestation and in the neonate. Am Rev Respir Dis. 1984;129:607–613. 28. Patel M, Mohr M, Lake D, et al. Clinical associations with immature breathing in preterm infants: part 2-periodic breathing. Pediatr Res. 2016;80:28–34. doi:10.1038/pr.2016.58. Epub 2016 Mar 22. 29. Kelly DH, Stellwagen LM, Kaitz E, et al. Apnea and periodic breathing in normal full-term infants during the first twelve months. Pediatr Pulmonol. 1985;1:215–219. 30. Wilkinson MH, Skuza EM, Rennie GC, et al. Postnatal development of periodic breathing cycle duration in term and preterm infants. Pediatr Res. 2007;62:331–336. 31. Edwards BA, Sands SA, Berger PJ. Postnatal maturation of breathing stability and loop gain: the role of carotid chemoreceptor development. Respir Physiol Neurobiol. 2013;185:144–155. doi:10.1016/j.resp.2012.06.003. Epub 2012 Jun 13. 32. Kinney HC, Filiano JJ, Harper RM. The neuropathology of the sudden infant death syndrome. A review. J Neuropathol Exp Neurol. 1992;51:115– 126. 33. Krous HF, Beckwith JB, Byard RW, et al. Sudden infant death syndrome and unclassified sudden infant deaths: a definitional and diagnostic approach. Pediatrics. 2004;114:234–238.
ARTICLE IN PRESS Kathleen Wermke, et al
Inspiratory Phonation in Baby Voice
34. Kinney HC, Richerson GB, Dymecki SM, et al. The brainstem and serotonin in the sudden infant death syndrome. Annu Rev Pathol. 2009;4:517– 550. 35. Emery MJ, Krous HF, Nadeau-Manning JM, et al. Serum testosterone and estradiol in sudden infant death. J Pediatr. 2005;147:586–591.
7
36. Marschik PB, Kaufmann WE, Sigafoos J, et al. Changing the perspective on early development of Rett syndrome. Res Dev Disabil. 2013;34:1236– 1239. doi:10.1016/j.ridd.2013.01.014. Epub 2013 Feb 9. 37. de Bona C, Zappella M, Hayek G, et al. Preserved speech variant is allelic of classic Rett syndrome. Eur J Hum Genet. 2000;8:325–330.