Sex Differences in Velopharyngeal Anatomy of 9- and 10-Year-Old Children

Abstract

Objective:

Understanding the normal anatomy of velopharyngeal (VP) mechanism and the emergence of sexual dimorphism provides valuable insights into differences of VP anatomy among males and females. The purpose of this study is to examine sex differences in VP anatomy in a large data set of 3,248 9- and 10-year-old children.

Method:

Static three-dimensional magnetic resonance imaging was used to compare five VP characteristics including velar length, velar thickness, effective velar length, levator veli palatini muscle length, and pharyngeal depth between age-matched males (n = 1,670) and females (n = 1,578). Additionally, these dimensions were used to determine the VP ratio and effective VP ratio.

Results:

Males showed significantly larger dimensions for all VP distances and significantly lower ratios of velar length and effective velar length to pharyngeal depth (p < .05). The magnitude of these effect sizes was small to medium, with Cohen's d values ranging from 0.12 to 0.63. Additionally, the VP ratio and effective VP ratio are lower among males compared to females (p < .05).

Conclusions:

Results suggest the presence of sexual dimorphism in the VP mechanism among 9- and 10-year-old children. These findings emphasize the necessity of using different normative data for males and females when making comparisons to patients with cleft palate.

Up to 40% of children with a repaired cleft palate will develop hypernasal speech due to velopharyngeal insufficiency (VPI; Kane et al., 2002; Kotlarek et al., 2023; Perry et al., 2018; Perry, Snodgrass, et al., 2022; Tian et al., 2010). The decision for surgery to treat VPI is based on a speech assessment and imaging. Velopharyngeal (VP) magnetic resonance imaging (MRI) is rapidly becoming a clinical entity, primarily because it provides a reliable means to quantify key VP structures and muscles (Ettema et al., 2002; Haenssler et al., 2020; Kollara et al., 2016; Perry et al., 2018; Scott et al., 2014; Sitzman et al., 2022). A recent survey showed that 93% of cleft clinics in the United States feel that MRI is of significant value to the clinical assessment process (Mason, 2022). To determine whether differences in patient anatomy are of clinical significance, comparisons of patient anatomy are often made to normal anatomy. However, there are gaps in our current understanding of normative anatomy, particularly in the pediatric population.

Perry et al. (2014, 2016) documented significant differences in the VP structures and muscles based on race and sex among an adult population. The authors proposed that these findings suggest that comparisons of patient anatomy should therefore be race and sex specific. Although these differences among the adult population have been consistently demonstrated across the literature (Bae et al., 2011; Perry et al., 2014, 2016), sex effects among the pediatric population have been less clear and studies have presented with conflicting results. Vorperian et al. (2011) demonstrated significant prepubertal sexual dimorphism in the oropharyngeal and nasopharyngeal regions of the vocal tract among individuals between the ages of birth to 19 years. This study was followed by a series of studies that examined whether structures within those vocal tract regions further displayed sex effects. Kollara et al. (2016) demonstrated sex differences among 32 children 4–8 years of age and reported that males demonstrate a larger velar insertion distance than females. Expanding the sample size, Perry et al. (2018) reported significant differences between 43 males and 42 females ages 4–9 years in the primary muscle for velar elevation, that is, levator veli palatini (levator) muscle. Specifically, males 4–9 years of age demonstrated a significantly longer levator compared to females. Although adequate sample sizes were reached, based on an a priori power analysis, the cutoff in age was 9 years.

Perry et al. (2019) expanded the age range by examining 202 individuals between 4 and 21 years and observed no significant sex effect between VP variables within the age range of 4–10 years. However, a noted limitation by authors was the relatively small sample size across each age, particularly for those over 9 years. Additionally, the findings highlight the importance of examining sex effects with larger sample sizes and more discrete age groups, compared to the age span of 7 years used for grouping participants used by Perry et al. and Kotlarek et al. examined 200 infants between birth and 24 months of age and reported significant sex effects for several VP variables including angle of origin, effective velar length, levator length, and velar insertion distance. These findings raise question to results from earlier studies among the child population. More specifically, it is an unexpected finding that sex effects would be evident in infancy (Kotlarek et al., 2022), become absent during childhood (Kollara et al., 2016; Perry et al., 2018), and emerge later in adulthood (Perry et al., 2014, 2019). These observations further highlight the importance of assessing sex effects in small discrete age groups with adequate sample sizes.

Further examination of sex effects on key aspects of VP anatomy is needed, particularly above the age of 9 years, to understand whether observations by Perry et al. (2019) using smaller sample sizes are observed using larger sample sizes. If significant sex effects are apparent, normative data that are sex specific are vitally important to the clinical process. That is, comparisons of patient anatomy, such as clinical data obtained on a VP MRI for a child with VPI, should be sex specific. The purpose of this study is to examine sex differences in VP anatomy of 9- and 10-year-old children in a large data set of over 3,200 children.

Method

Study Cohort

For this study, we utilized existing data collected as part of the National Institutes of Health–funded Adolescent Brain Cognitive Development (ABCD; https://abcdstudy.org) study. This involves a decade-long 21-site longitudinal MRI study (Jernigan et al., 2018). A total of 3,248 participants, including 1,670 males and 1,578 females (self-identified) aged 9 and 10 years old were selected from the National Institute of Mental Health Data Archive. Data that were pulled from the ABCD database included whole head structural MRI data of participant at rest and demographic data to collect information related to age, race, sex, and height. Participants selected from the database and then used for this study were included if they self-identified as White, based on U.S. Health and Human Services designated categories and were either 9 or 10 years of age at the time of the structural MRI. All participants spoke English, which was an inclusion criterion for the ABCD study cohort. Additionally, only children who were determined to have normal anatomy were included in this study. Inclusion of participants as part of the ABCD study (https://abcdstudy.org) met several inclusion and exclusion criteria, which were used to determine the child had VP anatomy that was likely representative of normal anatomy. These specific criteria included absent history of major neurological disorders, tumors, brain abnormalities, medical or neurological conditions or syndromes, normal gestational age with no major birth complications, and absent diagnosis of disabilities or disorders (see https://abcdstudy.org for comprehensive listing). Speech recordings were not part of the database and, thus, resonance was not able to be assessed formally and no data are included on these participants.

MRI

Participants were imaged across multiple MRI sites using the same MRI protocol and sequence parameters (Casey et al., 2018). For this study, transverse relaxation time (T2)–weighted static three-dimensional MRI data were used due to the high contrast between the muscle and other tissue structures. All MRI data included the whole head with the child awake, nonsedated, in the supine position, and without contrast. MRI data were obtained across Siemens, Philips, and General Electric scanners with identical matrix size (256 × 256), field of view (256 × 256), percent field of view phase (100%), and image resolution (1.0 mm × 1.0 mm × 1.0 mm). This resulted in an overall image acquisition that was similar across all scanners with images T2 weighted, producing a high contrast between tissue types. The impact of gravity while in the supine position has shown to have a negligible impact on VP structures (Kollara & Perry, 2014). During the scan, the velum was at rest, in a relaxed and lowered position. Participants were excluded if motion was observed in the image or image artifacts obscured the view of the anatomy of interest.

Image Analyses

Digital Imaging and Communication in Medicine (DICOM) files were analyzed using Amira 4 Visualization Volume Modeling software (Visage Imaging GmbH). The DICOM support system enables the data to preserve original geometry. Midsagittal image plane was determined as the image most clearly depicting the velar midline as noted by the presence of the anterior nasal spine, posterior nasal spine, maximum velar length, fourth ventricle, and genu of the corpus callosum (Ettema et al., 2002). The oblique coronal image plane used for measurements was determined by rotating the oblique slice placing the slice through the bulk of the velar eminence and in the plane, which shows the levator muscle from origin to insertion (Ettema et al., 2002). Measures and rations are described and illustrated in Table 1. Across all participants, we measured five VP characteristics including velar length (curvilinear line from posterior nasal spine to tip of uvula), velar thickness (distance from velar knee to velar dimple), effective velar length (distance between the posterior nasal spine and the point of insertion of the levator muscle into the velum as observed from the intersection of the oblique coronal image overlaid onto the midsagittal image), levator muscle length (from origin at the base of the skull to the insertion of the muscle into the velum and recorded as mean of right and left muscle bundles), and pharyngeal depth (distance from posterior nasal spine to posterior pharyngeal wall drawn parallel to the palatal plane, a reference line from anterior nasal spine through posterior nasal spine) and used these dimensions to calculate the VP ratio (velar length divided by pharyngeal depth) and effective VP ratio (effective velar length divided by pharyngeal depth). These variables were selected because they are critical to normal VP function, that is, the velum must be of adequate length and thickness to create velar stretch to make contact against the posterior pharyngeal wall. Additionally, the levator muscle is the primary muscle responsible for elevation and retraction of the velum to overcome the pharyngeal depth and create an airtight seal to separate the oral and nasal cavities during oralized sound production. These variables were also selected, because previous studies have shown differences in these variables among males and females (Perry et al., 2018; Perry et al., 2019) and are thus of particular interest for this study. Last, studies have suggested that these variables are likely linked to improved outcomes in VPI surgical planning and are of clinical importance in VPI assessments (Kao et al., 2008; Perry, Snodgrass, et al., 2022; Perry, Williams, et al., 2022).

Table 1.

Demonstration of the five velopharyngeal (VP) variables and two ratios calculated from the (A) sagittal image and (B) oblique coronal image. Note that the oblique coronal image plane is obtained by sampling a plane from the sagittal image as shown in the figure legend on Image B, which courses through the body of the levator muscle.

Demonstration of measure	Variable	Description
	Velar length (combined red and white line in Image A)	Curvilinear line drawn from the posterior nasal spine to the tip of the uvula
	Velar thickness (line perpendicular to the velar length in Image A)	Distance from the velar knee to the velar dimple, obtained at the point where the levator muscle inserts into the velum
	Effective velar length (red line in Image A)	Distance between the posterior nasal spine and the point of insertion of the levator muscle into the velum as observed from the intersection of the oblique coronal image overlaid onto the midsagittal image
	Levator veli palatini length (labeled “levator” in Image B)	Length of the levator veli palatini from its origin to its insertion at the velum, taken as mean of right and left muscle length
	Pharyngeal depth (labeled as “PNS–PPW” [posterior nasal spine–posterior pharyngeal wall] in Image A)	Distance from the PNS to the PPW along a plane continuous from the palatal plane (line coursing through anterior nasal spine through posterior nasal spine)
	VP ratio	Calculation obtained from dividing velar length by pharyngeal depth
	Effective velopharyngeal ratio	Calculation obtained from dividing effective velar length by pharyngeal depth

Note. VP = velopharyngeal; EVP = effective velopharyngeal.

Statistical Analyses

MRI DICOM images across the 3,248 participants were pulled from the MRI database and used to collect measures representing key VP structures at rest. A rater with over 5 years of experience in evaluating VP MRI data performed measures across the participants. Males and females showed no differences in mean age (p = .59). For statistical analysis, independent samples t tests were used to compare means of each VP measurement between males and females. For added assurance, we also ran Wilcoxon rank-sum tests for the two ratios, because these typically shown nonnormal distributions. Nominal statistical significance was a set at p ≤ .05. We set our study-wide adjusted significance at p ≤ .007 (.05/7, based on a Bonferroni correction by the number of comparisons made for the seven study variables), which is conservative given that the variables are likely correlated to some degree. We calculated effect sizes (Cohen's d) to help facilitate interpretation of the statistical results.

Reliability

Intrarater reliability was conducted by remeasuring 20% of the data between 5 and 7 months following the initial measurements. A two-way interclass correlation coefficient mixed effects model was used to evaluate intrarater reliability across the five VP variables. Results demonstrated intrarater reliability ranged from .82 to .94 (r = .89 for levator length, r = .87 for velar length, r = .94 for posterior nasal spine [PNS] to posterior pharyngeal wall [PPW], r = .84 thickness of velum, and r = .82 for effective velar length). Within the same rater, two repeated measured agreed 89% for levator length, with an absolute mean difference of 1.333 mm (SD = 1.332), agreed 87% for velar length, with an absolute mean difference of 0.032 mm (SD = 2.18), agreed 94% for PNS to PPW, with an absolute mean difference of 1.055 mm (SD = 2.022), agreed 84% for thickness of velum, with an absolute mean difference of 0.181 mm (SD = 0.913), and agreed 81% for effective velar length, with an absolute mean difference of 0.93 mm (SD = 2.103).

To assess interrater reliability, a second rater with 4 years of experience in MRI data analyses randomly selected and remeasured 20% of the data 5 months after the initial measure. Results demonstrated interrater reliability ranged from .74 to .89 (r = .74 for levator length, r = .84 for velar length, r = .89 for PNS to PPW, r = .76 thickness of velum, and r = .80 for effective velar length). Absolute mean differences in measurements were α = 0.76 for levator length, α = 0.92 for velar length, α = 0.92 for PNS to PPW, α = 0.78 for thickness of velum, and α = 0.84 for effective velar length. Two raters agreed 74% for levator length, with an absolute mean difference of 0.889 mm (SE = 0.818); agreed 84% for velar length, with an absolute mean difference of 1.68 mm (SE = 0.872); agreed 89% for PNS to PPW, with an absolute mean difference of 1.26 mm (SE = 1.043); agreed 76% for thickness of velum, with an absolute mean difference of 0.419 mm (SE = 0.371); and agreed 80% for effective velar length, with an absolute mean difference of 0.933 mm (SE = 0.69). Interrater and intrarater reliability ranged from r = .74 to .94 across VP variables. Interrater and intrarater reliability ranged from r = .74 to .94 across all VP variables.

Results

Males (M = 119.6 months; SD = 7.36 months) and females (M = 119.46 months; SD = 7.37 months) showed no differences in mean age (p = .59). Height of participants at the time of the MRI were collected from the ABCD database and were found to be similar, t_{(df = 3116.5)} = 0.0066609, p = .9947, between males (M = 55.35 in., SD = 2.79 in.) and females (M = 55.35 in., SD = 3.23 in.). In six males, the image quality was too poor (due to motion artifacts) to reliably measure the levator length, which also impacted the measure of effective velar length in one male participant. Additionally, velar thickness was unable to be measured in one male and two females due to motion artifacts creating poor image quality to delineate the exact boundary of the velum. These measures were removed from the analysis, as noted in Table 2.

Table 2.

Descriptive and test statistics.

Subjects/measures			t	df	p	Effect size (d)c
Subjects: (n) Velar length: M (SD)	Male: (1,670)	Female: (1,578)
	27.33 mm (3.83)	26.74 mm (3.66)	−4.465	3245.6	8.279e−06	−0.23
Subjects: (n) Effective velar length: M (SD)	Male: (1,669)	Female: (1,578)
	9.65 mm (3.48)	9.36 mm (3.4)	−2.424	3241.2	0.015*	−0.12
Subjects: (n) Pharyngeal depth: M (SD)	Male: (1,670)	Female: (1,578)
	17.9 mm (5.2)	16.8 mm (5.16)	−6.024	3238.3	1.886e−09	−0.30
Subjects: (n) Levator veli palatini muscle length: M (SD)	Male: (1,664)	Female: (1,578)
	38.93 mm (3.29)	37.54 mm (3)	−12.509	3235.1	< 2.2e−16	−0.63
Subjects: (n) Velar thickness: M (SD)	Male: (1,669)	Female: (1,576)
	8.78 mm (1.41)	8.53 mm (1.39)	−5.107	3237.3	3.447e−07	−0.25
Subjects: (n) VP ratio: M (SD)	Male: (1,670)	Female: (1,578)
	1.73 (1.01)	1.82 (1.06)	2.493a	3210	0.012*	0.12
Subjects: (n) EVP ratio: M (SD)	Male: (1,669)	Female: (1,578)
	0.58 (0.33)	0.61 (0.39)	2.614b	3076.9	0.008*	0.13

Note. df = degrees of freedom; VP = velopharyngeal; EVP = effective velopharyngeal.

^*a < .05.

^aWilcoxon test statistics: W = 1418383, p = .0001418.

^bWilcoxon test statistics: W = 1378520, p = .01916.

^cNegative d values indicate males larger than females; positive d values indicate males smaller than females.

Males showed significantly larger dimensions for all VP distances and significantly lower VP and effective VP ratios (p < .05) as see in Table 2 and graphically displayed in Figure 1. Extreme outliers were observed for levator muscle length (one extreme outlier) and velar thickness (two extreme outliers). In both cases, statistical tests were run with and without outliers and the removal of outliers did not impact the test observations; therefore, outliers were included in the final analysis. Differences for four of the VP dimensions also exceeded our strict study-wide significance threshold (p < .007), which was selected based on a Bonferroni correction for the number of comparisons made across the seven selected variables. The magnitude of these effect sizes was small to moderate, with Cohen's d values ranging from 0.12 to 0.63. Specifically, males showed a significantly longer (males, mean velar length = 27.33 mm; females, mean velar length = 26.74 mm) and thicker velum (males, mean velar length = 8.78 mm; females, mean velar length = 8.5 mm), both showing a small effect size. Males also displayed a greater effective velar length (males, M = 9.65 mm; females, M = 9.36 mm), pharyngeal depth (males, M = 17.9 mm; females, M = 16.8 mm), both with small effect size. The levator muscle was significantly longer in males (M = 38.93 mm) compared to females (M = 37.54 mm), showing a medium effect size. Last, VP ratio (W = 1418383, p = .0001418) and effective velopharyngeal (EVP) ratio (W = 1378520, p = .01916) were significantly lower for males than for females using both parametric (t tests) and nonparametric (Wilcoxon test) statistics, due to variations observed in the velum and greater pharyngeal depth.

Figure 1.

Box plot display of sex differences between velopharyngeal (VP) measurements and VP ratios. For the boxes, the horizontal lines represent the median, the lower whiskers represent the 25th percentile (−1.5* interquartile range), and the upper whiskers represent the 75th percentile (+1.5* interquartile range). Individual points are outliers. EVP = effective velopharyngeal.

Discussion

Pervasive sex differences are present in the VP region in 9- and 10-year-old children. Males showed significantly larger dimensions for all distances, while females showed significantly higher VP and EVP ratios. Although the observed sex differences were statistically significant using a conservative p value, the magnitude of sex effect on VP measures was generally small. However, even small differences (e.g., less than 1 mm), may be clinically relevant for the VP mechanism, where a small gap can produce VPI. To illustrate this point, Sitzman et al. (2023) compared 40 patients with nonsyndromic cleft palate and VPI to 40 participants with normal anatomy and speech. Using five of the same VP variables as used in this study, differences between groups were under 1 mm for three of the five variables, specifically velar thickness, pharyngeal depth, and effective VP ratio with statistically significant differences between groups for velar length and effective VP ratio. This study highlights the likelihood that differences in VP anatomy among those with and without VPI can be very small, but that such small changes can be directly related to perceived hypernasality among those with VPI. Future research is needed to fully understand the impact of these subtle variations in VP anatomy and the influence of sex and ancestry.

It is unlikely that the VP sex effects observed in this study are driven by general body size differences. Centers for Disease Control and Prevention (CDC) height charts show no difference in standing height for males and females at this age (CDC, 2017). Comparisons of our sample also reflect a nonsignificant difference in height between males and females. Samples are also well matched for age, so this is unlikely to drive any of the observed sex effects.

Findings in this study are consistent, in part, with other studies examining sexual dimorphism of key VP variables in subadults. Specifically, Perry et al. (2018) analyzed VP measures in 85 children 4–9 years of age and observed sexual differences only for the levator length, with males showing a longer muscle compared to females. Our findings among 3,248 children 9 and 10 years of age confirmed this result. Unlike this study, Perry et al. (2019) found no significant sex effect for additional VP variables in 103 children between 4 and 10 years of age. These prior results, however, may have been impacted by combining children across such a wide age range. Results from this study use a much narrower age range and demonstrate that sexual differences are present for all VP variables by the age of 9 and 10 years. However, our study is still not able to pinpoint the age at which these sex differences initially emerge. To accomplish this, well-powered longitudinal studies spanning the full range of childhood and adolescence will be needed.

Our findings suggest that many characteristic sex differences in the VP apparatus may be present prior to the onset of puberty. The ages represented in this study are likely on the cusp of puberty, with multiple studies documenting the trend toward earlier puberty worldwide (Eckert-Lind et al., 2020). Nevertheless, as prior studies have documented sex differences in the morphology of the VP region and the craniofacial complex more broadly in young children (Kesterke et al., 2016; Matthews et al., 2016; Perry et al., 2018), this would seem to point to factors operating prior to the onset of puberty. Numerous studies have shown that hormonal differences between males and females are present prior to puberty and that these are associated with prepubertal sexual dimorphism in body composition (Garnett et al., 2004; Ortega-Avila et al., 2022). Although we are unable to test for such relationships in this study, prepubertal hormonal difference should be considered as one possible factor driving early sex differences in VP morphology.

Results from this study also provide normative values for key VP variables in children 9 and 10 years of age, which will provide valuable insights into VP MRI methods used to assess VPI in patients. This age range has been underrepresented in the literature as prior studies have reported small sample sizes within this age range (Perry et al., 2019). The observation of sex effects in this population further underscores the need for patient comparisons to be sex specific when using MRI to assess VP structures.

The differences observed between males and females in this study could have implications on the dynamic function of the VP mechanism. For example, the sex differences observed may impact VP closure patterns and velar stretch for speech. Previous research has indicated that individuals with longer velums are more likely to present with a coronal closure pattern for speech, while individuals with shorter velums are more likely to present with circular closure patterns, meaning males at this age may be more likely to present with a coronal closure pattern while females are more likely to have a circular closure pattern at this age (Jordan et al., 2017). Previous literature has also indicated individuals with larger effective VP ratios require less velar stretch to achieve VP closure (Tian & Redett, 2009). This could indicate females would require more velar stretch than males at this age to achieve complete VP closure.

The presence of sexual dimorphism in the VP complex among children in this age range suggests that different normative data should be used for males and females when making comparisons with clinical populations for VP MRI procedures. Clinical VP MRI is becoming more common. As proposed by Perry, Snodgrass, et al. (2022), VP MRI can be used in children who present with VPI to determine the structural and functional causes of the perceived differences in the individual's speech (e.g., hypernasality and/or nasal air emission). Additionally, VP MRI provides valuable information about the different features of the levator muscle, VP closure pattern, VP gap size, and adenoid pad size (Perry, Snodgrass, et al., 2022). All of this anatomic information influences surgical decisions and can have a positive impact on the outcomes of VPI surgery. Although static MRI has limitations in evaluating an individual's speech, authors propose the use of normative data as a source of comparison to guide the cleft team's understanding of how patient anatomy differs from control data. Furthermore, VP atlases have been constructed and proposed for assessing VP function, which rely on control data to construct a data bank, which is then used to create a normalized model of VP structure (Xing et al., 2021). In all cases, normative data are proposed as the comparison to patient anatomy to further understand variations in cleft speech. This study emphasizes the importance of ensuring data at these young ages are sex matched. Further researcher is needed to examine these same-sex effects with a large sample size at other ages and determine if these sex differences are noted in clinical populations, such as those with a history of cleft palate and/or VPI, given these differences could play a role in the surgical decision-making process. For example, findings from this study indicate that males have longer velums, which has previously correlated with coronal closure pattern for speech (Jordan et al., 2017). Individuals with coronal closure patterns typically are not ideal candidates for certain surgeries, such as the pharyngeal flap (Jordan et al., 2017). It is also noted that individuals with greater EVP ratios require less stretch than their counterparts, and smaller VP ratios have been hypothesized to negatively impact VP competence in patients with VPI (Jordan et al., 2017; Tian & Redett, 2009). EVP ratio is an important factor when considering which patients with VPI are good candidates for palatal re-repair for VPI management, where it is hypothesized individuals with longer effective velums and typical pharynx size (e.g., do not have the profile of “deep pharynx” that is associated with 22q11.2 deletion syndrome) are more ideal candidates for palatal re-repair surgeries (Williams et al., 2023).

Although the results of this study are informative, they are not free of limitations. First, it should be noted that, in addition to the small-effect sizes, the observed mean between sex differences was smaller than the image resolution (1 mm/pixel). This underscores the importance of reliability and examining measurement error, which should be included in all MRI studies examining VP anatomy where differences may be small. This also highlights the importance of reporting the voxel size and specifications used when conducting VP MRI studies to ensure that image resolution is adequate to detect the level of difference that is observed across groups. Another limitation is that this study examined only self-reported Whites from the United States. Future studies are needed to examine sexual dimorphism of VP variables across different ethnic, racial, and ancestry groups due to the known differences in the timing of maturation across populations (Sun et al., 2002). In addition, static MRI was used to evaluate the VP variables. Due to the limitations using static MRI (Mason & Perry, 2017), future studies are needed to measure VP variables using dynamic MRI to assess musculature changes of VP variables during speech tasks. Because this database did not include dynamic MRI data, only static assessments could be performed. There are limited insights regarding the articulation patterns or behaviors among participants used in this study since data were not prospectively collected. Because this study used only static rest condition, this is likely of limited value. However, future studies examining dynamic VP function should carefully consider and collect speech data. The sex differences in dynamic VP activities during speech should be a focus for future investigations. Due to the absence of speech and resonance recordings in this database, it is important to consider this limitation when interpreting the findings of the study.

Conclusions

In summary, this study highlights sexual dimorphism in the VP region in 9- and 10-year-old children. Results from this study demonstrate larger dimensions of velar length, velar thickness, effective velar length, levator veli palatini muscle length, and pharyngeal depth among males compared to females. Additionally, the VP ratio and EVP ratio are lower among males compared to females. Future studies are needed to examine sex differences of VP variables among different racial groups using dynamic MRI and to determine magnitude of sex differences in clinical populations (e.g., cleft palate and VPI).

Data Availability Statement

The data underlying this article will be shared on reasonable request to the corresponding author.

Funding Statement

The data used in the preparation of this article were obtained from the Adolescent Brain Cognitive Development (ABCD) Study (https://abcdstudy.org), held in the National Institute of Mental Health Data Archive. This is a multisite, longitudinal study designed to recruit more than 10,000 children ages 9–10 years and follow them over 10 years into early adulthood. The ABCD Study is supported by the National Institutes of Health (NIH) and additional federal partners under awards U01DA041048, U01DA050989, U01DA051016, U01DA041022, U01DA051018, U01DA051037, U01DA050987, U01DA041174, U01DA041106, U01DA041117, U01DA041028, U01DA041134, U01DA050988, U01DA051039, U01DA041156, U01DA041025, U01DA041120, U01DA051038, U01DA041148, U01DA041093, U01DA041089, U24DA041123, and U24DA041147.