Restless senses, restless sleep: Exploring the sensory–sleep link in Phelan-McDermid syndrome. A Cross-Sectional Exploratory Study

Abstract

Background

Phelan-McDermid syndrome (PMS) is a rare genetic condition caused by SHANK3 gene haploinsufficiency. Individuals with PMS often exhibit developmental delays, language deficits, and neurological and psychiatric symptoms that frequently overlaps with autism. Sleep disturbances and sensory reactivity differences are also highly prevalent in this population, yet their relationship remains unexplored. Given prior research in autism, which has identified links between sensory sensitivities and sleep difficulties, this study aims to investigate the association between sensory processing differences and sleep disturbances in children with PMS.

Methods

We used 2 standardised questionnaires: The Children’s Sleep Habit Questionnaire (CSHQ) and the Child Sensory Profile 2 (SP-2) to measure sleep behaviour and sensory reactivity differences in families with children with PMS (n=40) (median age of caregivers = 42.95, median age of children = 10.52 years). We performed descriptive statistics, spearman correlations and stepwise regressions to examine relationships between sensory reactivity and sleep quality.

Results

Findings revealed significant correlations between sleep difficulties and sensory sensitivity, sensory registration, and sensory seeking. Further analysis of specific sensory modalities identified correlations between sleep difficulties and touch, movement and oral processing. Stepwise regression demonstrated that sensory seeking and oral sensitivity were significant predictors of sleep disturbances, explaining 19.9% and 34.6% of the variance, respectively.

Conclusions

This study highlights a significant relationship between sensory reactivity differences and sleep disturbances in children with PMS. Findings suggest that sensory processing differences, particularly oral sensitivity and sensory seeking, contribute to sleep difficulties. These insights underscore the need for targeted interventions to improve sleep quality in individuals by addressing sensory sensitivities.

Introduction

Phelan-McDermid syndrome (PMS) is a genetic disorder caused by haploinsufficiency of the SHANK3 gene, resulting either from a deletion at 22q13.33 that includes SHANK3 or from a pathogenic variant within the gene. Clinically, PMS is characterized by global developmental delay, severe language impairment, and a broad spectrum of neurological and psychiatric features [1–3]. In addition, clinical observations in humans consistently report difficulties in sleep initiation and maintenance, suggesting a strong association between SHANK3 haploinsufficiency and altered sleep patterns [1, 3, 4]. Although the underlying mechanisms in humans are not yet fully understood, experimental evidence indicates that SHANK3 plays a role in the regulation of sleep and circadian-related processes, providing a potential biological basis for these clinical observations [5].

PMS is also strongly associated with autism spectrum disorder (ASD), with approximately 70–80% of individuals meeting Diagnostic and Statistical Manual of Mental Disorders (DSM) criteria for autism, although estimates vary depending on assessment methods [6]. In fact, PMS is considered one of the most common monogenic forms of ASD and may account for up to 2% of ASD cases [7, 8]. Despite this substantial overlap, PMS presents a distinct neurobehavioral and sensory phenotype compared to idiopathic autism [9] which highlight the relevance and the limitations of using ASD as a comparative framework when studying PMS.

Recent studies indicate that more than 40% of individuals with PMS present with at least one of the following: neurological disturbances, sleep disorders, or neuroanatomical differences [10]. Sleep is essential for neural development, and inadequate sleep can lead to attentional, mood, and behavioral disturbances, affecting daily performance and participation in day-to-day activities [11, 12]. Sleep disturbances are common in PMS, with studies reporting prevalence rates ranging from 44% to 46% based on clinical and cohort data [6, 12]. However, higher rates have been reported in studies using detailed parent-reported measures. For example, Bro and colleagues [4] found that 89.9% of PMS participants exhibited sleep disturbances, including bedtime resistance, nocturnal awakenings lasting more than 15 min, parasomnias, and difficulties falling back asleep. These discrepancies likely reflect differences in assessment methods and definitions of sleep disturbances. Similarly, Smith-Hicks et al. [11] identified that individuals with PMS experience significant sleep disturbances related to nocturnal awakenings, parasomnias, and daytime sleepiness. Despite these findings, the specific types of sleep disturbances and their contributing factors remain underexplored in the context of PMS.

Children with PMS often face sensory integrative challenges, including sensory reactivity differences such as hyperreactivity (aka sensory over-responsivity, sensory sensitivity, sensory avoiding), hyporeactivity (aka sensory under-responsivity, low registration), and sensory seeking (aka sensory craving). Sensory reactivity is crucial for maintaining alertness and focus on tasks. Although sensory seeking behaviours may sometimes be perceived as pleasurable, sensory reactivity differences can also interfere with daily living activities and social interactions [12–17]. Sensory reactivity profiles in PMS typically show more hyporeactivity compared to autistic children and more sensory seeking compared to typically developing children [9].

In autism, sensory reactivity differences have been associated with sleep disturbances [18]. The variability of sleep and sensory differences in PMS underscores the need for further research into their interrelation. Although there is a lack of literature specifically addressing the relationship between sensory differences and sleep disturbances in PMS, as mentioned studies in autism provide valuable insights. Manelis-Baram et al. [19] examined the relationship between sleep and sensory reactivity in a sample of pre-school children with autism spectrum disorder (ASD), with a mean age of approximately 3 years at baseline and 4.5 years at follow-up, identifying a strong correlation between sensory sensitivity and both nocturnal sleep time and total sleep time. They concluded that sleep disturbances and the severity of sensory reactivity differences were related and possibly associated through a common physiological mechanism.

Similarly, Mazurek and Petroski [20] reported that sensory hyperreactivity was associated with delayed sleep onset, sleep duration, and nighttime awakenings in young children (2–5 years), while in older children (6–18 years), it was related to various sleep issues except nighttime awakenings. Looking within specific sensory systems, tactile sensitivity appears particularly relevant. Tzischinsky et al. [21] observed that differences in tactile and oral sensitivity were linked to total sleep disturbances, with tactile hyperreactivity explaining 24% of the variance in total sleep disturbance scores in autistic children. Jamioł-Milc et al. [22] investigated tactile modulation conditions, finding that tactile hypo-reactivity may also contribute to insomnia and poor-quality sleep, although more research is needed due to small sample sizes. Differences in sensory reactivity in autism encompass a range of findings, such as hyper- and hypo-reactivity and sensory seeking. Researchers report differences in auditory filtering and taste and smell differences [23], tactile hypo-reactivity [22, 24], vestibular and oral hyper-reactivity [25], vestibular hypo-reactivity [24], and tactile and oral hyper-reactivity (Tzischinsky et al., 2018) [21]. In adults with autism, Hohn et al. [26] suggested that visual sensory sensitivity is a driver of insomnia. Given the phenotypic overlap between PMS and idiopathic autism, particularly in sensory and sleep domains, these findings provide a relevant framework for investigating similar associations in PMS, while acknowledging important neurobiological differences between the conditions.

Sleep difficulties and sensory processing differences in Phelan-McDermid Syndrome (PMS) are highly variable. Current evidence suggests that the sensory profile of children with PMS is characterized by greater hyporeactivity and fewer hyperreactivity and sensory seeking symptoms compared to children with idiopathic autism. Sensory hyporeactivity may be related to sleep difficulties because reduced responsivity to internal and external sensory cues (e.g., light or circadian signals) may impair arousal regulation and the detection of sleep–wake transitions, leading to different sleep patterns (Mazurek & Petroski, 2015) [20]. In this context, the present study aims to explore the relationship between sleep and sensory reactivity in this population.

Given the limited research on this topic, an exploratory approach was adopted. Rather than testing specific hypotheses, the objective is to examine potential associations between sensory processing differences and sleep disturbances. Several explanatory frameworks may account for these associations, including shared underlying physiological mechanisms or the influence of third variables, such as overall phenotypic severity. Accordingly, findings should be interpreted as preliminary and hypothesis-generating.

Methods

Participants

In total 43 participants took part in this study, which was reduced to 40 after excluding those who did not meet inclusion criteria or had incomplete questionnaire data. Inclusion criteria required be parents of children aged 2–18 years with a confirmed PMS diagnosis. Although inclusion criteria required that children had a confirmed diagnosis of Phelan-McDermid syndrome (PMS), detailed information regarding the specific genetic diagnostic methods (e.g., Chromosomal Microarray Analysis [CMA], karyotyping, or Whole Exome Sequencing [WES]) was not systematically collected. This is due to the nature of the recruitment strategy (i.e., anonymous online survey distributed via patient organisations and social media), which limited access to detailed clinical and genetic records. Participants were recruited using opportunity sampling through advertisements distributed via Phelan-McDermid Syndrome Foundation (PMSF) charities such as PMSF UK, PMSF US, PMS Spain, Seaver Autism Centre and social media platforms on PMS based Facebook groups or displayed on Instagram or flyer advertisements. Interested individuals were provided a link to an online survey hosted on REDCap.

Recruitment materials described the study as focusing on sleep behaviours and sensory processing in children with PMS. Participation was voluntary and recruitment occurred through patient organisations and social media platforms across multiple countries; therefore, the total number of individuals exposed to study advertisements is unknown, and response rates could not be calculated. As a result, families with greater concerns about sleep or sensory difficulties may have been more likely to participate, introducing potential selection bias.

Demographic data was collected including caregiver and child age, gender, child medication use, and current treatments to characterise the sample and provide context for interpreting the findings. All responses were anonymised and linked using REDCap ID codes. The final sample consisted of 18 males and 22 females. The mean age of caregivers was 42.95 (SD = 8.850), while for children it was 10.52 years (SD = 5.786).

Procedure

Before participation, individuals interested reviewed an information sheet detailing the study aims and procedures. Informed consent was then obtained through an online consent form, ensuring participants understood how their data was going to be collected, processed, and stored. Data collection took place between April 2024 and June 2024.

Parents of children with Phelan-McDermid Syndrome completed a series of quantitative online questionnaires assessing their children’s sensory characteristics and sleep hygiene habits. Clinical variables such as medication use, such as melatonin, were collected (6 children were using melatonin). No other detailed clinical variables such as comorbid conditions (e.g., epilepsy, anxiety, ADHD), or level of intellectual functioning were systematically collected as part of the dataset. Specifically, the study utilised the Child Sensory Profile 2 and the Children’s Sleep Habits Questionnaire (CSHQ). These assessments were designed to evaluate sensory reactivity differences and sleep behaviours in children with Phelan McDermid Syndrome. Parents were asked to reflect on their child’s sleep behaviours occurring over a ‘typical’ week recently.

Measures

The Children’s Sleep Habits Questionnaire (CSHQ) [27]: is a parent reported questionnaire designed to assess several key sleep domains relevant to children. These domains include bedtime behaviour and sleep onset; sleep duration; anxiety around sleep; behaviour occurring during sleep and night wakings; sleep-disordered breathing; parasomnias; and morning waking/daytime sleepiness. Parents were asked to recall sleep behaviours occurring over a “typical” recent week. The CSHQ has been widely used in numerous studies examining sleep behaviour and has demonstrated acceptable psychometric properties [28] as well as having been validated and translated across different cultural contexts [29–35]. A total score above 41 is considered the clinical cut-off for identifying children with clinically significant sleep difficulties or a possible paediatric sleep disorder.

The Child Sensory Profile 2 [36]: is an 86-item questionnaire that evaluates sensory characteristics, such as: ‘My child becomes irritated by wearing shoes or socks’. This widely recognised tool is frequently used to measure sensory reactivity differences, particularly in children with neurodevelopmental conditions such as autism and PMS [13]. Participants completed the questionnaire online, which took approximately 10 min. The child sensory profile has demonstrated both good internal consistency across quadrant (αs = 0.89-0.95) and factor scores (αs = 0.82-0.93) [37].

Ethical considerations

All participants provided informed consent prior to participating in the study. Each participant first received an information sheet detailing the aims, procedures, and data management plan of the research, and only after reviewing it did, they give their consent. For children involved in the study, consent was obtained from their parent or legal guardian. Patients and their families were fully informed about the purpose and scope of the research, and participation was entirely voluntary, with the right to withdraw at any point and request removal of their data.

The study was conducted in accordance with the Declaration of Helsinki and was approved by the Ethics Committee Research of Reading University (2023-222-TT) and Rey Juan Carlos University (ENM103/202412202101122, 26 January 2022). Given the nature of the study—an anonymous online survey with no clinical intervention—it does not fall under the definition of a clinical trial; therefore, no trial registration was required or conducted.

Data management and privacy

Data collection was carried out using REDCap data collection software. All information was anonymised, with participant ID numbers used to pseudonymise identities and link data across questionnaires. Data were securely stored in a password-protected database and preserved in the University of Reading’s Research Data Archive (UOR Research Data Archive). Although pseudonymised, data may be accessed by authorised individuals beyond the principal investigator, in line with institutional data governance protocols.

Personal data will not be retained longer than necessary for the study’s aims. Periodic reviews are scheduled to assess whether data can be safely deleted.

Analysis

All data analyses were conducted using IBM SPSS^®. Individuals with incomplete assessment data were excluded prior to analysis. Descriptive statistics were computed for all demographic variables. The significance level was set at α = 0.05 for all analyses.

The aim of the study was to explore whether sleep difficulties in Phelan-McDermid syndrome (PMS) are associated with sensory reactivity differences, as previously observed in autistic populations. To examine these relationships, Spearman correlation analyses were conducted between sensory reactivity and sleep subscale scores.

To further explore potential associations, stepwise regression analyses were performed to identify sensory variables associated with sleep outcomes (CSHQ subscales). Variables included in regression models were selected based on prior correlation analyses.

Given the exploratory nature of this study and the relatively small sample size, regression analyses were conducted to identify potential associations rather than to test predefined hypotheses. Although the stepwise approach was used to reduce the number of predictors, this method has known limitations, including potential bias in coefficient estimation and model instability. Therefore, results from these analyses should be interpreted cautiously and considered hypothesis-generating.

No correction for multiple comparisons was applied, as analyses were exploratory in nature and aimed to identify potential associations rather than test predefined hypotheses. Given the relatively large number of correlations examined, this approach increases the risk of Type I error; therefore, findings should be interpreted with caution and considered hypothesis-generating. Confidence intervals (95%) were calculated for both correlation coefficients and regression parameters to provide estimates of effect size precision. Confidence intervals (95%) were calculated for both correlation coefficients and regression parameters to provide estimates of effect size precision.

To further characterise relationships within each instrument, within-form correlations among sensory and sleep subscales were examined. In addition, multicollinearity among predictors included in regression models was assessed using Variance Inflation Factors (VIFs).

Results

Descriptive statistics

Overall, 40 participants took part excluding outliers. Descriptive statistics (Table 1) summarise raw sensory and sleep scores, while Table 2 presents the distribution of sensory scores relative to normative data. Descriptive statistics (Table 1) were computed to summarise the central tendency and variability of the variables in the study. Results showed seeking total score (M = 46.95, SD = 21.36), avoiding total score (M = 36.60, SD = 17.93), sensitivity total score (M = 42.60, SD = 17.39), registration total score (M = 53.25, SD = 19.02), auditory total score (M = 18.25, SD = 8.93), and visual total score (M = 11.70, SD = 7.12). These values provide insight into the distribution of PMS symptoms and sleep quality within the sample. Furthermore, Table 2 contextualises these sensory profiles relative to normative data. Sensory seeking was more prevalent in 50% of individuals with PMS compared to a normed sample, sensory hyperreactivity (as reflected in sensory avoiding) was more common in 25%, and sensory hyporeactivity (as reflected in low registration scores) was prominent in 70% of the sample.

Table 1.

Means and standard deviations on sensory reactivity scores in participants Note: CHSQ: Children Sleep Habits Questionnaire

Sensory Profile		N	Mean	Std. Deviation
	Seeking Total Score	40	46.95	21.36
	Avoiding Total Score	40	36.60	17.93
	Sensitivity Total Score	40	42.60	17.38
	Registration Total Score	40	53.25	19.02
	Auditory Total Score	40	18.25	8.92
	Visual total score	40	11.70	7.11
	Touch total score	40	28.30	11.46
	Oral total score	40	16.38	12.02
	Movement total score	40	21.10	9.92
	Body position total score	40	15.85	8.17
	Valid N (listwise)	40
Children Sleep Habits Questionnaire	CSHQ total score	40	68.43	9.45
	Bedtime Resistance	40	8.65	2.98
	Sleep Onset Delay	40	1.70	0.75
	Sleep Duration	40	5.02	1.94
	Sleep Anxiety	40	5.97	1.73
	Night Wakings	40	5.47	1.90
	Parasomnias	40	10.77	2.13
	Sleep Disordered Breathing	40	3.92	1.40
	Daytime Sleepiness	39	10.64	2.68
	Valid N (listwise)	39

CHSQ Children Sleep Habits Questionnaire

Table 2.

Percentages of people who score above the ‘majority of others’ on Sensory Profile Subscales

	% Much Less Than Others	% Less Than Others	% Just Like the Majority of Others	% More Than Others	% Much More Than Others
Seeking/Seeker	2.5	10.0	37.5	25.0	25.0
Avoiding/Avoider	5.0	12.5	57.5	15.0	10.0
Sensitivity/Sensor	2.5	2.5	42.5	30.0	22.5
Registration/Bystander	0.0	2.5	27.5	30.0	40.0
Auditory	2.5	10.0	67.5	15.0	5.0
Visual	20.0	15.0	40.0	15.0	10.0
Touch	0.0	5.0	20.0	22.5	52.5
Movement	2.5	5.0	27.5	25.0	40.0
Body Position	2.5	5.0	47.5	12.5	32.5
Oral	*	27.5	50.0	12.5	10.0

* indicates missing or non-applicable data

Based on the Children’s Sleep Habits Questionnaire (CSHQ), a total score above 41 is considered indicative of sleep difficulties. In the current sample, all participants exceeded this threshold (M = 68.43, SD = 9.45), suggesting that every child in the study exceeded the cutoff score, indicating a potential paediatric sleep disorder. These values provide insight into the distribution of PMS symptoms and sleep quality within the sample.

Moreover, the Shapiro-Wilk test confirmed marginally non-normal distribution for CSHQ total scores (W(40) = 0.945, p = .051). For seeking scores, the data was normally distributed (W(40) = 0.983, p = .806). This was the same for sensitivity total scores which also confirmed normality, (W(40) = 0.988, p = .938). Registration scores did not significantly deviate from normality (W(40) = 0.972, p = .413). Given the borderline non-normal distribution of CSHQ scores and the relatively small sample size, Spearman’s rank correlation was used as a conservative non-parametric approach that does not assume normality and is less sensitive to outliers.

Sensory patterns and sleep in the CSHQ: Correlation and Regression Analyses

A Spearman correlation analysis was conducted to explore the relationships between sensory reactivity differences (based on sensory patterns) and sleep quality by looking at the 8 subscales scores from the 33-item CSHQ questionnaire (Table 3).

Table 3.

Spearman correlations between total sensory patterns and CSHQ Subscale Scores (n = 40)

			Bedtime Resistance	Sleep Onset Delay	Sleep Duration	Sleep Anxiety	Night Wakings	Parasomnias	Sleep Disordered Breathing	Daytime Sleepiness
Spearman’s rho	Seeking total score	Correlation Coefficient	.318*	.235	.236	.419**	.321*	.557**	.068	− .161
		Sig. (2-tailed)	.046	.145	.142	.007	.044	< .001	.676	.328
		N	40	40	40	40	40	40	40	39
	Sensitivity total score	Correlation Coefficient	.295	.175	.302	.484**	.385*	.159	.141	.236
		Sig. (2-tailed)	.065	.279	.058	.002	.014	.327	.387	.148
		N	40	40	40	40	40	40	40	39
	Registration total score	Correlation Coefficient	.259	.195	.324*	.402*	.399*	.211	− .129	− .041
		Sig. (2-tailed)	.106	.227	.041	.010	.011	.191	.428	.805
		N	40	40	40	40	40	40	40	39
	Avoiding total score	Correlation Coefficient	.077	− .068	.252	.249	.306	− .119	.154	.043
		Sig. (2-tailed)	.637	.678	.117	.121	.055	.465	.344	.797
		N	40	40	40	40	40	40	40	39

* Correlation is significant at the 0.05 level (2-tailed)

**Correlation is significant at the 0.01 level (2-tailed)

For seeking total scores, significant positive correlations were found with bedtime resistance [r(40) = 0.318, p = .046], sleep anxiety [r(40) = 0.419, p = .007], night wakings [r(40) = 0.321, p = .044], and parasomnias [r(40) = 0.557, p = .001]. For sensitivity total scores, significant positive correlations were observed with sleep anxiety [r(40) = 0.484, p = .002] and night wakings [r(40) = 0.385, p = .014]. Similarly, for registration total scores, significant positive correlations were found with sleep duration [r(40) = 0.324, p = .041], sleep anxiety [r(40) = 0.402, p = .010], and night wakings [r(40) = 0.399, p = .011].

Lastly, no significant correlations were found between avoiding total scores and any of the 8 subscales. Therefore overall, we can see that there is a significant correlation in seeking, sensitivity and registration on CHSQ subscale scores.

A stepwise regression analysis was conducted to examine whether sensory seeking, sensitivity, and registration were positively associated the subscale scores of the CSHQ sleep questionnaire (Table 4). Regression results are presented with 95% confidence intervals to reflect the precision of effect size estimates. Model fit indices (R² and F statistics) are reported descriptively. Seeking was positively associated bedtime resistance, explaining 18.9% of the variance (R² = 0.189, F(1,38) = 8.832, p = .005), with a standardized beta of β = 0.434 (p = .005). Registration was positively associated sleep duration, accounting for 12.5% of the variance (R² = 0.125, F(1,38) = 5.441, p = .025), with β = 0.354 (p = .025). Sensitivity was positively associated sleep anxiety, explaining 35% of the variance (R² = 0.350, F(1,38) = 20.453, p < .001), with β = 0.592 (p < .001), while seeking and registration were excluded from this model due to non-significant contributions (see Table 5). Sensitivity also was positively associated night wakings, explaining 11.5% of the variance (R² = 0.115, F(1,38) = 4.935, p = .032), with β = 0.339 (p = .032), with seeking and registration excluded.

Table 4.

Summary of stepwise regression analyses predicting CSHQ subscale scores (n = 40)

Outcome Variable	Predictor	b	SE b	β	95% CI	R²	Adjusted R²	F (df)
Bedtime Resist.	Seeking	0.061	0.020	0.434	[0.019, 0.102]	0.189	0.168	8.832 (1,38)
Sleep Duration	Registration	0.036	0.015	0.354	[0.005, 0.067]	0.125	0.102	5.441 (1,38)
Sleep Anxiety	Sensitivity	0.059	0.013	0.592	[0.033, 0.085]	0.350	0.333	20.453 (1,38)
Night Wakings	Sensitivity	0.037	0.017	0.339	[0.003, 0.071]	0.115	0.092	4.935 (1,38)
Parasomnias	Seeking	0.050	0.014	0.502	[0.022, 0.078]	0.252	0.232	12.828 (1,38)

Adjusted R² values are reported to provide a more accurate estimate of explained variance accounting for model complexity and sample size

Table 5.

Summary of excluded variables in stepwise regression analyses for sleep anxiety and night wakings (n = 40)

Outcome Variable	Variable	β	t	p	Partial Correlation	Tolerance
Sleep Anxiety	Seeking	0.165	1.009	0.320	0.164	0.642
Sleep Anxiety	Registration	0.127	0.615	0.542	0.101	0.406
Night Wakings	Registration	0.101	0.419	0.678	0.069	0.406
Night Wakings	Seeking	0.197	1.033	0.308	0.167	0.642

Finally, seeking was positively associated parasomnia scores, explaining 25.2% of the variance (R² = 0.252, F(1,38) = 12.828, p < .001), with β = 0.502 (p < .001).

Multicollinearity diagnostics for sensory pattern predictors indicated that Variance Inflation Factors (VIFs) ranged from 2.11 to 4.52 (seeking = 2.11; avoiding = 3.07; sensitivity = 4.52; registration = 3.09). These values suggest moderate but acceptable levels of multicollinearity, with no values exceeding commonly accepted thresholds.

Sensory modalities and sleep in the CSHQ: Correlation and Regression Analyses

A Spearman correlation analysis was conducted to examine the relationships between sensory reactivity differences (based on sensory modalities) and sleep quality, based on the eight subscale scores of the 33-item Children’s Sleep Habits Questionnaire (CSHQ) (see Table 6).

Table 6.

Spearman correlations between total sensory modalities and CSHQ Subscale Scores (n = 40)

			Bedtime Resistance	Sleep Onset Delay	Sleep Duration	Sleep Anxiety	Night Wakings	Parasomnias	Sleep Disordered Breathing	Daytime Sleepiness
Spearman’s rho	Auditory total score	Correlation Coefficient	.152	.075	.213	.309	.306	.008	− .044	.171
		Sig. (2-tailed)	.350	.644	.187	.053	.055	.963	.786	.299
		N	40	40	40	40	40	40	40	39
	Visual total score	Correlation Coefficient	.117	− .067	.071	.393*	.174	.238	− .273	.102
		Sig. (2-tailed)	.471	.680	.664	.012	.283	.140	.088	.539
		N	40	40	40	40	40	40	40	39
	Touch total score	Correlation Coefficient	.175	.067	.229	.437**	.403**	.289	.149	.011
		Sig. (2-tailed)	.280	.683	.156	.005	.010	.071	.358	.948
		N	40	40	40	40	40	40	40	39
	Movement total score	Correlation Coefficient	.301	.152	.215	.459**	.197	.436**	− .066	− .140
		Sig. (2-tailed)	.059	.349	.183	.003	.223	.005	.684	.397
		N	40	40	40	40	40	40	40	39
	Body position total score	Correlation Coefficient	.178	.072	.110	.348*	.194	.277	− .150	− .169
		Sig. (2-tailed)	.272	.659	.498	.028	.231	.083	.357	.303
		N	40	40	40	40	40	40	40	39
	Oral total score	Correlation Coefficient	.372*	.198	.394*	.396*	.499**	.235	.164	.053
		Sig. (2-tailed)	.018	.222	.012	.012	.001	.144	.312	.747
		N	40	40	40	40	40	40	40	39

* Correlation is significant at the 0.05 level (2-tailed)

**Correlation is significant at the 0.01 level (2-tailed)

No significant correlations were found for auditory total scores. However, for visual scores, a significant positive correlation was identified with sleep anxiety [r(40) = 0.393, p = .012].

Regarding touch total scores, significant positive correlations were observed with sleep anxiety [r(40) = 0.437, p = .005] and night wakings [r(40) = 0.403, p = .010]. For movement total scores, significant positive correlations were found with sleep anxiety [r(40) = 0.459, p = .003] and parasomnias [r(40) = 0.436, p = .005]. In relation to body position total scores, a significant positive correlation was found with sleep anxiety [r(40) = 0.348, p = .028]. Lastly, for oral total scores, significant positive correlations were identified with bedtime resistance [r(40) = 0.372, p = .018], sleep duration [r(40) = 0.394, p = .012], sleep anxiety [r(40) = 0.396, p = .012], and night wakings [r(40) = 0.499, p = .001].

Overall, these findings suggest that sensory processing domains—particularly visual, touch, movement, body position, and oral reactivity—are significantly associated with various sleep disturbances in children, as measured by the CSHQ.

To provide a comprehensive overview of the relationships across all sensory and sleep variables, a heatmap representing the full Spearman correlation matrix is presented in Fig. 1. The heatmap illustrates the direction and strength of the correlations across all sensory processing and sleep variables, with dark purple representing negative correlations, green indicating correlations close to zero, and yellow representing stronger positive correlations. This visualisation facilitates interpretation of the overall pattern of associations across both sensory patterns and modalities, including within- and between-form relationships.

Fig. 1.

Heatmap of Spearman correlations between sensory reactivity and sleep subscale scores

A stepwise regression analysis was conducted to examine whether oral, visual, touch, movement, and body position total scores were positively associated the eight subscale scores of the CSHQ sleep questionnaire (Table 7). Regression results are presented with 95% confidence intervals to reflect the precision of effect size estimates. Model fit indices (R² and F statistics) are reported descriptively. Oral total scores were positively associated bedtime resistance, explaining 31.3% of the variance (R² = 0.313, F(1,38) = 17.328, p < .001), with a standardized beta of β = 0.560 (p < .001). Oral total scores also were positively associated sleep duration, accounting for 15.6% of the variance (R² = 0.156, F(1,38) = 7.020, p = .012), with β = 0.395 (p = .012). Movement and touch total scores were positively associated sleep anxiety, explaining 34.3% of the variance (R² = 0.343, F(2,37) = 11.6, p = .002). Movement (β = 0.484, p = .002) and touch (β = 0.420, p = .018) were retained in the model, while oral, visual, and body position scores were excluded due to non-significant contributions (see Table 8). Oral total scores were positively associated night wakings, explaining 25.0% of the variance (R² = 0.250, F(2,37) = 12.691, p < .001), with β = 0.500 (p < .001), while touch was excluded from the model.

Table 7.

Summary of stepwise regression analyses predicting CSHQ subscale scores (n = 40)

Outcome Variable	Predictor	b	SE b	β	95% CI	R²	Adjusted R²	F (df)
Bedtime Resist.	Oral	0.139	0.033	0.560	[0.071, 0.206]	0.313	0.295	17.328 (1,38)
Sleep Duration	Oral	0.064	0.024	0.395	[0.015, 0.112]	0.156	0.134	7.020 (1,38)
Sleep Anxiety	Movement	0.084	0.025	0.484	[-0.021, 0.099]	0.343	0.307	11.6 (2,37)
	Touch	0.063	0.026	0.420	[0.012, 0.115]
Night Wakings	Oral	0.079	0.022	0.500	[0.034, 0.124]	0.250	0.210	12.691 (2,37)
Parasomnias	Movement	0.077	0.033	0.359	[0.011, 0.143]	0.129	0.106	5.621 (1,38)

Adjusted R² values are reported to provide a more accurate estimate of explained variance accounting for model complexity and sample size

Table 8.

Summary of excluded variables in stepwise regression analyses for sleep anxiety and night wakings (n = 40)

Outcome Variable	Variable	β	t	p	Partial Correlation | Tolerance
Sleep Anxiety	Oral total score	0.419	2.643	0.012	0.399 | 0.695
Sleep Anxiety	Visual total score	0.212	1.312	0.198	0.211 | 0.758
Sleep Anxiety	Body position	0.289	1.575	0.124	0.251 | 0.577
Night Wakings	Touch	0.133	0.687	0.496	0.112 | 0.536

Lastly, movement was positively associated parasomnia scores, explaining 12.9% of the variance (R² = 0.129, F(1,38) = 5.621, p = .023), with β = 0.359 (p = .023).

Multicollinearity diagnostics for sensory modality predictors indicated that VIFs ranged from 1.68 to 2.54 (auditory = 2.10; visual = 1.68; touch = 2.27; movement = 2.54; body position = 1.95; oral = 2.46). These values indicate low to moderate multicollinearity across predictors.

Discussion

Both sleep difficulties and differences in sensory reactivity are common in Phelan-McDermid syndrome (PMS). Currently, there is no conclusive evidence on the possible underlying causes of sleep problems in this population; however, the literature supports the role of haploinsufficiency of the SHANK3 gene as a factor involved in sleep disturbances in people with PMS with pathogenic variants and deletions [38, 39]. To our knowledge, this is the first study to examine associations between sensory reactivity subscale scores and sleep-related outcomes in children with PMS. Given the correlational design and the use of questionnaire subscales, these findings should be interpreted as associations between measured dimensions rather than evidence of independent or discrete constructs. Building on this, SHANK3 haploinsufficiency may provide a common neurobiological basis linking sensory processing differences and sleep disturbances in PMS. SHANK3 plays a key role in synaptic function, and its disruption has been associated with altered neural connectivity and excitatory–inhibitory balance [40]. These changes may affect brain systems involved in both sensory processing and arousal regulation, potentially contributing to atypical sensory reactivity as well as difficulties in sleep initiation and maintenance. From this perspective, sensory and sleep disturbances may reflect shared underlying mechanisms rather than independent features [5].

A characteristic pattern of sensory reactivity has now been identified in PMS, characterised by increased hyporeactivity and heightened sensory seeking [9], as well as altered sleep patterns including increased sleep resistance, nocturnal awakenings and daytime sleepiness [10, 11, 41]. Although there were no previous studies specifically exploring the relationship between these sensory patterns and sleep in people with PMS, research in children with idiopathic autism has reported significant associations between sensory disturbances and sleep problems [19, 20]. These findings are consistent with previous research reporting a high prevalence of sleep disturbances in PMS. For example, Della Bro et al. [4] reported that up to 90% of individuals exceed clinical thresholds on the CSHQ. In our sample, all participants scored above the clinical cutoff, further supporting the notion that sleep disturbances are a highly prevalent feature of PMS. However, differences in sampling methods and study design should be considered when comparing prevalence estimates.

The results of our study identified significant associations between sensitivity (sensory hyperreactivity), registration (hyporreactivity) and sensory seeking profiles with sleep subscales on the CSHQ questionnaire. Specifically, correlational analysis showed that sensory seeking scores were positively associated with sleep resistance, sleep-related anxiety, nocturnal awakenings and parasomnias. Likewise, sensitivity was related to sleep anxiety and nocturnal awakenings, while sensory registration was associated with sleep duration, anxiety and awakenings. In contrast, the pattern of sensory avoidance did not show significant associations with any sleep subscale.

Findings from regression analyses suggest that sensory subscale scores are associated with sleep disturbances in children with PMS. Heightened sensory sensitivity may increase levels of hyperalertness [42, 43], hindering the transition to rest, sleep reconciliation and sleep maintenance, as previously observed in children with idiopathic autism [44].

Similarly, previous research in children with sensory processing difficulties has reported that sensory seeking profiles are significantly associated with disrupted sleep patterns [42]. These results are consistent with the findings of Tyagi et al. [43], who identified a critical association between sleep disruption and increased sensory seeking. Manelis-Baram et al. [19] also identified significant associations between sensory sensitivity patterns and alterations in sleep duration and quality. In line with this literature, our findings also support an association between sensory seeking and sleep disturbances. However, in our sample, sensory seeking emerged as one of the strongest and most consistent predictors across multiple sleep domains. This suggests a potentially more prominent role of sensory seeking within the sensory profile of children with PMS, which may reflect the distinct sensory phenotype of this population compared to children with idiopathic autism or typical development [9].

On the other hand, the analysis of different modalities of sensory reactivity with different sleep difficulties in children with PMS. Our results support the idea that differences in tactile processing play an important role in sleep regulation in this population, in line with current scientific evidence that has identified difficulties in tactile reactivity as a relevant factor associated with sleep disturbances [21, 42].

Specifically, the significant correlations between tactile processing and sleep-related anxiety, as well as nighttime awakenings, reinforce the results of the regression analysis, where tactile processing, together with motion processing, explained 34.3% of the variance in sleep anxiety levels, with tactile processing being a significant predictor. The particularly strong associations observed with sleep anxiety suggest that anxiety-related mechanisms may play an important role in modulating the relationship between sensory reactivity and sleep disturbances. Heightened sensory reactivity may contribute to increased physiological and cognitive arousal, which can interfere with the ability to initiate and maintain sleep. In this context, sleep difficulties may be partially mediated by elevated anxiety levels, a mechanism that has been previously proposed in neurodevelopmental populations [20]. This is consistent with the study by Tzischinsky et al. [21], who found that tactile hyperreactivity explained up to 24% of the variance in sleep problems, particularly sleep difficulties and nighttime awakenings, highlighting the importance of considering tactile processing in the assessment and intervention of sleep disorders in PMS.

Significant correlations were also found between oral processing scores and bedtime resistance, sleep duration, sleep anxiety and nighttime awakenings. Furthermore, stepwise regression analysis indicated that oral processing was the strongest predictor of bedtime resistance, explaining 31.3% of the variance, and significantly predicting sleep duration and nighttime awakenings. These findings are in line with studies by Tzischinsky et al. [21] and Hartman et al. [42], which indicated that oral sensitivity can significantly contribute to sleep disturbances. Similarly, Jamiol-Milc et al. [22] noted that oral hypersensitivity in children with autism is associated with increased sleep disturbance, probably due to difficulties in sensory self-regulation necessary to achieve a calm state prior to sleep. These findings reaffirm that oral processing challenges represent an essential component of sleep disorders in PMS and should be prioritised in clinical approaches.

Unlike the sensory seeking, sensitivity, and registration profiles, this sample’s sensory avoiding pattern did not show important associations to indicators of sleep disturbances. Researchers found this prominent for sensory avoidance links to difficulties with sleep onset as well as frequent night awakenings in other neurodevelopmental disorders. In Autism and Attention-Deficit/Hyperactivity Disorder (ADHD), we see this relationship, possibly because a heightened hypervigilance responds to those environmental stimuli [45, 46]. One possible explanation is that, in the specific case of PMS, sensory avoidance is not a predominant feature of the sensory profile, unlike hyporeactivity and sensory seeking, which are more characteristic [9]. In this sense, the lower prevalence of sensory avoidance behaviors may have limited their statistical power as predictors of sleep in this sample. Alternatively, it is possible that avoidance behaviors in PMS manifest in contexts that do not directly interfere with sleep routines, thereby reducing their specific impact on the domains assessed by the CSHQ.

Finally, significant correlations were identified between movement scores and sleep-related anxiety, parasomnias as well as a marginal correlation with bedtime resistance. These findings, supported by stepwise regression analyses, indicated that movement processing in combination with tactile processing are significant predictors of sleep disturbances in PMS. These results could suggest that movement and touch processing may be related to an elevated physiological arousal pattern during sleep [46] which may facilitate the occurrence of atypical behaviours such as parasomnias, as well as contribute to a state of alertness that interferes with sleep onset and maintenance [47].

These findings have important implications for clinical practice, suggesting that targeting sensory processing differences may contribute to improved sleep quality and, consequently, enhance quality of life for individuals with PMS and their families. Behavioural sleep interventions, such as structured bedtime routines and sleep hygiene strategies, are commonly recommended in neurodevelopmental populations and may be beneficial in this group. In addition, interventions aimed at reducing sensory hyperreactivity or supporting sensory regulation may represent a particularly relevant approach. Strategies targeting sensory profiles identified in this study—especially sensory seeking and oral processing—may help reduce physiological arousal and facilitate sleep initiation and maintenance. Integrating behavioural and sensory-informed approaches may therefore represent a promising avenue for improving sleep outcomes in this population.

Limitations and future directions

While the findings of this study provide novel evidence regarding the relationship between sensory reactivity and sleep difficulties in children with PMS, several considerations should guide their interpretation and inform future research.

First, although the use of validated parent-report questionnaires allowed for the systematic characterisation of both sensory reactivity and sleep behaviours, future studies would benefit from incorporating objective measures of sleep, such as actigraphy or physiological recordings, to provide a more comprehensive assessment.

Second, the cross-sectional design enabled the identification of meaningful associations but does not allow conclusions regarding directionality or developmental trajectories. Longitudinal studies are therefore needed to better understand how sensory processing differences and sleep disturbances evolve over time.

In addition, aspects related to sampling and clinical characterisation should be considered. The use of a convenience sample and the lack of information on response rates limit the ability to determine how representative the sample is of the broader PMS population. Moreover, the overall clinical characterisation was limited, as the dataset did not include systematically collected information on key variables such as comorbid conditions (e.g., epilepsy, anxiety, ADHD), or level of intellectual functioning, and demographic and clinical data were only partially captured. Together with the wide age range and variability in functional ability, these factors may have introduced heterogeneity and potential confounding influences on both sensory and sleep outcomes. As such, the observed associations should be interpreted with caution, and their generalisability may be restricted.

Finally, several analytical considerations should be noted. The absence of correction for multiple comparisons increases the risk of type I error. In addition, the combination of correlation-based variable selection and stepwise regression limits the interpretability of p-values and may introduce bias in parameter estimates, increasing the risk of identifying spurious predictors. Accordingly, the findings should be considered exploratory and hypothesis-generating, with greater emphasis placed on effect sizes and their confidence intervals rather than statistical significance. Future research should employ theory-driven models and more robust statistical approaches in larger samples.

Despite these limitations, this study represents a meaningful contribution to the existing literature and lays the groundwork for future research exploring targeted sensory interventions to improve sleep outcomes in individuals with PMS.

Authors’ contributions

Conceptualisation: T.T., S.S.-T.; Data curation: N.G.; Formal analysis: N.G.; Investigation: T.T., S.S.-T.; Methodology: T.T., S.S.-T.; Project administration: T.T., S.S.-T.; Resources: T.T., S.S.-T.; Software: N.G.; Supervision: T.T., S.S.-T.; Writing—original draft: T.T., S.S.-T.; Writing—review & editing: N.G., T.T., S.S.-T. All authors have read and agreed to the published version of the manuscript.

Funding

This research has received no funding.

Data availability

The data that support the findings of this study are not openly available due to reasons of sensitivity and are available from the corresponding author upon reasonable request. Data are located in controlled access data storage at Reading University and Universidad Rey Juan Carlos.

Declarations

Ethics approval and consent to participate

All participants provided informed consent prior to participating in the study. In the case of participants under the age of 16, informed consent to participate was obtained from their parents or legal guardians. The study was conducted in accordance with the Declaration of Helsinki and approved by the Ethics Committee Research of Reading University (2023-222-TT) and Rey Juan Carlos University (ENM103/202412202101122, 26 January 2022). Patients and, when applicable, their parents were directly informed about the aims of the research project, and written informed consent was obtained from all participants.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.