Beyond Aging, Sex and Insomnia Disorder Shape NREM Brain Oscillations

Objectives: Chronic insomnia (INS) is particularly prevalent in older adults and females. Sex- and age-related differences in neurophysiological markers of sleep quality (sleep spindles and slow-wave activity [SWA]) may underlie differential vulnerability to INS. This study investigated the effects of sex and insomnia on spindle and SWA beyond aging, to better understand the mechanistic differences contributing to the higher prevalence of INS in females.

After a habituation night, one night of sleep assessed with polysomnography was analyzed in 222 adults (aged 18-82) including 119 INS (71% female) and 103 healthy sleepers (HS; 61% female). Spindle density, slow oscillation (SO) density, relative sigma power and SWA were derived during NREM sleep. Age, group, sex, and group-by-sex interactions were examined, with age as a covariate.

Age, insomnia, and sex each contributed uniquely to NREM oscillatory activity. INS primarily reduced spindle and SO density, while sex accounted for differences in SWA. While SWA was higher in females overall, sex differences were not significant within the INS or HS groups. Female INS reported highest rates of insomnia severity as well as lower sigma power than males in the INS group. Spindle and SO density deficits were also present in female INS relative to female HS, as well as male INS relative to male HS.

The combination of reduced sigma power in females with insomnia relative to their male counterparts, as well as less spindle and SO density compared to female healthy sleepers may contribute to greater insomnia severity in females.

Sex, gender, insomnia disorder, sleep, spindle, sigma, slow wave activity, delta, slow oscillation, EEG .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 19, 2026. ; https://doi.org/10.64898/2026.03.17.712450doi: bioRxiv preprint Statement of Significance Insomnia is a growing public health concern that is more commonly reported in females, yet the neural mechanisms underlying this sex difference remain poorly understood. Our findings suggest that specific markers of sleep quality are disproportionately disrupted in females with insomnia, potentially contributing to greater vulnerability and symptom severity. These results provide new insight into how sex influences the neurophysiology of insomnia disorder and identify oscillatory markers that could serve as targets for personalized interventions. Future research should investigate whether these alterations represent persistent dysfunction or reversible changes, which could advance understanding of the biological basis of insomnia and inform strategies to improve sleep health in at-risk populations. .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 19, 2026. ; https://doi.org/10.64898/2026.03.17.712450doi: bioRxiv preprint

Chronic insomnia is a growing public health concern, affecting approximately 16% of Canadians, and individuals worldwide 1,2. Chronic insomnia is defined by significant daytime impairments and reduced quality of life3–5 as a result of persistent difficulty initiating or maintaining sleep, at least three nights a week, for at least three months despite adequate opportunity for rest. Although chronic insomnia diagnosis is based on self-reported complaints, altered objective sleep measures can also be commonly observed within this population. Polysomnography (PSG) sleep studies show that, relative to healthy sleepers, some individuals with chronic insomnia exhibit prolonged sleep onset latency, increased time in lighter stages of non-rapid eye movement (NREM) sleep (i.e., NREM1 and NREM2), frequent nocturnal awakenings, and reduced sleep efficiency6–8. Beyond conventional macro-architectural measures, NREM sleep is defined by specific neural oscillations that reflect the integrity and restorative properties of sleep at a neurophysiological level and may therefore provide a more sensitive marker of insomnia- related sleep disruption 9–11. Sleep spindles are brief (0.5 to 3s) oscillating bursts of 11–16Hz sigma activity that can be observed frontally, and over central–parietal regions 12,13. Slow wave activity (SWA) is composed of delta waves oscillating between 1-4Hz as well as high-voltage biphasic waves called slow oscillations (SO; <1.25Hz) 14–16. These rhythms, while mainly known for their implications in sleep-dependent memory processes 17,18, also play a central role in maintaining sleep continuity and depth 19–21. Reduced spindle activity is a reliable marker of disrupted sleep 12,22,23, and some studies have shown that insomnia is also associated with reduced delta and SO power, increased fast-frequency EEG activity, and blunted slow-wave rebound following sleep deprivation, patterns consistent with hyperarousal and altered sleep homeostasis9,16,24,25. Importantly, sleep quality and its underlying neurophysiological oscillations are shaped by key biological moderators, most notably sex and age. In healthy populations females generally exhibiting greater sleep spindle density, and higher absolute sigma and delta power compared to males 26. However, findings across studies have been mixed and appear to vary depending on methodology, age, and analytic approach26,27. Contrasting to the female sleep advantage in some NREM brain oscillations, epidemiological studies consistently demonstrate that chronic insomnia is more prevalent in females than males, with females also showing a tendency to perceive and experience insomnia as more distressing and impairing 1,28. Importantly, while sex differences in sleep oscillations are well-documented among healthy sleepers 26,29,30, it remains unclear how these differences manifest in individuals with disturbed sleep, such as those with chronic insomnia. In addition to sex, aging also influences sleep physiology, further shaping the quality and stability of sleep across the lifespan. Both SWA and sleep spindle characteristics decline with age due to structural and functional brain changes, including neuronal loss, neurobiological deterioration, reduced thalamocortical connectivity, diminished GABAergic signaling, and cortical thinning 31–33. Natural age-related changes in sex hormones further modulate these processes. For instance, estradiol and progesterone enhance spindle activity via GABAergic mechanisms34, while SWA appears less sensitive to hormonal fluctuations, likely due to its regulation by sleep homeostasis 35–37. Testosterone may also influence cortical structures .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 19, 2026. ; https://doi.org/10.64898/2026.03.17.712450doi: bioRxiv preprint involved in spindle and SWA, but its effects remain understudied 36,38,39. These neurobiological and hormonal age-related changes contribute to shallower, more fragmented sleep and higher insomnia risk in older adults 33,40. Moreover, sex disparities in insomnia prevalence become apparent during puberty and persist across the lifespan, with higher risks during reproductive events (i.e., menarche, pregnancy, post-partum, perimenopause) 41–44. The highest incidence of insomnia occurs in midlife and older females1,4. During perimenopause and post-menopause (i.e., total cessation of menstruation), rates of insomnia increase likely due to hormonal changes such as estrogen decline, although comorbid health issues and psychosocial stressors may also contribute 45,46. In healthy sleepers, age is associated with decreases in delta power, sigma power, and centro-parietal spindle density in both sexes 31,47,48, but even with these declines, females tend to exhibit higher spindle density than males throughout the lifespan31. To date, most research on sex differences in sleep neurophysiology has focused on healthy individuals across the lifespan, leaving gaps in our understanding of how the presence of insomnia may impact sleep micro-architecture. Given the critical role of sleep spindles, delta waves and SOs in sleep quality and maintenance, sex differences in brain oscillations may contribute to differential vulnerability for insomnia and shape its clinical presentation. The current study aims to investigate the effects of age, biological sex and presence of chronic insomnia on spindle and SWA by identifying unique contributions of each factor and their interactions in a large dataset of individuals with and without insomnia, accounting for the influence of age. We hypothesize that above and beyond age, individuals with chronic insomnia will exhibit reduced spindle and SO density, as well as reduced sigma power and SWA relative to healthy sleepers. However, when stratified by sex, we expect females with chronic insomnia to show lower spindle and SWA than males with chronic insomnia given the higher severity of complaints and neurobiological/hormonal changes in females across the lifespan.

Participants Data used for the current study were collected in the scope of six different projects (published 11,49–52 or registered [ISRCTN13983243, NCT04024787, ISRCTN12645581] elsewhere) investigating sleep in individuals with chronic insomnia and healthy sleepers within the Sleep, Cognition, and Neuroimaging Laboratory in Montreal, Canada. All groups had a similar recruitment process, and participants were recruited through online and print advertisements, community postings, and physician referrals. Initial eligibility was assessed via a telephone screening, followed by a semi-structured in-person clinical interview, which most studies included the Structured Clinical Interview for DSM (SCID) to assess comorbid mental disorders. Eligibility criteria varied slightly across studies (e.g., age). Healthy sleepers were self- identified good sleepers, reported no sleep complaints, and had no history of sleep disorders. Participants with insomnia met diagnostic criteria for chronic insomnia disorder according to the Diagnostic and Statistical Manual of Mental Disorders (DSM-5 4) and the International Classification of Sleep Disorders, 3rd edition (ICSD-3 3). Criteria included self-reported difficulty initiating or maintaining sleep, or early morning awakenings occurring ≥ 3 nights per .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 19, 2026. ; https://doi.org/10.64898/2026.03.17.712450doi: bioRxiv preprint week for ≥ 3 months, with associated daytime impairment. General exclusion criteria for both groups included being outside the eligible age range (< 18 years), current neurological or psychiatric disorders other than anxiety or depression, medical conditions likely to affect sleep (e.g., epilepsy, multiple sclerosis, Parkinson’s disease, chronic pain, active cancer), untreated thyroid disorders, or major cardiovascular events (e.g., myocardial infarction, stroke). Other exclusion criteria included the presence of sleep disorders identified through moderate to severe sleep apnea (apnea-hypopnea index; AHI >5-15/h), and periodic limb movement (index >15/h). Participants were excluded for current use of medications known to affect sleep (e.g., hypnotics, antidepressants, and over-the-counter medication), inability to abstain from such medications for at least 1 week prior to the study, frequent alcohol (>10 drinks/week), cannabis or illicit drug use. Some studies included additional restrictions, such as differences in abstinence periods from prior medications or tobacco use. All participants provided written informed consent prior to participation. Procedures were approved by the Concordia University Human Research Ethics Committee and by the Comité d’Ethique de la Recherche of the Centre de Recherche de l’Institut de Gériatrie de Montréal (CRIUGM). Insomnia Severity Insomnia symptoms were assessed using the Insomnia Severity Index (ISI)53, a seven-item self- report instrument evaluating the nature, severity, and impact of insomnia over the preceding two weeks. Items assess difficulties with sleep onset, sleep maintenance, and early morning awakenings, as well as satisfaction with current sleep patterns, interference with daytime functioning, perceived impairment, and distress associated with sleep problems. Each item is rated on a five-point Likert scale (0–4), yielding a total score ranging from 0 to 28. Established interpretive ranges classify scores of 0–7 as no clinically significant insomnia, 8–14 as subthreshold insomnia, 15–21 as moderate clinical insomnia, and 22–28 as severe clinical insomnia 53. The ISI demonstrates strong content, concurrent, and predictive validity. In this study, it was controlled forthe analyses between insomnia groups. Procedure All participants underwent a clinical interview and screening for confounding sleep disorders (e.g., obstructive sleep apnea, restless legs syndrome, periodic limb movements, REM sleep behavior disorder, narcolepsy). Eligible participants then completed an overnight screening in lab PSG to rule out any primary sleep disorders (other than insomnia). That first PSG night also served as a habituation night. Participants returned for at least one experimental sleep assessment involving comprehensive PSG. When multiple nights were available (e.g., experimental night and/or intervention vs control night), the control night was selected for analysis. Participants slept in private, sound- attenuated bedrooms, with continuous video and physiological monitoring by a trained research assistant stationed in an adjacent control room. Measures Polysomnographic (PSG) Recording .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 19, 2026. ; https://doi.org/10.64898/2026.03.17.712450doi: bioRxiv preprint PSG recordings included electroencephalography (EEG), electrooculography (EOG), electromyography (EMG), and electrocardiography (ECG). Respiratory effort (thoracic and abdominal belts), airflow (nasal-oral thermocouple), and oxygen saturation (finger pulse oximetry) were also recorded during the first screening night. EEG electrodes were positioned according to the international 10–20 system, including at minimum Fz, F3, F4, Cz, C3, C4, Pz, P3, P4, O1, O2, and mastoids (M1–M2). Signals were recorded using SOMNOmedics amplifiers (Somnomedics GmbH, Germany), sampled at 512Hz, referenced online to Pz, and re-referenced offline to the contralateral mastoids (joint M1-M2). Polysomnographic Analyses Sleep staging (NREM1, NREM2, NREM3, REM, wake) and arousals were scored manually by two independent raters blind to participant group, following standard American Academy of Sleep Medicine 54,55 criteria, using the Wonambi Python toolbox ( https://wonambi- python.github.io)56. Sleep macroarchitecture variables were derived from standard scoring procedures and are reported to provide descriptive context for microarchitectural analyses. Extracted sleep parameters included total sleep time (TST), time in bed (TIB), sleep onset latency (SOL), wake after sleep onset (WASO; i.e., how long participants are awake after lights out), sleep stage proportions including % of wake (i.e., how much of the recording is scored as wake), sleep efficiency (SE; i.e., TST/TIB × 100), and sleep fragmentation index (SFI; i.e., # shifts to wake and lighter stages (e.g., NREM1 or NREM2) from NREM3 or REM)/TST [hrs] 57). Artefacts and poor-quality epochs and electrodes were detected manually. Oscillatory activities (event-based and spectral power) were automatically analyzed from Fz, Pz, and Cz channels using the seapipe pipeline ( https://github.com/nathanecross/seapipe), an open- source Python-based package58. Spectral power EEG spectrum power average (30s of time resolution with artefact excluded) was calculated with a 0.2Hz resolution, by applying a Fast Fourier Transformation (FFT; 50% overlapping, 5s windows, Hanning filter). Mean power was calculated for each 0.25Hz bins between 0.25Hz and 30Hz and for the following frequency bands: SO (0.25-1.25Hz), SWA (i.e., combined delta and SO: 0.25-4Hz), sigma (11.25-16Hz). Given that age differences have been shown to only emerge when isolating SO (<1.25Hz), and as combining SO with delta (1–4Hz) can mask SO-specific alterations 9, both SO and delta bands were extracted. Both absolute (µV²) and relative power (band/total power; %) were calculated for each band. Event detection Prior to spindle detection, we performed a data-driven selection of participant-specific sigma- band boundaries using the specparam algorithm to distinguish spectral peaks from aperiodic

activity59 within 9-16Hz across combined NREM2 and NREM3 sleep epochs. Such participant-driven sigma peak detection is robust against the inter-individual variability of spindle peak frequencies60,61. We observed a typical anteroposterior spindle frequency gradient62 with slower sigma peaks on frontal electrodes (Fz) and faster sigma peaks on central and .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 19, 2026. ; https://doi.org/10.64898/2026.03.17.712450doi: bioRxiv preprint posterior electrodes (Cz, Pz). For each participant, we used the highest peak, in the 9-13Hz range for Fz, 9-16Hz for Cz and 13-16Hz for Pz, to centre the sigma frequency band (with a 4Hz bandwidth) in spindle detection. Sleep spindles were detected using a consensus-based pipeline designed to reduce algorithm- specific bias by integrating multiple established spindle detection methods within a unified framework 63.First, we applied 4 validated automatic methods to detect spindles on Fz, Cz, Pz, including algorithms described by Ferrarelli et al, Mölle et al, Ray et al, and Lacourse and colleagues63–66. They were used with their described default settings but using the participant- specific sigma frequency band. EEG segments overlapping annotated artefacts or arousals were excluded prior to detection. To obtain a unified set of spindle events, detections from all algorithms were combined using an additive consensus approach. Overlapping detections across algorithms were merged and considered a single spindle event, whereas non-overlapping detections identified by only one algorithm were retained as individual spindles. Consensus spindles failing minimum duration criteria (0.5-3 s) were discarded, and duplicate events were removed. Final consensus spindles were defined as unified spindle events for downstream analyses. The use of an additive consensus threshold was chosen to maximise sensitivity to spindle events while retaining robustness through multi-algorithm detection. SOs were detected automatically on Fz, Cz and Pz following the procedure proposed by Staresina et al. (2015) 67. In brief, it involves i) filtering the participant’s SO-band signal (0.5- 1.25Hz); ii) identifying the events with a positive-to-negative zero crossing and a subsequent negative-to-positive zero crossing separated by 0.8-2s; iii) retaining the top 25% of events with the largest trough-to-peak amplitudes. For each participant and electrodes (Fz, Cz and Pz), we extracted SOs and spindle density (number/30s-epoch) across combined NREM2 and NREM3 sleep. Analyses were conducted at Fz as the primary electrode of interest for SWA measures, and both Fz and Cz for spindle measures. Corresponding measures at either Fz or Cz and Pz were included to evaluate the spatial distribution of effects and are described in the Results and Discussion. Fz and Cz findings that are presented in the Tables. Pz findings that are presented in Supplemental Material . Relative power spectral is presented. Absolute spectral power as well as spindle and SO amplitude can be found in the Supplemental Materials. Tables and figures present estimated marginal means, while raw means for all dependent variables can be found in Supplemental Table 1, 2, and 3. Statistical Analyses Sample size estimation was based on Buysse et al. (2008) 68, who reported a significant group- by-sex interaction in sleep microarchitecture. Accordingly, a minimum of 73 participants (48 with chronic insomnia, 25 healthy sleepers) were targeted. However, G*Power analysis indicated that 180 participants (90 insomnia, 90 healthy sleepers, balanced by sex) would be an adequate sample to detect medium-sized effects (f = .25) with 80% power at α = .05. Statistical analyses were conducted using SPSS 31.0.0.0. Participants were considered outliers if they showed extreme values (±3.29 SD) on three or more dependent variables; this conservative approach ensured cases were not excluded due to chance variation on a single measure. .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 19, 2026. ; https://doi.org/10.64898/2026.03.17.712450doi: bioRxiv preprint Assumptions of normality and homogeneity of variance were assessed using Shapiro–Wilk tests, skewness and kurtosis z-scores (±1.96), and Levene’s tests. Several dependent variables showed deviations from these assumptions; therefore, we used Generalized Linear Models (GLMs) with Huber–White/Sandwich robust estimators of variance and bootstrapped (1,000 samples) bias- corrected 95% confidence intervals to provide robust inference under unequal variances. GLM models included two categorical factors (Sex: female/male; Group: insomnia/healthy sleepers). Group × Sex interactions were examined, along with main effects of Sex and Group controlling for age. Age was entered as a continuous covariate to adjust for potential differences in age distribution between groups. This approach enabled us to test age effects as well as age-adjusted main effects of Sex, and Group on brain oscillations. All analyses used an α level of 0.05. Bootstrapped confidence intervals and Benjamini–Hochberg FDR correction were applied across analyses to account for multiple comparisons69. Analyses used listwise deletion for missing data. Together, these methods provide robust parametric inference while mitigating the influence of non-normality and heteroscedasticity.

Demographics The final sample (N = 222, age range 18-82) included 119 individuals with chronic insomnia (INS; 54%) and 103 healthy sleepers (HS; 46%). The sample comprised 67% females (n = 148), with 33% males ( n = 74). When stratified by both sleep group and biological sex, the sample comprised 34 male INS (MINS; 15%), 85 female INS (FINS; 38%), 40 male HS (MHS; 18%), and 63 female HS (FHS; 29%). This distribution of higher representation of females in both the INS and HS groups is consistent with prior findings of sex distributions in insomnia research 26,41. There was a statistically significant age difference across all four groups (Wald χ ²(3) = 38.79, p < .001; Supplemental Table 4). Bootstrapped pairwise comparisons indicated that FINS were older than FHS ( β = 13.99, p < .001), MINS ( β = -8.47, p = .005), and MHS ( β = 15.77, p < .001). MHS were also significantly younger than MINS ( β = 7.29, p= 0.04). Female and male HS did not differ in terms of age. Hence, all analyses were controlled for age. Within the INS group, insomnia severity (based on the ISI53) ranged from 8 (mild) to 27 (severe) with an average moderate severity score (16.56 ± 3.96). There was a significant difference between FINS and MINS (Wald χ ²(1) = 5.65, p= .017), with females reporting more severe insomnia (17.09 ± 0.42) than males (15.24 ± 0.66, β = -1.86, p < .05). Insomnia severity did not differ by age (Wald χ ²(1) = 0.01, p= .942). Sleep Macroarchitecture Descriptive sleep macroarchitecture for the sample is presented in Supplemental Tables 1-4 . These metrics provide context for subsequent analyses of NREM oscillatory activity. No hypotheses were tested for these variables. .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 19, 2026. ; https://doi.org/10.64898/2026.03.17.712450doi: bioRxiv preprint Age Effect in Sleep Macroachitecture Across All Participants Increased age was associated with a significant reduction of TST, SE, and time spent in NREM2, NREM3, and REM. Increases in time awake, WASO, and sleep fragmentation were also observed with increased age, but not TIB, SOL, or time spent in NREM1 (all p > 0.05; Supplemental Table 1). Effect of Sleep Group on Sleep Macroachitecture After controlling for age, INS spent significantly more time awake, WASO, and less time in N R E M 1 , N R E M 2 , N R E M 3 , R E M , T S T , a n d S E c o m p a r e d t o H S . H o w e v e r , T I B , S O L , a n d sleep fragmentation were not impacted by the presence of insomnia (all p > 0.05; Supplemental Table 2). Effect of Biological Sex on Sleep Macroachitecture While age was controlled, males had more time spent in NREM1, had more WASO, and sleep fragmentation compared to females. Sex did not significantly impact other sleep architecture variables (i.e., time spent awake, in NREM2, NREM3, REM, TIB, TST, SOL, and SE; all p > 0.05; Supplemental Table 3). Group by Sex Interaction on Sleep Macroachitecture After controlling for age, we found group by sex differences in measures of sleep macro- architecture (Supplemental Table 4 ) . Bo t h H S g r o u ps ha d hig he r T S T , S E , t im e i n N RE M1 , NREM2, NREM3 and REM. They also had lower WASO, and time awake compared with both INS groups. There were no group differences in SOL. Sleep fragmentation differed significantly between all 4 groups and appeared to mainly be driven by sex (Wald χ ²(3) = 27.13, p < .001) with males’ sleep appearing more fragmented than females. FINS showed a reduced NREM1, NREM2, NREM3, and REM percentage relative to FHS. The only significant sex difference within insomnia groups was for percentage of time spent in NREM3, where MINS displayed less NREM3 compared to FINS ( β = 6.74, 95% CI [3.58, 9.92]). NREM Sleep Microarchitecture Age Effect in Measures of Spindle and SWA Across All Participants In Cz, increasing age predicted significant reductions in spindle density ( β = –0.02, p < .001, 95% CI [–0.025, –0.013]) and in Fz for relative SWA (β = –0.001, p < .001, 95% CI [–0.002, –0.001]). SO density and relative sigma power were preserved across age in all channels ( Table 1). The main effects of age were seen across electrodes (Supplemental Table 8). Effect of Sleep Group on Measures of Spindle and SWA After controlling for age, INS showed significantly lower central spindle density ( β = –0.80, p < .001, 95% CI [–1.00, –0.60]), and frontal SO density ( β = –2.72, p <.001, 95% CI [–3.52, – 2.01]) compared to HS (Table 2) . This effect was seen across all electrodes. Groups did not differ in central sigma power or frontal relative SWA across electrodes (Supplemental Table 9). .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 19, 2026. ; https://doi.org/10.64898/2026.03.17.712450doi: bioRxiv preprint However, frontal relative sigma power (β = -0.01, p < .001, 95% CI [-0.01, -0.007]) was lower in INS than HS. Effect of Biological Sex on Measures of Spindle and SWA After controlling for age, frontal relative SWA ( β = –0.02, p = .032, 95% CI [–0.038, -0.003]) was lower in males than females, as well as posterior ( Table 3 and Supplemental Table 10 ). No significant sex differences were found for central spindle density ( β = –0.10, p = .329, 95% CI [–0.29, 0.10]) or frontal SO density ( β = –0.09, p = .836, 95% CI [–0.951, 0.703]) across electrodes (Supplemental Table 10). Central relative sigma power was equal between sexes ( β = 0.003, p = .135, 95% CI [-0.001, 0.007]), but frontal relative sigma power (β = 0.004, p = .007, 95% CI [0.002, 0.007]) was higher in males than in females. Similar results were fo und in Pz (Supplemental Table 10). Group by Sex Interaction on Measures of Spindle and SWA Concerning sigma power, we found a Group-by-sex interaction in central sigma power (Wald χ ²(3) = 9.68, p = .022; Table 4). FINS showed lower central relative sigma power than MINS (β = -0.006, 95% CI [-0.001, -0.012], p = .030). This relationship was maintained even after controlling for ISI score ( β = 0.005, p = .014) and was also found in Fz, but not Pz (Supplemental Table 11, 12, 13 ). MINS displayed higher central relative sigma power compared to MHS group (Table 4 and Figure 1B ). Within the HS group, FHS exhibited lower posterior relative sigma power than MHS ( β = 0.006, 95% CI [0.001, 0.011], p = .025), but not in Fz or Cz (all p > 0.05; Supplemental Table 11, 12, 13). For spindle density, we found a Group-by-sex interaction (Wald χ ²(3) = 77.11, p < .001) that was driven by MHS exhibiting less frontal spindle density than FHS ( β = 0.30, 95% CI [0.011, 0.611], p = .054), but not central or posterior spindle density (Table 4). While there were no sex differences within the INS group, both FINS and MINS exhibited lower central spindle density relative to the HS sex counterparts (FINS: β = 0.88, p < .001; MINS: β = -0.64, p < .001; Table 4 and Figure 1A ). The same relationship was seen for frontal and parietal spindle density Supplemental Table 11, 12, 13). For SWA, after adjusting for age, there were no Group-by-sex interactions in relative SWA in Fz (Wald χ ²(3) = 5.85, p =.119) or across other electrodes ( Table 4 and Supplemental Table 11, 12, 13). There was, however, Group-by-sex differences in SO density within frontal ( Wald χ ²(3) = 51.04, p < .001), central and parietal. There were no differences between either sex with INS on SO density across all electrodes. Relative to the HS sex counterparts, both FINS and MINS exhibited lower frontal SO density (FINS: β = 2.64, p < .001; MINS: β = -2.88, p < .001; Table 4 and Figure 2). See a summary of the results in Table 5. Additional group by sex analyses on spindle and SO amplitude, SO power, and absolute sigma and SWA can be found in Supplemental Material. .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 19, 2026. ; https://doi.org/10.64898/2026.03.17.712450doi: bioRxiv preprint Figure 1. Significant Group by Biological Sex Interaction on Spindle Activity A) Estimated marginal means ± SE from the generalized linear model of spindle density (detected on Cz) per Group and Sex B) Estimated marginal means ± SE from the generalized linear model of relative sigma power (detected on Cz; 11.25-16Hz) per Group and Sex * p < 0.05 ** p < 0.01 *** p .05 but became significant after bias-corrected confidence intervals Figure 2. Significant Group by Biological Sex Interaction on SWA A) Estimated marginal means ± SE from the generalized linear model of SO density (detected on Fz; 0.25-1.25 Hz) per Group and Sex B) Estimated marginal means ± SE from the generalized linear model of relative SWA (detected on Fz; 0.25-4 Hz) per Group and Sex * p < 0.05 ** p < 0.01 *** p < 0.001

Summary This study investigated how age, the presence of chronic insomnia disorder and biological sex, individually and jointly, influence key NREM oscillatory markers of sleep quality. We tested whether sex and insomnia exert effects beyond aging, and whether females with insomnia, who typically report greater symptom severity, show disproportionate alterations in NREM microarchitecture. Age, insomnia, and sex each contributed uniquely to SWA and spindle activity. Insomnia primarily drove reductions in spindle and SO density above and beyond age or sex effects, while sex accounted for differences in relative SWA. While relative SWA was higher in females across the full sample, sex differences were not significant within insomnia. Both insomnia and sex influenced central relative sigma power, with females showing lower sigma than males experiencing insomnia. In addition, both females and males with insomnia exhibited deficits in most measures compared to healthy sleepers. These patterns underscore that, independent of age, both sex and insomnia uniquely shape NREM oscillatory activity, and are dependent on the specific measures used. The Impact of Age on NREM Neural Activity In our sample, age selectively affected certain brain oscillations, while other NREM features were resilient. Specifically, typical age-related reductions were present in Cz for spindle density, .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 19, 2026. ; https://doi.org/10.64898/2026.03.17.712450doi: bioRxiv preprint and in frontal, central and parietal regions for relative SWA, likely due to structural and functional alterations in thalamocortical networks 7,23,31,70,71, as well as cortical thinning and diminished large-scale neural synchrony as a natural process of aging 72–74. The frontal SWA decline is consistent with known prefrontal cortical thinning, whereas the parietal SWA reduction may reflect broader network-level changes in NREM synchrony. Interestingly, the Age effect on spindle density was restricted to Cz, which may relate to our participant-specific sigma-band detection approach. The adapted frequency range at Cz likely captured both slow and fast spindles, thereby reflecting a more general spindle activity measure. In contrast, the adapted detection ranges at frontal and parietal electrodes may have preferentially targeted slow (Fz) and fast (Pz) spindles, respectively. Age-related changes in general spindle measures at central sites may therefore not extend to specific fast or slow spindle effects. However, relative sigma power as well as SO density in Cz appeared fairly consistent across adulthood, and were more selectively influenced by insomnia and/or sex, contrasting reports of strong age-related declines in spindle and general measures of SWA 31,73,75. Such discrepancies may reflect differences in sample composition and methodology such as the oscillatory feature and its detection method (e.g., spectral power or event detection). Many previous studies did not explicitly control for disturbed sleep or sex differences, factors that our results show can influence spindle and SWA measures 73. Additionally, our use of participant-specific sigma-band detection, combined with averaging across NREM2 and NREM3, likely mitigated some age- related declines that may have been overestimated in prior studies using fixed frequency bands or absolute power measures. This approach, together with separate analyses of frontal and centro- parietal spindles, may explain why we observed selective age effects rather than widespread age- related reductions. The Impact of Insomnia on NREM Neural Activity Within Fz, relative SWA appeared largely preserved in insomnia, whereas SO and spindle density, and relative frontal sigma power (but not central) emerged as key markers of sleep quality that were disrupted by insomnia disorder. This pattern highlights distinct measure- specific effects, emphasizing how event-based and spectral indices capture complementary aspects of cortical synchronization and NREM homeostasis. Insomnia was the primary driver of reductions in SO density, consistent with hyperarousal and shallower sleep depth 6,76 in insomnia, a pattern further supported by elevated sleep fragmentation indices and reduced NREM3 in our insomnia sample. In contrast, relative SWA was not impacted by insomnia, aligning with prior work indicating that insomnia-related SWA deficits may be more reliably detected in SO- specific measures 8,9 rather than broadband delta power 8,9,26,68,77–79. In fact, delta power can decrease24, remain relatively stable or even increase during sleep deprivation even when other markers of sleep depth are disrupted 80. Event-based measures of low-frequency (0.25-1.5 Hz) cortical synchrony, such as SO density, may be particularly sensitive to sleep fragmentation and hyperarousal 81 that are often present in individuals with insomnia. This is because coordinated activity across widespread cortical networks are strongly influenced by arousal-related subcortical structures such as the thalamus 82. In contrast, SWA is a broader and less temporally specific measure, averaging spectral power (up to 40Hz) across entire epochs. As a result, SWA can appear relatively preserved even when the fine-grained frequency or organization of individual SO events is disrupted 80. It is also quite possible that certain insomnia subtypes (e.g., insomnia with physiological hyperarousal) may disrupt slow-wave microstructure, without altering global measures of homeostatic sleep regulation 8,23,68,77,83. .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 19, 2026. ; https://doi.org/10.64898/2026.03.17.712450doi: bioRxiv preprint SOs also represent coordinated switches between cortical states that influence the timing and expression of other sleep rhythms, and play a critical role in coordinating spindles, supporting memory consolidation, and facilitating cellular recovery 81. Therefore, it is unsurprising that the presence of insomnia was also associated with robust and consistent alterations in spindle characteristics. These findings align with prior research 26,84,85 and studies demonstrating that reduced spindle activity is a reliable marker of disrupted sleep 12,22,23. In this context, insomnia emerged as the primary driver of reductions in frontal and central spindle density and contributed to decreases in relative frontal sigma power, underscoring the particular vulnerability of spindle- generating mechanisms in insomnia. Notably, studies reporting reductions in spindle activity similar to those observed here employed frontal and central EEG electrodes and were of good methodological quality 84,85. Although findings regarding spindle impairment in insomnia have been mixed, this variability is likely attributable to methodological heterogeneity 86 and the complexity of the disorder. Specifically, differences in study design may explain some inconsistencies: for example, some studies have found increased sigma power in insomnia 77,87 but these findings were often limited by factors such as narrow age ranges, restricted spindle frequency bands (e.g., 12.5–16 Hz) and localization exclusively to the left central region (e.g., C3), which reduces sensitivity to frontal spindles. Despite these differences, our results are consistent with broader literature highlighting the role of spindles in promoting sleep stability, and buffering stress 11,88,89. Moreover, they support evidence that spindle activity may serve as a useful predictor of response to Cognitive Behavioural Therapy for Insomnia 90(CBTi; the gold standard treatment for insomnia91). The Impact of Sex on NREM Neural Activity in Insomnia Contrary to much of the existing literature in healthy populations 26,29,92, we did not observe sex differences in spindle density or SO density in any group. This was unexpected, as females have been shown to exhibit higher spindle density and, in some studies, greater SWA 26. One possibility is that sex differences in event-based measures are partially contingent on age distribution, hormonal status, or topographic specificity, factors that may differ across studies. In this context, the absence of spindle and SO density differences in our sample suggests that sex effects in discrete oscillatory events may not be as robust across samples as spectral findings. Indeed, we found that relative frontal sigma power and SWA emerged as the primary sleep features differentiating the sexes, with sex exerting a primary influence on relative SWA. Although healthy males and females as well as male sand females with insomnia did not differ in relative SWA, a sex differences emerged only when sleep groups were combined. Males showed lower relative SWA overall, consistent with findings on absolute SWA 29,48,93–102, but not relative delta power95,97,98,103,104. When examined within groups, reduced sample size may have limited statistical power to detect small effects. Collapsing across sleep groups increases power and may have allowed a subtle overall sex difference in SWA to become detectable. Importantly, this pattern suggests that sex differences in SWA are not uniquely driven by healthy sleep or preserved in insomnia per se but instead reflect a modest global shift that becomes visible only when statistical power permits. Hormonal Modulation of Sigma Activity: Implications for Female Sleep Stability In insomnia, the typical female advantage in sigma power 26 observed in healthy sleepers disappeared with females exhibiting lower relative frontal and central sigma power than males, supporting the interpretation that insomnia may selectively disrupt mechanisms supporting .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 19, 2026. ; https://doi.org/10.64898/2026.03.17.712450doi: bioRxiv preprint female sleep stability. Although age was statistically controlled for, the average age of the female insomnia sample fell within the peri- to postmenopausal range (>45 years old), when there is progressive ovarian follicle loss and reduced ovarian responsiveness 105. Estradiol and progesterone levels decline, and residual hormonal variability during this transition has been linked to difficulties initiating sleep and increased nocturnal awakenings 106–108. These hormonal changes may be particularly relevant for spindle activity, given the neurophysiological mechanisms underlying their generation. Sleep spindles arise from reciprocal interactions between inhibitory neurons (e.g., GABA) in the thalamic reticular nucleus interacting with excitatory thalamocortical neurons (e.g., glutamate), producing the rhythmic activity that travels through thalamocortical circuits 34. Estradiol and progesterone are thought to modulate GABAergic and serotonergic signaling, thereby supporting thalamocortical excitability and spindle generation109. Consistent with hormonal fluctuation, spindle density has been shown to increase during the luteal phase and with oral contraceptive use during the reproductive years as well as menopausal hormone therapy 110,111 likely via progesterone-enhanced GABAergic transmission within cortical and thalamic networks. Given the central role of spindles in maintaining sleep continuity and depth 19–21, reductions in sigma activity may represent a key mechanism contributing to insomnia vulnerability9–11, and overall higher prevalence and severity rates of insomnia in females. Accordingly, the lower sigma power observed in females with insomnia may reflect the combined influence of female sex-steroid hormones on thalamocortical networks and the clinical expression of sleep disruption, highlighting a plausible mechanistic pathway underlying sex-specific vulnerability to insomnia. In contrast to the sex-specific effects observed for relative sigma power, no differences between males and females with insomnia were observed for frontal or central spindle density, SO density, or relative SWA. To our knowledge, no previous studies have explicitly examined sex differences or sex-by-group interactions in spindle or SO density in insomnia 26. For spindle density in Fz and Cz, both insomnia groups showed lower values than healthy sleepers, suggesting that insomnia broadly affects spindle generation but that the sex-specific vulnerability in females may be most apparent in sigma power rather than event counts. Prior literature supports that spectral sigma power and spindle density do not always correlate62,86. Sigma power reflects the overall energy in the sigma frequency range, whereas spindle density quantifies only the number of clearly defined spindles. Functional studies of developmental sleep 112, and other sleep disorders86,113 have similar divergences. Specifically, sigma power correlates with aging and cognitive performance, but spindle density does not systematically, indicating that these measures capture overlapping but distinct aspects of thalamocortical activity. The absence of sex differences in relative SWA among individuals with insomnia also aligns with the limited existing literature using both relative and absolute measures68,77,79,87,114,115. From a mechanistic perspective, this pattern indicates that while thalamocortical sigma activity may be sensitive to sex-specific processes, such as hormonal fluctuations, the cortical networks generating SO and delta activity seem to be similarly vulnerable to the effects of insomnia across sex. This may be alternatively attributable to a potential compensatory neuroadaptive mechanisms that help maintain sleep homeostasis in the face of gradual hormonal changes. These findings demonstrate that spindle and SWA metrics do not always change in parallel, highlighting the value of integrating both event-based and spectral measures, as each provides distinct and complementary insights into how insomnia and sex shape NREM sleep physiology that a single metric alone might miss. .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 19, 2026. ; https://doi.org/10.64898/2026.03.17.712450doi: bioRxiv preprint Insomnia-Related Sleep Changes in Females Some deficits in spindle and SWA measures were apparent only when females with insomnia were compared to female healthy sleepers, indicating that insomnia may involve deviations from normative female sleep patterns, such as reduced spindle and SO density. This finding is consistent with the higher insomnia severity reported by females in our sample, suggesting that their subjective complaints may reflect disruptions relative to typical female sleep, which can be missed if studies simply compare females to male insomnia profiles. Nevertheless, longitudinal research is required to determine the emergence of these alterations, whether they develop as a consequence of chronic insomnia, or contribute to subjective ratings of sleep quality, and the extent to which they persist over time. Methodological Impacts on Sex Differences in Insomnia Our results revealed both region-specific and global robust patterns across the scalp in how sex and insomnia affect NREM sleep. Sleep group and sex effects in spindle, SO density, and relative SWA were consistent across electrodes. These findings indicate that insomnia-related and sex-related changes in oscillatory activity, were widespread across the cortex rather than restricted to particular regions, suggesting these effects reflect global alterations in oscillatory activities. Frontal channels were particularly sensitive to sigma activity in insomnia, while frontal and posterior channels were sensitive to sex differences in sigma activity and relative SWA. Lastly, frontal and central channels were particularly sensitive to sigma power, with male insomnia participants showing higher frontal and central sigma than female insomnia participants, reflecting more global thalamocortical network activity impacts. Overall, these findings highlight that while some sleep features are regionally specific, others are robust across the scalp, emphasizing the importance of both electrode coverage and individualized analysis in insomnia research. Finally, our findings highlight important distinctions between absolute and relative EEG power when interpreting sex differences in insomnia. Consistent with systematic reviews 24, which suggested that relative power measures may be more sensitive at detecting power spectral differences during NREM sleep in those with insomnia, we observed that sex differences within insomnia varied depending on the power metric used. For SWA, females with insomnia exhibited higher absolute power than males, but not relative power, suggesting that absolute measures may be further confounded by sexually dimorphic anatomical factors such as skull thickness and tissue conductivity 73,94,116,117, rather than true differences in frequency-specific neural processes. In contrast, sigma power showed more nuanced patterns: relative frontal sigma power was lower in females with insomnia compared with males but showed the opposite relationship for posterior absolute power. Taken together, these results suggest that absolute power reflects global amplitude differences influenced by anatomy and age, while relative power, which normalizes power within each individual, may provide a more sensitive and physiologically meaningful index of sex-specific alterations in insomnia. Strengths and Limitations Our study has several methodological strengths that enhance the reliability and validity of the observed patterns. First, we conducted high-quality and low risk of bias research 118, evident by the large, justified sample size, rigorous ascertainment of exposure (i.e., diagnosis via structured .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 19, 2026. ; https://doi.org/10.64898/2026.03.17.712450doi: bioRxiv preprint clinical interview, PSG screening for comorbid sleep disorders, and use of validated clinical measures), careful comparability of groups, objective and validated outcome assessment, and appropriate statistical testing. We also used a merged spindle-detection approach from validated detectors63–66, reducing the bias inherent in any single algorithm and maximizing the robustness of event identification. Detection of brain oscillations was further optimized by applying adapted frequency bands, allowing the algorithm to accommodate well-established inter-individual variability119. Importantly, we examined relative and absolute power to understand true differences in neural activity 116,117. Lastly, this study was the first to examine sex differences in individuals with insomnia across spindle and SO events, and their sleep group × sex interactions. Several limitations of our study warrant acknowledgement. First, even though statistical models adjusted for unequal group sizes, the number of males with insomnia and healthy sleepers was smaller than females, which may limit precision for sex-specific estimates. However, the present sample constitutes one of the largest available cohorts including males with insomnia and healthy sleepers 26, strengthening the interpretability. Second, the sample was drawn from multiple studies that employed harmonized methodology but differed in age-related inclusion criteria, introducing potential heterogeneity in lifespan-related effects. This heterogeneity also represents a strength, allowing examination of oscillatory activity across a broad adult age range, and age was explicitly modeled to reduce confounding. Additionally, one of the four contributing insomnia datasets included participants with comorbid anxiety or depression. While this limits causal attribution, psychiatric comorbidity is common in chronic insomnia and thus enhances ecological validity. Moreover, medication use was carefully controlled across studies, reducing the likelihood that observed effects were driven by pharmacological influences. Lastly, this study did not have information on menstrual cycle phase, menopausal status or hormonal dosage, all of which may have improved our understanding of sex differences in brain oscillations. Future longitudinal and multimodal studies integrating additional oscillatory and clinical measures will be critical for clarifying the mechanisms responsible for these complaints Clinical and Translational Relevance These findings have important clinical and translational implications for improving the prevention and treatment of chronic insomnia. As insomnia is more prevalent in females and contributes substantially to lost productivity, worsening mental and physical health, and health- care utilization 120,121, identifying the mechanisms behind sex- and age-specific vulnerabilities is valuable. By demonstrating that NREM oscillatory features vary systematically by age, sex, and insomnia status, this study highlights physiological markers that could help personalize care. For instance, lower spindle density has been found to predict poorer response to CBT/i1 I 90. Therefore, individuals with insomnia who show impaired spindle synchrony (patterns that appear more common in females based on our current findings), may have an increase vulnerability that maintains symptoms89, and may not experience sufficient physiological change from treatments that primarily target behavioral and cognitive contributors to hyperarousal 122 rather than the underlying oscillatory mechanisms. This subgroup may require CBT-I augmented with physiology-focused interventions that downregulate arousal, strengthen sleep stability, and meaningfully influence neural activity (e.g., exercise, TMS, rocking stimulation) 83,123–129. These profiles therefore offer a pathway toward biomarker-informed treatment matching rather than relying solely on subjective symptom reports. Understanding these sex- and age-dependent .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 19, 2026. ; https://doi.org/10.64898/2026.03.17.712450doi: bioRxiv preprint mechanisms also supports the adaptation of preventive strategies that address biopsychosocial contributors to insomnia (such as genetic risk, caregiving stress, hormonal transitions, and higher rates of anxiety and depression in females) 130–132 ultimately strengthening efforts to promote sleep health across the lifespan133.

Taken together, our findings indicate that, beyond well-established age effects, meaningful sex differences exist in insomnia-related NREM oscillatory activity. These results highlight the need for sleep research and clinical practice to adopt sex- and gender-informed methodologies and carefully stratify samples by age. Future studies should explicitly examine sex-specific and insomnia-related effects on sleep microarchitecture, including meta-analytic approaches to strengthen the evidence base. An important next step is to determine whether first-line treatments such as CBT-I induce sex-specific responses and which domains are most affected (e.g., NREM oscillations, mood, memory, or subjective sleep quality), as evidence in this area remains incomplete and mixed. Understanding oscillatory signatures that differ by sex and age can guide targeted assessment and personalized interventions, moving the field toward mechanism-based, rather than symptom-based, treatment planning. Larger, well-powered, prospectively designed studies with uniformly characterized samples will be essential to refine these conclusions and clarify how age, sex, and insomnia jointly shape sleep physiology across the lifespan. Acknowledgments We would like to thank all participants and volunteers for their work on these studies. A special thank you to our volunteer Isabella Di Matteo for her assistance with data entry and quality checking. We acknowledge the contributions of the following team members who assisted in participants’ recruitment, data collection and data preprocessing: Lukia Tarelli, Kirsten Gong, Margaret McCarthy, Ophelia Fontaine, Sam Gillman, Jean-Louis Zhao, and postdoctoral fellow Mathilde Reyt. We also thank our sleep technologists Madeline Dickson and Elinah Mozhentiy from the Clinique SomnoMed for their contribution to the setup of sleep recordings, Financial disclosure TDV has received consultant and speaker fees from Eisai, Idorsia, Paladin Labs, Takeda and Axsome, as well as research grants from Jazz Pharmaceuticals and Paladin Labs. Non-financial disclosure None to report. Funding This work was funded by grants to TDV from the Natural Sciences and Engineering Research Council of Canada (NSERC), the Canadian Institutes of Health Research (MOP 142191, .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 19, 2026. ; https://doi.org/10.64898/2026.03.17.712450doi: bioRxiv preprint PJT153115) and the Fonds de Recherche du Québec (FRQ) and supported by a grant to NAW No. 767-2023-2413 from the Social Sciences and Humanities Research Council. Data availability All code used in this study was developed by the authors for the execution of the study and is available online at https://github.com/nathanecross/seapipe. Raw data are restricted due to legal/ethical considerations. However, they may be shared with other investigators upon reasonable request and evaluation of such request by our local ethics review board: [email protected] .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 19, 2026. ; https://doi.org/10.64898/2026.03.17.712450doi: bioRxiv preprint

1. Morin CM, Vézina-Im LA, Chen SJ, et al. Prevalence of insomnia and use of sleep aids among adults in Canada. Sleep Med. 2024;124:338-345. doi:10.1016/j.sleep.2024.09.044 2. Benjafield AV, Sert Kuniyoshi FH, Malhotra A, et al. Estimation of the global prevalence and burden of insomnia: a systematic literature review-based analysis. Sleep Med Rev. 2025;82:102121. doi:10.1016/j.smrv.2025.102121 3. American Academy of Sleep Medicine. International Classification of Sleep Disorders, Third Edition (ICSD-3). Darien, IL.; 2014. 4. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders. 5th ed. 2013. Accessed June 8, 2022. https://dsm.psychiatryonline.org/doi/book/10.1176/appi.books.9780890425596 5. Espie CA, Kyle SD, Hames P, Cyhlarova E, Benzeval M. The daytime impact of DSM-5 insomnia disorder: comparative analysis of insomnia subtypes from the Great British Sleep Survey. J Clin Psychiatry. 2012;73(12):e1478-1484. doi:10.4088/JCP.12m07954 6. Baglioni C, Regen W, Teghen A, et al. Sleep changes in the disorder of insomnia: a meta- analysis of polysomnographic studies. Sleep Med Rev. 2014;18(3):195-213. doi:10.1016/j.smrv.2013.04.001 7. Di Marco T, Scammell TE, Sadeghi K, et al. Hyperarousal features in the sleep architecture of individuals with and without insomnia. J Sleep Res. 2025;34(1):e14256. doi:10.1111/jsr.14256 8. Kao CH, D’Rozario AL, Lovato N, et al. Insomnia subtypes characterised by objective sleep duration and NREM spectral power and the effect of acute sleep restriction: an exploratory analysis. Sci Rep. 2021;11(1):24331. doi:10.1038/s41598-021-03564-6 9. Hogan SE, Delgado GM, Hall MH, et al. Slow-oscillation activity is reduced and high frequency activity is elevated in older adults with insomnia. J Clin Sleep Med JCSM Off Publ Am Acad Sleep Med. 2020;16(9):1445-1454. doi:10.5664/jcsm.8568 10. Perlis M, Pigeon W, Gehrman P, Findley J, Drummond S. Neurobiological Mechanisms In Chronic Insomnia. Sleep Med Clin. 2009;4(4):549-558. doi:10.1016/j.jsmc.2009.07.002 11. Dang-Vu TT, Salimi A, Boucetta S, et al. Sleep spindles predict stress-related increases in sleep disturbances. Front Hum Neurosci. 2015;9:68. doi:10.3389/fnhum.2015.00068 12. Fernandez LMJ, Lüthi A. Sleep Spindles: Mechanisms and Functions. Physiol Rev. 2020;100(2):805-868. doi:10.1152/physrev.00042.2018 13. Mölle M, Bergmann TO, Marshall L, Born J. Fast and slow spindles during the sleep slow oscillation: disparate coalescence and engagement in memory processing. Sleep. 2011;34(10):1411-1421. doi:10.5665/SLEEP.1290 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 19, 2026. ; https://doi.org/10.64898/2026.03.17.712450doi: bioRxiv preprint 14. Massimini, M., Huber, R., Ferrarelli, F., Hill, S., & Tononi, G. The Sleep Slow Oscillation as a Traveling Wave. Journal of Neuroscience. 2004;24(31):6862-6870. doi:10.1523/JNEUROSCI.1318-04.2004 15. Rechtschaffen, A., & Kales, A. A Manual of Standardized Terminology, Techniques and Scoring System for Sleep Stages of Human Subjects. Public Health Service, US Government Printing Office,; 1968. 16. Léger D, Debellemaniere E, Rabat A, Bayon V, Benchenane K, Chennaoui M. Slow-wave sleep: From the cell to the clinic. Sleep Med Rev. 2018;41:113-132. doi:10.1016/j.smrv.2018.01.008 17. Latchoumane CFV, Ngo HVV, Born J, Shin HS. Thalamic Spindles Promote Memory Formation during Sleep through Triple Phase-Locking of Cortical, Thalamic, and Hippocampal Rhythms. Neuron. 2017;95(2):424-435.e6. doi:10.1016/j.neuron.2017.06.025 18. Marshall L, Cross N, Binder S, Dang-Vu TT. Brain Rhythms During Sleep and Memory Consolidation: Neurobiological Insights. Physiol Bethesda Md. 2020;35(1):4-15. doi:10.1152/physiol.00004.2019 19. Dang-Vu TT, McKinney SM, Buxton OM, Solet JM, Ellenbogen JM. Spontaneous brain rhythms predict sleep stability in the face of noise. Curr Biol. 2010;20(15):R626-R627. doi:10.1016/j.cub.2010.06.032 20. Walker MP. The Role of Slow Wave Sleep in Memory Processing. J Clin Sleep Med JCSM Off Publ Am Acad Sleep Med. 2009;5(2 Suppl):S20-S26. 21. Dijk DJ. Regulation and Functional Correlates of Slow Wave Sleep. J Clin Sleep Med JCSM Off Publ Am Acad Sleep Med. 2009;5(2 Suppl):S6-S15. 22. Cox R, Hofman WF, Talamini LM. Involvement of spindles in memory consolidation is slow wave sleep-specific. Learn Mem. 2012;19(7):264-267. doi:10.1101/lm.026252.112 23. Chen C, Wang K, Belkacem AN, et al. A comparative analysis of sleep spindle characteristics of sleep-disordered patients and normal subjects. Front Neurosci. 2023;17. doi:10.3389/fnins.2023.1110320 24. Zhao W, Van Someren EJW, Li C, et al. EEG spectral analysis in insomnia disorder: A systematic review and meta-analysis. Sleep Med Rev. 2021;59:101457. doi:10.1016/j.smrv.2021.101457 25. Pigeon WR, Perlis ML. Sleep homeostasis in primary insomnia. Sleep Med Rev. 2006;10(4):247-254. doi:10.1016/j.smrv.2005.09.002 26. Walsh N, Phillips EM, Paez A, Cross N, Dang-Vu T, Perrault AA. A Systematic Review and Meta-Analysis of Biological Sex Differences in Sleep Spindles and Slow Wave Activity in Adults with and without Insomnia. bioRxiv. Preprint posted online January 23, 2026:2026.01.19.700353. doi:10.64898/2026.01.19.700353 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 19, 2026. ; https://doi.org/10.64898/2026.03.17.712450doi: bioRxiv preprint 27. Chapman R, Najima S, Tylinski ST, et al. (PDF) Sex differences in electrical activity of the brain during sleep: a systematic review of electroencephalographic findings across the human lifespan. ResearchGate. Published online August 10, 2025. doi:10.1186/s12938-025- 01354-z 28. Sidani S, Guruge S, Fox M, Collins L. Gender Differences in Perpetuating Factors, Experience and Management of Chronic Insomnia. J Gend Stud. 2019;28(4):402-413. doi:10.1080/09589236.2018.1491394 29. Djonlagic I, Mariani S, Fitzpatrick AL, et al. Macro and micro sleep architecture and cognitive performance in older adults. Nat Hum Behav. 2021;5(1):123-145. doi:10.1038/s41562-020-00964-y 30. Luca G, Haba Rubio J, Andries D, et al. Age and gender variations of sleep in subjects without sleep disorders. Ann Med. 2015;47(6):482-491. doi:10.3109/07853890.2015.1074271 31. Purcell SM, Manoach DS, Demanuele C, et al. Characterizing sleep spindles in 11,630 individuals from the National Sleep Research Resource. Nat Commun. 2017;8(1):15930. doi:10.1038/ncomms15930 32. Fogel SM, Smith CT. The function of the sleep spindle: a physiological index of intelligence and a mechanism for sleep-dependent memory consolidation. Neurosci Biobehav Rev. 2011;35(5):1154-1165. doi:10.1016/j.neubiorev.2010.12.003 33. Mander BA, Winer JR, Walker MP. Sleep and Human Aging. Neuron. 2017;94(1):19-36. doi:10.1016/j.neuron.2017.02.004 34. de Andrés I, Garzón M, Reinoso-Suárez F. Functional Anatomy of Non-REM Sleep. Front Neurol. 2011;2:70. doi:10.3389/fneur.2011.00070 35. Baker FC, Willoughby AR, Sassoon SA, Colrain IM, De Zambotti M. Insomnia in women approaching menopause: Beyond perception. Psychoneuroendocrinology. 2015;60:96-104. doi:10.1016/j.psyneuen.2015.06.005 36. Mong JA, Baker FC, Mahoney MM, et al. Sleep, rhythms, and the endocrine brain: influence of sex and gonadal hormones. J Neurosci Off J Soc Neurosci. 2011;31(45):16107-16116. doi:10.1523/JNEUROSCI.4175-11.2011 37. Smith PC, Mong JA. Neuroendocrine Control of Sleep. Curr Top Behav Neurosci. 2019;43:353-378. doi:10.1007/7854_2019_107 38. Tobiansky DJ, Wallin-Miller KG, Floresco SB, Wood RI, Soma KK. Androgen Regulation of the Mesocorticolimbic System and Executive Function. Front Endocrinol. 2018;9:279. doi:10.3389/fendo.2018.00279 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 19, 2026. ; https://doi.org/10.64898/2026.03.17.712450doi: bioRxiv preprint 39. Ismailogullari S, Korkmaz C, Peker Y, Bayram F, Karaca Z, Aksu M. Impact of Long-Term Gonadotropin Replacement Treatment on Sleep in Men with Idiopathic Hypogonadotropic Hypogonadism. J Sex Med. 2011;8(7):2090-2097. doi:10.1111/j.1743-6109.2010.02143.x 40. Varga AW, Ducca EL, Kishi A, et al. Effects of aging on slow-wave sleep dynamics and human spatial navigational memory consolidation. Neurobiol Aging. 2016;42:142-149. doi:10.1016/j.neurobiolaging.2016.03.008 41. Zeng LN, Zong QQ, Yang Y, et al. Gender Difference in the Prevalence of Insomnia: A Meta-Analysis of Observational Studies. Front Psychiatry. 2020;11:577429. doi:10.3389/fpsyt.2020.577429 42. Zhang J, Chan NY, Lam SP, et al. Emergence of Sex Differences in Insomnia Symptoms in Adolescents: A Large-Scale School-Based Study. Sleep. 2016;39(8):1563-1570. doi:10.5665/sleep.6022 43. Johnson EO, Roth T, Schultz L, Breslau N. Epidemiology of DSM-IV insomnia in adolescence: lifetime prevalence, chronicity, and an emergent gender difference. Pediatrics. 2006;117(2):e247-256. doi:10.1542/peds.2004-2629 44. Benge E, Pavlova M, Javaheri S. Sleep health challenges among women: insomnia across the lifespan. Front Sleep. 2024;3:1322761. doi:10.3389/frsle.2024.1322761 45. Krishnan V, Collop NA. Gender differences in sleep disorders. Curr Opin Pulm Med. 2006;12(6):383-389. doi:10.1097/01.mcp.0000245705.69440.6a 46. Mallampalli, M. P., Carter, C. L. Exploring Sex and Gender Differences in Sleep Health: A Society for Women’s Health Research Report. Journal of Women’s Health. 2014;23:553-562. 47. Schwarz JFA, Åkerstedt T, Lindberg E, Gruber G, Fischer H, Theorell-Haglöw J. Age affects sleep microstructure more than sleep macrostructure. J Sleep Res. 2017;26(3):277- 287. doi:10.1111/jsr.12478 48. Carrier J, Land S, Buysse DJ, Kupfer DJ, Monk TH. The effects of age and gender on sleep EEG power spectral density in the middle years of life (ages 20-60 years old). Psychophysiology. 2001;38(2):232-242. 49. Cross NE, Pomares FB, Nguyen A, et al. An altered balance of integrated and segregated brain activity is a marker of cognitive deficits following sleep deprivation. PLOS Biol. 2021;19(11):e3001232. doi:10.1371/journal.pbio.3001232 50. Cross N, Paquola C, Pomares FB, et al. Cortical gradients of functional connectivity are robust to state-dependent changes following sleep deprivation. NeuroImage. 2021;226:117547. doi:10.1016/j.neuroimage.2020.117547 51. Perrault AA, Pomares FB, Smith D, et al. Effects of cognitive behavioral therapy for insomnia on subjective and objective measures of sleep and cognition. Sleep Med. 2022;97:13-26. doi:10.1016/j.sleep.2022.05.010 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 19, 2026. ; https://doi.org/10.64898/2026.03.17.712450doi: bioRxiv preprint 52. Weiner OM, O’Byrne J, Cross NE, et al. Slow oscillation-spindle cross-frequency coupling predicts overnight declarative memory consolidation in older adults. Eur J Neurosci. 2024;59(4):662-685. doi:10.1111/ejn.15980 53. Bastien CH, Vallières A, Morin CM. Validation of the Insomnia Severity Index as an outcome measure for insomnia research. Sleep Med. 2001;2(4):297-307. doi:10.1016/s1389- 9457(00)00065-4 54. American Academy of Sleep Medicine. International Classification of Sleep Disorders: Diagnostic and Coding Manual (2nd Ed.). Westchester, IL: American. Academy of Sleep Medicine; 2005. 55. Iber, C., & American Academy of Sleep Medicine. The AASM Manual for the Scoring of Sleep and Associated Events: Rules, Terminology and Technical Specifications. American Academy of Sleep Medicine; 2007. 56. Van Rossum G, Drake FL. Python 3 Reference Manual. Published online 2009. 57. Haba-Rubio J, Ibanez V, Sforza E. An alternative measure of sleep fragmentation in clinical practice: the sleep fragmentation index. Sleep Med. 2004;5(6):577-581. doi:10.1016/j.sleep.2004.06.007 58. Cross N. nathanecross/seapipe. Published online October 27, 2025. Accessed November 4, 2025. https://github.com/nathanecross/seapipe 59. Donoghue T, Haller M, Peterson EJ, et al. Parameterizing neural power spectra into periodic and aperiodic components. Nat Neurosci. 2020;23(12):1655-1665. doi:10.1038/s41593-020- 00744-x 60. Finelli LA, Achermann P, Borbély AA. Individual “fingerprints” in human sleep EEG topography. Neuropsychopharmacol Off Publ Am Coll Neuropsychopharmacol. 2001;25(5 Suppl):S57-62. doi:10.1016/S0893-133X(01)00320-7 61. Cross N, O’Byrne J, Weiner OM, Giraud J, Perrault AA, Dang-Vu TT. Phase-Amplitude Coupling of NREM Sleep Oscillations Shows Between-Night Stability and is Related to Overnight Memory Gains. Eur J Neurosci. 2025;61(7):e70108. doi:10.1111/ejn.70108 62. Cox R, Schapiro AC, Manoach DS, Stickgold R. Individual Differences in Frequency and Topography of Slow and Fast Sleep Spindles. Front Hum Neurosci. 2017;11:433. doi:10.3389/fnhum.2017.00433 63. Lacourse K, Delfrate J, Beaudry J, Peppard P, Warby SC. A sleep spindle detection algorithm that emulates human expert spindle scoring. J Neurosci Methods. 2019;316:3-11. doi:10.1016/j.jneumeth.2018.08.014 64. Ferrarelli F, Huber R, Peterson MJ, et al. Reduced sleep spindle activity in schizophrenia patients. Am J Psychiatry. 2007;164(3):483-492. doi:10.1176/ajp.2007.164.3.483 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 19, 2026. ; https://doi.org/10.64898/2026.03.17.712450doi: bioRxiv preprint 65. Mölle M, Marshall L, Gais S, Born J. Grouping of spindle activity during slow oscillations in human non-rapid eye movement sleep. J Neurosci Off J Soc Neurosci. 2002;22(24):10941- 10947. doi:10.1523/JNEUROSCI.22-24-10941.2002 66. Ray LB, Sockeel S, Soon M, et al. Expert and crowd-sourced validation of an individualized sleep spindle detection method employing complex demodulation and individualized normalization. Front Hum Neurosci. 2015;9:507. doi:10.3389/fnhum.2015.00507 67. Staresina BP, Bergmann TO, Bonnefond M, et al. Hierarchical nesting of slow oscillations, spindles and ripples in the human hippocampus during sleep. Nat Neurosci. 2015;18(11):1679-1686. doi:10.1038/nn.4119 68. Buysse DJ, Germain A, Hall ML, et al. EEG Spectral Analysis in Primary Insomnia: NREM Period Effects and Sex Differences. Sleep. 2008;31(12):1673-1682. doi:10.1093/sleep/31.12.1673 69. Benjamini Y, Hochberg Y. Controlling the False Discovery Rate: A Practical and Powerful Approach to Multiple Testing. J R Stat Soc Ser B Methodol. 1995;57(1):289-300. doi:10.1111/j.2517-6161.1995.tb02031.x 70. Simon KC, Cadle C, Shuster AE, Malerba P. Sleep Across the Lifespan: A Neurobehavioral Perspective. Curr Sleep Med Rep. 2025;11(1):7. doi:10.1007/s40675-025-00322-2 71. Gaudreault PO, Gosselin N, Lafortune M, et al. The association between white matter and sleep spindles differs in young and older individuals. Sleep. 2018;41(9):zsy113. doi:10.1093/sleep/zsy113 72. Dubé J, Lafortune M, Bedetti C, et al. Cortical thinning explains changes in sleep slow waves during adulthood. J Neurosci Off J Soc Neurosci. 2015;35(20):7795-7807. doi:10.1523/JNEUROSCI.3956-14.2015 73. Rosinvil T, Bouvier J, Dubé J, et al. Are age and sex effects on sleep slow waves only a matter of electroencephalogram amplitude? Sleep. 2021;44(3):zsaa186. doi:10.1093/sleep/zsaa186 74. Ujma PP, Simor P, Steiger A, Dresler M, Bódizs R. Individual slow-wave morphology is a marker of aging. Neurobiol Aging. 2019;80:71-82. doi:10.1016/j.neurobiolaging.2019.04.002 75. Gao C, Scullin MK. Longitudinal trajectories of spectral power during sleep in middle-aged and older adults. Aging Brain. 2023;3:100058. doi:10.1016/j.nbas.2022.100058 76. Feige B, Baglioni C, Spiegelhalder K, Hirscher V, Nissen C, Riemann D. The microstructure of sleep in primary insomnia: an overview and extension. Int J Psychophysiol Off J Int Organ Psychophysiol. 2013;89(2):171-180. doi:10.1016/j.ijpsycho.2013.04.002 77. Spiegelhalder K, Regen W, Feige B, et al. Increased EEG sigma and beta power during NREM sleep in primary insomnia. Biol Psychol. 2012;91(3):329-333. doi:10.1016/j.biopsycho.2012.08.009 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 19, 2026. ; https://doi.org/10.64898/2026.03.17.712450doi: bioRxiv preprint 78. Lunsford-Avery JR, Edinger JD, Krystal AD. Optimizing computation of overnight decline in delta power: Evidence for slower rate of decline in delta power in insomnia patients. Clin Neurophysiol. 2021;132(2):545-553. doi:10.1016/j.clinph.2020.12.004 79. Wu YM, Pietrone R, Cashmere JD, et al. EEG power during waking and NREM sleep in primary insomnia. J Clin Sleep Med JCSM Off Publ Am Acad Sleep Med. 2013;9(10):1031- 1037. doi:10.5664/jcsm.3076 80. Ishii T, Taweesedt PT, Chick CF, O’Hara R, Kawai M. From macro to micro: slow-wave sleep and its pivotal health implications. Front Sleep. 2024;3:1322995. doi:10.3389/frsle.2024.1322995 81. Neske GT. The Slow Oscillation in Cortical and Thalamic Networks: Mechanisms and Functions. Front Neural Circuits. 2015;9:88. doi:10.3389/fncir.2015.00088 82. Halassa MM, Sherman SM. Thalamocortical Circuit Motifs: A General Framework. Neuron. 2019;103(5):762-770. doi:10.1016/j.neuron.2019.06.005 83. Kang JM, Cho SE, Moon JY, Kim SI, Kim JW, Kang SG. Difference in spectral power density of sleep electroencephalography between individuals without insomnia and frequent hypnotic users with insomnia complaints. Sci Rep. 2022;12(1):2117. doi:10.1038/s41598- 022-05378-6 84. Shao Y, Zou G, Tabarak S, et al. Spindle-related brain activation in patients with insomnia disorder: An EEG-fMRI study. Brain Imaging Behav. 2022;16(2):659-670. doi:10.1007/s11682-021-00544-2 85. Benbir Ş enel G, Aydın Ö, Tanrıöver Aydın E, Bayar MR, Karadeniz D. Changes in sleep structure and sleep spindles are associated with the neuropsychiatric profile in paradoxical insomnia. Int J Psychophysiol Off J Int Organ Psychophysiol. 2021;168:27-32. doi:10.1016/j.ijpsycho.2021.07.626 86. Weiner OM, Dang-Vu TT. Spindle Oscillations in Sleep Disorders: A Systematic Review. Neural Plast. 2016;2016:7328725. doi:10.1155/2016/7328725 87. Krystal AD, Edinger JD, Wohlgemuth WK, Marsh GR. NREM sleep EEG frequency spectral correlates of sleep complaints in primary insomnia subtypes. Sleep. 2002;25(6):630- 640. 88. Blanco-Duque C, Bond SA, Krone LB, et al. Oscillatory-Quality of sleep spindles links brain state with sleep regulation and function. Sci Adv. 2024;10(36):eadn6247. doi:10.1126/sciadv.adn6247 89. Normand MP, St-Hilaire P, Bastien CH. Sleep Spindles Characteristics in Insomnia Sufferers and Their Relationship with Sleep Misperception. Neural Plast. 2016;2016:1-10. doi:10.1155/2016/6413473 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 19, 2026. ; https://doi.org/10.64898/2026.03.17.712450doi: bioRxiv preprint 90. Dang-Vu TT, Hatch B, Salimi A, et al. Sleep spindles may predict response to cognitive- behavioral therapy for chronic insomnia. Sleep Med. 2017;39:54-61. doi:10.1016/j.sleep.2017.08.012 91. Qaseem A, Kansagara D, Forciea MA, Cooke M, Denberg TD, Clinical Guidelines Committee of the American College of Physicians. Management of Chronic Insomnia Disorder in Adults: A Clinical Practice Guideline From the American College of Physicians. Ann Intern Med. 2016;165(2):125-133. doi:10.7326/M15-2175 92. Gaillard JM, Blois R. Spindle density in sleep of normal subjects. Sleep. 1981;4(4):385-391. doi:10.1093/sleep/4.4.385 93. Hertenstein E, Gabryelska A, Spiegelhalder K, et al. Reference Data for Polysomnography- Measured and Subjective Sleep in Healthy Adults. J Clin Sleep Med JCSM Off Publ Am Acad Sleep Med. 2018;14(4):523-532. doi:10.5664/jcsm.7036 94. Dijk DJ, Beersma DG, Bloem GM. Sex differences in the sleep EEG of young adults: visual scoring and spectral analysis. Sleep. 1989;12(6):500-507. doi:10.1093/sleep/12.6.500 95. Latta F, Leproult R, Tasali E, et al. Sex differences in nocturnal growth hormone and prolactin secretion in healthy older adults: relationships with sleep EEG variables. Sleep. 2005;28(12):1519-1524. doi:10.1093/sleep/28.12.1519 96. Latreille V, Carrier J, Montplaisir J, Lafortune M, Gagnon JF. Non-rapid eye movement sleep characteristics in idiopathic REM sleep behavior disorder. J Neurol Sci. 2011;310(1- 2):159-162. doi:10.1016/j.jns.2011.06.022 97. Munch M, Knoblauch V, Blatter K, Wirz-Justice A, Cajochen C. Is homeostatic sleep regulation under low sleep pressure modified by age? Sleep. 2007;30(6):781-792. doi:10.1093/sleep/30.6.781 98. Armitage R. The distribution of EEG frequencies in REM and NREM sleep stages in healthy young adults. Sleep. 1995;18(5):334-341. doi:10.1093/sleep/18.5.334 99. Colrain IM, Turlington S, Baker FC. Impact of alcoholism on sleep architecture and EEG power spectra in men and women. Sleep. 2009;32(10):1341-1352. doi:10.1093/sleep/32.10.1341 100. Fukuda N, Honma H, Kohsaka M, et al. Gender difference of slow wave sleep in middle aged and elderly subjects. Psychiatry Clin Neurosci. 1999;53(2):151-153. doi:10.1046/j.1440-1819.1999.00508.x 101. Massicotte-Marquez J, Carrier J, Décary A, et al. Slow-wave sleep and delta power in rapid eye movement sleep behavior disorder. Ann Neurol. 2005;57(2):277-282. doi:10.1002/ana.20373 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 19, 2026. ; https://doi.org/10.64898/2026.03.17.712450doi: bioRxiv preprint 102. Jones CW, Larson O, Basner M, Dinges DF. The dynamic responses of mood and sleep physiology to chronic sleep restriction and subsequent recovery sleep. Sleep. 2024;47(9):zsae091. doi:10.1093/sleep/zsae091 103. Landolt HP, Gillin JC. Similar sleep EEG topography in middle-aged depressed patients and healthy controls. Sleep. 2005;28(2):239-247. doi:10.1093/sleep/28.2.239 104. Plante DT, Landsness EC, Peterson MJ, et al. Sex-related differences in sleep slow wave activity in major depressive disorder: a high-density EEG investigation. BMC Psychiatry. 2012;12:146. doi:10.1186/1471-244X-12-146 105. Santoro N, Randolph JF. Reproductive hormones and the menopause transition. Obstet Gynecol Clin North Am. 2011;38(3):455-466. doi:10.1016/j.ogc.2011.05.004 106. Haufe A, Leeners B. Sleep Disturbances Across a Woman’s Lifespan: What Is the Role of Reproductive Hormones? J Endocr Soc. 2023;7(5):bvad036. doi:10.1210/jendso/bvad036 107. Shechter A, Boivin DB. Sleep, Hormones, and Circadian Rhythms throughout the Menstrual Cycle in Healthy Women and Women with Premenstrual Dysphoric Disorder. Int J Endocrinol. 2010;2010:259345. doi:10.1155/2010/259345 108. Del Río JP, Alliende MI, Molina N, Serrano FG, Molina S, Vigil P. Steroid Hormones and Their Action in Women’s Brains: The Importance of Hormonal Balance. Front Public Health. 2018;6:141. doi:10.3389/fpubh.2018.00141 109. Kuhl H. Influence of ovarian cycle on the central nervous system. Therapeutische Umschau. 59(4). Accessed March 17, 2026. https://econtent.hogrefe.com/doi/epdf/10.1024/0040- 5930.59.4.175 110. Plamberger CP, Van Wijk HE, Kerschbaum H, et al. Impact of menstrual cycle phase and oral contraceptives on sleep and overnight memory consolidation. J Sleep Res. 2021;30(4):e13239. doi:10.1111/jsr.13239 111. Antonijevic IA, Stalla GK, Steiger A. Modulation of the sleep electroencephalogram by estrogen replacement in postmenopausal women. Am J Obstet Gynecol. 2000;182(2):277- 282. doi:10.1016/s0002-9378(00)70211-0 112. McClain IJ, Lustenberger C, Achermann P, Lassonde JM, Kurth S, LeBourgeois MK. Developmental Changes in Sleep Spindle Characteristics and Sigma Power across Early Childhood. Neural Plast. 2016;2016:3670951. doi:10.1155/2016/3670951 113. Beaudin AE, Younes M, Gerardy B, et al. Association between sleep microarchitecture and cognition in obstructive sleep apnea. Sleep. 2024;47(12):zsae141. doi:10.1093/sleep/zsae141 114. Bader K, Schäfer V, Nissen L, Schenkel M. Heightened beta EEG activity during nonrapid eye movement sleep in primary insomnia patients with reports of childhood maltreatment. J Clin Neurophysiol Off Publ Am Electroencephalogr Soc. 2013;30(2):188-198. doi:10.1097/WNP.0b013e3182767c4a .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 19, 2026. ; https://doi.org/10.64898/2026.03.17.712450doi: bioRxiv preprint 115. Hall M, Buysse DJ, Nowell PD, et al. Symptoms of stress and depression as correlates of sleep in primary insomnia. Psychosom Med. 2000;62(2):227-230. doi:10.1097/00006842- 200003000-00014 116. Cave AE, Barry RJ. Sex differences in resting EEG in healthy young adults. Int J Psychophysiol Off J Int Organ Psychophysiol. 2021;161:35-43. doi:10.1016/j.ijpsycho.2021.01.008 117. Panov G. Quantitative EEG Analysis: Introduction and Basic Principles | Springer Nature Experiments. Comput Neurosci. Published online 2023:85-91. 118. Wells G, Shea B, O’Connell D, et al. The Newcastle–Ottawa Scale (NOS) for Assessing the Quality of Non-Randomized Studies in Meta-Analysis. ᅟ . 2000;ᅟ . 119. Saletin JM, van der Helm E, Walker MP. Structural brain correlates of human sleep oscillations. NeuroImage. 2013;83:658-668. doi:10.1016/j.neuroimage.2013.06.021 120. Bolge SC, Doan JF, Kannan H, Baran RW. Association of insomnia with quality of life, work productivity, and activity impairment. Qual Life Res Int J Qual Life Asp Treat Care Rehabil. 2009;18(4):415-422. doi:10.1007/s11136-009-9462-6 121. Chalet FX, Saskin P, Ahuja A, et al. The Associations between Insomnia Severity and Health Outcomes in the United States. J Clin Med. 2023;12(6):2438. doi:10.3390/jcm12062438 122. Mitchell LJ, Bisdounis L, Ballesio A, Omlin X, Kyle SD. The impact of cognitive behavioural therapy for insomnia on objective sleep parameters: A meta-analysis and systematic review. Sleep Med Rev. 2019;47:90-102. doi:10.1016/j.smrv.2019.06.002 123. Plante DT, Goldstein MR, Cook JD, et al. Effects of oral temazepam on sleep spindles during non-rapid eye movement sleep: A high-density EEG investigation. Eur Neuropsychopharmacol. 2015;25(10):1600-1610. doi:10.1016/j.euroneuro.2015.06.005 124. Zhang J, Yetton B, Whitehurst LN, Naji M, Mednick SC. The effect of zolpidem on memory consolidation over a night of sleep. Sleep. 2020;43(11):zsaa084. doi:10.1093/sleep/zsaa084 125. Park I, Díaz J, Matsumoto S, et al. Exercise improves the quality of slow-wave sleep by increasing slow-wave stability. Sci Rep. 2021;11(1):4410. doi:10.1038/s41598-021-83817-6 126. Brown RE, Spratt TJ, Kaplan GB. Translational approaches to influence sleep and arousal. Brain Res Bull. 2022;185:140-161. doi:10.1016/j.brainresbull.2022.05.002 127. Khosravi M, Khosravi R. Effectiveness of non-invasive brain stimulation in improving sleep quality in insomnia: a systematic review. Sleep Sci Pract. 2025;9. doi:10.1186/s41606- 025-00137-8 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 19, 2026. ; https://doi.org/10.64898/2026.03.17.712450doi: bioRxiv preprint 128. Perrault AA, Khani A, Quairiaux C, et al. Whole-Night Continuous Rocking Entrains Spontaneous Neural Oscillations with Benefits for Sleep and Memory. Curr Biol CB. 2019;29(3):402-411.e3. doi:10.1016/j.cub.2018.12.028 129. Perrault AA, Cross NE, Vu TTD, Schwartz S, Bayer L. Whole-night gentle rocking improves sleep in poor sleepers with insomnia complaints. bioRxiv. Preprint posted online April 5, 2025:2025.03.31.646264. doi:10.1101/2025.03.31.646264 130. Caruso D, Masci I, Cipollone G, Palagini L. Insomnia and depressive symptoms during the menopausal transition: theoretical and therapeutic implications of a self-reinforcing feedback loop. Maturitas. 2019;123:78-81. doi:10.1016/j.maturitas.2019.02.007 131. Spielman AJ, Caruso LS, Glovinsky PB. A behavioral perspective on insomnia treatment. Psychiatr Clin North Am. 1987;10(4):541-553. 132. Suh S, Cho N, Zhang J. Sex Differences in Insomnia: from Epidemiology and Etiology to Intervention. Curr Psychiatry Rep. 2018;20(9):69. doi:10.1007/s11920-018-0940-9 133. Nowakowski S, Garland S, Grandner MA, Cuddihy L. Adapting Cognitive Behavioral Therapy for Insomnia. Elsevier; 2021. .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 19, 2026. ; https://doi.org/10.64898/2026.03.17.712450doi: bioRxiv preprint Table 1- Age Differences in Measures of Spindle and SWA Across All Participants (Fz) Wald χ ² (df= 1) p Coefficient SE p 95% Lower Upper Difference Event Detection Spindle Density 41.55 < .001 -0.02 0.003 <.001 -0.022 -0.012 ↓ SO Density 0.63 .427 -0.009 0.011 .444 -0.032 0.012 - Power Spectral Analysis Relative Sigma power 2.67 .102 5.70 × 10-5 3.59 × 10-5 .113 -6.10 × 10-6 0.000 - Relative SWA 33.24 < .001 -0.001 0.000 <.001 -0.002 -0.001 ↓ Cz) Event Detection Spindle Density 43.78 < .001 -0.02 0.003 <.001 -.025 -.013 ↓ SO Density 1.47 .225 -0.014 0.011 .222 -0.037 0.008 - Power Spectral Analysis Relative Sigma power 7.21 .007 0.000 1.09 x 10-6 .008 3.58 x 10-5 .000 - Relative SWA 36.75 < .001 -0.002 0.000 < .001 -.002 -.001 ↓ Note. Bias-corrected confidence intervals are presented. p = adjusted p. Significant results are bolded. Sigma power (11.25-16Hz), SWA (0.25-4Hz), SO power (0.25-1.25 Hz), NREM2 + NREM3 Table 2- Age Controlled Group Differences in Measures of Spindle and SWA (Fz) INS (n= 119) HS (n= 103) Wald χ ² (df= 1) p Coefficient SE p 95% Lower Upper Difference Event Detection Spindle Density 2.13 ± .05 2.91 ± .08 74.27 < .001 -0.77 0.09 < .001 -0.96 -0.60 INS < HS SO Density 5.12 ± .14 7.85 ± .37 49.61 <.001 –2.72 0.399 <.001 –3.52 –2.01 INS < HS Power Spectral Analysis Relative Sigma Power .02 ± .001 .03 ± .001 50.63 <.001 -0.01 0.001 <.001 -0.01 -0.007 INS < HS Relative SWA .79 ± .01 .79 ± .01 .125 .723 –0.003 0.008 .731 –0.018 0.013 - Cz) Event Detection Spindle Density 2.33 ± .06 3.13 ± .09 59.10 <.001 -0.80 0.102 <.001 –1.00 -0.60 INS < HS SO Density 5.04 ± .14 7.72 ± .38 47.51 <.001 –2.68 0.390 <.001 –3.45 -1.92 INS < HS Power Spectral Analysis Relative Sigma Power .03 ± .001 .027 ± .001 5.17 .023 0.004 0.002 .027 0.00 0.008 - Relative SWA .75 ± .01 .76 ± .01 0.51 .477 –0.006 0.008 .483 –0.022 0.010 - Note. Values are reported as estimated marginals means (age controlled) ± standard error with bias-corrected confidence intervals. p = adjusted p. Significant results are bolded. Sigma power (11.25-16Hz), SWA (0.25-4Hz), SO power (0.25-1.25 Hz), NREM2 + NREM3, INS = insomnia group; HS = healthy sleeper group. .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 19, 2026. ; https://doi.org/10.64898/2026.03.17.712450doi: bioRxiv preprint Table 3- Age Controlled Sex Differences in Measures of Spindle and SWA (Fz) Male (n= 74) Female (n= 148) Wald χ ² (df= 1) p Coefficient SE p 95% Lower Upper Difference Event Detection Spindle Density 2.44 ± .07 2.60 ± .05 3.19 .074 -0.16 .09 0.065 -0.34 0.004 - SO Density 6.44 ± .34 6.53 ± .23 .048 .827 –0.09 .409 .836 –0.951 0.703 - Power Spectral Analysis Relative Sigma power .03 ± .001 .02 ± .001 9.30 .002 0.004 .001 .007 0.002 0.007 M > F Relative SWA .78 ± .01 .80 ± .004 5.49 .019 –0.02 .009 .032 –0.038 -0.003 M < F Cz) Event Detection Spindle Density 2.68 ± .08 2.78 ± .06 0.94 .333 –0.10 .101 .329 –0.29 0.10 - SO Density 6.31 ± .34 6.45 ± .24 0.12 .733 –0.14 .423 .738 –0.98 0.70 - Power Spectral Analysis Relative Sigma power .03 ± .002 .03 ± .001 2.24 .134 0.003 .002 .135 –0.001 0.007 - Relative SWA .75 ± .01 .76 ± .004 4.35 .037 –0.018 .009 .048 –0.036 0.000 - Note. Values are reported as estimated marginals means (age controlled) ± standard error with bias-corrected confidence intervals. p = adjusted p. Significant results are bolded. Sigma power (11.25-16Hz), SWA (0.25-4Hz), SO power (0.25-1.25 Hz), NREM2 + NREM3, M = male; F = female. .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 19, 2026. ; https://doi.org/10.64898/2026.03.17.712450doi: bioRxiv preprint Table 4- Age Controlled Group by Sex Differences in Spindle and SWA Measures (Fz) Male INS (n= 34) Female INS (n= 85) Male HS (n= 40) Female HS (n= 63) Wald χ ² (df= 3) p Coefficient SE p 95% Lower Upper Difference Event Detection Spindle 2.15 ± .07 2.17 ± .06 2.74 ± .12 3.04 ± .10 77.11 <.001 -0.87 0.12 <.001 -1.10 -0.64 FINS < FHS Density 0.89 0.12 <.001 0.662 1.145 MINS < FHS 0.59 0.14 <.001 0.319 0.869 MINS < MHS -0.57 0.14 <.001 -0.845 -0.30 FINS < MHS 0.30 0.15 .054 0.011 0.611 MHS < FHS SO Density 5.00 ± .19 5.21 ± .15 7.87 ± .61 7.84 ± .46 51.04 <.001 2.64 0.47 <.001 1.74 3.58 FINS < FHS -2.88 0.65 <.001 -4.17 -1.65 MINS < MHS -2.85 0.50 <.001 -3.85 -1.90 MINS < FHS -2.67 0.64 <.001 -3.98 -1.49 FINS FINS Power 0.008 0.002 <.001 0.003 0.012 MINS < MHS 0.005 0.002 .028 0.001 0.009 MINS < FHS -0.01 0.002 <.001 -0.013 -0.007 FINS < FHS -0.013 0.002 <.001 -0.016 -0.010 FINS < MHS Relative SWA .78 ± .01 .80 ± .01 .78 ± .01 .80 ± .01 5.85 .119 - - - - - - Cz) Event Detection Spindle 2.37 ± .10 2.35 ± .06 3.01 ± .12 3.23 ± .12 60.01 <.001 -0.64 0.17 <.001 -0.980 -0.332 MINS < MHS Density -0.86 0.15 <.001 -1.14 -0.572 MINS < FHS 0.66 0.15 <.001 0.382 0.957 FINS < MHS 0.88 0.14 <.001 0.604 1.182 FINS < FHS SO Density 4.94 ± .20 5.13 ± .14 7.68 ± .62 7.77 ± .47 48.12 <.001 -2.83 0.51 <.001 –3.82 –1.85 MINS < FHS -2.73 0.63 .002 -3.99 -1.47 MINS < MHS 2.55 0.67 .002 1.30 3.81 FINS < MHS 2.65 0.47 <.001 1.74 3.62 FINS FINS Power -0.009 0.00 .011 -0.002 -0.015 MINS > MHS -0.008 0.00 .011 -0.002 -0.014 MINS > FHS Relative SWA .75 ± .01 .76 ± .01 .75 ± .01 .77 ± .01 5.43 .143 - - - - - - Note. Values are reported as estimated marginals means (age controlled) ± standard error with bias-corrected confidence interva ls. Only significant post-hoc analyses are presented. p = adjusted p. Results in bold are FDR-corrected comparisons p > .05 but became significant after bias-corrected confidence intervals. Sigma power (11.25-16Hz), SWA (0.25-4Hz), SO power (0.25-1.25Hz), NREM2 + NREM3 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 19, 2026. ; https://doi.org/10.64898/2026.03.17.712450doi: bioRxiv preprint Table 5- Summary of Results Outcome Age Effects Sex Effects* Insomnia Effects Key Interactions Primary Drivers Insomnia Severity X Females > Males ⎯ Female INS > Male INS Sex Sleep Architecture NREM1 X Females < Males X Female INS < Female HS Sex, Sex × Insomnia Both INS groups < than sex-matched HS NREM2, NREM3, REM Declined with age X INS < HS Female INS < Female HS Age, Insomnia, Sex× Insomnia Both INS groups Male INS (in NREM3 only) Declined with age X INS < HS Both INS groups < than sex-matched HS Age, Insomnia Increased with age Females HS Both INS groups > than sex-matched HS Age, Insomnia Increased with age Females < Males X SFI driven by sex Age, Sex X X X X X Relative Sigma Power Fz X Females < Males INS < HS Females INS < Male INS Sex, Insomnia, Sex × Insomnia Both INS groups < than sex-matched HS Relative Sigma Power X X X Females INS Male HS Spindle Density Fz Declined with age X INS < HS Both INS groups < than sex-matched HS Age, Insomnia Spindle Density Cz Declined with age X INS < HS Both INS groups Males X X Age, Sex X X INS < HS Both INS groups < than sex-matched HS Insomnia *Combined INS and HS groups, — Not examined, X no effect, INS= insomnia group, HS= healthy sleeper group, TIB = time in bed; TS T = total sleep time; SOL = sleep onset latency; WASO = wake after sleep onset; SE= Sleep efficiency; NREM = non-rapid eye movement sleep; REM = rapid eye movement sleep; SFI = sleep fragmentation index; .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 19, 2026. ; https://doi.org/10.64898/2026.03.17.712450doi: bioRxiv preprint Group by Sex Female HS Male HS Female INS Male INS Relative Sigma Power (band/total power; µV²) .06 . 05 .04 .03 .02 .01 .00 7 8 9 148 115 Interaction Effects of Group by Sex on Relative Sigma Power (Fz) 1 2 * * *** *** A B Group Group by Sex Female HS Male HS Female INS Male INS Relative Sigma Power (band/total power; µV²) .08 .06 .04 .02 .00 Interaction Effects of Group by Sex on Relative Sigma Power (Cz) * * 0 1 2 3 4 1 2 Spindle Density (events per 30sec epoch) INS HS Interaction Effects of Group by Sex on Spindle Density (Cz) Male Female *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** * .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 19, 2026. ; https://doi.org/10.64898/2026.03.17.712450doi: bioRxiv preprint 0 2 4 6 8 10 1 2 SO Density (events per 30sec epoch) INS HS Interaction Effects of Group by Sex on SO Density (Fz) Males Females****** *** * ** Group Group by Sex Female HS Male HS Female INS Male INS Relative Delta Power (band/total power; µV²) 1.00 .80 .60 .40 .20 .00 Interaction Effects of Group by Sex on Relative Delta Power (Fz) A B Group by Sex Female HS Male HS Female INS Male INS Relative SWA (band/total power; µV²) 1.00 .75 .50 Interaction Effects of Group by Sex on Relative SWA (Fz) .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 19, 2026. ; https://doi.org/10.64898/2026.03.17.712450doi: bioRxiv preprint