Introduction

At the dawn of time, there was neither light nor noise pollution [1–4], but a natural environment for Homo sapiens to evolve in [5, 6]. Under these conditions, humans gravitated towards diurnal activity when sunlight was available and sought rest under the gentle cloak of night [7]. Driven by the relentless forces of natural selection [8], humanity developed in perfect harmony with the slow cadence of life, seamlessly orchestrating the delicate balance between activity and sleep [9]. Thus, through evolution, light became the synchronizer of circadian rhythms [10–12], while sounds served both as a communication tool and a warning of potential danger, triggering survival instincts and biological responses such as increased levels of cortisol (CORT, stress hormone) [13, 14].

However, over the past three centuries, rapid technological advancements have disrupted this delicate balance [15, 16]. The “Industrial Revolution” accelerated urbanization, introduced shift work, and exposed people to increasingly intrusive artificial sounds from machinery [15], later from home appliances [16], and from rising traffic [17, 18]. Simultaneously, the discovery and widespread use of electricity led to pervasive artificial light at night (ALAN) [9, 19]. Both noise and ALAN pollution have since significantly disrupted human circadian rhythms and sleep patterns [20, 21].

Derived from the Latin word circā diēm (about a day), the circadian system is a self-sustained [22] bio-oscillator generating a daily cycle of approximately 24.18 h. This non-24-hour period can vary due to genetic differences [22, 23]. Aligning the human circadian rhythm to a 24-hour cycle is achievable through the entrainment of environmental synchronizers, including light, body temperature, social interactions, and behaviors such as metabolic intake and rest-activity patterns [24–26]. By providing consistent and predictable environmental cues, these synchronizers help properly stage circadian rhythms, supporting sleep-wake cycles and activity routines [22].

In humans, the circadian time-keeping system is governed by the suprachiasmatic nucleus (SCN), a bilateral structure located in the hypothalamus (HYP) [6, 24]. The SCN is synchronized in response to light input, which is recognized as the primary zeitgeber (the time-giver) [10–12]. Functioning as the “master clock” [8, 25, 26], the SCN controls a network of peripheral clocks present in mammalian tissues [26–28], predominantly in peripheral organs such as the liver, spleen, kidneys, heart, and lungs [29]. This regulation occurs through chemical and/or neuronal intermediates, e.g., glucocorticoid receptors, retinoic acid receptors, G-protein-coupled receptors, tyrosine kinase receptors, and Ca²⁺ channels [26, 27]. Additionally, body temperature and feeding are also recognized as dominant zeitgebers for peripheral clocks [26, 30, 31].

The circadian rhythms regulate several key functions, including behavioral patterns such as mood and sleep-wake cycles, as well as physiological processes, e.g., metabolic activities [22, 32–34]. They also govern intricate hormonal rhythms, i.e., melatonin (MLT) [33] and CORT [35], which serve as markers of the circadian day-night cycle [26]. MLT levels peak during the night, while CORT peaks at the beginning of the circadian cycle, preparing the body for wakefulness and activity [26, 31]. Therefore, shifts in the circadian cycle are often associated with changes in body temperature and the timing of MLT secretion relative to an endogenous circadian oscillation reference point [22].

As an integral part of human chronobiology [36], the circadian system plays a crucial role in regulating sleep [22, 37]. Normal sleep involves two distinct states: rapid eye movement (REM) and non-REM (NREM). NREM is intricately categorized into four stages, further classified as NREM stages 1 (N1) to N3, with N3 encompassing stages three and four (Table 1), which are commonly recognized as slow-wave sleep (SWS) [38]. Sleep typically starts with NREM, progressing through these stages before transitioning into REM sleep [38]. This cyclical progression between stages continues throughout the night [39], delineating what is often referred to as sleep architecture [40], as described in Table 1 [38–41].

Empirical evidence highlights that circadian dysfunction and impaired sleep can stem from numerous factors. These include substance use, which disrupts REM sleep [42–45], triggers nightmares, causes insomnia [46], and induces circadian rhythm shifts [37]. Mental health conditions also significantly impact sleep quality [47–49]. Additionally, circadian disruptions, insomnia, sleep breathing disorders, and hypersomnia [50] are linked to viral infections [51], cardio-respiratory disorders [51–55], such as COVID-19 [56], and degenerative diseases [57], i.e., Parkinson’s and Alzheimer’s [58, 59]. Behavioral factors like shift work [60], sleep deprivation [61], irregular sleep-wake patterns [62], and jet lag [63] can also cause “circadian misalignment” (the discrepancies between the circadian rhythm and circadian entrainment). Socioeconomic status [64], demographic factors [65, 66], and environmental stressors [67] may also play a role.

Scholars have increasingly recognized urban-related stressors as significant contributors to disturbed sleep patterns and circadian function. Notably, ALAN often termed light pollution [9, 19, 68, 69], and noise pollution [21, 70–73] have garnered attention. These factors have been linked to disturbances in the secretion of key hormones, i.e., MLT [74, 75] and CORT [35, 76], further underscoring their impact on health, sleep, and the circadian system as a whole [8, 25, 37, 77–79].

Albeit being vigorously discussed individually, the literature lacks comprehensive insights into the synergistic interplay between ALAN and noise, which are common attributes of real-life exposure in urban environments. This gap in knowledge limits understanding of their overall impact on public health, including crucial aspects such as sleep quality and well-being.

This review aims to deepen the scientific understanding of the effects of ALAN and noise on sleep, MLT, and health while elucidating their intricate interplay through three key steps: first, exploring the anatomical pathways supporting vision and hearing; second, examining the individual effects of each stressor on sleep, MLT, CORT and health; and finally, analyzing the combined impact of ALAN and noise on sleep.

Methodology

To deepen the understanding of the acute effects of uncontrolled exposure to ALAN and noise, this review synthesizes both theoretical and empirical findings. It also clarifies the characteristics of these exposures, as identified by scholars. Although it deviates from traditional systematic review methodology and structure, a similar approach was employed. Initially, a comprehensive survey of the literature was conducted using key terms and their combinations, including “circadian”, “sleep”, “ALAN”, “noise”, “health effect”, “eye”, “cones”, “rods”, “melanopsin”, “MLT”, “CORT” and “ear”. This survey encompassed papers published in Google Scholar from 1970 through 2024. Studies were then screened based on criteria such as publication date (favoring recent studies), relevance (human-based studies), peer review, and methodological robustness.

This resulted in the inclusion of 19 studies on ALAN, 19 studies on noise, and 7 studies on the combined effects of ALAN and noise on sleep. Figure 1 presents a breakdown of the studies by category and research type. The full list of studies, organized by category (ALAN, noise, and combined effect) and publication year, is provided in Table 2. This table includes details on methodologies, main results, and limitations. Most of the studies on ALAN and noise were conducted under controlled laboratory conditions, comprising 68.4% and 47.4%, respectively. These studies are recognized for their limitations, including small cohort sizes [80] and difficulties in reflecting real-life conditions [80, 81]. Field-based research accounts for only 21.1% of ALAN studies and 31.6% of noise studies, typically involving small cohorts and the use of expensive, complex, and intrusive gold-standard equipment (e.g., [82, 83]). Moreover, studies on the combined effect of ALAN and noise are scarce, accounting for only 15.6% of all reviewed studies, with only two studies conducted in real-life field conditions and only one study [54] introducing real-time measurements. This underscores the pressing need for large-cohort field research conducted in real-world settings.

Display full size

Figure 1. 

Classification of studies by category and research type. This classification excludes reviews, letters, and books, focusing solely on studies investigating the relationship between artificial light at night (ALAN) and noise with the circadian cycle, sleep, and melatonin (MLT). Additionally, the “laboratory” category encompasses research conducted in medical institutions, research facilities, and controlled clinical settings. Furthermore, the “ecological” category includes both individual-based and population-based epidemiological studies. The full list of studies is detailed in Table 2. Eco: ecological; Lab: laboratory; Retro: retrospective

The photic pathway

In the early 20th century, a new perspective on the mechanics of light emerged through Einstein’s quantum theory [84]. This groundbreaking work revealed light as electromagnetic radiation with a captivating duality [85], existing as both particles and waves [84]. Consequently, light can be quantified through various means, including its illuminance intensity, measured in lux, lumens, or candles, as well as its spectral composition, i.e., the amount of energy (number of photons) as a function of wavelength, assessed in nanometers (nm) [86].

The daily light and dark cycles govern the regulation of circadian rhythms [87] and are calibrated by the eye in response to changes in light intensity and spectral composition [24, 25, 77, 88]. The human eye is recognized as one of the most complex organs in the body [89]. Light enters the eye and meets the cornea, which refracts and transmits the light to the lens, continuing to the inner surface and up to the retina [89]. The retina encompasses three types of photoreceptor cells: rods, cones, and melanopsin [90] (see Figure 2A). Rods facilitate scotopic vision, enabling sight in low-light conditions and discerning black-and-white imagery [91], with maximal sensitivity at 507 nm [24]. Alternatively, cones specialize in photopic vision, enabling the perception of colors [92]. There are three distinct types of cone cells responsive to short (S), medium (M), and long (L) wavelengths, with peak sensitivities at 420 nm, 534 nm, and 546 nm, respectively [24, 33].

Display full size

Figure 2. 

Schematic representation of the photic pathway. Light reaches the retina, which houses rods for low-light vision [94] and cones sensitive to short (S), medium (M), and long (L) wavelengths of light, enabling color differentiation (A) [92, 96]. These photoreceptors transmit signals to retinal ganglion cells (RGCs), which relay visual information via the optic nerve to the lateral geniculate nucleus (LGN) [33, 92, 96] (B). From the LGN, signals are processed in the occipital cortex, including the primary visual cortex (V1), and continue to higher-order brain regions [97] (B). Concurrently, signals from rods and cones reach intrinsic photosensitive RGCs (ipRGCs) [100], which contain melanopsin that peaks at 480 nm [92]. IpRGCs directly transmit signals to the suprachiasmatic nucleus (SCN) as part of the non-visual pathway [6, 92] (C). Melanopsin signals also influence visual perception in V1 [100, 101] (B). The SCN activates various neuronal clusters involved in neurotransmitter secretion, including the locus coeruleus (LC) for noradrenaline (NA), the lateral hypothalamus (LH) for orexin (ORX), the raphe nuclei (RN) for serotonin (5-HT), the tuberomammillary nucleus (TN) for histamine (H), the adrenal gland (AG) for cortisol (CORT) [10, 102, 104], and the pineal gland (PG) for melatonin (MLT) [74, 103, ] (C)

These traditional photoreceptors [33] are connected to retinal ganglion cells (RGCs), which facilitate the transmission of visual stimuli through the optic nerve to the lateral geniculate nucleus (LGN) along the optic chiasm [24, 33, 89]. Information processed within the LGN then travels to the primary visual cortex (V1) and onward to the secondary visual cortices and higher-order brain regions for further processing [93] (Figure 2B).

Alternatively, as suggested by Blume et al. [33] and Mirjam et al. [24], signals from rods and cones are also transmitted to intrinsic photosensitive RGCs (ipRGCs) situated in the inner retina [94]. The ipRGCs contain the light-sensitive photopigment melanopsin [95], which peaks at 480 nm [24]. Signals from the ipRGCs are directly transmitted to the SCN, constituting the non-visual pathway (Figure 2C). Furthermore, melanopsin signals reach V1, where they contribute to visual perception [96, 97] (Figure 2B).

The SCN coordinates various diurnal brain activities, including thermoregulation, heart rate (HR) control, and activation of key neuronal clusters such as the locus coeruleus (LC), lateral HYP (LH), raphe nuclei (RN), tuberomammillary nucleus (TN), and adrenal gland (AG). These regions are instrumental in producing wake-promoting neurotransmitters, including noradrenaline (NA), which enhances alertness but can impair cognition and contribute to the development of Alzheimer’s and Parkinson’s diseases [98]; orexin (ORX), which is involved in the maintenance of arousal, emotions, and behavior [99]; serotonin (5-HT), which affects emotion, reward, pain, cognition, and sleep-wake patterns [100]; histamine (H), which plays a role in the immune response [101]; and CORT, which influences stress levels, blood sugar regulation, and immune system function [102]. Furthermore, the rhythmic activity of the SCN governs the pineal gland (PG), stimulating the secretion of MLT [74, 103, 104] (see Figure 2C).

Noteworthy is the recognition by Mirjam et al. [24] of undefined neurons, organs, and cells contributing to the sleep-wake cycle [24]. Additionally, several scholars have suggested that light can penetrate the skull and reach the HYP, thereby amplifying SCN exposure to light and affecting circadian rhythms and MLT secretion [105–108]. While most studies have been conducted on animals, it should be acknowledged that the level of penetration varies with wavelength and intensity and may differ among species and across ages [109], hence stressing the need for further research.

ALAN and sleep

Considering the intricate anatomy of the eye and the subsequent sensitivity of the SCN to ALAN, researchers have delved into their interconnected dynamics. In an early study, Wams et al. [110] found that exposure to ALAN affected sleep architecture, modulated circadian function related to sleep, and influenced homeostatic sleep pressure [110]. Furthermore, Cho et al. [10] discovered that exposure to dim 40 lux fluorescent light during sleep resulted in longer N1 sleep, shorter SWS, and increased arousal, leading to shallow and unstable sleep [10]. These studies emphasize the profound effect of ALAN intensity during sleep.

Additionally, Wallace-Guy et al. [111] examined the association between ALAN exposure, sleep, and depressed mood. Their results indicated that pre-sleep exposure (three hours before sleep, with a median of 24 lux) had no significant correlation with sleep or depression. However, average daily exposure to ALAN was found to be statistically significant, with shorter sleep (P < 0.01), reduced wake within sleep (P < 0.05), and increased depressed mood (P < 0.01) [111]. In contrast, Didikoglu et al. [112] found a correlation between pre-sleep ALAN exposure and longer sleep onset, particularly when the exposure occurred 30 min before sleep (P < 0.05) [112]. Although poor sleep may be influenced by MLT and the circadian phase-response curve to ALAN [12, 113], discrepancies in results highlight the need for further research. This research should aim to elucidate the circadian response curve by comparing daily exposure with pre-sleep exposure—considering cumulative vs. direct effects—while addressing factors such as exposure onset, duration, and dosage (spectra and intensity).

In another study, Chellappa et al. [114] compared the effects of three polychromatic light sources: 2,500 K (27 lux), 3,000 K (29.8 lux), and 6,500 K (27.6 lux). They observed that exposure to the blue-enriched 6,500 K source for 2 h led to lower NREM electroencephalographic (EEG) density (0.8% and 0.2%), reduced SWS activity (0.4% and 0.3%), and a shorter average total sleep time by 4 and 12.7 min compared to the 2,500 K warm light and 3,000 K classic light, respectively, suggesting an acute effect of blue-spectrum light on sleep architecture [114]. Additionally, Green et al. [115] investigated the impact of short wavelength light (SWL) on sleep. They examined the effect of light emitted from computer screens simulating exposure to different compositions: 80 lux and 350 lux, representing low and high intensities, respectively, combined with wavelengths of 460 nm and 650 nm, representing SWL and long-wavelength spectra. Volunteers were exposed to different combinations of light, which were randomly changed daily. The exposure started at 21:00 and ended at 23:00, after which volunteers were allowed to sleep. They found that SWL exposure correlated with shorter average sleep duration (8.6 min to 16.4 min), an increased number of wakes after sleep onset (WASO, 2.1% and 3.1%), and decreased sleep efficiency (SE, 2.1% and 3.1%, P < 0.01). Moreover, SWL was found to reduce the proportion of SWS in the sleep architecture (6.5% and 7.7%, P < 0.05) when comparing low and high intensities of light [115], highlighting the potential health concerns linked to SWL emitted from computer screens, smartphones, and tablets [115].

ALAN and MLT

Discovered by the American physician Aaron Lerner [116], N-acetyl-5-methoxy-tryptamine commonly known as MLT, is often referred to as the “hormone of darkness” [117]. It is primarily secreted by the PG, though not exclusively [118]. Lewy et al. [119] were the first to observe that exposure to light suppresses MLT secretion in humans. They found that exposure to 2,500 lux for an hour reduced MLT concentrations to nearly daytime levels [119]. Subsequent studies have further elucidated this phenomenon. For example, Zeitzer et al. [12] discovered that plasma MLT suppression began at 10 lux and increased with light intensity, with complete suppression occurring above 200 lux [12]. Moreover, several studies have highlighted the acute impact of SWL, specifically in the range of 470–525 nm [120] and even at a lower wavelength of 460 nm [115], underscoring its association with melanopsin [121, 122]. Recent research by St. Hilaire et al. [123] has shed further light on the intricate interplay between spectral exposure and MLT secretion. Their findings suggest an initial influence of cone photoreceptors in driving MLT suppression and phase resetting, followed by the dominance of melanopsin, which is particularly responsive to light at 481 nm, during prolonged exposures [123].

Interestingly, MLT suppression is not exclusively associated with normal vision. Czeisler et al. [124] demonstrated a 69% decrease in plasma MLT concentration in 3 of 11 blind patients after exposure to bright light, mirroring findings from healthy volunteers who experienced an average 66% decrease [124]. In a later study, Zaidi et al. [125] observed that exposure to monochromatic blue light at 460 nm reduced MLT secretion by 57%, delayed the MLT circadian rhythm by 1.2 h, and increased alertness, as indicated by EEG tests among blind volunteers with non-functional retinas [125]. Hull et al. [126] further confirmed these results, noting a reduction of more than 33% in MLT concentration among five blind patients following exposure to bright light [126].

This evolving understanding underscores the profound impact of light, both in intensity and spectrum, on MLT dynamics, shaping the sleep-wake cycles [127]. Additionally, it highlights the intrinsic role of the ipRGCs via the non-visual pathway and sensitivity to SWL, which suppress MLT, elevate alertness, and impede sleep [125, 126].

MLT concentration fluctuates throughout the day, typically rising one to two hours before bedtime and sharply declining in response to light exposure [128]. Moreover, Gooley et al. [129] discovered that the timing of light exposure influences MLT secretion. They found that pre-sleep exposure to 200 lux delayed MLT onset by 94 min and shortened its duration by 92 min (P < 0.05) compared to exposure to dim light (less than 3 lux) [129]. Therefore, MLT response to ALAN involves both a dose-response inhibition of MLT [123, 130] and a phase response in the temporal modulation of the rhythm [11, 120].

Swop et al. [131] pointed out that MLT sensitivity to light may differ based on factors such as age, gender, and the presence of mental or physical disorders [131]. Furthermore, Higuchi et al. [132] emphasized MLT’s responsiveness to changes in daily light patterns due to seasonal variations. These variations inform the biology of the time of year, helping to synchronize reproductive and biological activities accordingly [132, 133].

MLT exhibits anti-inflammatory, immunomodulatory, and antioxidant properties [134, 135]. It plays a significant role in immune system function [136], promotes tissue regeneration [137], acts as an anti-aging agent [138], reduces macromolecular damage by scavenging harmful free radicals [139], protects DNA integrity against malicious bindings, and helps suppress cancer resistance to treatments [140]. Consequently, suppression of MLT due to ALAN exposure could potentially compromise immune function, leading to an increased risk of early mortality [141].

Health effect of ALAN on MLT, and sleep

Given SCN function as the regulator of key neurotransmitters i.e., NA, ORX, 5-HT, H, CORT, and MLT [74, 93, 103, 104], alongside its central role as the circadian pacemaker, scholars have directed their attention toward unraveling this intricate mechanism shedding light on the health implication of ALAN exposure and consequently MLT suppression for human well-being.

Numerous studies have linked ALAN exposure to hormone-dependent cancer, including breast cancer [142–145] and prostate cancer [146, 147], both of which are associated with MLT suppression [9, 148]. Additionally, MLT suppression and circadian rhythm disruption have been associated with overweight, obesity [149–151], and diabetes [130, 152, 153]. Moreover, MLT plays a crucial role in the synthesis, secretion, and action of insulin [154]. Therefore, circadian misalignment can lead to late-night snacking, contributing to metabolic dysfunction [155]. Furthermore, ALAN has been associated with fatigue [156], circadian rhythm phase shifts [157], and molecular changes in blood vessels and the heart [158], which can contribute to cardiovascular disease development [130, 152, 153, 159]. These conditions collectively have the potential to disrupt sleep [6, 54, 70, 71].

Furthermore, insufficient exposure to sunlight has been linked to vitamin D deficiency, which also suppresses MLT and impedes sleep, leading to similar health dysfunctions [160–163]. The most recent study by Sirhan-Atalla et al. [149] identified ALAN, rather than vitamin D, as a more significant predictor of these effects (P < 0.05).

Notably, the effect of light exposure may vary with age [164], influenced by age-related degradation of the eye, manifested by yellowing of the lens [165]. Nevertheless, invasive interventions, e.g., cataract surgery, have demonstrated a positive effect on sleep [166, 167]. Moreover, as suggested by Shishegar et al. [168], daily exposure to varying color temperatures of light, from 6,500 K in the morning to 2,700 K in the evening, at intensities of 500 lux and 100 lux, respectively, can enhance sleep quality and reduce disturbance among the elderly compared to a constant 2,700 K lighting setup [168]. Additionally, Chellappa [169] identified several factors that may contribute to individual sensitivity to light and consequently affect sleep, including gender, chronotype, genetic haplotypes, individual circadian hallmarks, and retinal sensitivity [169]. Similarly, Xiao et al. [170] reported that elderly women (aged 50–71 years) sleep more than men, and the association between ALAN and shortened sleep is more pronounced in neighborhoods with higher poverty levels [170].

The auditory pathway

Sound is the result of waves and pressure fluctuations produced by mechanical movements and propagated through air or other media. It is characterized by frequency (Hz), wavelength (m), pressure (Pascal), and intensity, also known as sound pressure level (Lp), measured in decibels (dB) [171, 172]. Perception of sound varies subjectively [3], with noise typically classified as unwanted sound [73]. The human hearing threshold ranges between 20 Hz and 20 kHz but can vary with age and health [73, 171]. The sound pressure levels reported herein are A-weighted, which adjusts for the human ear’s varying sensitivity to different frequencies. This provides a decibel reading that more accurately reflects perceived loudness, denoted as dB(A) [173].

The human ears are situated within the temporal bone on both sides of the skull, comprising three primary sections: the outer, middle, and inner ear. They are asymmetric, thus enabling the localization of sound in space [174]. Incoming sound waves first reach the external pinna (auricle). The pinna collects incoming sound waves and channels them through the concha—a shell-shaped cavity—towards the opening of the ear canal. Subsequently, the sound waves travel through the ear canal, which is “S”-shaped, and reach the tympanic membrane, also known as the eardrum, separating the outer ear from the middle ear and vibrating in response to pressure variations carried by sound waves [174] (see Figure 3A). Within the ear canal lies the cartilaginous structure, where cerumen (wax) is secreted and together with hair follicles, aids in skin moisturization and trapping airborne particles [175]. This intricate design of the outer ear facilitates the amplification of sound pressure by approximately 10 dB(A) to 15 dB(A) and the protection of the tympanic membrane [176].

Display full size

Figure 3. 

Schematic representation of the auditory pathway. Sound waves enter through the outer ear, where the pinna and concha collect the sound waves, which are then directed to the tympanic membrane (eardrum) [174] (A). Within the middle ear, acoustic pressure is transduced into amplified mechanical pressure through the three ossicles (malleus, incus, and stapes) [174, 176, 177] (B). This amplified pressure is then transferred to the inner ear through the oval window. In the inner ear, mechanical pressure is transformed into electrical signals by the cochlea. These signals, combined with vestibular inputs, are transmitted via the vestibulocochlear nerve to the brainstem nuclei and the central auditory nervous system (CANS) [174, 176] (C). An indirect pathway involves subcortical areas encompassing the amygdala (AMYG), hippocampus (HP), and hypothalamus (HYP), influencing behavior, cognition, and cortisol (CORT) secretion [178–182] (D)

Within the tympanic cavity, nestled in the middle ear, reside the three smallest bones in the human skeleton: the malleus (hammer), the incus (anvil), and the stapes (stirrup) [177]. Together, they form the ossicle chain, which transfers subtle air pressure fluctuations captured by the tympanic membrane to the oval window, demarcating the boundary between the middle and the inner ear [174] (see Figure 3B). This mechanical arrangement serves as a transducer, bridging the gap between the air-filled environments of the outer ear and the fluid-filled inner ear. The mechanical motion of these bones increases acoustic pressure, effectively addressing the impedance mismatch between the air-filled and fluid-filled media [176].

The inner ear consists of two distinct fluids—the perilymph, rich in sodium, and the endolymph, high in potassium. This chemical contrast establishes an electrical potential difference supporting the sensory function of the inner ear organs [174]. Positioned in the upper region of the inner ear, the vestibular system governs balance, while the cochlea houses the sensory organs responsible for hearing—the organs of Corti. Hair cells within the organs of Corti translate mechanical movements of the cochlear fluids into electric signals, which are then transmitted via the auditory nerves [176]. The vestibulocochlear nerve, formed by the auditory and vestibular nerves, conveys this collective data through the brainstem to the central auditory nervous system (CANS) in the brain, where auditory sensation is processed [174] (see Figure 3C).

Recent studies [178–182] indicate the projection of auditory inputs into non-auditorial subcortical regions via neuronal connections [178]. These regions include the amygdala (AMYG), the hippocampus (HP) [183], and the HYP [179]. The AMYG, nestled in the limbic region, plays a key role in regulating emotions and consequent behaviors. Therefore, noise exposure can evoke feelings of fear, alter human behavior [184], and subsequently impact sleep patterns [185]. Another integral limbic structure, the HP, is primarily associated with learning, cognition [186], and memory consolidation during sleep [187]. Finally, the HYP, one of the oldest and smallest parts of the brain, governs various body functions, including metabolism, thermoregulation, SCN function, and sleep (see explanation above—“the photic pathway”) [188], extending to the regulation of CORT secretion by the AG (see Figure 3D).

Noise, sleep, and CORT

In their report, the World Health Organization (WHO) identified noise as a significant health stressor, with traffic highlighted as the most potent source [189]. Specifically, Ising and Braun [190] shed light on the detrimental effects of noise pollution on the human body, indicating that exposure to continuous sound levels equal to or exceeding 30 dB(A) and maximal noise exceeding 55 dB(A) during sleep elevated CORT and NA levels, contributing to stress-related diseases [190]. Later studies by Babisch et al. [191] and Maschke et al. [192] have corroborated these results.

Noise can affect sleep architecture, alter sleep behavior, and potentially induce insomnia. Vallet et al. [193] indicated a significant phase shift in sleep architecture at average continuous sound levels of 48.5 dB(A), accompanied by awakening events observed at 50.3 dB(A) (P < 0.01) [193]. Additionally, Eberhardt et al. [194] identified changes in sleep patterns at 45 dB(A) (P < 0.01) and an arousal response at continuous sound levels exceeding 55 dB(A) (P < 0.05) [194], while Okada et al. [195] noted changes in sleep patterns even at lower noise intensities of 40 dB(A) [195]. Aasvang et al. [196] revealed a negative association between exposure to railway noise above 50 dB(A) and REM sleep (P = 0.02) [196]. Moreover, noise peaks during the night correlated (P < 0.05) with increased awakening, higher REM sleep, and reduced episodes of SWS and sleep spindles, both considered sleep-protective processes [197]. Insomnia was found to be significantly associated with noise levels exceeding 55 dB(A) (P < 0.05), as established by Halonen et al. [198].

Noise in hospitals has been recognized as a significant stressor in public health, potentially impeding sleep, slowing recovery, and affecting medical staff performance [199, 200]. Thus, Buxton et al. [72] investigated changes in sleep architecture in response to hospital noise stimuli and found an increase in arousal during N2 sleep. Moreover, this noise caused a greater and more sustained elevation in instantaneous HR, suggesting stress and annoyance [72]. In another study, Park et al. [201] assessed the effect of noise on sleep using the Pittsburgh Sleep Quality Index (PSQI) [202]. They discovered that an average daily noise level of 63.5 dB(A) measured in hospital wards reduced sleep quality among 86% of the patients and was significantly associated with sleep disturbance (P < 0.05) [201]. Similarly, Simons et al. [203] indicated that daily exposure to 54 dB(A) had a negative impact on sleep quality (P < 0.05). Guisasola-Rabes et al. [204] further elucidated these findings, highlighting that for every 1 dB(A) increase in noise, a hospitalized patient's score on the Richards-Campbell Sleep Questionnaire (RCSQ) [205] decreased by 3.92 points, indicating poorer sleep quality (P < 0.05) [204].

Prolonged exposure to noise has been found to acutely affect sleep. Gitanjali and Ananth [206] showed that exposure to continuous occupational noise exceeding 75 dB(A) for eight hours during the day reduced SE. Additionally, this exposure significantly decreased the duration of both REM sleep and SWS while also delaying the onset of REM sleep (P < 0.01) [206]. Similarly, a later study by Lin et al. [207] found consistent results, revealing that individuals exposed to an average noise level of 76.8 dB(A) had a lower percentage of SWS and decreased SE (P < 0.05) compared to a low-noise group with an average level of 61.0 dB(A). Furthermore, post-exposure assessments showed associations with increased CORT levels, heightened sympathetic activity, and elevated blood pressure (P < 0.05) [207]. These results underscore the association between pre-sleep exposure to noise and its detrimental effects on sleep quality and physiological responses, which can lead to psychological outcomes such as stress and mental dysfunction.

Röösli et al. [208] suggested that the timing of noise exposure during sleep may influence sleep patterns. They found that exposure to noise within the first four hours of sleep increased sleep latency by 5.6 min per 10 dB(A). Additionally, an increase of 10 dB(A) during the three hours before waking reduced SE by 2–3% (P < 0.05) [208].

Age-related differences in noise sensitivity were highlighted by Öhrström et al. [209]. They compared children (mean age 10.8 years) with their parents (mean age 41.9 years) and found better sleep performance among the children. They also observed an acute effect on sleep metrics at continuous sound levels of 60 dB(A) to 64 dB(A) over 24 h, affecting both children and parents [209].

Bedroom window location may contribute to better sleep, as indicated by Bodin et al. [210], particularly in rooms facing quiet yards, water, or green spaces (P < 0.05). However, a later study by Bartels et al. [211] found no significant differences in self-reported poor sleep between bedroom windows facing streets with no traffic and those facing low-traffic streets. The researchers acknowledged that “streets with no traffic” do not necessarily guarantee a quiet façade. Despite this limitation, sleeping in bedrooms with windows facing streets with no traffic still yields improved sleep quality [211].

Due to the proliferation of wind turbines, noise frequency has been recognized as a potential risk factor. Initial studies suggested that low-frequency noise might affect humans more significantly than infrasound [212]. However, recent research contradicts this notion, highlighting the adverse impact of infrasound [213]. Additionally, low-frequency noise has been linked to increased CORT secretion [214, 215] and is subjectively associated with changes in sleep quality, mood, and annoyance [216–219]. Nevertheless, its correlation with objective sleep metrics remains inconclusive due to insufficient empirical support [216, 220–222].

The combined effect of ALAN and noise

As discussed earlier, both ALAN and noise are recognized as significant disruptors of sleep with negative health consequences. In everyday life, these stressors continuously emanate from myriad sources, including electronic devices, household appliances, advertising billboards, road illumination, motor traffic, commercial areas, and recreational spaces. Consequently, individuals are exposed to a constant barrage of these stressors, similar to uncontrolled emissions stimulating the senses—eyes and ears—with varying intensities and frequencies, encompassing both ALAN and noise.

Moreover, ALAN and noise share intrinsic mechanisms and exhibit comparable characteristics. They affect similar sub-cortical structures, such as the HYP, and contribute to hormonal imbalances, including MLT suppression and CORT hypersecretion. They also lead to behavioral changes, cognitive impairment, stress induction, and a weakened immune response [58, 71, 77, 156–160]. As a result, their combined effects may exacerbate the impact of each individual stressor. Therefore, understanding their combined impact has become increasingly urgent and essential.

In an earlier study, Ohayon and Milesi [223] investigated the relationship between exposure to outdoor ALAN and sleep patterns within a cohort of 19,136 individuals aged 18 to 99 years. Leveraging questionnaires, they collected sleep data and potential confounding factors, including self-reported noise levels. External assessments of ALAN were derived from Defense Meteorological Satellite Program data, using NASA’s Visible Infrared Imaging Radiometer Suite (VIIRS) instruments [224]. They found that increased exposure to outdoor ALAN correlated with delayed bedtimes, later wake-up times, shorter sleep durations, and increased daytime sleepiness (P < 0.01). Intriguingly, noise levels showed no statistically significant correlation with sleep metrics [223].

In contrast, Zhong et al. [153] amalgamated sleep data from the California Teacher Study cohort [225] with assessments of ALAN exposure from the New World Atlas [226] and noise evaluations derived from the US National Park Service Sound Map [227]. Their study revealed that decreased sleep duration was linked to increased ALAN exposure and found a significant correlation between noise levels and sleep latency (P < 0.05) [153].

Cleary-Gaffney et al. [228] explored the subjective interplay of ALAN and noise with psychological distress and sleep using a self-reported questionnaire and the PSQI questionnaire. They found that the perception of ALAN was significantly associated with poorer sleep quality (P < 0.002), while individuals reporting noise in their bedroom had higher PSQI scores (P = 0.018), suggesting poorer sleep. However, noise within the bedroom showed no significant correlation with sleep duration [228].

Schnelle et al. [83] tested a behavioral intervention aimed at limiting exposure to ALAN and noise among nursing home residents. They measured ALAN and noise levels using professional instruments—a light meter and a microphone—and recorded sleep metrics using actigraphy under controlled conditions. While their behavioral intervention yielded significant reductions in ALAN and noise exposure compared to the control group (P < 0.01), no significant changes were observed in sleep patterns. The authors suggest that the sleep results may have been influenced by individuals’ daytime napping and medical conditions, which were not considered in the study. Additionally, despite a significant decrease, the noise intensity after the intervention remained relatively high, with peaks exceeding 50 dB(A).

In another study, Younis et al. [229] explored the effects of ALAN and noise on patients sleeping in intensive care units using a light meter and a sound meter. Sleep quality was assessed through a self-reported questionnaire. The study revealed a mean noise level of 63.9 dB(A), with peak noise reaching 102 dB(A). The measured ALAN intensity averaged 104.1 lux, with peaks reaching 271 lux. Both stressors were identified as significant predictors of poor sleep (P < 0.001) [229].

These studies offer insights into the relationship between ALAN and noise and their effects on sleep. However, they do not thoroughly explore the mechanisms underlying their combined impact. Additionally, while most studies indicate a statistical association, these studies lack precise quantitative measures of this relationship. To address this gap, Gabinet and Portnov [70] conducted a comprehensive examination of the association between ALAN and noise exposure and their combined effects on sleep, controlling for various confounding factors, including age, gender, behavior, health, education, and income [71].

In the first study, Gabinet and Portnov [70] juxtaposed outdoor ALAN exposure derived from NASA’s VIIRS archive [224] with noise data interpolated from road densities based on road maps, comparing these with sleep data collected through questionnaires. Their findings revealed that both increased road density and ALAN intensity were linked to reductions in sleep duration of 4.5% and 3%, respectively, and heightened reported sleep difficulty of 3.5% and 11%, respectively (P < 0.01). Moreover, they uncovered an interaction effect between ALAN and noise. Specifically, they noted that an increase in ALAN within the high-noise exposure group enhanced sleep duration but concurrently increased reported sleep difficulty (P < 0.01) [70]. This correlation may be attributed to the exposure mechanism, where ALAN, constantly emitted into the environment, may infiltrate sleeping rooms, especially in places where residents are reluctant to lower their shades due to hot and humid weather conditions [230].

In their subsequent empirical study, Gabinet and Portnov [71] collected data from 72 volunteers who used smartphones to record ALAN and noise exposures. Sleep metrics were tracked via smartwatches. They found that exposure to ALAN and noise before bedtime led to a significant reduction in sleep duration by 9.7% to 9.9% and SE by 2.4% to 2.8%, respectively (P < 0.01). These results are coherent with earlier studies linking pre-sleep exposure with MLT suppression [12, 120] and CORT hypersecretion [207]. Interestingly, while ALAN exposure during sleep time showed no significant impact on sleep, noise exerted a dominant influence, resulting in a notable reduction in sleep duration by 13.5% and SE by 4.4% [71]. This is similar to the findings observed by Foo et al. [231]. Additionally, the combined effect of ALAN and noise led to a substantial decrease in sleep metrics. Concurrently, during the pre-sleep phase, sleep duration and SE were reduced by 15.3% and 8%, while during the sleep phase, reductions were 14% and 9%, respectively. Based on these findings, the authors conclude that ALAN and noise tend to amplify one another, potentially exacerbating health-related issues [71].

Conclusions

Throughout their lifespan, healthy individuals typically devote roughly one-third of their lives to the essential activity of sleeping [39]. Recognizing the profound significance of sleep for human health [232], sleep hygiene has emerged as a focal point among scholars and medical professionals. It serves as a proactive therapeutic approach aimed at ensuring longevity and maintaining a high quality of life [233].

As urbanization expands and population density increases, urban habitats are inundated with ALAN [19] and disharmony of unnatural sounds from technological marvels such as light emitted diodes (LEDs), smartphones, and tablets, which serve as sources of SWL [6, 115], along with an increased volume of motor traffic [3, 21]. These stressors are now recognized for their detrimental effects, supported by a growing body of research (see, inter alia, [9, 18, 76, 144, 146, 149, 190, 198, 234–243]). Empirical studies presented herein have elucidated how both ALAN and noise disrupt the natural circadian rhythm [25, 33, 38, 244–249], interfere with sleep patterns [10, 111, 112, 114, 192–195, 209], suppress the secretion of MLT [12, 119, 120], trigger the hypersecretion of stress hormones including CORT and NA [190–192], and lead to various health and mental impairments, such as nausea and fatigue [250, 251], cardiovascular disease [82, 130, 152, 153, 159, 252], hearing loss [239, 248], hormone-dependent cancers [9, 142–145, 147, 149, 253–255], and obesity and overweight [149–151].

Both ALAN and noise often emanate from similar sources and persistently emit into the environment. They share similar intrinsic mechanisms, potentially influencing each other’s effects [24, 74, 93, 178–182]. Several scholars have tried to disentangle this interconnected association by studying both ALAN and noise, yielding varied results. While some have suggested that both stressors contribute to poor sleep [70, 153, 228], others have proposed that only ALAN is associated with sleep disturbances [223]. However, these findings often stem from ecological studies that rely on satellite data to estimate nighttime exposures in sleeping environments. These studies typically lack control for individual-level confounders and may suffer from misclassification biases [256]. Personal dosimetry is ideal, as exemplified by Gabinet and Portnov [71], who conducted a real-life study using real-time measurements. This study underscores that pre-sleep exposure to both ALAN and noise acts as a potent stressor with established potential to disrupt sleep patterns. Interestingly, during sleep, noise emerged as the predominant influence on sleep. Furthermore, these stressors frequently synergize, amplifying one another’s adverse effects [71], which can influence overall health.

The present review pioneers the exploration of the combined effects of ALAN and noise on sleep, along with their intricate interaction, shedding light on the prevailing mechanisms underlying these stressors. In doing so, it offers novel insights into the existing body of knowledge. Furthermore, it meticulously outlines recent research on sleep and health dysfunctions linked to these stressors, thereby providing a necessary framework for reshaping current policies and standards.

While comprehensive, it is important to acknowledge the possibility of overlooking pertinent studies, which could result in a knowledge gap and limit the scope of the review. To address this issue, the author has implemented a methodological approach aimed at minimizing the likelihood of missing relevant publications.

Despite the well-established findings outlined, there is a clear need for further research to enrich current understanding [24]. This research can utilize advanced approaches such as smart technology [71] and citizen science principles [257–259] to enable unobtrusive, large-cohort field studies. Such studies would enhance comprehension of the underlying physical processes and stressors associated with sleep [220–222], refine existing models, elucidate the mechanisms of synergy [71], and introduce other potential physical variables, e.g., temperature [260], humidity [261], and air pollution [67]. Furthermore, these research endeavors are fundamental in fostering the development of innovative solutions and approaches, including the use of ALAN and noise as therapeutic modalities [262–276].

Translating research into actionable strategies for driving sustainable environmental change demands a comprehensive approach. This approach includes limiting exposure while advocating for healthy sleep hygiene practices [233, 277]. It can be achieved through strategically integrating policies that engage key stakeholders, foster improved communication about environmental challenges, minimize resistance or barriers to change, and establish common ground that encourages economic investment [278].

Such policies may necessitate the enforcement of more stringent environmental standards [235], e.g., adjusting noise and light thresholds to facilitate recovery and healing in hospitals and nursing homes [83, 229, 279], or securing a healthy environment in urban areas [280, 281]. This could involve advocating for legislative requirements for acoustic materials to ensure a noise-free environment [236, 237], as well as promoting technological advancements, which may include the development of monitoring and control systems [238]. Additionally, initiatives like smart lighting, which adjusts the intensity to align with residents’ needs while preserving environmental health [282], can be encouraged.

Lastly, investing in public awareness campaigns is crucial for promoting responsible behavior [240, 241], and for cultivating a sustainable environment conducive to quality sleep and overall well-being.