A narrative review of mechanisms linking romantic relationship experiences to sleep quality

Abstract

Although a significant body of research has revealed associations between romantic relationship experiences and sleep quality, there has not been clarity regarding the mechanisms underlying such associations. Toward this end, we review the existing studies that have tested mechanisms linking romantic relationship experiences to sleep quality. Guided by both theory and existing research, we organize our review around five key categories of mechanisms that may explain associations between romantic relationship experiences and sleep quality: emotional/affective responses, self-perceptions, social perceptions, self-regulation, and biological functioning. Our review of the literature indicates strong evidence in support of the mediating effects of emotional/affective mechanisms (e.g. emotions and mood states) in explaining associations between various aspects of romantic relationships (e.g. relationship satisfaction, partner conflicts, and attachment orientation) and sleep quality. Although there is ample theoretical support for the other mechanisms proposed, and although all proposed mechanisms have been separately linked to both romantic relationship experiences and sleep quality, few studies have directly tested them, pointing to profitable directions for future research. Understanding underlying mechanisms will enable the development of wise, process-based interventions that target specific mechanisms to improve couple members’ sleep quality and romantic relationship functioning.

Statement of Significance.

This review focuses on sleep quality in the context of romantic relationships and highlights mechanisms that link romantic relationship experiences (e.g. relationship satisfaction, partner conflicts, and attachment orientation) to sleep quality. Most current evidence supports the mediating effects of emotional/affective mechanisms (e.g. emotions and mood states). Other proposed mechanisms (e.g. self-regulation and biological functioning) require more empirical testing in future research. Understanding underlying mechanisms is important for informing the development of wise-to-process interventions that target specific mechanisms to improve sleep quality and romantic relationship functioning.

Sleep is a crucial aspect of daily life. However, many people in America are troubled by sleep problems such as difficulties falling asleep, difficulties remaining asleep overnight, and difficulties waking up in the morning. For instance, in the United States, about 35% of adults report sleeping, on average, less than the recommended seven hours per night [1], and almost half of all Americans say that they feel sleepy during the day between 3 and 7 days per week [2]. At the same time, abundant research has shown that low-quality sleep, such as difficulty falling asleep and short sleep duration, has been associated with many negative health consequences such as increased risks for cardiovascular and metabolic disease, impaired psychological experiences, and ultimately shortened lifespan [3].

Given the importance of sleep for health and well-being, social scientists have investigated factors that may contribute to better or worse sleep quality in the context of romantic relationships, given that most adults sleep with a romantic bedpartner [4]. Researchers have found significant bidirectional associations between romantic relationship experiences and sleep quality [5–7]. Specifically, research shows that people who experienced more negative partner interactions (e.g. conflicts), fewer positive partner interactions (e.g. support provision), and lower overall romantic relationship satisfaction reported experiencing worse sleep quality overnight [5, 8]. Reciprocally, people who experienced worse sleep quality at night reported experiencing more negative partner interactions (e.g. conflicts) and fewer positive partner interactions (e.g. support provision) the next day [9, 10].

Especially lacking in the literature is clarity and understanding regarding mechanisms underlying the associations between romantic relationship experiences and sleep quality. While some researchers have proposed theoretical models delineating possible mechanisms [7], there are no current reviews that summarize the existing research that has tested mechanisms using empirical data. Thus, guided by theory and existing research, we identify five key categories of mechanisms that may explain associations between romantic relationship experiences and sleep quality, as summarized in Figure 1, and we review supporting evidence for each. We first describe how researchers operationally define “sleep quality,” and we discuss theoretical foundations for the proposed mechanisms.

Figure 1.

Theoretical mediation model linking romantic relationship experiences to sleep quality.

Definitions and Validated Measures of Sleep Quality

In the literature we review, researchers typically define and measure sleep quality in standardized and well-validated ways. Sleep quality includes several components that reflect specific aspects of sleep. The Pittsburgh Sleep Quality Index (PSQI) [11] is frequently used to assess self-reports of each aspect of sleep quality, including (1) subjective sleep quality: measured by asking participants to rate their sleep quality overall, (2) sleep latency: measured by asking participants how long (in minutes) it has taken them to fall asleep each night, (3) sleep duration: measured by asking participants how many hours of actual sleep they get at night, (4) sleep efficiency: calculated by dividing participants’ total sleep time by their time in bed, (5) awakening: measured by asking participants how often they have had trouble sleeping because they wake up in the middle of the night or early morning, (6) medication use: measured by asking the participants how often they have taken medicine to help them sleep, and (7) alertness/sleepiness: measured by asking participants to report how often they have had trouble staying awake and keeping up enthusiasm to get things done during the day. The PSQI score represents a summary of these components, with higher scores indicating worse sleep quality. Sometimes researchers focus on specific aspects of sleep quality, such as sleep duration and subjective sleep quality. For example, in daily diary studies, researchers often assess subjective sleep quality using a single item: “How would you rate your sleep quality last night?”. Although most existing studies in the domain of this review use subjective measures of sleep quality, the aspects of sleep quality (e.g. sleep duration, sleep latency, and sleep efficiency) are also assessed using objective methods such as actigraphy, which continuously records body movements for 24 hours each day of the assessment period. When reviewing the literature on mechanisms linking romantic relationship experiences to sleep quality, we indicate the aspects of sleep quality assessed.

Theoretical Foundation for Potential Mechanisms

Attachment theory provides a strong theoretical foundation for understanding mechanisms linking romantic relationship experiences to sleep quality. The theory [12] posits that humans are driven by an innate biological need for felt-security that is fulfilled through interactions with close others (attachment figures) who are responsive to one’s needs. People in secure, satisfying romantic relationships have partners who are supportive and reliable, and their attachment needs are met promptly. Thus, they experience a series of positive effects psychologically (e.g. feel safe, secure, supported, and cared for; experience decreased negative affect and increased positive affect) and biologically (e.g. attenuated neural activation in brain regions associated with threat; decreased heart rate and blood pressure) [13] that could, in turn, positively influence sleep quality [14, 15]. In contrast, people in insecure, unsatisfying romantic relationships often find their partners to be unresponsive (e.g. neglecting and rejecting) or inconsistently responsive to their attachment needs. Thus, they experience a series of negative effects psychologically (e.g. feel unsafe, unworthy, unlovable, neglected, unsupported, and isolated; experience increased negative affect and decreased positive affect) and biologically (e.g. have higher daily cortisol levels) [13] that could, in turn, negatively impact sleep quality [16, 17].

The relationship experiences that attachment theory posits to be key in impacting attachment security (e.g. receiving encouragement and support from romantic partners) also should influence individuals’ self-efficacy (the belief in one’s own ability to reach personal goals and manage tasks) and self-control (one’s capacity to exert control over impulses, emotions, and other dominant responses), which in turn may induce changes in sleep health behaviors (e.g. having a fixed bedtime and wake-up time) and sleep quality. Likewise, individuals’ sleep experiences affect their self-efficacy and self-control, which in turn can affect relationship behaviors (e.g. interacting with their partner in a cheerful and positive manner) and romantic relationship experiences. This is consistent with social cognition theory [18], which proposes that self-efficacy and self-control are affected by the social environment and impact relational and health behaviors.

From this theoretical basis, we identify and review research in support of five key categories of mechanisms that explain associations between romantic relationship experiences and sleep quality. These include emotional (affective) responses, self-perceptions, social (relational) perceptions, self-regulation, and biological functioning. Because women are more sensitive to both the positive and negative aspects of marriage [19], and they experience poorer sleep quality than men overall [20], there could exist potential gender moderation effects. Although not all studies have tested gender differences for a variety of reasons (e.g. recruitment of all-female samples, small sample size, and gender difference not being a focal interest), we indicate the results for studies that did so in this review.

Emotional/affective response mechanisms

Emotional/affective responses to relationship experiences should mediate links between romantic relationship experiences and sleep quality. Emotional/affective response variables that have theoretical and empirical support as a mechanism include emotions and mood states, stress, and felt-security.

Emotions and mood states.

Affective mechanisms are the most rigorously investigated mechanisms underlying the associations between romantic relationship experiences and sleep quality in the current literature. This makes sense because emotions and mood states (e.g. anxiety, depression, and loneliness) are strongly linked with romantic relationship experiences [21–25], and sleep quality is both affected by and affects emotions and mood states [15, 17, 26, 27].

Several studies have directly investigated emotions and mood states as mechanisms underlying the associations between various aspects of romantic relationships and sleep quality in different populations. In daily-diary [10, 28] and cross-sectional [29, 30] studies conducted with community couples and adults in romantic relationships, negative affect mediated the significant associations between romantic relationship experiences (e.g. daily marital interactions, relationship satisfaction, and attachment orientation) and self-reported sleep quality (assessed using PSQI questionnaire or single item measure for subjective sleep quality) and sleep problems (assessed using sleep disturbance and sleep-related impairment subscales from the Patient-Reported Outcomes Measurement Information System). One daily diary study conducted with older couples also found positive affect to be a mechanism linking subjective sleep quality with marital interactions [31].

Among a variety of mood states, researchers have especially focused on the mediating effects of anxiety and depression. In both longitudinal [32, 33] and cross-sectional [34, 35] studies, anxiety and depression mediated associations between negative aspects of romantic relationships (e.g. psychological abuse, conflict with partner, and attachment anxiety) and both self-reported and actigraphy-assessed sleep quality (e.g. subjective sleep quality and sleep disturbances, actigraphy-assessed sleep efficacy, and sleep latency) among married or cohabiting couples. Generally, negative aspects of romantic relationships (e.g. greater psychological abuse, greater conflict with partner, and higher attachment anxiety) were related to greater symptoms of anxiety and depression, which in turn were associated with worse sleep quality. Also, research [36] has focused on the mediating effects of another mood state, loneliness, which is defined as a subjective, negative feeling related to one’s own experience of deficient social relations [37]. This daily diary study using a sample of military-connected couples found that loneliness mediated the associations between romantic capitalization support (i.e. responsive support for positive life events) and subjective sleep quality [36]. Specifically, this study revealed both actor and partner effects including indirect effects of poor capitalization support on own worse sleep outcomes via own perceived loneliness for both veterans and spouses (actor effects), and indirect effects of veteran’s poor capitalization support on spouse’s worse sleep quality via spouse’s perceived loneliness (partner effects) [36].

Additional studies, one correlational and one longitudinal, showed that increased anger accounts for the association between poorer overall sleep quality (assessed using PSQI or a one-item measure for subjective sleep quality) and reduced relationship quality [38]. Researchers also partially replicated this finding in a quasi-experiment: they experimentally induced varying affective states among romantically involved undergraduates and found that poor sleepers reported decreased perceptions of relationship quality when exposed to anger induction. However, they also found that poor sleep quality was associated with greater baseline anger across the emotion induction conditions, suggesting that poor sleep may increase anger in general [38].

Generally, the existing studies found that more positive romantic relationship experiences (e.g. capitalization support) and fewer negative romantic relationship experiences (e.g. conflicts) are linked with better sleep quality via lower levels of negative emotions and mood states and higher levels of positive affect. However, one study of osteoarthritis (OA) patients (≥50 years old) and their spouses did not find negative affect to mediate the association between subjective sleep quality and refreshing sleep (i.e. how refreshed or rested they felt after the previous night’s sleep) and marital tension [39]. This study found that morning anger resulted from unrefreshing sleep; however, morning anger did not predict marital tension throughout the rest of the day. It may be that older adults quickly recover from negative emotions experienced in the morning such that they do not influence marital interactions throughout the day. It is also possible that couples coping with a chronic illness may have romantic relationships and sleep experiences that are more strongly impacted by the chronic illness itself. In addition, older couples coping with a chronic illness might have developed a higher tolerance threshold for negative affect or better skills in regulating it. As a result, other mechanisms might play a more substantial role in mediating links between sleep quality and marital functioning in older couples coping with a chronic illness. It will be important for future research to replicate and explain this study’s findings with additional empirical data.

Similarly, another daily diary study conducted with older couples [10] did not find mediating effects of positive affect linking daily partner interactions to subjective sleep quality. This may be explained by different measures used to assess positive affect across studies, as research shows that specific types of positive affect influence sleep quality differently [40]. For example, high-activation positive affect (e.g. excitement) might cause physiological arousal before bed and interfere with sleep quality [41], while low-activation positive affect (e.g. calm) might help people relax and benefit their sleep [40]. Positive interpersonal emotions (e.g. compassion) and positive non-interpersonal emotions (e.g. calm and happy) might also exert different impacts on sleep quality. It will be important for future research to test the mediating effects of specific types of positive affect to discover how the various forms (e.g. calm, content, grateful, happy, excited, compassionate, and sympathetic) may differentially link romantic relationship experiences to sleep quality.

Of the three daily diary studies that tested for gender differences in the links between daily marital interactions and sleep quality, one study found no significant moderation effects of gender [28]; one study found negative affect to be a more salient mechanism for husbands [10]; and one study found positive affect to be a more salient mechanism for wives [31]. Of the two studies that tested for gender differences in the links between psychological abuse and sleep quality, one longitudinal study found depression to be a more salient mechanism for wives [33], while the other cross-sectional study found anxiety to be a more salient mechanism for wives [34]. Overall, the studies testing gender differences show the strongest mechanistic effects for wives; however, more research is needed to determine a clear pattern.

Stress.

Theory and research also support stress as an affective mechanism linking relationship experiences and sleep quality. Although stress has been conceptualized in a variety of ways [42, 43], researchers have predominately defined it as appraisals that environmental demands exceed one’s coping resources [44–46]. Included as a stress mechanism are extremely stressful states resulting from shocking or dangerous life events, known as post-traumatic stress, which are experienced as flashbacks and overreactions to reminders of trauma [47]. Stress (and post-traumatic stress) have been significantly associated with both romantic relationship experiences [48, 49] and sleep quality [50, 51].

One longitudinal study has directly investigated stress as a mechanism underlying the associations between romantic relationship experiences and sleep quality [52]. The researchers recruited middle-aged to older participants and found that relationship effectiveness (i.e. the degree to which a person has attained competence in romantic relationships) positively predicted subjective sleep quality (but not self-reported sleep duration) via reduced stress exposure [52]. Relatedly, another longitudinal study utilized a sample of couples randomly assigned to a couple relationship education program and found that both men’s and women’s improved core couple skills (i.e. conflict management and self-care skills) predicted better overall sleep quality assessed using PSQI through lower stress levels [53]. Dyadically, men’s and women’s better conflict management skills were associated with their partner’s better overall sleep quality via partner’s lower stress level; and men’s better self-care skills were associated with their partner’s better overall sleep quality via partner’s lower stress level [53]. In addition, two cross-sectional studies have shown that post-traumatic stress disorder (PTSD) symptoms (e.g. stressful dreams, flashbacks, physical reactions, and easily startled) mediated the links between romantic relationship experiences and self-reported overall sleep quality assessed using PSQI [54, 55]. Conducted with women who experienced domestic violence, these studies revealed significant associations between intimate partner violence and self-reported sleep quality (i.e. worse global sleep quality and greater sleep disturbances), and PTSD symptomatology partially mediated the associations [54, 55].

It is surprising that there have not been more empirical investigations testing stress as a mechanism linking romantic relationship experiences to sleep quality given that stress has been separately linked to relationship experiences and sleep quality in many studies. For example, researchers have found that people in healthy and satisfying relationships (characterized by more responsive support and positive interactions) have lower levels of perceived stress [49], while those in abusive relationships (e.g. experiencing emotional and physical violence from partners) are likely to develop symptoms of PTSD [56]. Likewise, stress, even from daily hassles, has a negative impact on relationship quality [48, 57]. Moreover, the links between sleep quality and stress are also well-documented. For example, overnight sleep deprivation significantly increases perceived stress [16], and perceived stress and PTSD symptoms predict worse sleep quality. Research has shown that even exposure to a stressful video prior to sleep caused poorer sleep, particularly if it activated memories of stressful life events [50]. Additionally, PTSD patients have worse sleep quality overall (e.g. more light sleep and less slow wave sleep) compared to people without PTSD [51]. The strong research showing separate links between stress and both relationship experiences and sleep quality points to the need for more research to empirically establish it as a mechanism.

Felt-security.

A major postulate of attachment theory [12] is that people, across the lifespan, are driven by the fundamental need for felt-security (i.e. feeling safe and secure). Attachment theory [12] postulates that core relationships in which attachment figures are available and supportive (sensitively responsive), especially during times of stress or difficulties, promote felt-security. Consistent with the theory, felt-security has been linked with positive relationship cognitions [58, 59] and increased relationship satisfaction [60]. Moreover, felt-security plays a crucial role in one’s ability to fall asleep because sleep is a defenseless condition and people need to feel secure and safe to reduce vigilance and fall asleep [14, 61]. Corroborating this, researchers have shown that increased felt-security is associated with reports of better sleep quality [29, 62, 63]. Thus, current theoretical and empirical work [60, 61] points to felt-security as an important mechanism underlying the associations between romantic relationship experiences and sleep quality.

Despite its theoretical importance and links to both romantic relationship experiences [60] and sleep quality [61], only one empirical publication, to our knowledge, has directly examined felt-security as a mechanism linking romantic relationship experiences to sleep quality. This cross-sectional research with a sample of younger to middle-aged adult couples (study 1) and a sample of older adult couples (study 2) showed that relationship-specific felt-security mediated the association between one’s general attachment orientation and self-reported sleep quality assessed using PSQI [29]. Specifically, couple members with insecure attachment (i.e. higher attachment anxiety and/or attachment avoidance) experienced worse sleep quality via lower levels of relationship-specific felt-security [29]. This indicates that, despite the dearth of research directly testing it thus far, felt-security is likely to be a strong mechanism linking romantic relationship processes to sleep quality. This mechanism requires further empirical testing in future research to replicate and establish its effects.

Self-perception mechanisms (self-efficacy)

Self-efficacy (i.e. the belief in one’s own ability to reach personal goals and to manage tasks) [18] should be a mechanism linking romantic relationship functioning to sleep quality given its strong links to both romantic relationship functioning [64] and sleep quality [65]. First, research has shown that supportive behaviors by a romantic partner in challenging situations predicted the partner’s self-efficacy, conceptualized in these studies as perceived capability [66, 67], and confidence in approaching challenging activities [68]. Moreover, relationship factors (e.g. investment, support, and satisfaction) significantly predicted the self-efficacy reports of patients with type 2 diabetes [64]. Reciprocally, greater self-efficacy has been linked with greater relationship satisfaction and relationship maintenance behaviors [69].

Self-efficacy also has been linked to sleep quality. Research has shown that sleep problems (e.g. insomnia) were associated with lower self-efficacy [65]. Moreover, perceptions of self-efficacy are theorized to promote individuals’ well-being, including their sleep quality, by increasing feelings of competence, autonomy, and control, and by reducing stress associated with failing to meet important goals [70, 71]. Supporting this, researchers have found that self-efficacy for managing chronic disease was correlated with better nighttime sleep quality [72].

One cross-sectional study that tested self-efficacy as a mechanism (using a sample of long-term breast cancer survivors) found that greater social constraints from other people (e.g. making people feel that they cannot talk about their breast cancer because it makes others uncomfortable) were associated with greater self-reported sleep disturbances through lower self-efficacy for symptom management [73]. Relatedly, a daily diary study with a sample of older married couples revealed that people with greater spousal support reported more goal progress (i.e. were more self-efficacious in pursuing goals), and goal progress in turn predicted better subjective sleep quality, with no significant gender differences [74]. Although these studies point to self-efficacy as an important mechanism linking romantic relationship experiences to sleep quality, additional studies are needed to fully establish the mediating role.

Social (relational) perception mechanisms

People’s perceptions of their social world, particularly their romantic partners, should mediate links between romantic relationship experiences and sleep quality. Two social perception variables that have theoretical and empirical support as mechanisms include perceived partner responsiveness (PPR) and perceived intimacy.

Perceived partner responsiveness.

PPR refers to the extent to which people believe that their romantic partner cares about, understands, and validates their thoughts and feelings [75]. It has been closely associated with both romantic relationship functioning [76] and sleep quality [77]. First, research shows that relationship variables (e.g. secure attachment) predict PPR [78], and that PPR affects relational outcomes. For example, higher levels of PPR promote emotional expression toward partners [79], positive relationship feelings [79], relationship satisfaction [80], and sexual satisfaction [75].

There is also research supporting associations between sleep quality and PPR. For example, several studies have linked lower perceptions of available social support with poor sleep quality [77, 81, 82]. Theoretically, an inability to perceive one’s social network (particularly a core attachment figure) as supportive and responsive to needs should disrupt all domains of well-being including sleep. Likewise, people who experience poor sleep quality may have difficulty recognizing, appreciating, and perceiving responsive support from their partners when it is provided.

Only one study, to our knowledge, has directly tested PPR as a mechanism linking romantic relationship experiences to sleep quality. This dyadic daily diary study with older adult couples showed that husbands’ higher negative partner interactions and lower positive partner interactions were indirectly associated with their own and their wives’ decreased subjective sleep quality that night via their own lower PPR and subsequent increased negative affect [10]. While this is a rigorous dyadic study that supports PPR as an important mechanism, additional studies are needed to empirically establish the mediating role of PPR in linking romantic relationship experiences to sleep quality. It will also be important to determine if the stronger pattern driven by husbands’ relational behaviors replicates in other studies.

Perceived intimacy.

Perceived intimacy (i.e. subjective perceptions of closeness to one’s romantic partner) is another social perception variable that is supported as a mechanism linking romantic relationship experiences to sleep quality. Perceived intimacy has been linked to both romantic relationship functioning [83] and sleep quality [84]. For example, couple members who report high relationship satisfaction report higher levels of intimacy in their relationships [83], and perceived intimacy is a strong predictor of romantic relationship satisfaction [85]. In addition, studies linking perceived intimacy to sleep quality have shown that daytime intimacy with romantic partners was positively associated with nighttime sleep quality [7, 86]. Relatedly, research has shown that intimate relationship behaviors such as self-disclosure [8] and affectionate touch [87] are associated with better sleep quality.

One empirical study has directly tested perceived intimacy as a mechanism linking romantic relationship experiences to sleep quality. This dyadic daily diary study using a sample of military-connected couples indicated that perceived intimacy mediated the associations between capitalization support (i.e. responsive support for positive life events) and subjective sleep quality [36]. This study revealed both actor and partner effects including indirect effects of capitalization support on own sleep outcomes via own perceived intimacy for both veterans and spouses (actor effects), and indirect effects of veteran’s capitalization support on spouse’s sleep quality via spouse’s perceived intimacy (partner effects) [36]. Although this study supports perceived intimacy as a mechanism linking romantic relationship experiences to sleep quality, future research is needed to build on this foundation to empirically establish the mediating effects in additional studies and samples.

Self-regulation mechanisms

One’s ability to regulate impulses, emotions, thoughts, and behaviors should mediate links between romantic relationship experiences and sleep quality. Two self-regulation variables that have empirical support as a mechanism include self-control and presleep cognition.

Self-control.

Self-control is defined as the capacity to exert control over one’s impulses, emotions, and other dominant responses [18]. Emotion regulation is a specific type of self-control that involves attempts to monitor and modulate emotional experiences [88]. Past research has shown that self-control (e.g. emotion regulation) is related to both romantic relationship functioning [89] and sleep quality [84].

Empirical research supports self-control as a mechanism linking romantic relationship functioning to sleep quality. A cross-sectional study with a sample of married adults revealed that self-reported sleep problems undermined self-control, and lower self-control increased self-reported physical and psychological marital aggression [90]. Relatedly, two cross-sectional studies [91, 92] and one longitudinal study [93] that directly tested the mediating role of emotion regulation found that emotion regulation mediated the relationship between self-reported sleep quality (measured using PSQI) or self-reported sleep disturbance (measured using Youth Self-Rating Insomnia Scale) and interpersonal aggression, including anger, hostility, physical aggression, verbal aggression, and relationship violence perpetration. Of the two studies that tested gender as a potential moderator, no significant gender differences were found [90, 93], indicating that the mediation pathway was consistent across genders.

Additional studies are needed to assess how self-control might mediate the associations between other romantic relationship experiences (e.g. supportive interactions) and sleep quality, given that research has also shown significant associations between positive romantic relationship experiences and self-control. For example, studies have found that self-control is related to increased levels of perspective-taking [94], forgiveness [95], and relationship satisfaction [96]. There is also abundant evidence in support of links between self-control and sleep quality. For example, sleep deprivation and poor sleep quality cause problems with self-control (e.g. emotion regulation) [84, 97, 98]. Reciprocally, decreased self-control can negatively impact sleep quality, causing difficulties in falling asleep and remaining asleep (e.g. via unhealthy behaviors such as going to bed later than intended vs. keeping daily routines) [99]. More empirical research is needed to establish self-control as a mechanism linking not only negative romantic relationship experiences (e.g. marital aggression) but also positive romantic relationship experiences (e.g. supportive interactions) to sleep quality.

Presleep cognition.

Presleep cognition is a specific self-regulation mechanism aimed at controlling bedtime thoughts. The period of time before people go to sleep (while in bed) is known as sleep onset latency [100]. When a person lays down for bed, they often do not immediately begin to sleep. What people think before they sleep (presleep cognition) is crucial in affecting sleep onset. Some people may focus on getting to sleep, while others may think about positive or negative events that happened during the day. People who fail to regulate their cognitive activity may experience rumination, which is defined as repetitive thinking about negative personal concerns and/or about the implications, causes, and meanings of them [101]. Rumination is a negative thought process and is significantly associated with both romantic relationship experiences [102] and sleep quality [103]. Given its role in both relationship and sleep functioning, presleep cognition should be a key mechanism underlying the associations between romantic relationship experiences and sleep quality.

Several studies have found that presleep cognition mediated the associations between relationship-relevant variables (e.g. social isolation, gratitude, and forgiveness) and sleep quality [104–107]. Specifically, one cross-sectional study found that gratitude (e.g. “I am grateful to a variety of people”) predicted better sleep quality assessed using PSQI (including greater subjective sleep quality and sleep duration, and less sleep latency and daytime dysfunction) via more positive presleep cognitions and fewer negative presleep cognitions [104]. Another cross-sectional study found that ostracism (i.e. being ignored or excluded by others) predicted worse subjective sleep quality, assessed using the Sleep Quality Scale [108], via greater presleep cognitive arousal [105]. Moreover, in both cross-sectional and prospective studies, rumination mediated the relationship between both situational forgiveness (forgiveness toward a particular offender) and dispositional forgiveness (the general tendency or willingness to forgive an offender) and subjective sleep quality [106], as well as the relationship between ostracism (social isolation) and insomnia [107] as assessed by the Insomnia Severity Index [109].

These studies provide evidence that general relationship experiences influence subjectively assessed sleep quality via presleep cognition. However, no studies have directly tested the mediating role of presleep cognition linking romantic relationship experiences to sleep quality, which is surprising given its strong links to both romantic relationship experiences and sleep quality. For example, insecure attachment orientation (attachment anxiety and attachment avoidance) has been associated with self-reported presleep cognitive arousal (e.g. worry about falling asleep, cannot shut off thoughts) [110] and higher levels of rumination [111]. Moreover, individuals’ own jealousy about a rival and their partner’s expression of jealousy stimulated rumination [102, 112, 113]. Reciprocally, the tendency to ruminate impaired an individual’s ability to maintain positive feelings about their romantic partner in the face of explicit or implicit reminders of relationship-threatening events [114]. There is also strong evidence linking presleep cognition to sleep quality. For example, researchers have found that presleep cognitive arousal was associated with greater sleep-onset latency, fewer hours of sleep, and greater difficulty in sleeping [103, 115, 116]. Researchers have also found that rumination in the presleep period following a stressful event had a negative impact on sleep quality [117]. Thus, there is strong evidence pointing to presleep cognition as a powerful mechanism. However, additional research is needed to establish the mediating effects of presleep cognition in the associations between romantic relationship experiences and sleep quality in samples of romantic couples.

Biological mechanisms

Specific biological responses to changes in one’s social environment should mediate links between romantic relationship experiences and sleep quality. Four biological variables that have theoretical and empirical support as mechanisms include oxytocin, hypothalamic-pituitary-adrenal (HPA) axis, inflammation, and autonomic nervous system (ANS).

Oxytocin.

Oxytocin is widely known as the “love hormone” due to its role in promoting maternal, pair-bonding, and sexual behaviors [118–120]. Because of its strong association with romantic relationship experiences such as closeness and affectionate touch [121], as well as its association with sleep quality [122, 123], oxytocin should be a significant biological pathway linking romantic relationship experiences to sleep quality.

First, oxytocin has been associated with attachment behaviors and affiliative processes, such as positive communication in male–female pair bonding [124], social touch [125], and positive partner interactions [126]. Moreover, researchers have found that oxytocin administration increased positive communication [124], reduced cortisol and autonomic arousal during couple conflict [127], and increased people’s appraisal of positive aspects of their relationships [128]. Sleep quality has also been strongly linked to oxytocin. For example, sleeping with a partner can promote the release of oxytocin through affiliative (e.g. hugging and holding hands) and sexual behaviors [118, 120], and studies have shown positive effects of oxytocin administration on sleep quality [122, 123].

Despite oxytocin’s strong separate associations with both romantic relationship experiences and sleep quality, no studies to our knowledge have investigated oxytocin as a mechanism linking romantic relationship experiences to sleep quality. However, a few studies have tested and found a moderating effect of oxytocin in the associations between romantic relationship experiences (e.g. everyday couple interaction and social support) and self-reported sleep quality [86, 129]. Specifically, one study found that subjective closeness and positive couple interaction were positively associated with subjective sleep quality, and the associations were stronger for participants who were randomly assigned to self-administer intranasal oxytocin compared to participants who were randomly assigned to self-administer a placebo [86]. Another study found that for women with high oxytocin, support from friends was associated with better sleep quality assessed using PSQI; however, for women with low oxytocin, support from friends was associated with poorer sleep quality because women with low oxytocin may not effectively interpret and utilize available support resources [129]. Future research must consider both the moderating and mediating roles of oxytocin in the associations between romantic relationship experiences and sleep quality to clearly elucidate its effects. We view this as a fruitful area for future research.

HPA axis.

The HPA axis is another powerful candidate as a biological pathway linking romantic relationship experiences and sleep quality as it is one of the most crucial physiological systems [130]. It produces the stress hormone cortisol, and the actions of this physiological system normally are tightly regulated to ensure that the body can quickly respond to stressful events and return to a normal state [131]. Regulation of the HPA axis is strongly related to both romantic relationship experiences [132] and sleep quality [133].

First, researchers have found significant links between romantic relationship functioning and HPA axis functioning. For example, among a sample of mothers, positive marital relationship functioning was associated with higher morning cortisol levels and a steeper decline in cortisol across the day, which is a healthy diurnal pattern [134]. Reciprocally, researchers found that elevated adrenocorticotropic hormone (ACTH) during a conflict discussion in the first year of marriage predicted decreased marital satisfaction 10 years later, and elevated norepinephrine predicted divorce [135]. Both ACTH and norepinephrine are associated with the regulation of the HPA axis. There also exist strong links between sleep quality and HPA axis functioning. Research has shown that sleep has an inhibitory influence on the HPA axis, whereas insomnia increases HPA-axis activity [136]. Moreover, successful versus unsuccessful regulation of the HPA axis impacts sleep quality. Individuals whose HPA axis was constantly activated could not get into a calm state that allowed them to fall asleep and remain asleep [137].

Although no studies to the best of our knowledge have directly tested the mediating role of the HPA axis underlying the associations between romantic relationship experiences and sleep quality, two studies have provided preliminary supporting evidence [138, 139]. In one study, researchers randomly assigned couples to either a single night of total sleep deprivation or a night of normal sleep, and all couples were instructed to discuss a topic of recurrent conflict after the experimental night. Results revealed higher cortisol levels during the conflict discussion for sleep-deprived couples compared to couples in the control condition [138], indicating that sleep-deprived couples felt more tension during the discussions. Greater cortisol levels and tension during conflict discussions have been associated with couple members dealing with the conflicts in a way that was harmful to their relationships [140, 141]. The other study also found significant associations among conflicts, sleep, and HPA axis. Specifically, greater quarreling was associated with worse subsequent sleep quality as assessed using PSQI as well as a dampened cortisol awakening response in highly anxiously attached women the next morning [139]. While these results point to the HPA axis as a strong mechanism, future research is needed to directly assess and establish the mediating function. For example, it is likely that greater conflict, poor handling of conflict, or unresolved conflict between couple members result in a continually activated HPA axis and increased release of cortisol, which in turn negatively affect sleep quality overnight.

Inflammation.

Inflammation is another physiological system that should mediate the links between romantic relationship experiences and sleep quality. Although we are under constant exposure to dangers such as bacteria, we do not always get ill because we are protected by the inflammation system. Like the HPA axis, the actions of the inflammation system normally are tightly regulated to ensure that the body can respond quickly to anything that might endanger health and return to a normal state. Given associations between sleep quality, relationship experiences, and inflammation in prior research [142], inflammation is likely to be an important pathway linking romantic relationship experiences to sleep quality.

First, researchers have found that spending more time co-present with a romantic partner in everyday life was associated with lower C-reactive protection (CRP), which suggests regulated inflammation [143]. Marital stress also has been associated with immune dysregulation, including increased production of interleukin 6 (IL-6) [141, 144, 145]. Significant associations between sleep quality and inflammation also have been well-documented. Sleep problems and poor subjective sleep quality have been shown to contribute to elevated inflammation [146, 147]. Moreover, dysregulation of the inflammation system influences sleep quality: biologically, sleep is regulated by circadian and homeostatic processes, and pathological inflammation disrupts the chemical signaling required to maintain a healthy sleep profile [148].

Despite inflammation being another strong candidate for a biological mechanism linking romantic relationship experiences to sleep quality given its strong relations to both, no studies to our knowledge have directly tested its mediating role. However, in other ways, researchers have investigated the complicated associations among sleep disturbance, inflammation, and social relationship experiences. One study found that self-reported sleep disturbance was associated with heightened systemic inflammation among the general population over a 5-year follow-up, and this association was significantly stronger in those who reported feelings of social isolation [149]. Another related study found that lower levels of plasma IL-6 were predicted by greater objectively assessed sleep efficiency and by more positive social relations among aging women [150]. Also, aging women with the highest IL-6 levels were those with both poor objectively assessed sleep efficiency and poor social relations [150]. Moreover, among individuals who reported low levels of social support, poor sleep quality assessed using PSQI was associated with higher CRP and IL-6, which suggests elevation in the inflammation system [142]. Finally, couple members’ shorter sleep duration was related to higher stimulated cytokine production after marital conflict [151]. Although these studies provide evidence in support of the significant links among inflammation, social relationship experiences, and sleep quality, it remains for future research to directly assess and determine the mediating role of inflammation underlying the associations between romantic relationship experiences and sleep quality.

Autonomic nervous system.

The ANS is another important biological system. It involves the sympathetic nervous system, which is responsible for fight or flight responses and the parasympathetic nervous system, which is responsible for rest and digest [152]. The sympathetic nervous system activates to speed up heart rate and to deliver more blood to areas of the body that need more oxygen to facilitate responses that assist in getting out of danger (e.g. fight or flight responses) [152]. The parasympathetic nervous system kicks in after periods of stress or danger to relax one’s body [152]. Given ANS associations with both romantic relationship experiences [153] and sleep quality [154], it has strong potential as a biological mechanism.

To elaborate, research has shown that relationship behaviors are related to ANS activity. For example, romantic relational aggression is associated with exaggerated fight or flight responses to a conflict discussion (sympathetic activation and parasympathetic withdrawal) [153], whereas relationship depth and low conflict are associated with greater nocturnal diastolic blood pressure dipping [155]. Likewise, sleep quality and the ANS are intimately connected [156]. For example, it has been shown that shifts in the sympathovagal balance toward sympathetic dominance during sleep contribute to decreased sleep maintenance and more disturbances in the sleep architecture of healthy individuals [157]. Reciprocally, poor sleep quality has been linked with greater ANS activation, such as elevated blood pressure and increased heart rate [154].

Although no studies to our knowledge have directly tested the mediating role of ANS activation underlying the associations between romantic relationship experiences and sleep quality, two studies point to it as a mechanism [158, 159]. One cross-sectional study [159], which assessed the association between couple members’ sleep–wake concordance and blood pressure, showed that higher sleep–wake concordance between couple members was associated with lower sleep systolic and diastolic blood pressure and with lower wake systolic blood pressure for both couple members. However, this association was stronger for women, suggesting that women may be more sensitive to the potential benefits of sleep–wake concordance. This study also showed that poorer marital adjustment was associated with higher sleep systolic blood pressure. Relatedly, a longitudinal study found that in recently separated adults, greater sleep complaints predicted significant increases in both systolic and diastolic blood pressure three months later [158]. These findings indicate that romantic relationship experiences, sleep quality, and ANS are all closely related. However, it remains for future research to directly assess and determine the mediating role of ANS underlying the associations between romantic relationship experiences and sleep quality.

Conclusions and Future Directions

We reviewed five major categories of mechanisms that may link romantic relationship experiences to sleep quality, including (1) emotional/affective responses (i.e. emotions and mood states, stress, and felt security), (2) self-perceptions (i.e. self-efficacy), (3) social/relational perceptions (i.e. PPR, perceived intimacy), (4) self-regulation (i.e. self-control, presleep cognition), and (5) biological functioning (i.e. oxytocin, HPA axis, inflammation, and ANS activation). The current literature has predominantly focused on testing emotional (affective response) mechanisms, showing strong evidence in support of the mediating roles of emotions and mood states. There is also empirical evidence in support of the mediating roles of social perceptions (i.e. PPR, perceived intimacy) and self-regulation (i.e. self-control, presleep cognition). There is a lack of empirical research that directly tests the mediating roles of self-perceptions (i.e. self-efficacy) and biological functioning (i.e. oxytocin, HPA axis, inflammation, and ANS activation), although these are strong candidates for mechanisms linking romantic relationship experiences to sleep quality given their separate links to each.

Because romantic relationship experiences strongly impact sleep quality, and sleep quality reciprocally impacts romantic relationship experiences [7], it is important to understand the mechanistic pathways underlying these associations for the purposes of developing effective, wise-to-process interventions to improve both sleep and relationship quality. For example, to target the emotional/affective mechanisms that have gained strong empirical support in the existing literature, we can ask couples to complete brief writing prompts before they go to sleep, instructing them to reflect upon the happy experiences in their marriage and describe situations in which they feel supported by their romantic partners. Completing such brief exercises can theoretically increase their positive affect and decrease their negative affect, which in turn will improve both their sleep quality and relationship quality.

When reviewing the existing evidence for mechanisms, we also noted when there were moderating effects of gender, given that some research has shown that women, overall, sleep more poorly than men [20], and that women are more sensitive to both the positive and negative aspects of romantic relationships [19]. Most existing studies that have tested gender differences found either no significant gender differences or stronger effects for women, consistent with research showing that women are more reactive than men to relationship experiences [160]. However, one study found that only older husbands’ relationship behaviors impacted both their own and their wives’ sleep quality via their own negative affect, consistent with research showing that men in traditional, heterosexual relationships tend to set the emotional tone in their relationships [10]. In addition to considering gender, future research might also consider other individual difference factors (e.g. personality traits such as cynical hostility) [161] that might strengthen or lessen the mediating effects of the proposed mechanisms. This may enable the design of effective interventions for specific populations.

Also, although we proposed the mechanisms to be bidirectional based on theory, there is not currently sufficient evidence in support of them working both ways. Most of the existing studies have tested and found evidence for the proposed mechanisms mediating the effects of romantic relationship experiences on sleep quality, but have not tested the other way around. Specifically, most current studies that have tested the proposed mechanisms have found that individuals’ romantic relationship experiences influence the proposed mediators (e.g. emotions and mood states, perceived stress), which in turn influence their sleep experiences. More studies are needed to assess the other direction (how sleep experiences influence romantic relationship functioning via the proposed mechanisms) and to compare the mechanistic strength of both directions.

This review highlights a number of important directions for future research. First, there is a paucity of research examining biological mechanisms, and we see that as a priority in future work aimed at establishing mechanistic pathways. Second, it is important to note that all proposed mechanisms are likely to be interrelated, and to interact in predicting outcomes. It will be important for future research to not only establish each mechanistic pathway, but to consider how mechanisms interact with one another in determining outcomes. Ideally, researchers should test multiple mechanisms in the same study, which will allow them to assess the interactions among mechanisms and also compare the relative strength of the mechanisms. Third, most of the existing studies that test mechanisms have used cross-sectional or correlational designs, most likely due to the difficulty of experimentally manipulating relationship and sleep experiences. It will be important for future research to employ more rigorous tests of mechanistic pathways that involve experimental manipulations. For example, researchers could randomly assign couples to have conflict conversations or neutral conversations, and then measure how the proposed mediators (e.g. negative affect, perceived stress) change, and then how their sleep quality changes overnight. Reciprocally, researchers could randomly assign participants to conditions of sleep deprivation or normal sleep overnight, and then measure how the proposed mediators change, and subsequently how their romantic relationship experiences are affected the next day. Greater use of ecological momentary assessment designs or prospective longitudinal designs also would enable stronger conclusions regarding directionality.

It will also be important for future studies to assess sleep both subjectively and objectively within the same study. Most existing studies we reviewed used self-report measures to assess sleep quality. However, with the advancement of technology, we can easily track sleep quality via actigraphy [162] or other wearable devices such as the Apple Watch [163] and Fitbit [164]. Obtaining multiple assessments of sleep quality would enable conclusions regarding the alignment between subjective reports and objective measures. Finally, it will be important for future research to recruit more diverse samples to enable generalizability of effects. In the existing studies reviewed, researchers have mainly focused on younger and middle-aged populations, white Americans, and heterosexual couples. There is a strong need for future research to investigate racial and cultural influences to understand if the same conclusions apply broadly.

In conclusion, this review summarizes the current theoretical and empirical evidence in support of five categories of mechanisms underlying the associations between romantic relationship experiences and sleep quality, including emotional/affective responses, self-perceptions, social/relational perceptions, self-regulation, and biological functioning. It also identifies both strengths and limitations of the existing literature and points to profitable directions for future research to fill important gaps. We hope this review will provide a foundational springboard for researchers to conduct needed studies to systematically fill important gaps in the literature. This will be important for informing the development of wise-to-process interventions that target specific mechanisms to improve sleep and relationship quality.

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Disclosure Statement

The authors declare that there is no conflict of interest.

Author Contributions

Yuxi Xie (Conceptualization [equal], Investigation [equal], Writing—original draft [Lead], Writing—review & editing [equal]), and Brooke C. Feeney (Conceptualization [equal], Investigation [equal], Supervision [lead], Writing—review & editing [equal])