Addressing the first-night effect: it is more than the environment

The study of sleep, particularly in laboratory settings, has long grappled with the first-night effect (FNE)—a phenomenon where sleep quality and other sleep parameters are typically poorer during the first night of recording. First recognized in 1964 [1], the FNE has been a critical topic in sleep research due to its potential to bias sleep results. It is primarily characterized by prolonged sleep onset latency, reduced total sleep time and REM sleep time, lower sleep efficiency, and increased awakenings during the initial sleep night compared to subsequent nights [2]. However, the manifestation and magnitude of the FNE have been inconsistent across studies, with variations attributed to factors such as recording settings (i.e. sleep laboratory vs. individuals’ home environment), and participants’ characteristics, including gender and age.

In an article published in this issue of SLEEP, Wick et al. [3] aimed to shed new light on the FNE. In the first experiment, 45 healthy participants spent two weekly separated nights in the sleep laboratory, while in a separate experiment, thirty healthy participants spent two nonconsecutive nights in the sleep laboratory and two nights at home. By employing such an experimental design, the authors were able to demonstrate that the FNE occurs on nonconsecutive nights and that it affects especially sleep duration and sleep continuity parameters. Furthermore, these findings provided evidence that the current practice in sleep research of using adaptation nights several days before the actual experiment is effective in controlling for sleep impairments on the first night, and that adaptation nights remain important in both unfamiliar (e.g. sleep lab) and familiar (e.g. home recording) environments.

Overall, this work offers important, novel insight into the role of FNE in sleep research, but also raises unresolved methodological issues and future areas that should be explored to fully characterize such effects. For example, a caveat in the interpretation of differences between in-lab and at-home recordings was the rigid control of sleep timing imposed by the researchers for the former. Participants were required to arrive at the sleep laboratory between 08:15 and 08:30 pm, with lights out enforced between 10:30 pm and 11:00 pm. This standardized schedule, while convenient for researchers, fails to account for individual differences in circadian rhythms and habitual sleep patterns. For many participants, this imposed bedtime occurs significantly earlier or later than their usual sleep onset, potentially inducing a form of circadian misalignment affecting key parameters involved in FNE, such as sleep onset latency. Furthermore, the study protocol dictated that lights be turned on exactly eight hours after sleep onset, regardless of the participant’s usual wake-up time. This choice likely affected individuals’ sleep cycles, especially for morning types who might naturally wake earlier or evening types who typically sleep later. Such forced awakenings could skew measurements of total sleep time and sleep efficiency, parameters crucial to assessing the FNE. Therefore, future work should apply more flexible protocols that accommodate individual sleep patterns.

Wick et al.‘s study [3] breaks new ground by examining the FNE in home environments and across nonconsecutive nights, addressing a significant gap in the literature. Their finding that FNEs persist in familiar settings and over weekly intervals is noteworthy. However, the requirement for participants to come to the laboratory for polysomnography (PSG) set-up, even for nights when they were sleeping at home, introduces a significant confound. The process of traveling to the lab, undergoing electrode placement, and then returning home could itself induce stress or arousal that may worsen sleep quality and negatively affect the “familiar” environment that the study aims to investigate. Future work should therefore strive to create more ecologically valid study designs. This could involve the use of less intrusive recording technologies that allow for more natural sleep patterns. Actigraphy combined with home-based electroencephalogram (EEG)/PSG recordings, for instance, might provide a better balance between data quality and minimal disruption to normal sleep habits.

Wick et al. showed how the FNE can significantly affect standard sleep metrics, including sleep onset latency and total sleep time [3], but did not examine the microstructure of sleep, including sleep oscillations. Future work should therefore assess possible FNE on sleep spindles and slow-wave activity parameters, thus providing more nuanced insights into sleep quality and adaptation processes. Sleep spindles, for instance, have been associated with sleep stability and protection against external disturbances [4]. Analyzing how spindle density, amplitude, and duration change across nights and environments could offer valuable information about the process of sleep adaptation. Similarly, a detailed examination of slow-wave activity, including its power, slope, and distribution across the night, could reveal subtle changes in sleep depth and recuperative processes that might not be captured by traditional sleep stage scoring.

Wick et al. also investigated interhemispheric asymmetry in slow-wave activity during the FNE, and reported a consistent asymmetry across both nights, with higher slow-wave activity (SWA) in the right frontal hemisphere compared to the left within slow-wave sleep (SWS) of the first sleep cycle throughout the nights. While interesting, these findings are inconsistent with previous research [5, 6]. Tamaki et al. [5] reported left hemisphere reduction in delta-activity during SWS only on the first night, while another recent study [6] found lower SWA in right frontal/prefrontal regions during the first night across entire non-rapid eye movement sleep. Additionally, Wick et al. found no correlation between asymmetry and sleep onset latency, contradicting Tamaki et al.‘s findings [5]. These discrepancies raise intriguing questions about the nature, consistency, and spatial specificity of hemispheric asymmetry in sleep. Wick et al. suggests that the asymmetry may have both trait-like and state-like components, as indicated by moderate intraclass correlation coefficients across nights. One way to further investigate this issue and assess this hypothesis would be to examine not only the presence of asymmetry but also how it evolves over the course of the night and across multiple nights, as well as potential methodological factors that might account for inconsistent findings. For instance, the use of high-density EEG may provide more spatially resolved data to better understand the regional specificity of these effects [6].

Another critical area for future research is the impact of individual differences on the FNE. Age, chronotype, personality traits, and prior sleep history are all factors that can significantly influence how individuals adapt to new sleep environments or recording procedures. Consistent with this assumption, Ding et al. [2] conducted a meta-analysis showing that the FNE varied across different age groups, with young adults showing the least pronounced effects. Large-scale studies that can account for these variables are therefore needed to develop a more comprehensive understanding of the FNE phenomenon.

In conclusion, the study by Wick et al. [3] provides valuable insights into the persistence of the FNE across nonconsecutive nights and in both familiar and unfamiliar environments. Building on these findings, future work should include: (1) employing less intrusive recording technologies, (2) analyzing sleep microstructure elements like spindles and slow-wave activity, (3) exploring interhemispheric asymmetry with high-density EEG, and (4) accounting for individual differences such as age and chronotype. By addressing these issues, future studies can help enhance our understanding of the FNE and its implications for sleep research and clinical practice.

Funding

This research was funded by the National Institute of Mental Health (NIMH), grant number R01MH130376 awarded to Fabio Ferrarelli.

Disclosure Statement

Financial disclosure: The authors have no financial arrangements or connections to report. Nonfinancial disclosure: The authors do not have any conflicts of interest to disclose.