Investigating the effects of partial sleep deprivation on creative ability: An experiment using Bayesian two-way analysis of variance

Investigating the effects of partial sleep deprivation on creative ability: An experiment using Bayesian two-way analysis of variance | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Investigating the effects of partial sleep deprivation on creative ability: An experiment using Bayesian two-way analysis of variance Yuki Motomura, Toshihiro Iwayama, Tamaki Ueda, Mayu Kajihara, and 1 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7562866/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Background This study investigated the impact of partial sleep deprivation on creativity by comparing performance changes between the partial sleep deprivation and normal sleep groups before and after engaging in creative tasks. Methods Participants were divided into two groups: those with partial sleep deprivation and those with normal sleep patterns. Both groups completed creativity assessments, including the Japanese version of the Divergent Association Task (DAT) and Tangram puzzles, before and after the sleep intervention. Results ​The results indicated that despite increased subjective sleepiness and decreased sustained attention in the partial sleep deprivation group, there were no significant differences between the two groups in terms of performance on the DAT and tangram tasks. Bayesian two-way analysis of variance supported the hypothesis that partial sleep deprivation does not adversely affect these aspects of creativity.​ Conclusions These findings suggest that mild partial sleep deprivation may not impair certain creative abilities such as verbal divergent thinking and insight problem solving. However, notably, while creative performance is maintained, other cognitive functions and overall health may be compromised by insufficient sleep. Therefore, individuals, especially those in creative professions, should be cautious of the potentially broader impact of sleep deprivation.​ Sleep loss creative cognition divergent thinking insight Figures Figure 1 Figure 2 Figure 3 Figure 4 Background Many famous creative individuals are said to have sacrificed their sleep for their work. Creative professionals, such as designers, researchers, and artists, are generally expected to experience sleep deprivation. Although no scientific study focusing exclusively on sleep duration among creative professionals could be found, a survey conducted on designers in Japan via social media reported that over half of the respondents slept for less than 6 h, with approximately 23% sleeping for less than 5 h ( https://amix-design.com/asoboad/stat/graph-9003/ ). Some studies have suggested that higher levels of divergent thinking are associated with insomnia symptoms [1] and that highly creative children often experience sleep disorders [2]. However, these findings seem to contradict previous research indicating that sleep deprivation impairs fundamental functions underlying creativity. Many aspects of the relationship between sleep and creativity remain unexplored. Creativity encompasses multiple abilities. This study adopted an approach that focuses on a specific cognitive domain to scientifically capture the highly heterogeneous concept of creativity. Dietrich and Kanso [3] classified creativity into three components: insight, divergent thinking, and artistic creativity. Insight is considered a productive type of thinking that generates new solutions to problems rather than a reproductive form of thought that applies previously experienced solutions [4]. Divergent thinking is the ability to generate multiple solutions to problems with no upper limits or definitive conclusions [5]. Here, we focus on insight and divergent thinking. Sleep deprivation adversely affects cognitive and motor performance. Dinges et al. [6] restricted the sleep duration of participants to 33% less than their habitual sleep time (resulting in an average of 4.95 h of sleep per night) and had them perform task-related exercises. Their results suggested that even a single night’s sleep restriction was associated with a decline in performance. Studies examining the relationship between sleep deprivation and brain function have suggested that the prefrontal cortex, which is involved in arousal, attention, and the ability to make and process appropriate judgments, is particularly vulnerable to sleep deprivation, thereby affecting goal-directed behavior and sustained attention [7,8]. Therefore, a decline in the cognitive abilities underlying creativity due to sleep deprivation is expected to lead to a decrease in creativity. Even one night of total sleep deprivation has been suggested to impair the flexibility aspect of divergent thinking [9]. Fluency, originality, and flexibility scores in the Language Form of the Torrance Tests of Creative Thinking have been shown to deteriorate following partial sleep deprivation [10]. Conversely, Landmann et al. [11], who investigated the effects of sleep on memory consolidation and reorganization, reported that sleep deprivation was beneficial for memory restructuring and solving insight problems. Participants were divided into groups based on their sleep or waking conditions (nighttime sleep, nighttime sleep deprivation, and daytime waking) and performed tasks before and after the respective conditions. The Compound Remote Associates Task, which is a creativity test in which participants are asked to find a word that can be linked to each of three seemingly unrelated words, was implemented. The nighttime sleep deprivation group showed significantly shorter response times to problems that were not solved during the first administration of the task than the night-time sleep and daytime waking groups [11]. Furthermore, Lacaux et al. [12] reported that drowsiness, the state between sleep and wakefulness, may promote creativity. In their experiment, participants solved mathematical insight problems before and after taking a 20-minute break in a dark room. During the break, participants who transitioned to non-REM stage N1 sleep were 2.7 times more likely to detect a hidden problem-solving strategy than those who did not fall asleep. The relationship between sleep deprivation and creativity has been inconsistent across studies, and further investigation is needed. A systematic review by Lim et al. [13] identified eight studies that investigated creative task performance under sleep-deprived conditions. Overall, the review concluded that sleep deprivation tends to impair creative ability. In particular, divergent linguistic thinking appears to be more vulnerable. However, owing to the limited quality and small sample sizes of these studies, the conclusions are far from definitive. Moreover, all studies employed interventions based on total sleep deprivation (i.e., staying awake overnight). In everyday life, people rarely experience a full night of sleep deprivation, but rather often accumulate sleep debts because of shortened sleep. Although Nelson et al. [10] utilized partial sleep deprivation in their study, their protocol involved only a 30-minute nap during a 24-hour wake period, which represents a relatively intense form of partial sleep deprivation. Whether similar effects occur with mild partial sleep deprivation that typically occurs in daily life, remains unknown. In this study, we focused on the aspects within the definitions of insight and divergent thinking that pertain to generating new solutions to a given problem and developing novel ideas under specific conditions, to examine their relationship with partial sleep deprivation. The participants were divided into groups based on their sleep conditions (normal sleep: 8 h; partial sleep deprivation: 2 h), and creativity tasks were administered before and after sleep manipulation. The creativity tasks included a tangram task, a type of insight problem [14], and a Divergent Association Task (DAT) that assesses linguistic divergent thinking. Aims This study investigated the effects of partial sleep deprivation on creative performance, focusing on verbal divergent thinking and insight problem-solving abilities. This study sought to clarify whether a single night of restricted sleep (2 h) significantly impairs these aspects of creativity compared with a full night of sleep (8 h). The following hypotheses were tested: Partial sleep deprivation leads to decreased DAT performance, indicating reduced verbal divergent thinking ability. Partial sleep deprivation results in fewer correct and diverse solutions to tangram-based insight problems, reflecting reduced problem-solving flexibility. Significance The findings of this study offer new insights into the relationship between sleep and creativity under conditions that closely resemble everyday experiences of mild sleep loss. Unlike previous studies that primarily focused on total sleep deprivation, this study contributes to the understanding of how partial sleep deprivation, a more common and ecologically valid form of sleep restriction, affects creative cognition. Method Ethics All participants provided informed consent and the study was conducted in accordance with the Declaration of Helsinki. The study protocol was approved by the Ethics Committee of the Faculty of Design at Kyushu University (approval number 373). Preliminary Survey The study was conducted between October 2020 and October 2021. A total of 46 young male participants (mean age 21.8 ± 2.0 years) from the Faculty of Arts and Design at Kyushu University took part in the preliminary survey. Prior to the experiment, participants were provided with explanations of the experimental procedures and important considerations. Consent for participation was obtained via an online program (because the experiment was conducted during the COVID-19 pandemic). The eligibility criteria were that the participants had no current or past psychiatric or sleep disorders, were not taking any medications that affected their sleep, and were able to adhere to the study schedule. Before the experiment, an email containing a link to the online program and tangram task was sent to each participant. The deadline for completing the task was set within two weeks of receiving the email. Prior to the experiment, participants completed questionnaires regarding their sleep habits and mental health through an online program. On the day of the experiment, between 10:00 a.m. and 12:00 a.m., the participants answered a questionnaire on sleepiness via an online program at home and completed the tangram task. After completing the task, participants were asked to send the results of the creative task via email. As this experiment focused on the impact of everyday sleepiness on creativity, sleep was not controlled. Questionnaires on Preliminary Survey The Pittsburgh Sleep Quality Index (PSQI) is a questionnaire designed to measure sleep health by assessing items such as sleep quality, sleep onset latency, sleep duration, sleep efficiency, sleep disturbances, the use of sleeping medications, and daytime dysfunction [15]. The participants were asked to complete the PSQI between receiving the URL for the online program and performing the tangram task. The K6 is a screening tool for depression and anxiety disorders. Its optimal cutoff score is 13, with a score of 13 or higher indicating potential depression or anxiety disorder [16]. The participants were required to complete the K6 during the period between receiving the URL for the online program and undertaking the tangram task. The Stanford Sleepiness Scale (SSS) is a 7-point subjective measure of sleepiness, with 1 representing the highest level of alertness and 7 representing the lowest [17,18] The SSS scores were used as an index of subjective alertness in this study. The SSS was completed by the participants before performing the tangram task between 10:00 a.m. and 12:00 a.m. The results of these questionnaire were collected for use in the subsequent main experiment. Tangram Task for the Preliminary Survey In this study, a tangram task, which is a type of insight problem, was employed as a creativity task. A tangram is a mathematical puzzle in which seven pieces (triangles and quadrilaterals cut from a square) are arranged to form various shapes. This has been used in previous studies as a measure of insight [14]. The participants were sent an email with an attached file containing the image shown in Figure 1A, and were instructed to print the file before the experiment and cut the figure into seven pieces. During the task, participants were instructed to use all seven pieces shown in Figure 1A to create each tangram, as shown in Figure 1B (a total of four tangrams). A time limit of 600 seconds (10 minutes) was set for each tangram and the time taken to complete each tangram was recorded. Moreover, because one of the purposes of this experiment was to select participants for the second experiment, all participants were instructed to perform the task under the same conditions (solving the tangrams in the order of (a), (b), (c), and (d), as shown in Figure 1B) and to work continuously without taking any breaks or sleeping. The results of the tangram task from the preliminary survey were used to select participants for the main experiment, in which a task was planned to examine the number of different combinations that could be created within the time limit for a single tangram. To reduce learning effects, a task different from the tangram task used in the main experiment was administered during the preliminary survey; however, it was designed such that the participants could approach the tangram in a similar manner. Main Experiment Participants The main experiment included 27 participants (22.6 ± 1.9 years old) selected from the preliminary survey. The PSQI and K6 scores obtained from the preliminary survey were used to screen participants for the main experiment. For participants whose interval between the preliminary survey and the main experiment exceeded one month, the PSQI and K6 were administered again. Furthermore, to equalize the participants’ ability to solve the tangram task, individuals were sequentially assigned to two groups (sleep and partial sleep deprivation), starting with those whose average tangram rank from the preliminary survey was closest to the overall mean. A power analysis conducted using G*Power (https://www.psychologie.hhu.de/arbeitsgruppen/allgemeine-psychologie-und-arbeitspsychologie/gpower) set the target sample size at 28 participants. Ultimately, 27 participants volunteered and were divided into two groups, such that variations in age, tangram-solving ability (measured by the average rank in the preliminary survey), and K6 scores were balanced. Additionally, efforts were made to distribute university majors (e.g., ergonomics, industrial design, acoustic design, art, and information design) as evenly as possible between the groups. The sleep group comprised 13 participants (22.4 ± 1.8 years old), and the partial sleep deprivation group included 14 participants (22.8 ± 1.9 years old). The average tangram rank was 18.2 ± 3.2 for the sleep group and 17.6 ± 3.7 for the partial sleep deprivation group, and the average K6 scores were 5.3 ± 2.7 and 4.9 ± 2.9, respectively. Experimental Procedure For the two days preceding the experiment, the participants were instructed to sleep for 8 h, with their bedtime set 4 h before and their wake-up time 4 h after the individual sleep midpoint, as determined by the PSQI. Additionally, for any participant whose bedtime would otherwise fall later than 1:00 a.m., sleep was regulated so that they went to bed at 1:00 a.m. and woke up at 9:00 a.m. This was applied to seven participants in the sleep group and three in the partial sleep deprivation group. Moreover, during the two-day sleep regulation period prior to the experiment, participants were prohibited from taking naps and heavy drinking. On the day before the experiment, all participants underwent a polymerase chain reaction (PCR) test as a preventive measure against COVID-19. Only patients who tested negative on the first day of the experiment were included. The experiment was conducted over three consecutive days (Figure 2A). During the experimental period, participants were not permitted to consume tobacco, caffeine, or alcohol, or engage in vigorous exercise, exciting activities, puzzle games, or napping. However, to minimize stress, activities that did not fall under these prohibitions, such as working on class assignments, reading, playing games, using the internet, or conversing with experimenters or other participants, were allowed at the discretion of the test conductors. During non-task periods, the participants stayed in the Human Factors Laboratory on the fourth floor of Building 1 at Kyushu University’s Ohashi Campus, where the room temperature was maintained at 26 ± 2°C. Although the basic set temperature was 26°C, the room temperature was increased if any participant complained of feeling cold. On the first day of the experiment, the participants were required to arrive at the experimental room at least one hour before their scheduled bedtime during the sleep regulation period. In the experimental room, after being fitted with the wearable device Fitbit Inspire2 (Fitbit, Inc.), the participants were instructed to sleep for 8 h under the same bedtime and wake-up times as those established during the sleep regulation period. On the second day, 1.5 h after their wake-up time, electrodes for electroencephalography (EEG) were attached, and the participants completed a series of tasks: an eyes-open resting period, the Psychomotor Vigilance Test (PVT), the DAT, and the tangram task (using shapes labeled Heart and Square [shapes1]). EEG measurements were conducted in a soundproof and radiofrequency-shielded room on the fourth floor of Building 1 at the campus. During the task performance, the participants wore short-sleeved shirts and shorts. The room conditions were strictly controlled: temperature at 26°C, humidity at 50%, and desk illuminance between 550 and 650 lux. On the night of the second day, after donning the wearable device, the participants were allowed either 8 (sleep group) or 2 (partial sleep deprivation group) h of sleep, and the partial sleep deprivation group was required to go to bed 2 h before their scheduled wake-up time. On the third day, 1.5 h after the wake-up time, EEG electrodes were attached again, and the participants performed the following tasks: an eyes-open resting period, the PVT, the DAT, and the tangram task (using shapes labeled as Heart and Square [shapes1] and Bird and Pot [shapes2]) (Figure 2B). On each experimental day, a maximum of two participants were tested simultaneously. In cases where two participants were scheduled concurrently, the second participant’s session commenced immediately after the first participant completed their session; consequently, the start time for the second participant was approximately 2.5–3 h after the wake-up time. In total, 15 participants (7 in the sleep group and 8 in the partial sleep deprivation group) participated in the first session and 12 participants (6 in each group) participated in the second session. However, the EEG results are not reported here. Questionnaires The State-Trait Anxiety Inventory (STAI) [19] is a 20-item questionnaire designed to measure state anxiety. To assess subjective sleepiness and mood, participants were instructed to complete the SSS and STAI via an online program immediately after the start of the experimental session. Psychomotor Vigilance Task The PVT was administered to objectively assess sleepiness. It measures sustained attention and is an objective indicator of decreased alertness [20]. In this study, a five‑minute PVT was conducted on a tablet device (FFF SMART LIFE CONNECTED, FFF‑TAB7) using the open-source software Vigilance Buddy 1.53 (https://researchbuddies.com/). In this task, a stimulus displayed as a video of incrementing numbers was presented on the screen and the participants were instructed to tap the screen as quickly as possible upon noticing the stimulus. The reaction time from the onset of the stimulus to the screen tap was measured. The primary outcome measure analyzed was the reciprocal of the mean reaction time, with a lower value indicating a decline in sustained attention [21]. Creativity Tasks Divergent Association Task The DAT was used to measure linguistic divergent thinking. It requires participants to list ten nouns that are as different as possible in terms of meaning or use within a four-minute period [22]. It was possible to objectively quantify divergent thinking abilities by calculating the semantic distances between these nouns. The DAT performance is correlated with other divergent thinking tasks (e.g., the Alternative Uses Task). In this study, a Japanese version of the DAT was developed based on an algorithm published by Olson et al. [22] (https://osf.io/bm5fd/) and was employed as the task. The DAT scores were calculated using Google Colaboratory as the Jupyter notebook execution environment and a pre-trained Japanese word vector file (.gz) was obtained online (https://fasttext.cc/docs/en/crawl-vectors.html). Following the scoring method of the English version, the first 7 valid responses out of the 10 nouns provided by the participants were used to compute 21 pairwise semantic distances. The DAT score was defined as 100 times the average semantic distance. Notably, a significant correlation was observed between the scores of the original and Japanese versions ( r = 0.50, p < 0.001), confirming that the Japanese DAT score is a useful index of creativity. For this task, the participants were given an A4 sheet that included task instructions, precautions, and an answer section, and were asked to write their responses using a ballpoint pen. Although the time limit for the task was four minutes, the task was concluded once the participant recorded 10 nouns. Furthermore, as the DAT was administered twice (either before and after sleep or before and after sleep deprivation) on the third day, participants were required to provide responses using a different set of nouns from those used on the second day. Tangram Task The tangram task was used as a creativity task to measure insight. The tangrams used in the second experiment were modeled using the 3D CAD software “Fusion 360” (.stl files). From these models, 3D printer files (.gx files) were created using the slicing software “FlashPrint.” Subsequently, using a 3D printer (“FLASHFORGE Adventurer3”), 22 sets of squares—each composed of 7 pieces and measuring 100 mm (height) × 100 mm (width) × 10 mm (thickness)—were produced. In the tangram task, participants were required to determine the number of different combinations that could be created for a given silhouette. Each silhouette was allotted a time limit of 7 minutes. Two types of comparisons were performed for the tangram task. In this experiment, on the second and third days, the participants solved the same silhouette (shapes1) to examine the incubation effect. Additionally, by having the participants solve shapes1 on the second day and a different silhouette (shapes2) on the third day, the impact of different sleep conditions on their responses to a new creativity task was investigated. Considering that individual differences in proficiency with a single silhouette might affect performance, the experiment used two silhouettes on the second day (Heart and Square: shapes1); on the third day, we used the same silhouettes (Heart and Square: shapes1) and two new silhouettes (Bird and Pot: shapes2). This design was used to average the responses for each pair of silhouettes, thereby neutralizing the effects of individual strengths or weaknesses. Given that the Heart and Pot shapes were bilaterally symmetrical, creating one combination might inadvertently lead to the formation of a symmetrical duplicate. Therefore, the participants were instructed not to produce bilaterally symmetrical combinations. In this experiment, the tangram task was performed using the 3D-printed tangrams. To account for order effects, the experiment was conducted in four different orders: two orders on the second day (Square → Heart or Heart → Square) and two orders on the third day (Bird → Heart → Pot → Square or Square → Pot → Heart → Bird), with counterbalancing among the participants. Statistical Analysis Statistical analyses were performed using R version 3.6.1 and JASP version 0.19.3. Classical statistical hypothesis testing based on p -values alone cannot adequately support the hypothesis that there is no difference between groups. Therefore, a combined approach was employed that reported both Bayesian hypothesis testing using Bayes factors (BFs), as recommended by Keysers et al. [23], and conventional p -values. Bayesian hypothesis testing calculates the subjective probability that a hypothesis is correct, thereby supporting the hypothesis that there is no difference between groups. First, to confirm that there were no differences in demographic data between the groups, Bayesian hypothesis tests were conducted on the K6 scores, tangram response rankings from the preliminary survey, and age. In this study, creativity tasks and questionnaires were administered before and after sleep in both sleep and partial sleep deprivation groups. If the changes from pre- to post-sleep differed between the groups, it was regarded as evidence of a sleep effect. Accordingly, statistical methods were used to test for interactions between the groups (two levels: sleep group vs. partial sleep deprivation group) and date (two levels: Day 2 vs. Day 3). Two separate analyses were performed for the tangram task, corresponding to the following research questions: Examining changes in performance on the same task (i.e., comparing performance on the same tangram task—Heart and Square shapes—on Days 2 and 3). Examining the generalization of performance to a different task (i.e., comparing performance on different tangram tasks: on Day 2, Heart and Square shapes and on Day 3, Bird and Pot shapes). In each analysis, the mean number of responses was used as the outcome measure. For the STAI, SSS, PVT, DAT, and the number of responses in the tangram tasks with the same silhouette and those with different silhouettes, a mixed-design two-way analysis of variance (ANOVA) (group [two levels: sleep group, partial sleep deprivation group] × date [two levels: Day 2, Day 3]) was conducted. Both BFs and p -values were calculated. When the assumption of sphericity was violated, the Greenhouse–Geisser correction was applied to adjust the degrees of freedom. Post-hoc tests were conducted using two-tailed t -tests with adjustments for multiple comparisons using the modified sequentially rejective Bonferroni method. One participant in the partial sleep deprivation group who did not complete the SSS was excluded from the analysis, resulting in 13 participants each in the sleep and partial sleep deprivation groups. Additionally, one participant in the sleep group whose PVT data could not be recorded was excluded, resulting in data from 12 and 14 participants in the sleep and partial sleep deprivation groups, respectively, being analyzed. In this study, following Jeffreys’ [24] criteria, BF was interpreted as follows: A BF between 1 and 3 indicates little anecdotal evidence in favor of the alternative hypothesis. A BF value between 3 and 10 indicates moderate support for the alternative hypothesis. A BF of 10 or greater indicates strong support for the alternative hypothesis. Conversely, a BF of less than one indicates support for the null hypothesis. Specifically, a BF between 0.1 and approximately 0.333 indicates moderate support for the null hypothesis, and a BF of 0.1 or below is interpreted as strong evidence in favor of the null hypothesis. These criteria were used to evaluate the strength of evidence for the hypotheses in each statistical test. Results Demographic Data Demographic data are presented in Table 1. A Bayesian t -test was conducted on age, K6 scores, PSQI scores, and average tangram rankings between the sleep and partial sleep deprivation groups. The resulting BFs ranged from 0.358 to 0.4, providing modest support for the hypothesis that there were no differences between the groups (age: BF = 0.4; K6: BF = 0.38; PSQI: BF = 0.358; and tangram: BF = 0.382). Table 1. Demographic data Condition Age Rank of tangram task K6 (depressive symptoms) PSQI (sleep quality) Mean SD Mean SD Mean SD Mean SD Sleep control 22.385 1.968 18.179 3.874 5.308 2.983 5.077 2.303 Sleep deprivation 22.786 1.850 17.607 3.279 4.857 2.840 5.071 1.706 Abbreviation: SD = standard deviation, PSQI = Pittsburgh sleep quality index State Anxiety For the STAI results, a two-way ANOVA (group [two levels] × date [two levels]) was performed (Figure 3A). The analysis revealed no significant main effects of group ( F (1, 25) = 0.28, p = 0.600) or date ( F (1, 25) = 1.71, p = 0.202). Moreover, no significant group × date interaction was observed ( F (1, 25) = 0.15, p = 0.700). The Bayesian two-way ANOVA results supported the hypothesis of no group effect (BF = 0.314), provided modest support for the hypothesis of no time effect (BF = 0.434), and strongly supported the hypothesis of no group × time interaction (BF = 0.163). Subjective Sleepiness A two-way ANOVA (group [two levels] × date [two levels]) was conducted on the SSS results (Figure 3B). The analysis revealed a significant main effect of group ( F (1, 24) = 14.9, p < 0.001), with the partial sleep deprivation group exhibiting higher SSS scores than the sleep group. A significant main effect of date was also observed ( F (1, 24) = 12.7, p = 0.002), with SSS scores being higher on Day 3 than on Day 2. Furthermore, a significant group × date interaction was observed ( F (1, 24) = 14.9, p < 0.001). Post-hoc tests indicated that on Day 3, the partial sleep deprivation group had significantly higher SSS scores ( F (1, 24) = 23.6, p < 0.001), and within the partial sleep deprivation group, the SSS scores on Day 3 were significantly higher ( F (1, 12) = 18.7, p = 0.001). The Bayesian two-way ANOVA results supported the hypotheses of a group effect (BF = 3.612), a time effect (BF = 10.583), and no group × time interaction (BF = 9.263). Objective Sleepiness For the PVT results, a two-way ANOVA (group [two levels] × date [two levels]) was conducted (Figure 3C). Although there was a trend toward a group effect ( F (1, 24) = 2.94, p = 0.099), no significant main effect of date was observed ( F (1, 24) = 1.44, p = 0.242). However, a significant group × date interaction was observed ( F (1, 24) = 9.75, p = 0.005). Post-hoc tests revealed that on Day 3, the partial sleep deprivation group had a significantly lower reciprocal of mean reaction time ( F (1, 24) = 5.50, p = 0.028), indicating poor sustained attention, whereas in the sleep group, the reciprocal of mean reaction time was significantly higher on Day 3 ( F (1, 11) = 11.2, p = 0.007). Additionally, in the partial sleep deprivation group, the reciprocal of the mean reaction time was significantly lower on Day 3 ( F (1, 13) = 5.89, p = 0.031). The Bayesian two-way ANOVA results supported the hypotheses of a group effect (BF = 3.357) and group × time interaction (BF = 7.128) but provided inconclusive evidence of a time effect (BF = 2.331). Divergent Association Task A two-way ANOVA (group [two levels] × date [two levels]) was conducted on the DAT results (Figure 4A). The analysis revealed no significant main effects of group ( F (1, 25) = 0.26, p = 0.611) or date ( F (1, 25) = 0.46, p = 0.502). Furthermore, no significant group × date interaction was observed ( F (1, 25) = 1.50, p = 0.231). The Bayesian two-way ANOVA results provided modest support for the hypotheses of no group effect (BF = 0.350), no time effect (BF = 0.266), and no group × time interaction (BF = 0.197). Tangram Task (Same Shapes on Days 2 and 3) For the tangram task results using the same silhouette on both days (Figure 4B), a two-way ANOVA (group [two levels] × date [two levels]) was performed. The analysis showed no significant main effect of group ( F (1, 24) = 0.04, p = 0.849); however, a significant main effect of date was observed ( F (1, 24) = 23.2, p < 0.001). No significant group × date interaction was observed ( F (1, 24) = 0.27, p = 0.608). The Bayesian two-way ANOVA results provided marginal support for the hypothesis of no group effect (BF = 0.334), very strong support for the hypothesis of a time effect (BF = 745.720), and inconclusive evidence regarding the group × time interaction (BF = 0.406). Tangram Task (Different Shapes on Days 2 and 3) For the tangram task results using different silhouettes on Days 2 and 3 (Figure 4C), a two-way ANOVA (group [two levels] × date [two levels]) was conducted. The analysis revealed no significant main effect of group ( F (1, 24) = 7.00 × 10⁻⁴, p = 0.979) and a significant main effect of date ( F (1, 24) = 96.0, p < 0.001). No significant group × date interaction was observed ( F (1, 24) = 0.04, p = 0.839). The Bayesian two-way ANOVA results provided modest support for the hypothesis of no group effect (BF = 0.352), very strong support for the hypothesis of a time effect (BF = 8.905 × 10⁹), and marginal support for the hypothesis of no group × time interaction (BF = 0.382). Discussion In this study, we examined the effects of sleep deprivation on creativity by administering creativity tasks before and after a partial sleep deprivation simulation trial. Furthermore, we compared the changes in performance between the partial sleep deprivation and normal sleep groups. The following discussion considers these findings. First, regarding the participant groups, we confirmed that there were no significant differences in the demographic data between the partial sleep deprivation and normal sleep groups. Based on the measures of age, tangram-solving ability, depression, and anxiety, no group differences were observed. Variance analysis using BFs provided modest support for the hypothesis that there were no differences between groups. Consequently, the sleep deprivation and partial sleep deprivation groups were effectively matched in terms of age, tangram-solving ability, and depression or anxiety. Second, the results regarding state anxiety, subjective sleepiness, and objective sleepiness are discussed. Regarding state anxiety, as measured by the STAI, neither a significant main effect nor a significant group × time interaction was observed; the Bayesian two-way ANOVA strongly supported the absence of an interaction between group and time. This indicates that the sleep conditions employed in the study did not affect the participants’ anxiety levels. Although previous studies have reported that multiple days of partial sleep deprivation are associated with increased STAI scores [25,26] this study’s use of a single day of partial sleep deprivation, which is a milder intervention, may account for this discrepancy. Conversely, the SSS showed a significant group × date interaction, with the partial sleep deprivation group exhibiting a marked increase in subjective sleepiness relative to the normal sleep group. In addition, the PVT results revealed a significant group × date interaction. Specifically, the partial sleep deprivation group demonstrated a decline in sustained attention following the intervention, as indicated by a lower reciprocal of the mean reaction time compared to the normal sleep group. These findings suggest that, although sleep manipulation did not alter state anxiety, it effectively increased the subjective and objective measures of sleepiness. To examine the effect of partial sleep deprivation on linguistic divergent thinking, we investigated whether there was a group × time interaction in DAT performance. Regarding the DAT results, no significant main effects of group or time were observed and there was no significant group × time interaction. The Bayesian two-way ANOVA supported the hypothesis that there was no interaction. These findings suggest that neither short sleep duration (2 h) nor normal sleep (2 h) differentially affects linguistic divergent thinking. This is the first study to demonstrate the impact of mild partial sleep deprivation, which is typically experienced in daily life, on creativity. A review of previous studies reported that, compared with performance on visuospatial tasks, linguistic divergent thinking is more adversely affected by sleep deprivation [13]; therefore, the absence of a difference in our study was unexpected. This discrepancy may be due to differences in the forms of sleep deprivation. All the reviewed studies considered total sleep deprivation (i.e., staying awake overnight). While partial sleep deprivation tends to preserve deep and rapid eye movement (REM) sleep, both of which are considered important for recovery because of sleep homeostasis, total sleep deprivation results in a loss in all sleep stages [27]. Our results suggested that the preservation of deep and REM sleep may have contributed to the recovery of divergent thinking. Although the intervention in this study was relatively mild compared with previous research, both subjective sleepiness and objective measures of sustained attention deteriorated significantly, but creative performance was maintained despite a decline in other aspects of cognitive functioning. Finally, we discussed the results of the tangram task. In this study, we addressed two research questions: (1) changes in performance on the same task, and (2) the generalization of performance to a different task. For Research Question 1, we compared the performance on the same tangram task between Days 2 and 3, and for Research Question 2, we compared the performance on different tangram tasks between Days 2 and 3. Regarding Research Question 1, no significant group × time interaction or main effect of group was found; however, a significant main effect of time was observed. Bayesian two-way ANOVA provided modest support for the hypothesis of no group effect, strong support for a time effect, and inconclusive evidence regarding a group × time interaction. These results indicate that it is difficult to attribute differences in the ability to repeatedly solve the same tangram task to the effects of sleep and partial sleep deprivation. However, the significant main effect of time, strongly supported by the Bayesian analysis, suggests that performance improved when the same tangram task was repeated after sleep or partial sleep deprivation. This improvement was likely attributable to the learning effect of solving the same task twice. Another possibility is that sleep deprivation shortens response times for problems that have previously been solved [11]. Thus, the learning effect may have been offset by the detrimental effects of partial sleep deprivation on the overall performance and learning, leading to a lack of detectable group differences. However, the current results do not allow a detailed examination of this possibility. Regarding Research Question 2, similar to Research Question 1, when comparing the performance on different tangram tasks between Days 2 and 3, no significant group × time interaction or main effect of group was observed; however, a significant main effect of time was evident. The Bayesian two-way ANOVA provided modest support for the hypothesis of no group effect, strong support for a time effect, and modest support for a group × time interaction. These findings also suggest a learning effect, with the improvement being generalized to different tangram tasks. In addition, the results provide modest support for the hypothesis that partial sleep deprivation does not influence this change. It is conceivable that participants were better able to combine elements from their memory of the shapes solved on Day 2 (shapes1) when tackling the same type of task (shapes2) on Day 3, leading to an increased number of combinations. Alternatively, differences in the task difficulty may have contributed to this finding. The difficulty of the “shapes2” task could have been inherently lower than that of the “shapes1” task for the participants, irrespective of any learning or sleep effects. Limitations The study had several limitations. First, although creativity tasks were administered twice–before and after sleep, it was not possible to completely eliminate the learning effects or control for task difficulty across days. Despite the two tasks being designed to measure the same aspects of creativity, it is inherently difficult to create tasks that are both difficult and free from learning effects when measuring creative performance. Second, in the tangram task, it was difficult to objectively determine whether the participants experienced a fixation or impasse, which is a key component of insight problem solving. If participants had not experienced an impasse, the task might not have functioned effectively as an insight problem. However, if the task difficulty increased significantly, most participants would likely remain stuck and unable to find solutions. This highlights the importance of carefully adjusting the task difficulty when using insight problems to measure creativity in an experimental setting. Furthermore, whether one experiences an impasse in a particular insight problem depends heavily on individual traits or prior experience, making it particularly challenging to use insight-based tasks uniformly across participants. A control condition involving a waking interval was necessary to separate the effects of learning from those of sleep. However, implementing a waking condition of an equivalent length would result in total sleep deprivation, making it difficult to isolate the learning effects. Another limitation is that the DAT used was a Japanese version for which translation and cross-linguistic validity have not been formally verified. Finally, the generalizability of the results is limited because the study included only male participants. Conclusion Despite the partial sleep deprivation group reporting greater subjective sleepiness and showing a decline in sustained attention compared with the sleep group, no group differences were observed in the linguistic divergent thinking task. Although BF did not meet Jeffreys’ [24] threshold for strong evidence, the results of the tangram task tended to support the absence of a sleep effect. These findings do not support those of previous studies that reported a decline in creative ability due to sleep deprivation but are consistent with the findings suggesting that performance on complex tasks may be more resilient to the effects of sleep loss. The brain may maintain a certain level of creative function under sleep-deprived conditions, similar to that when fully rested [28,29]. Furthermore, Sasmita et al. [30] suggested that the attention function could be preserved during sleep deprivation when a reward is involved. Since creativity tasks often involve moments of insight and problem solving, they may engage the brain’s reward system, which could help maintain performance despite declines in subjective and objective alertness. Given that the DAT has been reported to be more enjoyable than other creativity tasks [22], it may be the most robust against the effects of partial sleep deprivation. Our results suggested that mild sleep deprivation does not impair linguistic idea generation or performance on insight-based problems. This could partly explain why individuals in creative professions who are often sleep deprived may still function at a highly creative level. However, continuing to work while sleep deprived may come at the cost of other cognitive functions and health issues. Additionally, previous studies have shown that individuals with high creativity are more prone to insomnia and sleep disturbances [1,2] Thus, this study provides valuable evidence that highlights the need to raise awareness about the importance of maintaining sleep health among creative professionals. Abbreviations ANOVA: analysis of variance BFs: Bayes factors EEG: Electroencephalography DAT: Divergent Association Task PCR: Polymerase chain reaction PSQI: Pittsburgh Sleep Quality Index PVT: Psychomotor Vigilance Test REM: rapid eye movement SSS: Stanford Sleepiness Scale STAI: State-Trait Anxiety Inventory Declarations Ethics approval and consent to participate All participants provided informed consent and the study was conducted in accordance with the Declaration of Helsinki. The study protocol was approved by the Ethics Committee of the Faculty of Design at Kyushu University (approval number 373). Availability of data and materials The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request. Competing interests The authors declare that they have no competing interests. Funding This work was supported by KAKENHI(19K16244). Authors' contributions YM and TI analysed and interpreted the data and were major contributors of writing the manuscript. TI, TU, MK and KM performed the sleep experiment. All authors read and approved the final manuscript. Acknowledgments We would like to acknowledge the participants and staff involved in the experiment. References Beaty RE, Silvia PJ, Nusbaum EC, Vartanian O. Tired minds, tired ideas? Exploring insomnia and creativity. Thinking Skills Creativity. 2013;9:69–75. Healey D, Runco MA. Could creativity be associated with insomnia? Creativity Res J. 2006;18:39–43. Dietrich A, Kanso R. A review of EEG, ERP, and neuroimaging studies of creativity and insight. Psychol Bull. 2010;136:822–48. Mayer RE. Thinking, problem solving, cognition. W H Freeman/Times Books/Henry Holt and company; 1992. Guilford JP. The nature of human intelligence. New York: McGraw-Hill; 1967. Dinges DF, Pack F, Williams K, Gillen KA, Powell JW, Ott GE, et al. Cumulative sleepiness, mood disturbance, and psychomotor vigilance performance decrements during a week of sleep restricted to 4–5 h per night. Sleep. 1997;20:267–77. Boddez Y, Buabang EK, Zenses AK, Descheemaeker M. Commentary: Sleep deprivation promotes habitual control over goal-directed control: Behavioral and neuroimaging evidence. Front Behav Neurosci. 2018;12:82. Borragán G, Guerrero-Mosquera C, Guillaume C, Slama H, Peigneux P. Decreased prefrontal connectivity parallels cognitive fatigue-related performance decline after sleep deprivation. An optical imaging study. Biol Psychol. 2019;144:115–24. Wimmer F, Hoffmann RF, Bonato RA, Moffitt AR. The effects of sleep deprivation on divergent thinking and attention processes. J Sleep Res. 1992;1:223–30. Nelson CS, Dell’Angela K, Jellish WS, Brown IE, Skaredoff M. ‘Residents’ performance before and after night call as evaluated by an indicator of creative thought. J Am Osteopath Assoc. 1995;95:600–3. Landmann N, Kuhn M, Maier JG, Feige B, Spiegelhalder K, Riemann D, et al. Sleep strengthens but does not reorganize memory traces in a verbal creativity task. Sleep. 2016;39:705–13. Lacaux C, Andrillon T, Bastoul C, Idir Y, Fonteix-Galet A, Arnulf I, et al. Sleep onset is a creative sweet spot. Sci Adv. 2021;7:eabj5866. Lim AR, Williams BJ, Bullock B. The effect of sleep deprivation on creative cognition: A systematic review of experiment-based research. Creativity Res J. 2024:1–11. Takeuchi N, Mori T, Suzukamo Y, Izumi SI. Activity of prefrontal cortex in teachers and students during teaching of an insight problem. Mind Brain Educ. 2019;13:167–75. Buysse DJ, Reynolds CF, Monk TH, Berman SR, Kupfer DJ. The Pittsburgh Sleep Quality Index: A new instrument for psychiatric practice and research. Psychiatry Res. 1989;28:193–213. Furukawa TA, Kawakami N, Saitoh M, Ono Y, Nakane Y, Nakamura Y, et al. The performance of the Japanese version of the K6 and K10 in the World Mental Health Survey Japan. Int J Methods Psychiatr Res. 2008;17:152–8. Hoddes E, Zarcone V, Smythe H, Phillips R, Dement WC. Quantification of sleepiness: A new approach. Psychophysiology. 1973;10:431–6. Maclean AW, Fekken GC, Saskin P, Knowles JB. Psychometric evaluation of the Stanford sleepiness scale. J Sleep Res. 1992;1:35–9. Spielberger CD, Gorsuch RL, Lushene RE.. STAI manual for the State-Trait Anxiety Inventory (“Self-Evaluation Questionnaire”). Consulting Psychologists Press; 1970. https://ci.nii.ac.jp/ncid/BA66835301. Accessed 27 March 2025. Basner M, Mollicone D, Dinges DF. Validity and sensitivity of a brief psychomotor vigilance test (PVT-B) to total and partial sleep deprivation. Acta Astronaut. 2011;69:949–59. Basner M, Dinges DF. Maximizing sensitivity of the psychomotor vigilance test (PVT) to sleep loss. Sleep. 2011;34:581–91. Olson JA, Nahas J, Chmoulevitch D, Cropper SJ, Webb ME. Naming unrelated words predicts creativity. Proc Natl Acad Sci U S A. 2021;118:e2022340118. Keysers C, Gazzola V, Wagenmakers EJ. Using Bayes factor hypothesis testing in neuroscience to establish evidence of absence. Nat Neurosci. 2020;23:788–99. Jeffreys H. Theory of probability; 1961. https://global.oup.com/academic/product/the-theory-of-probability-9780198503682. Accessed 21 March 2025. Motomura Y, Kitamura S, Oba K, Terasawa Y, Enomoto M, Katayose Y, et al. Sleep debt elicits negative emotional reaction through diminished amygdala–anterior cingulate functional connectivity. PLOS One. 2013;8:e56578. Motomura Y, Katsunuma R, Yoshimura M, Mishima K. ‘Two days’ sleep debt causes mood decline during resting state via diminished amygdala–prefrontal connectivity. Sleep. 2017;40:zsx133. Banks S, Dinges DF. Behavioral and physiological consequences of sleep restriction. J Clin Sleep Med. 2007;3:519–28. Harrison Y, Horne JA. The impact of sleep deprivation on decision making: A review. J Exp Psychol Appl. 2000;6:236–49. Lim J, Dinges DF. A meta-analysis of the impact of short-term sleep deprivation on cognitive variables. Psychol Bull. 2010;136:375–89. Sasmita K, Massar SAA, Lim J, Chee MWL. Reward motivation normalises temporal attention after sleep deprivation. J Sleep Res. 2019;28:e12796. Additional Declarations No competing interests reported. Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7562866","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":523148865,"identity":"e228b2e3-d4dc-495e-9c55-fabb1424d518","order_by":0,"name":"Yuki Motomura","email":"data:image/png;base64,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","orcid":"","institution":"Kyushu University","correspondingAuthor":true,"prefix":"","firstName":"Yuki","middleName":"","lastName":"Motomura","suffix":""},{"id":523148866,"identity":"9dcaaff8-0451-46be-95d8-aabd2f3d2134","order_by":1,"name":"Toshihiro Iwayama","email":"","orcid":"","institution":"Kyushu University Graduate School of Integrative Frontier Science","correspondingAuthor":false,"prefix":"","firstName":"Toshihiro","middleName":"","lastName":"Iwayama","suffix":""},{"id":523148867,"identity":"b5e0e36e-0229-4289-980f-e438d81d8caf","order_by":2,"name":"Tamaki Ueda","email":"","orcid":"","institution":"Railway Technical Research Institute","correspondingAuthor":false,"prefix":"","firstName":"Tamaki","middleName":"","lastName":"Ueda","suffix":""},{"id":523148868,"identity":"bfc9d6b6-6b46-4081-9b24-d468a10b66fe","order_by":3,"name":"Mayu Kajihara","email":"","orcid":"","institution":"Kyushu University Graduate School of Integrative Frontier Science","correspondingAuthor":false,"prefix":"","firstName":"Mayu","middleName":"","lastName":"Kajihara","suffix":""},{"id":523148869,"identity":"81a3b80f-fb23-4373-bc44-17dc4c5689b9","order_by":4,"name":"Takayuki Momoi","email":"","orcid":"","institution":"Kyushu University Graduate School of Integrative Frontier Science","correspondingAuthor":false,"prefix":"","firstName":"Takayuki","middleName":"","lastName":"Momoi","suffix":""}],"badges":[],"createdAt":"2025-09-08 10:08:19","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7562866/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7562866/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":93335721,"identity":"2bc09b39-346d-478b-b8fa-c5300711bbb4","added_by":"auto","created_at":"2025-10-12 14:00:38","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":54745,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript0916.docx","url":"https://assets-eu.researchsquare.com/files/rs-7562866/v1/c7159521504a8247120a520e.docx"},{"id":93335724,"identity":"303e72a7-18c3-4a6f-a95c-db5d4c76a6cf","added_by":"auto","created_at":"2025-10-12 14:00:38","extension":"json","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":6631,"visible":true,"origin":"","legend":"","description":"","filename":"41f6aa06d6584283b3afb44f9c774ac3.json","url":"https://assets-eu.researchsquare.com/files/rs-7562866/v1/dac78e492817f0e07c4b95bd.json"},{"id":93335723,"identity":"ebc5435f-a40b-43d2-99a9-2be0dd95d34d","added_by":"auto","created_at":"2025-10-12 14:00:38","extension":"xml","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":82500,"visible":true,"origin":"","legend":"","description":"","filename":"41f6aa06d6584283b3afb44f9c774ac31enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-7562866/v1/23e89a300db98522c1215b77.xml"},{"id":93337276,"identity":"6d1c7df5-2169-4b84-ab9a-9bcc214da760","added_by":"auto","created_at":"2025-10-12 14:08:38","extension":"pptx","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":154498,"visible":true,"origin":"","legend":"","description":"","filename":"Figures.pptx","url":"https://assets-eu.researchsquare.com/files/rs-7562866/v1/be7036c7a146c6ab9616e7b6.pptx"},{"id":93337275,"identity":"069be82d-e175-4819-874d-3f0054679a80","added_by":"auto","created_at":"2025-10-12 14:08:38","extension":"xml","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":81004,"visible":true,"origin":"","legend":"","description":"","filename":"41f6aa06d6584283b3afb44f9c774ac31structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-7562866/v1/704501cddf2231b6b63eea90.xml"},{"id":93335726,"identity":"3fc81a8f-d7ec-4ff2-9776-fe54636130b6","added_by":"auto","created_at":"2025-10-12 14:00:38","extension":"html","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":91785,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-7562866/v1/3c68c25bc0934c32825328be.html"},{"id":93337274,"identity":"11f9959b-1d40-4e2b-87d7-ac743966322a","added_by":"auto","created_at":"2025-10-12 14:08:38","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":64286,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eTangram task for the preliminary survey\u003c/strong\u003e\u003cbr\u003e\nTangram materials and sample tasks used in the preliminary survey. A) Standard set of seven geometric pieces derived from a square, used for all tangram tasks. B) a–d represent the specific tangram problems presented to participants during the preliminary phase, which required them to reconstruct each shape using all seven pieces. These tasks were used to assess baseline insight problem-solving ability.\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-7562866/v1/6505e27d34b92e9ea2f4e223.png"},{"id":93335718,"identity":"c6397b1d-f699-4dbf-aebb-846756f8d880","added_by":"auto","created_at":"2025-10-12 14:00:38","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":91621,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eExperimental protocol\u003c/strong\u003e\u003cbr\u003e\nA) The timeline of the three-day experimental protocol for the sleep (n = 13) and partial sleep deprivation (n = 14) groups. The participants completed a sequence of tasks each morning, including filling the questionnaire, the EEG setup, eyes-open resting state, the Psychomotor Vigilance Task (PVT), the Divergent Association Task (DAT), and the tangram-based insight tasks.\u003c/p\u003e\n\u003cp\u003eB) Tangram task conditions: on Day 2, the participants solved two silhouettes from “shapes1” (Heart and Square), and on Day 3, they either repeated the same two shapes (shapes1) to assess the incubation effects or solved for two new shapes (shapes2: Pot and Bird) to examine transfer and generalization.\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-7562866/v1/b04a71cbc903990c18c8ef92.png"},{"id":93335720,"identity":"01768fc7-da26-405e-ad28-d1f1e1e2c4c3","added_by":"auto","created_at":"2025-10-12 14:00:38","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":41792,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSubjective mood and Subjective or objective sleepiness\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eSubjective and objective sleepiness and state anxiety scores across the two experimental days.\u003c/p\u003e\n\u003cp\u003e(A) State Anxiety Index. No significant differences were observed between groups or across days.\u003c/p\u003e\n\u003cp\u003eB) Stanford Sleepiness Scale score. On Day 3, the partial sleep deprivation group reported significantly higher subjective sleepiness than the sleep deprivation group (***\u003cem\u003ep\u003c/em\u003e \u0026lt; .001).\u003c/p\u003e\n\u003cp\u003eC) Reciprocal mean reaction time (1/s) for the Psychomotor Vigilance Task. A significant interaction was observed, with the sleep group showing improved performance and the partial sleep deprivation group showing a decline in performance on Day 3 (*\u003cem\u003ep\u003c/em\u003e \u0026lt; .05, **\u003cem\u003ep\u003c/em\u003e \u0026lt; .01).\u003c/p\u003e\n\u003cp\u003eError bars represent the standard deviation.\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-7562866/v1/25c2e51178d8b456bfe31a4f.png"},{"id":93335722,"identity":"724aaf6c-5a8a-45fe-84c8-f3852d29e469","added_by":"auto","created_at":"2025-10-12 14:00:38","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":58008,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eCreativity tasks\u003c/strong\u003e\u003cbr\u003e\nCreativity task performance across two experimental days.\u003c/p\u003e\n\u003cp\u003eA) Scores on the Divergent Association Task (DAT). No significant differences were observed between the sleep group and the partial sleep deprivation group across days.\u003c/p\u003e\n\u003cp\u003eB) Mean number of completions for repeated tangram shapes (shapes1) on Day 2 and Day 3. Performance improved on Day 3 for both groups, suggesting a learning or incubation effect, but no group differences were found.\u003c/p\u003e\n\u003cp\u003eC) Mean number of completions for novel tangram shapes (shapes2) on Day 3. Both groups generated more completions compared to Day 2 shapes1, with no significant group differences, suggesting possible generalization of creative strategies.\u003c/p\u003e\n\u003cp\u003eError bars represent standard deviations.\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-7562866/v1/8567395f326686055e49f42d.png"},{"id":93338074,"identity":"7d9abb2a-43d5-4598-9875-b31a736e0311","added_by":"auto","created_at":"2025-10-12 14:16:39","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":962216,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7562866/v1/634a35c3-6321-4f40-9cfb-71887d82eed7.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Investigating the effects of partial sleep deprivation on creative ability: An experiment using Bayesian two-way analysis of variance","fulltext":[{"header":"Background","content":"\u003cp\u003eMany famous creative individuals are said to have sacrificed their sleep for their work. Creative professionals, such as designers, researchers, and artists, are generally expected to experience sleep deprivation. Although no scientific study focusing exclusively on sleep duration among creative professionals could be found, a survey conducted on designers in Japan via social media reported that over half of the respondents slept for less than 6 h, with approximately 23% sleeping for less than 5 h (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://amix-design.com/asoboad/stat/graph-9003/\u003c/span\u003e\u003cspan address=\"https://amix-design.com/asoboad/stat/graph-9003/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). Some studies have suggested that higher levels of divergent thinking are associated with insomnia symptoms [1] and that highly creative children often experience sleep disorders [2]. However, these findings seem to contradict previous research indicating that sleep deprivation impairs fundamental functions underlying creativity. Many aspects of the relationship between sleep and creativity remain unexplored.\u003c/p\u003e\u003cp\u003eCreativity encompasses multiple abilities. This study adopted an approach that focuses on a specific cognitive domain to scientifically capture the highly heterogeneous concept of creativity. Dietrich and Kanso [3] classified creativity into three components: insight, divergent thinking, and artistic creativity. Insight is considered a productive type of thinking that generates new solutions to problems rather than a reproductive form of thought that applies previously experienced solutions [4]. Divergent thinking is the ability to generate multiple solutions to problems with no upper limits or definitive conclusions [5]. Here, we focus on insight and divergent thinking.\u003c/p\u003e\u003cp\u003eSleep deprivation adversely affects cognitive and motor performance. Dinges et al. [6] restricted the sleep duration of participants to 33% less than their habitual sleep time (resulting in an average of 4.95 h of sleep per night) and had them perform task-related exercises. Their results suggested that even a single night\u0026rsquo;s sleep restriction was associated with a decline in performance. Studies examining the relationship between sleep deprivation and brain function have suggested that the prefrontal cortex, which is involved in arousal, attention, and the ability to make and process appropriate judgments, is particularly vulnerable to sleep deprivation, thereby affecting goal-directed behavior and sustained attention [7,8]. Therefore, a decline in the cognitive abilities underlying creativity due to sleep deprivation is expected to lead to a decrease in creativity.\u003c/p\u003e\u003cp\u003eEven one night of total sleep deprivation has been suggested to impair the flexibility aspect of divergent thinking [9]. Fluency, originality, and flexibility scores in the Language Form of the Torrance Tests of Creative Thinking have been shown to deteriorate following partial sleep deprivation [10]. Conversely, Landmann et al. [11], who investigated the effects of sleep on memory consolidation and reorganization, reported that sleep deprivation was beneficial for memory restructuring and solving insight problems. Participants were divided into groups based on their sleep or waking conditions (nighttime sleep, nighttime sleep deprivation, and daytime waking) and performed tasks before and after the respective conditions. The Compound Remote Associates Task, which is a creativity test in which participants are asked to find a word that can be linked to each of three seemingly unrelated words, was implemented. The nighttime sleep deprivation group showed significantly shorter response times to problems that were not solved during the first administration of the task than the night-time sleep and daytime waking groups [11]. Furthermore, Lacaux et al. [12] reported that drowsiness, the state between sleep and wakefulness, may promote creativity. In their experiment, participants solved mathematical insight problems before and after taking a 20-minute break in a dark room. During the break, participants who transitioned to non-REM stage N1 sleep were 2.7 times more likely to detect a hidden problem-solving strategy than those who did not fall asleep.\u003c/p\u003e\u003cp\u003eThe relationship between sleep deprivation and creativity has been inconsistent across studies, and further investigation is needed. A systematic review by Lim et al. [13] identified eight studies that investigated creative task performance under sleep-deprived conditions. Overall, the review concluded that sleep deprivation tends to impair creative ability. In particular, divergent linguistic thinking appears to be more vulnerable. However, owing to the limited quality and small sample sizes of these studies, the conclusions are far from definitive. Moreover, all studies employed interventions based on total sleep deprivation (i.e., staying awake overnight). In everyday life, people rarely experience a full night of sleep deprivation, but rather often accumulate sleep debts because of shortened sleep. Although Nelson et al. [10] utilized partial sleep deprivation in their study, their protocol involved only a 30-minute nap during a 24-hour wake period, which represents a relatively intense form of partial sleep deprivation. Whether similar effects occur with mild partial sleep deprivation that typically occurs in daily life, remains unknown.\u003c/p\u003e\u003cp\u003eIn this study, we focused on the aspects within the definitions of insight and divergent thinking that pertain to generating new solutions to a given problem and developing novel ideas under specific conditions, to examine their relationship with partial sleep deprivation. The participants were divided into groups based on their sleep conditions (normal sleep: 8 h; partial sleep deprivation: 2 h), and creativity tasks were administered before and after sleep manipulation. The creativity tasks included a tangram task, a type of insight problem [14], and a Divergent Association Task (DAT) that assesses linguistic divergent thinking.\u003c/p\u003e\n\u003ch3\u003eAims\u003c/h3\u003e\n\u003cp\u003eThis study investigated the effects of partial sleep deprivation on creative performance, focusing on verbal divergent thinking and insight problem-solving abilities. This study sought to clarify whether a single night of restricted sleep (2 h) significantly impairs these aspects of creativity compared with a full night of sleep (8 h).\u003c/p\u003e\u003cp\u003eThe following hypotheses were tested:\u003c/p\u003e\u003cp\u003e\u003col\u003e\u003cspan\u003e\u003cli\u003e\u003cp\u003ePartial sleep deprivation leads to decreased DAT performance, indicating reduced verbal divergent thinking ability.\u003c/p\u003e\u003c/li\u003e\u003c/span\u003e\u003cspan\u003e\u003cli\u003e\u003cp\u003ePartial sleep deprivation results in fewer correct and diverse solutions to tangram-based insight problems, reflecting reduced problem-solving flexibility.\u003c/p\u003e\u003c/li\u003e\u003c/span\u003e\u003c/ol\u003e\u003c/p\u003e\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\u003ch2\u003eSignificance\u003c/h2\u003e\u003cp\u003eThe findings of this study offer new insights into the relationship between sleep and creativity under conditions that closely resemble everyday experiences of mild sleep loss. Unlike previous studies that primarily focused on total sleep deprivation, this study contributes to the understanding of how partial sleep deprivation, a more common and ecologically valid form of sleep restriction, affects creative cognition.\u003c/p\u003e\u003c/div\u003e"},{"header":"Method","content":"\u003cp\u003e\u003cstrong\u003e\u003cem\u003eEthics\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAll participants provided informed consent and the study was conducted in accordance with the Declaration of Helsinki. The study protocol was approved by the Ethics Committee of the Faculty of Design at Kyushu University (approval number 373).\u003c/p\u003e\n\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003ePreliminary Survey\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe study was conducted between October 2020 and October 2021. A total of 46 young male participants (mean age 21.8 \u0026plusmn; 2.0 years) from the Faculty of Arts and Design at Kyushu University took part in the preliminary survey. Prior to the experiment, participants were provided with explanations of the experimental procedures and important considerations. Consent for participation was obtained via an online program (because the experiment was conducted during the COVID-19 pandemic). The eligibility criteria were that the participants had no current or past psychiatric or sleep disorders, were not taking any medications that affected their sleep, and were able to adhere to the study schedule.\u003c/p\u003e\n\u003cp\u003eBefore the experiment, an email containing a link to the online program and tangram task was sent to each participant. The deadline for completing the task was set within two weeks of receiving the email.\u003c/p\u003e\n\u003cp\u003ePrior to the experiment, participants completed questionnaires regarding their sleep habits and mental health through an online program. On the day of the experiment, between 10:00 a.m. and 12:00 a.m., the participants answered a questionnaire on sleepiness via an online program at home and completed the tangram task. After completing the task, participants were asked to send the results of the creative task via email. As this experiment focused on the impact of everyday sleepiness on creativity, sleep was not controlled.\u003c/p\u003e\n\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eQuestionnaires on Preliminary Survey\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe Pittsburgh Sleep Quality Index (PSQI) is a questionnaire designed to measure sleep health by assessing items such as sleep quality, sleep onset latency, sleep duration, sleep efficiency, sleep disturbances, the use of sleeping medications, and daytime dysfunction [15]. The participants were asked to complete the PSQI between receiving the URL for the online program and performing the tangram task. \u003c/p\u003e\n\u003cp\u003eThe K6 is a screening tool for depression and anxiety disorders. Its optimal cutoff score is 13, with a score of 13 or higher indicating potential depression or anxiety disorder [16]. The participants were required to complete the K6 during the period between receiving the URL for the online program and undertaking the tangram task.\u003c/p\u003e\n\u003cp\u003eThe Stanford Sleepiness Scale (SSS) is a 7-point subjective measure of sleepiness, with 1 representing the highest level of alertness and 7 representing the lowest [17,18] The SSS scores were used as an index of subjective alertness in this study. The SSS was completed by the participants before performing the tangram task between 10:00 a.m. and 12:00 a.m.\u003c/p\u003e\n\u003cp\u003eThe results of these questionnaire were collected for use in the subsequent main experiment.\u003c/p\u003e\n\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eTangram Task for the Preliminary Survey\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eIn this study, a tangram task, which is a type of insight problem, was employed as a creativity task. A tangram is a mathematical puzzle in which seven pieces (triangles and quadrilaterals cut from a square) are arranged to form various shapes. This has been used in previous studies as a measure of insight [14].\u003c/p\u003e\n\u003cp\u003eThe participants were sent an email with an attached file containing the image shown in Figure 1A, and were instructed to print the file before the experiment and cut the figure into seven pieces. During the task, participants were instructed to use all seven pieces shown in Figure 1A to create each tangram, as shown in Figure 1B (a total of four tangrams). A time limit of 600 seconds (10 minutes) was set for each tangram and the time taken to complete each tangram was recorded. Moreover, because one of the purposes of this experiment was to select participants for the second experiment, all participants were instructed to perform the task under the same conditions (solving the tangrams in the order of (a), (b), (c), and (d), as shown in Figure 1B) and to work continuously without taking any breaks or sleeping.\u003c/p\u003e\n\u003cp\u003eThe results of the tangram task from the preliminary survey were used to select participants for the main experiment, in which a task was planned to examine the number of different combinations that could be created within the time limit for a single tangram. To reduce learning effects, a task different from the tangram task used in the main experiment was administered during the preliminary survey; however, it was designed such that the participants could approach the tangram in a similar manner.\u003c/p\u003e\n\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eMain Experiment\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eParticipants\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eThe main experiment included 27 participants (22.6 \u0026plusmn; 1.9 years old) selected from the preliminary survey. The PSQI and K6 scores obtained from the preliminary survey were used to screen participants for the main experiment. For participants whose interval between the preliminary survey and the main experiment exceeded one month, the PSQI and K6 were administered again. Furthermore, to equalize the participants\u0026rsquo; ability to solve the tangram task, individuals were sequentially assigned to two groups (sleep and partial sleep deprivation), starting with those whose average tangram rank from the preliminary survey was closest to the overall mean. A power analysis conducted using G*Power (https://www.psychologie.hhu.de/arbeitsgruppen/allgemeine-psychologie-und-arbeitspsychologie/gpower) set the target sample size at 28 participants.\u003c/p\u003e\n\u003cp\u003eUltimately, 27 participants volunteered and were divided into two groups, such that variations in age, tangram-solving ability (measured by the average rank in the preliminary survey), and K6 scores were balanced. Additionally, efforts were made to distribute university majors (e.g., ergonomics, industrial design, acoustic design, art, and information design) as evenly as possible between the groups. The sleep group comprised 13 participants (22.4 \u0026plusmn; 1.8 years old), and the partial sleep deprivation group included 14 participants (22.8 \u0026plusmn; 1.9 years old). The average tangram rank was 18.2 \u0026plusmn; 3.2 for the sleep group and 17.6 \u0026plusmn; 3.7 for the partial sleep deprivation group, and the average K6 scores were 5.3 \u0026plusmn; 2.7 and 4.9 \u0026plusmn; 2.9, respectively.\u003c/p\u003e\n\n\u003cp\u003e\u003cem\u003eExperimental Procedure\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eFor the two days preceding the experiment, the participants were instructed to sleep for 8 h, with their bedtime set 4 h before and their wake-up time 4 h after the individual sleep midpoint, as determined by the PSQI. Additionally, for any participant whose bedtime would otherwise fall later than 1:00 a.m., sleep was regulated so that they went to bed at 1:00 a.m. and woke up at 9:00 a.m. This was applied to seven participants in the sleep group and three in the partial sleep deprivation group. Moreover, during the two-day sleep regulation period prior to the experiment, participants were prohibited from taking naps and heavy drinking.\u003c/p\u003e\n\u003cp\u003eOn the day before the experiment, all participants underwent a polymerase chain reaction (PCR) test as a preventive measure against COVID-19. Only patients who tested negative on the first day of the experiment were included.\u003c/p\u003e\n\u003cp\u003eThe experiment was conducted over three consecutive days (Figure 2A). During the experimental period, participants were not permitted to consume tobacco, caffeine, or alcohol, or engage in vigorous exercise, exciting activities, puzzle games, or napping. However, to minimize stress, activities that did not fall under these prohibitions, such as working on class assignments, reading, playing games, using the internet, or conversing with experimenters or other participants, were allowed at the discretion of the test conductors. During non-task periods, the participants stayed in the Human Factors Laboratory on the fourth floor of Building 1 at Kyushu University\u0026rsquo;s Ohashi Campus, where the room temperature was maintained at 26 \u0026plusmn; 2\u0026deg;C. Although the basic set temperature was 26\u0026deg;C, the room temperature was increased if any participant complained of feeling cold.\u003c/p\u003e\n\u003cp\u003eOn the first day of the experiment, the participants were required to arrive at the experimental room at least one hour before their scheduled bedtime during the sleep regulation period. In the experimental room, after being fitted with the wearable device Fitbit Inspire2 (Fitbit, Inc.), the participants were instructed to sleep for 8 h under the same bedtime and wake-up times as those established during the sleep regulation period.\u003c/p\u003e\n\u003cp\u003eOn the second day, 1.5 h after their wake-up time, electrodes for electroencephalography (EEG) were attached, and the participants completed a series of tasks: an eyes-open resting period, the Psychomotor Vigilance Test (PVT), the DAT, and the tangram task (using shapes labeled Heart and Square [shapes1]). EEG measurements were conducted in a soundproof and radiofrequency-shielded room on the fourth floor of Building 1 at the campus. During the task performance, the participants wore short-sleeved shirts and shorts. The room conditions were strictly controlled: temperature at 26\u0026deg;C, humidity at 50%, and desk illuminance between 550 and 650 lux. On the night of the second day, after donning the wearable device, the participants were allowed either 8 (sleep group) or 2 (partial sleep deprivation group) h of sleep, and the partial sleep deprivation group was required to go to bed 2 h before their scheduled wake-up time.\u003c/p\u003e\n\u003cp\u003eOn the third day, 1.5 h after the wake-up time, EEG electrodes were attached again, and the participants performed the following tasks: an eyes-open resting period, the PVT, the DAT, and the tangram task (using shapes labeled as Heart and Square [shapes1] and Bird and Pot [shapes2]) (Figure 2B). On each experimental day, a maximum of two participants were tested simultaneously. In cases where two participants were scheduled concurrently, the second participant\u0026rsquo;s session commenced immediately after the first participant completed their session; consequently, the start time for the second participant was approximately 2.5\u0026ndash;3 h after the wake-up time. In total, 15 participants (7 in the sleep group and 8 in the partial sleep deprivation group) participated in the first session and 12 participants (6 in each group) participated in the second session. However, the EEG results are not reported here.\u003c/p\u003e\n\n\u003cp\u003e\u003cem\u003eQuestionnaires\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eThe State-Trait Anxiety Inventory (STAI) [19] is a 20-item questionnaire designed to measure state anxiety. To assess subjective sleepiness and mood, participants were instructed to complete the SSS and STAI via an online program immediately after the start of the experimental session.\u003c/p\u003e\n\n\u003cp\u003e\u003cem\u003ePsychomotor Vigilance Task \u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eThe PVT was administered to objectively assess sleepiness. It measures sustained attention and is an objective indicator of decreased alertness [20]. In this study, a five‑minute PVT was conducted on a tablet device (FFF SMART LIFE CONNECTED, FFF‑TAB7) using the open-source software Vigilance Buddy 1.53 (https://researchbuddies.com/). In this task, a stimulus displayed as a video of incrementing numbers was presented on the screen and the participants were instructed to tap the screen as quickly as possible upon noticing the stimulus. The reaction time from the onset of the stimulus to the screen tap was measured. The primary outcome measure analyzed was the reciprocal of the mean reaction time, with a lower value indicating a decline in sustained attention [21].\u003c/p\u003e\n\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eCreativity Tasks\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eDivergent Association Task\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eThe DAT was used to measure linguistic divergent thinking. It requires participants to list ten nouns that are as different as possible in terms of meaning or use within a four-minute period [22]. It was possible to objectively quantify divergent thinking abilities by calculating the semantic distances between these nouns. The DAT performance is correlated with other divergent thinking tasks (e.g., the Alternative Uses Task). \u003c/p\u003e\n\u003cp\u003eIn this study, a Japanese version of the DAT was developed based on an algorithm published by Olson et al. [22] (https://osf.io/bm5fd/) and was employed as the task. The DAT scores were calculated using Google Colaboratory as the Jupyter notebook execution environment and a pre-trained Japanese word vector file (.gz) was obtained online (https://fasttext.cc/docs/en/crawl-vectors.html). Following the scoring method of the English version, the first 7 valid responses out of the 10 nouns provided by the participants were used to compute 21 pairwise semantic distances. The DAT score was defined as 100 times the average semantic distance. Notably, a significant correlation was observed between the scores of the original and Japanese versions (\u003cem\u003er\u003c/em\u003e = 0.50, \u003cem\u003ep\u003c/em\u003e \u0026lt; 0.001), confirming that the Japanese DAT score is a useful index of creativity.\u003c/p\u003e\n\u003cp\u003eFor this task, the participants were given an A4 sheet that included task instructions, precautions, and an answer section, and were asked to write their responses using a ballpoint pen. Although the time limit for the task was four minutes, the task was concluded once the participant recorded 10 nouns. Furthermore, as the DAT was administered twice (either before and after sleep or before and after sleep deprivation) on the third day, participants were required to provide responses using a different set of nouns from those used on the second day.\u003c/p\u003e\n\n\u003cp\u003e\u003cem\u003eTangram Task\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eThe tangram task was used as a creativity task to measure insight. The tangrams used in the second experiment were modeled using the 3D CAD software \u0026ldquo;Fusion 360\u0026rdquo; (.stl files). From these models, 3D printer files (.gx files) were created using the slicing software \u0026ldquo;FlashPrint.\u0026rdquo; Subsequently, using a 3D printer (\u0026ldquo;FLASHFORGE Adventurer3\u0026rdquo;), 22 sets of squares\u0026mdash;each composed of 7 pieces and measuring 100 mm (height) \u0026times; 100 mm (width) \u0026times; 10 mm (thickness)\u0026mdash;were produced.\u003c/p\u003e\n\u003cp\u003eIn the tangram task, participants were required to determine the number of different combinations that could be created for a given silhouette. Each silhouette was allotted a time limit of 7 minutes. Two types of comparisons were performed for the tangram task. In this experiment, on the second and third days, the participants solved the same silhouette (shapes1) to examine the incubation effect. Additionally, by having the participants solve shapes1 on the second day and a different silhouette (shapes2) on the third day, the impact of different sleep conditions on their responses to a new creativity task was investigated.\u003c/p\u003e\n\u003cp\u003eConsidering that individual differences in proficiency with a single silhouette might affect performance, the experiment used two silhouettes on the second day (Heart and Square: shapes1); on the third day, we used the same silhouettes (Heart and Square: shapes1) and two new silhouettes (Bird and Pot: shapes2). This design was used to average the responses for each pair of silhouettes, thereby neutralizing the effects of individual strengths or weaknesses. Given that the Heart and Pot shapes were bilaterally symmetrical, creating one combination might inadvertently lead to the formation of a symmetrical duplicate. Therefore, the participants were instructed not to produce bilaterally symmetrical combinations.\u003c/p\u003e\n\u003cp\u003eIn this experiment, the tangram task was performed using the 3D-printed tangrams. To account for order effects, the experiment was conducted in four different orders: two orders on the second day (Square \u0026rarr; Heart or Heart \u0026rarr; Square) and two orders on the third day (Bird \u0026rarr; Heart \u0026rarr; Pot \u0026rarr; Square or Square \u0026rarr; Pot \u0026rarr; Heart \u0026rarr; Bird), with counterbalancing among the participants.\u003c/p\u003e\n\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eStatistical Analysis\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eStatistical analyses were performed using R version 3.6.1 and JASP version 0.19.3.\u003c/p\u003e\n\u003cp\u003eClassical statistical hypothesis testing based on \u003cem\u003ep\u003c/em\u003e-values alone cannot adequately support the hypothesis that there is no difference between groups. Therefore, a combined approach was employed that reported both Bayesian hypothesis testing using Bayes factors (BFs), as recommended by Keysers et al. [23], and conventional \u003cem\u003ep\u003c/em\u003e-values. Bayesian hypothesis testing calculates the subjective probability that a hypothesis is correct, thereby supporting the hypothesis that there is no difference between groups.\u003c/p\u003e\n\u003cp\u003eFirst, to confirm that there were no differences in demographic data between the groups, Bayesian hypothesis tests were conducted on the K6 scores, tangram response rankings from the preliminary survey, and age.\u003c/p\u003e\n\u003cp\u003eIn this study, creativity tasks and questionnaires were administered before and after sleep in both sleep and partial sleep deprivation groups. If the changes from pre- to post-sleep differed between the groups, it was regarded as evidence of a sleep effect. Accordingly, statistical methods were used to test for interactions between the groups (two levels: sleep group vs. partial sleep deprivation group) and date (two levels: Day 2 vs. Day 3).\u003c/p\u003e\n\u003cp\u003eTwo separate analyses were performed for the tangram task, corresponding to the following research questions:\u003c/p\u003e\n\u003col start=\"1\" type=\"1\"\u003e\n\u003cli\u003eExamining changes in performance on the same task (i.e., comparing performance on the same tangram task\u0026mdash;Heart and Square shapes\u0026mdash;on Days 2 and 3). \u003c/li\u003e\n\u003cli\u003eExamining the generalization of performance to a different task (i.e., comparing performance on different tangram tasks: on Day 2, Heart and Square shapes and on Day 3, Bird and Pot shapes).\u003c/li\u003e\n\u003c/ol\u003e\n\u003cp\u003eIn each analysis, the mean number of responses was used as the outcome measure. For the STAI, SSS, PVT, DAT, and the number of responses in the tangram tasks with the same silhouette and those with different silhouettes, a mixed-design two-way analysis of variance (ANOVA) (group [two levels: sleep group, partial sleep deprivation group] \u0026times; date [two levels: Day 2, Day 3]) was conducted. Both BFs and \u003cem\u003ep\u003c/em\u003e-values were calculated. When the assumption of sphericity was violated, the Greenhouse\u0026ndash;Geisser correction was applied to adjust the degrees of freedom. Post-hoc tests were conducted using two-tailed \u003cem\u003et\u003c/em\u003e-tests with adjustments for multiple comparisons using the modified sequentially rejective Bonferroni method. One participant in the partial sleep deprivation group who did not complete the SSS was excluded from the analysis, resulting in 13 participants each in the sleep and partial sleep deprivation groups. Additionally, one participant in the sleep group whose PVT data could not be recorded was excluded, resulting in data from 12 and 14 participants in the sleep and partial sleep deprivation groups, respectively, being analyzed.\u003c/p\u003e\n\u003cp\u003eIn this study, following Jeffreys\u0026rsquo; [24] criteria, BF was interpreted as follows: A BF between 1 and 3 indicates little anecdotal evidence in favor of the alternative hypothesis. A BF value between 3 and 10 indicates moderate support for the alternative hypothesis. A BF of 10 or greater indicates strong support for the alternative hypothesis. Conversely, a BF of less than one indicates support for the null hypothesis. Specifically, a BF between 0.1 and approximately 0.333 indicates moderate support for the null hypothesis, and a BF of 0.1 or below is interpreted as strong evidence in favor of the null hypothesis. These criteria were used to evaluate the strength of evidence for the hypotheses in each statistical test.\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003e\u003cstrong\u003e\u003cem\u003eDemographic Data\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eDemographic data are presented in Table 1. A Bayesian \u003cem\u003et\u003c/em\u003e-test was conducted on age, K6 scores, PSQI scores, and average tangram rankings between the sleep and partial sleep deprivation groups. The resulting BFs ranged from 0.358 to 0.4, providing modest support for the hypothesis that there were no differences between the groups (age: BF = 0.4; K6: BF = 0.38; PSQI: BF = 0.358; and tangram: BF = 0.382).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eTable 1. Demographic data\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"590\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 81px;\"\u003e\n \u003cp\u003eCondition\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003eAge\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003eRank of tangram task\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 81px;\"\u003e\n \u003cp\u003eK6 (depressive symptoms)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003ePSQI (sleep quality)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 81px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003eMean\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003eSD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003eMean\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003eSD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 81px;\"\u003e\n \u003cp\u003eMean\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003eSD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003eMean\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003eSD\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 81px;\"\u003e\n \u003cp\u003eSleep control\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003e22.385\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003e1.968\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003e18.179\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003e3.874\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 81px;\"\u003e\n \u003cp\u003e5.308\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003e2.983\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003e5.077\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003e2.303\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 81px;\"\u003e\n \u003cp\u003eSleep deprivation\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003e22.786\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003e1.850\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003e17.607\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003e3.279\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 81px;\"\u003e\n \u003cp\u003e4.857\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003e2.840\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003e5.071\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003e1.706\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eAbbreviation: SD = standard deviation, PSQI = Pittsburgh sleep quality index\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eState Anxiety\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eFor the STAI results, a two-way ANOVA (group [two levels] \u0026times; date [two levels]) was performed (Figure 3A). The analysis revealed no significant main effects of group (\u003cem\u003eF\u003c/em\u003e(1, 25) = 0.28, \u003cem\u003ep\u003c/em\u003e = 0.600) or date (\u003cem\u003eF\u003c/em\u003e(1, 25) = 1.71, \u003cem\u003ep\u003c/em\u003e = 0.202). Moreover, no significant group \u0026times; date interaction was observed (\u003cem\u003eF\u003c/em\u003e(1, 25) = 0.15, \u003cem\u003ep\u003c/em\u003e = 0.700). The Bayesian two-way ANOVA results supported the hypothesis of no group effect (BF = 0.314), provided modest support for the hypothesis of no time effect (BF = 0.434), and strongly supported the hypothesis of no group \u0026times; time interaction (BF = 0.163).\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eSubjective Sleepiness\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eA two-way ANOVA (group [two levels] \u0026times; date [two levels]) was conducted on the SSS results (Figure 3B). The analysis revealed a significant main effect of group (\u003cem\u003eF\u003c/em\u003e(1, 24) = 14.9, \u003cem\u003ep\u003c/em\u003e \u0026lt; 0.001), with the partial sleep deprivation group exhibiting higher SSS scores than the sleep group. A significant main effect of date was also observed (\u003cem\u003eF\u003c/em\u003e(1, 24) = 12.7, \u003cem\u003ep\u003c/em\u003e = 0.002), with SSS scores being higher on Day 3 than on Day 2. Furthermore, a significant group \u0026times; date interaction was observed (\u003cem\u003eF\u003c/em\u003e(1, 24) = 14.9, \u003cem\u003ep\u003c/em\u003e \u0026lt; 0.001). Post-hoc tests indicated that on Day 3, the partial sleep deprivation group had significantly higher SSS scores (\u003cem\u003eF\u003c/em\u003e(1, 24) = 23.6, \u003cem\u003ep\u003c/em\u003e \u0026lt; 0.001), and within the partial sleep deprivation group, the SSS scores on Day 3 were significantly higher (\u003cem\u003eF\u003c/em\u003e(1, 12) = 18.7, \u003cem\u003ep\u003c/em\u003e = 0.001). The Bayesian two-way ANOVA results supported the hypotheses of a group effect (BF = 3.612), a time effect (BF = 10.583), and no group \u0026times; time interaction (BF = 9.263).\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eObjective Sleepiness\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eFor the PVT results, a two-way ANOVA (group [two levels] \u0026times; date [two levels]) was conducted (Figure 3C). Although there was a trend toward a group effect (\u003cem\u003eF\u003c/em\u003e(1, 24) = 2.94, \u003cem\u003ep\u003c/em\u003e = 0.099), no significant main effect of date was observed (\u003cem\u003eF\u003c/em\u003e(1, 24) = 1.44, \u003cem\u003ep\u003c/em\u003e = 0.242). However, a significant group \u0026times; date interaction was observed (\u003cem\u003eF\u003c/em\u003e(1, 24) = 9.75, \u003cem\u003ep\u003c/em\u003e = 0.005). Post-hoc tests revealed that on Day 3, the partial sleep deprivation group had a significantly lower reciprocal of mean reaction time (\u003cem\u003eF\u003c/em\u003e(1, 24) = 5.50, \u003cem\u003ep\u003c/em\u003e = 0.028), indicating poor sustained attention, whereas in the sleep group, the reciprocal of mean reaction time was significantly higher on Day 3 (\u003cem\u003eF\u003c/em\u003e(1, 11) = 11.2, \u003cem\u003ep\u003c/em\u003e = 0.007). Additionally, in the partial sleep deprivation group, the reciprocal of the mean reaction time was significantly lower on Day 3 (\u003cem\u003eF\u003c/em\u003e(1, 13) = 5.89, \u003cem\u003ep\u003c/em\u003e = 0.031). The Bayesian two-way ANOVA results supported the hypotheses of a group effect (BF = 3.357) and group \u0026times; time interaction (BF = 7.128) but provided inconclusive evidence of a time effect (BF = 2.331).\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eDivergent Association Task\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eA two-way ANOVA (group [two levels] \u0026times; date [two levels]) was conducted on the DAT results (Figure 4A). The analysis revealed no significant main effects of group (\u003cem\u003eF\u003c/em\u003e(1, 25) = 0.26, \u003cem\u003ep\u003c/em\u003e = 0.611) or date (\u003cem\u003eF\u003c/em\u003e(1, 25) = 0.46, \u003cem\u003ep\u003c/em\u003e = 0.502). Furthermore, no significant group \u0026times; date interaction was observed (\u003cem\u003eF\u003c/em\u003e(1, 25) = 1.50, \u003cem\u003ep\u003c/em\u003e = 0.231). The Bayesian two-way ANOVA results provided modest support for the hypotheses of no group effect (BF = 0.350), no time effect (BF = 0.266), and no group \u0026times; time interaction (BF = 0.197).\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eTangram Task (Same Shapes on Days 2 and 3)\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eFor the tangram task results using the same silhouette on both days (Figure 4B), a two-way ANOVA (group [two levels] \u0026times; date [two levels]) was performed. The analysis showed no significant main effect of group (\u003cem\u003eF\u003c/em\u003e(1, 24) = 0.04, \u003cem\u003ep\u003c/em\u003e = 0.849); however, a significant main effect of date was observed (\u003cem\u003eF\u003c/em\u003e(1, 24) = 23.2, \u003cem\u003ep\u003c/em\u003e \u0026lt; 0.001). No significant group \u0026times; date interaction was observed (\u003cem\u003eF\u003c/em\u003e(1, 24) = 0.27, \u003cem\u003ep\u003c/em\u003e = 0.608). The Bayesian two-way ANOVA results provided marginal support for the hypothesis of no group effect (BF = 0.334), very strong support for the hypothesis of a time effect (BF = 745.720), and inconclusive evidence regarding the group \u0026times; time interaction (BF = 0.406).\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eTangram Task (Different Shapes on Days 2 and 3)\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eFor the tangram task results using different silhouettes on Days 2 and 3 (Figure 4C), a two-way ANOVA (group [two levels] \u0026times; date [two levels]) was conducted. The analysis revealed no significant main effect of group (\u003cem\u003eF\u003c/em\u003e(1, 24) = 7.00 \u0026times; 10⁻⁴, \u003cem\u003ep\u003c/em\u003e = 0.979) and a significant main effect of date (\u003cem\u003eF\u003c/em\u003e(1, 24) = 96.0, \u003cem\u003ep\u003c/em\u003e \u0026lt; 0.001). No significant group \u0026times; date interaction was observed (\u003cem\u003eF\u003c/em\u003e(1, 24) = 0.04, \u003cem\u003ep\u003c/em\u003e = 0.839). The Bayesian two-way ANOVA results provided modest support for the hypothesis of no group effect (BF = 0.352), very strong support for the hypothesis of a time effect (BF = 8.905 \u0026times; 10⁹), and marginal support for the hypothesis of no group \u0026times; time interaction (BF = 0.382).\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eIn this study, we examined the effects of sleep deprivation on creativity by administering creativity tasks before and after a partial sleep deprivation simulation trial. Furthermore, we compared the changes in performance between the partial sleep deprivation and normal sleep groups. The following discussion considers these findings.\u003c/p\u003e\n\u003cp\u003eFirst, regarding the participant groups, we confirmed that there were no significant differences in the demographic data between the partial sleep deprivation and normal sleep groups. Based on the measures of age, tangram-solving ability, depression, and anxiety, no group differences were observed. Variance analysis using BFs provided modest support for the hypothesis that there were no differences between groups. Consequently, the sleep deprivation and partial sleep deprivation groups were effectively matched in terms of age, tangram-solving ability, and depression or anxiety.\u003c/p\u003e\n\u003cp\u003eSecond, the results regarding state anxiety, subjective sleepiness, and objective sleepiness are discussed. Regarding state anxiety, as measured by the STAI, neither a significant main effect nor a significant group \u0026times; time interaction was observed; the Bayesian two-way ANOVA strongly supported the absence of an interaction between group and time. This indicates that the sleep conditions employed in the study did not affect the participants\u0026rsquo; anxiety levels. Although previous studies have reported that multiple days of partial sleep deprivation are associated with increased STAI scores [25,26] this study\u0026rsquo;s use of a single day of partial sleep deprivation, which is a milder intervention, may account for this discrepancy.\u003c/p\u003e\n\u003cp\u003eConversely, the SSS showed a significant group \u0026times; date interaction, with the partial sleep deprivation group exhibiting a marked increase in subjective sleepiness relative to the normal sleep group. In addition, the PVT results revealed a significant group \u0026times; date interaction. Specifically, the partial sleep deprivation group demonstrated a decline in sustained attention following the intervention, as indicated by a lower reciprocal of the mean reaction time compared to the normal sleep group. These findings suggest that, although sleep manipulation did not alter state anxiety, it effectively increased the subjective and objective measures of sleepiness.\u003c/p\u003e\n\u003cp\u003eTo examine the effect of partial sleep deprivation on linguistic divergent thinking, we investigated whether there was a group \u0026times; time interaction in DAT performance. Regarding the DAT results, no significant main effects of group or time were observed and there was no significant group \u0026times; time interaction. The Bayesian two-way ANOVA supported the hypothesis that there was no interaction. These findings suggest that neither short sleep duration (2 h) nor normal sleep (2 h) differentially affects linguistic divergent thinking. This is the first study to demonstrate the impact of mild partial sleep deprivation, which is typically experienced in daily life, on creativity. A review of previous studies reported that, compared with performance on visuospatial tasks, linguistic divergent thinking is more adversely affected by sleep deprivation [13]; therefore, the absence of a difference in our study was unexpected. This discrepancy may be due to differences in the forms of sleep deprivation. All the reviewed studies considered total sleep deprivation (i.e., staying awake overnight). While partial sleep deprivation tends to preserve deep and rapid eye movement (REM) sleep, both of which are considered important for recovery because of sleep homeostasis, total sleep deprivation results in a loss in all sleep stages [27]. Our results suggested that the preservation of deep and REM sleep may have contributed to the recovery of divergent thinking. Although the intervention in this study was relatively mild compared with previous research, both subjective sleepiness and objective measures of sustained attention deteriorated significantly, but creative performance was maintained despite a decline in other aspects of cognitive functioning.\u003c/p\u003e\n\u003cp\u003eFinally, we discussed the results of the tangram task. In this study, we addressed two research questions: (1) changes in performance on the same task, and (2) the generalization of performance to a different task. For Research Question 1, we compared the performance on the same tangram task between Days 2 and 3, and for Research Question 2, we compared the performance on different tangram tasks between Days 2 and 3. Regarding Research Question 1, no significant group \u0026times; time interaction or main effect of group was found; however, a significant main effect of time was observed. Bayesian two-way ANOVA provided modest support for the hypothesis of no group effect, strong support for a time effect, and inconclusive evidence regarding a group \u0026times; time interaction. These results indicate that it is difficult to attribute differences in the ability to repeatedly solve the same tangram task to the effects of sleep and partial sleep deprivation. However, the significant main effect of time, strongly supported by the Bayesian analysis, suggests that performance improved when the same tangram task was repeated after sleep or partial sleep deprivation. This improvement was likely attributable to the learning effect of solving the same task twice. Another possibility is that sleep deprivation shortens response times for problems that have previously been solved [11]. Thus, the learning effect may have been offset by the detrimental effects of partial sleep deprivation on the overall performance and learning, leading to a lack of detectable group differences. However, the current results do not allow a detailed examination of this possibility.\u003c/p\u003e\n\u003cp\u003eRegarding Research Question 2, similar to Research Question 1, when comparing the performance on different tangram tasks between Days 2 and 3, no significant group \u0026times; time interaction or main effect of group was observed; however, a significant main effect of time was evident. The Bayesian two-way ANOVA provided modest support for the hypothesis of no group effect, strong support for a time effect, and modest support for a group \u0026times; time interaction. These findings also suggest a learning effect, with the improvement being generalized to different tangram tasks. In addition, the results provide modest support for the hypothesis that partial sleep deprivation does not influence this change. It is conceivable that participants were better able to combine elements from their memory of the shapes solved on Day 2 (shapes1) when tackling the same type of task (shapes2) on Day 3, leading to an increased number of combinations. Alternatively, differences in the task difficulty may have contributed to this finding. The difficulty of the \u0026ldquo;shapes2\u0026rdquo; task could have been inherently lower than that of the \u0026ldquo;shapes1\u0026rdquo; task for the participants, irrespective of any learning or sleep effects.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eLimitations\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe study had several limitations. First, although creativity tasks were administered twice\u0026ndash;before and after sleep, it was not possible to completely eliminate the learning effects or control for task difficulty across days. Despite the two tasks being designed to measure the same aspects of creativity, it is inherently difficult to create tasks that are both difficult and free from learning effects when measuring creative performance.\u003c/p\u003e\n\u003cp\u003eSecond, in the tangram task, it was difficult to objectively determine whether the participants experienced a fixation or impasse, which is a key component of insight problem solving. If participants had not experienced an impasse, the task might not have functioned effectively as an insight problem. However, if the task difficulty increased significantly, most participants would likely remain stuck and unable to find solutions. This highlights the importance of carefully adjusting the task difficulty when using insight problems to measure creativity in an experimental setting. Furthermore, whether one experiences an impasse in a particular insight problem depends heavily on individual traits or prior experience, making it particularly challenging to use insight-based tasks uniformly across participants.\u003c/p\u003e\n\u003cp\u003eA control condition involving a waking interval was necessary to separate the effects of learning from those of sleep. However, implementing a waking condition of an equivalent length would result in total sleep deprivation, making it difficult to isolate the learning effects. Another limitation is that the DAT used was a Japanese version for which translation and cross-linguistic validity have not been formally verified. Finally, the generalizability of the results is limited because the study included only male participants.\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eDespite the partial sleep deprivation group reporting greater subjective sleepiness and showing a decline in sustained attention compared with the sleep group, no group differences were observed in the linguistic divergent thinking task. Although BF did not meet Jeffreys\u0026rsquo; [24] threshold for strong evidence, the results of the tangram task tended to support the absence of a sleep effect. These findings do not support those of previous studies that reported a decline in creative ability due to sleep deprivation but are consistent with the findings suggesting that performance on complex tasks may be more resilient to the effects of sleep loss. The brain may maintain a certain level of creative function under sleep-deprived conditions, similar to that when fully rested [28,29]. Furthermore, Sasmita et al. [30] suggested that the attention function could be preserved during sleep deprivation when a reward is involved. Since creativity tasks often involve moments of insight and problem solving, they may engage the brain\u0026rsquo;s reward system, which could help maintain performance despite declines in subjective and objective alertness. Given that the DAT has been reported to be more enjoyable than other creativity tasks [22], it may be the most robust against the effects of partial sleep deprivation.\u003c/p\u003e\n\u003cp\u003eOur results suggested that mild sleep deprivation does not impair linguistic idea generation or performance on insight-based problems. This could partly explain why individuals in creative professions who are often sleep deprived may still function at a highly creative level. However, continuing to work while sleep deprived may come at the cost of other cognitive functions and health issues. Additionally, previous studies have shown that individuals with high creativity are more prone to insomnia and sleep disturbances [1,2] Thus, this study provides valuable evidence that highlights the need to raise awareness about the importance of maintaining sleep health among creative professionals.\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003cp\u003eANOVA: analysis of variance\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eBFs: Bayes factors\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eEEG: Electroencephalography\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eDAT: Divergent Association Task\u0026nbsp;\u003c/p\u003e\n\u003cp\u003ePCR: Polymerase chain reaction\u0026nbsp;\u003c/p\u003e\n\u003cp\u003ePSQI: Pittsburgh Sleep Quality Index\u0026nbsp;\u003c/p\u003e\n\u003cp\u003ePVT: Psychomotor Vigilance Test\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eREM: rapid eye movement\u0026nbsp;\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eSSS: Stanford Sleepiness Scale\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eSTAI: State-Trait Anxiety Inventory\u0026nbsp;\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eEthics approval and consent to participate\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAll participants provided informed consent and the study was conducted in accordance with the Declaration of Helsinki. The study protocol was approved by the Ethics Committee of the Faculty of Design at Kyushu University (approval number 373).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAvailability of data and materials\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare that they have no competing interests.\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis work was supported by KAKENHI(19K16244).\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthors\u0026apos; contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eYM and TI analysed and interpreted the data and were major contributors of writing the manuscript. TI, TU, MK and KM performed the sleep experiment. All authors read and approved the final manuscript. \u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgments\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe would like to acknowledge the participants and staff involved in the experiment.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eBeaty RE, Silvia PJ, Nusbaum EC, Vartanian O. Tired minds, tired ideas? Exploring insomnia and creativity. Thinking Skills Creativity. 2013;9:69\u0026ndash;75.\u003c/li\u003e\n\u003cli\u003eHealey D, Runco MA. Could creativity be associated with insomnia? Creativity Res J. 2006;18:39\u0026ndash;43.\u003c/li\u003e\n\u003cli\u003eDietrich A, Kanso R. A review of EEG, ERP, and neuroimaging studies of creativity and insight. Psychol Bull. 2010;136:822\u0026ndash;48.\u003c/li\u003e\n\u003cli\u003eMayer RE. Thinking, problem solving, cognition. W H Freeman/Times Books/Henry Holt and company; 1992.\u003c/li\u003e\n\u003cli\u003eGuilford JP. The nature of human intelligence. New York: McGraw-Hill; 1967.\u003c/li\u003e\n\u003cli\u003eDinges DF, Pack F, Williams K, Gillen KA, Powell JW, Ott GE, et al. Cumulative sleepiness, mood disturbance, and psychomotor vigilance performance decrements during a week of sleep restricted to 4\u0026ndash;5 h per night. Sleep. 1997;20:267\u0026ndash;77.\u003c/li\u003e\n\u003cli\u003eBoddez Y, Buabang EK, Zenses AK, Descheemaeker M. Commentary: Sleep deprivation promotes habitual control over goal-directed control: Behavioral and neuroimaging evidence. Front Behav Neurosci. 2018;12:82.\u003c/li\u003e\n\u003cli\u003eBorrag\u0026aacute;n G, Guerrero-Mosquera C, Guillaume C, Slama H, Peigneux P. Decreased prefrontal connectivity parallels cognitive fatigue-related performance decline after sleep deprivation. An optical imaging study. Biol Psychol. 2019;144:115\u0026ndash;24.\u003c/li\u003e\n\u003cli\u003eWimmer F, Hoffmann RF, Bonato RA, Moffitt AR. The effects of sleep deprivation on divergent thinking and attention processes. J Sleep Res. 1992;1:223\u0026ndash;30.\u003c/li\u003e\n\u003cli\u003eNelson CS, Dell\u0026rsquo;Angela K, Jellish WS, Brown IE, Skaredoff M. \u0026lsquo;Residents\u0026rsquo; performance before and after night call as evaluated by an indicator of creative thought. J Am Osteopath Assoc. 1995;95:600\u0026ndash;3.\u003c/li\u003e\n\u003cli\u003eLandmann N, Kuhn M, Maier JG, Feige B, Spiegelhalder K, Riemann D, et al. Sleep strengthens but does not reorganize memory traces in a verbal creativity task. Sleep. 2016;39:705\u0026ndash;13.\u003c/li\u003e\n\u003cli\u003eLacaux C, Andrillon T, Bastoul C, Idir Y, Fonteix-Galet A, Arnulf I, et al. Sleep onset is a creative sweet spot. Sci Adv. 2021;7:eabj5866.\u003c/li\u003e\n\u003cli\u003eLim AR, Williams BJ, Bullock B. The effect of sleep deprivation on creative cognition: A systematic review of experiment-based research. Creativity Res J. 2024:1\u0026ndash;11.\u003c/li\u003e\n\u003cli\u003eTakeuchi N, Mori T, Suzukamo Y, Izumi SI. Activity of prefrontal cortex in teachers and students during teaching of an insight problem. Mind Brain Educ. 2019;13:167\u0026ndash;75.\u003c/li\u003e\n\u003cli\u003eBuysse DJ, Reynolds CF, Monk TH, Berman SR, Kupfer DJ. The Pittsburgh Sleep Quality Index: A new instrument for psychiatric practice and research. Psychiatry Res. 1989;28:193\u0026ndash;213.\u003c/li\u003e\n\u003cli\u003eFurukawa TA, Kawakami N, Saitoh M, Ono Y, Nakane Y, Nakamura Y, et al. The performance of the Japanese version of the K6 and K10 in the World Mental Health Survey Japan. Int J Methods Psychiatr Res. 2008;17:152\u0026ndash;8.\u003c/li\u003e\n\u003cli\u003eHoddes E, Zarcone V, Smythe H, Phillips R, Dement WC. Quantification of sleepiness: A new approach. Psychophysiology. 1973;10:431\u0026ndash;6.\u003c/li\u003e\n\u003cli\u003eMaclean AW, Fekken GC, Saskin P, Knowles JB. Psychometric evaluation of the Stanford sleepiness scale. J Sleep Res. 1992;1:35\u0026ndash;9.\u003c/li\u003e\n\u003cli\u003eSpielberger CD, Gorsuch RL, Lushene RE.. STAI manual for the State-Trait Anxiety Inventory (\u0026ldquo;Self-Evaluation Questionnaire\u0026rdquo;). Consulting Psychologists Press; 1970. https://ci.nii.ac.jp/ncid/BA66835301. Accessed 27 March 2025.\u003c/li\u003e\n\u003cli\u003eBasner M, Mollicone D, Dinges DF. Validity and sensitivity of a brief psychomotor vigilance test (PVT-B) to total and partial sleep deprivation. Acta Astronaut. 2011;69:949\u0026ndash;59.\u003c/li\u003e\n\u003cli\u003eBasner M, Dinges DF. Maximizing sensitivity of the psychomotor vigilance test (PVT) to sleep loss. Sleep. 2011;34:581\u0026ndash;91.\u003c/li\u003e\n\u003cli\u003eOlson JA, Nahas J, Chmoulevitch D, Cropper SJ, Webb ME. Naming unrelated words predicts creativity. Proc Natl Acad Sci U S A. 2021;118:e2022340118.\u003c/li\u003e\n\u003cli\u003eKeysers C, Gazzola V, Wagenmakers EJ. Using Bayes factor hypothesis testing in neuroscience to establish evidence of absence. Nat Neurosci. 2020;23:788\u0026ndash;99.\u003c/li\u003e\n\u003cli\u003eJeffreys H. Theory of probability; 1961. https://global.oup.com/academic/product/the-theory-of-probability-9780198503682. Accessed 21 March 2025.\u003c/li\u003e\n\u003cli\u003eMotomura Y, Kitamura S, Oba K, Terasawa Y, Enomoto M, Katayose Y, et al. Sleep debt elicits negative emotional reaction through diminished amygdala\u0026ndash;anterior cingulate functional connectivity. PLOS One. 2013;8:e56578.\u003c/li\u003e\n\u003cli\u003eMotomura Y, Katsunuma R, Yoshimura M, Mishima K. \u0026lsquo;Two days\u0026rsquo; sleep debt causes mood decline during resting state via diminished amygdala\u0026ndash;prefrontal connectivity. Sleep. 2017;40:zsx133.\u003c/li\u003e\n\u003cli\u003eBanks S, Dinges DF. Behavioral and physiological consequences of sleep restriction. J Clin Sleep Med. 2007;3:519\u0026ndash;28.\u003c/li\u003e\n\u003cli\u003eHarrison Y, Horne JA. The impact of sleep deprivation on decision making: A review. J Exp Psychol Appl. 2000;6:236\u0026ndash;49.\u003c/li\u003e\n\u003cli\u003eLim J, Dinges DF. A meta-analysis of the impact of short-term sleep deprivation on cognitive variables. Psychol Bull. 2010;136:375\u0026ndash;89.\u003c/li\u003e\n\u003cli\u003eSasmita K, Massar SAA, Lim J, Chee MWL. Reward motivation normalises temporal attention after sleep deprivation. J Sleep Res. 2019;28:e12796.\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Sleep loss, creative cognition, divergent thinking, insight","lastPublishedDoi":"10.21203/rs.3.rs-7562866/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7562866/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground\u003c/h2\u003e\u003cp\u003eThis study investigated the impact of partial sleep deprivation on creativity by comparing performance changes between the partial sleep deprivation and normal sleep groups before and after engaging in creative tasks.\u003c/p\u003e\u003ch2\u003eMethods\u003c/h2\u003e\u003cp\u003eParticipants were divided into two groups: those with partial sleep deprivation and those with normal sleep patterns. Both groups completed creativity assessments, including the Japanese version of the Divergent Association Task (DAT) and Tangram puzzles, before and after the sleep intervention.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e\u003cp\u003e​The results indicated that despite increased subjective sleepiness and decreased sustained attention in the partial sleep deprivation group, there were no significant differences between the two groups in terms of performance on the DAT and tangram tasks. Bayesian two-way analysis of variance supported the hypothesis that partial sleep deprivation does not adversely affect these aspects of creativity.​\u003c/p\u003e\u003ch2\u003eConclusions\u003c/h2\u003e\u003cp\u003eThese findings suggest that mild partial sleep deprivation may not impair certain creative abilities such as verbal divergent thinking and insight problem solving. However, notably, while creative performance is maintained, other cognitive functions and overall health may be compromised by insufficient sleep. Therefore, individuals, especially those in creative professions, should be cautious of the potentially broader impact of sleep deprivation.​\u003c/p\u003e","manuscriptTitle":"Investigating the effects of partial sleep deprivation on creative ability: An experiment using Bayesian two-way analysis of variance","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-10-12 14:00:33","doi":"10.21203/rs.3.rs-7562866/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"cc9fc968-5232-4d7e-aa20-ac106eaec5c2","owner":[],"postedDate":"October 12th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2026-04-03T11:08:37+00:00","versionOfRecord":[],"versionCreatedAt":"2025-10-12 14:00:33","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7562866","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7562866","identity":"rs-7562866","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}