Sleep Deprivation Increases Belief Change and Suppresses Confirmation Bias in Decision Tasks

Sleep Deprivation Increases Belief Change and Suppresses Confirmation Bias in Decision Tasks | 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 Article Sleep Deprivation Increases Belief Change and Suppresses Confirmation Bias in Decision Tasks Charlotte Anckaert, Philippe Peigneux, Wim Gevers This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-5289682/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 Belief updating, the process of revising one’s beliefs in light of new information, is subject to confirmation bias - the tendency to prioritize information that supports existing beliefs. How this process is affected by sleep deprivation (SD), which is known to impair cognitive function, remains underexplored. Here, we examined how SD alters belief change and confirmation bias in response to peer advice during a decision-making task. Across two studies, we found that SD substantially increased the likelihood of belief changes, with participants being more likely to change their beliefs following SD compared to after a full night of sleep. Even after SD, beliefs were still changed using logical reasoning, as participants continued to consider their initial confidence. At the same time, SD eliminated confirmation bias. Indeed, the typical confirmation bias asymmetry (i.e., a stronger increase in confidence when in agreement compared to a smaller drop in confidence when in disagreement) was no longer observed under SD. Our findings suggest that sleep loss undermines belief stability and dampens the cognitive biases that typically reinforce belief consistency. As a result, SD contexts may exert a significant impact on decision making processes, eventually leading to deleterious consequences in high stake situations. Social science/Psychology Social science/Psychology/Human behaviour Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Introduction Sleep deprivation (SD) is a common condition among high-stakes decision-makers across various fields, including military, healthcare, and corporate environments (e.g., Associated Press, 2020; Bamford, 2001). SD is known to impair cognition and decision-making (Killgore et al., 2007; Landolt et al., 2014), potentially leading to altered behaviours in high-stakes situations (Killgore, 2010). Despite extensive research on the cognitive effects of SD – including suggestibility (Blagrove & Akehurst, 2000), increased threat perception (Zenses et al., 2019), and other altered cognitive processes (Kassin & Kiechel, 1996) – its specific impact on belief updating remains underexplored (Achtziger et al., 2014; Poinas et al., 2010). Belief updating is a critical process for accurate understanding and informed decision-making in modern information-rich societies (Sommer et al., 2023). SD, known to impair cognitive function and reduce motivation to reconcile inconsistencies (Blagrove, 1996), may hinder effective belief updating, particularly in individuals with decisional power like policymakers (Frenda et al., 2016). This impaired motivation could manifest in two contrasting ways: sleep-deprived individuals might either adopt new beliefs without critical evaluation (the "follow along" hypothesis) or rigidly adhere to their existing beliefs, downplaying the significance of new information (the "rigid" hypothesis, Dickinson & Drummond, 2008; Harrison & Horne, 1999). Belief updating, the modification of beliefs in response to new information, is a cognitive process well known to be influenced by confirmation bias, where individuals tend to favour information that supports their existing beliefs (Nickerson, 1998). Given its prominent role in belief updating, confirmation bias is a suited mechanism to explore how altered states of vigilance, such as SD, might influence this process. SD could either amplify confirmation bias, leading to increased rigidity and fewer belief changes (supporting the rigid hypothesis, Harrison & Horne, 1999), or diminish it, resulting in more belief changes in response to disconfirming information (supporting the follow-along hypothesis, Blagrove, 1996; Frenda et al., 2016). The current studies examine how SD affects two key aspects of belief updating: belief change and confidence update. These markers, which measure the extent to which individuals modify their beliefs and adjust their confidence in those beliefs, have been widely studied under normal conditions (e.g., Lefebvre et al., 2021; Rollwage et al., 2020; Rollwage & Fleming, 2021). Typically, belief change is associated with a significant drop in confidence (Pescetelli et al., 2021), and confidence increases when new information aligns with existing beliefs but decreases when it contradicts them. The present studies address these questions through two experiments using a binary decision task related to real estate property prices, adapted from Kappes et al. (2019). In this task, participants must assess their level of confidence in their belief, and receive feedback from a partner, who agreed or disagreed with their belief. In Study 1, participants could maintain or change their initial belief and update confidence in their final belief. Study 2 replicated Study 1 and was extended with a condition where participants could only update confidence in their belief, but no longer change it. We measured the confidence update and the frequency of belief changes. Confidence update is measured as the difference between the final (i.e., after incorporating the partner’s information) and initial confidence rating (i.e., confidence in the first belief about the price of a property). A positive confidence update reflects an increase in confidence. Belief change, on the other hand, is the act of switching from one to the other belief after incorporating the partner’s information. The task was performed twice in a counterbalanced within-subjects design: all participants performed the task once after a night of normal sleep and once after a SD night. With these experiments, we aim at clarifying how and to what extent SD impacts belief change frequency and confidence updating. If SD amplifies confirmation bias (the rigid hypothesis), we expect to see fewer belief changes and smaller confidence updates, particularly after receiving disconfirming information (Dickinson & Drummond, 2008; Harison & Horne, 1999). Alternatively, if SD diminishes confirmation bias (the follow-along hypothesis), we expect to see more frequent belief changes and larger confidence updates (Blagrove, 1996; Frenda et al., 2016). Study 1 Methods Participants 36 participants aged 19–25 years (M age = 21.25, SD age = 1.48, 13 female) were recruited through social media announcements and flyers. We aimed for a sample size of 32 based on the recommendations by Brysbaert and Stevens (2018) for mixed model analysis for a predicted group effect. Inclusion criteria were satisfying usual sleep quality (Pittsburgh Sleep Quality Index [PSQI] < 7, Buysse et al., 1989), a score < 7 at the shortened version of the Beck Depression Inventory (BDI, Rosner, 2015), and moderate or neutral chronotype (30 < score < 70) on the morningness-eveningness questionnaire (Horne & Östberg, 1976). Participants received monetary compensation and gave their informed consent to participate in this experiment approved by the local Ethical Committee (agreement #684/2022) of the Faculty of Psychology and Educational Sciences of the Université libre de Bruxelles. Procedure To increase power and control for interindividual variability in susceptibility to belief updating and influences of SD, we used a within-subjects design. All participants were randomly assigned to the SD or normal/regular sleep (RS) condition as a first session, followed by the other condition in a session held one week later (Fig. 1 ). At each session, both phases of the experiment lasted about 50 minutes each (see task details below). From Day1 to the end of the experiment, participants wore a wrist actimeter (wGT3XBT Monitor, ActiGraph, Pensacola, FL, USA) and filled in every morning a short questionnaire about their sleep quality on the previous night (except after the SD night, Ellis et al., 1981). Visual inspection of actimetric recordings was used in conjunction with sleep questionnaires to check the regularity of the sleep-wake cycle during the entire experiment. In the SD condition, participants stayed in the laboratory from ca. 8pm (after they finished Phase 1) to ca. 9am (after they finished Phase 2). They stayed awake the entire night under the supervision of two experimenters and were allowed to engage in quiet activities (e.g., reading, listening to music). High-arousal physical activities were not allowed. Light meals were offered throughout the night with continuous water access. The Karolinska Sleepiness Scale (KSS; Åkerstedt and Gillberg, 1990) was filled in every hour during the SD night to gauge the evolution of drowsiness. In the RS condition, participants spent a night of sleep at home in between Phase 1 and Phase 2. Smoking, alcohol, and energy drinks were not allowed on experimental days. Task The design is adapted from the Kappes et al.’s (2019) study experimental design, consisting of a belief induction phase (Phase 1) followed by a belief updating phase (Phase 2). Both phases of the study consisted of five practice trials followed by 200 experimental trials (Fig. 2 ). In Phase 2, information from another participant (namely, the pseudo-partner) was presented during the task. This information was pseudo-randomly generated, and data was collected in groups of minimum six subjects to create the impression that the partner's information was obtained from a genuine participant. During Phase 1 (Day3 evening in each condition), real estate images were depicted on screen, together with price indications taken from a Brussels-based real estate website. Since all participants underwent both conditions, we developed two databases of Brussels properties. Participants saw images of real estate properties, with prices manipulated to be either 20% higher or 20% lower than the actual market value. Using the keyboard, participants indicated their belief about whether the real price of the houses was higher or lower than the indicated price by pressing the up or down arrow, then had to provide their confidence rating in their decision using a slider on a 0-100 scale (Initial Confidence). In Phase 2 (Day4 morning), participants saw the same set of real estate images, with identical prices as in the first session. They also saw their initial opinion and level of confidence, as well as their pseudo-partner's opinion (but not their confidence, at variance with Kappes et al., 2019) for each property, resulting in agreement trials (i.e., the partner has the same opinion) and disagreement trials (i.e., the partner has the opposite opinion) with an equal number of trials for each. Using the up or down arrow (see Fig. 2 ) participants could then modify their initial decision (Belief Change) while seeing the new information on the screen, and then enter their confidence level on this second opinion (Confidence Update) on a 0-100 scale. Confidence Update was measured as the difference between the final and initial confidence rating of the participant. Data analysis Statistical analyses were conducted using R v4.4.1 (R Core Team, 2012). Belief Change and Confidence Update were statistically analysed using (generalised) linear mixed-effects models using the lme4 package v1.1-28 (Bates et al., 2015). Belief Change refers to the probability of changing beliefs on all trials. Confidence Update was calculated as the difference between the Final and Initial Confidence reported (final minus initial). A positive Confidence Update indicates an increase in confidence. Initial Confidence rating was controlled for, as participants with lower initial confidence have more room to increase their confidence after agreement and less room to decrease after disagreement, and vice versa for those with higher initial confidence. Planned comparisons (one-sided t-tests) were done using the emmeans package v1.7.2 (Lenth, 2019); false discovery rate (FDR) p -values are reported. Results Belief change A generalized linear mixed-effects model was used to investigate the influence of Initial Confidence (ranging between 0 and 100), Partner Information (agreement/ disagreement), and Sleep Condition (RS/ SD) on the probability of belief changes. The model included fixed effects for Initial Confidence, Partner Information, and Sleep Condition, as well as their interaction. A random intercept and slopes for Partner Information and Sleep Condition were specified for participants to account for individual differences. The model was fitted using the Laplace approximation with a binomial family and a logit link function. The analysis of the model revealed significant main effects of Initial Confidence, Partner Information, and Sleep Condition on the probability of belief change. Higher initial confidence level was associated with a lower probability of belief change (Estimate = -0.02, z = -16.83, p < .001). Not surprisingly, trials involving disagreement information increased the probability of belief change compared to agreement trials (Estimate = 1.75, z = 7.26, p < .001). Importantly, being sleep deprived strongly increased the probability of belief change compared to the RS condition (Estimate = 0.74, z = 4.13, p < .001). Specifically, the likelihood of belief changes in the RS condition was 5.4%, whereas in the sleep-deprived condition it increased to 11.6%. These results indicate that SD strongly increases the likelihood of participants changing their beliefs. The interaction effect between Partner Information and Sleep Condition was not observed (Estimate = 0.18, z = 1.33, p = 0.18), suggesting that the increase in frequency of belief changes is similar for agreement and disagreement trials when sleep deprived (Fig. 3 ). In conclusion, lower initial confidence, facing disagreement, and being sleep deprived all increase the probability of changing beliefs. Confidence Update A linear mixed-effects model was used to examine the effects of Partner Information and Sleep Condition on the Confidence Update. The model included fixed effects for Initial Confidence, Partner Information, and Sleep Condition, and their interaction, as well as a random intercept per participant and slopes for Partner Information and Sleep Condition to account for individual variability. The model was fitted using the Bound Optimization by Quadratic Approximation (“bobyqa”) optimizer with a maximum of 5000 function evaluations. The linear mixed-effects model revealed several significant effects on Confidence Update, controlled for Initial Confidence. Partner Information showed a significant main effect (χ2(1) = 47.81, β = −11.24, SE = 1.63, p < .001), with disagreement trials associated with a more negative confidence update compared to agreement trials. Sleep Condition also had a significant main effect (χ2(1) = 17.83, β = −4.12, SE = 0.98, p < .001), with participants in the SD condition showing a smaller confidence update compared to when they did the RS condition. The interaction between Sleep Condition and Partner Information (χ2(1) = 11.16, β = 1.84, SE = 0.55, p < .001) further illustrates the overall lower confidence after SD: on agreement trials, confidence increased less when sleep deprived compared to after RS (EMM sleep deprived = 3.62, EMM normal sleep = 7.74, : t (35) = 4.22, p < .001), while on disagreement trials, confidence decreased more in SD compared to RS (EMM sleep deprived = -5.78, EMM normal sleep = -3.50, : t (35) = 2.34, p < .05). As can be seen in Fig. 4 , the interaction reveals that confirmation bias is eliminated by SD. A stronger increase in confidence after agreement compared to the decrease after disagreement is only observed after RS ( t (35) = 6.91 p < .001). Interim discussion The findings from this first study reveal significant insights into the cognitive processes underlying belief updating and decision-making, particularly under SD influence. Our research indicates that SD strongly increases the likelihood of belief change. This suggests that a lack of sleep significantly alters the stability of one's beliefs. This increased frequency of belief changes appears to be intricately linked to lower confidence levels when individuals are sleep deprived. A Welch two-samples t-test showed that there was a significant difference in mean confidence between the two sleep conditions, suggesting that confidence is significantly higher after RS compared to after SD (M regular sleep = 65.39, M sleep deprivation = 60.02, t(35) = 13.3, p < .001). As confidence wanes, people seem more prone to reconsider their beliefs, highlighting the role of confidence as a critical mediator in belief stability (e.g., Gradassi et al., 2022). Our results also suggest that even under SD, a rational decision-making process persists: initial confidence levels continue to serve as a key factor influencing belief change. When participants enter a decision-making scenario with low initial confidence, they are more likely to change their beliefs, and this remains true when sleep deprived. This finding underscores the robustness of confidence as a determinant of belief stability, suggesting that it operates consistently across varying states of alertness. A second intriguing aspect of Study 1 is that confirmation bias, although present in well-rested individuals, disappears when they are sleep deprived. Under RS conditions, participants exhibited a tendency to increase their confidence more after receiving confirming information than they decreased it after receiving disconfirming information, demonstrating classic confirmation bias (Klayman, 1995). However, when deprived of sleep, this bias vanished. This suggests that SD may impair the cognitive mechanisms that typically drive confirmation bias, possibly leading to a more balanced evaluation of evidence, or simply to a different decision-making strategy when under the cognitive strain of SD. When receiving agreement information, the increase in confidence is weaker after SD. When receiving disconfirming information, the decrease in confidence is stronger after SD. Both these tendencies align well with the finding of a lower overall confidence after SD. However, the difference in decrease in confidence between RS and SD after receiving disconfirming information is less pronounced and borderline significant. A possible explanation for this phenomenon is that when sleep-deprived, participants may be less inclined to significantly lower their confidence after receiving disconfirming information because doing so triggers a belief change. In this context, the act of changing one's mind may be seen as a more adaptive response than sticking with an opinion that leads to a substantial decrease in confidence. Essentially, rather than holding onto a belief that suddenly seems much less certain, individuals may prefer to revise their belief altogether, thereby avoiding the discomfort associated with lowered confidence. We pre-registered this hypothesis prior to conducting the follow-up study (OSF, Anckaert & Gevers, 2023). We designed an experiment aimed at disentangling the relationship between belief change and confidence update under RS and SD conditions. In this study, we introduced a condition where participants are allowed only to update their confidence levels without giving them the option to change their beliefs. We hypothesised that confidence would decrease stronger after receiving disagreement information under SD and a belief change was not allowed compared to under RS, and compared to when a belief change was allowed. Study 2 Methods Participants 50 new participants aged 19–29 years (M age = 22.19, SD age = 2.09, 38 female) were recruited through social media announcements and flyers. Inclusion criteria and procedure were the same as Study 1. One participant failed to show up for the normal sleep condition and was therefore excluded. This follow-up experiment was conducted under approval of the local Ethical Committee (agreement #684/2022) of the Faculty. Task Phase 1 was identical in both Studies. Phase 2 partially differed: it consisted of two blocks of 100 trials. One block was identical to the second phase of Study 1, meaning that participants could change or maintain their initial belief while seeing the new information on the screen and update confidence level in their second opinion (Change Allowed block, CA). In the other, new block, participants could update their confidence but were not given the option to change their opinion (Fig. 5, No Change Allowed block, NCA). All participants performed both blocks in counterbalanced order 1 . Results Change Allowed block: Belief Change We used the same model as in Study 1 to explore how Partner Information, Sleep Condition, and their interaction affected Belief Change, while controlling for Initial Confidence. This model also included a random intercept for each participant and allowed the effects of Partner Information and Sleep Condition to vary by individual. The results of Study 1 were fully replicated. A significant negative effect of Initial Confidence was found (Estimate = − 0.03, z = − 15.35, p < .001), indicating that higher confidence was associated with lower odds of belief change. There was a main effect of Partner Information, with disagreement trials associated with higher odds of belief change compared to agreement trials (Estimate = 2.45, z = 9.27, p < .001). The main effect of Sleep Condition was observed again. SD associated with higher odds of changing beliefs (15.7%) compared to RS (10.9%, estimate = 0.532, z = 1.99, p = .047). Like Study 1, the interaction between Partner Information and Sleep Condition was not observed (Estimate = − 0.23, z = − 1.21, p = .23). Change Allowed vs. No Change Allowed block: Confidence Update We used a linear mixed-effects model to investigate the influence of having the possibility to change beliefs on confidence updating. We assessed the influence of Partner Information, Sleep Condition, and Change Condition (CA vs. NCA) on Confidence Update by including a main effect for Initial Confidence (to control for its influence) and a three-way interaction effect between Partner Information, Sleep Condition, and Change Condition. The model also included a random slope for Partner Information and Sleep Condition and a random intercept per participant. The analysis revealed a significant three-way interaction effect (χ2(7) = 234.92, p < .001). Figure 7 illustrates eight key comparisons, with statistical significance highlighted for clarity. Within-Group Comparisons: Sleep Condition and Partner Information • Change Allowed (Left Panel): Replicating Study 1, after a RS night, confirmation bias was observed. Participants showed a significant larger increase in confidence after agreement (EMM = 9.64) compared to the decrease after disagreement (EMM = -1.89), t (48) = 7.39, p < .001. SD did not significantly affect the confidence increase after agreement compared to RS. However, the confidence decrease following disagreement was larger under SD (EMM = -4.57) compared to RS (EMM = -1.89), t (48) = 2.59, p = .016. • No Change Allowed (Right Panel): Similarly, under RS, participants exhibited a confirmation bias, with a significant larger confidence increase after agreement (EMM = 10.22) compared to the decrease after disagreement (EMM = -3.62), t (48) = 8.88, p < .001. Under SD, participants increased their confidence less after agreement (EMM = 7.78) compared to RS (EMM = 10.22), t (48) = 2.35, p = .028. The confidence decrease after disagreement was more pronounced after SD (EMM = -8.77) than after RS (EMM = -3.62), t (48) = 5.01, p < .001. Between-Group Comparisons: Change Conditions • Comparing Change Allowed vs. No Change Allowed: When participants were not allowed to change their belief, confidence drops were more substantial across both sleep conditions. After RS, confidence dropped more in the NCA condition by a magnitude of 1.73, t (48) = 3.76, p < .001. After SD, this difference was even greater, with a larger confidence drop of 4.21 in the NCA condition, t (48) = 9.16, p < .001. An independent t-test was conducted to compare the magnitude of confidence drops between the SD and RS condition. The drop in confidence after disagreement between CA and NCA after SD was significantly larger compared to after RS ( t (48) = -3.44, p < .001). SD leads to a greater decrease in confidence when participants are not allowed to change their minds. Overall, these results consistently show that SD diminishes or suppresses confirmation bias, as seen in both conditions. The newly introduced NCA condition further amplifies the effects of SD, particularly by enlarging the confidence drop when participants face disagreement and are unable to change their beliefs. [1] Due to human error during data collection, the counterbalancement was not optimal. 34 participants completed one order, 16 the other. Discussion In two studies, we investigated the impact of SD on belief updating, specifically examining how alterations in sleep affect both the likelihood of belief changes and the associated confidence updates. Across two studies, our results consistently demonstrate that SD alters belief updating dynamics, making individuals more susceptible to belief change while also diminishing their confidence in those beliefs. In Study 2, we replicated key findings from Study 1, further highlighting how sleep loss disrupts cognitive biases such as confirmation bias. Under well-rested conditions (RS), our findings align with previous research on belief updating. Participants were more likely to change beliefs when initially low in confidence (e.g., Pescetelli & Yeung, 2020; Gradassi et al., 2022), and they exhibited the classic confirmation bias. In line with the literature (Klayman, 1995; Nickerson, 1998), the change in confidence was asymmetric: confidence increased more after receiving confirming information than it decreased following disconfirming information. One of the most striking findings is that SD increased the likelihood of belief changes non selectively, occurring both after agreement and disagreement. SD impairs cognitive functions such as memory and attentional control (Killgore et al., 2007; Landolt et al., 2014). The heightened rate of belief changes observed here may stem from these cognitive impairments, which weaken the stability of initial beliefs. Notably, in our studies, this effect was closely tied to confidence. Sleep-deprived participants exhibited lower confidence overall, particularly after receiving disconfirming information. This finding aligns with research suggesting that lower confidence increases the likelihood of belief change (Pescetelli et al., 2021). By reducing confidence, SD may undermine the cognitive stability necessary to sustain an initial belief, thereby prompting belief shifts more readily. This supports the “follow along” hypothesis, which posits that sleep-deprived individuals may have less motivation or cognitive resources to critically evaluate new information (Blagrove, 1996). Nonetheless, the intact role of initial confidence in predicting belief change under SD suggests that core mechanisms of belief stability are still operational (Baranski, 2007). Regarding confidence updates, SD reduced the typical confidence boost following agreement with a partner and amplified the confidence drop after disagreement, significantly attenuating confirmation bias. Rather than favouring confirmatory over disconfirmatory evidence, sleep-deprived participants appeared to "equalize" their responses to both types of information. This reduction in confirmation bias may be linked to findings that SD diminishes individuals' ability to weigh new information appropriately (Dickinson & Drummond, 2008) and reduces their ability to extract relevant information from stimuli (Johnson et al., 2020). The combination of lower overall confidence and impaired information processing likely contributes to an increased focus on disconfirming evidence, thereby reducing the strength of confirmation bias. The absence of confirmation bias under SD can be interpreted in several ways. It may reflect a weakening of the cognitive resources necessary to sustain bias (e.g., attentional control and memory; Harrison & Horne, 1999), or it may signal a shift in decision-making strategies. Sleep-deprived individuals may adopt a more balanced but less confident evaluation of new information, giving equal weight to confirming and disconfirming evidence. This reduction in bias could have both positive and negative implications: while it may lead to greater flexibility in updating beliefs, it may also result in erratic decision-making when consistency is required. A key insight from Study 2 is that when belief change was not allowed (NCA condition), participants experiencing sleep deprivation (SD) exhibited a significantly sharper decline in confidence after receiving disconfirming information. This suggests that, in the absence of the option to change beliefs, SD amplifies the psychological impact of disconfirming evidence, leading to heightened belief instability. Under these circumstances, individuals appear more vulnerable to disconfirming information, resulting either in frequent belief changes or steep declines in confidence when belief change is constrained. These findings underscore the critical role of confirmation bias in belief updating, particularly under conditions of SD. Typically, confirmation bias serves as a cognitive filter that helps stabilize beliefs by allowing individuals to give more weight to evidence supporting their existing beliefs while discounting contradictory information (Klayman, 1995). In well-rested individuals, this bias acts as a gatekeeper, buffering against confidence drops. However, when this gatekeeping function weakens under SD, the usual asymmetry between confidence gains after agreement and confidence drops after disagreement diminishes. Consequently, individuals become less able to maintain confidence in their beliefs, leading to increased flexibility in changing their beliefs or a greater likelihood of confidence collapse. Conclusion In conclusion, this research sheds light on how SD affects belief updating by increasing the frequency of belief changes and diminishing confirmation bias. Our findings have several important implications for understanding decision-making in high-stakes environments, where SD is common (e.g., military, medical, political, and corporate settings). Sleep-deprived decision-makers may be more prone to changing their beliefs in response to new information, which could be either advantageous or detrimental depending on the context. In fast-paced environments requiring rapid adaptation, this increased flexibility may be beneficial. However, in scenarios requiring stable, consistent decision-making, the reduced cognitive gatekeeping due to diminished confirmation bias could lead to erratic or suboptimal choices. Future research should explore the neural mechanisms underlying these effects and investigate whether similar patterns are observed under other forms of cognitive strain, such as stress or time pressure. Understanding these dynamics can lead to strategies to mitigate the adverse effects of SD on decision-making, particularly in critical settings where consistent and accurate judgments are essential. References Achtziger, A., Alós-Ferrer, C., Hügelschäfer, S., & Steinhauser, M. (2014). The neural basis of belief updating and rational decision making. Social Cognitive and Affective Neuroscience , 9 (1), 55–62. https://doi.org/10.1093/scan/nss099 Åkerstedt, T., & Gillberg, M. (1990). Subjective and Objective Sleepiness in the Active Individual. International Journal of Neuroscience , 52 (1–2), 29–37. https://doi.org/10.3109/00207459008994241 Associated Press. (2020, 11 december). EU Leaders Agree to Reduce Emissions After All-night Talks. Voice Of America . https://www.voanews.com/a/europe_eu-leaders-agree-reduce-emissions-after-all-night-talks/6199453.html Bamford, J. (2001). Body of Secrets: Anatomy of the Ultra-Secret National Security Agency . Anchor Books. Baranski, J. V. (2007). Fatigue, sleep loss, and confidence in judgment. Journal Of Experimental Psychology Applied , 13 (4), 182–196. https://doi.org/10.1037/1076-898x.13.4.182 Bates, D., Maechler, M., Bolker, B., & Walker, S. (2015, 7 april). CRAN - Package lme4 . Comprehensive R Archive Network (CRAN). Geraadpleegd op 6 mei 2022, van https://cran.r-project.org/web/packages/lme4/index.html Blagrove, M. (1996). Effects of length of sleep deprivation on interrogative suggestibility. Journal of Experimental Psychology: Applied , 2 (1), 48–59. https://doi.org/10.1037/1076-898x.2.1.48 Blagrove, M., & Akehurst, L. (2000). Effects of sleep loss on confidence–accuracy relationships for reasoning and eyewitness memory. Journal of Experimental Psychology: Applied , 6 (1), 59–73. https://doi.org/10.1037/1076-898x.6.1.59 Brysbaert, M., & Stevens, M. C. (2018). Power Analysis and Effect Size in Mixed Effects Models: A Tutorial. Journal of Cognition , 1 (1). https://doi.org/10.5334/joc.10 Buysse, D. J., Reynolds, C. F., 3rd, Monk, T. H., Berman, S. R., & Kupfer, D. J. (1989). The Pittsburgh Sleep Quality Index: a new instrument for psychiatric practice and research. Psychiatry Res, 28 (2), 193–213. Dickinson, D. L., & Drummond, S. P. (2008). The effects of total sleep deprivation on bayesian updating. Judgment and Decision Making , 3 (2), 181–190. https://doi.org/10.1017/s1930297500001522 Ellis, B. W., Johns, M. W., Lancaster, R., Raptopoulos, P., Angelopoulos, N., & Priest, R. G. (1981). The St. Mary's Hospital sleep questionnaire: a study of reliability. Sleep , 4(1), 93–97. Frenda, S. J., Berkowitz, S. S., Loftus, E. F., & Fenn, K. M. (2016). Sleep deprivation and false confessions. Proceedings of the National Academy of Sciences of the United States of America , 113 (8), 2047–2050. https://doi.org/10.1073/pnas.1521518113 Gradassi, A., Van Den Bos, W., & Molleman, L. (2022). Confidence of others trumps confidence of self in social information use. PsyArXiv . https://doi.org/10.31234/osf.io/mqyu2 Harrison, Y., & Horne, J. (1999). One Night of Sleep Loss Impairs Innovative Thinking and Flexible Decision Making. Organizational Behavior and Human Decision Processes , 78 (2), 128–145. https://doi.org/10.1006/obhd.1999.2827 Landolt, H., Holst, S. C., Sousek, A., Bassetti, C., Dogas, Z., & Peigneux, P. (2014). Effects of acute and chronic sleep deprivation. In Sleep Medicine Textbook: Second Edition (Vols. 67–80, pp. 49–61). https://doi.org/10.5167/uzh-107182 Horne, J., & Östberg, O. (1976). A self-assessment questionnaire to determine morningness- eveningness in human circadian rhythms. International Journal of Chronobiology , 4 (2), 97–110. https://ci.nii.ac.jp/naid/20001543451 Johnson, D. J., Stepan, M. E., Cesario, J., & Fenn, K. M. (2020). Sleep Deprivation and Racial Bias in the Decision to Shoot: A Diffusion Model Analysis. Social Psychological And Personality Science , 12 (5), 638–647. https://doi.org/10.1177/1948550620932723 Kappes, A., Harvey, A. K., Lohrenz, T., Montague, P. R., & Sharot, T. (2019). Confirmation bias in the utilization of others’ opinion strength. Nature Neuroscience , 23 (1), 130–137. https://doi.org/10.1038/s41593-019-0549-2 Kassin, S. M., & Kiechel, K. L. (1996). The Social Psychology of False Confessions: Compliance, Internalization, and Confabulation. Psychological Science , 7 (3), 125–128. https://doi.org/10.1111/j.1467-9280.1996.tb00344.x Killgore, W. D. (2010). Effects of sleep deprivation on cognition. In Progress in Brain Research (pp. 105–129). Elsevier BV. https://doi.org/10.1016/b978-0-444-53702-7.00007-5 Killgore, W. D., Killgore, D. B., Day, L., Li, C. I., Kamimori, G. H., & Balkin, T. J. (2007). The Effects of 53 Hours of Sleep Deprivation on Moral Judgment. Sleep , 30 (3), 345–352. https://doi.org/10.1093/sleep/30.3.345 Klayman, J. (1995). Varieties of Confirmation Bias. In The Psychology of learning and motivation (pp. 385–418). https://doi.org/10.1016/s0079-7421(08)60315-1 Lefebvre, G., Summerfield, C., & Bogacz, R. (2021). A Normative Account of Confirmation Bias During Reinforcement Learning. Neural Computation , 34 (2), 307–337. https://doi.org/10.1162/neco_a_01455 Lenth, R. V. (2019, 27 maart). CRAN - Package emmeans . emmeans: estimated marginal means, aka least-squared means. Geraadpleegd op 6 mei 2022, van https://cran.r- project.org/web/packages/emmeans/index.html Nickerson, R. (1998). Confirmation Bias: A Ubiquitous Phenomenon in Many Guises. Review of General Psychology , 2 (2), 175–220. https://doi.org/10.1037/1089-2680.2.2.175 OSF . Anckaert & Gevers, 2023. https://doi.org/10.17605/OSF.IO/MS2ZW Pescetelli, N., Hauperich, A., & Yeung, N. (2021). Confidence, advice seeking and changes of mind in decision making. Cognition , 215 , 104810. https://doi.org/10.1016/j.cognition.2021.104810 Pescetelli, N., & Yeung, N. (2020). The effects of recursive communication dynamics on belief updating. Proceedings Of The Royal Society B Biological Sciences , 287 (1931), 20200025. https://doi.org/10.1098/rspb.2020.0025 Poinas, F., Rosaz, J., & Roussillon, B. (2010). Updating Beliefs with Imperfect Signals: Experimental Evidence. Social Science Research Network . https://doi.org/10.2139/ssrn.1731549 R Core Team (2012). R: A Language and Environment for Statistical Computing . Vienna: R foundation for Statistical Computing. http://www.R-project.org/ Rollwage, M., & Fleming, S. M. (2021). Confirmation bias is adaptive when coupled with efficient metacognition. Philosophical Transactions of the Royal Society B , 376 (1822), 20200131. https://doi.org/10.1098/rstb.2020.0131 Rollwage, M., Loosen, A. M., Hauser, T. U., Moran, R., Dolan, R. J., & Fleming, S. M. (2020). Confidence drives a neural confirmation bias. Nature Communications , 11 (1). https://doi.org/10.1038/s41467-020-16278-6 Rosner, R. I. (2015). Beck Depression Inventory (BDI). New Jersey: John Wiley & Sons, Inc. https://doi.org/10.1002/9781118625392.wbecp261 Sommer, J., Musolino, J., & Hemmer, P. (2023). Updating, evidence evaluation, and operator availability: A theoretical framework for understanding belief. Psychological Review, 131 (2), 373–401. https://doi.org/10.1037/rev0000444 WGT3X-BT | ActiGraph Wearable Devices . https://theactigraph.com/actigraph-wgt3x-bt Zenses, A., Lenaert, B., Peigneux, P., Beckers, T., & Boddez, Y. (2019). Sleep deprivation increases threat beliefs in human fear conditioning. Journal of Sleep Research , 29 (3). https://doi.org/10.1111/jsr.12873 Additional Declarations There is NO Competing Interest. 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-5289682","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":367768216,"identity":"29ce7124-c89a-48dd-ade9-af1bf7eeb83f","order_by":0,"name":"Charlotte Anckaert","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABCElEQVRIiWNgGAWjYDACCSB+AOcdABHMBxgeNjAk4NWSwGAggdDCxpbAkEiiFh4DvFr4Zzc/+5BQ8aeOn4H52YMfZ+zkzOf3fJNI3MGQx4/LkjvHjGcknDGQkGxgMzfsuZFsLHOMd5tE4hmGYskG7FoMJBKMGRLbDCQMDjCYSTN8YE6cwQbS0saQuOEALi3pn8Fa7A+wfwNqqa+fwcbzDKxlP04tOVBbGHiAttw4nCDBxsMGsQWXX27kFDMknDGWnHGYp0yy58xxwxlsacYWiWckiiVw2MI/I30zw4cKOX7+9vZtEj+OVctLMB9+eOPjDps8fhzeRwBmNOsJqR8Fo2AUjIJRgAcAALXvVlwA1vUSAAAAAElFTkSuQmCC","orcid":"https://orcid.org/0000-0001-9503-5940","institution":"Université libre de Bruxelles","correspondingAuthor":true,"prefix":"","firstName":"Charlotte","middleName":"","lastName":"Anckaert","suffix":""},{"id":367768217,"identity":"484bd3ac-8208-4a0d-b615-630a98abc4b6","order_by":1,"name":"Philippe Peigneux","email":"","orcid":"https://orcid.org/0000-0003-4745-1434","institution":"Universite Libre de Bruxelles (ULB)","correspondingAuthor":false,"prefix":"","firstName":"Philippe","middleName":"","lastName":"Peigneux","suffix":""},{"id":367768218,"identity":"218c234e-edfd-4ea0-bd9c-a7b282d7c40a","order_by":2,"name":"Wim Gevers","email":"","orcid":"","institution":"Université libre de Bruxelles","correspondingAuthor":false,"prefix":"","firstName":"Wim","middleName":"","lastName":"Gevers","suffix":""}],"badges":[],"createdAt":"2024-10-18 13:15:19","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-5289682/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-5289682/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":67117230,"identity":"d7ed2f28-e714-43eb-ae0b-2b0f1beae45b","added_by":"auto","created_at":"2024-10-21 10:42:11","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":46143,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eTimeline.\u003c/strong\u003eIn both SD and RS conditions, participants wore an actimeter starting from Day1 and filled in a questionnaire probing their sleep quality starting from Day2. On Day3 in the evening in each condition, participants were administered the Phase 1 task in which they had to decide about the actual price of a shown property and provide their confidence rating. Half of the participants went home to sleep after Phase 1, the other half stayed in the lab for a supervised SD night. On Day 4 in the morning, participants were administered the Phase 2 task in which they were informed on the opinion of the partner and had opportunity to change their belief and confidence rating. There was a one-week interval between the two conditions (counterbalanced across participants).\u003c/p\u003e","description":"","filename":"image1.png","url":"https://assets-eu.researchsquare.com/files/rs-5289682/v1/9a45da0e562a24d423896caf.png"},{"id":67117233,"identity":"17c6bce1-c2ab-43ec-b57f-c6d8dd0c459d","added_by":"auto","created_at":"2024-10-21 10:42:11","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":166689,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eTask.\u003c/strong\u003eIn Phase 1 (evening session), participants were shown the image of a property with the tentative price of that property shown beneath the image. The participants’ task was to decide if the real asking price was higher or lower than the price on screen. Participants entered their choice by pressing the “up” or “down” arrow on the keyboard. Thereafter, they rated their confidence using the mouse, on a scale ranging from 0 to 100. In Phase 2 (morning session), participants were presented with their original choice and confidence level, along with the choice of their so-called partner. Participants could update both their confidence level and change their belief. All participants performed this task in a SD and RS condition, with a one-week interval, in a counterbalanced order. In this example, the partner disagrees with the participant. The participant changed beliefs and rated a higher confidence level compared to the initially indicated confidence level.\u003c/p\u003e","description":"","filename":"image2.png","url":"https://assets-eu.researchsquare.com/files/rs-5289682/v1/a20e02f7209e3b92577c26e9.png"},{"id":67117235,"identity":"12b08fde-5240-4ab7-8f71-c27b54053024","added_by":"auto","created_at":"2024-10-21 10:42:12","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":82057,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003ePredicted Probabilities of Belief Changes. \u003c/strong\u003ePredicted probability of belief change as a function of initial confidence under disagreement conditions, comparing RS and SD. The left panel displays the predicted probability of belief change for participants who had a normal night of sleep, while the right panel shows the same predictions for sleep-deprived participants. The shaded areas represent 95% confidence intervals. Both panels show a negative relationship between initial confidence and the likelihood of belief change, with lower confidence levels associated with a higher probability of belief change. However, SD increases the overall likelihood of belief change.\u003c/p\u003e","description":"","filename":"image3.png","url":"https://assets-eu.researchsquare.com/files/rs-5289682/v1/05538287d38ebb96caf55e3e.png"},{"id":67117236,"identity":"41535407-9677-4a08-b05e-df7de6573391","added_by":"auto","created_at":"2024-10-21 10:42:12","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":41954,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eEstimated Marginal Means of Confidence Update.\u003c/strong\u003e The estimated marginal means of confidence update across Partner Information and Sleep Conditions. Confidence increased when in agreement and decreased when in disagreement. This confidence update was modulated by sleep condition. The increase when in agreement was larger after RS compared to after SD. The decrease when in disagreement was smaller after RS compared to after SD. This results in SD eliminating confirmation bias. \u0026nbsp;The error bars indicate the standard error of the within-subject means, and asterisks indicate significant pairwise comparisons (*\u003cem\u003ep\u003c/em\u003e \u0026lt; .05, ***\u003cem\u003ep\u003c/em\u003e \u0026lt; .001).\u003c/p\u003e","description":"","filename":"image4.png","url":"https://assets-eu.researchsquare.com/files/rs-5289682/v1/c7c07fd6664bcb52a597845d.png"},{"id":67117232,"identity":"7444703c-a3f0-4237-9364-dc9a0f37933f","added_by":"auto","created_at":"2024-10-21 10:42:11","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":284029,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eTask.\u003c/strong\u003e The top panel illustrates the \"belief change allowed\" condition. In this condition, participants were presented with the peer's opinion regarding property prices and had the opportunity to revise their initial belief by pressing the up or down arrow. After confirming their belief change (if any), they proceeded to rate their confidence in this revised belief. The bottom panel illustrates the \"no belief change allowed\" condition. Here, participants also viewed the partner’s opinion, but unlike the previous condition, they were instructed to press the space bar to continue to the next stage. They were not given the option to change their belief and could only proceed to the confidence rating screen, where they rated how confident they were in their initial belief without having the opportunity to change it. In the NCA condition in this example (bottom), the partner disagrees with the participant. As a response, the participant decreased confidence from 45 to 12.\u003c/p\u003e","description":"","filename":"image5.png","url":"https://assets-eu.researchsquare.com/files/rs-5289682/v1/f3ad2cafd14695fc50928130.png"},{"id":67117231,"identity":"e35c528d-5dec-49ae-887a-babb8ac3b4a7","added_by":"auto","created_at":"2024-10-21 10:42:11","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":82738,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003ePredicted Probabilities of Belief Changes in Study 2. \u003c/strong\u003ePredicted probability of belief change as a function of initial confidence under disagreement conditions, comparing RS and SD. Like Study 1, disagreement, low initial confidence, and SD all increased the likelihood of belief changes. The shaded areas represent 95% confidence intervals.\u003c/p\u003e","description":"","filename":"image6.png","url":"https://assets-eu.researchsquare.com/files/rs-5289682/v1/975ace5196f0a3d5817b85b1.png"},{"id":67117972,"identity":"8cbf9c44-e279-4d48-8ce6-e36cae308d69","added_by":"auto","created_at":"2024-10-21 10:50:11","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":55083,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eEstimated marginal means of confidence update based on Partner Information, Change Condition, and Sleep Condition. \u003c/strong\u003eThe x-axis categorizes partner information into \"agreement\" and \"disagreement,\" while the y-axis measures the estimated marginal mean of confidence update. The error bars indicate the standard error of the within-subject means. Confidence updates are affected by the ability to change beliefs, the nature of the partner's information, and sleep condition. The largest drop in confidence is observed after disagreement, when a belief change was not allowed, under SD.\u003c/p\u003e","description":"","filename":"image7.png","url":"https://assets-eu.researchsquare.com/files/rs-5289682/v1/f529fc9f95a44485dce759bc.png"},{"id":68847513,"identity":"08251453-bb50-4c05-a458-e675beb2824e","added_by":"auto","created_at":"2024-11-12 16:20:12","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1250118,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5289682/v1/8f0cc8ff-f935-45b5-a4b8-91192557db35.pdf"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"Sleep Deprivation Increases Belief Change and Suppresses Confirmation Bias in Decision Tasks","fulltext":[{"header":"Introduction","content":"\u003cp\u003eSleep deprivation (SD) is a common condition among high-stakes decision-makers across various fields, including military, healthcare, and corporate environments (e.g., Associated Press, 2020; Bamford, 2001). SD is known to impair cognition and decision-making (Killgore et al., 2007; Landolt et al., 2014), potentially leading to altered behaviours in high-stakes situations (Killgore, 2010). Despite extensive research on the cognitive effects of SD \u0026ndash; including suggestibility (Blagrove \u0026amp; Akehurst, 2000), increased threat perception (Zenses et al., 2019), and other altered cognitive processes (Kassin \u0026amp; Kiechel, 1996) \u0026ndash; its specific impact on belief updating remains underexplored (Achtziger et al., 2014; Poinas et al., 2010).\u003c/p\u003e \u003cp\u003eBelief updating is a critical process for accurate understanding and informed decision-making in modern information-rich societies (Sommer et al., 2023). SD, known to impair cognitive function and reduce motivation to reconcile inconsistencies (Blagrove, 1996), may hinder effective belief updating, particularly in individuals with decisional power like policymakers (Frenda et al., 2016). This impaired motivation could manifest in two contrasting ways: sleep-deprived individuals might either adopt new beliefs without critical evaluation (the \"follow along\" hypothesis) or rigidly adhere to their existing beliefs, downplaying the significance of new information (the \"rigid\" hypothesis, Dickinson \u0026amp; Drummond, 2008; Harrison \u0026amp; Horne, 1999).\u003c/p\u003e \u003cp\u003eBelief updating, the modification of beliefs in response to new information, is a cognitive process well known to be influenced by confirmation bias, where individuals tend to favour information that supports their existing beliefs (Nickerson, 1998). Given its prominent role in belief updating, confirmation bias is a suited mechanism to explore how altered states of vigilance, such as SD, might influence this process. SD could either amplify confirmation bias, leading to increased rigidity and fewer belief changes (supporting the rigid hypothesis, Harrison \u0026amp; Horne, 1999), or diminish it, resulting in more belief changes in response to disconfirming information (supporting the follow-along hypothesis, Blagrove, 1996; Frenda et al., 2016).\u003c/p\u003e \u003cp\u003eThe current studies examine how SD affects two key aspects of belief updating: belief change and confidence update. These markers, which measure the extent to which individuals modify their beliefs and adjust their confidence in those beliefs, have been widely studied under normal conditions (e.g., Lefebvre et al., 2021; Rollwage et al., 2020; Rollwage \u0026amp; Fleming, 2021). Typically, belief change is associated with a significant drop in confidence (Pescetelli et al., 2021), and confidence increases when new information aligns with existing beliefs but decreases when it contradicts them.\u003c/p\u003e \u003cp\u003eThe present studies address these questions through two experiments using a binary decision task related to real estate property prices, adapted from Kappes et al. (2019). In this task, participants must assess their level of confidence in their belief, and receive feedback from a partner, who agreed or disagreed with their belief. In Study 1, participants could maintain or change their initial belief and update confidence in their final belief. Study 2 replicated Study 1 and was extended with a condition where participants could only update confidence in their belief, but no longer change it. We measured the confidence update and the frequency of belief changes. Confidence update is measured as the difference between the final (i.e., after incorporating the partner\u0026rsquo;s information) and initial confidence rating (i.e., confidence in the first belief about the price of a property). A positive confidence update reflects an increase in confidence. Belief change, on the other hand, is the act of switching from one to the other belief after incorporating the partner\u0026rsquo;s information. The task was performed twice in a counterbalanced within-subjects design: all participants performed the task once after a night of normal sleep and once after a SD night. With these experiments, we aim at clarifying how and to what extent SD impacts belief change frequency and confidence updating. If SD amplifies confirmation bias (the rigid hypothesis), we expect to see fewer belief changes and smaller confidence updates, particularly after receiving disconfirming information (Dickinson \u0026amp; Drummond, 2008; Harison \u0026amp; Horne, 1999). Alternatively, if SD diminishes confirmation bias (the follow-along hypothesis), we expect to see more frequent belief changes and larger confidence updates (Blagrove, 1996; Frenda et al., 2016).\u003c/p\u003e"},{"header":"Study 1","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eMethods\u003c/h2\u003e \u003cdiv id=\"Sec4\" class=\"Section3\"\u003e \u003ch2\u003eParticipants\u003c/h2\u003e \u003cp\u003e36 participants aged 19\u0026ndash;25 years (M\u003csub\u003eage\u003c/sub\u003e = 21.25, SD\u003csub\u003eage\u003c/sub\u003e = 1.48, 13 female) were recruited through social media announcements and flyers. We aimed for a sample size of 32 based on the recommendations by Brysbaert and Stevens (2018) for mixed model analysis for a predicted group effect. Inclusion criteria were satisfying usual sleep quality (Pittsburgh Sleep Quality Index [PSQI]\u0026thinsp;\u0026lt;\u0026thinsp;7, Buysse et al., 1989), a score\u0026thinsp;\u0026lt;\u0026thinsp;7 at the shortened version of the Beck Depression Inventory (BDI, Rosner, 2015), and moderate or neutral chronotype (30\u0026thinsp;\u0026lt;\u0026thinsp;\u003cem\u003escore\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;70) on the morningness-eveningness questionnaire (Horne \u0026amp; \u0026Ouml;stberg, 1976). Participants received monetary compensation and gave their informed consent to participate in this experiment approved by the local Ethical Committee (agreement #684/2022) of the Faculty of Psychology and Educational Sciences of the Universit\u0026eacute; libre de Bruxelles.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e\n\u003ch3\u003eProcedure\u003c/h3\u003e\n\u003cp\u003eTo increase power and control for interindividual variability in susceptibility to belief updating and influences of SD, we used a within-subjects design. All participants were randomly assigned to the SD or normal/regular sleep (RS) condition as a first session, followed by the other condition in a session held one week later (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). At each session, both phases of the experiment lasted about 50 minutes each (see task details below). From Day1 to the end of the experiment, participants wore a wrist actimeter (wGT3XBT Monitor, ActiGraph, Pensacola, FL, USA) and filled in every morning a short questionnaire about their sleep quality on the previous night (except after the SD night, Ellis et al., 1981). Visual inspection of actimetric recordings was used in conjunction with sleep questionnaires to check the regularity of the sleep-wake cycle during the entire experiment. In the SD condition, participants stayed in the laboratory from ca. 8pm (after they finished Phase 1) to ca. 9am (after they finished Phase 2). They stayed awake the entire night under the supervision of two experimenters and were allowed to engage in quiet activities (e.g., reading, listening to music). High-arousal physical activities were not allowed. Light meals were offered throughout the night with continuous water access. The Karolinska Sleepiness Scale (KSS; \u0026Aring;kerstedt and Gillberg, 1990) was filled in every hour during the SD night to gauge the evolution of drowsiness. In the RS condition, participants spent a night of sleep at home in between Phase 1 and Phase 2. Smoking, alcohol, and energy drinks were not allowed on experimental days.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e\n\u003ch3\u003eTask\u003c/h3\u003e\n\u003cp\u003eThe design is adapted from the Kappes et al.\u0026rsquo;s (2019) study experimental design, consisting of a belief induction phase (Phase 1) followed by a belief updating phase (Phase 2). Both phases of the study consisted of five practice trials followed by 200 experimental trials (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). In Phase 2, information from another participant (namely, the pseudo-partner) was presented during the task. This information was pseudo-randomly generated, and data was collected in groups of minimum six subjects to create the impression that the partner's information was obtained from a genuine participant. During Phase 1 (Day3 evening in each condition), real estate images were depicted on screen, together with price indications taken from a Brussels-based real estate website. Since all participants underwent both conditions, we developed two databases of Brussels properties. Participants saw images of real estate properties, with prices manipulated to be either 20% higher or 20% lower than the actual market value. Using the keyboard, participants indicated their belief about whether the real price of the houses was higher or lower than the indicated price by pressing the up or down arrow, then had to provide their confidence rating in their decision using a slider on a 0-100 scale (Initial Confidence). In Phase 2 (Day4 morning), participants saw the same set of real estate images, with identical prices as in the first session. They also saw their initial opinion and level of confidence, as well as their pseudo-partner's opinion (but not their confidence, at variance with Kappes et al., 2019) for each property, resulting in agreement trials (i.e., the partner has the same opinion) and disagreement trials (i.e., the partner has the opposite opinion) with an equal number of trials for each. Using the up or down arrow (see Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e) participants could then modify their initial decision (Belief Change) while seeing the new information on the screen, and then enter their confidence level on this second opinion (Confidence Update) on a 0-100 scale. Confidence Update was measured as the difference between the final and initial confidence rating of the participant.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003eData analysis\u003c/h2\u003e \u003cp\u003eStatistical analyses were conducted using R v4.4.1 (R Core Team, 2012). Belief Change and Confidence Update were statistically analysed using (generalised) linear mixed-effects models using the lme4 package v1.1-28 (Bates et al., 2015). Belief Change refers to the probability of changing beliefs on all trials. Confidence Update was calculated as the difference between the Final and Initial Confidence reported (final minus initial). A positive Confidence Update indicates an increase in confidence. Initial Confidence rating was controlled for, as participants with lower initial confidence have more room to increase their confidence after agreement and less room to decrease after disagreement, and vice versa for those with higher initial confidence. Planned comparisons (one-sided t-tests) were done using the emmeans package v1.7.2 (Lenth, 2019); false discovery rate (FDR) \u003cem\u003ep\u003c/em\u003e-values are reported.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eResults\u003c/h3\u003e\n\u003cdiv id=\"Sec9\" class=\"Section2\"\u003e \u003ch2\u003eBelief change\u003c/h2\u003e \u003cp\u003eA generalized linear mixed-effects model was used to investigate the influence of Initial Confidence (ranging between 0 and 100), Partner Information (agreement/ disagreement), and Sleep Condition (RS/ SD) on the probability of belief changes. The model included fixed effects for Initial Confidence, Partner Information, and Sleep Condition, as well as their interaction. A random intercept and slopes for Partner Information and Sleep Condition were specified for participants to account for individual differences. The model was fitted using the Laplace approximation with a binomial family and a logit link function.\u003c/p\u003e \u003cp\u003eThe analysis of the model revealed significant main effects of Initial Confidence, Partner Information, and Sleep Condition on the probability of belief change. Higher initial confidence level was associated with a lower probability of belief change (Estimate = -0.02, z = -16.83, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;.001). Not surprisingly, trials involving disagreement information increased the probability of belief change compared to agreement trials (Estimate\u0026thinsp;=\u0026thinsp;1.75, z\u0026thinsp;=\u0026thinsp;7.26, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;.001). Importantly, being sleep deprived strongly increased the probability of belief change compared to the RS condition (Estimate\u0026thinsp;=\u0026thinsp;0.74, z\u0026thinsp;=\u0026thinsp;4.13, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;.001). Specifically, the likelihood of belief changes in the RS condition was 5.4%, whereas in the sleep-deprived condition it increased to 11.6%. These results indicate that SD strongly increases the likelihood of participants changing their beliefs. The interaction effect between Partner Information and Sleep Condition was not observed (Estimate\u0026thinsp;=\u0026thinsp;0.18, z\u0026thinsp;=\u0026thinsp;1.33, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.18), suggesting that the increase in frequency of belief changes is similar for agreement and disagreement trials when sleep deprived (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). In conclusion, lower initial confidence, facing disagreement, and being sleep deprived all increase the probability of changing beliefs.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eConfidence Update\u003c/h3\u003e\n\u003cp\u003eA linear mixed-effects model was used to examine the effects of Partner Information and Sleep Condition on the Confidence Update. The model included fixed effects for Initial Confidence, Partner Information, and Sleep Condition, and their interaction, as well as a random intercept per participant and slopes for Partner Information and Sleep Condition to account for individual variability. The model was fitted using the Bound Optimization by Quadratic Approximation (\u0026ldquo;bobyqa\u0026rdquo;) optimizer with a maximum of 5000 function evaluations.\u003c/p\u003e \u003cp\u003eThe linear mixed-effects model revealed several significant effects on Confidence Update, controlled for Initial Confidence. Partner Information showed a significant main effect (χ2(1)\u0026thinsp;=\u0026thinsp;47.81, \u003cem\u003eβ\u003c/em\u003e = \u0026minus;11.24, SE\u0026thinsp;=\u0026thinsp;1.63, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;.001), with disagreement trials associated with a more negative confidence update compared to agreement trials. Sleep Condition also had a significant main effect (χ2(1)\u0026thinsp;=\u0026thinsp;17.83, \u003cem\u003eβ\u003c/em\u003e = \u0026minus;4.12, SE\u0026thinsp;=\u0026thinsp;0.98, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;.001), with participants in the SD condition showing a smaller confidence update compared to when they did the RS condition. The interaction between Sleep Condition and Partner Information (χ2(1)\u0026thinsp;=\u0026thinsp;11.16, \u003cem\u003eβ\u003c/em\u003e\u0026thinsp;=\u0026thinsp;1.84, SE\u0026thinsp;=\u0026thinsp;0.55, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;.001) further illustrates the overall lower confidence after SD: on agreement trials, confidence increased less when sleep deprived compared to after RS (EMM\u003csub\u003esleep deprived\u003c/sub\u003e = 3.62, EMM\u003csub\u003enormal sleep\u003c/sub\u003e = 7.74, : \u003cem\u003et\u003c/em\u003e(35)\u0026thinsp;=\u0026thinsp;4.22, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;.001), while on disagreement trials, confidence decreased more in SD compared to RS (EMM\u003csub\u003esleep deprived\u003c/sub\u003e = -5.78, EMM\u003csub\u003enormal sleep\u003c/sub\u003e = -3.50, : \u003cem\u003et\u003c/em\u003e(35)\u0026thinsp;=\u0026thinsp;2.34, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;.05). As can be seen in Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e, the interaction reveals that confirmation bias is eliminated by SD. A stronger increase in confidence after agreement compared to the decrease after disagreement is only observed after RS (\u003cem\u003et\u003c/em\u003e(35)\u0026thinsp;=\u0026thinsp;6.91 \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;.001).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003eInterim discussion\u003c/h2\u003e \u003cp\u003eThe findings from this first study reveal significant insights into the cognitive processes underlying belief updating and decision-making, particularly under SD influence. Our research indicates that SD strongly increases the likelihood of belief change. This suggests that a lack of sleep significantly alters the stability of one's beliefs. This increased frequency of belief changes appears to be intricately linked to lower confidence levels when individuals are sleep deprived. A Welch two-samples t-test showed that there was a significant difference in mean confidence between the two sleep conditions, suggesting that confidence is significantly higher after RS compared to after SD (M \u003csub\u003eregular sleep\u003c/sub\u003e = 65.39, M \u003csub\u003esleep deprivation\u003c/sub\u003e = 60.02, t(35)\u0026thinsp;=\u0026thinsp;13.3, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;.001). As confidence wanes, people seem more prone to reconsider their beliefs, highlighting the role of confidence as a critical mediator in belief stability (e.g., Gradassi et al., 2022). Our results also suggest that even under SD, a rational decision-making process persists: initial confidence levels continue to serve as a key factor influencing belief change. When participants enter a decision-making scenario with low initial confidence, they are more likely to change their beliefs, and this remains true when sleep deprived. This finding underscores the robustness of confidence as a determinant of belief stability, suggesting that it operates consistently across varying states of alertness.\u003c/p\u003e \u003cp\u003eA second intriguing aspect of Study 1 is that confirmation bias, although present in well-rested individuals, disappears when they are sleep deprived. Under RS conditions, participants exhibited a tendency to increase their confidence more after receiving confirming information than they decreased it after receiving disconfirming information, demonstrating classic confirmation bias (Klayman, 1995). However, when deprived of sleep, this bias vanished. This suggests that SD may impair the cognitive mechanisms that typically drive confirmation bias, possibly leading to a more balanced evaluation of evidence, or simply to a different decision-making strategy when under the cognitive strain of SD.\u003c/p\u003e \u003cp\u003eWhen receiving agreement information, the increase in confidence is weaker after SD. When receiving disconfirming information, the decrease in confidence is stronger after SD. Both these tendencies align well with the finding of a lower overall confidence after SD. However, the difference in decrease in confidence between RS and SD after receiving disconfirming information is less pronounced and borderline significant.\u003c/p\u003e \u003cp\u003eA possible explanation for this phenomenon is that when sleep-deprived, participants may be less inclined to significantly lower their confidence after receiving disconfirming information because doing so triggers a belief change. In this context, the act of changing one's mind may be seen as a more adaptive response than sticking with an opinion that leads to a substantial decrease in confidence. Essentially, rather than holding onto a belief that suddenly seems much less certain, individuals may prefer to revise their belief altogether, thereby avoiding the discomfort associated with lowered confidence.\u003c/p\u003e \u003cp\u003eWe pre-registered this hypothesis prior to conducting the follow-up study (OSF, Anckaert \u0026amp; Gevers, 2023). We designed an experiment aimed at disentangling the relationship between belief change and confidence update under RS and SD conditions. In this study, we introduced a condition where participants are allowed only to update their confidence levels without giving them the option to change their beliefs. We hypothesised that confidence would decrease stronger after receiving disagreement information under SD and a belief change was not allowed compared to under RS, and compared to when a belief change was allowed.\u003c/p\u003e \u003c/div\u003e "},{"header":"Study 2","content":" \u003cdiv id=\"Sec13\" class=\"Section3\"\u003e \u003ch2\u003eMethods\u003c/h2\u003e \u003cdiv id=\"Sec14\" class=\"Section4\"\u003e \u003ch2\u003eParticipants\u003c/h2\u003e \u003cp\u003e50 new participants aged 19\u0026ndash;29 years (M\u003csub\u003eage\u003c/sub\u003e = 22.19, SD\u003csub\u003eage\u003c/sub\u003e = 2.09, 38 female) were recruited through social media announcements and flyers. Inclusion criteria and procedure were the same as Study 1. One participant failed to show up for the normal sleep condition and was therefore excluded. This follow-up experiment was conducted under approval of the local Ethical Committee (agreement #684/2022) of the Faculty.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003eTask\u003c/h2\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003ePhase 1 was identical in both Studies. Phase 2 partially differed: it consisted of two blocks of 100 trials. One block was identical to the second phase of Study 1, meaning that participants could change or maintain their initial belief while seeing the new information on the screen and update confidence level in their second opinion (Change Allowed block, CA). In the other, new block, participants could update their confidence but were not given the option to change their opinion (Fig.\u0026nbsp;5, No Change Allowed block, NCA). All participants performed both blocks in counterbalanced order\u003ca class=\"FNLink\" href=\"#Fn1\" id=\"#FNLinkFn1\"\u003e1\u003c/a\u003e.\u003c/p\u003e\u003c/div\u003e\n\u003ch3\u003eResults\u003c/h3\u003e\n\u003cdiv id=\"Sec17\" class=\"Section2\"\u003e \u003ch2\u003eChange Allowed block: Belief Change\u003c/h2\u003e \u003cp\u003eWe used the same model as in Study 1 to explore how Partner Information, Sleep Condition, and their interaction affected Belief Change, while controlling for Initial Confidence. This model also included a random intercept for each participant and allowed the effects of Partner Information and Sleep Condition to vary by individual. The results of Study 1 were fully replicated. A significant negative effect of Initial Confidence was found (Estimate\u0026thinsp;=\u0026thinsp;\u0026minus;\u0026thinsp;0.03, z\u0026thinsp;=\u0026thinsp;\u0026minus;\u0026thinsp;15.35, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;.001), indicating that higher confidence was associated with lower odds of belief change. There was a main effect of Partner Information, with disagreement trials associated with higher odds of belief change compared to agreement trials (Estimate\u0026thinsp;=\u0026thinsp;2.45, z\u0026thinsp;=\u0026thinsp;9.27, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;.001). The main effect of Sleep Condition was observed again. SD associated with higher odds of changing beliefs (15.7%) compared to RS (10.9%, estimate\u0026thinsp;=\u0026thinsp;0.532, z\u0026thinsp;=\u0026thinsp;1.99, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;.047). Like Study 1, the interaction between Partner Information and Sleep Condition was not observed (Estimate\u0026thinsp;=\u0026thinsp;\u0026minus;\u0026thinsp;0.23, z\u0026thinsp;=\u0026thinsp;\u0026minus;\u0026thinsp;1.21, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;.23).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec18\" class=\"Section2\"\u003e \u003ch2\u003eChange Allowed vs. No Change Allowed block: Confidence Update\u003c/h2\u003e \u003cp\u003eWe used a linear mixed-effects model to investigate the influence of having the possibility to change beliefs on confidence updating. We assessed the influence of Partner Information, Sleep Condition, and Change Condition (CA vs. NCA) on Confidence Update by including a main effect for Initial Confidence (to control for its influence) and a three-way interaction effect between Partner Information, Sleep Condition, and Change Condition. The model also included a random slope for Partner Information and Sleep Condition and a random intercept per participant.\u003c/p\u003e \u003cp\u003eThe analysis revealed a significant three-way interaction effect (χ2(7)\u0026thinsp;=\u0026thinsp;234.92, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;.001). Figure\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e7\u003c/span\u003e illustrates eight key comparisons, with statistical significance highlighted for clarity.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec19\" class=\"Section2\"\u003e \u003ch2\u003eWithin-Group Comparisons: Sleep Condition and Partner Information\u003c/h2\u003e \u003cdiv id=\"Sec20\" class=\"Section3\"\u003e \u003ch2\u003e\u0026bull; Change Allowed (Left Panel):\u003c/h2\u003e \u003cp\u003eReplicating Study 1, after a RS night, confirmation bias was observed. Participants showed a significant larger increase in confidence after agreement (EMM\u0026thinsp;=\u0026thinsp;9.64) compared to the decrease after disagreement (EMM = -1.89), \u003cem\u003et\u003c/em\u003e(48)\u0026thinsp;=\u0026thinsp;7.39, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;.001. SD did not significantly affect the confidence increase after agreement compared to RS. However, the confidence decrease following disagreement was larger under SD (EMM = -4.57) compared to RS (EMM = -1.89), \u003cem\u003et\u003c/em\u003e(48)\u0026thinsp;=\u0026thinsp;2.59, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;.016.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec21\" class=\"Section2\"\u003e \u003ch2\u003e\u0026bull; No Change Allowed (Right Panel):\u003c/h2\u003e \u003cp\u003eSimilarly, under RS, participants exhibited a confirmation bias, with a significant larger confidence increase after agreement (EMM\u0026thinsp;=\u0026thinsp;10.22) compared to the decrease after disagreement (EMM = -3.62), \u003cem\u003et\u003c/em\u003e(48)\u0026thinsp;=\u0026thinsp;8.88, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;.001. Under SD, participants increased their confidence less after agreement (EMM\u0026thinsp;=\u0026thinsp;7.78) compared to RS (EMM\u0026thinsp;=\u0026thinsp;10.22), \u003cem\u003et\u003c/em\u003e(48)\u0026thinsp;=\u0026thinsp;2.35, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;.028. The confidence decrease after disagreement was more pronounced after SD (EMM = -8.77) than after RS (EMM = -3.62), \u003cem\u003et\u003c/em\u003e(48)\u0026thinsp;=\u0026thinsp;5.01, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;.001.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec22\" class=\"Section2\"\u003e \u003ch2\u003eBetween-Group Comparisons: Change Conditions\u003c/h2\u003e \u003cdiv id=\"Sec23\" class=\"Section3\"\u003e \u003ch2\u003e\u0026bull; Comparing Change Allowed vs. No Change Allowed:\u003c/h2\u003e \u003cp\u003eWhen participants were not allowed to change their belief, confidence drops were more substantial across both sleep conditions. After RS, confidence dropped more in the NCA condition by a magnitude of 1.73, \u003cem\u003et\u003c/em\u003e(48)\u0026thinsp;=\u0026thinsp;3.76, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;.001. After SD, this difference was even greater, with a larger confidence drop of 4.21 in the NCA condition, \u003cem\u003et\u003c/em\u003e(48)\u0026thinsp;=\u0026thinsp;9.16, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;.001. An independent t-test was conducted to compare the magnitude of confidence drops between the SD and RS condition. The drop in confidence after disagreement between CA and NCA after SD was significantly larger compared to after RS (\u003cem\u003et\u003c/em\u003e(48) = -3.44, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;.001). SD leads to a greater decrease in confidence when participants are not allowed to change their minds. Overall, these results consistently show that SD diminishes or suppresses confirmation bias, as seen in both conditions. The newly introduced NCA condition further amplifies the effects of SD, particularly by enlarging the confidence drop when participants face disagreement and are unable to change their beliefs.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003c/div\u003e\u003cp\u003e[1] Due to human error during data collection, the counterbalancement was not optimal. 34 participants completed one order, 16 the other.\u003c/p\u003e\n\u003cp\u003e\u003cbr\u003e\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eIn two studies, we investigated the impact of SD on belief updating, specifically examining how alterations in sleep affect both the likelihood of belief changes and the associated confidence updates. Across two studies, our results consistently demonstrate that SD alters belief updating dynamics, making individuals more susceptible to belief change while also diminishing their confidence in those beliefs. In Study 2, we replicated key findings from Study 1, further highlighting how sleep loss disrupts cognitive biases such as confirmation bias.\u003c/p\u003e \u003cp\u003eUnder well-rested conditions (RS), our findings align with previous research on belief updating. Participants were more likely to change beliefs when initially low in confidence (e.g., Pescetelli \u0026amp; Yeung, 2020; Gradassi et al., 2022), and they exhibited the classic confirmation bias. In line with the literature (Klayman, 1995; Nickerson, 1998), the change in confidence was asymmetric: confidence increased more after receiving confirming information than it decreased following disconfirming information.\u003c/p\u003e \u003cp\u003eOne of the most striking findings is that SD increased the likelihood of belief changes non selectively, occurring both after agreement and disagreement. SD impairs cognitive functions such as memory and attentional control (Killgore et al., 2007; Landolt et al., 2014). The heightened rate of belief changes observed here may stem from these cognitive impairments, which weaken the stability of initial beliefs. Notably, in our studies, this effect was closely tied to confidence. Sleep-deprived participants exhibited lower confidence overall, particularly after receiving disconfirming information. This finding aligns with research suggesting that lower confidence increases the likelihood of belief change (Pescetelli et al., 2021). By reducing confidence, SD may undermine the cognitive stability necessary to sustain an initial belief, thereby prompting belief shifts more readily. This supports the \u0026ldquo;follow along\u0026rdquo; hypothesis, which posits that sleep-deprived individuals may have less motivation or cognitive resources to critically evaluate new information (Blagrove, 1996). Nonetheless, the intact role of initial confidence in predicting belief change under SD suggests that core mechanisms of belief stability are still operational (Baranski, 2007).\u003c/p\u003e \u003cp\u003eRegarding confidence updates, SD reduced the typical confidence boost following agreement with a partner and amplified the confidence drop after disagreement, significantly attenuating confirmation bias. Rather than favouring confirmatory over disconfirmatory evidence, sleep-deprived participants appeared to \"equalize\" their responses to both types of information. This reduction in confirmation bias may be linked to findings that SD diminishes individuals' ability to weigh new information appropriately (Dickinson \u0026amp; Drummond, 2008) and reduces their ability to extract relevant information from stimuli (Johnson et al., 2020). The combination of lower overall confidence and impaired information processing likely contributes to an increased focus on disconfirming evidence, thereby reducing the strength of confirmation bias.\u003c/p\u003e \u003cp\u003eThe absence of confirmation bias under SD can be interpreted in several ways. It may reflect a weakening of the cognitive resources necessary to sustain bias (e.g., attentional control and memory; Harrison \u0026amp; Horne, 1999), or it may signal a shift in decision-making strategies. Sleep-deprived individuals may adopt a more balanced but less confident evaluation of new information, giving equal weight to confirming and disconfirming evidence. This reduction in bias could have both positive and negative implications: while it may lead to greater flexibility in updating beliefs, it may also result in erratic decision-making when consistency is required.\u003c/p\u003e \u003cp\u003eA key insight from Study 2 is that when belief change was not allowed (NCA condition), participants experiencing sleep deprivation (SD) exhibited a significantly sharper decline in confidence after receiving disconfirming information. This suggests that, in the absence of the option to change beliefs, SD amplifies the psychological impact of disconfirming evidence, leading to heightened belief instability. Under these circumstances, individuals appear more vulnerable to disconfirming information, resulting either in frequent belief changes or steep declines in confidence when belief change is constrained.\u003c/p\u003e \u003cp\u003eThese findings underscore the critical role of confirmation bias in belief updating, particularly under conditions of SD. Typically, confirmation bias serves as a cognitive filter that helps stabilize beliefs by allowing individuals to give more weight to evidence supporting their existing beliefs while discounting contradictory information (Klayman, 1995). In well-rested individuals, this bias acts as a gatekeeper, buffering against confidence drops. However, when this gatekeeping function weakens under SD, the usual asymmetry between confidence gains after agreement and confidence drops after disagreement diminishes. Consequently, individuals become less able to maintain confidence in their beliefs, leading to increased flexibility in changing their beliefs or a greater likelihood of confidence collapse.\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eIn conclusion, this research sheds light on how SD affects belief updating by increasing the frequency of belief changes and diminishing confirmation bias. Our findings have several important implications for understanding decision-making in high-stakes environments, where SD is common (e.g., military, medical, political, and corporate settings). Sleep-deprived decision-makers may be more prone to changing their beliefs in response to new information, which could be either advantageous or detrimental depending on the context. In fast-paced environments requiring rapid adaptation, this increased flexibility may be beneficial. However, in scenarios requiring stable, consistent decision-making, the reduced cognitive gatekeeping due to diminished confirmation bias could lead to erratic or suboptimal choices. Future research should explore the neural mechanisms underlying these effects and investigate whether similar patterns are observed under other forms of cognitive strain, such as stress or time pressure. Understanding these dynamics can lead to strategies to mitigate the adverse effects of SD on decision-making, particularly in critical settings where consistent and accurate judgments are essential.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eAchtziger, A., Al\u0026oacute;s-Ferrer, C., H\u0026uuml;gelsch\u0026auml;fer, S., \u0026amp; Steinhauser, M. (2014). The neural basis of belief updating and rational decision making. \u003cem\u003eSocial Cognitive and Affective Neuroscience\u003c/em\u003e, \u003cem\u003e9\u003c/em\u003e(1), 55\u0026ndash;62. https://doi.org/10.1093/scan/nss099\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003e\u0026Aring;kerstedt, T., \u0026amp; Gillberg, M. (1990). Subjective and Objective Sleepiness in the Active Individual. \u003cem\u003eInternational Journal of Neuroscience\u003c/em\u003e, \u003cem\u003e52\u003c/em\u003e(1\u0026ndash;2), 29\u0026ndash;37. https://doi.org/10.3109/00207459008994241\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAssociated Press. (2020, 11 december). EU Leaders Agree to Reduce Emissions After All-night Talks. \u003cem\u003eVoice Of America\u003c/em\u003e. https://www.voanews.com/a/europe_eu-leaders-agree-reduce-emissions-after-all-night-talks/6199453.html\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBamford, J. (2001). \u003cem\u003eBody of Secrets: Anatomy of the Ultra-Secret National Security Agency\u003c/em\u003e. Anchor Books.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBaranski, J. V. (2007). Fatigue, sleep loss, and confidence in judgment. \u003cem\u003eJournal Of Experimental Psychology Applied\u003c/em\u003e, \u003cem\u003e13\u003c/em\u003e(4), 182\u0026ndash;196. https://doi.org/10.1037/1076-898x.13.4.182\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBates, D., Maechler, M., Bolker, B., \u0026amp; Walker, S. (2015, 7 april). \u003cem\u003eCRAN - Package lme4\u003c/em\u003e. Comprehensive R Archive Network (CRAN). Geraadpleegd op 6 mei 2022, van https://cran.r-project.org/web/packages/lme4/index.html\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBlagrove, M. (1996). Effects of length of sleep deprivation on interrogative suggestibility. \u003cem\u003eJournal of Experimental Psychology: Applied\u003c/em\u003e, \u003cem\u003e2\u003c/em\u003e(1), 48\u0026ndash;59. https://doi.org/10.1037/1076-898x.2.1.48\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBlagrove, M., \u0026amp; Akehurst, L. (2000). Effects of sleep loss on confidence\u0026ndash;accuracy relationships for reasoning and eyewitness memory. \u003cem\u003eJournal of Experimental Psychology: Applied\u003c/em\u003e, \u003cem\u003e6\u003c/em\u003e(1), 59\u0026ndash;73. https://doi.org/10.1037/1076-898x.6.1.59\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBrysbaert, M., \u0026amp; Stevens, M. C. (2018). Power Analysis and Effect Size in Mixed Effects Models: A Tutorial. \u003cem\u003eJournal of Cognition\u003c/em\u003e, \u003cem\u003e1\u003c/em\u003e(1). https://doi.org/10.5334/joc.10\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBuysse, D. J., Reynolds, C. F., 3rd, Monk, T. H., Berman, S. R., \u0026amp; Kupfer, D. J. (1989). The Pittsburgh Sleep Quality Index: a new instrument for psychiatric practice and research. \u003cem\u003ePsychiatry Res, 28\u003c/em\u003e(2), 193\u0026ndash;213.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDickinson, D. L., \u0026amp; Drummond, S. P. (2008). The effects of total sleep deprivation on bayesian updating. \u003cem\u003eJudgment and Decision Making\u003c/em\u003e, \u003cem\u003e3\u003c/em\u003e(2), 181\u0026ndash;190. https://doi.org/10.1017/s1930297500001522\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEllis, B. W., Johns, M. W., Lancaster, R., Raptopoulos, P., Angelopoulos, N., \u0026amp; Priest, R. G. (1981). The St. Mary's Hospital sleep questionnaire: a study of reliability. \u003cem\u003eSleep\u003c/em\u003e, 4(1), 93\u0026ndash;97.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFrenda, S. J., Berkowitz, S. S., Loftus, E. F., \u0026amp; Fenn, K. M. (2016). Sleep deprivation and false confessions. \u003cem\u003eProceedings of the National Academy of Sciences of the United States of America\u003c/em\u003e, \u003cem\u003e113\u003c/em\u003e(8), 2047\u0026ndash;2050. https://doi.org/10.1073/pnas.1521518113\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGradassi, A., Van Den Bos, W., \u0026amp; Molleman, L. (2022). Confidence of others trumps confidence of self in social information use. \u003cem\u003ePsyArXiv\u003c/em\u003e. https://doi.org/10.31234/osf.io/mqyu2\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHarrison, Y., \u0026amp; Horne, J. (1999). One Night of Sleep Loss Impairs Innovative Thinking and Flexible Decision Making. \u003cem\u003eOrganizational Behavior and Human Decision Processes\u003c/em\u003e, \u003cem\u003e78\u003c/em\u003e(2), 128\u0026ndash;145. https://doi.org/10.1006/obhd.1999.2827\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLandolt, H., Holst, S. C., Sousek, A., Bassetti, C., Dogas, Z., \u0026amp; Peigneux, P. (2014). Effects of acute and chronic sleep deprivation. In \u003cem\u003eSleep Medicine Textbook: Second Edition\u003c/em\u003e (Vols. 67\u0026ndash;80, pp. 49\u0026ndash;61). https://doi.org/10.5167/uzh-107182\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHorne, J., \u0026amp; \u0026Ouml;stberg, O. (1976). A self-assessment questionnaire to determine morningness- eveningness in human circadian rhythms. \u003cem\u003eInternational Journal of Chronobiology\u003c/em\u003e, \u003cem\u003e4\u003c/em\u003e(2), 97\u0026ndash;110. https://ci.nii.ac.jp/naid/20001543451\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJohnson, D. J., Stepan, M. E., Cesario, J., \u0026amp; Fenn, K. M. (2020). Sleep Deprivation and Racial Bias in the Decision to Shoot: A Diffusion Model Analysis. \u003cem\u003eSocial Psychological And Personality Science\u003c/em\u003e, \u003cem\u003e12\u003c/em\u003e(5), 638\u0026ndash;647. https://doi.org/10.1177/1948550620932723\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKappes, A., Harvey, A. K., Lohrenz, T., Montague, P. R., \u0026amp; Sharot, T. (2019). Confirmation bias in the utilization of others\u0026rsquo; opinion strength. \u003cem\u003eNature Neuroscience\u003c/em\u003e, \u003cem\u003e23\u003c/em\u003e(1), 130\u0026ndash;137. https://doi.org/10.1038/s41593-019-0549-2\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKassin, S. M., \u0026amp; Kiechel, K. L. (1996). The Social Psychology of False Confessions: Compliance, Internalization, and Confabulation. \u003cem\u003ePsychological Science\u003c/em\u003e, \u003cem\u003e7\u003c/em\u003e(3), 125\u0026ndash;128. https://doi.org/10.1111/j.1467-9280.1996.tb00344.x\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKillgore, W. D. (2010). Effects of sleep deprivation on cognition. In Progress in Brain Research (pp. 105\u0026ndash;129). Elsevier BV. https://doi.org/10.1016/b978-0-444-53702-7.00007-5\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKillgore, W. D., Killgore, D. B., Day, L., Li, C. I., Kamimori, G. H., \u0026amp; Balkin, T. J. (2007). The Effects of 53 Hours of Sleep Deprivation on Moral Judgment. \u003cem\u003eSleep\u003c/em\u003e, \u003cem\u003e30\u003c/em\u003e(3), 345\u0026ndash;352. https://doi.org/10.1093/sleep/30.3.345\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKlayman, J. (1995). Varieties of Confirmation Bias. In \u003cem\u003eThe Psychology of learning and motivation\u003c/em\u003e (pp. 385\u0026ndash;418). https://doi.org/10.1016/s0079-7421(08)60315-1\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLefebvre, G., Summerfield, C., \u0026amp; Bogacz, R. (2021). A Normative Account of Confirmation Bias During Reinforcement Learning. \u003cem\u003eNeural Computation\u003c/em\u003e, \u003cem\u003e34\u003c/em\u003e(2), 307\u0026ndash;337. https://doi.org/10.1162/neco_a_01455\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLenth, R. V. (2019, 27 maart). \u003cem\u003eCRAN - Package emmeans\u003c/em\u003e. emmeans: estimated marginal means, aka least-squared means. Geraadpleegd op 6 mei 2022, van https://cran.r- project.org/web/packages/emmeans/index.html\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNickerson, R. (1998). Confirmation Bias: A Ubiquitous Phenomenon in Many Guises. \u003cem\u003eReview of General Psychology\u003c/em\u003e, \u003cem\u003e2\u003c/em\u003e(2), 175\u0026ndash;220. https://doi.org/10.1037/1089-2680.2.2.175\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003e\u003cem\u003eOSF\u003c/em\u003e. Anckaert \u0026amp; Gevers, 2023. https://doi.org/10.17605/OSF.IO/MS2ZW\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePescetelli, N., Hauperich, A., \u0026amp; Yeung, N. (2021). Confidence, advice seeking and changes of mind in decision making. \u003cem\u003eCognition\u003c/em\u003e, \u003cem\u003e215\u003c/em\u003e, 104810. https://doi.org/10.1016/j.cognition.2021.104810\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePescetelli, N., \u0026amp; Yeung, N. (2020). The effects of recursive communication dynamics on belief updating. \u003cem\u003eProceedings Of The Royal Society B Biological Sciences\u003c/em\u003e, \u003cem\u003e287\u003c/em\u003e(1931), 20200025. https://doi.org/10.1098/rspb.2020.0025\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePoinas, F., Rosaz, J., \u0026amp; Roussillon, B. (2010). Updating Beliefs with Imperfect Signals: Experimental Evidence. \u003cem\u003eSocial Science Research Network\u003c/em\u003e. https://doi.org/10.2139/ssrn.1731549\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eR Core Team (2012). \u003cem\u003eR: A Language and Environment for Statistical Computing\u003c/em\u003e. Vienna: R foundation for Statistical Computing. http://www.R-project.org/\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRollwage, M., \u0026amp; Fleming, S. M. (2021). Confirmation bias is adaptive when coupled with efficient metacognition. \u003cem\u003ePhilosophical Transactions of the Royal Society B\u003c/em\u003e, \u003cem\u003e376\u003c/em\u003e(1822), 20200131. https://doi.org/10.1098/rstb.2020.0131\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRollwage, M., Loosen, A. M., Hauser, T. U., Moran, R., Dolan, R. J., \u0026amp; Fleming, S. M. (2020). Confidence drives a neural confirmation bias. \u003cem\u003eNature Communications\u003c/em\u003e, \u003cem\u003e11\u003c/em\u003e(1). https://doi.org/10.1038/s41467-020-16278-6\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRosner, R. I. (2015). Beck Depression Inventory (BDI). \u003cem\u003eNew Jersey: John Wiley \u0026amp; Sons, Inc.\u003c/em\u003e https://doi.org/10.1002/9781118625392.wbecp261\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSommer, J., Musolino, J., \u0026amp; Hemmer, P. (2023). Updating, evidence evaluation, and operator availability: A theoretical framework for understanding belief. \u003cem\u003ePsychological Review, 131\u003c/em\u003e(2), 373\u0026ndash;401. https://doi.org/10.1037/rev0000444\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003e\u003cem\u003eWGT3X-BT | ActiGraph Wearable Devices\u003c/em\u003e. https://theactigraph.com/actigraph-wgt3x-bt\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZenses, A., Lenaert, B., Peigneux, P., Beckers, T., \u0026amp; Boddez, Y. (2019). Sleep deprivation increases threat beliefs in human fear conditioning. \u003cem\u003eJournal of Sleep Research\u003c/em\u003e, \u003cem\u003e29\u003c/em\u003e(3). https://doi.org/10.1111/jsr.12873\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"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":"","lastPublishedDoi":"10.21203/rs.3.rs-5289682/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-5289682/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eBelief updating, the process of revising one\u0026rsquo;s beliefs in light of new information, is subject to confirmation bias - the tendency to prioritize information that supports existing beliefs. How this process is affected by sleep deprivation (SD), which is known to impair cognitive function, remains underexplored. Here, we examined how SD alters belief change and confirmation bias in response to peer advice during a decision-making task. Across two studies, we found that SD substantially increased the likelihood of belief changes, with participants being more likely to change their beliefs following SD compared to after a full night of sleep. Even after SD, beliefs were still changed using logical reasoning, as participants continued to consider their initial confidence. At the same time, SD eliminated confirmation bias. Indeed, the typical confirmation bias asymmetry (i.e., a stronger increase in confidence when in agreement compared to a smaller drop in confidence when in disagreement) was no longer observed under SD. Our findings suggest that sleep loss undermines belief stability and dampens the cognitive biases that typically reinforce belief consistency. As a result, SD contexts may exert a significant impact on decision making processes, eventually leading to deleterious consequences in high stake situations.\u003c/p\u003e","manuscriptTitle":"Sleep Deprivation Increases Belief Change and Suppresses Confirmation Bias in Decision Tasks","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-10-21 10:42:07","doi":"10.21203/rs.3.rs-5289682/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":"cc3bb9f8-562c-43c3-9e6d-5a63c90e9dcf","owner":[],"postedDate":"October 21st, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":39137606,"name":"Social science/Psychology"},{"id":39137607,"name":"Social science/Psychology/Human behaviour"}],"tags":[],"updatedAt":"2024-11-12T16:12:05+00:00","versionOfRecord":[],"versionCreatedAt":"2024-10-21 10:42:07","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-5289682","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-5289682","identity":"rs-5289682","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}