Original date of publication: 21-06-2019
University of Amsterdam
Abstract
During sleep the human brain consolidates and recombines memories. This recombination of memory can create new connections between separate experiences, leading to new insights. People have reported consciously experiencing new insights into daily-life problems during, and after, a bout of sleep. The specific characteristics of these sleep-related insights are currently not fully known. In this study it was found that variance in the characteristics of sleep-related insights is associated with differences in age, lucidity of dreams and chronotype. In addition, insights of emotional nature were found to be more often associated with dreams compared to insights of rational nature. These findings show that specific demographic- and lifestyle-factors may be linked to the way memories are processed during sleep, resulting in a differing expression of sleep-related insights.
Keywords: Sleep; Insights; SRI; Memory; Lifestyle
Introduction
During the human lifespan, a large part of our time is spent sleeping. It is assumed that sleep plays a major role in the maintenance of our physical health. However, sleep also seems to have a crucial role in the regulation of mental functions, such as memory. There have been reports of people experiencing insights into daily-life problems during, or shortly after, a period of sleep. For instance, famous artist Salvador Dalí reports in his book 50 Secrets of Magic Craftmanship (1992) that he purposely tried inducing a state between sleep and wakefulness to tap into the creative processes occurring during sleep. Otto Loewi, the pharmacologist that discovered neuronal chemical transmission, came up with the idea of chemical transmission during a night of intermittent sleep (McCoy and Tan 2014).Insights like these are potentially linked to mental processes that are specifically active during sleep. Currently little is known about these sleep-related insights (SRI’s), their occurrence and characteristics. This exploratory paper is on the frontier of what is known about SRI’s and reports on the correlation between SRI’s and various demographic- and lifestyle-factors.
During sleep the brain is actively consolidating memory (Diekelmann and Born 2010). It is reported that sleep is even involved in more complex creative processes where memories and related information are recombined. Research by Gupta, van der Meer et al. (2010) indicates that rodents experience neuronal replay & recombination activity during sleep after day-time training. In human participants relational memory was also improved upon incorporation of sleep (Ellenbogen, Hu et al. 2007). Findings like these indicate a tight link between sleep and memory in general. But more specific influence on SRI’s might be caused by memory recombination and associated dreams. Dreams can potentially be a source of insights into daily-life problems (Edwards, et al. 2013), with emotional experiences being consolidated during Rapid-Eye-Movement (REM) sleep in particular. Research by Cai, et al. (2009) also showed that recombination of information is largely active during REM sleep compared to non-REM sleep, suggesting recombination-caused insights are more prevalent during dreaming (which is more prevalent during REM sleep). However, research by Verleger, et al. (2013) also gave indications for the formation of insights during Slow Wave Sleep (SWS). Sleep-related insights could therefore be linked to both REM-sleep and NREM-sleep. As shown by Dresler, et al. (2012), people that consciously know they are dreaming (lucid dreaming) or have control over their dreams, have additional activation patterns in brain areas associated with memory. People that experience lucid dreams can therefore potentially consolidate SRI’s better into their conscious memory.
Furthermore, research by Nissen, et al. (2011) showed that Primary Insomnia patients suffered from both impaired procedural memory and declarative memory. Given the role of sleep in memory consolidation, SRI’s can potentially be influenced by sleep quality. Another significant contributor to both sleep quality and sleep habits is a person’s chronotype, the behavioral effect on bedtime and sleep-habits caused by physiological factors. People that identify as evening-persons tend to prefer later bedtimes and have lower sleep quality (Taillard, et al. 2002). Because of the comprehensive effect on sleep and sleep habits, a person’s chronotype is a potential influencer of the amount and types of SRI’s that are experienced. Besides the effect of behavioral differences, differences in the use of medication and drugs can affect both sleep quality and REM sleep (Naiman, R. 2017; Vogel, et al. 1990). Even alcohol shows signs of suppressing REM sleep (Ebrahim, et al. 2013). All these lifestyle factors can therefore potentially influence the occurrence and characteristics of SRI’s.
An initial pilot study on the occurrence of SRI’s and its association with dreams has been performed by Perdomo, Hofman & Talamini (2018). In this study a questionnaire was developed to investigate the occurrence and qualities of SRI’s in a group of university students. Here it was found that SRI’s occurred in up to 80 percent of the participants, with 40 percent experiencing them regularly. The SRI’s occurred in diverse states of sleep and were often associated with sleep mentation. This suggests that the SRI’s are linked to higher-order processes, such as memory recombination. However, this research was conducted on a group of uniform students with a similar age and educational background. Extrapolation of these results to the general population might therefore be inaccurate. Also, only few demographic- and lifestyle variables were taken into account.
This current study focused on elaborating on these previous findings using a broader test-population and more thorough investigation of lifestyle factors, which potentially influence the characteristics and occurrence of sleep-related insights. As this current study is mostly explorative, no strict hypothesis has been formed. However, it is suspected that SRI’s are negatively influenced by low sleep-quality and use of drugs or medication, as these factors are thought to impair memory recombination. The study was conducted using a comprehensive online questionnaire and focused on a broad representation of the Dutch population.
Materials & Methods
Subjects
In total 671 Dutch speaking participants (36.8% male and 63.2% female) finished the questionnaire. The participants’ age spanned from 18 to 90 years old (mean = 34.8, SD = 16.0). The educational background ranged from no education to university graduates (48.1% of participants either attend or finished university, 51.9% attended occupational education or a lower degree). Of these 671 participants, 173 joined the second test for the assessment of test-retest reliability. The study was approved by the Ethics Review Board of the University of Amsterdam and all participants signed for Informed Consent.
Materials
A new questionnaire was composed for this study, based on the Sleep-Related Insight questionnaire of the University of Amsterdam (L.M. Talamini and W.F. Hofman). Additional items for sleep-quality were added based upon the Dutch Sleep Quality Scale (Visser, P., et al. 1979). New original questions were added for use of medication, the use of food supplements, chronotype and sleeping habits for a total of 51 items. The questionnaire contained 3 demographic questions, 19 questions focused on the occurrence, characteristics and frequencies of SRI’s, 17 questions focused on subjective sleep quality, 6 questions on sleeping habits and 6 questions on the use of medication and (food)supplements. A complete overview of all the questions can be found in the original Dutch questionnaire in the appendix.
Procedure
The study recruited participants during a period of 15 weeks. The questionnaire was presented online using Qualtrics questionnaire software (2015), in the Dutch language. A weblink was distributed to all participants, their own devices (computer/tablet/phone) were used to fill in the questionnaire. The average time required to fill in the questionnaire was approximately 10 minutes. Participants were presented with the option to fill in the questionnaire a second time (for test-retest reliability assessment). The participants that volunteered for the second assessment received an automatic email invitation exactly 7 days after finishing the first questionnaire.
Recruitment
Participants were recruited online using Social Media networking websites, request placements on online fora, two promotional videos and advertisements. A participation request was placed on the social media accounts of all involved researchers and on both general and sleep-related fora. The first promotional video was posted on the social media accounts of the University of Amsterdam. A second promotional video was placed on the social media pages of a sleep-related local business (Slaaploos). Advertisements were run on Facebook and Instagram, targeted on people of 18 years and older and fluent in Dutch.
Data analysis
In total 1,042 participants started the questionnaire. The participants that didn’t finish the questionnaire were excluded, leaving 671 participants. One participant answered the questionnaire very fast (88 seconds), but as the answers seemed consistent the participant was not removed. The 671 participants that finished the questionnaire received a request to fill in the questionnaire a second time for test-retest reliability assessment, 7 days after the initial questionnaire. Participants that hadn’t finished the second questionnaire in 7-14 days after the initial questionnaire were excluded, leaving 173 participants for test-retest reliability assessment. Data analysis was performed using R Version 3.5.3. Test-retest reliability was tested using the non-parametric Spearman’s correlation test, as normality assumptions were not met for parametric testing. For the other analyses the assumptions for parametric testing were met, therefore parametric tests were performed*. Internal consistency was measured by Cronbach’s α. A composite score for SRI prevalence was calculated for every participant, consisting of the non-weighted average score of the questions on SRI prevalence and frequency. The influence of demographic- and lifestyle-factors on this composite SRI-score was calculated using R’s built-in Multiple Regression function. Distribution of insight-types over gender, sleep-phase and dreams were analyzed using Pearson’s Chi-squared test.
*The assumptions for the Multiple Regression Analysis were met, the residuals followed a linear pattern, there was no multicollinearity (the VIF’s were between 1 and 2) and the residuals had equal variance. The assumptions for Pearson’s Chi-squared test were met, participants were included in maximum one cell, all cells contained at least 5 entries and all cells contained frequencies.
Results
Test-Retest reliability and internal consistency
In total 173 participants finished the second questionnaire for the test-retest reliability assessment. Test-retest reliability was calculated using Spearman’s Correlation Test. All items showed sufficient reliability, ranging from a correlation of .47 to .75. In Table 1 the correlation for all SRI-questions is displayed. The internal consistency of the Sleep-related insights questions (items 5 to 23) was calculated using Cronbach’s α, which gave an internal consistency of .757.
| Question | Correlation coefficient (rho) | Question | Correlation coefficient (rho) |
| 5 | .67 | 13 | .58 |
| 6 | .46 | 14 | .75 |
| 7 | .52 | 15 | .54 |
| 8 | .66 | 17 | .75 |
| 9 | .56 | 18 | .53 |
| 10 | .65 | 21 | .64 |
| 11 | .57 | 22 | .68 |
| 12 | .63 | 23 | .59 |
Table 1 – Test-Retest reliability – Test-Retest correlation of questionnaire items 5-18 and 21-23, measured at an interval of 7-14 days. All correlations are significant (p<0.001) and range from 0.47 to 0.75.
Prevalence of SRI’s
In total 84.5% of the participants have experienced a SRI at least once, with 14.5% experiencing SRI’s weekly, 24.4% monthly, 30.3% at least once a year and 15.5% never. Of the participants reporting SRI’s, 78.1% reported their insight(s) to be useful in daily life, and 15.4% have tried consciously inducing SRI’s by contemplating a problem prior to sleeping. Multiple Regression analysis was performed to determine what demographic- and lifestyle-factors contribute to the prevalence of SRI’s. This resulted in a significant regression equation (F(15,571) = 2.94, p<0.001), with a R2 of 0.071. The prevalence of SRI’s is correlated with age (p=0.002), lucid dreaming (p=0.006) and chronotype (p=0.001). Age has a negative relationship with SRI’s, with older people experiencing less SRI’s. Lucid dreaming has a positive relationship with SRI’s, with people that experience lucid dreams also experiencing more SRI’s. Regarding the chronotype, people having the subjective experience of being a morning person experience more SRI’s, and evening persons experience less SRI’s. The other measured demographic- and lifestyle-factors did not show a significant correlation with SRI prevalence. In Table 2 the β-coefficients (standardized), squared structure coefficient and p-values for all measured independent variables are displayed.
| Independent variable effect on SRI-score | β-coefficient standardized | Rs2 | P-value |
| Gender | -0.064 | 0.119 | 0.135 |
| Age | -0.142 | 0.187 | 0.002* |
| Education | 0.058 | 0.113 | 0.187 |
| Sleep quality | -0.059 | 0.070 | 0.290 |
| Zinc | 0.030 | 0.003 | 0.458 |
| Sleep supplements | -0.041 | 0.001 | 0.350 |
| Alcohol | 0.066 | 0.043 | 0.121 |
| Cannabis | -0.002 | <0.001 | 0.957 |
| Antidepressants | 0.038 | 0.036 | 0.369 |
| Tranquilizers | 0.009 | 0.001 | 0.828 |
| Lucid dreaming | 0.114 | 0.207 | 0.006* |
| Chronotype | -0.144 | 0.140 | 0.001* |
| Awakenings | 0.070 | 0.077 | 0.184 |
| Bedtime | -0.014 | 0.031 | 0.752 |
| Irregular sleep | 0.003 | <0.001 | 0.950 |
Table 2 – SRI-score correlates – Effect of demographic- and lifestyle-factors on the composite SRI-score, given the fit of (F(15,571) = 2.94, p<0.001), with a R2 of 0.071. As displayed, the variables age, lucid dreaming and chronotype have a statistically significant correlation with the SRI-score. *significant p-value.
Distribution of SRI’s
Of the SRI’s that the participants experienced, 57.2% consisted of emotional insights (personal in nature) and 42.8% consisted of rational insights (theoretical in nature). Emotional and rational insights were not evenly distributed over dream- and non-dream phases of the night. Emotional insights were more often associated with dreams (p<0.001) compared to rational insights. No significant differences were found for the distribution of emotional and rational insights over gender or time of the night & sleep-wake state.
Discussion
The objective of this study was to research the occurrence and characteristics of sleep-related insights in a broad population. This study recruited participants of all ages and socio-economic backgrounds, giving a better representation of the population compared to previous studies conducted on SRI’s. Using an elaborated SRI-questionnaire, the demographic- and lifestyle-factors potentially influencing SRI’s was investigated. This led to the findings that SRI’s are negatively correlated with age, positively correlated with lucid dreaming and that the chronotype ‘morning person’ is positively correlated with experiencing SRI’s. Furthermore, emotional insights are more often associated with dreams compared to rational insights.
This study showed that SRI’s are very frequent (84.5%) in the Dutch population, confirming previous research (Perdomo, et al. 2018). Of these insights, both emotional insights (57.2%) and rational insights (42.8%) are common. Insights of emotional nature are more frequently associated with dreams. This could potentially be explained by the increase in emotional memory consolidation during REM-sleep (Edwards, et al. 2013), in which vivid (hallucinogenic) dreams are more frequent (Fosse, et al. 2004). Whereas insights of a rational nature might be more tightly associated with non-REM sleep dreams (Fosse, et al. 2004), where hippocampus dependent forms of memory are encoded (Wamsley, et al. 2010), or SWS where less conscious dreams are formed (Verleger, et al., 2013). The use of drugs or medications did not seem to influence this process, as no significant correlations between SRI’s and the use of these substances was found. Prior it was expected that drugs would influence SRI’s, due to drugs affecting the duration of REM-sleep, in which memory is consolidated and recombined. However, this did not seem to be the case. One possible explanation is that REM-sleep is less crucial to SRI-formation or memory consolidation as is currently considered, as proposed by Vertes, et al. (2000). This could potentially explain why SRI’s are not influenced by drugs that decrease the time spent in REM-sleep. As this study did not directly measure changes in REM-sleep, it cannot be concluded that the users of medications or drugs did in fact experience a reduction in REM-sleep. A direct study on the influence of REM-sleep duration on SRI would be required. It is also possible that the effect size of these medications is very low, thus leading to a failure to reject the null-hypothesis of no effect of drugs on SRI’s.
The prevalence of SRI’s correlated with age, lucid dreaming and self-reported chronotype. Age displayed a negative correlation with SRI’s, potentially explained by a decrease in sleep quality (Ohayon, et al. 2004), dream recall, REM sleep and general memory performance in older adults (Mangiaruga, et al. 2018). Participants that experienced lucid dreams showed an increase in overall SRI-score. This might be related to better dream recall, the ability to actively solve problems in dreams due to its conscious nature or that people that are actively lucid dreaming have stronger intentions towards the practical usage of the dream-state. Participants that reported being a ‘morning person’ were associated with a higher SRI-score, indicating that morning persons experienced more SRI’s compared to evening persons. This could be caused by the overall higher sleep-quality in morning persons compared to evening persons (Vitale, et al. 2014), or because of the higher attentional performance of morning persons in the morning (Facer-Childs, et al. 2019), which is required for the remembrance of insights.
As can be reported from this study SRI’s are a very frequent occurrence, both for emotional insights and rational insights. These insights are thought to be associated with memory consolidation and recombination processes during various sleep phases. These unconscious processes during sleep lead to tangible and useful insights, in a majority of the Dutch population. The characteristics of these insights are correlated with demographic- and lifestyle-factors, indicating a relationship between these factors and memory processing during sleep.
References
Cai, D. J., Mednick, S. A., Harrison, E. M., Kanady, J. C., & Mednick, S. C. (2009). REM, not incubation, improves creativity by priming associative networks. Proceedings of the National Academy of Sciences, 106(25), 10130-10134. doi:10.1073/pnas.0900271106
Dali, S., & Chevalier, H. (1992, p. 33-38). 50 Secrets of Magic Craftsmanship. North Chelmsford, MA: Courier Corporation.
Dresler, M., Wehrle, R., Spoormaker, V. I., Koch, S. P., Holsboer, F., Steiger, A., … Czisch, M. (2012). Neural Correlates of Dream Lucidity Obtained from Contrasting Lucid versus Non-Lucid REM Sleep: A Combined EEG/fMRI Case Study. Sleep, 35(7), 1017-1020. doi:10.5665/sleep.1974
Ebrahim, I. O., Shapiro, C. M., Williams, A. J., & Fenwick, P. B. (2013). Alcohol and Sleep I: Effects on Normal Sleep. Alcoholism: Clinical and Experimental Research, 37(4), 539`-549. doi:10.1111/acer.12006
Edwards, C. L., Ruby, P. M., Malinowski, J. E., Bennett, P. D., & Blagrove, M. T. (2013). Dreaming and insight. Frontiers in Psychology, 4. doi:10.3389/fpsyg.2013.00979
Ellenbogen, J. M., Hu, P. T., Payne, J. D., Titone, D., & Walker, M. P. (2007). Human relational memory requires time and sleep. Proceedings of the National Academy of Sciences, 104(18), 7723-7728. doi:10.1073/pnas.0700094104
Facer-Childs, E. R., Campos, B. M., Middleton, B., Skene, D. J., & Bagshaw, A. P. (2019). Circadian phenotype impacts the brain’s resting-state functional connectivity, attentional performance, and sleepiness. Sleep, 42(5). doi:10.1093/sleep/zsz033
Fosse, R., Stickgold, R., & Hobson, J. A. (2004). Thinking and hallucinating: Reciprocal changes in sleep. Psychophysiology, 41(2), 298-305. doi:10.1111/j.1469-8986.2003.00146.x
Naiman, R. (2017). Dreamless: the silent epidemic of REM sleep loss. Annals of the New York Academy of Sciences, 1406(1), 77-85. doi:10.1111/nyas.13447
Nissen, C., Kloepfer, C., Feige, B., Piosczyk, H., Spiegelhalder, K., Voderholzer, U., & Riemann, D. (2011). Sleep-related memory consolidation in primary insomnia. Journal of Sleep Research, 20(1pt2), 129-136. doi:10.1111/j.1365-2869.2010.00872.x
Ohayon, M. M., Carskadon, M. A., Guilleminault, C., & Vitiello, M. V. (2004). Meta-Analysis of Quantitative Sleep Parameters From Childhood to Old Age in Healthy Individuals: Developing Normative Sleep Values Across the Human Lifespan. Sleep, 27(7), 1255-1273. doi:10.1093/sleep/27.7.1255
Perdomo, V.L., Hofman, W.F., Talamini, L.M. (2018). Sleep Fosters Insights Into Real-Life Problems. Archives Italiennes de Biologie, 156(3), 87-98. doi: 10.12871/00039829201831.
Qualtrics XM // The Leading Experience Management Software. (2015, October 13). Retrieved 2019, from http://www.qualtrics.com/
Schoch, S. F., Cordi, M. J., Schredl, M., & Rasch, B. (2018). The effect of dream report collection and dream incorporation on memory consolidation during sleep. Journal of Sleep Research, 28(1), e12754. doi:10.1111/jsr.12754
Taillard, J., Philip, P., & Bioulac, B. (2002). Morningness/eveningness and the need for sleep. Journal of Sleep Research, 8(4), 291-295. doi:10.1046/j.1365-2869.1999.00176.x
Verleger, R., Rose, M., Wagner, U., Yordanova, J., & Kolev, V. (2013). Insights into sleep’s role for insight: Studies with the number reduction task. Advances in Cognitive Psychology, 9(4), 160-172. doi:10.5709/acp-0143-8
Vertes, R. P., & Eastman, K. E. (2000). The case against memory consolidation in REM sleep. Behavioral and Brain Sciences, 23(6), 867-876. doi:10.1017/s0140525x00004003
Vitale, J. A., Roveda, E., Montaruli, A., Galasso, L., Weydahl, A., Caumo, A., & Carandente, F. (2014). Chronotype influences activity circadian rhythm and sleep: Differences in sleep quality between weekdays and weekend. Chronobiology International, 32(3), 405-415. doi:10.3109/07420528.2014.986273
Vogel, G., Buffenstein, A., Minter, K., & Hennessey, A. (1990). Drug effects on REM sleep and on endogenous depression. Neuroscience & Biobehavioral Reviews, 14(1), 49-63. doi:10.1016/s0149-7634(05)80159-9
Visser P. et al. (1979, p. 135-145) Sleep and Mood: Measuring the Sleep Quality. In: Priest R.G., Pletscher A., Ward J. (eds) Sleep Research. Springer, Dordrecht
Wamsley, E. J., Tucker, M., Payne, J. D., Benavides, J. A., & Stickgold, R. (2010). Dreaming of a Learning Task Is Associated with Enhanced Sleep-Dependent Memory Consolidation. Current Biology, 20(9), 850-855. doi:10.1016/j.cub.2010.03.027