Hirshkowitz Et Al. 2015)

[Adriana Gowen]

Thankyou for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

Academic performance and cognitive activities like learning are influenced by sleep and mood or emotion. This review discusses the roles of sleep and mood/emotion in learning and academic performance.

Since there is some overlap between the descriptions of mood and emotion, in the context of the core content of this review, here, mood and emotion have not been differentiated based on their theoretical/psychological definitions. This is because the aim of the review is not to distinguish between the effects of mood and emotion on learning. Thus, these have been referred to as general affective states; essentially specific states of mind that affect learning. Also, these have been addressed in the context of the study being discussed and cited in that specific place in the review.

Like sleep, mood and emotions determine our physical and mental health. Depressive disorders have prevailed as one of the leading causes of health loss for nearly 30 years (James et al., 2018). Increased incidence of mood disorders amongst the general population has been observed (Walker et al., 2020), and there is an increase in such disorders amongst students (Auerbach et al., 2018). These have further risen during the COVID-19 pandemic (Son et al., 2020; Wang et al., 2020).

The relationship between sleep, mood and cognition/learning is far more complex than perceived. Therefore, this review aims to recognise the interrelationships between the aforementioned trio. It critically examines the effects of sleep and mood on cognition, learning and academic performance (Fig. 1). Furthermore, it discusses how various regulatory factors can directly or indirectly influence cognition and learning. Factors discussed here are age, gender, diet, hydration level, obesity, sex hormones, daytime nap, circadian rhythm, and genetics (Fig. 1). The effect of sleep and mood on each other is also addressed. Pictorial models that hypothesise learning on an emotion scale and vice-versa have been proposed.

Sleep and mood/emotion affect cognition and academic achievement. Their effects can be additionally influenced by other factors like diet, metabolic disorders (e.g., obesity), circadian rhythm, daytime nap, hydration level, age, gender, and genetics. The figure presents the interrelationships and highlights the complexity emerging from the interdependence between factors, action of multiple factors on a single factor or vice-versa and the bidirectional nature of some associations. These associations collectively determine learning and thereby, academic achievement. Direction of the arrow represents effect of a factor on another.

Such effects of sleep deprivation are obvious because it can impair procedural and declarative learning (Curcio et al., 2006; Kurniawan et al., 2016), decrease alertness (Alexandre et al., 2017), and impair memory consolidation (Hagewoud et al., 2010), attention and decision making (Alhola and Polo-Kantola, 2007). It can increase low-grade systemic inflammation and hinder cognitive functionality (Choshen-Hillel et al., 2020). Hippocampus is the region in the brain that plays the main role in learning, memory, social cognition, and emotion regulation (Y. Zhu et al., 2019). cAMP signalling plays an important role in several neural processes such as learning and memory, cellular excitability, motor function and pain (Lee, 2015). A brief 5-hour period of sleep deprivation interferes with cAMP signalling in the hippocampus and impairs its function (Vecsey et al., 2009). Thus, optimal academic performance is hindered, if there is a sleep disorder (Hershner and Chervin, 2014).

Despite the clear significance of appropriate sleep quality and quantity in cognitive processes, there are some caveats to drawing definitive conclusions in certain areas. First, there are uncertainties around how much sleep is optimal and how to measure sleep quality. This is further confounded by the dependence of sleep quality and quantity on various genetic and environmental factors (Roenneberg, 2013). Moreover, although sleep enhances emotional memory, during laboratory investigations, this effect has been observed only under specific experimental conditions. Also, the experiments conducted have differed in the methods used and in considering parameters like timing and duration of sleep, age, gender and outcome measure (Lipinska et al., 2019). This orientates conclusions to be specific to those experimental conditions and prevents the formation of generic opinions that would be applicable to all circumstances.

The reason for such inconclusive and unanticipated results could be that sleep is not the sole determinant of learning. Learning is affected by various other factors that may alter, exacerbate, or surpass the influence of sleep on learning (Fig. 1). These factors have been discussed in the subsequent sections.

Based on the examples and discussion so far, a direct relationship between emotions and learning could be hypothesised, whereby positive emotions would promote creative learning strategies and academic success, whereas negative emotions would lead to cognitive impairment (Fig. 2a). However, this relationship is far more complex and different than perceived.

Emotions have been shown on a hypothetical learning scale. a Usually, positive and negative emotions are perceived to match with optimal and poor learning, respectively. b Emotions that lead to sub-optimal/poor and optimal/better learning have been shown on the hypothetical learning scale. Here, distinct from (a), both negative emotions and high arousal positive emotions have been implicated in poorer learning compared with low-intensity positive emotion like pleasantness; the latter is believed to lead to optimal learning. The question mark reflects that some negative emotions like shame might stimulate learning, but the exact intensity of such emotions and whether these would facilitate better or worse learning than high arousal positive emotions or pleasantness need to be investigated.

Although positive mood favours the recall of learnt words, it correlates with increased distraction and poor planning (Martin and Kerns, 2011). High levels of positive emotions like excitedness and elatedness may decrease achievement (Fig. 2b) (Valiente et al., 2012). It may be surprising to know that negative emotions such as shame and anxiety can arouse cognitive activity (Miller et al., 2018). Along similar lines, it has been observed that participants exposed to sad and neutral moods performed similarly in visual statistical (learning) tasks but those who experienced sad stimuli showed high conscious access to the acquired statistical knowledge (Bertels et al., 2013). Dysphoria is a state of dissatisfaction that may be accompanied by anxiety and depression. Participants with dysphoria have shown more sensitivity to temporal shifts in outcome contingencies than those without dysphoria (Msetfi et al., 2012), and these participants reiterated the depressive realism effect and were quicker in endorsing the connection between negative words and ambiguous statements, demonstrating a negative bias (Hindash and Amir, 2012). Likewise, not the positive emotion but negative emotion has been shown to influence the learning outcomes, and it increased the efficiency and precision of learning morphosyntactic instructions involving morphology and syntax of a foreign language (X. Liu et al., 2018). Thus, negative emotions can allow, and at times, stimulate or facilitate learning (Figs. 2 and 3). Further investigation is needed on the intensity of these emotions, whether these would facilitate better or worse learning than high-intensity positive emotions and whether the results would be task specific.

The figure depicts that low-to-medium intensity positive emotion like pleasantness leads to optimal learning, whereas high-intensity emotions, either positive or negative, may lead to suboptimal or comparatively poorer learning. The model considers the apparently unexpected data that negative emotions may stimulate learning. However, which negative emotions these would be, their intensities and their corresponding level of learning are not known, and so these are not shown in the figure. Also, the figure shows bias towards positive emotions in mediating optimal learning. This information is based on the literature so far. Note that the figure represents concepts only and is not prescriptive. It shows inequality and differences between the impacts of high arousal positive and high arousal negative emotions. This concept needs to be investigated. Therefore, the figure may/may not be an accurate mathematical representation of learning with regards to the intensities of positive and negative emotions. In actuality, the scaling and intensities of emotions on the negative and positive sides of the scale may not be equal, particularly in reference to the position of optimal learning on the scale. Furthermore, upon plotting the 3rd dimension, which could be one or more of the regulatory factors discussed here might alter the position and shape of the optimal learning peak.

Interestingly, some studies have indicated that emotions do not play a significant role in context. For example, a study showed that there was no evidence that negative emotions in depressed participants showed negative interpretations of ambiguous information (Bisson and Sears, 2007). In another study, improvements in visuomotor skills happened regardless of perturbation or mood states (Kaida et al., 2017). Thus, mood can either impair, enhance or have no effect on cognition. The effect of mood on cognition and learning can be variable and depend on the complexity of the task (Martin and Kerns, 2011) and/or other factors. Some of these factors have been discussed in the following section. The discrepancies in the data on the effects of mood on cognition and learning may be partly attributed to the influence of these factors on cognitive functions.

3a8082e126