4. Discussion
In this experimental study, adolescents who consumed a low GI breakfast and performed 30 min exercise showed greater improvements in their mathematics score and computation time on a simple arithmetic test compared with those who remained fasted and sedentary over a morning. In addition, adolescents in the combined intervention had temporary mid-morning improvements in reaction times for congruent, incongruent and control trials during an interference (Stroop Color-Word) test and in an attention (Go/No-Go) task. However, these faster reactions did not translate into improved task accuracy and the improvement in reaction time was obviated by morning end. This investigation builds on the work of one previously published study examining acute differences in cognitive outcomes in youth given a high- or low-GI breakfast in combination with exercise [25]. We believe that our study is the first to demonstrate an academic performance outcome after combining breakfast with exercise using an acute experimental design. Nevertheless, it is important to quantify that we did not see any clear effect of the combined intervention over individual interventions, despite increases for mathematical score and computation time in the combined group being greater than those observed in the breakfast or exercise groups alone.
Breakfast intake in children has declined in many countries. In one large multi-country comparison involving >190,000 adolescents aged 11–15 years a frequency of daily breakfast consumption of >70% was found in only 4 of 41 countries [41]. Other cross-sectional studies also indicate that breakfast skipping ranges from 10% to 30% among youth [42,43]. Childhood physical inactivity is a growing issue in many countries and despite consensus that time spent in physical activity does not hinder academic performance, the perception persists that time spent on non-academic pursuits negatively impacts schoolwork [44]. Our data showing improved mathematical accuracy and increased processing speed with the combined intervention may therefore be important in overcoming such beliefs. Moreover, it could encourage parents, educators and policy makers to implement healthy lifestyle programs for better classroom outcomes.
Previous evidence demonstrates the benefits of both breakfast and exercise on improving academic outcomes and cognitive function in adolescents but we are aware of only one investigation examining their combined effects on cognition alone [25]. Similar to that previous study, we used a Stroop Color-Word Test as a measure of selective attention and inhibition and also found improved response times but not accuracy in the combined breakfast and exercise condition compared with adolescents who were fasted and sedentary. The faster response times observed were for all Stroop Test trials, including during interference in the incongruent trials, demonstrating that even complex responses were improved. These observations agreed with those of our test of attentional control, the Go/No-Go Task, where again response time was faster but accuracy was not improved. We did not see any effect on our test of working memory (Digit-Span Test). Our findings are an important confirmation of the previous work [25] and reassuringly similar given that our participants were in late adolescence when growth and brain development are more similar to that of adulthood.
That the cognitive improvement was only for reaction times may not be surprising. Literature on the effects of breakfast and exercise on cognitive tasks is inconsistent. For example, breakfast can have domain-specific effects on areas of cognition such as memory but this is influenced by factors such as the cognitive load imposed and even gender [1]. Acute exercise may also impact different cognitive domains differently. One suggestion is that exercise induces increases of neurotransmitters (norepinephrine and dopamine) that have a positive effect on processing speeds but which could induce neural noise in the brain hindering working memory tasks [45]. Certainly, the influence on reaction times is further supported by one meta-analysis which showed that moderate intensity exercise, such as that used here, results in increased processing speeds [45]. In the same analysis, the authors suggest that effects on task accuracy are limited because the tests used are not sufficiently complex to measure exercise-induced changes in accuracy performance [45]. Future studies should consider including more complex tasks to evaluate this.
An important point to highlight was that the improvement in response time was at test-series2 immediately after the combined intervention. The improvement did not last until test-series3 when reaction times for both the Stroop Color-Word Test and Go/No-Go Task were similar in B-E with all other conditions. In part, this could be from progressive waning of the acute influences of breakfast and exercise session on these cognitive processes. Transient effects on cognition with breakfast intake (see breakfast mechanisms below) are known but the duration of any effect varies among studies. Similarly, meta-analysis demonstrates that post-exercise cognitive task performance often improves but the size of any effect on performance diminishes over time [11]. The neurotransmitter catecholamine, noradrenaline, only has a half-life in the peripheral circulation of 3 min and it has been argued that this may be one explanation [45]. However, this explanation has been challenged as the half-life of brain (central) catecholamine concentrations is considerably longer at 8–12 h [46]. An alternative explanation is the influence of exercise on the protein brain-derived neurotrophic factor (BDNF) which acts as a stimulus for signaling pathways involved in neurogenesis and neuroplasticity. The effect of BDNF on downstream pathways takes time and so is most likely observed post-exercise [45]. However, BDNF increases tend to be greater for higher intensity exercise protocols and heart rate and RPE data from our exercise indicate that these were of a more moderate intensity [11]. Thus, any effect on BDNF may have been limited.
We believe the present work is the first to show the combined benefit of breakfast and exercise on performance of simple mathematical calculation accuracy and processing speed—an important demonstration from an education perspective. The fact that this improvement occurred with a single morning’s intervention affirms previous literature that some of the improvement elicited from breakfast and exercise is from the acute effects rather than long-term changes in nutritional status or physical fitness [1,2,3,4,5,11,12,13,14]. Our study design offers a novel perspective to previous work by providing a direct comparison with a morning where individuals were fasted and sedentary as well as the individual interventions themselves. One observation is that the period of improvement with the mathematics test (test-series3) did not correspond with that of the cognitive tests (test-series2). Decreased task sensitivity with repeated testing for the cognitive tests may explain this but certainly the improvement in mathematics accuracy and processing speed remained within the time frame previously reported for breakfast or exercise [1,11,46]. Despite the important demonstration of increased mathematics accuracy and processing speed with the combined intervention, we are cautious about the extent of the improvement achieved. Although the mean percentage change in the B-E group for mathematics score and computational time was more than double the percentage observed in the F-S group, the raw scores at the end of the morning did not differ between groups suggesting the improvement was from different starting points. Nevertheless, no significant difference among groups at the start of the morning was seen and we do not believe that the improved scores in the B-E group are simply an artefact as similar improvements were not observed in the F-E or B-S groups. Similarly, we draw attention to the extent of the absolute mathematics improvements in the combined group over the morning which were smaller than that of the differences among groups at baseline. Thus, the overall effect of the intervention is limited. To some extent this is reassuring as no real improvement in mathematics ability would be expected in response to acute breakfast and exercise over a morning but rather the improvements reflect the limited effects of the intervention themselves.
A systematic review including 34 studies supports the assertion that acute breakfast intake improves cognition in both in well-nourished children and those considered nutritionally at risk. Breakfast has a transient beneficial effect on cognition for up to 4 h post-ingestion, particularly for tasks related to attention, executive function and memory [1]; a time period within that of the present study. Several mechanisms are postulated to explain the effect of breakfast intake on academic and cognitive performance including satiation and hunger alleviation [4,7], improved fuel and nutrient provision for the central nervous system [1,6], and an increase in the synthesis or levels of brain neurotransmitters [1,6]. Evidence indicates that one or several of these mechanisms were present in our study including lower subjective hunger and appetite, greater subjective fullness and satiety, and higher glucose concentrations. Some of this may stem from the low GI meal we employed rather than a high GI meal [2,3,4,6,8]. One note of caution here, however, relates to the subjective measures of hunger in our study, which although improved on the B-E compared with fasted groups over the morning, differed at the outset. We told participants at the start of the morning upon reporting to the laboratory which group they were in instead of blinding them until the treatment was given. In retrospect, this may have inadvertently biased their ratings on these measures. Behavioral explanations are another alternative possibility for the effects of breakfast on cognitive performance as short-term breakfast consumption can heighten subjective feelings of alertness and motivation to concentrate and learn [1]. However, we did not see greater motivation or arousal with B-E compared with fasting. Instead, feeling (good/pleasure) improved in B-E compared with the other three groups and mental effort in B-E dropped compared with the F-E group from the start to middle morning. These results suggest that the combination of breakfast and exercise is beneficial to enhance good feeling and make adolescents feel ready to engage in a task. Finally, there is recent evidence that intake of cocoa/cocoa products, as a source of dietary polyphenols (primarily flavonols), can acutely improve cognitive abilities in young adults and children [47]. This includes in areas such as attention, working memory and processing speed with improved cerebral blood flow or cerebral blood oxygenation postulated as potential mechanisms for improvement. Cocoa was present in both the Milo (7%) and hazelnut spread (7.4%) used in the breakfast in the present investigation and may have contributed to the acute improvements in cognition observed. The extent to which this occurred is speculative without a non-cocoa control breakfast for comparison. Moreover, other studies showed effects of low GI breakfasts, with or without the presence of exercise, on similar cognitive outcomes [8,25]. Nevertheless, the addition of cocoa products to potentiate the effects of breakfast on cognition in adolescents is a potentially exciting area for future research.
The exercise protocol used was a combination of a short period of aerobic exercise and bilateral coordinative ball exercises for several reasons. Firstly, current evidence suggests a relationship between both cardiorespiratory fitness and aerobic physical activity with performance on cognitive tests and academic achievement tests [17,21]. Secondly, physical education lessons in many countries, such as Singapore, emphasize a range of activities that combine coordinative exercise and physical workout and our protocol part-imitated both demands. Finally, the bilateral exercises employed emphasize the ability to balance, react, adjust and differentiate to different stimuli within a short time. These exercises were observed to elicit greater post-exercise increases in attention and concentration for adolescents, compared with their counterparts instructed to undertake moderate intensity exercise without a specific coordinative emphasis in a physical education class [24]. It is postulated that bilateral exercises activate the cerebellum and areas of the prefrontal lobe related to mathematics processing and reading comprehension in children to a greater extent than exercise without the same coordinative emphasis [24]. The improvements observed with the combined intervention confirm this previous observation that this type of physical activity may have a role in eliciting short-term improvements in cognition and academic performance. Importantly, the bilateral coordinative exercises in the intervention were rated as “light” (RPE of 11) by the participants suggesting that exercise does not need to be of a more vigorous nature to achieve short-term improvements in these areas.
There are several limitations to the present study. The parallel group study design means that individual differences in the rate of learning may have affected our results. However, the randomization process should have limited this impact and potential issues with learning and ceiling effects on tests prevented use of a repeated measures design. Although the combined intervention improved mathematics computation time and reaction time on the Go/No-Go Task compared with exercise alone, most measures for the combined intervention did not differ from the breakfast or exercise given in isolation. Our initial power analysis suggested we would require 128 adolescents, or 32 per group, to elicit a worthwhile change from the control intervention. Thus, we cannot rule out the possibility of a type II error when comparing our combined intervention with the individual interventions. Given the consensus within the existing literature, however, it seems likely that both breakfast and exercise are beneficial for cognition and academic performance in adolescents. Nevertheless, the primary aim of the study was to examine the improvements brought about by the combined intervention, rather than the individual interventions themselves where a substantial body of literature already exists, which is why the B-E group was used as the reference. We did nevertheless conduct an exploratory analysis using F-S as the control intervention and this confirmed that the only difference observed for our major outcomes was in the B-E group compared with F-S, with no differences between F-S with the two individual interventions.
We did not record the usual breakfast consumption of the adolescents. It is possible that the acute effects of breakfast on cognitive and academic outcome measures vary depending on the extent to which individuals are habituated to breakfast intake. We also chose to give the exercise before the breakfast for reasons of digestive comfort. The sequence of the interventions may be important in eliciting the changes we observed and, in reality, children may eat breakfast before actively transporting themselves to school. Finally, practical reasons such as participants transporting to the laboratory and performing baseline measures meant that, in real-time, exercise (09:30 a.m.) and breakfast (10:00 a.m.) were relatively late in the morning. This extended the period of fasting and meant that breakfast feeding probably did not occur in conjunction with the normal daily routine, both of which could have affected our outcome measures.
CC BY 4.0. "Breakfast and Exercise Improve Academic and Cognitive Performance in Adolescents" by Kawabata M, Lee K, Choo HC et al., Nutrients (2021). https://doi.org/10.3390/nu13041278. Licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Peer-reviewed and open access; full text from PubMed Central (PMC8068805). Headings and layout adapted for reading; figures and tables are in the original.