Decline in executive control during acute bouts of exercise as a function of exercise intensity and fitness level

Abstract

Studies on the effects of acute bouts of cardiovascular exercise on cognitive performances show contradictory findings due to methodological differences (e.g., exercise intensity, cognitive function assessed, participants’ aerobic fitness level, etc.). The present study assessed the acute effect of exercise intensity on cognition while controlling for key methodological confounds. Thirty-seven participants (M_age = 23. 8 years; SD = 2.6) completed a computerized modified-Stroop task (involving denomination, inhibition and switching conditions) while pedalling at 40%, 60% and 80% of their peak power output (PPO). Results showed that in the switching condition of the task, error rates increased as a function of exercise intensity (from 60% to 80% of PPO) in all participants and that lower fit individuals showed increased reaction time variability. This suggests that acute bouts of cardiovascular exercise can momentarily alter executive control and increase performance instability in lower fit individuals.

Introduction

The effect of acute bouts of cardiovascular exercise on cognition has been a subject of increasing interest over the past few years. These studies can help understand the impact of physical efforts on cognition as required in many everyday life situations (e.g., housekeeping, home repairs, as well as some employment such as firefighters and construction workers). The ever advancing quest for better athletic performances, the valorisation of the sports industry as well as the evolution of neuroimaging equipment and techniques (Acevedo & Ekkekakis, 2006) also contribute to stimulate research in this field.

So far, studies on the effects of acute cardiovascular exercise on cognition have lead to inconclusive results. Some studies have shown that an acute bout of exercise can momentarily enhance cognitive performances (McGlynn et al., 1977, McMorris and Graydon, 1997, McMorris et al., 1999, McMorris et al., 2003, Pesce and Audiffren, 2011, Pesce et al., 2007), while others showed deleterious effects of exercise on cognition (Audiffren et al., 1998, Audiffren et al., 2009, Del Giorno et al., 2010, Dietrich and Sparling, 2004, McMorris et al., 2009, McMorris and Keen, 1994, Pontifex and Hillman, 2007). Some studies have showed mixed results, with both beneficial and negative effects or no impact of exercise (Chmura et al., 1994, Davranche et al., 2009, Davranche and McMorris, 2009). Finally, some reports suggest that acute exercise exerts no effects on cognition (Fery, Ferry, Vom Hofe, & Rieu, 1997). Theoretical explanations have been put forward to account for changes in cognitive performances during acute bouts of cardiovascular exercise. Cognitive-energetic models suggest an explanation for both positive and negative effects of exercise on cognition (Audiffren, 2009). Moderate duration exercise is thought to improve cognition either by increasing the amount of available resources (Kahneman, 1973) or by increasing arousal and/or activation (Humphreys and Revelle, 1984, Sanders, 1983). Given the limited nature of the brain’s resources, the negative impact of exercise could be due to resources sharing between the exercise and the cognitive task (Hockey, 1997, Kahneman, 1973, Sanders, 1983).

In line with this view, the transient hypofrontality theory suggests that strenuous exercise causes a change in physiological state, which momentarily disrupts brain homeostasis. In reaction, the brain modifies its resources allocation. In fact, maintaining high intensity physical exercise increases neural resources demand in some brain regions, which can reduce essential metabolic resources (e.g., oxygen and glucose) availability in other brain regions. According to this theory, the prefrontal cortex does not play a critical role in maintaining high intensity exercise and would therefore be massively affected by the reduced resources availability (Dietrich, 2006, Dietrich, 2009, Dietrich and Audiffren, 2011, Dietrich and Sparling, 2004).

Brain imaging studies partly support the transient hypofrontality theory and suggest that additional physiological mechanisms are involved. In fact, near infrared spectroscopy’s (NIRS) studies generally show a decrease in oxygenation in the prefrontal lobes during intense exercise (Bhambhani et al., 2007, Ekkekakis, 2009). A possible explanation for this phenomenon is that at high intensities, hyperventilation causes a decrease in arterial carbon dioxide pressure (PaCO₂) which in turn causes a constriction of cerebral blood vessels and ultimately a decrease in cerebral perfusion (Linkis et al., 1995). Consequently, acutely increasing exercise intensity should impair executive control, which relies on the functional integrity of the frontal cortex. Contemporary conceptions of executive control functions suggest that elementary mechanisms such as inhibition (i.e., inhibition of a salient response), switching (alternating mentally between sets of rules) and updating (information updating and monitoring in working memory) play a critical role in controlling and modulating other cognitive processes such as memory and reasoning (Levine et al., 2008, Miyake et al., 2000). However, few studies have distinguished these elementary mechanisms when assessing the effects of acute exercise on cognition.

Finally, some researchers suggest that neuroendocrinological changes could also be potential underpinnings of the effect of acute cardiovascular exercise on cognition via the influence of catecholamines (i.e., dopamine, noradrenaline and adrenaline) and cortisol. At high exercise intensities, an increase in cortisol level can lead to an important increase in arousal, which might then lead to impaired cognitive performances. The catecholamine increase can lead to a preferential activation of the limbic system at the expense of the prefrontal lobes and therefore cause a breakdown in performance in executive control tasks (McMorris, 2009).

Whether these explanations are complementary or not will necessitate additional empirical data. So far, even rigorous meta-analytic and narrative reviews of studies assessing cognitive performances during acute bouts of cardiovascular exercise report contradictory findings (Brisswalter et al., 2002, Chang et al., 2012, Lambourne and Tomporowski, 2010, McMorris et al., 2011, Tomporowski, 2003, Tomporowski and Ellis, 1986). For example, most of them state that the moment at which the cognitive task is performed (during, immediately after or after a delay following exercise) has an impact on cognition. Yet, others (Chang et al., 2012) claim the exact opposite and therefore run counter to most theories & hypotheses (e.g., THT) as well as many individual well structured empirical studies which have clearly predicted or shown differences in neurocognitive performances according to this variable.

Important methodological differences across studies can at least partly explain diverging results. A first methodological concern has to do with the various cognitive processes that have been tested during exercise. For example, Audiffren, Tomporowski, and Zagrodnik (2008) and McMorris et al. (2003) have shown positive effects of acute exercise on processing speed using simple and choice reaction time tasks (RT) while others (Davranche and McMorris, 2009, McMorris et al., 2009, Pontifex and Hillman, 2007) have reported deleterious effects on inhibition tasks. Cognitive processes underlying the completion of the cognitive task thus appears a major factor that needs to be taken into account when studying the impact of acute bouts of cardiovascular exercise on cognitive performances.

A second methodological concern is the experimental design with regard to physical exercise. Some researches (McMorris et al., 2003, Paas and Adam, 1991) reported improvement in processing speed at high intensities during a continuous cardiovascular exercise. In other studies that included rest periods between bouts of exercise, processing speed was slower during exercise compared to rest (Audiffren et al., 1998). Therefore, even when the same cognitive process is being tested (e.g., processing speed), different experimental designs (e.g., whether baseline performances are assessed at rest or during exercise) can lead to diverging results.

The intensity at which exercise is performed is another source of confound across studies. Many studies assessed the effect of moderate exercise intensity on cognition, but few studies compared the impact of low and high intensity exercise (McMorris et al., 2011). Moreover, the index of exercise intensity greatly varies among studies: percentage of heart rate reserve, maximal heart rate (HR_max), peak power output (PPO), maximal oxygen uptake ($\dot{V}{O}_2\max$), lactate threshold, adrenaline threshold, ventilatory threshold, etc. Even among studies that used the same index (e.g., HR_max) (Dietrich and Sparling, 2004, Pontifex and Hillman, 2007), exercise intensity levels varies across studies (i.e., 70–80% vs. 60%, respectively).

Few studies have considered participants’ aerobic fitness level as a potential confounding factor in the relationship between acute cardiovascular exercise and cognition (Pesce, 2009). This factor has been underlined in recent meta-analyses (Lambourne and Tomporowski, 2010, McMorris et al., 2011). Previous reviews also strongly recommended that aerobic fitness be measured and considered into the analyses (Brisswalter et al., 2002, Tomporowski, 2003, Tomporowski and Ellis, 1986). In support for this, it has been observed that higher fit individuals show larger improvement in cognitive performances during high intensity exercise than lower fit individuals (Brisswalter, Legros, & Delignières, 1994). Nevertheless, the moderating role that fitness level might have on the effect of acute exercise on cognition is not fully understood. Some evidence suggests that a high fitness level is associated with a maintained oxygen saturation level in the prefrontal cortex during high intensity exercise. In fact, despite a decrease in PaCO₂, oxygen (O₂) saturation tends to increase in the prefrontal lobes of athletes during intense cycling (Nielsen, Boushel, Madsen, & Secher, 1999). A recent review (Rooks, Thom, McCully, & Dishman, 2010) confirms this observation and suggests that the drop in oxygen values in the prefrontal cortex observed at very hard intensities is seen in untrained individuals, but not in those who are aerobically trained. Therefore, this suggests that cognitive performances might be less impaired during high intensity exercise in higher fit individuals. Second, a difference in biochemical changes during moderate to intense exercise has also been hypothesized to explain the modulating effect of fitness level (Brisswalter, Arcelin, Audiffren, & Delignieres, 1997). In fact, urinary catecholamine concentration gets more important after the exercise in lower fit than in higher fit individuals (de Diego Acosta et al., 2001). As stated earlier, an important catecholamine secretion could preferentially activate the limbic system at the expense of the prefrontal lobes and therefore negatively affect cognitive performances in lower fit individuals. A third explanation combines the effect of fitness and emotional regulation. In fact, considering their scarce exposition to high intensity exercise and to the negative symptoms associated with it (i.e., hyperventilation and perhaps nausea) lower fit individual might be more anxious than their higher fit counterparts and therefore perform less well at higher intensities (McMorris, 2009). Even though these hypotheses still need to be empirically tested, all three lead to the same prediction and suggest that a high fitness level could attenuate deleterious effects of intense cardiovascular exercise on cognition.

In sum, variability among studies limits definitive conclusion on the effect of acute cardiovascular exercise on cognitive functions. Sources of variability include exercise intensity, cognitive processes underlying task completion, experimental design as well as participants’ aerobic fitness level. The primary goal of the present study was to assess effects of exercise intensity on cognition by comparing cognitive performances at different intensity levels (light, moderate and intense). The second aim was to assess the effect of exercise on cognition while taking into account key methodological factors by 1 – using a computerized modified-Stroop task that allows to dissociate executive control from processing speed within the same task and 2 – controlling for aerobic fitness level (based on a direct measure of $\dot{V}{O}_2\max$).

Based on the brief literature review, we expect that higher intensities should yield greater impact on cognition than lower workloads. Moreover, tasks relying more heavily on the prefrontal lobes (executive control condition) will be negatively altered during acute bouts of cardiovascular exercises. Finally, higher fit individuals should exhibit less deleterious effects than their lower fit counterparts.