Physical exercise positively impacts the body [1] and has several well-documented neurobiological effects, such as augmentations of brain vascularization and increases in proteins and neurotransmitters, which promote neurogenesis, neuronal survival, and angiogenesis overall brain volume enhancement [1, 2]. High-intensity intermittent exercise (HIIE) has recently been shown to be an effective alternative to aerobic exercise programs, with positive effects on cognitive functions, such as information processing and response inhibition [3, 4].

Although a substantial body of literature attempts to verify the acute effects of exercise on cognitive performance in children, young or older adults, findings are highly contradictory. For example, there are studies showing positive effects of exercise on cognitive performance [5, 6, 7, 8], while other studies show a negative effect of exercise [9, 10, 11, 12, 13]. In addition, some studies demonstrated no effect of acute exercise (e.g., [13, 14, 15]) on cognitive performance. In terms of cognitive functions, acute exercise primarily improves executive functions (EF) that are associated with the prefrontal cortex, including attention, working memory, problem solving, cognitive flexibility, verbal fluency, decision making, and inhibitory control [16]. These positive changes have been shown to occur at very low to very high exercise intensities [17], with effects lasting up to two hours after exercise [18]. These effects depend on several factors, such as the exercise protocol (intensity and duration), the timing of cognitive testing (during versus post-exercise), the complexity of the tasks, and the fitness level of the participants [10, 17, 19]. Typically, acute effects of exercise are measured on a variety of simple cognitive tasks that target, for example, attention/perception (e.g., reaction time tests) or EFs such as working memory and inhibition (e.g., Flanker task, Stroop Color task, n-back task [20]). The consensus from the literature on acute exercise effects on cognitive performance is a small but positive effect of moderate HIIE, particularly in older adults [21, 22, 23]. However, Etnier et al. [24] noted that we are not dealing with a strong effect.

In general, all the exercises used in these studies increased heart rate, which may influence exercise-induced psychological arousal levels and play an important role in improving EF, possibly by increasing neural activity in the brain [10]. The mechanisms underlying the discussed effects of acute exercise on cognition are still under debate, and different neurophysiological explanations, such as the reticular-activating hypofrontality model [25] or the catecholamines hypothesis [26], emerged to explain these findings. However, there is evidence that the positive effects on cognition are caused by changes in the concentration of specific extracellular neurochemicals (neurotransmitters, neurotrophins, and neuromodulators; [27, 28]). These neurotransmitters (e.g., dopamine, norepinephrine, serotonin, acetylcholine, GABA, and glutamate), neurotrophins such as brain-derived neurotrophic factor (BDNF), insulin-like growth factor-1 (IGF-1), and vascular endothelial growth factor (VEGF), and neuromodulators (e.g., endogenous opioids and endocannabinoids) in turn lead to increased plasticity along with altered synaptic transmission and induction of brain vascularization during physical exercises, particularly affecting prefrontal cortices, which are associated with thinking, decision making, and behavior (executive function hypothesis; [29, 30]). Although increases in various neurochemicals have been confirmed, it is not entirely clear how these neurochemicals are involved in the changes in cognitive function induced by acute exercise [31]. For example, the effect of BDNF on plasticity or vascularization does not immediately lead to changes in cognitive performance. Instead, changes in the factors mentioned above are acute or chronic effects, which are changes that can occur as a result of exercise over a certain duration. Furthermore, vascular remodeling is not an acute process, nor is synaptogenesis. However, the modulation of a neurotransmitter or activation of its receptor on an endothelial cell, which causes vasodilation, are acute effects that may have an immediate impact on cognitive function.

There is a wide variation in the protocols used to deliver acute exercise intervention [32]. Most studies examining the effect of exercise on cognitive function focused primarily on continuous forms of exercise such as running or cycling at a constant intensity (low, moderate, and/or high). The number of studies using an intermittent exercise (IE) protocol is limited [33], although acute HIIE protocols tend to increase concentrations of certain extracellular neurochemicals more significantly than low-intensity acute exercise protocols in healthy young adults [31, 34, 35] and late- and middle-aged older adults [36]. In a prospective randomized controlled trial, Winter et al. [37] demonstrated that a HIIE protocol (2 $\times$ 3 min sprints) had greater benefits for cognition (vocabulary learning; exercise accelerates learning and improves long-term retention of learned material) than moderate-intensity exercise (40 min running) or a sedentary control condition in healthy male sports students. However, this study was conducted in fit adult participants, so its results must be applied with caution to younger and older people, as age and physical activity are moderating variables in the relationship between exercise and cognition ([16], see also Cooper et al. [29]).

The question remains whether or not the benefits of HIIE exercise on cognition, particularly EF, occur in children. Several studies demonstrated positive effects of HIIE on cognitive function in children [8, 38, 39, 40]. Cooper et al. [38] stated that these findings in children are consistent with the literature for adults, suggesting that HIIE has a particularly positive effect on cognitive function. A recent systematic review summarized the existing evidence on the chronic effects of exercise interventions on cognitive function across the life span [41]. This review concluded that there is moderately strong evidence that moderate to high exercise intensities lead to improvements in cognition, especially processing speed, memory, and EF. According to this review article, the most substantial evidence is observed in children between 6 and 13 years and older adults over 50 years. Similar to this review, a meta-analysis by Chang et al. [17] investigating the effects of acute exercise on cognitive performance demonstrated that acute effects may have contributed to larger effects in children and older adults. Since these age groups are vulnerable to exercise, exercise’s acute and chronic positive effects may be observed. However, the meta-analysis by Chang et al. [17] highlighted the difficulty in comparing the intensity and duration of acute physical activity due to different physiological developmental processes in children, young adults, and older adults. Overall, the results of this meta-analysis showed that moderate exercise had a positive effect on cognitive performance during and immediately after exercise and after exercise with a delay. Limitations in the experimental literature on children are that many of the studies have not described the experimental procedure of the exercise interventions (e.g., intensity) in sufficient detail, are often characterized by low methodological rigor, and show procedural differences that are likely due to the usual classroom-based research approaches to exercise and cognition in children (i.e., Tomborowski et al. [42]).

However, acute exercise is likely to affect complex cognitive tasks that engage the executive control system due to the interaction between executive tasks and brain function and exercise-induced stress [43]. The decline in performance is particularly evident during moderate-intensity [44] and high-intensity exercise [45]. Tasks that require EF are related to cognitive flexibility, inhibitory control, and working memory [32] and are more involved in prefrontal cortex functions than other simple tasks such as recall or short-term memory, visual search, or simple and choice reaction time [46]. Such simple tasks require focusing on identifying relevant stimuli and then responding to comparatively simple, predetermined responses [47] and show limited involvement in the function of the prefrontal cortex [48]. In contrast, multiple object tracking (MOT; [49]), in which multiple moving targets are tracked simultaneously, requires continuous task performance with higher-level cognitive functions. Also, MOT tasks resemble various everyday tasks that require tracking multiple objects, such as playing or watching various team sports, crossing a busy street, or driving in traffic. The Multiple Object Tracking (MOT) paradigm, first developed by Pylyshyn and Storm [49], has been used extensively to study MOT in a laboratory setting. Faubert and Sidebottom [50] introduced a perceptive-cognitive training program for athletes called NeuroTracker (NT; CogniSens Athletic, Inc., Montreal, Quebec, Canada). This training program is of great importance for processing information that requires attention and visual-spatial working memory in a dynamic context [50]. It stimulates many brain networks that must work together during exercise, including complex motion integration, dynamic, sustained, and distributed attention processing, dynamic visual-spatial processing, and working memory [51]. 3D-MOT tasks are now widely used in research to investigate dynamic visual attention in different groups of individuals (elderly, healthy controls, and children with neurodevelopmental disorders). For example, a recent study demonstrated a positive relationship between 3D-MOT ability and task performance in elderly participants tested with the NeuroTracker and two driving simulator scenarios. Better NeuroTracker performance was significantly associated with fewer crashes and lane deviations [52]. In another study, older participants performed worse than younger participants on MOT tasks [53]. Pothier et al. [54] tested MOT while walking simultaneously with young and older adults. They found a decrease in performance on the MOT with increasing complexity of the MOT task. An age-related decrease in MOT and walking performance was found, with older adults’ performance impaired under conditions of high attentional load. Similar results were observed in children with neurodevelopmental disorders, in whom repeated 3D-MOT training resulted in better overall attentional performance on the Conners Continuous Performance Task [55].

In summary, much of the literature examining the exercise-cognition relationship focuses on young adults. However, there is a lack of information on children [56, 57], especially compared with young and older adults. In addition, most studies measure cognitive performance at rest but rarely during exercise, and only few studies use HIIE protocols. Although our previous study investigated perceptual-cognitive performance using the NT during physical exercise in terms of a dual-task (DT) paradigm [58], studies examining this during acute exercise across the lifespan are still lacking. To address these gaps, we focus on the effect of a HIIE protocol on perceptual-cognitive performance during exercise in children (CH), young adults (YA), and older adults (OA). In a review by Basso and Suzuki [31], acute exercise was defined as a single bout of physical activity, such as the HIIT protocol used in this study. Therefore, we examine whether the effects of a HIIE protocol on performance in a perceptual-cognitive task differ between CH, YA, and OA. We expected that (1) all groups improve their perceptual-cognitive task performance throughout the HIIE protocol and that (2) YA, in particular, demonstrate better performance in the perceptual-cognitive task than OA and CH at rest and during exercise.

This study aimed to examine the performance of CH, YA, and OA on a perceptual-cognitive task during an HIIE. Considering baseline measures, the group of YA performed best and achieved significantly higher absolute NT scores than CH and OA.

Legault et al. [63] reported similar results on young and older adults in a previous study. This pattern is also consistent with Kennedy et al. [73], who observed an age-related decline in tracking ability and suggested that these findings may indicate a decline in higher-level cognitive functions. In addition, Trick et al. [74] noted that, on average, YA could track four targets at once, whereas OA can track around three targets (see also Alvarez and Franconeri [75]). These results may be related to the development of functions thought to be involved in MOT, mainly working memory [76] and attention [77, 78]. In MOT tasks, participants must continuously monitor (i.e., track) the changing spatial positions of targets and actively maintain target representations over time, which requires visual working memory, another process involved in MOT. Drew et al. [79] showed neural activity, indicating that two separate mechanisms are involved in tracking: an indexing mechanism closely associated with visual working memory and a mechanism that tracks target locations. There appears to be a high degree of inter-individual variability in this tracking threshold, which may link to MOT expertise. For example, the ability of humans to track multiple objects simultaneously increases with age [80]. Various examples from daily life include monitoring children on a playground or in a swimming pool, tracking multiple vehicles and pedestrians while driving, or watching a soccer match. Oksama and Hyönä [81] examined individual differences in MOT performance and found that visual-spatial short-term memory capacity of 2 to 6 items was a significant predictor of MOT, indicating a role for memory in tracking. The well-documented performance limitations (e.g., speed, distances, number of targets) also suggest that a limited attentional resource is available to support tracking. In particular, increasing the difficulty of a MOT task can deplete the attentional resource to the point where a second object cannot be tracked [82].

Since both attention and working memory can be affected by age, it seems reasonable to expect age-related decrements in performance on MOT tasks [83]. Furthermore, the significant difference between CH and YA and OA in the present study is consistent with previous examinations of MOT performance in childhood and across the lifespan [53, 74]. Moreover, Trick et al. [74] and Harris et al. [84] suggest that regular exposure to certain real-life activities, such as action sports and videogames, is associated with higher MOT performance. Therefore, CH likely performs better than OA because they experience such gaming situations and complex dynamic scenes more frequently in their everyday lives. However, Legault et al. [85] pointed out that further experiments will be required to identify which functions may be affected by aging since 3D-MOT is a complex task sensitive to many factors.

Considering NT performance during HIIE, the present study includes DT conditions (i.e., running while MOT) that target participants’ ability to flexibly allocate attentional resources and switch between the goals of the tasks performed in parallel (cognitive flexibility). There is strong evidence that OA demonstrate greater DT costs than YA [54, 86] as they require more time to switch between two different tasks [87]. Thus, these decreases with increasing age and demands of the NT could explain the differences in the absolute task progression between YA and OA. The reticular-activating hypofrontality model (RAH; [48, 88]) hypothesizes that acute exercise shifts the brain’s metabolic resources away from specific regions such as the prefrontal cortex and instead favors structures that support exercise, such as the reticular formation and motor cortices. This process would facilitate sensory and motor task performances, whereas the associated hypofrontality would impair EF. Therefore, the effects of shifting metabolic resources can be greater in OA than CH, resulting in less MOT task progression, particularly in OA. The significant improvement in absolute NT scores in YA during HIIE seems consistent with the notion that acute exercise has a small positive effect on cognitive performance [5, 7]. Notably, since performance in CH and OA remains relatively constant, this may be indicative of age-related differences in potential effects of acute HIIE.

However, the present findings are inconsistent with learning effects in MOT tasks, as previous studies demonstrated continuous task progression in different populations [89], including YA and OA adults [85], concluding that both have a similar ability to improve with training. In the absence of a significant interaction time $\times$ group on NT scores normalized to baseline performance, the present results diverge further with respect to group differences in task progression. This may be related to the wide variability in study designs, experimental and stimulus characteristics, and statistical analyses, which are limitations of research investigating MOT during exercise. The NeuroTracker® must be further tested in terms of its psychometric properties (see Vater et al. [89]). Studies may also be biased by other moderating factors, such as the type, duration and intensity of exercise, physical fitness, and timing and type of cognitive tasks assessed [10, 17, 19, 33]. However, no significant effects of aerobic endurance performance on NT performance were found in our study. Also, this study does not address whether the changes observed in this study are due to training at all or whether they are simply to be expected due to the natural cognitive function that occurs throughout life. However, learning effects seem to have similar albeit different trajectories in CH, YA, and OA, with greater improvements in YA and OA. It is not appropriate to speak of comparison across the life span because of the absence of a middle-aged adult group. In fact, the choices and lifestyle in these middle years likely have the greatest impact on brain health (including cognitive performance) in old age [90].

In conclusion, the present study confirms previous findings on age-related differences in NT performance. Results show that YA perform better than CH and OA at baseline and during HIIE. However, in all three groups, we saw an improvement in NT performance for the absolute scores. Beneficial effects of acute HIIE on perceptual-cognitive performance may be confounded with learning effects and cannot be inferred from mere task progression; therefore, future studies should include a control group without HIIE but with perceptual-cognitive training. Based on the present findings, the effects of different exercise protocols (e.g., continuous vs. intermittent) seem to be a worthwhile subject for future investigations. Normalized speed thresholds should best capture improvement differences between groups to compare results across studies better, as pre-test values are taken as the baseline.

CH, children; HIIE, high-intensity intermittent; exercise; IE, intermittent exercise; MOT, multiple object tracking; NT, NeuroTracker; OA, old adults; YA, young adults.

TJK—Formal analysis, Writing-Original draft, Visualization; SYP—Conceptualization, Methodology, Investigation, Data Curation, Writing-Original draft; VB—Investigation, Writing-Reviewing and Editing; NS—Conceptualization, Methodology, Writing-Reviewing and Editing, Visualization, Supervision.

The study was conducted according to the guidelines of the Declaration of Helsinki. After consultation of the Ethics Committee of the University Stuttgart, the anonymous analysis of data of our participants, which were collected as part of intervention, needed no guidance after the Professional Code for Physicians in Germany (§15 (1)). There were no concerns of the commission about collecting, processing and publishing such data. Also, regarding the guidelines of the German Research Foundation (DFG), no ethics application was required to conduct this study. All younger and older adults and the children and their parents/caregivers gave their written informed consent.

The authors thank the volunteers who participated in the study.

This research received no external funding.

The authors declare no conflict of interest.