Exercise promotes brain health: a systematic review of fNIRS studies

Exercise can induce brain plasticity. Functional near-infrared spectroscopy (fNIRS) is a functional neuroimaging technique that exploits cerebral hemodynamics and has been widely used in the field of sports psychology to reveal the neural mechanisms underlying the effects of exercise. However, most existing fNIRS studies are cross-sectional and do not include exercise interventions. In addition, attributed to differences in experimental designs, the causal relationship between exercise and brain functions remains elusive. Hence, this systematic review aimed to determine the effects of exercise interventions on alterations in brain functional activity in healthy individuals using fNIRS and to determine the applicability of fNIRS in the research design of the effects of various exercise interventions on brain function. Scopus, Web of Science, PubMed, CNKI, Wanfang, and Weipu databases were searched for studies published up to June 15, 2021. This study was performed in accordance with the PRISMA guidelines. Two investigators independently selected articles and extracted relevant information. Disagreements were resolved by discussion with another author. Quality was assessed using the Cochrane risk-of-bias method. Data were pooled using random-effects models. A total of 29 studies were included in the analysis. Our results indicated that exercise interventions alter oxygenated hemoglobin levels in the prefrontal cortex and motor cortex, which are associated with improvements in higher cognitive functions (e.g., inhibitory control and working memory). The frontal cortex and motor cortex may be key regions for exercise-induced promotion of brain health. Future research is warranted on fluctuations in cerebral blood flow during exercise to elucidate the neural mechanism underlying the effects of exercise. Moreover, given that fNIRS is insensitive to motion, this technique is ideally suited for research during exercise interventions. Important factors include the study design, fNIRS device parameters, and exercise protocol. The examination of cerebral blood flow during exercise intervention is a future research direction that has the potential to identify cortical hemodynamic changes and elucidate the relationship between exercise and cognition. Future studies can combine multiple study designs to measure blood flow prior to and after exercise and during exercise in a more in-depth and comprehensive manner.


Introduction
Exercise intervention is a convenient and adaptive approach to effectively enhance the cognitive function and emotion of individuals (Verburgh et al., 2014;Kawagoe et al., 2017).Indeed, an increasing number of studies have demonstrated its beneficial effects on the healthy development of brain function (Mandolesi et al., 2018;Chen, 2020).Recent studies have predominantly focused on the variations in cognitive function and brain functional activity, such as cerebral blood flow, before and after exercise intervention (Fujihara et al., 2021;Kim et al., 2021;Zhang et al., 2021).Exploring real-time alterations in cerebral blood flow during exercise interventions can reveal hemodynamic changes (Endo et al., 2013;Eggenberger et al., 2016;Carius et al., 2020) and execution (Chen et al., 2017;Coetsee and Terblanche, 2017;Yang et al., 2020) and enhance our understanding of the mechanism underlying the effects of exercise on the brain.
The development of functional near-infrared spectroscopy (fNIRS) has enabled the exploration of hemodynamic changes in cerebral blood flow during exercise interventions.Specifically, it allows non-invasive monitoring of brain tissue oxygenation and hemodynamics (Hoshi, 2005) and possesses distinct advantages over other neuroimaging modalities, such as electroencephalography (EEG) and functional magnetic resonance imaging (fMRI).In addition, it balances both temporal resolution and spatial resolution and is comparatively less sensitive to motion (Leff et al., 2011;Scarapicchia et al., 2017).Previous exercise intervention studies using fNIRS devices largely focused on exercise interventions such as walking (Hamacher et al., 2015), posture, and walking (Herold et al., 2017), which are practical within the laboratory setting.Given the diversity in experimental designs, the effects of exercise on the brain exhibit substantial variability.
The application of fNIRS in the field of sport and exercise psychology is heterogeneous due to variations in the utilization of fNIRS and experimental design.Therefore, to improve uniformity across different studies investigating the influence of exercise on brain functional activity, this review aimed to examine studies that employed near-infrared spectroscopy to detect changes in brain hemodynamics before, during, and after exercise.The purpose of this review was as follows: (1) offer recommendations regarding study designs and research related to fNIRS technology in exercise intervention studies; (2) analyze the designs of various exercise protocols and compare the results obtained after or during exercise; and (3) evaluate the characteristics of changes in cerebral blood flow after and during exercise.Overall, the objective of this review was to investigate the effects of various exercise interventions on alterations in brain functional activity from different perspectives (before and after exercise vs. during exercise).

Methods
This systematic review was performed and reported in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) guidelines (Page et al., 2021) and the Cochrane Collaboration Handbook (Higgins et al., 2019).

. Search strategy
Two reviewers (J.M.H. and T.X.) conducted an independent literature search to screen related studies.The third reviewer, Q.Q.S., resolved disagreements by arbitration.
Scopus, Web of Science, PubMed, CNKI, Wanfang, and Weipu databases were searched from inception to June 15, 2021.The keywords were (Verburgh et al., 2014) exercise (physical activity, exercise, fitness, and sport) and ( 2) fNIRS (functional near-infrared spectroscopy).These terms were consistently applied across each database, serving as the main topic and free-text words in the title.

. Eligibility criteria
Studies were considered eligible if they fulfilled the following criteria: (1) the subjects were healthy; (2) the articles were published in the English language or Chinese language in peer-reviewed journals; (3) exercise-related intervention studies utilizing large muscle groups of the whole body; and (4) at least one cerebral cortical blood flow change was assessed using fNIRS.
Our review focused on the effect of exercise interventions on common healthy participants.The exclusion criteria were as follows: unclear exercise protocols, exercise protocols not designed to improve brain or cognitive health (e.g., exercise test to exhaustion), and studies involving combined interventions (e.g., nutrition and cognition).To ensure generalizability, research utilizing clinical samples (e.g., overweight/obese) and those examining special groups (athletes or people with long-term exercise habits) were excluded.

. Data extraction
Duplicated studies screened from the database search and reference lists were initially excluded.Next, the titles and abstracts were individually evaluated by two authors (J.M.H. and T.X.) to further exclude articles based on the eligibility criteria.Afterward, the two authors independently evaluated the articles.Disagreements were resolved by discussion and consensus among the three authors (Q.Q.S., J.M.H., and T.X.).
The two authors independently extracted the following data from eligible studies: (1) basic information, including the year of publication, participant characteristics, and study design; (2) study design, including study group or condition design, fNIRS state (resting-state or task-design), physiological outcome index, and behavioral outcome index; (3) fNIRS device parameters, including types of fNIRS devices, fNIRS sampling frequency, number of light emitting diodes, laser diodes, channels, fNIRS instrument location and area of interest, and position/arrangement and placement of the light source and detector; (4) the exercise intervention design, covering exercise type, exercise intervention period, frequency of exercise, exercise intensity, and single intervention duration; and (5) the primary endpoints of the studies.  .

Risk of bias assessment
The risk of bias in selected studies was independently assessed by two authors (J.M.H. and T.X.) using the Cochrane Collaboration Risk-of-Bias tool (Higgins et al., 2011(Higgins et al., , 2019)).Disagreements were resolved by discussion with another author (Q.Q.S.) to achieve consensus (see Table 1 and Figure 1).

Study selection and characteristics
The search process is detailed in a flow chart illustrated in Figure 2. The search strategy yielded 6,220 studies from the predefined databases.After excluding duplicates and reviewing the full text, 69 studies met the criteria based on the consensus reached by the reviewers.From these, 22 eligible articles were included in the first category (cerebral hemodynamics were measured before and after exercise) and 8 in the second category (cerebral hemodynamics were measured during exercise).Among them, one study was simultaneously in both categories.

. Quality of included studies
The details on the quality of the included studies in bias risk assessment are summarized in the supporting material.Of note, 28 studies did not provide details on selective reporting, 27 studies reported no other biases, 23 studies reported complete outcome data, 15 studies reported random sequence generation, 1 study reported allocation concealment, 1 study reported blinding of participants and personnel, and 1 study reported blinding of outcome assessment.

. Study design
Twenty-two studies measured cerebral hemodynamics before and after exercise interventions, eight studies (including only adults) documented cerebral hemodynamics during the exercise intervention, and one study recorded cerebral hemodynamics before, during, and after the exercise intervention.
Only one study measured hemodynamic changes and activity in the resting state.In this particular study, baseline brain activity was assessed in the seated position for 5 min (Endo et al., 2013).

. . Study design involving measurements during exercise interventions
In this category (see Table 3), eight studies presented data on cerebral hemodynamic activation during the exercise intervention.Of note, all studies used exercise tasks to investigate the task design.Most studies either used block designs or did not specify the design, whilst few studies provided a detailed description of the design of exercise tasks.As anticipated, studies adopting a block design employed relatively short durations for each block, similar to the cognitive task, ranging from 20 to 40 seconds.

. fNIRS devices . . Measurements before and after exercise interventions
Most included studies conducted fNIRS tests before and after a single, long-term exercise intervention using eleven different fNIRS devices.The device sampling frequency ranged from 1 to 50 Hz, with the majority of devices utilizing 16 emitting diodes and 16 laser diodes.The number of channels ranged from 2 to 48.

. . Measurements during exercise interventions
fNIRS was conducted during acute exercise interventions (see Table 5) using four distinct fNIRS devices were used.The device sampling frequency ranged from 1 to 1,000 Hz, and 8 emitting diodes and 8 laser diodes were employed in the majority of the studies.The number of channels ranged from 4 to 24.

. Exercise intervention
All exercise interventions were categorized into three types according to their frequency and duration, regardless of study design.In other words, they were measured before and after longterm exercise interventions (n = 5), measured before and after one-time exercise interventions (n = 17), and measured during one-time exercise interventions (n = 8).Among them, merely one study presented data before, during, and after acute exercise interventions.Major confounding factors adjusted for across these studies included exercise type, duration, intensity, frequency, and duration of activity.

. . Design of measurements before and after exercise interventions
Five studies investigated hemodynamic changes before and after long-term exercise interventions.Since before-after tests were used, the influence of exercise on fNIRS imaging results was not considered.A broad range of exercise interventions was implemented in these studies, including walking (Coetsee and Terblanche, 2017), Tai Chi Chuan (TCC) (Yang et al., 2020), Baduanjin mind-body (BMB) (Chen et al., 2017), tennis (Lai et al., 2020) or interactive cognitive-motor video game dancing (DANCE), and balance and stretching training (BALANCE) (Eggenberger et al., 2016).The exercise intervention period lasted 8 weeks in most studies (Eggenberger et al., 2016;Chen et al., 2017;Lai et al., 2020;Yang et al., 2020), with only one study extending to 16 weeks (Coetsee and Terblanche, 2017).The frequency of exercise ranged from 2 to 5 times a week.Exercise intensity was classified into three categories: low, moderate, and high.Most studies employed moderate exercise intensity, except for one study that did not report data on intensity (Chen et al., 2017) and one that used moderate-vigorous (Coetsee and Terblanche, 2017)  The duration of a single intervention ranged from 30 min to 90 min.Details are listed in Table 6.Seventeen studies measured cerebral blood flow before and after acute exercise interventions (see Table 7).Exercise types involved cycling and running in the majority of studies, with the exception of one study that incorporated push-ups (Miyashiro et al., 2021).The duration of a single intervention varied from 10 min to 30 min, with seven studies employing a 10-min duration (Yanagisawa, 2010;Hyodo, 2012;Byun et al., 2014;Wen et al., 2015a,b;Kujach, 2018;Kim et al., 2021), five studies opting for 15 min (Endo et al., 2013;Hashimoto et al., 2018;Ji et al., 2019;Stute et al., 2020;Fujihara et al., 2021), two studies using a 20min duration (Jiang and Wang, 2016;Miyashiro et al., 2021), one study implementing a duration of 25 min (Xu et al., 2019), and two studies extending to 30 min (Lambrick et al., 2016;Zhang et al., 2021).Lastly, exercise intensity was mostly moderate.

. . Design of measurements during acute exercise intervention
In the one-time exercise interventions, eight studies measured fNIRS during the exercise intervention (see Table 8).These studies mainly selected exercise interventions involving minimal head movement, such as cycling (Endo et al., 2013;Auger et al., 2016;Kriel et al., 2016;Monroe et al., 2016), basketball slalom dribbling (Carius et al., 2020), barbell squats (Kenville et al., 2017), and walking (Kurz et al., 2012;Herold et al., 2019).Moreover, most studies implemented cycling and walking interventions, while four studies used cycling (Endo et al., 2013;Auger et al., 2016;Kriel et al., 2016;Monroe et al., 2016), and two studies used walking (Kurz et al., 2012;Herold et al., 2019).The duration of the intervention ranged from 10 min to 25 min.While moderate intensity was used in most of the eight studies, some studies did not report exercise intensity and instead reported data on the exercise load.

. . Changes in brain functional activity before and after exercise interventions
Five studies investigated cerebral blood flow after long-term exercise interventions (see Table 9).One study measured oxy-Hb, deoxy-Hb, and total Hb levels (Coetsee and Terblanche, 2017), whilst the remaining four studies measured oxy-Hb levels (Eggenberger et al., 2016;Chen et al., 2017;Lai et al., 2020;Yang et al., 2020).After the long-term intervention, oxy-Hb levels were  increased in the left PFC during the flanker and N-back tasks.
Likewise, deoxy-Hb levels were increased in the left PFC during the Stroop task across almost all studies.One study used a walking task and described that oxy-Hb levels were higher in the left PFC and right PFC during walking.
During the acute intervention, eight studies explored changes in various exercise conditions.Three studies explored changes in different cycling intensities and noted that oxy-Hb levels were increased in bilateral PFC during exercise at the intensity of 60% EX max (Endo et al., 2013) and that under all three conditions of rest, 40%, and 80% intensity levels.The subjects at rest exhibited significantly lower extracerebral and cerebral deoxy-Hb levels compared to values measured at the 80% intensity level of exercise.Furthermore, another study detected significant alterations in the hemodynamic response in almost all channels, with increased oxy-Hb in the bilateral SPL and left PMC after short-distance channel regression (Kenville et al., 2017).At the same time, two studies explored the effects of different levels of exercise difficulties on cerebral hemodynamics.Four studies varied exercise intensity and found an increase in oxy-Hb levels in the bilateral PFC, bilateral M1, SSC, SMA, SPL, left IPL, and right PMC and a concurrent decrease in deoxy-Hb levels in the PFC.One study compared forward walking with backward walking and documented that oxy-Hb levels were higher in the SMA, PCG, and SPL, whereas deoxy-Hb levels were decreased in the SMA.Another study compared brain function in overground and treadmill conditions and observed an increase in oxy-Hb levels in the L-PFC, R-PFC, L-PMC, R-PMC, and B-SMA.One study explored the effect of active recovery on brain function and detected an increase in PFC activity.An earlier study employed BSDT to explore brain function under different basketball dribbling conditions and found that IL-M1 (deoxy-Hb levels) and contralateral PMC-SMA (deoxy-Hb levels) activities were decreased.One study investigated the effect of active recovery on brain function and evinced a significant interaction the initiation and acceleration of walking from the first to the seventh s of walking (t1-7); t10-25, the second timeframe t10-25 represented "steady-state" walking; t26-34, the end of walking with deceleration; t35-46 was chosen to analyze the decrease in oxy-Hb levels below the baseline during the rest phases.↑, compared with the former, the latter was higher than the former; ↓, compared with the former, the latter was lower than the latter.Nonsignificant changes in brain activity were not reported.Frontal_Sup_L: the frontal superior left area, Frontal_Inf_L: the frontal inferior left area, Frontal_Sup_R: the frontal superior right area, Frontal_Inf_R: the frontal inferior area right.
between condition and bout (every 30 seconds of high-intensity exercise is termed a bout) for mean changes in [HHb] across bouts.Within conditions, significant increases in mean [HHb] were observed across bouts, with values progressively increasing over time under HIITPASS and HIITACT conditions.

Study design
Most task-related fNIRS studies aimed to detect activation before and after exercise.Notably, relatively few studies have been conducted on resting-state hemodynamics in this field; only one study examined changes in resting-state hemodynamics before and after exercise (Endo et al., 2013) and did not identify significant changes.Studies performing measurements before and after exercise interventions provide evidence of the effects of exercise on an individual's neural mechanisms.
A minority of the included studies explored cerebral blood flow during exercise interventions (Kurz et al., 2012;Endo et al., 2013;Auger et al., 2016;Kriel et al., 2016;Monroe et al., 2016;Kenville et al., 2017;Herold et al., 2019;Carius et al., 2020), all of which were task-designed and used exercise type as an exercise task.Nevertheless, it is worth mentioning that there are limited available exercise-type options, and they are relatively fixed.Most exercise tasks in those studies adopted the same design as cognitive tasks and controlled the duration of each trial to approximately 30 to 40 seconds (Kurz et al., 2012;Kenville et al., 2017;Carius et al., 2020).However, from a practical perspective, exercise tasks lasting <30 seconds are challenging to implement.Thus, fNIRS should be implemented to develop an exercise task that combines the characteristics of the movement with the feasibility of fNIRS, and suitable methods should be selected to analyze the fNIRS data.Notably, although fNIRS is not as sensitive to motion artifacts as functional MRI and EEG, the rapid motion of any vibrating fiber may lead to substantial changes in the hemoglobin signal.Therefore, when designing exercise tasks, frequent and intense head movements should be avoided.4,5,6,7,9,11,12,13,and 14 (left IFG,PMC,rostral IPL,and SPL).According to the spatial map of the 23 subjects in the pre-exp condition and pre-ctrl condition, also found that channels 2, 3, 4, 5, 6, 7, 9, 10, and 13 (IFG, PMC, and rostral IPL), were activated during both action execution and observation.NR, not reported; pre, pretest; post, posttest; ex, exercise; con, control; exp, experimental; EXmax, maximum voluntary exercise; HIE, high-intensity intermittent exercise; CONT, an acute bout of continuous exercise; INT, an acute bout of intermittent exercise; PE, physical exercise; CE, cognitive exercise; exec, action execution; obsc, action observation.↑,compared with the former, the latter was higher than the former; ↓, compared with the former, the latter was lower than the former.Nonsignificant changes in brain activity were not reported.
Future studies can combine real-time changes in cerebral blood flow before, during, and after exercise in a more indepth and comprehensive manner to establish the relationship ↑, compared with the former, the latter was higher than the former; ↓, compared with the former, the latter was lower than the former.Nonsignificant changes in brain activity were not reported.
between exercise and brain plasticity.In addition, the benefits and mechanisms of exercise interventions on the health of individuals should be explored from multiple perspectives, combining behavioral, physiological, and cerebral assessments. .

fNIRS equipment and parameter settings
The included studies measured cerebral hemodynamics before and after exercise employing a relatively large number of channels.In studies designed to measure cerebral hemodynamics during exercise, the number of channels was relatively low, with the maximum number of channels in the included studies being 24 (Kurz et al., 2012).Attributed to its task specificity, studies that measured cerebral hemodynamics during exercise intervention all used portable devices.
Besides, the selection of ROIs also varied with the study design.Studies designed to measure cerebral hemodynamic changes before and after exercise activity explored the effects of exercise on cognition, with the prefrontal lobe being a key region.In contrast, studies designed to measure cerebral hemodynamics during exercise intervention targeted more locations in the sensorimotor areas (Endo et al., 2013;Auger et al., 2016;Kriel et al., 2016;Monroe et al., 2016;Kenville et al., 2017;Herold et al., 2019;Carius et al., 2020) or other brain areas.
Several exercise processes require a combination of physical activity and cognitive engagement.Hence, it is critical not only to place channels in the motor cortex but also to consider its impact on the prefrontal cortex.Advances in technology have led to an increase in the number of channels for portable devices, thus enabling the measurement of cerebral hemodynamics in multiple ROIs.Given the potential for signal quality issues for measurement during exercise, it is recommended to establish a short-distance channel and instruct participants to minimize head movement during the testing procedure.

. Exercise intervention
Among studies designed to measure fluctuations in cerebral hemodynamics before and after exercise interventions, both longterm exercise interventions and short-term exercise interventions were available.Conversely, few long-term interventions were investigated, most likely due to challenges in conducting the assessment over extended periods.Exercise protocols for long-term interventions were flexible, featuring a diverse range of exercise protocols without overlap between studies and a uniform exercise frequency of 2 to 3 sessions per week (Eggenberger et al., 2016;Coetsee and Terblanche, 2017;Lai et al., 2020;Yang et al., 2020).Only one study applied an exercise frequency of 5 times per week (Chen et al., 2017).Similarly, exercise intensity and duration were relatively consistent, with all being of moderate intensity (Eggenberger et al., 2016;Chen et al., 2017;Coetsee and Terblanche, 2017;Lai et al., 2020;Yang et al., 2020) and each session lasting over 30 min (Chen et al., 2017;Lai et al., 2020;Yang et al., 2020).
At present, studies focusing on the effects of long-term exercise interventions on brain functional activity using fNIRS predominantly aim to detect changes in cognitive-task-related brain functional activity before and after exercise interventions but do not involve the detection of brain functional activity during the exercise task.In studies designed to measure changes in cerebral hemodynamics during acute exercise intervention, all exercise types were relatively easy to implement in the laboratory, requiring minimal activity space and offering flexibility, such as cycling and running.These factors may account for the fact that only three exercise programs were used.In the future, it might be possible to further explore changes in cerebral blood flow during other commonly practiced exercises that involve minimal head movement, such as Tai Chi Chuan and yoga.
To explore hemodynamic changes during exercise, future studies may extend the duration of exercise sessions to 30 min to determine the effect of long exercise session durations on cerebral cortical blood flow.
Achieving consistent results with long-term exercise interventions was challenging due to the diversity of the tasks performed.Interestingly, although the adopted tasks were different, most studies identified changes in the left PFC during inhibitory control tasks (Chen et al., 2017;Coetsee and Terblanche, 2017;Yang et al., 2020).In general, higher hemoglobin activity was observed in the cortical region after the cessation of one-time exercise compared with baseline cortical activity, similar to changes in cortical activation during cognitive tasks.Besides, alterations in the prefrontal cortex and motor cortex were observed during one-time exercise sessions.
Considering that fNIRS signals are significantly impacted by physiological artifacts of the system (Caldwell et al., 2016), the influence of exercise on the cerebral cortex is assumed to mainly arise from physiological confounders after exercise.According to the findings of a methodological investigation, fNIRS signals are affected by systemic physiological artifacts for up to approximately 8 min after stopping cycling for 10 min (Byun et al., 2014).Therefore, the results of studies that performed fNIRS tests after exercise (< ∼8 min) should consider the effect of physiological confounds.However, a reasonable hypothesis suggests that some fNIRS signals observed post-exercise originated from neuronal activity.The entire prefrontal cortex should be impacted by systemic physiological changes if the greater cortical activity observed after exercise is primarily a result of systemic physiological artifacts.The fact that greater cortical activity was observed only in specific regions of the prefrontal cortex rather than the entire prefrontal cortex is significant because it supports the hypothesis that the fNIRS signals were at least partially derived from neuronal activity.The positive neurobehavioural correlation between cognitive ability and cortical activity in various regions of the prefrontal cortex further supports this notion.
The review indicated that exercise interventions alter oxygenated hemoglobin levels in the prefrontal cortex and motor cortex, which are associated with improvements in higher cognitive functions such as inhibitory control and working memory.These findings further support the hypothesis that exercise promotes changes in the prefrontal and motor cortex, which may be key regions for exercise to promote brain health.To deepen our understanding of the fundamental processes by which exercise modifies the brain, future studies should concentrate on alterations in cerebral blood flow during physical exertion.

. Limitations
Nevertheless, this review has limitations that merit acknowledgment.To begin, this review only explored studies involving exercise durations over 10 min, but future studies should expand their scope to explore durations of exercise that contribute to brain health.The primary purpose of this review was to conduct subgroup analysis based on different research designs (measuring fNIRS before, during, and after exercise intervention).Future analyses should aim to stratify group populations or exercise types into different subgroups.The systematic search, dual-author screening, eligibility assessment, and quality appraisal were employed to minimize biased selection of studies, whilst dual-author auditors ensured a thorough search.However, this review was not preregistered and not available for inspection by other researchers, and future reviews can be registered with PROSPERO in advance.A meta-analysis was not conducted due to the high degree of heterogeneity across studies and the low number of studies examining each outcome.

Conclusions and future perspectives
Overall, fNIRS is a promising neuroimaging tool that provides insights into changes in exercise-induced hemodynamics before, during, and after exercise interventions.
Studies using fNIRS to measure cerebral blood flow before and after exercise interventions have predominantly incorporated relatively few long-term interventions and predominantly featured short-term interventions.These studies have mainly focused on the effects of exercise on brain activity during cognitive tasks such as inhibitory control.However, the study equipment and study intervention protocols were less stringent and relatively more flexible.All studies on this topic have focused on changes in oxygenation in the prefrontal area of interest.
Few of the included studies measured hemodynamic changes during exercise, with all of them employing shortterm interventions, relatively fixed exercise intervention protocols, minimal head motion, and short exercise durations.Of note, the exercises were easy to execute in the laboratory setting, and the fNIRS systems had a limited number of channels.
Exercise protocols for long-term interventions encompassed a wide range of exercise protocols and an exercise frequency ranging from 2 to 3 sessions per week.The exercise intensity was moderate, and the duration of a single session exceeded 30 min.Exercise types for long-term interventions were those that necessitated minimal space and were flexible, such as cycling and running.The duration of a single exercise session did not exceed 30 min, with most sessions lasting for 10 min.Finally, exercise intensities were maintained at a moderate level.
The results of this review signaled that exercise interventions alter oxygenated hemoglobin levels in the prefrontal cortex and motor cortex, which are associated with improvements in higher cognitive functions such as inhibitory control and working memory.The frontal cortex and motor cortex may be key regions for exercise to promote brain health.Future research could further focus on changes in cerebral blood flow during exercise to better understand the underlying mechanisms by which exercise influences brain function.
Moreover, due to its insensitivity to motion, the fNIRS is ideally suited for research during exercise interventions.It is paramount to thoroughly scrutinize the study design, fNIRS device parameters, and exercise protocols.Cerebral blood flow during exercise intervention represents a future research direction that has the potential both to identify cortical hemodynamic changes and elucidate the relationship between exercise and cognition.
Future studies can combine two study designs that measure blood flow before, during, and after exercise in a more in-depth and comprehensive manner.Additionally, they can utilize multiple indicators, such as functional connectivity, to accurately reflect the effects of exercise on brain functional networks.
Activated in left IFG, PMC, rostral IPL, and SPL, except SPL during action execution.Action observation and execution in all the no-exercise conditions: the IFG, PMC, rostral IPL, and SPL were significantly activated.Only the activation of the SPL during action execution in the post-ctrl condition and the rostral IPL during action execution in the post-ctrl condition were not significantly activated ROI-based group analysis for the effect of moderateintensity exercise In action observation, during the post-sessions (exp/ctrl), there were significant differences between the exp and ctrl conditions in all four ROIs.In action execution, post-ctrl vs. post-exp: ↑left IFG Zhang et al.
intensity.TABLE Study design of measurement during exercise interventions.
slow, dominant right hand at slow walking pace; NDH slow, non-dominant left hand at slow walking pace; AH slow, alternating hands at slow walking pace; DH fast, dominant right hand at fast walking pace; NDH fast, non-dominant left hand fast walking pace; AH fast, alternating hands at fast walking pace; SIC, sprint interval cycling; CRC, constant resistance cycling; BS, barbell squat; HIITPASS, HIIT with passive recovery; HIITACT, HIIT with active recovery; REC, recovery; BL, baseline values; NR, not reported; NA, not applicable.
,TABLE The fNIRS devices used in the study design for measurement before and after exercise intervention.
TABLE The fNIRS devices used in the study design for measurement during exercise intervention.
TABLE Exercise protocol used in the study design for measurement before and after long-term exercise intervention.
TABLE Exercise protocol used in the study design for measurement before and after acute exercise intervention.
TABLE Exercise protocol used in the study design measurement during acute exercise intervention.
TABLE Changes of brain functional activity before and after long-term exercise interventions.