Introduction

The increasing aging of the world population and associated degenerative diseases have become public health challenges1. Cognitive decline is a prominent feature of population aging, and an increasing number of studies have investigated strategies to enhance cognitive performance2,3. Exercise has long been recognised as a crucial component of health promotion4. The benefits of exercise are well documented and encompass cardiovascular health5, muscle strength6, and weight management7.

An increasing body of research evidence indicates that exercise can be an effective intervention for improving cognitive performance8,9,10. The underlying neurophysiological mechanism by which exercise enhances cognitive performance is thought to be accelerated brain-derived neurotrophic factor (BDNF) synthesis causing activation of pathways to initiate neuroplasticity and neurogenesis in the hippocampus11,12. Furthermore, research indicates that exercise is associated with enhanced mood states and diminished stress and anxiety levels13,14,15, which in turn indirectly influence cognitive performance.

Prior research has demonstrated that distinct types of exercise modalities elicit disparate effects on cognitive performance. For instance, aerobic exercise demonstrated incremental enhancements in attention, executive function, and memory9. Resistance training exhibited favourable outcomes on overall cognition, cognitive impairment screening measures, and executive function, while exerting no influence on working memory16. In recent years, high-intensity interval training has been the subject of considerable research interest due to its notable benefits for cardiorespiratory and metabolic health17.

The primary appeal of HIIT is its capacity to achieve elevated energy expenditure and cardiorespiratory loads in a condensed timeframe through the alternation of high-intensity exercise with brief periods of rest. This approach offers the benefit of a relatively brief training duration while eliciting comparable physiological adaptations to those observed in longer traditional aerobic training regimens18. And cardiorespiratory fitness is thought to be associated with more effective cognitive function19. For this reason, HIIT has also been identified as a potential method of enhancing cognitive function20.

Northey et al. demonstrated that a HIIT intervention led to a notable enhancement in executive functioning in older adults21. Similarly, Liu et al. showed that a HIIT intervention resulted in a considerable improvement in executive functioning, even in young adults with the highest cognitive abilities22. Nevertheless, Chua et al. demonstrated that a HIIT intervention did not enhance memory in children23. The extant literature does not yield a consensus regarding the impact of HIIT on diverse domains of cognitive performance. The current evidence base is insufficient to draw definitive conclusions. Additionally, the age of participants and the cycle of intervention varied across studies, which may contribute to the inconsistency in findings.

Therefore, this study collected the existing literature on the effects of HIIT on cognitive performance. After screening and extraction, meta-analyses and subgroup analyses of participant age and intervention cycles were performed. The implications of these findings are discussed in order to synthesise them to provide clearer evidence and an evidence-based basis for future research and practice.

Methods

This study followed the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines24. It is registered in the Prospective International Registry for Systematic Reviews (PROSPERO) under the registration number CRD42024577003.

Data collection

This study searched the electronic databases PubMed, Embase, Cochrane Library, Web of Science, Scopus and EBSCO. The search strategy used Boolean operators (‘AND’ and ‘OR’) to link subject terms with free terms, and the following key words and related terms (MeSH) were used for the search: ‘high-intensity interval training’ ‘high-intensity intermittent exercise’ ‘sprint interval training’ ‘HIIT’ ‘cognition’ ‘executive function’ ‘reaction time’ ‘memory’ ‘intelligence’ ‘perception’ ‘cognitive performance’ ‘recall’ ‘mental’ ‘processing’ ‘randomised controlled trial’ ‘RCT’. The search period was from the inception of the respective databases to February 2024, and the language restriction was English. Details of the search methodology can be found in Supplementary Material S1.

Choice criterion

Literature was included if it included the effect of high intensity interval training on cognitive performance and if it was a randomised controlled trial. In this study, high-intensity interval training was defined as intermittent exercise performed at maximal effort, including short or long intervals (ranging from ≤ 45 s to 2 ~ 4 min) at a relative intensity of ≥ 80% VO2max, ≥ 80% HRreserve, or ≥ 85% HRmax, with short rest or active recovery intervals between sessions18. The definition of cognitive performance was based on recognised cognitive domains in the current cognitive psychology and neuropsychology literature, including attention, information processing, executive function, memory and reaction time25,26,27. Animal studies, reviews, conference papers, commentaries and articles for which the full text was not available and data could not be obtained from the original article were excluded. Initially, two authors, KHL and WZ, independently searched and determined the relevance of article titles and abstracts, and then independently reviewed the full text of potentially eligible articles. Any disagreements between the authors were resolved by discussion with HBW.

Cognitive task type

Cognitive performance is a broad and complex concept; therefore, assessing changes in cognitive performance from a holistic perspective may be one-sided and subjective, and may lead to significant heterogeneity problems. To better assess cognitive performance, we divided it according to the types of cognitive tasks identified by Lezak et al.26, see Table 1.

Table 1 Cognitive tasks and cognitive task categories.
Full size table

Data extraction and statistical analysis

Data from the included articles were examined and extracted, including study authors, year, country, sample size of experimental and control groups, sample characteristics, intervention characteristics, study design and cognitive measures. The results were calculated using standardised mean differences (SMDs) and 95% confidence intervals (CIs). The SMDs were used because the included studies used different cognitive tasks to measure cognitive performance. An improvement in results on memory tasks is reflected in an increase in memory scores. For the tasks measuring information processing and executive function, the improvement in scores was reflected in a reduction in reaction time. To correct for effects that were inconsistent with the direction of our meta-analyses, we multiplied the effect size values by -1 to ensure that all effects were in the same direction28. Effect sizes were classified as large (SMD > 0.8), medium (SMD 0.5 ~ 0.8), small (SMD 0.2 ~ 0.5) and moderate (SMD < 0.2)29. Significance levels were set at P < 0.05 and 95% confidence intervals. Due to the expected heterogeneity in the age of the sample, HIIT intervention (intensity and duration) between the included studies, the results of our analyses were analysed using a random effects model30. Heterogeneity between studies was summarised by the Q statistic calculated in a chi-square analysis, and I2 values were calculated to examine inconsistencies between the results of the included studies, with I2 values interpreted as low (25), medium (50), and high (75)31. Risk of bias was assessed for the included literature using the Cochrane Risk of Bias Assessment Tool. Data were analysed using Review Manager software (RevMan, 5.4, The Cochrane Collaboration, 2020).

Results

A total of 1462 studies were searched in all databases, 1442 were excluded based on the inclusion criteria, and a total of 20 studies were included in the final meta-analysis, using a flowchart to document the process of literature screening, see Fig. 1.

Fig. 1
figure 1

PRISMA flow diagram.

Full size image

Study characteristics

A total of 20 studies and data from 981 subjects were included in the meta-analysis. 19 studies reported a mean age of the sample of 35.20 ± 24.04 years, with the majority of effects coming from studies of middle-aged adults (30 ~ 60 years, n = 6 studies), and fewer studies of young adults (20 ~ 30 years, n = 5 studies), children and adolescents (8 ~ 20 years, n = 5 studies), and older adults ( 60 ~ years, n = 3 studies). The meta-analyses included 3 studies on attention measures with 2 attention task paradigms, 8 studies on information processing measures with 3 information processing task paradigms, 15 studies on executive function measures with 8 executive function task paradigms, 7 studies on memory measures with 10 memory task paradigms, and 1 study on reaction time measured in 1 study with 1 reaction time task paradigm. Table 2 documents the characteristics of the studies included in the meta-analysis.

Table 2 Characteristics of all studies included.
Full size table

Risk of bias

The Cochrane Risk of Bias Assessment Tool was used to assess the quality of the included literature49. The risk of bias was assessed in six aspects: selective bias (randomisation, allocation concealment), measurement bias, follow-up bias, reporting bias, implementation bias and other bias, and the level of risk of bias was categorised into three levels: low risk of bias, high risk of bias and unknown risk of bias. The literature included in this study had a low risk of bias, see Fig. 2, and the publication bias of the included literature was examined by drawing a funnel plot, see Fig S1, S2 and S3 in Supplementary Material S2.

Fig. 2
figure 2

Risk of bias.

Full size image

Meta-analysis results

Due to the limitations of the inclusion criteria, the included studies tested cognitive tasks focusing on information processing, executive function and memory. The number of studies testing attention34,40,48 and reaction time42 was small and heterogeneous (I2 = 98%), and performing meta-analyses would have led to unreliable results. Therefore, in the present study, meta-analyses were performed only for information processing, executive function and memory.

Information processing. The analysis of information processing showed a small improvement in information processing with HIIT compared to the control group (SMD = 0.33, 95% CI: 0.15 ~ 0.52, p = 0.0005). The results are shown in Fig. 3.

Fig. 3
figure 3

Forest diagram of information processing. A: Trail Making Test A; B: Digit Symbol Substitution Test; C: Stroop Neutral Test.

Full size image

Executive function. Analysis of executive function showed a small improvement in executive function with HIIT compared to controls (SMD = 0.38, 95% CI: 0.26 ~ 0.50, p < 0.00001). Results are shown in Fig. 4.

Fig. 4
figure 4

Forest diagram of executive function. A: Trail Making Test B; B: Flanker Test; B1: Congruent; B2: Incongruent; C: Groton Maze Learning Task; D: Stroop Congruent; E: Stroop Interference; F: Stroop Incongruent; G: Digital Conversion Tasks.

Full size image

Memory. Analyses of memory showed a small improvement in memory with HIIT compared to controls (SMD = 0.21, 95% CI: 0.07 ~ 0.35, p = 0.004). The results are shown in Fig. 5.

Fig. 5
figure 5

Forest diagram of memory. A: One Card Learning Task; B: One Back Task; C: Two Back Task; D: Digit Span Forward; E: Wechsler Letter-Number Sequencing; F: Digit Span Backward; G: Sternberg Paradigm Test; G1: One-item; G2: Three-item; G3: Five-item; H: Verbal Learning Memory Test Recall; J: International Shopping List Task; K: Rey Osterrieth Fgure Immediate Recall; L: Rey Osterrieth Fgure Delayed Recall; M: Brief Visuospatial Memory Test; N: Wechsler Logical Memory Immediate Recall; P: Wechsler Logical Memory Delayed Recall; Q: International Shopping List Task Recall; R: Rey Auditory Verbal Memory Recall Test.

Full size image

Subgroup analysis

To gain further insight into the impact of HIIT on cognitive performance across different age groups and intervention cycles, the present study analysed subgroups of participants based on age and intervention cycle. In accordance with the delineation method employed in previous studies, the age categories were defined as follows: 8 ~ 20 years, 20 ~ 30 years, 30 ~ 60 years, and 60 ~ years10. With regard to the intervention cycle, the classification was as follows: acute (i.e., completed either immediately or within one day), ≤ 8 weeks, and > 8 weeks17. In order to better accommodate the heterogeneity between the subgroups, a random effects model was employed in the subsequent subgroup analyses.

Age. Subgroup analyses were performed according to participant age. Compared to controls, HIIT had a moderate improvement in information processing for participants aged 60 ~ years (SMD = 0.59, 95% CI: 0.34 ~ 0.84, p < 0.00001). However, there was no statistically significant effect on information processing for participants aged 8 ~ 20 years and 30 ~ 60 years. See Fig. 6. HIIT showed small improvements in executive function for participants aged 8 ~ 20 years (SMD = 0.37, 95% CI: 0.11 ~ 0.62, P = 0.004), 20 ~ 30 years (SMD = 0.42, 95% CI: 0.08 ~ 0.76, P = 0.01), 30 ~ 60 years (SMD = 0.41, 95% CI: 0.14 ~ 0.67, P = 0.003) and 60 ~ years (SMD = 0.35, 95% CI: 0.07 ~ 0.62, P = 0.01). See Fig. 7. In contrast, HIIT showed only a small improvement in memory for participants aged 30 ~ 60 years (SMD = 0.38, 95% CI: 0.19 ~ 0.57, p < 0.0001). There was no statistically significant effect on memory for participants in other age groups. See Fig. 8.

Fig. 6
figure 6

Forest plot of the age-specific subgroups of the participants (information processing). A: Trail Making Test A; B: Digit Symbol Substitution Test; C: Stroop Neutral Test.

Full size image
Fig. 7
figure 7

Forest plot of the age-specific subgroups of the participants (executive function). A: Trail Making Test B; B: Flanker Test; B1: Congruent; B2: Incongruent; C: Groton Maze Learning Task; D: Stroop Congruent; E: Stroop Interference; F: Stroop Incongruent; G: Digital Conversion Tasks.

Full size image
Fig. 8
figure 8

Forest plot of the age-specific subgroups of the participants (memory). A: One Card Learning Task; B: One Back Task; C: Two Back Task; D: Digit Span Forward; E: Wechsler Letter-Number Sequencing; F: Digit Span Backward; G: Sternberg Paradigm Test; G1: One-item; G2: Three-item; G3: Five-item; H: Verbal Learning Memory Test Recall; J: International Shopping List Task; K: Rey Osterrieth Fgure Immediate Recall; L: Rey Osterrieth Fgure Delayed Recall; M: Brief Visuospatial Memory Test; N: Wechsler Logical Memory Immediate Recall; P: Wechsler Logical Memory Delayed Recall; Q: International Shopping List Task Recall; R: Rey Auditory Verbal Memory Recall Test.

Full size image

Intervention cycle. A subgroup analysis of HIIT intervention cycles showed that performing HIIT for > 8 weeks had a moderate improvement in participants’ information processing compared to the control group (SMD = 0.59, 95% CI: 0.34 ~ 0.84, p < 0.00001). Acute and ≤ 8 weeks of HIIT had no statistically significant effect on participants’ information processing. See Fig. 9. Acute (SMD = 0.33, 95% CI: 0.08 ~ 0.58, P = 0.01), ≤ 8 weeks (SMD = 0.45, 95% CI: 0.23 ~ 0.67, P < 0.0001) and > 8 weeks (SMD = 0.40, 95% CI: 0.17 ~ 0.63, P = 0.0008) of HIIT all showed small improvements. See Fig. 10. Acute HIIT had no statistically significant effect on participants’ memory, ≤ 8 weeks (SMD = 0.23, 95% CI: 0.01 ~ 0.44, P = 0.04) and > 8 weeks (SMD = 0.46, 95% CI: 0.16 ~ 0.75, P = 0.002) of HIIT had small improvements in participants’ memory. See Fig. 11.

Fig. 9
figure 9

Forest plot of the subgroups of the intervention cycles (information processing). A: Trail Making Test A; B: Digit Symbol Substitution Test; C: Stroop Neutral Test.

Full size image
Fig. 10
figure 10

Forest plot of the subgroups of the intervention cycles (executive function). A: Trail Making Test B; B: Flanker Test; B1: Congruent; B2: Incongruent; C: Groton Maze Learning Task; D: Stroop Congruent; E: Stroop Interference; F: Stroop Incongruent; G: Digital Conversion Tasks.

Full size image
Fig. 11
figure 11

Forest plot of the subgroups of the intervention cycles (memory). A: One Card Learning Task; B: One Back Task; C: Two Back Task; D: Digit Span Forward; E: Wechsler Letter-Number Sequencing; F: Digit Span Backward; G: Sternberg Paradigm Test; G1: One-item; G2: Three-item; G3: Five-item; H: Verbal Learning Memory Test Recall; J: International Shopping List Task; K: Rey Osterrieth Fgure Immediate Recall; L: Rey Osterrieth Fgure Delayed Recall; M: Brief Visuospatial Memory Test; N: Wechsler Logical Memory Immediate Recall; P: Wechsler Logical Memory Delayed Recall; Q: International Shopping List Task Recall; R: Rey Auditory Verbal Memory Recall Test.

Full size image

Discussion

The meta-analysis examined the effects of HIIT on information processing, executive function and memory in 20 studies from 6 electronic databases, while also analysing the moderating effects of two factors, age and intervention cycle. The results showed small improvements in information processing, executive function and memory with HIIT compared with controls. Of particular note was a slightly greater effect on improving executive function, a higher cognitive ability. These results provide evidence for HIIT as a highly effective exercise modality for improving cognitive function.

Our results showed small improvements in executive function with HIIT in people of all ages. In particular, young people were able to improve even at their highest cognitive age. We also found that the effect of HIIT on executive function improved as the duration of the intervention increased. Executive function is a higher-order cognitive function that includes inhibitory control, working memory and cognitive flexibility, and has important implications across the lifespan50. Hsieh et al. showed improvements in inhibitory control with acute HIIT, and improvements in both inhibitory control and working memory with chronic HIIT51. Consistent with our findings, the benefits of chronic HIIT on executive function were greater compared to acute HIIT.

We also found that HIIT moderately improved information processing in older adults over the age of 60. This may be because the neuroplasticity of the brain declines with age, and HIIT, as a strong stimulus, may have induced more significant changes in neuroplasticity in older adults, resulting in a positive effect52. Foong et al. found that information processing mediates psychosocial stress to some extent in older adults, and that a decline in information processing may further reduce their higher-level cognitive abilities53. Therefore, HIIT may be an intervention to improve information processing and reduce stress in older adults. Notably, acute and less than 8 weeks of HIIT had little effect on information processing, and more than 8 weeks of HIIT showed moderate improvement.Kendall et al. concluded that acute HIIT only increased levels of central arousal and had no effect on processing time42. Further research is needed to investigate the mechanisms by which chronic HIIT affects information processing.

In addition, the study found a small improvement in memory with HIIT in middle-aged people aged 30 ~ 60 years, with no significant effect in other age groups. Acute HIIT had no effect on memory, less than 8 weeks of HIIT had a small improvement in memory, and more than 8 weeks of HIIT improved memory more significantly. This suggests that chronic HIIT is more effective at improving memory. In addition, Iuliano et al. found that exercise intensity was an important factor in improving memory, and that low-intensity exercise of similar duration did not improve memory54. It is important to prevent and delay age-related memory decline, and high-intensity exercise can alleviate age-related decline, induce higher levels of stress hormones, and increase the production of new neurons in the hippocampus to improve hippocampus-controlled learning and memory skills55.

There is a growing body of research on exercise interventions for cognition, and the potential mechanisms by which HIIT, as an emerging form of exercise, produces benefits are not fully understood, but some researchers have suggested that HIIT may work by producing different mechanisms to those found in other forms of exercise. HIIT induces the secretion of pro-angiogenic vascular endothelial growth factor (VEGF), a pro-angiogenic neovascular growth factor that promotes angiogenesis and the formation of new blood vessels56. Exercise-induced VEGF secretion and angiogenesis have also been shown to promote vasodilation via nitric oxide (NO), which plays a role in vascular remodelling. In addition, HIIT also promotes the production of free radicals and related reactive oxygen/nitrogen species (ROS/RNS), which play a role in maintaining cerebrovascular oxygen homeostasis at physiological levels and upregulate the expression of antioxidant enzymes, vascular endothelial growth factor (VEGF), brain-derived neurotrophic factor (BDNF) and insulin growth factor-1 (IGF-1)57. These pleiotropic factors promote neurogenesis, neuronal survival, neuroplasticity and cognitive performance58.

Overall, HIIT provided unexpected benefits on cognitive performance, especially in middle-aged and older populations with age-related cognitive decline. In addition, the intervention cycle was an important factor that had a profound effect on cognitive performance. The physiological mechanisms that produce benefits may also differ between intervention cycles. A single session of acute exercise is usually associated with transient improvements, i.e. increased endorphin release and increased cerebral blood flow, but with concomitant peripheral and central fatigue and therefore no significant improvement in cognitive performance59. Chronic exercise, on the other hand, produces physiological adaptations in the body and improves brain structure and function60. Therefore, chronic and sustained HIIT is feasible for non-pharmacological improvement of cognitive performance. Our findings contribute to a deeper understanding of the relationship between HIIT and cognitive performance and promote the use of HIIT in cognitive health, such as in school-based physical education programmes and clinical rehabilitation therapy.

Limitations and future works

Our meta-analysis has some limitations. There are relatively few studies of HIIT on cognitive performance, resulting in fewer included studies, smaller sample sizes in some studies, and the need for more studies with larger samples to support HIIT if it is to be extended to the general population. In addition, inconsistencies in the intervention methods (exercise protocol, exercise intensity, exercise frequency, exercise duration) of the included studies, inconsistencies in the health status of the participants, and inconsistencies in the testing methods of the outcome indicators are differences that may affect the reliability of the results. Therefore, future studies could investigate the different effects and biological mechanisms of HIIT programmes of different intensities and durations on specific domains of cognitive performance in populations with different health conditions, and identify and develop appropriate exercise programmes for different populations to improve cognitive performance and human health.

Conclusion

This meta-analysis showed that HIIT had a small effect on cognitive performance. In particular, the effect on executive function, a high-level cognitive ability, was greater. For children and adolescents, HIIT is effective in improving executive function, and higher levels of executive function allow for flexible coordination, optimisation and control of cognitive problem-solving processes, leading to improved thinking skills and academic performance, which is of great interest to educators. For middle-aged and older adults, HIIT improves information processing, executive function and memory, which helps to reduce cognitive degenerative diseases. Additionally, chronic HIIT has been demonstrated to exert a more pronounced influence on cognitive performance than acute HIIT. It is recommended that HIIT, as an emerging and highly regarded exercise intervention, should have a greater focus on cognitive and mental health in the context of the current research hotspots in the field of health that focus on “adolescents” and “aging” as objects of study.