In recent years, the prevalence of obesity has become increasingly severe. Research suggests that by 2030, the global rate of overweight and obesity among adults is expected to reach 58%, making it a major public health concern of the 21st century (Kelly et al., 2008). Obesity is associated with a range of physiological conditions, including diabetes, hypertension, and coronary heart disease, contributing to higher morbidity and all-cause mortality rates, and posing a substantial economic burden on society (Friedemann et al., 2012). More concerning is the potential detrimental impact of obesity on cognitive functioning (Miller and Spencer, 2014). Regarding the causes of overweight and obesity, Wood et al. emphasize the critical role of eating behaviors and food-related decision-making (Wood et al., 2016), particularly the excessive intake of high-calorie foods, as key factors contributing to weight gain. For example, the weight gain observed in individuals with obesity has been associated with an automatic approach tendency toward food (Carnell and Wardle, 2008), as well as heightened attentional focus on food cues (Guo et al., 2024).

Inhibitory control refers to the ability to suppress dominant or automatic responses, involving the regulation of attention, behavior, thoughts, or emotions in the face of strong internal or external stimuli. This ability facilitates goal-directed behavior and is considered a core subcomponent of executive function (Crone, 2009; Diamond, 2013). Neuroscientific evidence has shown that individuals with obesity exhibit systematic neurocognitive deficits in executive functioning, with impairments in inhibitory control being particularly pronounced (Yang et al., 2018). These deficits result in a tendency toward impulsive choices during food-related decision-making, characterized by enhanced attentional capture by high-calorie food cues (Do et al., 2024).

From the perspective of neuroplasticity induced by exercise interventions, enhancing inhibitory control in response to food-related stimuli has emerged as a novel target in obesity management (Smith et al., 2011; Miguet et al., 2023; Moschonis and Trakman, 2023). Existing literature suggests that acute exercise has a small-to-moderate positive effect on inhibitory function (Kempton et al., 2011; Chang et al., 2012; Guiney and Machado, 2013; Ludyga et al., 2016; Kao et al., 2018; Ligeza et al., 2018; Yang et al., 2018). However, some studies have reported null effects of acute exercise on inhibitory control (Lambourne and Tomporowski, 2010). Such inconsistencies may be attributed to variations in the intensity and duration of the exercise protocols used across studies (Olson et al., 2016; Wohlwend et al., 2017). Event-related potentials (ERPs) provide a useful tool for examining the neural underpinnings of inhibitory control. The N2 component, occurring approximately 200–350 ms after stimulus onset, is sensitive to conflict detection, with its amplitude increasing as inhibitory demands rise (Folstein and Van Petten, 2008; Larson et al., 2014). The P3 component, typically observed between 300 and 600 ms post-stimulus, is associated with attentional allocation and the engagement of inhibitory resources (Falkenstein et al., 1991; Gajewski and Falkenstein, 2013). In food-specific inhibitory control tasks, individuals with obesity exhibit a pronounced attentional bias toward food stimuli, particularly high-calorie foods. This is evidenced by significantly greater N2 amplitudes in response to food-related images compared to neutral stimuli, suggesting enhanced early attentional resource allocation to high-calorie food cues. Moreover, gender differences have been reported in the neural correlates of food-related cognitive control (Clayson et al., 2011; Melynyte et al., 2017). Specifically, obese females tend to rely more on the prefrontal-insular connectivity for inhibitory regulation, whereas obese males appear to engage the dorsolateral prefrontal–striatal pathways associated with reward processing. Compared to individuals with normal weight, those with obesity typically show attenuated N2 amplitudes and prolonged N2 latencies, indicating deficits in early conflict monitoring, which may compromise subsequent inhibitory control processes. Regarding the P3 component, studies have found that high-calorie food stimuli elicit smaller P3 amplitudes and longer latencies during NoGo trials in individuals with obesity relative to normal-weight counterparts, reflecting reduced recruitment of cognitive resources necessary for inhibition. Additionally, under resting conditions, P3 amplitudes during high-calorie food NoGo trials are significantly greater than those in Go trials, suggesting that inhibiting responses to high-calorie foods demands greater cognitive effort.

Recent studies have demonstrated that acute bouts of moderate-intensity aerobic exercise (MIAE) can indirectly influence eating behaviors by enhancing prefrontal cortical circuits related to inhibitory control, thereby reducing impulsive consumption of high-calorie foods (Carbine et al., 2024). This effect is reflected in increased N2 and P3 amplitudes in response to high-calorie food images. Notably, although traditional physical activity guidelines emphasize MIAE for promoting health benefits, adherence among adults with obesity tends to be low, which presents practical challenges and limits its feasibility in real-world applications (Bull et al., 2020). In addition to MIAE, high-intensity interval exercise (HIIE) has garnered considerable attention and preference, largely due to its “time-efficiency” (Chen and Zhu, 2024; Chen et al., 2025). Research has shown that HIIE may modulate eating behavior through lactate signaling, enhancing regulation along the hypothalamic–prefrontal pathway. Lactate enters the arcuate nucleus via monocarboxylate transporters, where it suppresses AgRP neuron activity and reduces the motivational drive to consume high-calorie foods (Jacob et al., 2023). Furthermore, lactate functions as a “metabolic buffer,” enhancing mitochondrial oxidative phosphorylation efficiency in the prefrontal cortex (PFC), thereby supporting more sustained inhibitory synaptic transmission (De Sousa Fernandes et al., 2022). These findings suggest the potential applicability of HIIE in enhancing food-specific inhibitory control.

Given that individuals with obesity differ significantly from those of normal weight in terms of metabolic status and responsiveness to exercise interventions (Van Galen et al., 2018), indirect evidence alone may be insufficient to formulate targeted recommendations for individuals with obesity. There is a clear need for direct empirical investigations into how exercise interventions modulate food-specific inhibitory control in this population. The aim of the present study is to investigate whether a single session of HIIE, compared to MIAE, serves as a time-efficient alternative—or potentially a superior strategy—for enhancing food-specific inhibitory control in obese adults. Furthermore, the study aims to explore whether the regulatory effects of these two exercise modalities differ by sex.

In terms of behavioral performance related to food-specific inhibitory control, both short-term high-intensity interval exercise (HIIE) and moderate-intensity aerobic exercise (MIAE) facilitated faster reaction times in obese adults. Notably, HIIE exerted a greater enhancement in response speed among obese males compared to MIAE. However, ERP data revealed no significant modulation in N2 or P3 amplitudes in this subgroup following HIIE. This dissociation between behavioral performance and ERP outcomes precludes strong mechanistic conclusions, and we thus interpret these findings with caution. One plausible explanation, based on previous neurophysiological literature, is that HIIE may induce non-inhibitory arousal- or motivation-related changes in neural functioning, potentially involving the prefrontal–striatal circuitry. Supporting this, earlier studies have reported that HIIE acutely elevates catecholamine levels (e.g., norepinephrine) and brain-derived neurotrophic factor (BDNF) in obese individuals (Athanasiou et al., 2023; Ceylan et al., 2023), which may influence motor-associative circuits or reward–motivation pathways in the ventral striatum (Gökçe et al., 2024). While these effects may contribute to faster behavioral responses, further investigation using targeted neurophysiological markers is necessary to clarify the underlying mechanisms (Gökçe et al., 2024). Such pathways can accelerate behavioral output without the recruitment of additional inhibitory resources. These findings align with the neural efficiency hypothesis, which posits that acute metabolic stress induced by HIIE activates the sympathetic–adrenal medullary axis, elevates arousal, and enhances motor cortex excitability, enabling faster responses with less cortical effort (Inoue et al., 2020). Given the regional specificity of N2 and P3 components across cognitive processing stages, with N2 at Fz typically reflecting conflict detection and P3 at Pz associated with higher-order cognitive processes such as attentional allocation, this study focused on examining amplitude changes at these critical sites. While inhibitory control is recognized as a product of distributed neural coordination, the regulatory role of individual regions—especially the frontal cortex—remains fundamental (Paret et al., 2019). The modified food-related Go/NoGo task used in this study incorporated high-calorie, low-calorie, and neutral food images to simulate real-life dietary decision-making contexts, thereby enabling more ecologically valid assessments of food-specific inhibition (Bartholdy et al., 2016). Results showed that high-calorie food images elicited significantly larger N2 amplitudes than both low-calorie and neutral stimuli in both male and female obese participants, reflecting increased conflict monitoring demands evoked by energy-dense cues (Hoyniak and Petersen, 2019). Additionally, NoGo trials elicited greater P3 amplitudes than Go trials, indicating that response inhibition required greater cognitive resources than execution, further supporting the validity of the Go/NoGo paradigm used in this context (Wang et al., 2022).

Given the limited number of studies and the inconsistent findings regarding the acute effects of exercise on food-related inhibitory control, further research using event-related potential (ERP) techniques is warranted. These investigations should aim to determine whether short-term exercise enhances food inhibition by strengthening inhibitory processes toward high-calorie, palatable foods, or by increasing motivational salience toward healthier, low-calorie food choices, thereby indirectly reducing high-calorie food intake (Lowe et al., 2016, 2019). Wang reported that, in males, acute exercise reduced activation in the orbitofrontal cortex (OFC) during food inhibition tasks, which was associated with lower self-reported hunger. This suggests a perceptual role of the OFC in food-related inhibitory processes, a pattern not observed in females, thereby indicating potential sex-specific modulation in food inhibition following exercise interventions. In the current study, under the MIAE condition, obese males exhibited no significant difference in N2 amplitude between Go and NoGo trials for low-calorie food images. Typically, the N2 component reflects conflict monitoring mediated by the anterior cingulate cortex (ACC) or early response inhibition via the prefrontal-parietal network (Folstein and Van Petten, 2008). In classical Go/NoGo paradigms, NoGo trials usually elicit larger N2 amplitudes due to the need to suppress a dominant response—an index of “response–non-response” conflict detection (Folstein and Van Petten, 2008). However, in this study, the Go trials with low-calorie images evoked greater N2 amplitudes than NoGo trials, which may reflect a task-specific redistribution of cognitive resources. When low-calorie images serve as Go stimuli and high-calorie images as NoGo stimuli, obese males—possibly due to prolonged exposure to high-calorie environments—may display an implicit approach bias, or an automatic response tendency toward such cues. MIAE, by enhancing cognitive control via the prefrontal–striatal pathway (Guiney et al., 2015), may prompt participants to allocate greater attentional resources during Go trials to overcome habitual neglect of low-calorie cues, which are typically less motivationally salient. Thus, the elevated N2 amplitude in Go trials may not reflect classical inhibition demands, but rather enhanced ACC-based conflict monitoring related to the initiation of a non-dominant response (i.e., approaching low-calorie food). Furthermore, MIAE may improve prefrontal oxygenation (Tsukamoto et al., 2016) and strengthen dorsolateral prefrontal–parietal connectivity (Legget et al., 2016), facilitating early attentional capture of low-calorie stimuli. The increased N2 in Go trials may also represent more refined perceptual processing—such as the recognition of health-related attributes—requiring ACC-DLPFC cooperation to suppress the automatic attentional bias toward high-calorie cues (Hare et al., 2009). In this context, the increased N2 amplitude reflects a strategic cognitive shift in obese males post-exercise—from passive neglect to active recognition of low-calorie food cues.

In the P3 amplitude analysis at the Fz electrode, female participants with obesity showed no significant difference between NoGo and Go trials under the MIAE condition when responding to high-calorie food images. Typically, NoGo trials elicit larger P3 amplitudes than Go trials, as they represent greater inhibitory demands. This is especially true for high-calorie stimuli, which are known to require elevated cognitive control due to their high reward salience. However, the absence of a significant NoGo–Go P3 difference in this study suggests that, under MIAE, obese females no longer recruited additional inhibitory resources when exposed to high-calorie cues. Previous research has found that acute MIAE attenuates neural responses to food-related stimuli, particularly compared to non-food stimuli, indicating that exercise may play a positive role in reducing attentional allocation to food cues. The present findings extend this perspective, suggesting that MIAE not only modulates attention to food cues, but may also enhance inhibitory responses to high-calorie foods. Moreover, the observed increase in N2 amplitude to low-calorie Go trials among obese females after MIAE implies enhanced conflict monitoring, likely mediated by the anterior cingulate cortex (ACC). This supports the hypothesis that MIAE improves prefrontal oxygenation and insulin sensitivity, which may optimize early ACC-based categorization of food stimuli (e.g., distinguishing low- vs. high-calorie foods). The disappearance of the NoGo–Go P3 difference in high-calorie conditions also indicates a reorganization of late-stage inhibitory resource allocation. This could result from: increased serotonergic transmission via upregulation of tryptophan hydroxylase activity, reducing the reward salience of high-calorie foods (Strasser et al., 2016). Enhanced hippocampal–prefrontal synaptic plasticity, promoting adaptive cognitive strategies such as implicit attentional avoidance, which reduce reliance on late-stage P3-linked cognitive effort (Crabtree et al., 2014). Additionally, MIAE may suppress excessive activation of striatal dopamine D2 receptors, thereby dampening the anticipated reward conflict associated with high-calorie cues. This mechanism could weaken the positive feedback loop in NoGo tasks that typically links inhibitory demand to resource allocation (Robison et al., 2018). In sum, the absence of a NoGo–Go P3 difference for high-calorie stimuli under MIAE may reflect a shift from active suppression to automatic processing in obese females. Although this interpretation is consistent with the neural efficiency hypothesis and previous literature, we acknowledge that it remains a speculative conclusion within the scope of the current data. The ERP results did not provide direct neural markers to validate a shift in inhibitory strategy, such as from active suppression to automatic processing. Therefore, this mechanism should still be regarded as a theoretical hypothesis that requires further validation. Future studies may incorporate additional neurophysiological tools, such as source localization or connectivity analyses, to further clarify the specific neural mechanisms underlying these ERP patterns. This supports the notion that MIAE can induce a neural efficiency reorganization of food-specific inhibitory control, rendering high-calorie stimuli less cognitively prioritized during evaluation, thereby balancing neural responses in the food-related Go/NoGo task.

It should be noted that ERP data were only recorded following each experimental condition (HIIE, MIAE, and resting control), and no pre-exercise baseline was collected. This decision was primarily based on concerns regarding participant fatigue and the potential for practice effects that may confound the comparability of ERP responses across sessions. However, the absence of a pre-intervention baseline introduces limitations in attributing observed ERP differences solely to exercise-induced changes, as session-to-session variability cannot be completely ruled out. Secondly, the initial sample size estimation was conducted using G*Power based on a typical three-condition within-subject design model, without incorporating sex as a predefined factor. As a result, the current sample size (16 participants per sex) may be underpowered to detect sex × condition interaction effects, and related findings should be interpreted as exploratory. Thirdly, although the sample size (n = 32) was statistically justified, the study population consisted exclusively of young obese adults from China, which limits the generalizability of the findings to other populations, such as older adults, individuals with comorbidities, or people from different cultural and ethnic backgrounds. In addition, the requirement that participants be capable of performing cycling exercise at 90% HR_max may have resulted in the inclusion of relatively fitter individuals with obesity. Therefore, the generalizability of the findings to individuals with lower fitness levels should also be interpreted with caution. Future studies are encouraged to recruit larger, more diverse samples and consider stratifying participants by fitness level to enhance ecological validity and external applicability.

High-intensity interval exercise (HIIE) may improve behavioral response speed in obese males by promoting non-inhibitory functional optimization of the prefrontal–striatal pathway, rather than engaging traditional inhibitory control systems. This effect reflects the characteristic mechanism of short-term exercise described by the neural efficiency hypothesis.