Alzheimer's disease (AD), the most prevalent neurodegenerative dementia, is characterized by progressive deterioration of memory, language, cognition, and daily functioning (Bäckman et al., 2004), and is frequently accompanied by neuropsychiatric symptoms, including apathy, depression, agitation, and psychosis (Creese and Lunnon, 2022). The disease is pathologically defined by extracellular β-amyloid (Aβ) plaque deposition and intracellular neurofibrillary tangles formed by hyperphosphorylated tau, together with neuronal loss, neuroinflammation, and oxidative damage (De Strooper and Karran, 2016; Tzioras et al., 2023). Neuronal loss is strongly associated with cognitive decline in both patients and animal models of AD, and accumulating evidence suggests that Aβ neurotoxicity is a major driver of neurodegeneration (Tzioras et al., 2023). These pathological changes are accompanied by glial activation, which accelerates neuropathological progression (Feng et al., 2023). Individuals aged 65 years and older with AD dementia have an average survival of four to eight years after diagnosis (Tom et al., 2015). The incidence and mortality of AD continue to rise, making it the fifth leading cause of death worldwide and imposing an increasing financial and societal burden (Alzheimer's Association, 2024; Ren et al., 2022). While effective pharmacological treatments for AD remain elusive (Alzheimer's Association, 2024), non-pharmacological interventions, particularly exercise-based therapies, hold promise in slowing disease progression.

Physical inactivity has been identified as an important risk factor for AD (De la Rosa et al., 2020). A substantial body of evidence demonstrates that regular physical exercise exerts multi-domain neuroprotective effects, including enhancement of synaptic plasticity (Casaletto et al., 2022), promoting hippocampal neurogenesis (Zhao et al., 2025), mitigating Aβ and tau pathology (Kim et al., 2023; Xu et al., 2022), attenuating neuroinflammatory responses (Wang et al., 2023a), and reducing blood-brain barrier damage(Liang et al., 2023). Clinical studies further indicate that exercise interventions improve cognitive function, daily activities, psychiatric symptoms, and physical capacity in patients with AD (López-Ortiz et al., 2021). These findings suggest that physical exercise serves as a viable intervention strategy for ameliorating AD progression. However, the therapeutic efficacy of exercise varies across intervention paradigms. Emerging evidence suggests that moderate-intensity aerobic exercise, resistance training, and combined exercise modalities can improve activities of daily living and physical function, with aerobic exercise appearing particularly beneficial for cognitive performance (López-Ortiz et al., 2021). In contrast, multimodal exercise programs incorporating aerobic, strength, balance, and flexibility components have been reported to enhance mobility and executive function in individuals with mild cognitive impairment, but not in patients with AD (de Oliveira Silva et al., 2019). These inconsistent findings highlight substantial heterogeneity in the effects of physical exercise on AD. Differences in exercise characteristics (including modality, intensity, duration, and frequency) may underlie this variability, indicating the need to identify exercise regimens that produce optimal therapeutic outcomes for AD.

Aerobic exercise, defined as sustained rhythmic activity relying on oxygen-dependent energy metabolism (Millstein, 2013). has shown benefits in reducing AD risk, improving cognition, alleviating depressive symptoms, and enhancing both basic and instrumental activities of daily living (López-Ortiz et al., 2021; Öhman et al., 2016). Running, a commonly studied aerobic activity, can be categorized into voluntary and forced paradigms. Voluntary exercise is self-initiated and intrinsically motivated, while forced exercise is an externally imposed, mandatory activity often linked to structured, dose-controlled training outcomes. This fundamental distinction may lead to differential effects on brain function and behavior. In adult male Sprague-Dawley rats, voluntary wheel running alleviated stress-induced behavioral depression (Moraska and Fleshner, 2001). Forced treadmill running appeared to trigger stress responses and increase anxiety-like behaviors, whereas voluntary wheel exercise may have an anti-anxiety effect (Leasure and Jones, 2008). Interestingly, voluntary, involuntary, and forced exercises showed equally effective in mitigating both cognitive and non-cognitive impairments in a rat model of AD (Belviranlı and Okudan, 2019), where involuntary exercise emphasizes passive and non-volitional attribute. When “involuntary” exercise refers to externally imposed, protocol-controlled physical activity, it is interchangeable with forced exercise in this context. Most published studies classify treadmill-driven exercise intervention under the category of forced exercise intervention (Sun et al., 2025; Svensson et al., 2016; Yuede et al., 2009). Although both voluntary wheel running and forced treadmill exercise increased hippocampal volume, voluntary exercise more effectively reduced amyloid plaque burden and memory deficits in Tg2576 AD mice (Yuede et al., 2009). However, head-to-head comparisons of voluntary, forced, and combined exercise interventions in AD (particularly with respect to behavioral outcomes, adult neurogenesis, serum metabolomic profiles, and brain activity) remain limited. Elucidating the differential mechanisms underlying these exercise modalities is essential for optimizing exercise-based interventions and improving their translational relevance.

In this study, we investigated the differential effects of voluntary wheel running (recreational-type), forced treadmill running (moderate-intensity) and combined intervention (voluntary and forced running) in treatment of Aβ oligomer-induced AD model mice, and explored the mechanisms by which distinct exercise modalities ameliorate Aβ-mediated neuropathology. In addition, to enhance the translational relevance of our preclinical findings, we conducted an acute, exploratory human study using portable functional near-infrared spectroscopy (fNIRS) to characterize modality-specific cortical activation patterns during voluntary and forced exercise in patients with AD. This complementary human experiment was designed to probe immediate brain network responses to different exercise paradigms, thereby providing a mechanistic link between the cumulative effects of long-term exercise intervention in animal models and the immediate effects of acute exercise intervention in human trials.