With the rapid acceleration of global population aging, maintaining the physical and mental health of older adults has become a crucial public health priority. Cognitive decline, a condition commonly observed in the aging population, is increasingly recognized not only as a natural consequence of aging but also as a potential early indicator of neurodegenerative diseases, such as dementia. It represents a significant threat to individual autonomy and social independence, necessitating urgent attention [1]. Cognitive decline spans multiple domains, including memory, concentration, judgment, and problem-solving abilities. Moreover, it is closely associated with a reduced ability to perform activities of daily living, underscoring the importance of effective prevention and intervention strategies [2].

In this context, regular physical activity or exercise has received attention as a nonpharmacological intervention with proven beneficial effects on cognitive function in older adults. Recent evidence suggests that individuals who fail to meet the recommended physical activity thresholds (150 min of moderate-intensity exercise or 75 min of vigorous-intensity exercise per week) are approximately 2.29 times more likely to experience cognitive impairment than those who meet these thresholds. Moreover, older adults who remain completely inactive during the week have an approximately 1.22-fold higher risk of cognitive impairment than those who engage in physical activity even once a week [3]. These cognitive benefits are thought to be mediated by exercise-induced increases in hippocampal volume, resulting in improved memory, executive function, and processing speed [4]. Moreover, physical activity increases the secretion of brain-derived neurotrophic factor, activates the cerebral cortex, and promotes neuroplasticity, thereby improving overall brain health [5]. Given its low risk of adverse effects and favorable cost-effectiveness compared to pharmacological therapies, regular physical activity has substantial potential for integration into clinical and community-based health programs.

Cognitive function can be evaluated using various neuroimaging modalities, such as electroencephalography (EEG), magnetic resonance imaging, computed tomography, and positron emission tomography (PET) [6]; however, these methods are often impractical because of their high costs and procedural complexity. Moreover, they offer limited utility in detecting subtle cognitive impairments in certain domains, such as memory, attention, language, and perception. To overcome these limitations, various neuropsychological tests have been developed and validated for use in older populations. Widely used tests include the Controlled Oral Word Association Test (COWAT) for assessing verbal fluency and executive function (with particular emphasis on frontal lobe function); the Trail Making Test (TMT) for assessing visuomotor speed, attention, working memory, and executive function; and the Digit Span Test (DST) for assessing attention and working memory [7]. In particular, patients with mild cognitive impairment (MCI) had an average score of approximately 7 points lower than the cognitively normal group on the COWAT phonemic fluency task, and the effect size was 0.61, indicating an intermediate level, suggesting that the test has the ability to detect early changes in cognitive function [8]. In addition, COWAT semantic fluency shows a decline from the early stages of MCI and tends to gradually worsen until the mild Alzheimer’s stage, so semantic fluency is considered an indicator that can respond sensitively to the early stages of cognitive decline [9].

Empirical findings have demonstrated the positive effects of exercise on cognitive performance. In one study, a 12-week walking exercise intervention significantly improved COWAT scores among 33 older adults with cognitive decline, indicating enhanced memory function [10]. Similarly, high-intensity aerobic exercise has been found to substantially improve TMT performance, indicating gains in cognitive-motor integration [11]. Moderate-to-high-intensity resistance training has also been found to enhance DST scores, supporting the role of physical exercise in promoting cognitive health among older adults [12].

As the prevalence of cognitive impairment increases with aging, proactive screening and evaluation are crucial. However, the high cost of medical testing poses a barrier for many older individuals. Among the available nonpharmacological strategies, regular physical activity is one of the most accessible and effective options in terms of both safety and affordability. Importantly, cognitive changes in older adults can be assessed using simple neuropsychological tools that can be feasibly implemented in community centers and dementia care facilities. Hence, the present study aimed to assess changes in cognitive function among older adults following a 12-week structured exercise program by using validated neuropsychological tests. The findings will serve as foundational evidence to facilitate the development and implementation of community-based health promotion programs for older women.

This study assessed changes in cognitive function among older adults aged ≥ 65 years following a 12-week structured exercise program by using standardized neuropsychological assessments. The findings revealed significant improvements in semantic fluency, as assessed by the COWAT, and in cognitive processing speed and attention, as assessed by the TMT, in the exercise group. However, no such improvements were noted in the control group.

The COWAT evaluates semantic and phonemic verbal fluency. In this study, the participants were instructed to generate as many words as possible in 1 min for the semantic categories “animals” and “items purchasable at a store” and for the phonemic categories using the Korean consonants “ㄱ,” “ㅇ,” and “ㅅ” [8]. Semantic fluency performance has been linked to frontal lobe atrophy and is considered a sensitive tool for distinguishing between different forms of cognitive impairment, including dementia and Alzheimer’s disease [16]. Moreover, diminished semantic fluency has been linked to the progression from mild cognitive impairment to dementia [17]. An important aspect of the COWAT is the generation of word clusters, which reflect underlying cognitive processes. Semantic clusters typically consist of consecutive words from the same subcategory (e.g., “fish,” “mackerel” or “tiger,” “leopard”), whereas phonemic clusters may contain rhyming words (e.g., gosari [bracken], goguma [sweet potato]), words beginning with the same phoneme (e.g., sagwa [apple], satang [candy]), or homophones (e.g., bam [night], bam [chestnut]) [18]. Troyer et al. [18] reported that clustering primarily reflects temporal lobe-mediated processes, such as lexical and semantic memory retrieval, whereas switching between clusters reflects frontal and frontocortical executive functions, such as cognitive flexibility and set shifting. Studies involving patients with neurological conditions, such as schizophrenia, Huntington’s disease, and Parkinson’s disease, have reported greater impairments in switching than in clustering; however, the opposite trend has been typically noted in Alzheimer’s disease. These distinctions suggest that performance patterns on the COWAT reflect the location and nature of brain dysfunction [19]. Notably, phonemic fluency begins to decline gradually in the 60s, with a marked deterioration observed by the 80s. As this age range coincides with the highest risk of cognitive disorders, such as mild cognitive impairment and Alzheimer’s disease, phonemic fluency is particularly crucial for the early detection of cognitive decline [20]. In this study, a significant improvement was specifically observed in semantic fluency in the exercise group. This finding may be attributed to enhanced functional connectivity between the default mode network (DMN), which is more active during passive or resting states, and the salience network (SN), which facilitates attention to relevant stimuli. Increased connectivity between the frontoparietal network (FPN) and the SN has been associated with memory enhancement [10]; this relationship is consistent with the observed improvements in both semantic fluency and TMT performance in this study.

The TMT is a widely used and validated neuropsychological tool for assessing brain injury and diagnosing neurodegenerative conditions, including Alzheimer’s disease [21]. Poor TMT performance, reflected by longer completion times, indicates impairments in cognitive domains, such as attention, working memory, visual perception, and cognitive flexibility. Although suboptimal TMT performance may also be attributed to physical limitations, such as reduced grip strength or fine motor coordination, the test cannot independently differentiate between cognitive and motor causes of poor performance [21]. Nevertheless, the significant improvements in TMT scores in the exercise group, along with the gains in semantic fluency, suggest that the intervention had a favorable impact on both cognitive and physical domains. Similarly, in a previous study, the TMT exhibited 100% sensitivity and 90% specificity in identifying abnormalities in COWAT performance [22]. There are two primary versions of the TMT: TMT-A, which requires connecting numbers in sequential order (e.g., 1-2-3…25), and TMT-B, which requires alternating between numbers and letters (e.g., 1-ㄱ-2-ㄴ-3…). TMT-A primarily assesses psychomotor speed, visuospatial search, and target-directed tracking [23]. Both versions have demonstrated meaningful correlations in patients with brain injury [24]. In contrast, in our study, no significant differences were observed in the DST scores between the exercise and control groups. The DST, a widely used clinical and research tool, assesses short-term verbal memory and attention [25]. DST performance relies more heavily on the integrity of the left cerebral hemisphere than on that of the right cerebral hemisphere or diffuse brain damage [26]. The lack of significant changes in DST performance in this study may be attributed to the exclusion of individuals with diagnosed dementia or cognitive impairment. Thus, the baseline cognitive health of the participants may have limited the potential for detectable improvements. Moreover, DST performance can vary based on different factors, such as anxiety, task familiarity, educational background, and literacy level. Practice effects may also contribute to performance variability, particularly in individuals with limited prior exposure to such assessments [25].

This study has several limitations. First, the sample exclusively consisted of cognitively healthy older women without diagnosed dementia or cognitive impairment. Future studies should assess the effects of structured exercise programs in populations with clinically diagnosed cognitive disorders. Second, although validated neuropsychological tests were employed, cognitive performance could be influenced by transient factors, such as mood, fatigue, and general physical condition on the day of testing. Future research should consider performing objective, instrument-based assessments of cognitive function to supplement neuropsychological outcomes. Lastly, as all participants were women, the findings may not be generalizable to older men. Future research should include male participants to assess potential sex-based differences in cognitive responses to exercise interventions.

In this study, a 12-week structured exercise program significantly improved semantic fluency, as assessed by the COWAT, and cognitive processing speed and attention, as assessed by the TMT, among older women. No significant changes were observed in semantic fluency in the control group or in DST performance in either group. These findings indicate that regular exercise may serve as an effective nonpharmacological strategy for preserving or enhancing cognitive function in older women, potentially preventing age-related cognitive decline and delaying the onset of neurodegenerative disorders.

In light of these findings, it is essential to develop and implement exercise programs based on comprehensive analyses of diverse factors influencing cognitive health in aging populations. Expanding the application and dissemination of such programs in community settings may be crucial for promoting successful aging and addressing the challenges of a rapidly aging society.