INTRODUCTION
Age-related cognitive decline has become a major public health issue owing to global population aging, particularly in super-aged societies, such as South Korea, where over 20% of the population is aged 65 years or older [1]. This demographic trend is expected to intensify in the coming decades. Cognitive impairment in aging populations reduces an individual’s quality of life and imposes significant social and economic burdens [2].
Mild cognitive impairment (MCI), an intermediate stage between normal aging and dementia, is characterized by measurable cognitive deficits with maintained functional independence [3]. Early intervention is considered crucial for delaying or preventing progression to dementia and reducing long-term care needs.
However, pharmacological treatments for MCI have demonstrated limited efficacy and are frequently associated with adverse effects [4-7]. Although approved for Alzheimer’s disease, common drugs, such as cholinesterase inhibitors and N-methyl-D-aspartate receptor antagonists, have yielded inconsistent results among those with MCI [4,6]. Furthermore, current clinical guidelines advise against their routine use [8]. Recently, antiamyloid therapies, such as Donanemab and Lecanemab, have been approved; however, high cost, complex administration protocols, and risk of serious side effects, including cerebral hemorrhage and brain edema [9,10], limit their utility in real-world settings, which reinforces the need for non-pharmacological interventions.
Thus, non-pharmacological interventions focusing on modifiable lifestyle behaviors have gained attention as low-risk, nonpharmacological alternatives. Among these, nature-based interventions (NBIs) have emerged as possible approaches for supporting cognitive health among older adults with MCI.
Shanahan et al. [11] defined NBIs as structured activities that aim to engage individuals with natural environments to promote physical, mental health, and overall well-being. Jiang et al. [12] further characterized NBIs as integrative programs that incorporated sensory stimulation, physical activity, and emotional engagement to enhance overall health. These interventions aim to harness the multifaceted components of natural environments in combination with bodily movement, social interaction, and psychological stimulation to foster cognitive, emotional, and physical wellness.
Recent studies reported that NBIs may enhance cognitive and emotional outcomes in older adults [13-15]. Natural settings may stimulate the prefrontal cortex and attentional networks, which may promote neuroplasticity and support cognitive restoration [16-18]. These effects may be particularly relevant for individuals with MCI, who often exhibit early impairments in attention and executive function. NBIs also integrate physical activity and social interaction, both of which contribute to cognitive resilience, and offer a potentially further effective approach than single-component interventions [19,20].
Despite increasing interest, few experimental studies have examined the effects of NBIs specifically in older adults with MCI. Jun et al. [21] found no significant changes in cognitive function following a forest experience program, while Baek et al. [15] reported positive outcomes in a single-group design that lacked a control group.
Considering the research gaps and mixed findings, this study aimed to empirically investigate the effects of a NBI on cognitive function in older adults with MCI via a control group design matched for age and sex. Furthermore, this study sought to contribute to evidence-based preventive strategies for promoting cognitive health in aging populations with MCI.
METHODS
Study design
This study adopted an age- and sex-matched experimental-control group design. Individuals aged 60 years or older with a clinical diagnosis of MCI were recruited. Assessments of cognitive and emotional function, blood pressure, and physical performance were conducted pre- and post-intervention. The NBI program was implemented from April to October 2022. The experimental group engaged in weekly sessions for four weeks, with 3-5 individuals per group. The control group maintained their usual daily routines (Figure 1). This study was approved by the Institutional Review Board of Chuncheon Sacred Heart Hospital, Republic of Korea (Approval No. CHUNCHEON 2021-08-011).
Participants
Participants were recruited from outpatient clinics at the Department of Psychiatry, Chuncheon Sacred Heart Hospital, and via outreach to a local dementia care center. Eligible individuals were older adults clinically diagnosed with MCI by a board-certified psychiatrist. Those who expressed interest underwent preliminary screening to assess dementia-related indicators and confirm their eligibility via Clinical Dementia Rating (CDR), Global Deterioration Scale (GDS), Short Blessed Test (SBT), and Instrumental Activities of Daily Living (IADL).
Inclusion criteria were those who: 1) met the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5) criteria for mild neurocognitive disorder [22], 2) had a clinical diagnosis of MCI based on Petersen’s criteria, subjective memory complaint, intact daily function, preserved general cognition, impaired memory for age, and absence of dementia [23], and 3) had a score of 0.5-1 or 2-4 points on the CDR [24] and GDS [25], respectively.
Exclusion criteria included those with 1) neurological disorders (e.g., epilepsy, Parkinson’s disease); 2) psychiatric disorders based on the DSM-5, which included bipolar disorder or substance/alcohol dependence; 3) severe medical conditions (e.g., cardiovascular or endocrine disease); 4) olfactory dysfunction; or 5) initiation, discontinuation, or dose adjustment of medications known to affect cognitive function during 3 months prior to consent or throughout the intervention period, including cognitive enhancers (e.g., donepezil, rivastigmine, galantamine, memantine) and major Central Nervous System-acting drugs (e.g., strong anticholinergics, antipsychotics, benzodiazepines/hypnotics, anticonvulsants). Written informed consent was obtained after participants were informed of the study’s purpose, procedures, and ethical considerations.
Participants were assigned to the experimental group, and the control group was matched for age and sex. This study included 50 participants, with 25 in each group.
Measurements
Cognitive performance was set as the primary focus to evaluate the effects of the NBI on older adults with MCI. Furthermore, emotional status, physical function, and physiological indicators were also assessed to examine additional outcomes.
Cognitive function
To assess cognitive function, this study used selected subtests from the Korean version of the Consortium to Establish a Registry for Alzheimer’s Disease Neuropsychological Battery, originally developed by a U.S. NIH-supported consortium [26] and later standardized in Korea by Lee et al. [27] It evaluated language, memory, visuospatial, and executive functions through subtests, such as the Mini-Mental State Examination (MMSE), Stroop Test, and semantic verbal fluency. Internal consistency was reported as Cronbach’s α=0.92 [27].
The MMSE-KC, Stroop Test, and semantic verbal fluency were selected as key indicators of general cognitive function. An arithmetic task (addition and subtraction) was included to assess numerical processing.
The MMSE, a widely used screening tool for neurocognitive disorders, measures various domains, such as orientation, memory, attention, and language. The total score ranges from 0 to 30 points. The MMSE is a reliable index of global cognition that is frequently used to monitor cognitive changes over time [28].
The Stroop Test evaluates attention, executive function, and inhibitory control via three subtasks: Word, Color, and Interference. Each is scored based on the number of correct responses within 1 minute [29,30].
Semantic verbal fluency was measured via an animal-naming task, which evaluated lexical retrieval and semantic memory access under time pressure [31]. Participants named as many animals as possible in 1 minute; invalid or repeated responses were excluded.
Arithmetic ability was assessed via two tasks: Addition & Subtraction_retrieval (ASr), which required direct recall of simple sums, and Addition & Subtraction_procedural (ASp), which involved multi-step calculations. These tasks served as task-based measures that reflected the cognitive demands of working memory and executive functions in arithmetic problem solving [32].
Emotional function
Emotional status was assessed via the Korean version of the Geriatric Depression Scale Short Form (SGDSSF-K), developed by Yesavage et al. [33] and translated by Kee [34]. This 15-item binary response scale yielded higher scores for more severe depressive symptoms. It demonstrated strong reliability in older adult populations (Cronbach’s α=0.88) [34].
Physical function
Physical performance was measured via the Short Physical Performance Battery (SPPB), which included tests of balance, gait speed, and chair stands. Scores range from 0 to 12 points, and lower scores indicate poorer physical function [35]. The SPPB demonstrated excellent internal consistency (α=0.86) and test-retest reliability (intraclass correlation [ICC]=0.92) [35]. Additionally, the Timed Up and Go (TUG) test was employed to assess mobility and balance. It measured the time taken to stand up, walk 3 m, return, and sit down [36]. It demonstrated good test-retest reliability (ICC=0.98) and internal consistency (α=0.81) [37].
Blood pressure
Blood pressure was measured as a physiological indicator. Systolic pressure below 120 mm Hg and diastolic pressure below 80 mm Hg were considered normal. In older adults, values up to 140/90 mm Hg may be acceptable due to age-related changes. Elevated blood pressure was associated with an increased risk of dementia [38].
Intervention
Intervention design
The NBI program comprised four weekly sessions (100- 110 minutes each), conducted in small groups of 3-5 participants. All sessions were led by certified forest therapy instructors who followed a standardized protocol. Group-based format aimed to promote social engagement and active participation.
Study site
Sessions were conducted at the National Center for Forest Education Chuncheon in Gangwon-do, South Korea. The site encompasses approximately 335 hectares of mixed coniferous and deciduous forest, where pine and oak species dominate, accompanied by diverse native shrubs and trees. The forest ecosystem also supports rich understory vegetation, herbaceous species, streams, birds, and insects, offering a high level of ecological diversity. Seasonal landscape changes, plant scents, and the natural sounds of flowing water and birds provide multisensory stimuli characteristic of a nature-based restorative environment suitable for cognitive recovery.
The main activity trail is located at an elevation of 188-227 m, with an average gradient of 2.3°-2.9°, forming a gently sloping 3-km round-trip path. Some sections are equipped with barrier-free deck trails, ensuring accessibility and walking safety for older adults. The facility also includes a visitor center, auditorium, cafeteria, and lawn plaza, allowing flexible indoor-outdoor program operation depending on weather conditions.
The site provided ecologically diverse environments and safe, barrier-free facilities. Rich sensory stimuli, natural scenery, seasonal aromas, and ambient sounds enhanced the setting’s suitability for cognitive recovery in older adults. The site was accessible through a memorandum of understanding with the National Center for Forest Education Chuncheon.
NBI program
The intervention was grounded in the theory that sensory stimulation and emotional relaxation in nature supported cognitive restoration [39]. Natural settings, such as forests, reduced emotional stress and enhanced attention and working memory [40]. To reflect these restorative effects, the program incorporated forest walking, nature-based exercises, sensory play, reminiscence activities, and structured observational tasks (Table 1).
Statistical analysis
All statistical analyses were conducted using SPSS WIN version 26.0 (IBM Corp.). Group homogeneity was tested via independent t-tests and chi-square tests. Fisher’s exact test was applied when the expected frequency was <5. Normality was assessed using the Shapiro-Wilk test. A Mann-Whitney U test was applied when normality was not satisfied. To evaluate intervention effects, a repeated-measures analysis of variance (ANOVA) was performed with group and time as the betweensubjects and within-subjects factors, respectively. Within-group pre-post comparisons were analyzed using paired samples ttests. Statistical significance was set at p<0.05.
RESULTS
Baseline demographic and clinical characteristics and homogeneity between groups
A total of 50 participants were assigned equally to the experimental and control groups, with matching by age and sex. Three participants in the experimental group withdrew due to incomplete intervention attendance, and three in the control group did not complete the post-assessment, yielding a final sample size of 44 (Table 2).
Cognitive effects
Repeated measures ANOVA on the MMSE scores revealed a statistically significant interaction between group and time (F=10.226, p=0.003), which indicated differing patterns of score changes between the experimental and control groups. Additional paired t-tests showed a significant increase in MMSE scores in the experimental group (t=-2.270, p=0.034), whereas the control group exhibited a significant decline (t=2.262, p=0.034). These results are visually summarized in Figure 2, which illustrates the divergent post-intervention trends between the two groups.
Although no significant interaction between time and group was observed in the ASr, ASp, and Stroop Color (SC) tasks, a significant main effect of time was observed. Additional within-group comparisons revealed significant pre-post improvements only in the experimental group post-intervention. For Stroop Word, the main effect of time was significant; however, the paired t-test revealed no significant difference between pre- and post-intervention scores. No significant effects were observed for Stroop Interference and Verbal Fluency_Semantic (Table 3).
Emotional effects
Analysis of the SGDS scores revealed a marginally significant interaction effect between group and time (F=3.885, p=0.055), although it did not reach statistical significance. Neither the main effects of group (F=1.222, p=0.275) nor time (F=3.803, p=0.058) were statistically significant. However, additional paired t-tests indicated a significant reduction in the SGDS scores within the experimental group (t=2.444, p=0.023), while no significant change was observed in the control group (t=-0.018, p=0.986) (Table 4).
Physical effects
Both the SPPB and TUG test revealed no significant main effects of group or time, nor any significant interaction effects between group and time (Table 5).
Blood pressure
No statistically or clinically significant changes in blood pressure were observed between the groups. Furthermore, no significant group, time, or interaction effects were observed (Table 6).
DISCUSSION
This study aimed to empirically examine the effects of an NBI on cognitive function in older adults with MCI and compared an intervention group with a control group. Results revealed that the MMSE, used as a representative indicator of cognitive function, significantly improved in the intervention group post-intervention compared with the control group, which demonstrated a statistically significant interaction effect between time and group. In contrast, emotional, physiological, and physical function indicators had no significant interaction effects.
Improvement in the MMSE scores was statistically significant and clinically meaningful. The MMSE is a widely used screening tool to diagnose neurocognitive disorders and evaluate intervention effects. It provides a structured quantification of overall cognitive status, and score changes may reflect actual functional improvements. Previous research revealed that improvements in MMSE scores were significantly associated with enhanced functional independence and better performance in activities of daily living [41-43].
Since its high test-retest reliability and inter-rater agreement make it suitable for tracking changes through repeated assessments, even short-term improvements in scores may serve as objective indicators of positive cognitive effects resulting from therapeutic or intervention programs [44,45].
Cummings [46] synthesized evidence from earlier studies and reported that a change of 1.2-2.6 points in the MMSE score may represent the Minimally Clinically Important Difference (MCID) for patients with MCI or mild Alzheimer’s disease. In this study, the MMSE score in the experimental group increased by approximately 1.2 points, close to the lower bound of the reported MCID and beyond the range of natural variability (±0.5 points), which suggested that the change may be interpreted as meaningful [45,47]. Therefore, significant improvement in the MMSE scores observed in the experimental group suggested that the NBI had a positive impact on cognitive function in older adults with MCI. Since MCI represents a critical window in which early intervention can delay progression to dementia, the clinical significance of interventions that promote cognitive improvement is particularly noteworthy [28].
Improvement in cognitive function may result from the combined effects of multiple factors inherent in natural environments [48]. Environmental stimuli are important contributors to improvements in attention and executive function, and natural settings provide sensory inputs that enhance both selective attention and executive functioning [40,49]. Sounds of nature and landscapes promote emotional stability and sustained attention [50-52]. According to Kaplan and Berman [53], natural environments facilitate the activation of involuntary attention, which enables recovery from attentional fatigue and improves direct attention and executive control. Considering these previous findings, NBIs may potentially influence cognitive function by engaging attention and executive processes. In this study, although no significant interaction effects were observed between the groups, within-group improvements were observed in the ASr, ASp, and SC tasks, which were closely related to attention and executive function. However, further research should clarify how NBIs influence attention and executive function and lead to cognitive improvements.
Another potential environmental factor is the influence of volatile organic compounds present in forest air, particularly phytoncides, known as therapeutic agents in forest environments [54]. According to recent research, inhalation of phytoncides in patients with MCI improved performance on the Stroop task. Furthermore, functional near-infrared spectroscopy revealed more efficient cognitive task performance in the phytoncide group, as evidenced by reduced oxygen consumption in the left ventrolateral prefrontal cortex [55]. Since participants in this study were repeatedly exposed to forest environments, similar neurophysiological mechanisms could have been at play. Although no significant between-group differences were observed, significant within-group improvements in Stroop performance in the intervention group suggested that phytoncide exposure may have contributed to enhanced prefrontal cortical activation.
Taken together, these findings suggest that restorative elements of natural environments—such as attention restoration and exposure to forest-derived chemical compounds—may contribute to improved cognitive function. However, these mechanistic explanations remain hypothetical and require targeted experimental investigation. Future research employing powered randomized controlled trials will be critical for confirming these preliminary findings and refining estimates of the observed effects.
Natural environments are considered complex stimulus settings that may indirectly contribute to changes in cognitive function by providing emotional stimuli that promote psychological stability as well as physical contexts that encourage bodily activity [14,56,57]. However, in this study, no significant between-group differences were observed in emotional, physical, or physiological indicators. Previous studies reported that activities in forest environments provided opportunities for sensory stimulation, physical movement, and social interaction, which positively influenced the reduction of depressive symptoms [14,57]. Although no significant differences were observed between groups, significant improvements in depressive symptoms were observed within the group that participated in NBI. This may be attributed to the higher baseline level of depression in the experimental group compared with the control group, which suggested that the intervention was more effective in reducing depressive symptoms among those with greater initial vulnerability. Therefore, future research should apply such interventions to populations with higher emotional vulnerability, such as those with clinical depression, for further refined analysis.
Regarding physical function, no significant changes were observed in either the SPPB or TUG test. The relatively short duration of the four-week intervention and its low physical intensity may not have provided sufficient physiological stimulation to induce measurable improvements. Nishiguchi et al. [58] reported significant improvements in SPPB scores after an eight-week forest-based aerobic exercise program, which suggested that longer and further intensive interventions may be necessary to produce observable effects on physical performance. Furthermore, this study employed gross functional assessments focused on gait and mobility. In contrast, Baek et al. [15] employed various physical assessments, which included body composition analysis via bioelectrical impedance, grip strength measurement via standardized resistance tests, balance assessment through postural sway analysis, and detailed gait evaluation that focused on stride length and walking speed. Their study demonstrated significant improvements in muscle mass, balance ability, and gait stability across most measured domains. These findings suggest that the absence of significant effects in this study may partly reflect limitations in both the sensitivity of the physical assessment tools and intervention dosage. Future studies should consider incorporating further sensitive and diverse measures of physical function and designing interventions with greater intensity and duration to better evaluate the full impact of nature-based programs on physical health.
Conclusively, this study is meaningful because it experimentally verified the effects of a NBI on cognitive improvement in older adults with MCI. Although MCI is characterized by noticeable cognitive decline, therapeutic options to prevent further deterioration remain limited. This study applied a non-pharmacological, non-invasive, and easily accessible intervention grounded in daily life and presented a practical method to enhance or maintain cognitive function in this population. Such an approach offers advantages regarding spatial accessibility and feasibility, which makes it a valuable option for real-life implementation. Therefore, continued research on systematic and long-term NBIs for older adults with MCI can substantially contribute to dementia prevention efforts and the development of community-based health promotion programs.
A limitation of this study is the intervention’s relatively short duration (4 weeks). Although nature-based activities demonstrated potential for positive changes in cognitive function, no significant between-group differences were observed in emotional, physiological, or physical function indicators. These health outcomes typically require more time to manifest measurable effects. Particularly, repeated stimulation, sufficient intensity of physical activity, and prolonged exposure are often necessary to induce meaningful change in physiological and physical function. This study employed an age- and sexmatched case-control design, with 25 participants per group. Within this design, the relatively small sample size reflects the exploratory nature of early-phase research. Prior studies have similarly used small samples: for example, Baek et al. [15] investigated 22 older adults with MCI without a control group, and Park et al. [59] examined 33 older adults with cognitive concerns using a controlled design. These precedents demonstrate that comparable or smaller sample sizes are common at this exploratory stage. However, small samples limit statistical precision and generalizability; therefore, the results should be interpreted cautiously. Given that the interpretation of mixed outcomes depends in part on how clearly environmental exposure and intervention intensity are defined, future research should standardize and quantify the reporting of study-site characteristics and intervention dose and intensity in alignment with experimental design principles.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Download Citation
Print
