Advances in Virtual Reality-Based Physical Rehabilitation for Neurodegenerative Diseases: A Systematic Review

Abstract

Background: Neurodegenerative diseases cause both progressive motor and cognitive impairments for which no curative treatments exist. Virtual reality (VR)-based rehabilitation has emerged as a promising strategy to enhance physical rehabilitation by offering immersive, engaging, and personalized environments. Methods: A systematic review was conducted following the PRISMA guidelines, examining studies published between 2020 and 2025 in the Web of Science and Scopus. Twelve studies met the inclusion criteria, focusing on VR-based physical rehabilitation in individuals with neurodegenerative diseases. Results: Most studies reported significant improvements in balance, gait, postural control, and motor function. Some studies also found benefits in processing speed, executive function, emotional well-being, and activities of daily living. Across different levels of immersion, VR interventions showed high usability and were well tolerated, even among older adults. However, methodological limitations, such as small sample sizes, a lack of control groups, and short intervention durations, were prevalent. Conclusions: VR has demonstrated to be an effective and well-tolerated tool for the rehabilitation of individuals with neurodegenerative diseases such as Parkinson’s disease and multiple sclerosis. However, the confirmation of its clinical efficacy and long-term impact necessitates the execution of randomized controlled trials with larger samples and extended follow-up periods.

1. Introduction

Neurodegenerative diseases are chronic, progressive conditions that primarily affect the central nervous system, leading to functional decline across the motor, cognitive, and emotional domains. Among the most prevalent of these disorders are Parkinson’s disease, Alzheimer’s disease and multiple sclerosis [1]. Despite advances in pharmacological management, no curative treatments currently exist. Consequently, rehabilitation strategies have become essential for mitigating symptom progression, maintaining functional independence, and improving quality of life. However, adherence to conventional rehabilitation programs is often low, and outcomes remain variable, highlighting the need for more engaging, adaptable, and effective therapeutic approaches [2].

In rehabilitation, traditional approaches typically include cognitive training, aimed at improving specific cognitive functions through repeated practice, behavioral therapy, which focuses on modifying maladaptive thoughts and behaviors, and physiotherapy, which targets motor recovery through physical exercise [3,4]. More recently, computer-based interventions have emerged as an accessible alternative, delivering cognitive and motor training via digital platforms such as desktop applications or web-based software. These programs allow for standardized content delivery, adjustable difficulty levels, and remote access, making them increasingly scalable and cost-effective [5].

Virtual Reality (VR) represents a next-generation advancement in computer-based rehabilitation, offering immersive, interactive environments that closely replicate real-world scenarios. Over the past decade, its use in clinical research and practice has expanded substantially, with studies supporting its effectiveness in the assessment, diagnosis, and treatment of neurological and neuropsychological disorders [6,7]. Unlike conventional tools, which often lack ecological validity, VR enables the simulation of dynamic, repeatable tasks in safe and controlled settings, allowing clinicians to capture complex functional behaviors with greater fidelity [8]. A key advantage of VR lies in its high level of sensory engagement, which enhances user motivation and adherence. Additionally, VR platforms facilitate real-time performance tracking, adaptive task difficulty adjustment, and customized feedback, thus enhancing the therapeutic process and facilitating individualized treatment planning [9,10]. These characteristics position VR as a versatile and potent instrument within neurorehabilitation frameworks.

The classification of VR technologies is typically based on the degree of immersion, user isolation from the physical surrounding, they provide [11]. Non-immersive systems utilize conventional screens and offer limited interaction. Semi-immersive systems, such as panoramic projections or CAVE environments, provide intermediate levels of immersion [12]. Fully immersive systems, which employ head-mounted displays and motion tracking, offer the highest levels of sensory engagement [10]. Each format offers distinct advantages depending on patient needs, therapeutic goals, and technological accessibility [13]. The choice of immersion level critically influences therapeutic outcomes and patient tolerability. Non-immersive systems offer high accessibility and minimal side effects, making them particularly suitable for older adults and early-stage assessments [3,14]. Semi-immersive CAVE systems provide increased engagement while minimizing adverse effects commonly associated with head-mounted displays [15]. Fully immersive systems deliver the highest sense of presence but require careful screening due to risks such as motion sickness and contraindications in sensitive populations [15,16]. Therapeutic efficacy varies across immersion levels and should be matched to individual rehabilitation objectives and neurological conditions [14].

In the domain of neurorehabilitation, the utilization of VR has been explored across a range of modalities. These include environmental reconstructions, which have been employed for the purpose of retraining navigation or instrumental activities of daily living [17,18]. Situational simulations have been employed to facilitate behavioral adaptation [19], and cognitive–motor dual-task training has been implemented as an effective strategy to improve the concurrent processing of motor and cognitive activities [20]. It has been demonstrated that VR can enhance various aspects of physical and mental well-being, including attention, gait, balance, executive functioning, and emotional health, particularly in patients with limited mobility or poor adherence to conventional therapies [21,22]. Specific platforms such as the CAREN system (Computer Assisted Rehabilitation Environment), integrate immersive VR with motion capture to deliver personalized feedback and optimize motor learning [23]. Similarly, exercises based on videogames, commonly known as exergames, have achieved notable popularity due to their motivational value, accessibility, and their capacity to provide repetitive, goal-oriented physical training [24].

The global burden of neurodegenerative diseases continues to rise, particularly in aging populations, highlighting an urgent need for more effective and accessible rehabilitation strategies [25]. In this context, VR has emerged as an innovative and increasingly viable solution for addressing the limitations of conventional rehabilitation approaches. While preliminary findings are promising, the clinical effectiveness, methodological rigor, and scalability of VR-based interventions remain areas of active investigation. Although VR-based neurorehabilitation is gaining attention, rigorous existing reviews have important gaps that limit their clinical use. Most prior reviews have examined VR across heterogeneous neurological populations, combining acute conditions such as stroke with progressive neurodegenerative diseases that require fundamentally different therapeutic approaches [3,7,10,13,26]. Additionally, earlier reviews included older studies, combining outdated VR technologies with contemporary systems [5,6,27], making it difficult to assess the true potential of current VR interventions. Furthermore, previous systematic evaluations have focused on motor outcomes [13,14,27] while overlooking critical factors such as user experience, technology acceptance, and the practical implementation considerations [1,4,11] that are essential for clinical adoption.

The present systematic review aims to provide the first comprehensive evaluation focused exclusively on VR-based physical rehabilitation in neurodegenerative diseases using contemporary evidence. Specifically, it synthesizes recent studies (2020–2025) to capture the latest technological advances, examines both clinical effectiveness and user-centered outcomes, and evaluates intervention feasibility across different levels of VR immersion. By focusing on neurodegenerative conditions specifically, this review aims to provide clinically relevant evidence to inform rehabilitation protocols and support the evidence-based implementation of VR in neurorehabilitation.

2. Materials and Methods

2.1. Search Strategy

This systematic review was conducted in accordance with the PRISMA guidelines (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) [28]. The literature search aimed to identify peer-reviewed articles that investigated the use of VR tools in the physical rehabilitation of individuals with neurodegenerative diseases. Searches were performed in Web of Science (WoS) and Scopus databases. The selection of these databases was made on the basis of their comprehensive multidisciplinary coverage of the clinical, psychological, and engineering domains. Preliminary searches in PubMed and IEEE Xplore suggested substantial overlap with the records already indexed in WoS and Scopus. The search was limited to the previous five years (2020–2025) to ensure the inclusion of the most recent evidence, as rapid technological advances in VR render older interventions less comparable to current approaches. The final search was completed on 30 March 2025 and included studies published between January 2020 and March 2025. The search strategy employed a combination of keywords, which were combined using the Boolean operator “AND”: “virtual reality” AND “rehabilitation” AND “neurodegenerative disease*”. No quotation marks were used to broaden the retrieval of relevant records. Filters were applied to restrict the search to the last five years. In WoS, the terms were searched within the “Topic” field (title, abstract, and keywords), excluding review articles and conference proceedings. Similarly, in Scopus, the search was conducted within “Article Title, Abstract, and Keywords,” with review articles and proceedings removed.

2.2. Eligibility Criteria and Study Selection

The eligibility criteria employed in this systematic review included the following: (1) original research articles, (2) studies written in the English language, (3) studies assessing the effects of a VR-based physical rehabilitation intervention, (4) studies focused on improving physical function in individuals diagnosed with a neurodegenerative disease and (5) studies with full text available. The following criteria were used to determine exclusion from the study: (1) case reports or conference proceedings, (2) animal studies or purely technical simulations without human cognitive data, (3) studies addressing cognitive but not physical rehabilitation outcomes, and (4) articles not employing VR technologies as part of the intervention. The selection of the studies was conducted independently by two reviewers, with any discrepancies addressed through consultation with a third reviewer.

In total, 75 records were identified from searches in the databases. During the preliminary screening phase, 21 duplicate records were eliminated. The titles, abstracts, and results of the remaining articles were then reviewed, leading to the exclusion of 42 records that did not meet the inclusion criteria. A total of 12 studies were incorporated into the final review. The PRISMA flowchart of the search process is presented in Figure 1.

This systematic review was registered after the data extraction phase had begun. The materials are available in the OSF repository (Registration: https://doi.org/10.17605/OSF.IO/ZTR3U, accessed on 8 September 2025).

3. Results

This systematic review synthesized findings from 12 studies conducted between 2020 and 2025. Table 1 summarizes the type of VR employed, sample and intervention characteristics, the assessment tests employed, and the primary outcome measures for each revised article. Additionally, Supplementary Table S1 deeply collects information about other external devices employed in the studies, as well as detailed information about VR tasks characteristics.

3.1. Study Characteristics

The type of VR technology employed exhibited significant variation across the studies. Five of the interventions employed non-immersive VR [29,30,31,32,33], whereas four interventions utilized semi-immersive VR technologies [32,34,35,36], and four employed immersive VR [37,38,39,40]. The study by Manuli et al. [32] employed a combination of semi-immersive and non-immersive VR tools. In addition, external tools and devices which were employed in the studies included motion controllers and sensors [29,30,35,37,38,39,40], balance platforms [29,30,31,33], infrared cameras [33], laptops [38], and robotic devices [32], as well as body weight support platforms and safety harnesses [32,33,36], among others.

Sample sizes varied widely, ranging from pilot studies with 6 to 20 participants [33,36,37,38,39] to larger studies with more than 20 participants [29,30,31,32,34,35,40].

Table 1. Overview of the reviewed studies employing VR technology for physical rehabilitation in individuals with neurodegenerative diseases.

Author(s), Year	VR *	Sample	Intervention	Assessment	Main Findings
Formica et al., 2023 [34]	Semi-immersive VR (CAREN * system)	Parkinson N = 31 (18 M *, 13 F *) No control group Age range = 18–73	Duration: 2 months (8 weeks) Frequency: 3/week Session length: 50 min	Gait and balance: 10MWT *, BBS * General cognitive functioning: MoCA * Executive functioning: FAB * Coping strategies: COPE * Fear of falling: FES-I * Depression: HRS-D *	Cognitive and emotional benefits (executive function, anxiety, depression); greater physical and cognitive effort in response to stress.
Honzíková et al., 2025 [37]	Immersive VR (Meta Quest 2)	Parkinson N = 19 (sex not specified) No control group Mean age = 64.2 ± 12.8	Duration: 1 month (4 weeks) Frequency: 2/week Session length: 20 min	Gait and balance: 10MWT, BBS Functional mobility: TUG * + dual task QoL *: PDQ-39 *	Improvements in stability, mobility, and quality of life.
Imbimbo et al., 2021 [35]	Semi-immersive VR (Nirvana system)	Parkinson N = 26 (22 M, 4 F) No control group Age range = 66.25–75.75	Duration: 6 weeks Frequency: 2/week Session length: 45 min	Gait and balance: 6MWT *, BBS Cognitive reserve: CRI-q *	Improved balance and gait in patients with higher cognitive reserve.
Kashif et al., 2022 [29]	Non-immersive VR (wall-mounted display, Wii controllers and Wii Fit board)	Parkinson N = 44 (25 M, 19 F) Experimental group: n = 22 Control group: n = 22 Age range: 50–80	Duration: 12 weeks Frequency: 3/week Session length: 60 min (40 min session + 20 min walking and cycling)	Motor function and balance: UPDRS *-III, BBS, ABC * ADLs *: UPDRS-II	Improved motor function, balance, and ADLs.
Kashif et al., 2024 [30]	Non-immersive VR (wall-mounted display, Wii controllers and Wii Fit board)	Parkinson N = 60 (33 M, 27 F) VR group: n = 20 MI * group: n = 20 PT * group: n = 20	Duration: 12 weeks Frequency: 3/week Session length: 60 min (20 min VR/MI + 40 min PT)	Motor function and balance: UPDRS-III, BBS, ABC ADLs: UPDRS-II	VR showed better outcomes in balance and motor function vs. MI and PT. Best results were obtained for the VR + PT combination.
Malisky et al., 2024 [31]	Non-immersive VR (wall-mounted display, Wii controllers and Wii Fit board)	Spinocerebellar ataxia N = 28 (20 M, 8 F) No control group Age range: 15–70	Duration: 10 weeks Frequency: 2/week Session length: 50 min	Balance: ABC ADLs: VADL *	Improved balance and gait, and reduced fall frequency.
Manuli et al., 2020 [32]	Semi-immersive VR (Nirvana system, CAREN system) Non-immersive VR (VRRS *)	Multiple sclerosis N = 84 (47 M, 37 F) No control group Age range: 18–75	Duration: 8 weeks Frequency: 3–5/week Session length: 60 min	System usability: SUS * Goal achievement: GAS * Well-being perception: MSQOL *	Improved perceived QoL (physical and mental) and achievement of therapeutic goals. High satisfaction and usability.
Mazzari et al., 2025 [33]	Non-immersive VR (Technobody devices)	Parkinson N = 18 (sex not specified) Patients: n = 9 Controls: n = 9 Age range: 55–85	Duration: 8 weeks Frequency: 2/week Session length: 60 min	Motor function and balance: TUG, BBS	Improvements in balance, gait, trunk flexion, pain threshold, and erector spinae displacement.
Pullia et al., 2023 [36]	Semi-immersive VR (C-Mill system)	Parkinson N = 20 (13 M, 7 F) Experimental group: n = 10 Control group: n = 10 Age range: 50–70	Duration: 5 weeks Frequency: 4/week Session length: 45 min	Motor function, gait, and balance: 10MWT, 6MWT, TUG, UPDRS-III, TS *, BBS Fear of falling: FES-I ADLs: FIM *	Both conventional and VR training improved motor function; treadmill + VR improved endurance and postural control.
Rodríguez-Fuentes et al., 2024 [38]	Immersive VR (Meta Quest 3)	Multiple sclerosis N = 18 (5 M, 13 F) Control group: n = 10 Experimental group: n = 8 Age range: 18–65	Duration: 5 weeks Frequency: 3/week Session length: 30 min	Motor function, gait, and balance: TS, TUG Functional mobility and strength: TUG + dual task, FTSST *, JHD * Fatigue: FSS * Reaction time: Rezzil software (1.9.0 version) VR-related assessments: SSQ *, SUS, GEQ, * ad hoc satisfaction questionnaire Perceived effort: Borg scale	High usability and satisfaction. Improved lower-limb endurance, functional mobility, and reduced fall risk.
Sánchez-Herrera-Baeza et al., 2020 [39]	Immersive VR (Oculus Rift 2, OR2-LMC)	Parkinson N = 6 (5 M, 1 F) No control group Age range: 69–80	Duration: 6 weeks Frequency: 3/week Session length: 30 min	Strength: JHD Manual dexterity, coordination, fine motor speed: BBT *, PPT * Upper-limb performance: ARAT * Global satisfaction: CSQ-8 *	Improved strength, gross/fine coordination, and speed of movement on the affected side.
Schuch et al., 2020 [40]	Immersive VR (VR Box)	Parkinson N = 23 (16 M, 7 F) Experimental group: n = 11 (mean age: 63 ± 2.80) Control group: n = 12 (mean age: 69 ± 2.3)	Duration: 5 weeks Frequency: 2/week Session length: 28 min (8 min warm-up + 20 min VR)	Motor function, gait, and balance: UPDRS-III, 10MWT, TUG. Physical activity: IPAQ * General cognition: MMSE * Memory functioning: RBMT-3 * Anxiety: STAI *	No significant improvements in balance, mobility, or cognition.

* Abbreviations (in alphabetical order): 6MWT = 6-Minute Walk Test; 10MWT = 10 Meter Walk Test; ABC = Activities-specific Balance Confidence scale; ADLs = Activities of Daily Living; ARAT = Action Research Arm Test; BBS = Berg Balance Scale; BBT = Box and Block Test; CAREN = Computer Assisted Rehabilitation Environment; COPE = Coping Orientation to Problems Experienced inventory; CRI-q = Cognitive Reserve Index questionnaire; CSQ-8 = Client Satisfaction Questionnaire; F = Females; FAB = Frontal Assessment Battery; FES-I = Falls Efficacy Scale; FIM = Functional Independence Measure; FSS = Fatigue Severity Scale; FTSST = Five Times Sit-to-Stand Test; GAS = Goal Attainment Scaling; GEQ = Game Experience Questionnaire; HRS-D = Hamilton Rating Scale for Depression; IPAQ = International Physical Activity Questionnaire; JHD = Jamar Hand Dynamometer; M = Males; MI = Motor Imagery; MMSE = Mini-Mental State Examination; MoCA = Montreal Cognitive Assessment; MSQOL = Multiple Sclerosis Quality of Life; PDQ-39 = Parkinson’s Disease Questionnaire; PPT = Purdue Pegboard Test; PT = Physical Therapy; QoL = Quality of Life; RBMT-3 = Rivermead Behavioral Memory Test third edition; SSQ = Simulator Sickness Questionnaire; STAI = State-Trait Anxiety Inventory; SUS = System Usability Scale; TS = Tinetti Scale; TUG = Timed Up and Go test; UPDRS = Unified Parkinson’s Disease Rating Scale; VADL = Vestibular disorders Activities of Daily Living scale; VR = Virtual Reality; VRRS = Virtual Reality Rehabilitation System.

Table 1. Overview of the reviewed studies employing VR technology for physical rehabilitation in individuals with neurodegenerative diseases.

Author(s), Year	VR *	Sample	Intervention	Assessment	Main Findings
Formica et al., 2023 [34]	Semi-immersive VR (CAREN * system)	Parkinson N = 31 (18 M *, 13 F *) No control group Age range = 18–73	Duration: 2 months (8 weeks) Frequency: 3/week Session length: 50 min	Gait and balance: 10MWT *, BBS * General cognitive functioning: MoCA * Executive functioning: FAB * Coping strategies: COPE * Fear of falling: FES-I * Depression: HRS-D *	Cognitive and emotional benefits (executive function, anxiety, depression); greater physical and cognitive effort in response to stress.
Honzíková et al., 2025 [37]	Immersive VR (Meta Quest 2)	Parkinson N = 19 (sex not specified) No control group Mean age = 64.2 ± 12.8	Duration: 1 month (4 weeks) Frequency: 2/week Session length: 20 min	Gait and balance: 10MWT, BBS Functional mobility: TUG * + dual task QoL *: PDQ-39 *	Improvements in stability, mobility, and quality of life.
Imbimbo et al., 2021 [35]	Semi-immersive VR (Nirvana system)	Parkinson N = 26 (22 M, 4 F) No control group Age range = 66.25–75.75	Duration: 6 weeks Frequency: 2/week Session length: 45 min	Gait and balance: 6MWT *, BBS Cognitive reserve: CRI-q *	Improved balance and gait in patients with higher cognitive reserve.
Kashif et al., 2022 [29]	Non-immersive VR (wall-mounted display, Wii controllers and Wii Fit board)	Parkinson N = 44 (25 M, 19 F) Experimental group: n = 22 Control group: n = 22 Age range: 50–80	Duration: 12 weeks Frequency: 3/week Session length: 60 min (40 min session + 20 min walking and cycling)	Motor function and balance: UPDRS *-III, BBS, ABC * ADLs *: UPDRS-II	Improved motor function, balance, and ADLs.
Kashif et al., 2024 [30]	Non-immersive VR (wall-mounted display, Wii controllers and Wii Fit board)	Parkinson N = 60 (33 M, 27 F) VR group: n = 20 MI * group: n = 20 PT * group: n = 20	Duration: 12 weeks Frequency: 3/week Session length: 60 min (20 min VR/MI + 40 min PT)	Motor function and balance: UPDRS-III, BBS, ABC ADLs: UPDRS-II	VR showed better outcomes in balance and motor function vs. MI and PT. Best results were obtained for the VR + PT combination.
Malisky et al., 2024 [31]	Non-immersive VR (wall-mounted display, Wii controllers and Wii Fit board)	Spinocerebellar ataxia N = 28 (20 M, 8 F) No control group Age range: 15–70	Duration: 10 weeks Frequency: 2/week Session length: 50 min	Balance: ABC ADLs: VADL *	Improved balance and gait, and reduced fall frequency.
Manuli et al., 2020 [32]	Semi-immersive VR (Nirvana system, CAREN system) Non-immersive VR (VRRS *)	Multiple sclerosis N = 84 (47 M, 37 F) No control group Age range: 18–75	Duration: 8 weeks Frequency: 3–5/week Session length: 60 min	System usability: SUS * Goal achievement: GAS * Well-being perception: MSQOL *	Improved perceived QoL (physical and mental) and achievement of therapeutic goals. High satisfaction and usability.
Mazzari et al., 2025 [33]	Non-immersive VR (Technobody devices)	Parkinson N = 18 (sex not specified) Patients: n = 9 Controls: n = 9 Age range: 55–85	Duration: 8 weeks Frequency: 2/week Session length: 60 min	Motor function and balance: TUG, BBS	Improvements in balance, gait, trunk flexion, pain threshold, and erector spinae displacement.
Pullia et al., 2023 [36]	Semi-immersive VR (C-Mill system)	Parkinson N = 20 (13 M, 7 F) Experimental group: n = 10 Control group: n = 10 Age range: 50–70	Duration: 5 weeks Frequency: 4/week Session length: 45 min	Motor function, gait, and balance: 10MWT, 6MWT, TUG, UPDRS-III, TS *, BBS Fear of falling: FES-I ADLs: FIM *	Both conventional and VR training improved motor function; treadmill + VR improved endurance and postural control.
Rodríguez-Fuentes et al., 2024 [38]	Immersive VR (Meta Quest 3)	Multiple sclerosis N = 18 (5 M, 13 F) Control group: n = 10 Experimental group: n = 8 Age range: 18–65	Duration: 5 weeks Frequency: 3/week Session length: 30 min	Motor function, gait, and balance: TS, TUG Functional mobility and strength: TUG + dual task, FTSST *, JHD * Fatigue: FSS * Reaction time: Rezzil software (1.9.0 version) VR-related assessments: SSQ *, SUS, GEQ, * ad hoc satisfaction questionnaire Perceived effort: Borg scale	High usability and satisfaction. Improved lower-limb endurance, functional mobility, and reduced fall risk.
Sánchez-Herrera-Baeza et al., 2020 [39]	Immersive VR (Oculus Rift 2, OR2-LMC)	Parkinson N = 6 (5 M, 1 F) No control group Age range: 69–80	Duration: 6 weeks Frequency: 3/week Session length: 30 min	Strength: JHD Manual dexterity, coordination, fine motor speed: BBT *, PPT * Upper-limb performance: ARAT * Global satisfaction: CSQ-8 *	Improved strength, gross/fine coordination, and speed of movement on the affected side.
Schuch et al., 2020 [40]	Immersive VR (VR Box)	Parkinson N = 23 (16 M, 7 F) Experimental group: n = 11 (mean age: 63 ± 2.80) Control group: n = 12 (mean age: 69 ± 2.3)	Duration: 5 weeks Frequency: 2/week Session length: 28 min (8 min warm-up + 20 min VR)	Motor function, gait, and balance: UPDRS-III, 10MWT, TUG. Physical activity: IPAQ * General cognition: MMSE * Memory functioning: RBMT-3 * Anxiety: STAI *	No significant improvements in balance, mobility, or cognition.

* Abbreviations (in alphabetical order): 6MWT = 6-Minute Walk Test; 10MWT = 10 Meter Walk Test; ABC = Activities-specific Balance Confidence scale; ADLs = Activities of Daily Living; ARAT = Action Research Arm Test; BBS = Berg Balance Scale; BBT = Box and Block Test; CAREN = Computer Assisted Rehabilitation Environment; COPE = Coping Orientation to Problems Experienced inventory; CRI-q = Cognitive Reserve Index questionnaire; CSQ-8 = Client Satisfaction Questionnaire; F = Females; FAB = Frontal Assessment Battery; FES-I = Falls Efficacy Scale; FIM = Functional Independence Measure; FSS = Fatigue Severity Scale; FTSST = Five Times Sit-to-Stand Test; GAS = Goal Attainment Scaling; GEQ = Game Experience Questionnaire; HRS-D = Hamilton Rating Scale for Depression; IPAQ = International Physical Activity Questionnaire; JHD = Jamar Hand Dynamometer; M = Males; MI = Motor Imagery; MMSE = Mini-Mental State Examination; MoCA = Montreal Cognitive Assessment; MSQOL = Multiple Sclerosis Quality of Life; PDQ-39 = Parkinson’s Disease Questionnaire; PPT = Purdue Pegboard Test; PT = Physical Therapy; QoL = Quality of Life; RBMT-3 = Rivermead Behavioral Memory Test third edition; SSQ = Simulator Sickness Questionnaire; STAI = State-Trait Anxiety Inventory; SUS = System Usability Scale; TS = Tinetti Scale; TUG = Timed Up and Go test; UPDRS = Unified Parkinson’s Disease Rating Scale; VADL = Vestibular disorders Activities of Daily Living scale; VR = Virtual Reality; VRRS = Virtual Reality Rehabilitation System.

The duration of the intervention program ranged from five to twelve weeks, with frequencies ranging from two and six sessions per week, and individual session lengths varying from twenty to sixty minutes. The physical interventions were mainly delivered in the form of exergames. Regarding VR tasks, they primarily targeted motor and cognitive rehabilitation, involving gait training, static and dynamic balance, coordination, strength, joint mobility, activities of daily living (ADLs), and cognitive components such as memory and motor planning. These tasks included exercises such as walking in immersive scenarios [34,36,37], performing specific movements to improve static and dynamic balance [29,30,31,33,34,35,36,40], and engaging in games that involve coordination and dexterity [29,30,31,33,34,35,39]. Some studies employed progressive, adaptive methodologies, adjusting task difficulty individually according to participants’ tolerance or performance [29,30,33,34,35,36,37,38], whereas others used standardized scenarios with systematically increasing complexity [34,35]. Moreover, dual-task training protocols, which involve both cognitive and motor aspects, were implemented in some studies [37,39]. Additionally, several interventions provided visual [29,30,31,32,33,34,35,36,37], auditory [29,34,35,36,37], and, in some cases, tactile feedback [34,35] to enhance the patient’s experience and performance.

Motor function, gait, and balance were assessed employing tasks such as the Unified Parkinson’s Disease Rating Scale (UPDRS) part III [29,30,36,40], the Timed Up and Go test (TUG) [33,36,38,40], the 10 Meter Walk Test (10MWT) [34,36,37,40], the 6-Minute Walk Test (6MWT) [35,36], the Tinetti scale [36,38], and the Berg Balance Scale (BBS) [29,33,34,35,36,37,39]. Balance confidence was also measured in some studies with the Activities-specific Balance Confidence scale (ABC) [29,30,31]. Functional mobility and physical activity were assessed in three studies [37,38,40]. Strength was also measured with a hand dynamometer in two of the reviewed studies [38,39], and Rodríguez-Fuentes et al. [38] also considered fatigue. Sánchez-Herrera-Baeza et al. [39] centered on the assessment of manual dexterity, coordination, fine motor speed, and upper-limb performance.

The cognitive function of the subjects was the focus of some of the reviewed studies. Specifically, the Montreal Cognitive Assessment (MoCA) [34] and the Mini-Mental State Examination (MMSE) [40] were employed for the assessment of general cognition; the Frontal Assessment Battery (FAB) [34] for the assessment of executive functions; the third edition of the Rivermead Behavioral Memory Test (RBMT-3) [40] for the assessment of memory functioning; and the Rezzil software (1.9.0 version) for the assessment of reaction time [38]. Interestingly, cognitive reserve was also considered in one study [35]. Other studies also collected psychological outcomes through the assessment of anxiety and depression symptoms [34,40], the utilization of coping strategies in response to stressful life events [34], the assessment of fear of falling [34,36], and the perception of the achievement of goals [32].

The Quality of Life (QoL) and ADLs were assessed in some studies with the Functional independence measure (FIM) [36], the Vestibular Disorders Activities of Daily Living Scale (VADL) [31], the standardized Parkinson’s Disease Questionnaire (PDQ-39) [37], the UPDRS part II [29,30], and the Multiple Sclerosis Quality of Life (MSQOL) [32]. Complementary, some studies obtained VR-related outcomes, such as the subjective usability of the VR device [32,38], the presence of cybersickness [38], and the feel after stopping use of the VR [38]; ad hoc satisfaction questionnaires were also completed [38]. Additionally, the global satisfaction with the intervention and the perceived effort expended during the intervention were measured in two studies [38,39].

3.2. Study Quality and Validity

The risk of bias was assessed using the JBI Critical Appraisal Checklists [41,42,43,44]. Cross-sectional and quasi-experimental studies [31,32,34,35,37] generally showed a moderate risk of bias due to small sample sizes, the absence of control groups, and the use of short pre–post designs. However, some studies reported no attrition and performed adequate statistical analyses [34,37]. Only Mazzari et al. [33] included a control group, strengthening internal validity. By contrast, Manuli et al. [32] applied a personalized treatment approach within a single group, enhancing ecological validity but limiting statistical comparability. Randomized controlled trials (RCTs) [29,30,36,38,40] demonstrated higher methodological rigor, particularly in outcome assessment. However, most were only partially blinded and provided limited details on allocation concealment [29,30,36,38], with Schuch et al. [40] representing the most robust design. Only one study employed a mixed-methods design that incorporated qualitative data [39]. Nevertheless, the theoretical stance of the researchers was not explicitly stated, limiting transparency in the interpretation of findings (for further details, see Supplementary Material Table S2).

The majority of the studies included small sample sizes, with numerous studies including fewer than 30 participants [31,33,35,36,37,38,39,40], a circumstance that diminishes the statistical power and the generalizability of the findings. The highest recorded sample size was 84 subjects [32].

Regarding the duration of the intervention, studies carried out programs which lasted between 4 [37] and 12 [29,30] weeks, with no studies including long-term follow-up. Most of the studies stablished an intervention duration between 5 and 8 weeks [32,33,34,35,36,38,39,40].

Although all studies employed validated instruments to assess cognitive or motor functions, some of them lacked standardized and validated measures to assess user satisfaction or perceived change [29,30,34,35,40].

While several studies employed controlled designs that could allow for rigorous randomization procedures [29,30,33,36] none of them provided specific information on how participants were randomized into different groups. Moreover, half of the reviewed studies [31,32,34,35,37,39] did not include a control group.

3.3. Intervention Effectiveness

3.3.1. Motor Function and Balance

Most of the reviewed studies reported significant improvements in balance, gait, and motor function [29,30,31,33,35,36,37]. These improvements were particularly evident in measures such as the BBS, the TUG, and the UPDRS-III. Interestingly, Kashif et al. [30] observed more improvements in balance and motor function in people with Parkinson’s Disease who received a VR-based intervention compared to those who received motor imagery and physical activity interventions separately. Moreover, the combination of physical therapy and VR was the intervention which obtained the best results.

3.3.2. Cognitive and Emotional Function

Among those studies which explored cognition, one of them observed benefits after the VR-based interventions, particularly in attention, inhibitory control, and processing speed [34]. In contrast, Schuch et al. [40] conducted correlational analyses and found no significant associations between memory functioning and intervention measures. Moreover, Rodríguez-Fuentes et al. [38] explored the efficacy of an immersive VR intervention in patients with multiple sclerosis. Participants were divided into two groups. Both groups received standard therapies at a center, but the experimental group also participated in an immersive VR program. Small improvements in reaction time were observed in both groups, though no significant differences were found between the groups after the intervention. Regarding psychological benefits, Formica et al. [34] found improvements in depressive symptoms and coping strategies. In contrast, the study of Schuch et al. [40] did not find significant correlations between anxiety symptoms and intervention outcomes.

3.3.3. Activities of Daily Living and Quality of Life

Some studies reported improvements in QoL. Specifically, two studies observed higher functional autonomy, higher perceived mobility, and better performance in ADLs over time [29,30,36,37]. Two other studies found improvements in subjective perceptions of physical, mental, and/or emotional well-being [32,37]. However, not all studies found significant improvements after a VR-based intervention [31].

3.3.4. Subjective Assessment and Usability

In the reviewed studies, VR devices were well tolerated by the participants, who also reported high levels of satisfaction and usability. Manuli et al. [32] observed that patients with multiple sclerosis perceived the robotic devices they used in the study, including semi-immersive and non-immersive VR tools, as highly usable and well-tolerated. Similarly, Rodríguez-Fuentes et al. [38] explored the feasibility and safety of an immersive VR device in multiple sclerosis patients, finding high usability rates and low rates of adverse symptoms such as disorientation and nausea (commonly referred to as cybersickness) [45]. Moreover, most patients perceived the VR-based intervention positively and generally recommended it. Additionally, the study of Sánchez-Herrera-Baeza et al. [39] employed immersive VR technology for upper-limb rehabilitation in people with Parkinson’s disease. The study observed a high degree of participant satisfaction and an absence of adverse side effects during the intervention. Overall, these results support the notion that VR devices are generally well tolerated and positively evaluated by users, regardless of the level of immersion.

4. Discussion

This systematic review assessed the benefits, limitations, and future challenges of immersive and non-immersive VR in neurodegenerative disease rehabilitation. Most studies reported improvements in motor function, balance, gait, cognition, and emotional well-being, particularly in Parkinson’s disease. These results support VR as an effective, innovative tool for enhancing physiological and psychological recovery and improving patient quality of life.

A substantial proportion of the articles reviewed emphasize the role of these physical therapies in improving sensorimotor deficits commonly observed in individuals with neurodegenerative diseases that affect motor-related brain systems (e.g., basal ganglia or cerebellum) [46]. Evidence from randomized controlled trials with high methodological quality supports the efficacy of VR-based interventions in neurorehabilitation. Improvements in balance, gait, and motor function (e.g., tremor, rigidity, bradykinesia) have been consistently reported [29,30,36,38,40], with VR contributing to the clinically meaningful maintenance of these gains. Several authors highlighted substantial improvements in stability and mobility in patients with Parkinson’s disease and spinocerebellar ataxia [31,35,36]. Mazzari et al. [33] showed that VR interventions can specifically address postural abnormalities, such as trunk flexion. Additionally, Honzíková et al. [37] emphasized the value of cognitive–motor VR training in promoting task automation and improving motor planning. Similar outcomes have also been described in multiple sclerosis, particularly in mobility and lower-limb strength [38]. However, Rodríguez-Fuentes et al. [38] found no evidence for greater long-term efficacy of VR compared to conventional therapy, although a reduction in perceived effort during VR sessions emerged as a key factor for patient adherence. Likewise, Schuch et al. [40] observed that, while VR was safe, feasible, and well tolerated, short-term interventions were insufficient to promote significant improvements in movement smoothness or cognitive outcomes in Parkinson’s disease. VR appears to be a safe and motivating tool for physical rehabilitation, enhancing patient engagement and facilitating motor learning through external sensory inputs; however, the long-term sustainability of its clinical benefits remains uncertain and should be established as a critical objective for future longitudinal trials.

Some small pilot and cross-sectional studies reported cognitive benefits in attention, inhibitory control, and processing speed [34], as well as increased cognitive reserve [35]. However, the presence of prior therapist involvement in both groups may have masked further differences. In contrast, quasi-experimental designs without control groups failed to demonstrate significant cognitive gains [37], and similar null findings were reported in randomized controlled trials such as that of Schuch et al. [40]. It is of note that only Rodríguez-Fuentes et al. [38] observed faster response times post-treatment across groups, suggesting that processing speed may be sensitive to physical training regardless of VR use. The authors highlighted the possibility that the implicit cognitive training (e.g., virtual slackline balance) incorporated in the motor task may have exerted an insufficient cognitive load. The one-month duration and the absence of booster sessions are likely contributors to the absence of significant cognitive changes. Maggio et al. [20] reported significant executive function improvements after a two-month, high-intensity intervention, emphasizing the importance of sustained protocols and long-term follow-up. These findings are consistent with those of Xiao et al. [47], who proposed that multisensory, embodied interventions enhance cognition through BDNF-mediated hippocampal plasticity. However, given the transient nature of these neuroplastic effects, they may dissipate in the absence of sustained stimulation. This phenomenon could provide a rationale for the limited short-term efficacy observed in brief dual-task training protocols.

Emotional benefits were mainly reported in small pilot studies without control groups. Formica et al. [34] observed reductions in anxiety and depressive symptoms, along with improvements in psychological well-being and quality of life. Similarly, some studies noted that functional progress, such as improved motor performance and greater independence in daily activities [29,30,36,37], were often accompanied by a more positive self-perception of physical, mental, and emotional well-being [32,37]. Although Malisky et al. [31] did not report statistically significant changes in ADLs, the reduction in VADL median scores suggested clinically relevant, though still emerging, improvements. Regarding RCTs, only Schuch et al. [40] evaluated the emotional domain, reporting no significant effects on anxiety. By contrast, most trials focused on functional outcomes in ADLs. Pullia et al. [36] found that patients receiving VR-based interventions reported greater perceived independence compared with conventional therapies, and Kashif et al. [29,30] further demonstrated that these clinical gains were sustained in the medium term. Importantly, Rodríguez-Fuentes et al. [38] emphasized the high usability and good tolerance of VR devices, with participants reporting high satisfaction and minimal adverse effects during rehabilitation. This distinction highlights that motor and functional outcomes exhibit robustness and consistency across trials, whereas evidence regarding the cognitive and emotional domains remains preliminary, reflecting the need for more rigorous and sustained interventions to clarify their VR impact.

While the evidence spans multiple domains, most VR interventions focused on changes in motor and functional tests. The clinical interpretation of outcomes is complicated by the heterogeneity of assessment tools. Performance-based tasks (10MWT, 6MWT, TUG, FTSST, ARAT, JHFT, BBT, PPT) provide objective and ecologically valid indices of function, though they are less sensitive to subtle changes in neurorehabilitation. Clinician-rated scales such as UPDRS, BBS, or TS offer standardized and sensitive measures of motor deficits. However, they are still partly influenced by subjective judgment. Finally, self-reported measures (ABCS, FES-I, Borg) capture patients’ perception of effort and functional capacity, adding a valuable patient-centered perspective but also reflecting motivational and affective biases.

VR-based rehabilitation enhances motor function across domains, improving gait endurance (6MWT), stability, TUG performance, lower-limb strength (FTSST), and upper-limb dexterity and grip (JHD, BBT, PPT) [35,36,38,39]. In contrast, gait speed (10MWT) shows more variable outcomes, likely due to lower sensitivity [34,36,37,40]. The importance of personalized interventions is emphasized by the presence of cognitive reserve and individual variability [35]. In line with objective performance-based measures, clinical observations indicate that physical rehabilitation using VR yields consistent improvements in motor function (UPDRS-III) [29,36], particularly bradykinesia [30], although Schuch et al. [40] reported no significant change in UPDRS-III. Similarly, gains in activities of daily living (UPDRS-II) [29,30] and balance (BBS) [29,30,33,34,36] are robust, with greater gains reported in patients with higher cognitive reserve, suggesting that personalized approaches may optimize outcomes [35]. However, Malisky et al. [31] observed no improvement in perceived balance safety, potentially reflecting disease progression and duration. Nevertheless, reductions in fall risk (TS) are more variable, indicating that the observed benefits may be context-dependent and influenced by individual patient profiles [36,38]. Self-reported measures (ABCS, FES-I, Borg scale) indicate improved balance confidence, reduced fear of falling, and lower perceived exertion [29,30,34,36], even in cases where performance-based tasks (e.g., ARAT) demonstrate minimal change [39]. This highlights the complementary value of patient-reported outcomes in capturing the motivational and experiential benefits of VR rehabilitation.

In addition to improving clinical outcomes, VR provides a powerful experimental framework for investigating and training the underlying processes of motor control. Unlike conventional rehabilitation tools, VR environments can isolate and manipulate the feedforward (predictive) and feedback-driven mechanisms of motor control. Recent evidence shows that performance tracking in immersive VR is influenced by the interplay between feedforward planning and real-time error correction and that these mechanisms are affected differently by aging and neurological status [48]. In the context of neurodegenerative diseases, this distinction is crucial. While improvements in standardized clinical scales (e.g., UPDRS and BBS) reflect functional recovery, VR tasks reveal whether gains derive from enhanced predictive control (e.g., anticipatory gait adjustments) or more efficient error correction (e.g., balance recovery after perturbations). This mechanistic perspective suggests that VR delivers therapy and serves as a laboratory for testing hypotheses about sensorimotor adaptation and internal model updating. This dual role of VR could guide the precise tailoring of interventions. For example, protocols could be designed that explicitly target predictive planning versus feedback responses, depending on the deficits of the patient.

Methodologically, the effectiveness of VR-based interventions may depend on individual factors such as cognitive reserve, disease stage, and patient engagement [32,35,38]. A consensus has been reached among authors regarding the critical role of personalized protocols and adapted virtual environments in the maximization of therapeutic outcomes. For instance, advanced VR systems such as CAREN and Nirvana have demonstrated high efficacy in controlled clinical settings for improving balance, cognitive–motor integration, gait, and executive functioning [32,35]. In contrast, semi-immersive platforms and exergaming approaches have proven to be more accessible and better tolerated by certain patient profiles, while still promoting significant functional and cognitive benefits.

A promising direction concerns the use of adaptive VR systems capable of stimulating different functional domains with varying intensities. As shown in several of the included studies (see Supplementary Table S1), interventions frequently integrated progressive adaptation strategies. For instance, Kashif et al. [29,30] employed Wii Fit-based exergames where difficulty levels were systematically increased according to patient performance [29,30]. Similarly, Mazzari et al. [33] progressively adjusted trunk- and gait-training tasks using Technobody devices, beginning with partial body-weight support and increasing the challenge over time [33]. Honzíková et al. [37] incorporated adaptive tolerance thresholds in dual-task modules, while Pullia et al. [36] enabled physiotherapists to modify treadmill-based VR scenarios in real time. Building upon these approaches, future developments could integrate automatic planning techniques to generate individualized stimulation schedules that dynamically balance motor, cognitive, and emotional demands. Such techniques would not only personalize therapy but also optimize training load, ensuring that patients receive stimulation tailored to their clinical profile and rehabilitation goals. This perspective highlights the potential of VR not only as a therapeutic delivery tool but also as a platform for intelligent, data-driven rehabilitation planning.

The versatility of VR environments has enabled the design of tasks that simultaneously engage the physical and cognitive domains. This offers a functionally meaningful and ecologically valid approach to addressing the multifaceted deficits seen in neurodegenerative diseases [34,35,37,38]. As indicated by several authors, the simultaneous targeting of motor and cognitive functions within rehabilitation paradigms is of particular importance [31,32,37]. These studies suggest that a dual-task strategy may be beneficial in promoting the generalization of skills.

In comparative studies between traditional and VR-based interventions, only the groups exposed to VR demonstrated significant improvements in real-world functional tasks [29,30,36,38]. This indicates that VR may facilitate the transfer of learned skills to real-world daily activities. This effect was especially pronounced in studies utilizing exergaming platforms, where increased motivation and game-based interaction appeared to enhance patient engagement and therapeutic adherence [30,33,38]. From a clinical perspective, the current evidence indicates that VR interventions are generally feasible and well tolerated, even among older adults, with low reports of adverse effects such as cybersickness. High levels of usability and satisfaction across diverse platforms [32,38,39] suggest that patient adherence is likely to be maintained when VR is implemented in real-world clinical settings. However, practical considerations remain critical for widespread adoption. First, the acquisition and maintenance costs of advanced VR systems, such as CAREN or Technobody devices, may limit their availability in public healthcare systems, whereas lower-cost non-immersive exergaming platforms appear more accessible and scalable. Second, the effective use of VR technologies requires specialized training for clinicians and rehabilitation staff, which may present organizational challenges. When compared to conventional physiotherapy, VR shows added value in terms of motivation, engagement, and potential for dual-task training. However, it is important to note that, at present, VR should be regarded as a complement to, rather than a replacement for, conventional physiotherapy. At this stage, the evidence does not yet justify a broad shift in resources toward VR-only rehabilitation. Instead, hybrid models that integrate VR with established therapies may offer the most effective and sustainable approach. Subsequent research endeavors must also address cost-effectiveness and implementation strategies to support the translation of VR interventions into standard clinical practices.

However, the findings should be interpreted with caution due to several notable methodological limitations. Despite the presence of controlled designs in numerous studies [29,30,33,36], the absence of transparency in randomization procedures hinders a precise assessment of methodological quality and gives rise to concerns regarding potential selection bias. Approximately half of the reviewed studies lacked control groups [30,32,34,35,37,39], thereby precluding randomization and hindering causal interpretation. Additionally, several studies were conducted in fields such as software engineering, rather than in clinical neuroscience, often overlooking crucial neuropsychological considerations. Moreover, none of the twelve studies included follow-up assessments, leaving the long-term effectiveness of VR interventions largely unknown [49,50].

Several high-quality randomized controlled trials provide strong evidence for the added value of VR over conventional rehabilitation, typically based on physiotherapy and cognitive exercises [29,30,36,38]. These studies report notable improvements in motor, cognitive, and emotional domains in patients with Parkinson’s disease, along with high levels of patient satisfaction and adherence during the rehabilitation process. Importantly, high acceptability and tolerability of VR technologies were observed, even among older adults aged between 69 and 80 years [39], a critical consideration for clinical translation given the prevalence of elderly populations in neurorehabilitation settings.

Several barriers to the widespread clinical adoption of VR in physical rehabilitation persist. The limited availability of VR equipment in public healthcare systems, the requirement for specialized technical training, and the high cost of advanced devices represent significant challenges. Despite these constraints, the current evidence supports VR as a safe and well-tolerated intervention with strong potential to become an integral component of neuropsychological rehabilitation. Future research should address key gaps, including larger sample sizes, randomized controlled trials with active control groups, and long-term follow-up assessments. The substantial heterogeneity in software platforms, hardware configurations, and intervention protocols currently limits the generalizability and clinical applicability of the findings, emphasizing the need for standardized, accessible, and evidence-based protocols. Moreover, few studies have objectively assessed the cognitive impact of dual-task VR training, particularly on specific cognitive domains. A more profound understanding of these effects is imperative for enhancing the efficacy of clinical and neuropsychological interventions. Subsequent research should include comparative studies across neurodegenerative disorders to identify disorder-specific cognitive, motor, and psychological response patterns. Such investigation would facilitate the development of interventions that are more precise in their alignment with the distinct clinical profiles and rehabilitative needs of each patient population.

In summary, this review presents some limitations that should be acknowledged when interpreting its findings. First, the search strategy was restricted to the Web of Science and Scopus, which, although ensuring broad interdisciplinary coverage, may have excluded relevant studies indexed exclusively in other databases. Likewise, the decision to limit the search to the past five years allowed us to focus on contemporary VR technologies but inevitably excluded earlier work that may provide valuable insights into historical trends in the field. Regarding the reviewed studies, most were characterized by small sample sizes, the absence of control groups, and a lack of detailed information on randomization procedures, which reduces methodological robustness, and limits the generalizability of the results. Moreover, none of the studies included long-term follow-up assessments, leaving the durability of the reported effects uncertain. Finally, the marked heterogeneity in the devices, software platforms, intervention protocols, and outcome measures hinders direct comparisons across studies and constrains the clinical applicability of the current evidence.

5. Conclusions

This systematic review provides evidence supporting the therapeutic potential of VR in the physical rehabilitation of neurodegenerative diseases, particularly Parkinson’s disease and multiple sclerosis. Across diverse levels of immersion and intervention protocols, VR interventions were consistently associated with improvements in motor function, including balance, gait, and postural control. In some cases, these interventions were also associated with enhancements in cognitive performance and emotional well-being. These benefits were frequently accompanied by high levels of user satisfaction, usability, and adherence, favorable factors for clinical implementation, even among older adults. Performance-based outcomes are corroborated by clinician observations and patient self-reports, reflecting context- and patient-dependent effects, with self-perception capturing additional motivational and experiential dimensions. However, the methodological limitations, including the small sample sizes, lack of control groups, and the absence of long-term follow-up, require cautious interpretation of these promising findings. It is imperative that research in this area moving forward prioritize the design of larger, methodologically rigorous randomized controlled trials, with standardized outcome metrics and extended follow-up periods to assess the durability of treatment effects. In the future, researchers should adopt multimodal assessment strategies that combine clinician-rated, self-reported, and performance-based outcomes to maximize both sensitivity and ecological validity, while better capturing the full therapeutic impact of VR interventions. Emphasis should be placed on cost-effectiveness, implementation in public health systems, and the personalization of VR protocols based on cognitive reserve, motivation, and functional capacity. Despite the present obstacles, VR emerges as a promising, well-tolerated, and increasingly feasible modality to complement conventional neurorehabilitation, with the potential to enhance autonomy and quality of life in neurodegenerative populations.