Traumatic brain injury (TBI) refers to focal or diffuse neurological damage resulting from various external mechanical forces, such as impact, rapid acceleration or deceleration, or penetrating injury (1). TBI affects an estimated 55 million people worldwide each year and remains a leading cause of injury-related mortality and long-term physical and cognitive disability, particularly among young adults (2–5). Given its substantial health and socioeconomic burden, TBI represents a major public health challenge (2). Patients with TBI commonly experience motor, cognitive, and affective impairments that significantly impact their quality of life and that of their caregivers alike (6). Among these, cognitive deficits are particularly concerning, as they hinder rehabilitation outcomes by limiting work capacity, social functioning, and independence in daily activities, thereby compounding the overall disease burden (7, 8).

Cognitive rehabilitation (CR) for TBI aims to restore impaired cognitive functions and develop compensatory strategies to improve daily functioning (9, 10). Over the past three decades, growing evidence has demonstrated that CR can significantly enhance functional outcomes in patients with TBI through both the recovery of lost abilities and the implementation of adaptive strategies (11–14). However, the practical delivery of CR presents notable challenges. These interventions often require substantial time, resources, and financial commitment, which can lead to reduced engagement and poor adherence to treatment protocols, ultimately diminishing their therapeutic effectiveness (15).

In recent years, technological advancements have enabled the development of innovative rehabilitation approaches, such as digital intervention including computer-assisted therapies and virtual reality (VR) interventions (16–18). These methods have demonstrated clinical effectiveness in treating various cognitive disorders, such as those associated with stroke and dementia (19–21). Conventional cognitive rehabilitation typically involves therapist-led interventions, delivered either individually or in group settings, often with support from family members or multidisciplinary teams (22). Technology-mediated interventions (e.g., VR, computer-based training) are increasingly being adopted as viable alternatives to traditional face-to-face cognitive therapy due to their multiple advantages (23).

Computer-based cognitive intervention (CCI) employs digital tools to enhance or restore cognitive functions, including memory, attention, problem-solving, and work-related skills, through targeted exercises and adaptive training programs (20). The potential benefits of CCI include ease of self-administration, improved accessibility, and greater cost-effectiveness (24). Additionally, CCI can boost participant engagement through varied formats (e.g., videos, gamification), unlimited responsiveness, and adaptive feedback, promoting a sense of interactivity and enjoyment (25). Most CCI programs are delivered through digital platforms or mobile applications, enhancing flexibility and convenience-especially for those with mobility challenges or limited access to rehabilitation services-thus significantly improving care accessibility (24). Traditional CCI mainly presented on a standard two-dimensional desktop or laptop monitor. Interaction is typically achieved via mouse, keyboard, or touchscreen. While delivered via computer, these programs offer a limited sense of immersion and presence.

VR is an emerging computer-based technology that provides users with dynamic, three-dimensional simulated environments in which they can interact as if in real physical spaces (26). It primarily employs head-mounted displays that provide a wide field of view and head-tracking, creating a strong sense of presence and immersion. The potential advantages of VR include: (1) dynamically adjusting stimulus intensity and task difficulty based on patient performance, (2) integrating cognitive and functional training to promote neuroplastic recovery, and (3) enabling objective quantification of user behavior and performance metrics (27, 28). As a result, VR enhances training specificity and patient engagement by reducing boredom and frustration through a more sophisticated and ecologically valid methodology (29). VR has demonstrated efficacy as both an assessment and therapeutic tool for addressing motor impairments and cognitive dysfunction, including executive function and functional activities (30, 31).

A systematic review by Alashram (32) reported that computerized cognitive training (CCT) improved various cognitive domains in patients with TBI and that combining CCT with other interventions produced greater benefits than CCT alone. However, the findings were qualitative rather than quantitative, and the review focused solely on CCT, excluding newer technologies such as VR. Several studies employing VR have shown that it can enhance cognitive and behavioral functioning in patients with TBI (16, 17). A review by Andrei et al. (33) further indicated that VR-based interventions significantly improved cognitive functions in TBI patients, particularly in the domains of attention, executive function, and visuospatial abilities. Nonetheless, current evidence supporting the use of VR in cognitive neurorehabilitation for TBI remains limited, and there is no clear clinical consensus regarding its therapeutic efficacy (16). To the best of our knowledge, no meta-analysis has yet examined the effects of digital (computer- and VR-based) cognitive intervention in TBI. Therefore, we conducted a meta-analysis to quantitatively assess the impact of these interventions on cognitive function and to further investigate differences in therapeutic efficacy and contributing factors when compared to passive controls and traditional rehabilitation methods.

This meta-analysis was conducted in strict accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines (34).

Based on the current evidence reviewed, our results demonstrate that digital cognitive intervention significantly improved global cognitive function, executive function, attention, and social cognitive function in patients with TBI. However, no significant improvements were observed in memory, processing speed, ADL, or psychosocial outcomes, including self-efficacy and anxiety/depression. Subgroup analyses revealed that VR-CI resulted in significantly greater improvements in global cognitive and executive functions compared to conventional face-to-face cognitive therapy. Moreover, a higher number of training sessions appeared to enhance cognitive benefits. VR-CI was also found to have beneficial effects on depression in TBI patients.

The manifestation and severity of cognitive dysfunction following TBI are influenced by various factors, including the nature of the injury, psychosocial circumstances, and individual patient characteristics (32). Common cognitive deficits post-TBI include memory impairments (90%), attention difficulties (82%), and executive dysfunction (75%) (40). Digital cognitive intervention leverage technology to enhance cognitive performance through targeted, engaging training paradigms. Computerized programs (e.g., Cogmed, Lumosity) use adaptive tasks to strengthen specific domains such as working memory and attention by reinforcing neural pathways through repetitive and progressively challenging exercises (41). VR interventions immerse users in simulated real-world environments, facilitating multisensory integration and ecological transfer by simultaneously engaging memory, executive functions, and visuospatial skills (42). A fundamental mechanism underlying the effectiveness of these interventions is neuroplasticity—the brain’s capacity to reorganize and form new synaptic connections in response to structured therapeutic stimuli (43, 44). By capitalizing on this neuroadaptive potential, digital interventions can induce cognitive improvements in task-relevant brain regions (e.g., the prefrontal cortex for executive functions) through structured training (45). Additionally, their engaging interfaces and real-time feedback mechanisms enhance patient motivation and adherence, further supporting behavioral and cognitive recovery. Consistent with previous systematic reviews by Alashram (32) and Manivannan et al. (17), our meta-analysis found that digital cognitive intervention significantly enhanced global cognitive function, highlighting their therapeutic promise for cognitive rehabilitation. Subgroup analyses also revealed that patients who completed a greater number of training sessions showed superior cognitive gains. Therefore, a moderate increase in the number of training sessions may help optimize and sustain long-term cognitive benefits.

Two studies involving underage patients with TBI were included in our meta-analysis. Shen et al. found that the VR-CI program was safe for children with TBI and showed promising, though statistically non-significant, benefits for executive function (38). Another study reported that CCI significantly improved working memory in children with TBI, but this benefit was not sustained during the follow-up period (46). Notably, there are substantial developmental differences in cognition between children and adults. The ongoing neuroplasticity of the developing brain may enhance responsiveness to cognitive interventions (32). Therefore, further research is warranted to investigate the safety and efficacy of digital cognitive interventions in pediatric TBI populations.

Our subgroup analyses showed that VR-CI led to significantly greater improvements in global cognitive function and executive functioning compared to traditional face-to-face cognitive therapy. VR technology enables the creation of interactive, multisensory, three-dimensional environments that support comprehensive behavioral monitoring, offering clinical assessment and rehabilitation capabilities beyond those of conventional psychoeducational approaches (26). Executive function encompasses a range of high-level cognitive skills, including goal setting and initiation, planning and organization, task execution, and performance monitoring and regulation (47). These skills are essential for executing goal-directed behaviors and play a critical role in activities of daily living (ADLs). For example, Jacoby et al. employed a virtual reality supermarket to enhance executive function in TBI patients, demonstrating that VR-acquired skills successfully transferred to real-world tasks (27). The multisensory stimulation and interactive nature of VR environments allow patients to repeatedly practice tasks in contexts that closely resemble real-life situations, a benefit not achievable through conventional cognitive rehabilitation. As such, VR-CI represents a promising alternative approach for improving executive function in TBI patients.

Our results showed that CCI, including rehabilitation gaming, did not produce significant improvements in executive function compared to usual care or passive control groups. A previous systematic review reported that CCI could enhance executive function in patients with acquired brain injury, including TBI (48). However, since only three studies in our analysis examined the effects of CCI on executive function, these findings should be interpreted with caution due to the limited evidence base. One of the included studies on gaming-based rehabilitation noted that outcomes could have been influenced by factors such as the participants’ broad age range, variability in game types, their attitudes toward the therapy, among others (49). The computerized interventions in the remaining two studies appeared to primarily target improvements in attention and information processing (50, 51). These findings suggest that personalized CCI programs tailored to the specific cognitive deficits of individual patients may improve treatment efficacy by better addressing their unique therapeutic needs.

Social cognition is a critical domain for individuals with TBI, as it underpins social reintegration, emotional cue interpretation, and interpersonal relationships-functions commonly impaired post-injury (52). Traditional cognitive interventions tend to emphasize non-social cognitive domains such as memory and executive function, often neglecting aspects of social cognition (47, 53). Our findings showed that CCI significantly improved social cognitive function. Although only two studies included social cognition outcomes, their results provide preliminary evidence supporting the therapeutic potential of CCI in this area.

Enhancement of attention is crucial for both functional recovery and active engagement in rehabilitation programs (54). Our meta-analysis found a statistically significant improvement in attention following digital cognitive intervention. Although only two studies included attentional outcomes, their findings provide preliminary evidence supporting the effectiveness of these interventions in this cognitive domain. Therefore, additional high-quality studies are needed to further explore the impact of digital cognitive interventions on attention and social cognition. Several studies have also reported a negative correlation between reductions in depressive symptoms and improvements in attention (55, 56). Given that depressive symptoms are critical determinants of neurorehabilitation efficacy, addressing them is essential. VR environments can incorporate exposure therapy and relaxation exercises, which have been shown to reduce depressive symptoms, phobias, and post-traumatic stress (57). In line with this, our subgroup analysis demonstrated a meaningful reduction in depressive symptoms following VR-CI. However, no significant reduction in anxiety symptoms was observed following digital cognitive interventions. This may be partially due to the limited number of eligible studies included. Additionally, interventions targeting anxiety in TBI patients often involve computer-based cognitive behavioral therapy, which was beyond the scope of our review (58).

Our pooled results did not show significant improvements in memory or processing speed following digital cognitive intervention. This may be attributed to a lack of targeted training focused on these specific domains during the interventions. We extracted results from all memory-related scales used in the included studies, which covered a broad range of memory subtypes, such as working memory, everyday memory, short-term memory, and delayed memory. The varying effects of digital interventions on different memory domains could contribute to the heterogeneity of findings. Furthermore, the lack of sensitivity of some assessment tools may have limited the detection of subtle memory improvements associated with digital training (59). While our findings differ from several previous reviews that reported improvements in memory-including working memory-following digital cognitive interventions in patients with TBI (60, 61). However, both Rodríguez-Rajo et al. (52) and Phillips et al. (62) found no statistically significant differences in memory outcomes between intervention and control groups. Therefore, future research should utilize more sensitive and standardized assessment tools, develop targeted interventions, and conduct larger, high-quality randomized controlled trials to elucidate the efficacy of digital interventions in specific cognitive domains such as memory and processing speed.

No significant improvements in ADL were observed in this review, which may be attributed to the limited number of studies (n = 3) that provided analyzable data using ADL-related scales. Notably, none of these three studies incorporated specific ADL training tasks within their cognitive intervention programs. Prior evidence suggests a well-established association between improvements in executive function and attention and enhanced performance in ADL tasks (63). ADL is a critical domain, as one of the primary goals of CI is to support patients’ reintegration into real-world functional activities. Therefore, future CI studies should include standardized assessments of ADL to better evaluate functional outcomes. Self-efficacy, on the other hand, has demonstrated a significant independent effect on the efficacy of cognitive rehabilitation (59). Encouragement and positive feedback during therapy can enhance patients’ perceived self-efficacy. However, our findings showed no significant improvement in self-efficacy following computer-based cognitive interventions. The study by Tam et al. (59) found that self-efficacy was significantly enhanced in participants who received targeted feedback interventions. As self-efficacy plays a key role in brain injury rehabilitation, future cognitive intervention studies should routinely include standardized measures of self-efficacy and consider integrating motivational elements to support its improvement.

VR-CI has also been demonstrated to improve global cognitive function, executive function and memory among stroke patients compared to control treatments (21). Alashram et al. reported that the efficacy of CCI for visual and verbal working memory in acquired brain injury (ABI) population (64). ABI, encompassing etiologies such as stroke, hypoxic–ischemic injury, and brain tumors, shares with TBI the common endpoint of disrupted neural networks and consequent cognitive impairment. A core sequelae of ABI is the disruption of inherent neuroplasticity, the brain’s ability to reorganize itself by forming new neural connections. Both focal (e.g., stroke) and diffuse (e.g., TBI, anoxia) injuries can impair this process. CCI is fundamentally engineered to harness and drive neuroplasticity through the principles of massed practice and intensity. This intensive practice is believed to promote synaptic strengthening and efficiency within damaged or alternative neural networks (65). The game-like, engaging nature of many immersive VR-CI applications can increase adherence to the high repetitions needed for neuroplasticity (66). Therefore, the efficacy of digital cognitive interventions is not limited to TBI alone but is potentially generalizable to a wide range of ABI patients.

Our meta-analysis quantitatively confirmed the beneficial effects of digital cognitive intervention on cognitive function in patients with TBI. The low heterogeneity observed among the included studies enhances the robustness of the findings. However, this meta-analysis has several limitations. First, the results for some outcomes may have been influenced by the limited number of studies available. Subgroup analyses exploring potential moderating factors were also constrained by the small sample of studies. Several covariates may influence treatment effects, including: (a) the severity and age range of TBI patients (from mild to severe), (b) the time elapsed since injury, (c) the duration of follow-up after the intervention, and (d) the presence or absence of one-on-one coaching during treatment. Second, there is a lack of standardization in cognitive outcome assessments across studies, which underscores the need for more precise and consistent evaluation methods. Third, this review only included English-language publications, which may have introduced geographic and cultural bias. To better understand the effectiveness and influencing factors of digital cognitive intervention in TBI, larger, high-quality randomized controlled trials with more rigorous designs are urgently needed.

Our meta-analysis demonstrated that digital (computer- and VR-based) cognitive intervention have a positive impact on global cognitive function, executive function, attention, and social cognitive function in patients with TBI. However, these interventions did not show significant effects on memory, processing speed, ADL, or psychosocial outcomes (self-efficacy, anxiety, and depression). Subgroup analyses revealed that VR-CI were significantly more effective than traditional cognitive therapy in improving global cognitive and executive functions. VR-CI also showed beneficial effects on depressive symptoms. Moreover, a greater number of training sessions may enhance the cognitive benefits achieved.

In clinical practice, digital cognitive intervention is recommended to enhance cognitive functions in both adult and pediatric patients with TBI. Although more further researches are warranted to investigate the safety and efficacy of digital cognitive interventions in pediatric TBI populations. In global cognitive function and execution function, VR-CI demonstrate superior efficacy compared to conventional cognitive intervention. Given the variability in cognitive deficit profiles and intervention protocols focused on different specific domains, we recommend individualized treatment strategies to more effectively address cognitive impairments in TBI patients. Further research is needed to determine the most appropriate digital cognitive intervention programs tailored to individual patient characteristics.