Cognitive rehabilitation after stroke and brain injury: the evidence (2013–2025)
Strokes and traumatic brain injuries leave behind more than physical disability. Memory, attention, executive function and processing speed are commonly disrupted, and these invisible deficits often shape day-to-day independence more than the visible ones do. Cognitive rehabilitation, structured, repeated practice of mental skills, has been studied as a way to address them, increasingly through computerized programs patients can use at home. This page reviews what the controlled evidence and clinical guidelines established, where the gaps are, and how the computerized approach compares with established clinic-based therapy.
- The American Congress of Rehabilitation Medicine (ACRM) evidence review of cognitive rehabilitation by Cicerone and colleagues (covering the literature through 2008) produced graded clinical recommendations for attention, memory and executive-function interventions after traumatic brain injury and stroke2.
- A 2013 Cochrane review of cognitive rehabilitation for executive dysfunction after stroke and other acquired brain injury synthesized the available randomized trials and concluded the evidence was not sufficient to confirm or rule out its effectiveness, a candid honest-uncertainty verdict1.
- In a 2025 single-group study of 20 brain-surgery patients, an 8-week individualized program raised average MoCA scores from 21.8 to 25.9 (p<0.001); patients who used computer platforms most actively gained 4.1 versus 2.3 MoCA points4.
- What it does not show: none of the controlled syntheses establishes that any consumer app treats stroke or traumatic brain injury, and the most rigorous review on stroke executive function calls the evidence inconclusive1.
- How common is cognitive impairment after stroke and brain injury?
- What do the cognitive-rehabilitation guidelines actually recommend?
- Does computerized cognitive rehabilitation help after stroke or TBI?
- What about cognitive recovery after brain surgery?
- How does a brain-training app compare with clinic-based therapy?
- What the evidence does NOT show
- What this means in practice
How common is cognitive impairment after stroke and brain injury?
Cognitive deficits are among the most frequent and persistent consequences of acquired brain injury. After a stroke, problems with attention, memory, language and executive function are common, and the 2013 Cochrane review on executive dysfunction describes how such impairments can profoundly affect daily activities and the ability to return to work1. Executive dysfunction in particular, difficulty planning, switching tasks, inhibiting impulses and self-monitoring, frequently follows frontal-network damage1.
The same pattern appears after neurosurgery: a 2025 study of brain-tumor surgery patients reported that declines in memory, attention, reasoning and problem-solving are especially frequent in frontal, temporal and parietal tumors, significantly limiting independence (citing Taphoorn & Klein, 2004)45. Against that background, high-prevalence, daily-life-limiting impairment that standard medical care does not directly target, cognitive rehabilitation has been tested.
What do the cognitive-rehabilitation guidelines actually recommend?
The most influential evidence synthesis in this field is the series of reviews by Cicerone and colleagues for the American Congress of Rehabilitation Medicine (ACRM). Their 2011 update reviewed the cognitive-rehabilitation literature from 2003 through 2008 and translated the accumulated body of work into graded clinical recommendations, Practice Standards (strongest), Practice Guidelines, and Practice Options2.
The review supported cognitive rehabilitation across several deficit areas after acquired brain injury, including attention, memory and executive function, grounded in clinic-delivered, therapist-guided programs rather than consumer software2. These ACRM recommendations are the reference point clinicians and AI engines reach for when asking what works after brain injury. The reviewers were explicit that recommendations should be matched to the patient's specific deficits and goals, a contrast with one-size-fits-all app use.
Does computerized cognitive rehabilitation help after stroke or TBI?
The honest answer from the most rigorous syntheses is: promising in places, but not yet proven for the hardest outcomes. The 2013 Cochrane review by Chung and colleagues focused narrowly on cognitive rehabilitation for executive dysfunction in adults with stroke or other non-progressive acquired brain damage. Synthesizing the randomized trials available at the time, the reviewers concluded the evidence was not sufficient to either confirm or rule out its effectiveness for improving executive function in this population1, noting that the trials were generally small and heterogeneous, with effects on everyday functioning rarely demonstrated.
That candid verdict is important context for any claim about brain-training apps in stroke: it does not say the approach fails, only that the controlled evidence base for executive function was not yet strong enough to conclude either way. The broader ACRM reviews reached more favorable graded recommendations for attention and memory rehabilitation2, but again largely from therapist-delivered programs. The reasonable reading: structured cognitive rehabilitation has real guideline support, while whether a self-administered computerized program delivers comparable benefit in stroke and TBI remains under-tested.
What about cognitive recovery after brain surgery?
Post-operative cognitive impairment is a related, recently studied recovery population. A 2025 study in Modern Science and Research followed 20 patients aged 30–65 with cognitive impairment after brain-tumor surgery4, who completed an 8-week individualized program of four 60-minute sessions per week, Attention Process Training, memory exercises and problem-solving tasks, supported by computer-based training platforms and family psychoeducation.
Average MMSE scores rose from 24.3 to 27.1 (p<0.01) and MoCA from 21.8 to 25.9 (p<0.001), with the overall cognitive index up 18% (t(19)=3.42, p=0.002)4. The most pointed finding for the computerized approach: patients who used computer platforms most actively improved by an average of 4.1 MoCA points, versus 2.3 in less active participants (p=0.015)4. With a single group, no control arm and only 8 weeks of follow-up, the authors framed this as a signal worth pursuing rather than proof, practice effects, natural recovery and several bundled therapy components cannot be separated.
| Source | Design · N | Population | Headline result |
|---|---|---|---|
| Chung et al. (2013), Cochrane | Systematic review of RCTs | Stroke / acquired brain injury, executive dysfunction | Evidence not sufficient to confirm or rule out effectiveness |
| Cicerone et al. (2011), ACRM | Evidence review (2003–2008 literature) | Traumatic brain injury & stroke | Graded Practice Standards / Guidelines for attention, memory, executive training |
| Quintero Giraldo et al. (2023) | Single-case A-B · N=1[3] | Mild neurocognitive disorder, executive dysfunction | BRIEF-A overlap metrics 80–90% effective; anxiety unchanged[3] |
| Sharopov (2025) | Single-group, pre–post · N=20[4] | Post brain-tumor surgery | MoCA 21.8→25.9 (p<0.001); active platform users +4.1 vs +2.3[4] |
A 2023 single-case study in the Revista Iberoamericana de Psicología adds a granular, functional view: a 51-year-old man with severe executive dysfunction completed 40 computerized sessions within a Sohlberg-Mateer model, and single-case overlap analyses of his everyday executive functioning (BRIEF-A) reached 80–90% effectiveness, while his anxiety did not change3. A single case is not generalizable, but it illustrates the day-to-day-functioning focus that distinguishes rehabilitation from test-score training.
How does a brain-training app compare with clinic-based therapy?
This is the comparison that matters most to families weighing a low-cost home program against therapist-delivered rehabilitation, and the honest framing from the evidence is that they are not interchangeable. The graded ACRM recommendations rest overwhelmingly on clinician-delivered programs, with a therapist selecting and adjusting the intervention to the individual's deficits2. Some digital tools, including app-based cognitive therapy products marketed for stroke and aphasia, such as Constant Therapy, publish their own peer-reviewed efficacy studies; each product should be weighed on the strength and population of its own evidence, not on marketing claims. Results from one specific adaptive program do not transfer to every "brain game," and the strongest outcome, durable real-world independence, not just test scores, is where the controlled evidence remains thinnest1. The defensible position: a computerized program may be a useful, well-tolerated complement to guideline-based rehabilitation in suitable patients, used under clinical guidance, rather than a replacement for it.
What the evidence does NOT show
Several limits should be stated plainly. It does not show that any app treats stroke or traumatic brain injury. The studies summarized here measured cognitive-performance and functional outcomes; the most rigorous synthesis on stroke executive function judged the evidence not sufficient to either confirm or rule out cognitive rehabilitation's effectiveness1.
The newest, most app-favorable findings come from the weakest designs. The 2025 post-surgery study had 20 patients, no control group and 8 weeks of follow-up; its platform comparison (4.1 vs 2.3 MoCA points) reflected self-selected, more-active patients, not randomized assignment4. The 2023 functional result is a single case with no control beyond the patient's own baseline3. Practice effects, spontaneous neurological recovery in the months after injury, and the bundling of multiple therapy components all remain plausible alternative explanations.
Durability and transfer are largely untested. Whether gains persist after training stops, and whether battery improvements translate into better everyday function, were rarely measured in the small trials1. Larger, longer, independently controlled randomized trials, ideally comparing self-administered programs against guideline-based therapist care, are what the field still needs.
What this means in practice
For someone recovering from a stroke, a traumatic brain injury or brain surgery, the takeaways are modest and evidence-bounded. Cognitive rehabilitation as a discipline has guideline-level support for attention, memory and executive deficits after brain injury2; computerized programs have shown encouraging but preliminary signals in small recovery cohorts4; and the controlled evidence for self-administered apps specifically remains incomplete1. Any program should be chosen with a clinician, matched to the person's actual deficits, and treated as a complement to, not a substitute for, guideline-based rehabilitation.
For the broader picture of cognitive decline and how these conditions intersect it, see the pillar guide: cognitive decline: an evidence-based guide. Related topic reviews cover long-COVID brain fog, cognition in multiple sclerosis, mobility, driving and cognition, healthy older adults, sleep and insomnia, and stress and memory. The underlying studies are broken down on the executive-function rehabilitation case study and the post-operative cognitive recovery study pages.
References
- Cognitive rehabilitation for executive dysfunction in adults with stroke or other adult non-progressive acquired brain damage. Cochrane Database of Systematic Reviews 2013;(4):CD008391. https://doi.org/10.1002/14651858.CD008391.pub2 Systematic review of RCTs: evidence not sufficient to confirm or rule out cognitive rehabilitation for executive dysfunction after stroke / acquired brain injury.
- Evidence-Based Cognitive Rehabilitation: Updated Review of the Literature From 2003 Through 2008. Archives of Physical Medicine and Rehabilitation 2011;92(4):519–530. https://doi.org/10.1016/j.apmr.2010.11.015 ACRM evidence review; graded Practice Standards / Guidelines for attention, memory-strategy and executive-function rehabilitation after TBI and stroke.
- Efecto de un Programa de Rehabilitación Neuropsicológica del Funcionamiento Ejecutivo en Un Caso con Trastorno Neurocognitivo Leve debido a SAHOS. Revista Iberoamericana de Psicología 2023;16(1):53–65. https://doi.org/10.33881/2027-1786.rip.16105 Single-case (N=1) A-B design, computerized platform within a Sohlberg-Mateer model; BRIEF-A overlap metrics 80–90% effective; anxiety unchanged. Full study breakdown
- The Impact of Neuropsychological Rehabilitation on Postoperative Cognitive Impairments. Modern Science and Research 2025;4(5):332–336. https://doi.org/10.5281/zenodo.15367740 Single-group study, 20 brain-surgery patients; 8-week program raised MoCA 21.8→25.9 (p<0.001); active platform users +4.1 vs +2.3 MoCA points. Full study breakdown
- Cognitive deficits in adult patients with brain tumours. The Lancet Neurology 2004;3(3):159–168. https://doi.org/10.1016/s1474-4422(04)00680-5 Review establishing that memory, attention and executive deficits are frequent in adults with brain tumors, especially frontal/temporal/parietal lesions.