What this research is about
A stroke damages part of the brain that used to send instructions to the body. Recovery happens because the brain reorganises itself — surviving regions gradually take on some of the work the damaged area used to do. Most stroke rehabilitation aims to encourage that reorganisation, mainly through practice: doing the movement again and again until the brain rebuilds the pathway (Nudo, 2006).
There is an obvious problem with this. In the weeks after a stroke, the limb that most needs practice is often the one that can barely move. You cannot practise reaching for a cup if you cannot lift your hand off your lap — the therapy that would help most is the hardest to actually do.
The studies listed on this page all circle around one question: can you get some of the benefit of practice by imagining the movement instead of performing it?
Why imagining a movement is not the same as doing nothing
The starting point for this research was a surprising finding from brain scanning. When a person lies still in a scanner and vividly imagines opening and closing their hand, many of the same brain areas become active as when they genuinely move it (Lotze et al., 1999). The imagined movement is not a picture the brain is looking at — it is a rehearsal the brain is running, using much of the same machinery it would use for the real thing (Jeannerod, 2001).
Something similar happens when we watch other people move: certain brain cells fire both when you perform an action and when you see someone else perform it (Rizzolatti & Craighero, 2004; Decety & Grèzes, 1999). This is part of the reasoning behind mirror therapy, where a mirror placed over the affected hand gives the brain the impression the weak side is moving normally (Altschuler et al., 1999).
Because imagined movement reaches the motor system without needing the limb to cooperate, one review described it as a possible backdoor into the motor system after stroke (Sharma et al., 2006). There is also direct evidence that this can change the brain: people who improved after a course of mental practice showed measurable shifts in the activity of their motor cortex on scanning afterwards (Page et al., 2009).
What the studies found
The early clinical studies were small and encouraging. Adding mental practice to ordinary therapy improved arm and hand function years after stroke (Page et al., 2007) and appeared feasible in the first weeks too (Page et al., 2001). Rehearsing everyday activities mentally — folding laundry, making a drink — helped people carry them out more independently, including tasks they had not specifically practised (Liu et al., 2004). Smaller studies suggested possible benefit for walking (Malouin & Richards, 2010), hand function in long-term stroke (Dijkerman et al., 2004), and neglect, where a person loses awareness of one side of space (Smania et al., 1997).
Then came the largest trial. A Scottish study followed 121 people with arm weakness within six months of stroke, comparing four weeks of supervised motor imagery against a matched non-movement rehearsal programme and against normal care, with the arm assessed by someone blind to group. There was no difference between the three groups on any measure (Ietswaart et al., 2011) — the single most important result on this page, and a negative one.
Where does that leave things? The most reliable summary is a Cochrane review pooling 25 trials and 676 people (Barclay et al., 2020): moderate-certainty evidence that mental practice added to other rehabilitation improves arm activity and movement more than that rehabilitation alone. Only low-certainty evidence that it improves everyday activities, and used instead of conventional therapy it did not clearly help. No study reported any harm.
A separate meta-analysis of 32 trials found benefits for balance, walking and arm function overall — but when the analysis was restricted to the highest-quality studies, those differences disappeared (Guerra et al., 2017). Earlier reviews had reached the same cautious verdict: promising, insufficiently proven (Braun et al., 2006; Zimmermann-Schlatter et al., 2008).
Mirror therapy has fared somewhat better. A Cochrane review of 62 studies and 1,982 people found moderate-quality evidence that it improves movement and everyday activities, at least when used alongside conventional rehabilitation (Thieme et al., 2018).
For language, the evidence is stronger and older. Speech and language therapy improves functional communication, reading, writing and speaking compared with no therapy at all, across 57 trials and 3,002 people with aphasia (Brady et al., 2016). More intensive therapy appeared to work better — but more people dropped out of the most intensive schedules, which matters when deciding what is realistic. Practice delivered at home over a screen has been trialled as one way of increasing the amount without increasing the burden (Bilda et al., 2015).
What the research does not prove
Four things should be said plainly.
First, mental practice is an addition to rehabilitation, not a replacement for it — the evidence applies to people also doing physical therapy, and there is no good evidence it works alone instead of conventional treatment (Barclay et al., 2020).
Second, the largest and most rigorous single trial found no benefit at all (Ietswaart et al., 2011), and across the literature the better-designed a study is, the smaller its reported effect (Guerra et al., 2017) — usually a sign the true effect is more modest than early enthusiasm suggested.
Third, nobody knows the right dose or the right timing. How many minutes, how many days a week, how soon after the stroke — the trials could not answer this (Barclay et al., 2020). Separate research offers theory, not proof: the body cycles between alertness and rest roughly every ninety minutes (Kleitman, 1963; Rossi & Nimmons, 1991), the drowsy state just before sleep is linked to unusually vivid, spontaneous imagery (Mavromatis, 1987; Schacter, 1976; Ghibellini & Meier, 2023), and sleep itself helps consolidate a physical skill learned earlier in the day (Walker et al., 2002; Stickgold & Walker, 2007). None of this has been tested with imagery practice after stroke — it explains why a session might be timed a certain way, not that doing so is proven to help.
Fourth, not everyone finds imagery equally easy, and the ability to generate a vivid mental image of movement can be affected by the stroke itself (Wu et al., 2013). A practice that depends on imagination will suit some people much better than others.
What it means
The honest summary is that mental practice is a reasonable thing to do alongside your rehabilitation, and an unreasonable thing to rely on instead of it. The likely benefit is real but modest, and the evidence is not yet settled.
What makes it worth considering anyway is the other side of the ledger: across every study reviewed, no harm was reported. It costs nothing, and needs no equipment, therapist, or working limb. It can be done at three in the morning when you cannot sleep, or in a hospital bed when nothing else is possible — and for people told there is nothing more to be done, having something to do is not a small thing, even before you count its effect on the arm.
This overview is a summary of published research, not medical advice. Decisions about your rehabilitation should be made with your own clinical team, who know your situation.
A
- (1999). Rehabilitation of hemiparesis after stroke with a mirror. The Lancet, 353(9169), 2035–2036.
B
- (2020). Mental practice for treating upper extremity deficits in individuals with hemiparesis after stroke. Cochrane Database of Systematic Reviews, 2020(5), CD005950.
- (2015). Intensive telerehabilitation with speech-language therapists versus an intensive telerehabilitation home training programme for aphasia: A randomised controlled trial. Journal of Telemedicine and Telecare, 21(1), 39–47.
- (2016). Speech and language therapy for aphasia following stroke. Cochrane Database of Systematic Reviews, 2016(6), CD000425.
- (2006). The effects of mental practice in stroke rehabilitation: A systematic review. Archives of Physical Medicine and Rehabilitation, 87(6), 842–852.
D
- (1999). Neural mechanisms subserving the perception of human actions. Trends in Cognitive Sciences, 3(5), 172–178.
- (2004). Does motor imagery training improve hand function in chronic stroke patients? A pilot study. Clinical Rehabilitation, 18(5), 538–549.
G
- (2023). The hypnagogic state: A brief update. Journal of Sleep Research, 32(1), e13719.
- (2017). Motor imagery training after stroke: A systematic review and meta-analysis of randomized controlled trials. Journal of Neurologic Physical Therapy, 41(4), 205–214.
I
- (2011). Mental practice with motor imagery in stroke recovery: Randomized controlled trial of efficacy. Brain, 134(5), 1373–1386.
J
- (2001). Neural simulation of action: A unifying mechanism for motor cognition. NeuroImage, 14(1), S103–S109.
K
- (1963). Sleep and wakefulness (Revised ed.). University of Chicago Press.
L
- (2004). Mental imagery for promoting relearning for people after stroke: A randomized controlled trial. Archives of Physical Medicine and Rehabilitation, 85(9), 1403–1408.
- (1999). Activation of cortical and cerebellar motor areas during executed and imagined hand movements: An fMRI study. Journal of Cognitive Neuroscience, 11(5), 491–501.
M
- (2010). Mental practice for relearning locomotor skills. Physical Therapy, 90(2), 240–251.
- (1987). Hypnagogia: The unique state of consciousness between wakefulness and sleep. Routledge & Kegan Paul.
N
- (2006). Mechanisms for recovery of motor function following cortical damage. Current Opinion in Neurobiology, 16(6), 638–644.
P
- (2001). A randomized efficacy and feasibility study of imagery in acute stroke. Clinical Rehabilitation, 15(3), 233–240.
- (2009). Cortical plasticity following motor skill learning during mental practice in stroke. Neurorehabilitation and Neural Repair, 23(4), 382–388.
- (2007). Mental practice in chronic stroke: Results of a randomized, placebo-controlled trial. Stroke, 38(4), 1293–1297.
R
- (2004). The mirror-neuron system. Annual Review of Neuroscience, 27, 169–192.
- (1991). The wave nature of consciousness: A new direction for the evolution of psychotherapy. American Journal of Clinical Hypnosis, 32(3), 178–206.
- (1991). The 20-minute break: Reduce stress, maximize performance, and improve health and emotional well-being using the new science of ultradian rhythms. Jeremy P. Tarcher.
S
- (1976). The hypnagogic state: A critical review of the literature. Psychological Bulletin, 83(3), 452–481.
- (2006). Motor imagery: A backdoor to the motor system after stroke? Stroke, 37(7), 1941–1952.
- (1997). Visuomotor imagery and rehabilitation of neglect. Archives of Physical Medicine and Rehabilitation, 78(4), 430–436.
- (2007). Sleep-dependent memory consolidation and reconsolidation. Sleep Medicine, 8(4), 331–343.
T
- (2018). Mirror therapy for improving motor function after stroke. Cochrane Database of Systematic Reviews, 2018(7), CD008449.
W
- (2002). Practice with sleep makes perfect: Sleep-dependent motor skill learning. Neuron, 35(1), 205–211.
- (2003). Sleep and the time course of motor skill learning. Learning & Memory, 10(4), 275–284.
- (2013). Assessing the effect of cognitive ability on upper-limb motor imagery in persons after stroke using the kinesthetic and visual imagery questionnaire. American Journal of Occupational Therapy, 67(3), 318–324.
Z
- (2008). Efficacy of motor imagery in post-stroke rehabilitation: A systematic review. Journal of NeuroEngineering and Rehabilitation, 5(1), 8.