Between one third and 63% of patients with Parkinson’s Disease (PD) suffer from Freezing of Gait (FOG), which results in falls, decreased independence and reduced quality of life (QoL) (Rutz and Benninger, 2020; Silva-Batista et al., 2020). FOG is characterized by an abrupt, erratic gait interruption (Peterson et al., 2014; Pozzi et al., 2019) despite the person intending to move forward (Nutt et al., 2011). This can result in hesitation starting a movement, stopping during turns (Plotnik and Hausdorff, 2008) and trembling in place (Moore et al., 2008; Nonnekes et al., 2019). While medication with levodopa has been shown to be partially effective, FOG is still sporadically observed when patients are in the clinical “ON”-medication state (Schaafsma et al., 2003a). FOG episode (Lees, 1989; Pozzi et al., 2019; Silva-Batista et al., 2020) severity correlates with disease progression (Paul et al., 2018). People with PD are considered to be in an ON-medication state when levodopa medication is effective in alleviating motor symptoms and in the OFF-medication state when there is not enough levodopa remaining in the brain to relieve motor symptoms sufficiently (Lees, 1989). Falls frequency is associated with FOG both in OFF-medication and ON-medication states but far greater in the OFF-medication state (Schaafsma et al., 2003b). Other non-invasive treatments which alleviate FOG severity include exercise rehabilitation programs (Cucca et al., 2016; Silva-Batista et al., 2020) but rehabilitation gains are not often retained (Lees, 1989; Gilat et al., 2021). If both rehabilitation and medication are not successfully alleviating FOG severity, then further treatment options should be explored.

Because FOG can occur during ON-medication states (Lewis and Barker, 2009), it may not be solely attributed to depleted dopamine (Schaafsma et al., 2003a; Lewis and Shine, 2016). Current theories as to the cause of FOG include:

An earlier theory proposes a disruption of supraspinal controls to the central pattern generators (CPGs) (Plotnik and Hausdorff, 2008; Marquez et al., 2020). These causes may not be exclusive of each other.

Freezing of Gait is experienced by individuals with PD at environmental boundaries such as doorways, or crossing roads, when in tight spaces, and when facing an obstacle such as furniture or trip hazards such as uneven flooring (Ramos et al., 2020). However, people with PD report fewer FOG episodes when navigating obstacles such as stairways where specific attentional strategies are used, e.g., subjects may be able to focus on keeping each step even following the step treads (Rutz and Benninger, 2020) or when using a pedestrian cross walk following evenly spaced lines (Ramos et al., 2020). Could a person who experiences FOG learn and utilize a similar strategy to regulate their gait without freezing at an environmental obstacle or boundary?

In healthy individuals, motor control is not just regulated by CPGs and the brainstem but refined with cortical commands (Pozzi et al., 2019). The neuronal circuits that comprise CPGs which result in rhythmic movements such as walking are always active, but modulated by higher brain centers (Marder and Bucher, 2001; Behrendt et al., 2013, 2014). The cerebral cortex modulates rhythms that originate in spinal networks. In the PD brain, the depletion of dopamine may result in not enough to act as a neurotransmitter. Competition is created between complementary neural circuits, both inhibitory and excitatory (Lewis and Barker, 2009) which leads to FOG.

Rehabilitation exercises developed specifically to improve FOG symptoms show benefits (Gilat et al., 2021), but as those benefits are not retrained, would need to be continuously practiced to preserve the improvement. A safe, but effective adjunct therapy could be utilized to facilitate this practice independently, without the need for additional support by staff (Heremans et al., 2012). In this article the authors will propose why Motor Imagery (MI) could support standard rehabilitation practice to prevent FOG at environmental boundaries.

Motor imagery is the mental rehearsal of an action without its actual physical performance. It can be “performed” in the first person or “viewed” as a third person. MI can be visual in nature or kinesthetic (Saimpont et al., 2013). MI has been used successfully in healthy adults, especially in sports, for some time (Saimpont et al., 2013). One of the draw backs is it can be mentally demanding (Abbruzzese et al., 2015) but it can be performed seated as a visualization only without safety risks, and, once the skill is learnt, can be practiced independently without a therapist present (Heremans et al., 2012). This would not replace other rehabilitation practices but adjunct them (Slimani et al., 2016; Da Nascimento et al., 2019). MI is now being used in training for healthy older adults as an adjunct to physiotherapy. Studies are conflicting, but research currently supports the idea that MI can improve the outcomes of therapeutic interventions in older populations (Saimpont et al., 2013). It is believed to preserve or stimulate the forward planning neural pathways (Ashley Fox, 2013). MI shows the most benefit in healthy elderly adults when in the third person, supported by auditory cues and kinesthetic in nature (Saimpont et al., 2013). Importantly, the pathways used in MI are partially the same as those used in motor execution (Snijders et al., 2011), specifically the SMA, premotor cortex, primary motor cortex, posterior parietal regions (e.g., the inferior and superior parietal lobes), the BG and cerebellum (Moran and O’Shea, 2020).

Motor imagery alone can cause a significant improvement in muscle force, as well as increasing the motor-activity related cortical potential of healthy older adults (Jiang et al., 2016). Mouthon et al. (2015) showed that during MI, cortico-spinal excitability is increased in healthy young adults during mental training of balance tasks, as evidenced by motor evoked potential facilitation. In healthy subjects, brain plasticity can be measured after MI, and motor performance is improved (Debarnot et al., 2014).

Traditional gait and balance therapy is usually delivered in an explicit teaching method. Explicit learning is a complex learning method requiring a high cognitive load, using working memory. As people age, and with the impact of disease, the ability to learn through explicit methods deteriorates. This is especially true in dual-task situations (Abbruzzese et al., 2015). People are expected to integrate and memorise instructions in traditional motor learning (Rutz and Benninger, 2020). This style of explicit learning is not retained in people with brain lesions in the motor pathways such as PD (Rutz and Benninger, 2020).

An important consideration for any clinician wishing to consider a patient for this therapy is their cognitive decline. People with PD may exhibit cognitive decline which varies with the stage and time of onset of the disease (Ding et al., 2015). Salient to gait rehabilitation are losses in executive function, short-term working memory and visual-spatial memory. Early cognitive impairment is difficult to diagnose. There is no clear pattern to the decline and the decline can only be assessed on an individual basis and is unique to each person (Ding et al., 2015). The gross progression of the pathology of PD will eventually include atrophy of several brain regions (Ding et al., 2015), some of which we can surmise could affect the person’s ability to engage in MI and overall gait rehabilitation.

The advantage of MI is that it does not require learning (Snijders et al., 2011), it uses working memory efficiently, and allows for greater training endurance (Debarnot et al., 2014; Moran and O’Shea, 2020). Gait re-education and the creation of new attentional strategies through means such as MI should allow for greater retention of motor learning in people with PD (Mirelman et al., 2013). MI can be mentally fatiguing (Abbruzzese et al., 2015) and the authors suggest that using MI as an intervention (Podda et al., 2020) should be considered on a case-by-case basis, and that an appropriate dose is considered for that individual. It may not always be an appropriate intervention with this condition.

There are however, a number of questions that need to be carefully considered before pursuing MI as a possible treatment modality for FOG. These are:

Subjects with FOG experience a worsened QoL and risk falls through the sudden interruption of gait. Although much has been discovered about the underlying causes, no treatment to date has been completely successful in eliminating FOG. FOG often occurs at specific moments or places, such as environmental boundaries e.g., a person with FOG may get “stuck” at the threshold of a doorway. FOG rehabilitation exercises show benefits, but those benefits are not retained without ongoing intervention. What if the person with FOG could independently continue those exercises in absolute safety inside their own minds, thus continuing to benefit from improved mobility without a therapist present–MI?

Because studies of MI in FOG are not homologous in design, it is difficult to compare studies to have a clear understanding if MI in PD is an effective treatment to reduce FOG and how best to apply it. The drawbacks are that MI may cause fatigue and that it has been difficult to assess in PD subjects. For the therapist some treatment factors are unknown, such as an effective dose. An individual may not be able to fully benefit due to cognitive decline. It also requires practice on the part of the therapist to teach the technique consistently. The proposed benefits are that MI of an action uses many of the same neural pathways as the physical movement, has the potential to produce motor neuron plasticity, is an efficient way to learn and has shown substantial treatment benefits in other populations. It should not be disregarded without complete study and a larger body of comparable and well controlled research should be carried out to establish if MI can mitigate FOG at environmental boundaries.

The original contributions presented in the study are included in the article/supplementary material, further inquiries can be directed to the corresponding author/s.

DH had the idea for the article with support from MB and drafted the article with inputs from HW and MB. DH and HW performed the literature search. All authors critically revised the work.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.