Matthew J. Fogarty
The goal of this opinion is to convince readers of the importance of nomenclature and natural history in understanding the physiology of the ageing neuromotor system. We may be a century removed from the golden age of description, where the classification by general clinical signs (physiology) and postmortem pathology defined various diseases and ailments. However one glaring omission to the present day is a satisfactory nomenclature for age-associated decline in neuromotor function, which may be both distinct and inclusive of striated muscle atrophy and weakness (i.e. sarcopenia), itself a term coined only in the last couple of decades. To aid in this endeavour we posit the term ‘mild motor impairment’, which we define as a disordered timing (when to recruit motor units, i.e. activation of motor units across different motor pools) and/or activation (motor unit type and discharge duration) of motor units, contributing to deficits in a desired motor behaviour. Like mild cognitive impairment (MCI) (Langa & Levine, 2014), ‘mild motor impairment’ may be a harbinger of more serious conditions, such as sarcopenia/frailty, that contribute to morbidity and mortality. Here our goal is threefold: (i) to convince the reader of the importance of age-related ‘mild motor impairment’, which is to motor neurons (MNs) and their synaptic inputs what MCI is to hippocampal neurons and their synapses; (ii) that a neurogenic critical framework of movement decline with age is consistent with ‘mild motor impairment’ as a concept and moreover accounts for ageing deficits unrelated to joint issues or muscle weakness (e.g. dysphagia); and (iii) some testable hypotheses that lend weight to the prior points and underscore ‘mild motor impairment’ as a neurogenic manifestation of ageing. We briefly touch on the broader clinical implications for the colloquial classification of motor impairments as ‘mild motor impairment’. Dementia, including the most common form, Alzheimer's disease (AD), is primarily characterised by deficits in memory, cognitive and executive functions (Knopman et al., 2021). It has been apparent for decades that age-associated dementia and AD are associated with extensive prodromal phases (Buchman & Bennett, 2011). The pre- or early-onset disturbances exist in a continuum with many other comorbidities, such as obstructive sleep apnoea (OSA), dysphagia, sarcopenia, decreased cardiorespiratory fitness, obesity and diabetes (Andrade et al., 2018; Humbert et al., 2010; Knopman et al., 2021). Despite this breadth of physiological disturbance in ageing, the pathophysiology of dementia and AD centres on the progressive loss of neuronal synapses, which leads to subtle changes that later precipitate full-blown neuronal death and the demented brain state (Knopman et al., 2021; Selkoe, 2002). MCI may be an intermediate phase between normal ageing and full-blown dementia, and thus a symptomatic manifestation of otherwise prodromal disease (Langa & Levine, 2014). Notably in dementia and AD the correctness of executive function decision making (i.e. the appropriateness of a particular behaviour) is impaired, not necessarily the actioned behaviour (Knopman et al., 2021). In the elderly declining motor performance is readily apparent in the absence of a specific disease condition, with pathophysiological similarities in the ageing neuromotor system to that of the MCI brain. For example subtle motor behavioural dysfunctions arise contemporaneously with altered synaptic inputs onto MNs from late middle age in rodent models (Fogarty, 2025). It takes some time before altered motor functions progress to the stage where behavioural failure due to weakness (contemporaneous with MN death and striated muscle atrophy) ensues. In humans these motor dysfunctions commonly present with a wide array of comorbidities (similar to MCI, dementia and AD), including decreased cardiorespiratory fitness, obesity, diabetes, arthritis and other joint dysfunctions (Larsson et al., 2019). However most gross measurements of motor behaviour are entirely devoted to outcomes such as strength or power measures (Hassan et al., 2021; Larsson et al., 2019). These outcomes hardly capture the full gamut of human or animal neuromotor function and are instead a proxy for motor unit recruitment and/or muscle strength. Importantly estimates of grip and upper limb strength do not encompass the major morbidity and mortality factors of ageing, upper airway infection (aeromotor dysfunction) or falls (torso, lower limb and proprioception dysfunction). Instead it could be posited that correct neuromotor behaviour relies on two factors: (i) the appropriate level of motor unit recruitment within a motor pool to reach the necessary level of force/torque and (ii) the correct orchestration of disparate motor pools to adequately perform the timing and movement patterning of a particular motor skill (co-ordination of motor unit activation across motor pools). It is inarguable that a substantial part of the former and almost the entirety of the latter (sans a portion constrained by joint flexibility) are undergirded by neural timing and activation. Failure of the former point is likely to constitute some degree of progression from ‘mild motor impairment’ to sarcopenia or frailty. Despite the importance of the neural control of movement for successful motor behaviours, the striated muscle has received an almost reverential level of attention, particularly with regard to the mechanisms of sarcopenia, which is defendable (Larsson et al., 2019), and ageing movement preservation/enhancement, which, based on translational results, has been fultile (Tsai, 2024). Although many groups worldwide are focused on ageing effects on motor units, it is often through a paradigm of motor unit number estimation (MUNE), motor unit recruitment or motor unit facilitation (delta F; Hassan et al., 2021). Each of these deals with EMG evaluations of motor units in a manner that relates to the total number of MNs (MUNE), the recruitment of the motor pool (more MNs recruited leading to more force) and the duration and frequency of motor unit firing (Hassan et al., 2021; Orssatto et al., 2023). Considering this intense attention on motor units, it may be slightly unreasonable to say that there is a lack of insight into the MN aspects of ageing. However these motor unit approaches are confounded by potential age-related neuromuscular junction (NMJ) changes and are mainly performed under static isometric (i.e. non-movement) conditions, without weight bearing, and each centre on strength, ignoring co-ordination entirely. If it were possible to perform these motor unit recordings in the context of short physical performance battery (SBBP) and timed get up and go (TUG) testing, then we would at least be evaluating motor unit contributions to ‘real-world’ scenarios. The concept of ‘mild motor impairment’ allows us a platform to view the MN as less of a discharge device and more of an important player in a complex circuit (i.e. giving credence to MN inputs and outputs). This importance of MN inputs and outputs results in the neuroenteric aspects of age-related motor dysfunction. Here we consider the neuromotor system to comprise the motor planning centres in the motor cortex, cerebellum and brainstem, the MNs, the NMJs and the striated muscle. The neurocentrism of impaired ageing motor function may be attributed to MN loss, motor cortical (programme), pattern (central pattern generator) or feedback (cerebellar) impairment and/or NMJ deficits. Pathophysiological similarities exist between dementia and the ageing neuromotor system. Early pathophysiological changes in MCI and dementia include neuronal hyperexcitability and synapse loss (Knopman et al., 2021; Scheff et al., 2006; Selkoe, 2002), and deficits in aerodigestive behaviours (Andrade et al., 2018; Humbert et al., 2010; Langa & Levine, 2014). Age-related motor cortex hyperexcitability (Viteri et al., 2025), alongside synapse loss and aerodigestive deficits (Fogarty, 2025), has been found from late middle age in ageing models. Later in more symptomatic AD neuronal death is a major contributor to symptoms, with significant loss of hippocampus and some cerebral neurons, which contributes to the major dementia behavioural phenotypes (Knopman et al., 2021). This is largely triggered by inflammation (in response to tau and plaques, etc.) associated with mitochondrial degeneration (mitochondrial fragmentation and mitophagy), leading to necroptosis (Cai & Tammineni, 2017; Knopman et al., 2021). Similarly with ageing MN death occurs at older ages (see later) when sarcopenia is evident. Despite the links between age-associated MN death and mitochondrial dysfunctions being relatively unexplored, it is highly implicated and highly characterised in the pathogenesis of abnormal striated muscle function with age (Larsson et al., 2019). Although the evidence for MN death (and axonal degeneration) is abundant in human and animal ageing (Fogarty, 2025; Fogarty & Sieck, 2023; Hashizume et al., 1988; Kawamura et al., 1977; Larsson et al., 2019; Tomlinson & Irving, 1977), the timeline of demise remains obscure. Modest evidence that exists suggests that spinal cord MN death in humans is progressive from middle age (Kawamura et al., 1977; Tomlinson & Irving, 1977), although in rodents, MN death (at least in the brainstem) is rapid from late middle age to old age (Fogarty, 2025). More incisively, aerodigestive motor dysfunctions, where activation and timing impairments, rather than weakness per se, are the major pathophysiological causes include OSA, dysphagia and dysphonia, all significant harbingers of both frailty and neurodegenerative diseases (Andrade et al., 2018; Fogarty, 2025; Humbert et al., 2010). Indeed brainstem MNs are under a plethora of diverse influences, from anatomical pathways to neurotransmitters, and undergird the life-sustaining behaviours of airway patency, airway defence and swallowing (Frazure et al., 2025; Pitts & Iceman, 2023). Strikingly these brainstem MN pools and target muscle may be exceptions to the available evidence, suggesting that fatigable units (larger MNs; Fogarty & Sieck, 2023; Hashizume et al., 1988) innervating type IIx (found in humans) and/or IIb muscle fibres (Grosicki et al., 2022; Horwath et al., 2025) are the most sensitive to ageing. Human tongue striated muscle mainly comprises slow and fast fatigue-resistant motor units, yet human tongue weakness in old age is prevalent (Crow & Ship, 1996) and increases the risk of many motor dysfunctions with age, including the aforementioned dysphagia, dysphonia and impaired airway defence (Andrade et al., 2018; Humbert et al., 2010). The interplay among diversity of brainstem MN inputs, motor unit type and selective vulnerability to neurodegeneration is of intense interest in ageing, ALS and symptomatic treatment of late-stage dementia. In different rat models of ageing, brainstem neuromotor dysfunctions largely parallel those found in ageing humans. We have shown that the capacity of MN circuit integration in swallow timing is greatly reduced at late middle age, prior to overt swallow pressure deficits and MN loss in old age (Fogarty, 2025). Ongoing work suggests this pattern is not exclusive to brainstem-innervated striated muscle, nor to aerodigestive or respiratory motor pools. Although aerodigestive behaviours occur largely in the absence of joint movement, nimbleness and ageing of the skeleton itself may affect the interpretation of limb behaviours. We hypothesise that early intervention (middle age) in arresting the prodromal synapse loss that characterises ‘mild motor impairment’ will improve the early timing of behavioural abnormalities and degenerative MN phenotype observed at late middle age. We hypothesise that later intervention (late middle age) will not fully ameliorate the prodromal synapse loss that characterises ‘mild motor impairment’ but will reduce and/or delay MN death and muscle weakness. If one accepts ‘mild motor impairment’ as a contributor to ageing neuromotor dysfunction and sarcopenia, a plethora of potential interventions exists, namely ALS and AD therapeutics. In experimental contexts ALS drugs seem highly amenable to abrogating MN loss. The major problem with ALS treatment is not the efficacy of individual pharmaceutical compounds against a particular pathogenic mechanism but rather the inability to get therapies in the hands of a patient at a time in the natural history of the disease where there is still a process to mitigate, rather than a fait accompli. In addition it is very difficult to predict individuals who will go on to develop ALS. However ageing is not under the same time constraint, with the ageing neurodegenerative process (‘mild motor impairment’) likely spanning decades. Additionally we have identifiable populations at risk of ‘mild motor impairment’, including patients earmarked for bedrest, chemotherapy candidates or OSA sufferers who may benefit from prophylactic intervention. To this end the utility of ‘failed’ ALS therapeutics on ageing-related MN degenerations may be an untapped area of promise. One such ‘failed’ ALS therapy, riluzole, has been shown to reduce synapse loss in an AD model and in cerebral neurons of ageing rats (Hunsberger et al., 2015; Pereira et al., 2014), proving as a proof of concept to test whether we can delay the onset of ‘mild motor impairment’ (hypothesis 1) or the conversion of ‘mild motor impairment’ to overt motor dysfunction and sarcopenia/frailty. By contrast a different problem exists with the preclinical development of AD therapies. Perhaps one of the reasons AD remains therapeutically intractable is the aggressiveness of the mutant models used, with all but one or two exhibiting a significant demented phenotype at ages that are too young to be considered ‘young’ in an ageing study. Thus the field seems to be fastidiously working to cure a specific genetic underpinning rather than the more generalised condition. Particularly an AD mutant model on the F344 background has been developed, with the benefit of exhibiting AD symptoms at an age approaching the clinically relevant late middle age (Cohen et al., 2013), although adoption of its use has been reluctant. Regardless as the major outcome measure for many preclinical AD therapeutics is the maintenance of dendritic spines, compounds such as SPG302 (Trujillo-Estrada et al., 2021), which maintain these in AD models, may be ideally suited for use in testing hypotheses 1 and 2. The rationale for testing ‘mild motor impairment’ is straightforward: if we allay the MN pre-death degenerative phenotype and preserve MNs, we expect denervation-induced weakness to be mitigated. Even if sarcopenia is not ameliorated, we may still improve the second facet of effective neuromotor behaviours, namely the timing and activations of surviving MNs. Thus even modest success would still be impactful. Obviously in the clinical space muscle evaluations may stratify those at risk of frailty who may benefit most from this type of pharmaceutical intervention. For the concept of ‘mild motor impairment’ to really gain traction, we need to disabuse ourselves of the notion that force and power are all the neuromotor system evaluations one needs. Indeed with the advent of the electric start lawnmower and the relative paucity of careers requiring precarious fingertip purchase on ledges, hand grip strength seems a particularly asinine evaluation compared to, for example, penmanship. Indeed grip strength and penmanship are unrelated to the major morbidity and mortality factors in old age, aspiration pneumonia and falls. For behaviours where force generation is not the limiting factor, for example brainstem aerodigestive activities such as speech and swallow, subtle differences may allow us to stratify patients and/or evaluate interventions in a highly sensitive manner. Through the lens of ‘mild motor impairment’, the gerontological field may come to embrace a fuller picture of motor unit timing, motor unit activation and force when considering motor dysfunction, sarcopenia and frailty. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article. None of the authors has any conflicts of interest, real or perceived, to disclose. M.F.: conception or design of the work; drafting the work or revising it critically for important intellectual content; final approval of the version to be published; agreement to be accountable for all aspects of the work. This work was supported by R01 AG086136 (M.J.F.) from the NIH.