The role of the cerebral cortex in postural responses to externally induced perturbations

Neuroscience and Biobehavioral Reviews 57 (2015) 142–155

Contents lists available at ScienceDirect

Neuroscience and Biobehavioral Reviews journal homepage: www.elsevier.com/locate/neubiorev

Review

The role of the cerebral cortex in postural responses to externally induced perturbations D.A.E. Bolton ∗ School of Psychology, Queen’s University Belfast, Belfast BT7 1NN, Northern Ireland, UK

a r t i c l e

i n f o

Article history: Received 25 April 2015 Received in revised form 9 August 2015 Accepted 25 August 2015 Available online 29 August 2015 Keywords: Cerebral cortex Balance Postural response Central set Compensatory balance response

a b s t r a c t The ease with which we avoid falling down belies a highly sophisticated and distributed neural network for controlling reactions to maintain upright balance. Although historically these reactions were considered within the sub cortical domain, mounting evidence reveals a distributed network for postural control including a potentially important role for the cerebral cortex. Support for this cortical role comes from direct measurement associated with moments of induced instability as well as indirect links between cognitive task performance and balance recovery. The cerebral cortex appears to be directly involved in the control of rapid balance reactions but also setting the central nervous system in advance to optimize balance recovery reactions even when a future threat to stability is unexpected. In this review the growing body of evidence that now firmly supports a cortical role in the postural responses to externally induced perturbations is presented. Moreover, an updated framework is advanced to help understand how cortical contributions may influence our resistance to falls and on what timescale. The implications for future studies into the neural control of balance are discussed. © 2015 Elsevier Ltd. All rights reserved.

Contents 1. 2. 3. 4.

5. 6.

7.

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143 What is a compensatory reaction? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143 Indirect support for a cortical role in postural responses to externally induced perturbations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 144 Direct measures of cortical neurophysiology related to externally induced postural responses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145 4.1. Functional near-infrared spectroscopy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145 4.2. Electroencephalography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145 4.2.1. Perturbation-evoked cortical responses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 146 4.2.2. Pre-perturbation cortical activity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147 4.3. Transcranial magnetic stimulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147 Evidence for accelerated speed of processing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149 5.1. Timing of cortical influence on the postural response . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151 Potential roles for the cerebral cortex in balance: Future directions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151 6.1. Building reference frames and putting the world into motor terms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151 6.2. Cortical roles in predicting instability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152 Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153

∗ Tel.: +44 0 28 9097 5476; fax: +44 0 28 9097 5486. E-mail address: [email protected] http://dx.doi.org/10.1016/j.neubiorev.2015.08.014 0149-7634/© 2015 Elsevier Ltd. All rights reserved.

D.A.E. Bolton / Neuroscience and Biobehavioral Reviews 57 (2015) 142–155

1. Introduction A broadly distributed neural network controls upright stability in humans. While the relative contribution of distinct parts of the nervous system to maintain balance remains unclear it is now well established that even the most advanced regions of the neural hierarchy play some role in balance control. Most remarkable is the accumulating evidence that the cerebral cortex plays an important role in balance including compensatory reactions to unexpected postural challenge. This represents an important departure from the historical framework that placed postural regulation largely in the domain of subcortical networks. A seemingly effortless ability to stay upright in healthy humans belies the sophisticated mechanisms acting to preserve an elevated centre of mass over a small bipedal base of support. However in many disease states, particularly those involving neurological disruption, the challenge of this task is exposed rendering numerous clinical populations vulnerable to falls. With significant societal and individual costs related to falls (e.g. severe injury and even death) this represents a major health care concern (Baker et al., 2011; Carroll et al., 2005; Kannus et al., 2005). Thus illuminating the underlying neural mechanisms for controlling balance is essential for developing targeted therapies to mitigate fall risk. As expected, considerable effort has gone into exploring factors that impact balance such as the role of different sensory cues in triggering corrective actions (Bolton and Misiaszek, 2009; Macpherson et al., 2007; Stapley et al., 2002) or spinal and brainstem mechanisms acting to stabilize the body (Bolton and Misiaszek, 2012; Honeycutt and Nichols, 2010; Macpherson and Fung, 1999; Mori, 1987). Conversely, much less research has investigated the role of the cerebral cortex in balance. This gap has likely been encouraged by the long-held belief that postural responses are mostly managed sub cortically (Magnus, 1926; Sherrington, 1910). Of course, this view has been reasonable given that reduced animal preparations retain an impressive capacity for generating complex righting reactions (Honeycutt and Nichols, 2010). Moreover, the comparably slow pace of sensory-cued voluntary acts versus the onset of automated postural responses has influenced the assumption that much of the neural processing related to generating balance reactions originates sub-cortically. While subcortical networks are critical in generating compensatory behaviour more recent investigations have demonstrated that the cerebral cortex makes a meaningful contribution to compensatory balance reactions. The major shift towards recognising an important role for the cerebral cortex in balance control has been discussed previously (Jacobs and Horak, 2007; Maki and McIlroy, 2007) and the reader is referred to these comprehensive reviews. A critical distillation of these past reviews is that the cerebral cortex can: (1) Modulate upcoming potential postural responses via central set based on intention and knowledge of perturbation or environmental characteristics, (2) provide online monitoring of balance status and (3) modulate late-phase or change-in-support responses characteristics perhaps through direct control. The present review extends upon past work to highlight some of the more recent advances. This includes updated information regarding cortical contributions to the perception (and prediction) of instability as well as a role in shaping the motor response. Where possible an indication of the impact of particular brain regions in responding to external perturbations will be provided. Moreover, compelling evidence now exists that postural threat is associated with accelerated engagement of cortical networks thus challenging previously assumed speed of transmission barriers to why the cerebral cortex could not play a role in rapid postural responses. This review will present the emerging evidence for a cortical role in reactive balance emphasizing research that directly measures cortical neurophysiology in association with externally induced postural

143

responses. Moreover, suggestions for future research are provided. Overall, an updated framework is advanced for how the cerebral cortex may influence both the online generation of compensatory reactions as well as contributing to a priori setting of the central nervous system (CNS) state to influence such control.

2. What is a compensatory reaction? Compensatory balance responses are highly sophisticated, whole-body reactions rapidly generated by the nervous system to resist a loss of equilibrium. The initial triggered stage of the response, the automatic postural reaction (APR), is intimately linked to the sensory volley from a fall, evident by the direction specificity of the resultant muscle pattern and the fact that the motor response scales with the perturbation magnitude (Macpherson and Horak, 2013). These postural responses are extremely robust in that a variety of sensory inputs, even from separate modalities, can trigger an appropriate corrective action. Furthermore, APRs are highly generalizable in the sense that many different forms of perturbation may elicit a similar directionappropriate counter response including surface translations or tilts (Moore et al., 1988; Nashner, 1977; Safavynia and Ting, 2012) slips or trips while walking (Eng et al., 1994; Schillings et al., 1996) and perturbations to the torso (Misiaszek and Krauss, 2005). Importantly, these responses are not a simple collection of segmental stretch reflexes but instead represent complex patterns of muscle action organized around the goal of maintaining upright balance (Macpherson and Horak, 2013). This is perhaps most obvious when one considers that many of the muscles engaged in the automated action are often remote from the site of perturbation. For example a perturbation to standing stability can trigger rapid corrective reactions in the upper limbs at similar onset latencies to the early leg responses (McIlroy and Maki, 1995). The fact that these responses even at the remote locations have a direction-specific nature (e.g. early arm responses that aim toward a handle) argues against a generic startle effect but rather suggests a more behaviourally relevant motor command (Gage et al., 2007; Maki and McIlroy, 1997). This last point seems remarkable given that APRs are delayed relative to an autogenic stretch reflex but faster than muscle onsets associated with standard measures of voluntary reaction time, which tends to indicate an important role for fast-acting subcortical networks in coordinating this class of behaviour. An important feature of APRs is that they persist even when subjects try to supplant these actions with separate motor commands or attempt to suppress them altogether (Burleigh and Horak, 1996; McIlroy and Maki, 1993; Weerdesteyn et al., 2008). Thus, to a certain degree the initial postural response is immutable, at least in terms of the directionally tuned response pattern and onset. However, intention and environmental context can allow some critical modulation in the gain over these early actions. A classic example of contextual modulation in a balance reaction involves the distinct response strategies that subjects adopt when perturbed while standing on a surface that imposes different biomechanical constraints (Horak and Nashner, 1986). Here, it has been shown that when producing a counter reaction to a rapid platform translation subjects will tend to generate an ankle torque to resist the fall. However, when exposed to the same rapid translation but standing on a narrow beam motor reactions are now engaged about the hip to counter the body sway given that the ankle torque is no longer contextually relevant to control balance. Thus, these responses are not entirely hard-wired reactions but rather can be modified in a goal-specific manner. Although an automated and stereotyped early postural response is critical when recovering balance this must be reinforced with continuing action to eventually secure upright posture. This is

144

D.A.E. Bolton / Neuroscience and Biobehavioral Reviews 57 (2015) 142–155

particularly true when faced with large perturbations where recovering balance from a fixed support base is unlikely and requires a change in support reaction (Maki et al., 2003). These compensatory actions are more elaborate as when navigating a limb to a new stable support base or prolonging stance phase in one leg while a recovery step in the other leg takes additional time to clear an obstacle. Though the rapid APR offers a valuable early foundation to oppose instability the complex array of scenarios where falls occur often requires a higher degree of movement sophistication, which stresses the need for cortical involvement. This includes, in the case of change in support reactions, the additional challenges of spatial navigation and addressing affordances in the surrounding environment. The challenge is elevated considering that the naturally occurring compensatory reactions transpire when subjects are not expecting a perturbation thus reducing the possibility of achieving such control by pre-setting parameters prior to the onset of instability. While convincing evidence exists for the importance of subcortical structures in compensatory balance the typical research models employed need to be carefully considered when drawing conclusions regarding the neural control of balance. Typically the studies on reduced animal models are restricted to fixed support actions in an attempt to focus on early phases of the postural response. In fact, even the aforementioned decerebrate studies often need to use some form of body fixation (e.g. at the head and/or tail) further restricting the type of postural reactions that can emerge. These are very reasonable approaches to contend with limitations imposed by a reduced preparation and also to address the specific research questions these studies are designed to answer. However it should be noted that drawing sweeping conclusions about the predominance of subcortical roles in compensatory balance would be skewed by an exclusive focus on reduced animal models. This is particularly striking when one considers the very simple manner of imposing perturbations and relatively obstruction-free response environment in these studies. In addition, generalizing such findings to human balance control may be tenuous given the increased biomechanical challenge for humans using a small bipedal support base to control a relatively high centre of mass. Fittingly, there is already a recognized disparity in terms of how dependant humans are on cortical drive for locomotion versus quadrupeds possibly related to this greater challenge (Nielsen, 2003).

3. Indirect support for a cortical role in postural responses to externally induced perturbations From earlier reviews on this topic (Jacobs and Horak, 2007; Maki and McIlroy, 2007) several lines of evidence offered indirect support for the cortical contribution to postural recovery. Namely, past research demonstrated that when the cerebral cortex is either damaged or preoccupied with a cognitively demanding task postural performance is disrupted. A summation of this work is that compensatory behaviour ultimately demands the participation of cortical networks. More recent findings provide additional indirect support for a cortical contribution including: (a) specific brain regions associated with balance control and falls risk, (b) particular deficits manifest in the postural responses of stroke patients, and (c) the role of higher brain processes in sustaining visual spatial information to guide compensatory change in support reactions. At the most basic level a role for the cerebral cortex in compensatory balance can be appreciated by recognizing there is some instability in reduced animal models as well as in humans following cortical damage. In humans this is evident in the abnormal corrective actions presented by stroke patients. Here, postural responses to surface translations are weakened in the paretic limb and there is a clear shift toward compensatory actions on the non-paretic limb

(Geurts et al., 2005). The presence of a weak leg poses a particularly difficult challenge for stroke patients when using a compensatory stepping action since this would require executing a rapid step with one limb to establish a new base of support while the other is tasked with single-limb support of the entire body. In general these patients avoid step initiation with the paretic limb even when the non-paretic limb is either pre-loaded or physically blocked during perturbation trials (Lakhani et al., 2011a; Mansfield et al., 2011, 2012), which indicates a reluctance to incorporate this affected limb into the rapid response. The greater challenge with using the paretic limb means adopting altered motor patterns to counter a fall however greater instability is an inevitable consequence. A critical point from this work is not only the resultant instability due to a weakened limb post-stroke but also the recognition that these participants had great difficulty in adapting their behaviour to environmental constraints even with full awareness of these constraints. Quickly and efficiently selecting the appropriate stepping limb based upon the environment is critical when avoiding a fall. Indeed, specific limitations in this ability to adapt compensatory stepping to the environment may be an important factor with the heightened fall risk following stroke (Mansfield et al., 2012). As further testament to the link between cortical function and balance control there is a well-established correlation between the incidence of falls in older adults and severity of cognitive decline (Ambrose et al., 2013; Herman et al., 2010; Muir et al., 2012). Given that cognitive decline is rooted in structural deterioration within the brain this suggests a more direct link between physical brain degeneration of cognitive networks and impaired balance control. In support of this idea imaging work has revealed that the activity in brain regions associated with response inhibition and selective attention, indicators of executive cognitive processing, is linked with risk of falling (Nagamatsu et al., 2011). In particular, Nagamatsu et al. (2011) revealed that baseline activation of left orbital frontocortex and anterior cingulate gyrus were both associated with falls risk in community-dwelling seniors. Since the earlier reports on the connection between cognitive performance and balance there now exists a vast collection of reports that clearly expose the link between reduced cognitive capacity and heightened falls risk (Ambrose et al., 2013; Chen et al., 2012; Herman et al., 2010; Holtzer et al., 2007; Liu-Ambrose et al., 2008; Mirelman et al., 2012; Morris et al., 1987; Muir et al., 2012) even when the cognitive decline is slight (Gleason et al., 2009). Beyond results from lesion studies and age-related deficits, the interaction between neural networks that contribute to balance and those involved with cognitive function has been demonstrated using dual-task interference. As a general rule when participants are exposed to postural perturbations while simultaneously engaged in a cognitively demanding task (e.g. visual-motor tracking task, performing arithmetic, etc.) performance is impaired in one or both tasks (Maki and McIlroy, 2007). In the case of postural deficits such an effect emerges only after the earliest response to perturbation (Norrie et al., 2002; Rankin et al., 2000) supporting a distinction between the initial, automated reaction versus subsequent stages more influenced by cognitive resources. While the phenomenon of dual-task interference is well established there has been progress in recent years using this approach to further expose the link between higher brain processes and postural recovery. For example, one recent study used a cognitive task emphasizing visual working memory capacity in place of more traditional working memory tasks that involve both information storage and processing (Little and Woollacott, 2015). As expected, the authors demonstrated an interference effect on both postural recovery and cognitive performance when the tasks were combined thereby connecting a more precise cognitive function with the neural control of balance. Based upon imaging work where behavioural performance on this cognitive task alone

D.A.E. Bolton / Neuroscience and Biobehavioral Reviews 57 (2015) 142–155

correlated with activity in posterior parietal and occipital cortices (McCollough et al., 2007; Vogel and Machizawa, 2004; Xu and Chun, 2006) these same cortical networks potentially play a role in the postural response. Another recent advance involves insight into the demand for cognitive resources in sustaining a visual spatial map to guide compensatory change of support reactions. In studies where participants are required to produce a compensatory reach to grasp (RTG) to a stable handle, if visual occlusion occurs with a delay that deprives subjects of the visual scene there is a concomitant decrease in reaching accuracy (Cheng et al., 2013). Furthermore engaging these subjects in a cognitively demanding task results in further performance decline of a compensatory reach. Overall this work emphasizes the cognitive role in not only guiding movement based upon the environment but also in holding a memory trace of the visual scene to guide such action. Although the indirect evidence for cortical contributions to compensatory balance has been informative several limitations exist. For example, research on populations with neurological injury pose a challenge in that (a) these subjects rarely have lesions that are isolated to a particular cortical location and (b) it is often unclear if the measured response reflects the specific role of the lesion site or some compensatory state of the CNS. Notably, participants in the aforementioned stroke studies were not specified to have lesions isolated to the cortex therefore such results do not exclude the possibility of altered subcortical function leading to an impaired postural response. Similarly, findings from dual-task interference studies and the observed deficits with ageing do not provide a clear, mechanistic view of exactly how the cerebral cortex influences the postural response. In fact, dual tasking is not purely a cortical task and it is possible that subcortical resources are competing during postural response tasks. Given the important link between cognitive function and falls it would appear that acquiring such a detailed understanding is imperative.

4. Direct measures of cortical neurophysiology related to externally induced postural responses The most direct evidence of potential cortical involvement comes from recording neural activity temporally coupled to induced instability and the onset of compensatory balance reactions. This has been shown in animal studies using direct recording from cortical neurons (Beloozerova et al., 2003b, 2005) and in human studies using brain imaging techniques such as EEG (Adkin et al., 2006; Jacobs et al., 2008; Marlin et al., 2014; Mochizuki et al., 2008, 2009, 2010; Quant et al., 2004a; Varghese et al., 2014) much of which has been discussed previously (Jacobs and Horak, 2007; Maki and McIlroy, 2007). Studies using direct recordings from the motor cortex of standing cats (Beloozerova et al., 2005) and rabbits (Beloozerova et al., 2003a) exposed to surface perturbation show a connection between cortical activation and muscle responses in the perturbed limbs. Thus, the cerebral cortex shows direct involvement in resisting perturbation even if only to augment existing subcortical mechanisms (although there is compelling evidence in both human and animal models that this trans-cortical response to segmental perturbation is much more sophisticated, Pruszynski and Scott, 2012). While the studies discussed below offer direct evidence for a cortical contribution to reactive balance control in humans this undoubtedly represents an incomplete picture. Given the infancy of using direct neurophysiological measures to address the cortical role in balance defining the brain regions involved in postural responses will only represent what is currently known based on a limited number of studies. In some cases the type of measurement used may restrict us to relatively easy-access cortical targets. For

145

example, an emphasis on regions such as the primary motor cortex (M1) partly reflects the relative ease of exploring this cortical node in the response network using techniques such as transcranial magnetic stimulation (TMS). Furthermore, even where good evidence exists for the involvement of a particular cortical generator in the response to perturbation, a precise function is often difficult to ascertain, at least to the exclusion of other possible roles. However as more studies begin to incorporate direct neurophysiological probes a greater understanding for these cortical roles will emerge. Since the major reviews on this topic in 2007 some notable advances include: (1) The use of functional near-infrared spectroscopy (fNIRS) to image brain activity following postural perturbation, (2) The use of non-invasive brain stimulation to temporarily disrupt cortical processing to investigate cortical contributions to compensatory arm reactions, and (3) greater insight into the cortical generators of perturbation-evoked potentials using electroencephalography including a revised view on what this cortical response may represent in terms of postural recovery.

4.1. Functional near-infrared spectroscopy To provide a direct measure of cortical activity associated with balance a mobile version of positron emission tomography has been used to demonstrate that various static (i.e. non-perturbed) balance tasks such as standing on one leg or standing with eyes closed engage distinct cortical regions depending on the particular task (Ouchi et al., 1999). Extending brain imaging to compensatory balance, fNIRS has been instrumental in monitoring cortical activity in response to platform translations (Mihara et al., 2008). Once again, a number of cortical structures become activated in response to both expected and unexpected perturbations including brain areas generally considered to be involved in executive control (i.e. prefrontal cortex, or PFC). Notably, such a result is highly congruent with the dual-task interference previously noted where executive neural regions would be involved. Follow-up studies by this same group using a similar approach to investigate reactive balance in stroke patients offered further evidence for cortical engagement in response to platform translations (Fujimoto et al., 2014; Mihara et al., 2012). Of particular interest from these results was the finding that increased activation of the supplementary motor area (SMA) and PFC following intensive rehabilitation was correlated with improved stability measures (e.g. Berg Balance scale) in stroke survivors. It is difficult to speculate whether increased SMA and PFC responsiveness following training reflects the return to a normal cortical contribution to balance or is due instead to some compensation. Either way this demonstrates a connection between cortical activity immediately following perturbation and stability measures.

4.2. Electroencephalography One caveat with the aforementioned imaging techniques is that these measures have poor temporal resolution and therefore do not provide insight into the time course of cortical involvement relative to a perturbation. To overcome such limitations electroencephalography (EEG) has been used to investigate cortical processes and with this technique several studies have demonstrated clear cortical events both before and after a perturbation (Adkin et al., 2006; Dietz et al., 1984, 1985; Jacobs et al., 2008; Little and Woollacott, 2015; Marlin et al., 2014; Mochizuki et al., 2008, 2009, 2010; Quant et al., 2004a,b; Smith et al., 2012; Varghese et al., 2014, 2015). Overall, these EEG potentials represent cortical processes time-locked to the perturbation such as early sensory intake and correlates of

146

D.A.E. Bolton / Neuroscience and Biobehavioral Reviews 57 (2015) 142–155

perception as well as motor-related events all with fine temporal precision while also providing some degree of spatial localization. 4.2.1. Perturbation-evoked cortical responses The most prominent early cortical potential related to a loss of balance is the N1, a negative potential with a peak occurring approximately 100–150 ms following a perturbation. The topographical location of this potential suggests cortical generators of frontocentral origin (Maki and McIlroy, 2007) with recent work offering further insight into the specific generators as discussed later in this section (Marlin et al., 2014) (See Table 1). The N1 occurs in response to multiple modes of perturbation and relates to general features of event detection rather than reflecting direct sensory processing or the explicit execution of motor responses (Mochizuki et al., 2009; Quant et al., 2004b). Indeed recent work has even revealed evidence for a small amplitude N1 response time-locked to naturally occurring moments of instability during quiet stance (Varghese et al., 2015). While the N1 is typically considered in terms of large, external perturbations it is of interest to note that this cortical response to instability is evident even when time-locked to much smaller levels of sway that naturally emerge during stance. This underscores a basic mechanism linking cortical activity with instability, one that scales with the postural challenge. Mochizuki et al. (2009) effectively demonstrated the fact that the N1 represents a general cortical response to instability. In their study the mode of perturbation (either seated or standing) and the means of motor response (arm versus leg response) greatly differed and yet the N1 was preserved in both timing and amplitude. Moreover, the fact that the topographical location of these cortical potentials was not lateralized for arm versus leg responses provided added evidence against any simple link with generating the motor output. This latter point is similar to past findings by Quant et al. (2004b) where the N1 persists even without motor responses coupled to the instability. While peak onset of the N1 is highly consistent across task conditions, the amplitude of this potential varies considerably. For example, even if perturbation magnitude is held constant, the N1 amplitude will vary with predictability of the perturbation (Adkin et al., 2006; Mochizuki et al., 2008) age of the participants (Duckrow et al., 1999) attention (Little and Woollacott, 2015;

Quant et al., 2004a) and the presence of concurrent sensory discharge (Staines et al., 2001). Remarkably, the N1 is attenuated when subjects are simultaneously engaged in a secondary cognitive task clearly exposing reliance upon attention-demanding cortical resources (Little and Woollacott, 2015; Quant et al., 2004a). Given this attention demand and the increased amplitude of the N1 with unexpected perturbations it has been proposed that this potential may in fact index some form of error detection and more specifically the detection of instability (Maki and McIlroy, 2007). Support for this idea comes from observing the similarities between N1 and the error-related negativity (ERN) that arises following errors in cognitive tasks (i.e. both potentials are maximal over frontalcentral electrode sites). With several lines of research indicating an Anterior Cingulate Cortex (ACC) generator for the ERN (Carter et al., 1998; Holroyd et al., 1998; Kerns et al., 2004; Miltner et al., 1997, 2003) a recent study explored whether a perturbation-evoked N1 response originated at the same cortical site as the ERN produced in a flanker task (Marlin et al., 2014). Unexpectedly, their results revealed the ERN and N1 arose from different neural generators. In particular, the ERN originated from the ACC as expected, but the N1 was localized at the SMA, a traditional motor area. While the dipole localising technique doesn’t necessarily exclude multiple cortical generators contributing to the N1 (including the ACC), it does at least indicate some task-related differences in the generation of the ERN based upon error detection in a cognitive task versus the perturbation-evoked N1. This study by Marlin et al. (2014) offers insight into the possible significance of the N1 in response to perturbation. In particular, these findings suggest that the N1 may reflect neural events related to motor planning and execution in the balance context. It is important to recognize that the later onset for the N1 peak relative to the early automatic postural response indicates that these cortical processes do not directly contribute to generating the early postural response, which is instead likely generated by spinal and subcortical networks (Jacobs and Horak, 2007). Subsequent phases of the compensatory balance response may then involve the cerebral cortex with the ongoing response modified to suit environmental constraints and individual goals. This would be consistent with dual-task interference results that demonstrate the earliest postural responses are unaffected but later stages of the compensatory

Table 1 Summary table illustrating cortical regions involved with the postural responses to externally induced perturbations. The proposed role for each brain region in the postural response is included. For the purpose of this table only human studies employing direct cortical neurophysiological probes are included. Abbreviations: DLPFC—Dorsolateral Prefrontal Cortex, FEF—Frontal Eye Fields, SMA—Supplementary Motor Area, PPC—Posterior Parietal Cortex, ACC—Anterior Cingulate Cortex, CNV—Contingent Negative Variation, ERD—Event Related Desynchronization. Measure

Brain regions

Reported role in response

Functional near-infrared spectroscopy

Prefrontal Cortex Supplementary Motor Area Posterior Parietal Cortex

Transcranial Magnetic Stimulation

Primary Motor Cortex

Electro-encephalography Post-perturbation

Supplementary Motor Area Anterior Cingulate Cortex

Electro-encephalography Pre-perturbation

Supplementary Motor Area Sensorimotor cortex

Bilateral DLPFC and FEF perturbation-evoked response Potential role in the allocation of visuospatial attention. Note: activation occurs even without pre-warning cue. Prediction of impending perturbation was associated with enhanced activation in the SMA (role in motor preparation) and right PPC (role in visual-spatial attention and representation of body schema) (Mihara et al., 2008) Perturbation-related activation in bilateral SMA in stroke patients following rehabilitation (role in the evoked postural response to perturbation) (Fujimoto et al., 2014) Motor cortical influence over the postural response in the perturbed lower limb (Taube et al., 2006, 2007). Note: the motor cortical influence on lower-limb responses appears specific to late-phase response characteristics. Motor cortical influence over the evoked postural response in the upper limb (Bolton et al., 2011, 2012) N1—Large, negative, perturbation-evoked potential located over frontocentral scalp electrodes. Speculated cortical origins from surface EEG include the ACC and SMA. Proposed role in event/error detection (Maki and McIlroy, 2007)Marlin et al. (2014) localized N1 to an SMA generator. The specific SMA locus suggests a contribution towards planning and execution of later stages of the evoked postural response CNV—negativity maximal at vertex with proposed cortical generators SMA and sensorimotor cortex Potential role in motor preparation for an imminent postural response. Note: evident only when perturbations are predictable (Jacobs et al., 2008; Mochizuki et al., 2010; Smith et al., 2012) ERD (beta)—Suggested SMA generator. Changes associated with perturbation magnitude for PD group but not healthy controls suggests a more direct cortical influence on set-related motor preparation for those with PD (Smith et al., 2012)

D.A.E. Bolton / Neuroscience and Biobehavioral Reviews 57 (2015) 142–155

balance response demand cognitive resources as indicated by a diversion of attention from the secondary task approximately 200 ms after reaction onset (McIlroy et al., 1999). Overall, Marlin et al. (2014) suggest that the N1 may represent the involvement of the SMA, potentially along with other regions such as the PFC, in generating a motor plan to shape the later stages of the compensatory balance response. However, further study is required to determine the specific function of the perturbation-evoked N1. Another interesting line of enquiry using EEG has investigated the frequency characteristics of the cortical responses to wholebody balance perturbations (Varghese et al., 2014). The results reveal evidence of EEG frequency modulation associated with the perturbation-evoked N1 response. Specifically, the findings suggest a relationship between synchronization of distinct frequency oscillations and the perturbation evoked N1. It appears the N1 may be a product of multiple complex waveforms rather than a single event from a particular brain region. Such phase resetting of the ongoing EEG oscillations has already been demonstrated to play an important role in the cortical responses to auditory and visual stimuli (Brandt, 1997; Jansen et al., 2003; Makeig et al., 2002) as well as in more high-level perceptual processes such as the ERN (Luu et al., 2004). This frequency modulation may be related to binding multiple cortical processes to the common task of maintaining stability. However, exactly how a neural signal coded via phase resetting of ongoing EEG oscillations applies to reactive postural control is presently unknown. 4.2.2. Pre-perturbation cortical activity In addition to a delayed online cortical response to instability evidence suggests that the cerebral cortex plays an important role in anticipatory changes throughout the CNS serving the control of balance (Jacobs and Horak, 2007). Preparatory neural modulation or ‘central set’ occurs prior to the onset of the perturbation and can manifest at multiple levels of the neural axis in task-dependant ways. Here, central set is broadly considered as the ‘modification of automatic motor responses based on expectation of stimulus characteristics’ (Horak et al., 1989). A major challenge to overcome when investigating central set in postural control is that overt behavioural measures such as ground reaction forces, motion capture, or muscle onsets are unable to capture neural modulation in advance. Conversely, EEG offers a means for measuring this preparatory cortical state in the timeframe before a perturbation. An early study to explore this association between anticipatory cortical activity and the response to externally triggered perturbations was conducted by Jacobs et al. (2008). Here, they monitored cortical activity before a postural perturbation in participants where the predictability of perturbation timing was manipulated. Their findings revealed a prominent cortical potential related to motor preparation, the contingent negative variation (CNV), evident only when perturbations were predictable. Of note, this study was the first to demonstrate the existence of CNV potentials prior to an externally induced postural perturbation. Given that the neural generators of the CNV potential’s components in voluntary motor control include the supplementary motor and primary sensorymotor cortex (Bares et al., 2007; Hamano et al., 1997; Lamarche et al., 1995) this offers compelling evidence for a cortical influence on the evoked postural response (Table 1). Importantly, this preparatory cortical activity paralleled cue-related improvements in postural recovery. These findings extend upon past research where preparatory neural changes have been inferred from the resultant behaviour. Here, instead of observing the consequences of central set, a measure that appears to represent central set more directly was offered by focusing on cortical activity in the time frame before anticipated perturbation. In a later study, Mochizuki et al. (2010) explored the relationship between these preparatory measures of central set and

147

perturbation-evoked cortical activity. Throughout their experiment perturbation timing remained predictable, however the magnitude of the perturbation varied along with the predictability of the perturbation magnitude. Their findings revealed that preperturbation cortical activity (i.e. CNV) scaled to the expected size of perturbation whereas the N1 always scaled to the actual size of perturbation (Fig. 1). For situations where perturbation magnitude could not be anticipated, a higher ‘default’ level of preparatory cortical activity was observed. In general their results reveal dissociation between preparatory and perturbation-evoked cortical markers with the N1 reflecting the consequence of instability and the CNV reflecting anticipatory set specific to postural context. Similarly, Adkin et al. (2008) noted differential effects of postural conditions on pre-perturbation versus perturbation-evoked cortical potentials when investigating the cortical responses to postural instability. In their study increased postural threat (achieved by perturbing individuals at an elevated height) produced greater N1 response amplitudes to unpredictable perturbations whereas the level of postural threat had no discernable impact on anticipatory potentials. Anticipatory cortical roles in relation to postural responses were subsequently investigated in patients with Parkinson’s disease (PD) (Smith et al., 2012). Given the diminished central set evident in these patients during postural recovery the authors hypothesized that the impaired response scaling with PD would be associated with deficient cortical preparatory activity. In addition to the aforementioned CNV they included event-related desynchronization (ERD) measures to identify more focal cortical processing related to movement. Similar to past reports, participants with PD revealed an impaired ability to scale their postural responses to a predictable magnitude of perturbation. Interestingly, the CNV prior to perturbation was present in both groups suggesting a retained capacity for cortical preparation in those with PD. Unexpectedly, their specific measure for response preparation within motor corticostriatal circuits (beta-ERD) correlated with postural responses in PD but not controls. As mentioned by the authors this finding possibly suggests a more direct influence on postural response scaling for those with PD. These results may reflect some form of cortical compensatory mechanism although an imperfect one given that scaling of the response amplitude to the expected magnitude of perturbation does not occur. 4.3. Transcranial magnetic stimulation Some important insights into the timing of cortical contributions to balance have also been afforded by the use of TMS. This technique induces a focal activation of underlying neural networks to either modify cortical excitability via repetitive TMS or as a measurement probe using single-pulse TMS. One common approach is to apply stimulation over M1 and then measure the resultant motor-evoked potential (MEP) in a target muscle downstream, which can be used to assess the cortico-spinal drive onto the muscle. Using a combination of single-pulse TMS and peripheral nerve stimulation an early cortical influence (<90 ms) on muscle activity of a rapidly stretched ankle muscle has been demonstrated immediately following sudden platform translation in standing subjects (Taube et al., 2006). These findings corroborate results previously discussed using recording electrodes within the motor cortex of animal models to demonstrate that M1 activity preceded postural muscle activity in the perturbed limb muscles (Beloozerova et al., 2003b, 2005). In fact, in a follow-up study by Taube and colleagues a correlation was noted between increased cortico-spinal drive onto perturbed ankle muscles and stability performance in subjects that were trained in a balance task (Taube et al., 2007). These studies therefore provide support for the rapid cortical contribution to counter a perturbation, at least in terms of resisting a

148

D.A.E. Bolton / Neuroscience and Biobehavioral Reviews 57 (2015) 142–155

Fig. 1. Electrophysiological and kinetic data comparing preparatory and perturbation-evoked responses following temporally predictable perturbations of different magnitudes. Single subject representation of the EEG (top panel), COP (middle panel), and stepping limb gastrocnemius EMG (bottom panel) averaged across all trials in the conditions where participants either (A) knew the perturbation magnitude in advance or (B) where perturbation magnitude was unpredictable. Perturbations were imposed using a lean and release apparatus. Solid lines depict large amplitude perturbations requiring a stepping response. Dotted lines depict small amplitude perturbations, which were managed by feet-in-place responses. The vertical dashed line at time zero represents perturbation onset. Adapted from Mochizuki et al. (2010).

D.A.E. Bolton / Neuroscience and Biobehavioral Reviews 57 (2015) 142–155

sudden segmental stretch. However, as stated previously compensatory reactions are more than mere segmental stretch responses so these findings do not necessarily point to a cortical role in triggering complex postural synergies. From the earlier sections it was presented that the automated reactions to maintain balance are not limited to the perturbed body part but in fact constitute a synchronized body-wide reaction with consistent response latencies even in muscles remote from the site of physical disturbance. This would suggest a more centralized coordinated command. Thus a relevant way to assess what central mechanisms drive the complex corrective action is to measure muscle responses in a body segment that is not directly perturbed. A recent line of inquiry by McIlroy and colleagues has adopted a rapid chair-tilt paradigm to help address this issue because it involves a compensatory arm response to avoid falling backward while seated (Bolton et al., 2011, 2012; Gage et al., 2007; Mochizuki et al., 2009). Notably, the earliest muscle onset latencies in response to the chair tilt are equivalent to arm response latencies reported in standing perturbation studies suggesting that the reactions being probed by this model may represent similar RTG synergies. In the original study by Gage et al. (2007) subjects were instructed to quickly grasp a handle in response to one of two distinct cues: either an auditory tone or a sudden tilt of the chair. Muscle activity in the right (response) arm was recorded along with arm kinematics and these measures revealed a preserved spatial-temporal reach pattern for both conditions hinting at a similar neural control network. What was remarkable however was the same reach pattern was produced approximately twice as fast in response to the perturbation even though participants were always instructed to reach as fast as possible for either cue and these trials were randomly intermingled. Given that the voluntary reaction task using the auditory cue would be assumed to transit a cortical route the expression of a common reach pattern would suggest that the same cortical networks may be engaged to produce the compensatory RTG. It is important to recognize however that producing the same motor pattern in itself does not necessarily require the same cortical origin. In fact, the same basic reaching synergies could well be housed within subcortical networks and prepared to release quickly upon reception of the cueing stimuli. Such a mechanism has been previously proposed as underlying rapid motor responses that are greatly accelerated when combined with a startling acoustic stimulus (Carlsen et al., 2004; Valls-Sole et al., 1999). Thus the cerebral cortex may be altogether bypassed when producing what may in fact be brainstem-mediated reactions. In a follow-up study this specific question of whether a cortical network was involved in producing both the auditory-cued and perturbation-evoked arm response was tested more directly by temporarily interrupting M1 excitability. Here, continuous theta burst stimulation (cTBS) was applied over the hand representation of M1 to transiently disrupt cortical activity in order to explore its role in the two reaching conditions (Bolton et al., 2011). Results confirmed the earlier findings of Gage et al. (2007) in that compensatory RTG involved a similar muscle activation pattern but at a much faster rate than the voluntary RTG. The important extension here was that in both cases there was a focal amplitude reduction of hand muscle responses following the application of cTBS (Fig. 2). This offered support for the notion that the compensatory arm response did in fact involve a motor cortical contribution. Additional support was provided in a second study, again applying cTBS over the same focal hand representation and exposing participants to sudden chair tilts (Bolton et al., 2012). In this case participants held onto a fixed handrail with both hands and were required to quickly pull themselves back to an upright, seated position using both arms upon chair release. This bilateral, fixed support compensatory balance paradigm had the advantage of a built-in control in that it allowed comparison between muscle responses in

149

the stimulated arm versus the non-stimulated arm. As before, the results revealed altered response amplitudes in focally stimulated hand muscles while non-stimulated arm muscles were unaffected. The combination of both studies offers support for a cortical (at least motor cortical) role in contributing to the compensatory arm response even when the perturbing stimulus arises from a remote site from the responding muscles. Some caution should be exercised when interpreting the above studies as it is unknown if the association between cortical excitability and postural responses generalizes across upper and lower limbs. This is particularly relevant given previous evidence suggesting postural responses of the upper limb correspond more with evoked cortical potentials than those of the lower limb (Quintern et al., 1985). Furthermore, while the above results suggest a cortical role in the evoked postural response there is the possibility that changes to cortical activity using cTBS indirectly affect subcortical networks such as the striatum via connections with the cortical target (Ko et al., 2008). Notably, recent evidence suggests that the application of transcranial direct current stimulation (tDCS) over motor regions modulated the excitability of subcortical structures resulting in accelerated onsets for brainstem-mediated movements (Nonnekes et al., 2014). Thus, the potential exists that brain stimulation techniques such as cTBS may result in an altered subcortical-mediated postural response rather than reflecting disrupted transmission through the motor cortex. Nevertheless, a focal drop in the intrinsic hand muscle response as noted above would not be consistent with the multi-segmental motor organization evident in many subcortical regions such as the basal ganglia (Wichmann and DeLong, 2013).

5. Evidence for accelerated speed of processing A major barrier to recognizing a cortical role in compensatory balance has been the temporal disparity between volitional versus compensatory movement onsets. Certainly a cortical contribution to postural reactions would not be consistent with processing delays associated with typical voluntary reactions. Yet from the chair-tilt studies just described one can see that when movements were triggered by the perturbation, muscle onsets were greatly accelerated while still traversing a motor cortical route. Thus, what may drive such an increased speed of cortical involvement? One possibility is that compensatory movements are somehow accelerated when linked with either the centrally perceived threat of a fall and/or the stimulus characteristics associated with the perturbation. Presumably, this may involve activating ascending arousal mechanisms originating in the brainstem that then boost cortical responsiveness (Jones, 2008; Sibley et al., 2014). It is notable that many brainstem networks involved in activating postural synergies via descending projections (Stapley and Drew, 2009) also project to the forebrain and therefore could simultaneously act to accelerate cortical processing (Jones, 2008). A recent series of studies have used the electro-dermal response (EDR) to assess physiological arousal following rapid chair tilt perturbations (Sibley et al., 2008, 2009, 2010a,b). One important finding to emerge from this work was that the EDR did not reflect sensory or motor drives alone but instead reflected the need to execute a rapid balance reaction within the context of instability (Sibley et al., 2009). That is to say that this marker of physiological arousal was sensitive to the overall context of compensatory balance versus a simple mapping to motor or sensory components of the reaction. Remarkably, it has been demonstrated that even non-compensatory movements (e.g. raising the heel off a footrest) when temporally coupled with the chair release produced equally accelerated reaction times despite the fact that this movement is unrelated to countering the fall (Lakhani et al., 2011b). Thus, the

150

D.A.E. Bolton / Neuroscience and Biobehavioral Reviews 57 (2015) 142–155

Fig. 2. Schematic representation of the reach-to-grasp paradigm with either chair perturbation or auditory tone as the response cue. TOP: Single subject data for individual trials, pre-cTBS (black line) versus post-cTBS (red line) muscle responses to perturbations and auditory cues (Individual bursts aligned to average pre-cTBS onsets to emphasize differences in response amplitude.). BOTTOM: Group average response amplitudes (IEMG over initial 100 ms of muscle burst). Data presenting post-cTBS values as a percentage of pre-cTBS values. * Significant difference at p < 0.05. Adapted from Bolton et al. (2011). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

movement need not be relevant to corrective balance per se but simply rides upon a phasic wave of cortical arousal provoked by unanticipated instability. Relating this back to the startle reflex it is clear that a more insistent physical stimulus, such as a loud sound, can evoke a prepared motor action faster than when movements are prompted by non-startling stimuli. A number of studies have investigated this phenomenon and found a consistent acceleration of many different movements, both simple and complex, when coupled with a startling acoustic stimulus (Carlsen et al., 2004; Valls-Sole et al., 1999). Because these movements were produced much faster than what could be elicited in the normal voluntary condition it has been proposed that brainstem networks may store the prepared movement for quicker release due to being linked with

the brainstem-mediated startle response. To test this idea a recent study used single-pulse TMS to transiently suppress M1 while subjects performed a rapid wrist movement in a standard startlereaction paradigm (Alibiglou and MacKinnon, 2012). Crucially, aptly timed TMS pulses delayed both normal auditory-cued and startle-evoked movements providing evidence that both movements traversed a cortical route despite a much faster response speed with startle-evoked movements. In fact, their results showed a clear dissociation between the wrist response and the startle reflex activity manifest in the sternocleidomastoid muscle, a muscle associated with the classic startle reflex. The authors proposed that rather than stored motor plans released from brainstem networks due to the acoustic startle these networks facilitated the cortex via ascending projections, which then triggered the

D.A.E. Bolton / Neuroscience and Biobehavioral Reviews 57 (2015) 142–155

pre-planned movement from M1. While there are clear differences between startle reactions and compensatory balance reactions the point is that using standard measures of voluntary reaction time as a gauge for how quickly the cerebral cortex can play a role in sensorimotor transformation would fail to account for such mechanisms that facilitate processing speed. Aside from arousal effects recent work has provided evidence that stimulus intensity alone prompts a faster motor response in a simple reaction time task (Lakhani et al., 2012). The important implication to compensatory balance is that a large, synchronous afferent volley from multiple sensory channels would presumably translate into a movement much quicker than some solitary, and comparably mild cueing stimulus used in typical voluntary reaction tasks. This is particularly relevant given that so many sensory pathways can evoke postural responses. Furthermore, when chair-tilt perturbations are paired with nonstartling auditory tones, subjects demonstrate a conditioned rise in arousal (measured by EDR) and faster muscle onset times in subsequent trials even in the absence of the actual perturbation (Lakhani et al., 2013). This suggests that the tone alone could trigger the physiological arousal and the faster motor reaction. Thus stimulus context also plays an important role in dictating speed of processing. It now appears that factors ranging from bottom-up effects such as the stimulus intensity can drive reaction speed, but also top-down expectation can incite the arousal needed to boost responsiveness. In any event, either mechanism can hasten cortical contributions to compensatory balance reactions in a manner that has not been fully appreciated to date. In the chair perturbation studies just described postural recovery is managed by an upper limb response leading one to question if upper and lower limb responses can be generalized. This is an important consideration given the many circumstances where rapid stepping reactions may be the only recourse to maintain stability. As noted by Gage et al. (2007) upper limb responses often demand much higher levels of precision for acquiring handhold targets to establish a new base of support. This may distinguish upper versus lower limbs with regard to control demands. Conceivably such heightened demands for upper limb control could result in differences in how the cerebral cortex influences the respective postural responses. Although this matter remains unresolved the results from Lakhani et al. (2011b) suggest that at least in terms of activation speed pre-set responses can be accelerated to a similar degree for both the arm and leg. Overall, cortically driven reactions appear to have a similar capacity to engage both upper and lower limb responses when coupled with arousal due to postural threat. However, it is difficult at present to know if this generalization applies beyond latency of response. 5.1. Timing of cortical influence on the postural response Consistent with previous reviews (Jacobs and Horak, 2007; Maki and McIlroy, 2007) it remains clear that the earliest postural responses following externally induced perturbations are likely subcortical in origin with increasing probability for a cortical influence as time elapses. For instances where instability can be anticipated lower regions of the neural axis can be primed to facilitate rapid and contextually relevant counter reactions. Additionally, there is now evidence for a potential cortical contribution to compensatory reactions much earlier than once assumed. Indeed beyond priming more basal response synergies housed in brain stem networks there is now reason to believe cortical responses can also be prepared to release programmed actions at an accelerated rate contingent upon factors such as arousal. Essentially deep brain structures may be energising output from cortical centres during (or just prior to) external perturbations. Although some foreknowledge would be needed to prepare sophisticated cortical counter

151

measures prior to a postural disturbance there are many instances in daily life where such anticipation may be afforded. 6. Potential roles for the cerebral cortex in balance: Future directions While there is a growing appreciation for a cortical contribution to balance control in humans a pronounced gap remains in understanding exactly how this works. Certainly, cortical mechanisms serving balance have not been explored with the same meticulousness as subcortical righting mechanisms. This is likely due to the comparably recent recognition of a significant cortical contribution to balance control but also the greater difficulty in directly exploring this role. For example a typical reduced animal model to probe the neural control of balance involves removing the cortex, possibly along with other higher structures, and then either recording activity in subcortical neural regions and/or stimulating these areas to monitor postural changes (Mori et al., 1982, 1992; Mori, 1987; Whelan, 1996). Such reduced preparations are clearly not possible when working from the highest level of the neural axis. Ultimately the number of studies that this review can rely upon for direct measurement of cortical roles during extrinsically induced postural responses is limited particularly when focused on human research. As pointed out recently by Jacobs (2014) this knowledge gap can be addressed with the application of more direct assessment methods such as tDCS, TMS, EEG, and fNIRS (Jacobs, 2014). In fact, these tools could offer a particularly powerful complement to the more traditional biomechanical measures of compensatory balance behaviour in an attempt to develop a comprehensive understanding of the neural control of balance. Consequently one important part of this review is to advance viable suggestions for how the cerebral cortex may be involved in compensatory balance control to help guide future research efforts. In this regard related research in areas of either non-perturbed balance or in normal voluntary motor control may offer useful hints regarding potential cortical roles in compensatory balance. It is critical to note however that these ideas await experimental confirmation within a reactive balance context. 6.1. Building reference frames and putting the world into motor terms Spatial reference frames of the body within the environment influence compensatory reactions. This is perhaps easiest to imagine when one considers change of support reactions where an arm or leg must be navigated through the environment to establish a new support base. The notion of a continuously updated egocentric spatial map that may direct rapid compensatory movements has been proposed previously (Maki and McIlroy, 2007) and exemplified in a paradigm where subjects were exposed to an unexpected, one-time perturbation while walking through an unfamiliar room (King et al., 2011; Maki and McIlroy, 2007; Maki et al., 2008) (See Fig. 3). When participants navigated a complex environment filled with potential obstacles and affordances (e.g. handrail), eye gaze measurements revealed no eye fixation to the handrail ultimately used in the recovery response. Instead it appears that visual information gathered from earlier saccades to the target handle was incorporated into some internalized frame of reference to direct the compensatory arm response. It is important to stress that this would involve continually updating a map of the external world without awareness of an imminent fall and yet this map seems to be available to guide a rapid and targeted postural adjustment. Overall, there is evidence that at least some of the influence of the cerebral cortex on the balance reaction is established prior

152

D.A.E. Bolton / Neuroscience and Biobehavioral Reviews 57 (2015) 142–155

6.2. Cortical roles in predicting instability

Fig. 3. Methodological details from King and colleagues (2011). (A) Schematic drawing of the large (2 m × 6 m) motion platform used to evoke the reach-to-grasp reactions. (B) Photograph showing the view seen by the subject after opening the door at the start of the trial (the telephone that the subject was instructed to find is located on the desk, next to the computer monitor). (C) Example eye-tracker scene-camera video image, showing the point-of-gaze cursor and the “gaze ellipses” corresponding to visual angles of 5◦ , 10◦ , 15◦ , 20◦ and 30◦ (in the displayed image, gaze is fixated on the computer monitor and the far end of the handrail is visible within a 15◦ visual angle). Adapted from King et al. (2011).

to perturbation. The notion of central set has been extensively studied and in the traditional sense it has been regarded as a key factor when people have prior knowledge of perturbation and/or response characteristics including temporal predictability of the disturbance. However even when a fall is completely unexpected, thus eliminating the potential for direct pre-planning, the cortex may still play a role in preparing the system to respond. Studies exploring non-speeded voluntary movements would suggest that simply viewing objects in the environment primes relevant potential actions geared towards interacting with these objects. Gibson originally proposed this concept of affordances (Gibson, 1979) and since that time both animal (Rizzolatti and Luppino, 2001) and human (Franca et al., 2012) studies have offered support for this idea. Consequently if our perception of the world is continuously being put into motor terms (Rizzolatti et al., 1997; Rizzolatti and Luppino, 2001) this has clear implications for how efficiently corrective actions can be quickly produced. In other words, perception of objects in the surrounding environment could bias particular responses even without explicit foreknowledge of a fall. As previously discussed, postural threat may drive cortical networks to play an earlier and more meaningful role in compensatory balance responses. However beyond the notion of simply driving the system faster it seems that the system itself is designed to operate with greater ethological efficiency than once thought. This more automated link between perception and action at a cortical level would have clear advantages in reactive balance control. Ultimately to elucidate if such perceptual priming occurs would require direct measures of cortical neurophysiology in the time period both before and after perturbation and preferably in conditions where perturbation is truly unexpected. Such study designs present many methodological challenges. However these efforts may be highly informative in linking some important aspects of a priori cortical processes and postural control.

The cerebral cortex is well equipped to learn patterns and predict future states (Hawkins and Blakeslee, 2004). Herein resides a potentially important contribution of cortical networks to maintaining balance, that is, prediction of upcoming events using incomplete, early sensory data. In a balance context this means the capacity to predict future instability based upon preliminary afferent information indicating an imminent fall. In this way we are not constrained to catch up with instability that has already occurred but rather we may be able to generate counter action against a predicted future instability. Such an idea is consistent with the fact that neural markers (via EEG) are strongly correlated to the eventual time to contact the boundary of stability but only weakly correlated to standard centre of pressure measures which are more about the location at that specific point in time (Slobounov et al., 2009). These markers measured as modulation of EEG activity within distinct frequency bands consistently presaged individuals reaching their margin of stability defined by the boundaries of their feet. Essentially this cortical EEG marker reflected the ‘virtual time-tocontact’, which was a forecast of the time when participants would reach stability limits thus potentially affording an early-warning signal to start corrective action to counter an imminent fall. If postural control is indeed a protective set of actions guarding an internal representation of uprightness then this predictive element is not only concerned with where the body is in space at that moment but when will it become unstable based upon the body’s own stability margins. More recent work shows a similar cortical marker detecting future instability in a challenged walking task prior to any overt signs that the body was falling (Sipp et al., 2013). In the aforementioned studies there are clear cortical indicators that occur prior to overt instability and appear to be predictably time-locked to the eventual loss of balance. This provides strong support for the notion that the brain is using an experienced based model of the emerging sensory data to predict when the boundaries of stability will be breached. It is however unclear if these markers index detection of instability, allocation of attention, or generation of the counter response. It is further unclear if such predictive measures play a role in the much more pronounced perturbations in reactive balance studies.

7. Conclusion Balance is managed through a distributed neural network and there is now appreciation for a meaningful role of the cerebral cortex within this network. Certainly, the cerebral cortex does not act in isolation when influencing postural recovery but relies heavily on interactions with cortico-basal ganglia and cortico-cerebellar loops (Jacobs and Horak, 2007) in addition to the earlier mentioned spinal and brainstem righting mechanisms. The increasing volume of studies that link cognitive function and falls risk provide evidence for not only a cortical role broadly speaking, but also the specific involvement of cognitive resources in preventing falls. Although several lines of evidence support a cortical role in compensatory balance the underlying mechanisms remain unclear. Some of the key cortical regions demonstrating a role in postural responses to externally induced perturbations are shown in Table 1. As mentioned previously by Jacobs and Horak (2007) other regions such as Parietal, Temporal, and Insular cortices appear to be involved in perceiving postural vertical, therefore such areas are also likely to strongly influence postural recovery. At present however, our current understanding based upon direct measures of cortical neurophysiology in relation to postural perturbation is limited; a point reflected in the sparsely populated summary table.

D.A.E. Bolton / Neuroscience and Biobehavioral Reviews 57 (2015) 142–155

Since the recognition that corrective reactions are not simply hardwired reflexive acts research has been in progress to use this insight to either train compensatory behaviours directly (Mansfield et al., 2010, 2011) or even to incorporate gains from resistance training into reactive stepping responses (Pijnappels et al., 2008). Consequently those individuals at risk could be equipped with a greater capacity to avoid a fall. Beyond direct attempts to develop compensatory motor skills another intriguing idea is the potential use of exercise to boost cognitive performance more broadly, which may then give rise to cognitively dependent improvements in resisting falls (Liu-Ambrose et al., 2013). In general, the success of rehabilitation programs tailored toward falls depends upon targeting the specific deficit. The point has been recently put forth that overcoming cognitive impairment may indeed be the rate-limiting step in cognitively impaired individuals that are at a high risk of falling (Montero-Odasso et al., 2012). To aid in this effort of targeting therapy and developing a more comprehensive understanding for the neural control of balance future work will need to reveal the cortical mechanisms that allow us to detect instability and react appropriately. Notably, the accessibility of the cerebral cortex for neuroplastic change with intervention offers good incentive for advancing our knowledge in this particular area (Jacobs, 2014). When studying the cortical contribution to reactive balance much of this role may occur prior to a perturbation with processes designed to set the stage for fall prevention. This suggests a timeframe where some essential cortical roles may be optimally revealed. Such mechanisms are likely to become more critical in complex environments therefore this has important implications for the types of paradigms scientists and clinicians use to test balance. Along with revealing cortical mechanisms in stabilizing the body there is important work ahead to determine how we can build a greater and more sophisticated response capacity to meet the complex and varying environmental challenges so pervasive in daily life.

Acknowledgements I would like to thank W. E. McIlroy for his insightful comments on an earlier draft of this manuscript.

References Adkin, A.L., Campbell, A.D., Chua, R., Carpenter, M.G., 2008. The influence of postural threat on the cortical response to unpredictable and predictable postural perturbations. Neurosci. Lett. 435, 120–125. Adkin, A.L., Quant, S., Maki, B.E., McIlroy, W.E., 2006. Cortical responses associated with predictable and unpredictable compensatory balance reactions. Exp. Brain Res. 172, 85–93. Alibiglou, L., MacKinnon, C.D., 2012. The early release of planned movement by acoustic startle can be delayed by transcranial magnetic stimulation over the motor cortex. J. Physiol. 590, 919–936. Ambrose, A.F., Paul, G., Hausdorff, J.M., 2013. Risk factors for falls among older adults: a review of the literature. Maturitas 75, 51–61. Baker, N.L., Cook, M.N., Arrighi, H.M., Bullock, R., 2011. Hip fracture risk and subsequent mortality among Alzheimer’s disease patients in the United Kingdom, 1988–2007. Age Ageing 40, 49–54. Bares, M., Nestrasil, I., Rektor, I., 2007. The effect of response type (motor output versus mental counting) on the intracerebral distribution of the slow cortical potentials in an externally cued (CNV) paradigm. Brain Res. Bull. 71, 428–435. Beloozerova, I.N., Sirota, M.G., Orlovsky, G.N., Deliagina, T.G., 2005. Activity of pyramidal tract neurons in the cat during postural corrections. J. Neurophysiol. 93, 1831–1844. Beloozerova, I.N., Sirota, M.G., Swadlow, H.A., Orlovsky, G.N., Popova, L.B., Deliagina, T.G., 2003a. Activity of different classes of neurons of the motor cortex during postural corrections. J. Neurosci. 23, 7844–7853. Beloozerova, I.N., Zelenin, P.V., Popova, L.B., Orlovsky, G.N., Grillner, S., Deliagina, T.G., 2003b. Postural control in the rabbit maintaining balance on the tilting platform. J. Neurophysiol. 90, 3783–3793. Bolton, D.A., Misiaszek, J.E., 2012. Effects of ankle extensor muscle afferent inputs on hip abductor and adductor activity in the decerebrate walking cat. J. Neurophysiol. 108, 3034–3042.

153

Bolton, D.A., Misiaszek, J.E., 2009. Contribution of hindpaw cutaneous inputs to the control of lateral stability during walking in the cat. J. Neurophysiol. 102, 1711–1724. Bolton, D.A., Patel, R., Staines, W.R., McIlroy, W.E., 2011. Transient inhibition of primary motor cortex suppresses hand muscle responses during a reactive reach to grasp. Neurosci. Lett. 504, 83–87. Bolton, D.A., Williams, L., Staines, W.R., McIlroy, W.E., 2012. Contribution of primary motor cortex to compensatory balance reactions. BMC Neurosci. 13, 102–2202. Brandt, M.E., 1997. Visual and auditory evoked phase resetting of the alpha EEG. Int. J. Psychophysiol. 26, 285–298. Burleigh, A., Horak, F., 1996. Influence of instruction, prediction, and afferent sensory information on the postural organization of step initiation. J. Neurophysiol. 75, 1619–1628. Carlsen, A., Chua, R., Inglis, J.T., Sanderson, D.J., Franks, I.M., 2004. Prepared movements are elicited early by startle. J. Mot. Behav. 36, 253–264. Carroll, N.V., Slattum, P.W., Cox, F.M., 2005. The cost of falls among the community-dwelling elderly. J. Manage. Care Pharm. 11, 307–316. Carter, C.S., Braver, T.S., Barch, D.M., Botvinick, M.M., Noll, D., Cohen, J.D., 1998. Anterior cingulate cortex, error detection, and the online monitoring of performance. Science 280, 747–749. Chen, T.Y., Peronto, C.L., Edwards, J.D., 2012. Cognitive function as a prospective predictor of falls. J. Gerontol. B: Psychol. Sci. Soc. Sci. 67, 720–728. Cheng, K.C., Pratt, J., Maki, B.E., 2013. Effects of spatial-memory decay and dual-task interference on perturbation-evoked reach-to-grasp reactions in the absence of online visual feedback. Hum. Mov. Sci. 32, 328–342. Dietz, V., Quintern, J., Berger, W., 1984. Cerebral evoked potentials associated with the compensatory reactions following stance and gait perturbation. Neurosci. Lett. 50, 181–186. Dietz, V., Quintern, J., Berger, W., Schenck, E., 1985. Cerebral potentials and leg muscle e.m.g. responses associated with stance perturbation. Exp. Brain Res. 57, 348–354. Duckrow, R.B., Abu-Hasaballah, K., Whipple, R., Wolfson, L., 1999. Stance perturbation-evoked potentials in old people with poor gait and balance. Clin. Neurophysiol. 110, 2026–2032. Eng, J.J., Winter, D.A., Patla, A.E., 1994. Strategies for recovery from a trip in early and late swing during human walking. Exp. Brain Res. 102, 339–349. Franca, M., Turella, L., Canto, R., Brunelli, N., Allione, L., Andreasi, N.G., Desantis, M., Marzoli, D., Fadiga, L., 2012. Corticospinal facilitation during observation of graspable objects: a transcranial magnetic stimulation study. PLoS ONE 7, e49025. Fujimoto, H., Mihara, M., Hattori, N., Hatakenaka, M., Kawano, T., Yagura, H., Miyai, I., Mochizuki, H., 2014. Cortical changes underlying balance recovery in patients with hemiplegic stroke. Neuroimage 85 (Pt 1), 547–554. Gage, W.H., Zabjek, K.F., Hill, S.W., McIlroy, W.E., 2007. Parallels in control of voluntary and perturbation-evoked reach-to-grasp movements: EMG and kinematics. Exp. Brain Res. 181, 627–637. Geurts, A.C., de Haart, M., van Nes, I.J., Duysens, J., 2005. A review of standing balance recovery from stroke. Gait Posture 22, 267–281. Gibson, J.J., 1979. The Ecological Approach to Visual Perception. Houghton Mifflin, Boston, MA. Gleason, C.E., Gangnon, R.E., Fischer, B.L., Mahoney, J.E., 2009. Increased risk for falling associated with subtle cognitive impairment: secondary analysis of a randomized clinical trial. Dement. Geriatr. Cogn. Disord. 27, 557–563. Hamano, T., Luders, H.O., Ikeda, A., Collura, T.F., Comair, Y.G., Shibasaki, H., 1997. The cortical generators of the contingent negative variation in humans: a study with subdural electrodes. Electroencephalogr. Clin. Neurophysiol. 104, 257–268. Hawkins, J., Blakeslee, S., 2004. On Intelligence. Times Books, New York City. Herman, T., Mirelman, A., Giladi, N., Schweiger, A., Hausdorff, J.M., 2010. Executive control deficits as a prodrome to falls in healthy older adults: a prospective study linking thinking, walking, and falling. J. Gerontol. A: Biol. Sci. Med. Sci. 65, 1086–1092. Holroyd, C.B., Dien, J., Coles, M.G., 1998. Error-related scalp potentials elicited by hand and foot movements: evidence for an output-independent error-processing system in humans. Neurosci. Lett. 242, 65–68. Holtzer, R., Friedman, R., Lipton, R.B., Katz, M., Xue, X., Verghese, J., 2007. The relationship between specific cognitive functions and falls in aging. Neuropsychology 21, 540–548. Honeycutt, C.F., Nichols, T.R., 2010. The decerebrate cat generates the essential features of the force constraint strategy. J. Neurophysiol. 103, 3266–3273. Horak, F.B., Diener, H.C., Nashner, L.M., 1989. Influence of central set on human postural responses. J. Neurophysiol. 62, 841–853. Horak, F.B., Nashner, L.M., 1986. Central programming of postural movements: adaptation to altered support-surface configurations. J. Neurophysiol. 55, 1369–1381. Jacobs, J.V., 2014. Why we need to better understand the cortical neurophysiology of impaired postural responses with age, disease, or injury. Front. Integr. Neurosci. 8, 69. Jacobs, J.V., Fujiwara, K., Tomita, H., Furune, N., Kunita, K., Horak, F.B., 2008. Changes in the activity of the cerebral cortex relate to postural response modification when warned of a perturbation. Clin. Neurophysiol. 119, 1431–1442. Jacobs, J.V., Horak, F.B., 2007. Cortical control of postural responses. J. Neural Transm. 114, 1339–1348.

154

D.A.E. Bolton / Neuroscience and Biobehavioral Reviews 57 (2015) 142–155

Jansen, B.H., Agarwal, G., Hegde, A., Boutros, N.N., 2003. Phase synchronization of the ongoing EEG and auditory EP generation. Clin. Neurophysiol. 114, 79–85. Jones, B.E., 2008. Modulation of cortical activation and behavioral arousal by cholinergic and orexinergic systems. Ann. N.Y. Acad. Sci. 1129, 26–34. Kannus, P., Sievanen, H., Palvanen, M., Jarvinen, T., Parkkari, J., 2005. Prevention of falls and consequent injuries in elderly people. Lancet 366, 1885–1893. Kerns, J.G., Cohen, J.D., MacDonald III, A.W., Cho, R.Y., Stenger, V.A., Carter, C.S., 2004. Anterior cingulate conflict monitoring and adjustments in control. Science 303, 1023–1026. King, E.C., Lee, T.A., McKay, S.M., Scovil, C.Y., Peters, A.L., Pratt, J., Maki, B.E., 2011. Does the “eyes lead the hand” principle apply to reach-to-grasp movements evoked by unexpected balance perturbations? Hum. Mov. Sci. 30, 368–383. Ko, J.H., Monchi, O., Ptito, A., Bloomfield, P., Houle, S., Strafella, A.P., 2008. Theta burst stimulation-induced inhibition of dorsolateral prefrontal cortex reveals hemispheric asymmetry in striatal dopamine release during a set-shifting task: a TMS-[(11)C]raclopride PET study. Eur. J. Neurosci. 28, 2147–2155. Lakhani, B., Mansfield, A., Inness, E.L., McIlroy, W.E., 2011a. Compensatory stepping responses in individuals with stroke: a pilot study. Physiother. Theory Pract. 27, 299–309. Lakhani, B., Miyasike-Dasilva, V., Vette, A.H., McIlroy, W.E., 2013. ‘Priming’ the brain to generate rapid upper-limb reactions. Exp. Brain Res. 230, 261–270. Lakhani, B., Van Ooteghem, K., Miyasike-Dasilva, V., Akram, S., Mansfield, A., McIlroy, W.E., 2011b. Does the movement matter?: Determinants of the latency of temporally urgent motor reactions. Brain Res. 1416, 35–43. Lakhani, B., Vette, A.H., Mansfield, A., Miyasike-daSilva, V., McIlroy, W.E., 2012. Electrophysiological correlates of changes in reaction time based on stimulus intensity. PLoS ONE 7, e36407. Lamarche, M., Louvel, J., Buser, P., Rektor, I., 1995. Intracerebral recordings of slow potentials in a contingent negative variation paradigm: an exploration in epileptic patients. Electroencephalogr. Clin. Neurophysiol. 95, 268–276. Little, C.E., Woollacott, M., 2015. EEG measures reveal dual-task interference in postural performance in young adults. Exp. Brain Res. 233, 27–37. Liu-Ambrose, T., Nagamatsu, L.S., Hsu, C.L., Bolandzadeh, N., 2013. Emerging concept: ‘central benefit model’ of exercise in falls prevention. Br. J. Sports Med. 47, 115–117. Liu-Ambrose, T.Y., Ashe, M.C., Graf, P., Beattie, B.L., Khan, K.M., 2008. Increased risk of falling in older community-dwelling women with mild cognitive impairment. Phys. Ther. 88, 1482–1491. Luu, P., Tucker, D.M., Makeig, S., 2004. Frontal midline theta and the error-related negativity: neurophysiological mechanisms of action regulation. Clin. Neurophysiol. 115, 1821–1835. Macpherson, J.M., Everaert, D.G., Stapley, P.J., Ting, L.H., 2007. Bilateral vestibular loss in cats leads to active destabilization of balance during pitch and roll rotations of the support surface. J. Neurophysiol. 97, 4357–4367. Macpherson, J.M., Fung, J., 1999. Weight support and balance during perturbed stance in the chronic spinal cat. J. Neurophysiol. 82, 3066–3081. Macpherson, J.M., Horak, F.B., 2013. Chapter 41: Posture. In: Kandel, E.R., Schwartz, J.H., Jessell, T.M., Siegelbaum, S.A., Hudspeth, A.J. (Eds.), Principles of Neural Science. , fifth ed. McGraw-Hill, New York, NY, pp. 935–959. Magnus, R., 1926. Cameron prize lectures on some results of studies in the physiology of posture. Lancet 208, 583–632. Makeig, S., Westerfield, M., Jung, T.P., Enghoff, S., Townsend, J., Courchesne, E., Sejnowski, T.J., 2002. Dynamic brain sources of visual evoked responses. Science 295, 690–694. Maki, B.E., Cheng, K.C., Mansfield, A., Scovil, C.Y., Perry, S.D., Peters, A.L., McKay, S., Lee, T., Marquis, A., Corbeil, P., Fernie, G.R., Liu, B., McIlroy, W.E., 2008. Preventing falls in older adults: new interventions to promote more effective change-in-support balance reactions. J. Electromyogr. Kinesiol. 18, 243–254. Maki, B.E., McIlroy, W.E., 2007. Cognitive demands and cortical control of human balance-recovery reactions. J. Neural Transm. 114, 1279–1296. Maki, B.E., McIlroy, W.E., 1997. The role of limb movements in maintaining upright stance: the “change-in-support” strategy. Phys. Ther. 77, 488–507. Maki, B.E., McIlroy, W.E., Fernie, G.R., 2003. Change-in-support reactions for balance recovery. IEEE Eng. Med. Biol. Mag. 22, 20–26. Mansfield, A., Inness, E.L., Komar, J., Biasin, L., Brunton, K., Lakhani, B., McIlroy, W.E., 2011. Training rapid stepping responses in an individual with stroke. Phys. Ther. 91, 958–969. Mansfield, A., Inness, E.L., Lakhani, B., McIlroy, W.E., 2012. Determinants of limb preference for initiating compensatory stepping poststroke. Arch. Phys. Med. Rehabil. 93, 1179–1184. Mansfield, A., Peters, A.L., Liu, B.A., Maki, B.E., 2010. Effect of a perturbation-based balance training program on compensatory stepping and grasping reactions in older adults: a randomized controlled trial. Phys. Ther. 90, 476–491. Marlin, A., Mochizuki, G., Staines, W.R., McIlroy, W.E., 2014. Localizing evoked cortical activity associated with balance reactions: does the anterior cingulate play a role? J. Neurophysiol. 111, 2634–2643. McCollough, A.W., Machizawa, M.G., Vogel, E.K., 2007. Electrophysiological measures of maintaining representations in visual working memory. Cortex 43, 77–94. McIlroy, W.E., Maki, B.E., 1995. Early activation of arm muscles follows external perturbation of upright stance. Neurosci. Lett. 184, 177–180. McIlroy, W.E., Maki, B.E., 1993. Changes in early ‘automatic’ postural responses associated with the prior-planning and execution of a compensatory step. Brain Res. 631, 203–211.

McIlroy, W.E., Norrie, R.G., Brooke, J.D., Bishop, D.C., Nelson, A.J., Maki, B.E., 1999. Temporal properties of attention sharing consequent to disturbed balance. Neuroreport 10, 2895–2899. Mihara, M., Miyai, I., Hatakenaka, M., Kubota, K., Sakoda, S., 2008. Role of the prefrontal cortex in human balance control. Neuroimage 43, 329–336. Mihara, M., Miyai, I., Hattori, N., Hatakenaka, M., Yagura, H., Kawano, T., Kubota, K., 2012. Cortical control of postural balance in patients with hemiplegic stroke. Neuroreport 23, 314–319. Miltner, W.H., Braun, C.H., Coles, M.G., 1997. Event-related brain potentials following incorrect feedback in a time-estimation task: evidence for a “generic” neural system for error detection. J. Cogn. Neurosci. 9, 788–798. Miltner, W.H., Lemke, U., Weiss, T., Holroyd, C., Scheffers, M.K., Coles, M.G., 2003. Implementation of error-processing in the human anterior cingulate cortex: a source analysis of the magnetic equivalent of the error-related negativity. Biol. Psychol. 64, 157–166. Mirelman, A., Herman, T., Brozgol, M., Dorfman, M., Sprecher, E., Schweiger, A., Giladi, N., Hausdorff, J.M., 2012. Executive function and falls in older adults: new findings from a five-year prospective study link fall risk to cognition. PLoS ONE 7, e40297. Misiaszek, J.E., Krauss, E.M., 2005. Restricting arm use enhances compensatory reactions of leg muscles during walking. Exp. Brain Res. 161, 474–485. Mochizuki, G., Boe, S., Marlin, A., McIlRoy, W.E., 2010. Perturbation-evoked cortical activity reflects both the context and consequence of postural instability. Neuroscience 170, 599–609. Mochizuki, G., Sibley, K.M., Cheung, H.J., Camilleri, J.M., McIlroy, W.E., 2009. Generalizability of perturbation-evoked cortical potentials: independence from sensory, motor and overall postural state. Neurosci. Lett. 451, 40–44. Mochizuki, G., Sibley, K.M., Esposito, J.G., Camilleri, J.M., McIlroy, W.E., 2008. Cortical responses associated with the preparation and reaction to full-body perturbations to upright stability. Clin. Neurophysiol. 119, 1626–1637. Montero-Odasso, M., Verghese, J., Beauchet, O., Hausdorff, J.M., 2012. Gait and cognition: a complementary approach to understanding brain function and the risk of falling. J. Am. Geriatr. Soc. 60, 2127–2136. Moore, S.P., Rushmer, D.S., Windus, S.L., Nashner, L.M., 1988. Human automatic postural responses: responses to horizontal perturbations of stance in multiple directions. Exp. Brain Res. 73, 648–658. Mori, S., 1987. Integration of posture and locomotion in acute decerebrate cats and in awake, freely moving cats. Prog. Neurobiol. 28, 161–195. Mori, S., Kawahara, K., Sakamoto, T., Aoki, M., Tomiyama, T., 1982. Setting and resetting of level of postural muscle tone in decerebrate cat by stimulation of brain stem. J. Neurophysiol. 48, 737–748. Mori, S., Matsuyama, K., Kohyama, J., Kobayashi, Y., Takakusaki, K., 1992. Neuronal constituents of postural and locomotor control systems and their interactions in cats. Brain Dev. 14, S109–S120. Morris, J.C., Rubin, E.H., Morris, E.J., Mandel, S.A., 1987. Senile dementia of the Alzheimer’s type: an important risk factor for serious falls. J. Gerontol. 42, 412–417. Muir, S.W., Gopaul, K., Montero Odasso, M.M., 2012. The role of cognitive impairment in fall risk among older adults: a systematic review and meta-analysis. Age Ageing 41, 299–308. Nagamatsu, L.S., Hsu, C.L., Handy, T.C., Liu-Ambrose, T., 2011. Functional neural correlates of reduced physiological falls risk. Behav. Brain Funct. 7, 37–46, 37-9081-7-37. Nashner, L.M., 1977. Fixed patterns of rapid postural responses among leg muscles during stance. Exp. Brain Res. 30, 13–24. Nielsen, J.B., 2003. How we walk: central control of muscle activity during human walking. Neuroscientist 9, 195–204. Nonnekes, J., Arrogi, A., Munneke, M.A., van Asseldonk, E.H., Oude Nijhuis, L.B., Geurts, A.C., Weerdesteyn, V., 2014. Subcortical structures in humans can be facilitated by transcranial direct current stimulation. PLoS ONE 9, e107731. Norrie, R.G., Maki, B.E., Staines, W.R., McIlroy, W.E., 2002. The time course of attention shifts following perturbation of upright stance. Exp. Brain Res. 146, 315–321. Ouchi, Y., Okada, H., Yoshikawa, E., Nobezawa, S., Futatsubashi, M., 1999. Brain activation during maintenance of standing postures in humans. Brain 122 (Pt 2), 329–338. Pijnappels, M., Reeves, N.D., Maganaris, C.N., van Dieen, J.H., 2008. Tripping without falling; lower limb strength, a limitation for balance recovery and a target for training in the elderly. J. Electromyogr. Kinesiol. 18, 188–196. Pruszynski, J.A., Scott, S.H., 2012. Optimal feedback control and the long-latency stretch response. Exp. Brain Res. 218, 341–359. Quant, S., Adkin, A.L., Staines, W.R., Maki, B.E., McIlroy, W.E., 2004a. The effect of a concurrent cognitive task on cortical potentials evoked by unpredictable balance perturbations. BMC Neurosci. 5, 18. Quant, S., Adkin, A.L., Staines, W.R., McIlroy, W.E., 2004b. Cortical activation following a balance disturbance. Exp. Brain Res. 155, 393–400. Quintern, J., Berger, W., Dietz, V., 1985. Compensatory reactions to gait perturbations in man: short- and long-term effects of neuronal adaptation. Neurosci. Lett. 62, 371–376. Rankin, J.K., Woollacott, M.H., Shumway-Cook, A., Brown, L.A., 2000. Cognitive influence on postural stability: a neuromuscular analysis in young and older adults. J. Gerontol. A: Biol. Sci. Med. Sci. 55, M112–M119. Rizzolatti, G., Fadiga, L., Fogassi, L., Gallese, V., 1997. The space around us. Science 277, 190–191. Rizzolatti, G., Luppino, G., 2001. The cortical motor system. Neuron 31, 889–901.

D.A.E. Bolton / Neuroscience and Biobehavioral Reviews 57 (2015) 142–155 Safavynia, S.A., Ting, L.H., 2012. Task-level feedback can explain temporal recruitment of spatially fixed muscle synergies throughout postural perturbations. J. Neurophysiol. 107, 159–177. Schillings, A.M., Van Wezel, B.M., Duysens, J., 1996. Mechanically induced stumbling during human treadmill walking. J. Neurosci. Methods 67, 11–17. Sherrington, C.S., 1910. Flexion-reflex of the limb, crossed extension-reflex, and reflex stepping and standing. J. Physiol. 40, 28–121. Sibley, K.M., Lakhani, B., Mochizuki, G., McIlroy, W.E., 2010a. Perturbation-evoked electrodermal responses are sensitive to stimulus and context-dependent manipulations of task challenge. Neurosci. Lett. 485, 217–221. Sibley, K.M., Mochizuki, G., Esposito, J.G., Camilleri, J.M., McIlroy, W.E., 2008. Phasic electrodermal responses associated with whole-body instability: presence and influence of expectation. Brain Res. 1216, 38–45. Sibley, K.M., Mochizuki, G., Frank, J.S., McIlroy, W.E., 2010b. The relationship between physiological arousal and cortical and autonomic responses to postural instability. Exp. Brain Res. 203, 533–540. Sibley, K.M., Mochizuki, G., Lakhani, B., McIlroy, W.E., 2014. Autonomic contributions in postural control: a review of the evidence. Rev. Neurosci. 25 (5), 687–697. Sibley, K.M., Mochizuki, G., McIlroy, W.E., 2009. Perturbation-evoked electrodermal activity responds to instability, not just motor or sensory drives. Clin. Neurophysiol. 120, 619–625. Sipp, A.R., Gwin, J.T., Makeig, S., Ferris, D.P., 2013. Loss of balance during balance beam walking elicits a multifocal theta band electrocortical response. J. Neurophysiol. 110, 2050–2060. Slobounov, S., Cao, C., Jaiswal, N., Newell, K.M., 2009. Neural basis of postural instability identified by VTC and EEG. Exp. Brain Res. 199, 1–16. Smith, B.A., Jacobs, J.V., Horak, F.B., 2012. Effects of magnitude and magnitude predictability of postural perturbations on preparatory cortical activity in older adults with and without Parkinson’s disease. Exp. Brain Res. 222, 455–470. Staines, W.R., McIlroy, W.E., Brooke, J.D., 2001. Cortical representation of whole-body movement is modulated by proprioceptive discharge in humans. Exp. Brain Res. 138, 235–242.

155

Stapley, P.J., Drew, T., 2009. The pontomedullary reticular formation contributes to the compensatory postural responses observed following removal of the support surface in the standing cat. J. Neurophysiol. 101, 1334–1350. Stapley, P.J., Ting, L.H., Hulliger, M., Macpherson, J.M., 2002. Automatic postural responses are delayed by pyridoxine-induced somatosensory loss. J. Neurosci. 22, 5803–5807. Taube, W., Gruber, M., Beck, S., Faist, M., Gollhofer, A., Schubert, M., 2007. Cortical and spinal adaptations induced by balance training: correlation between stance stability and corticospinal activation. Acta Physiol. (Oxf) 189, 347–358. Taube, W., Schubert, M., Gruber, M., Beck, S., Faist, M., Gollhofer, A., 2006. Direct corticospinal pathways contribute to neuromuscular control of perturbed stance. J. Appl. Physiol. (1985) 101, 420–429. Valls-Sole, J., Rothwell, J.C., Goulart, F., Cossu, G., Munoz, E., 1999. Patterned ballistic movements triggered by a startle in healthy humans. J. Physiol. 516 (Pt 3), 931–938. Varghese, J.P., Beyer, K.B., Williams, L., Miyasike-daSilva, V., McIlroy, W.E., 2015. Standing still: is there a role for the cortex? Neurosci. Lett. 590, 18–23. Varghese, J.P., Marlin, A., Beyer, K.B., Staines, W.R., Mochizuki, G., McIlroy, W.E., 2014. Frequency characteristics of cortical activity associated with perturbations to upright stability. Neurosci. Lett. 578, 33–38. Vogel, E.K., Machizawa, M.G., 2004. Neural activity predicts individual differences in visual working memory capacity. Nature 428, 748–751. Weerdesteyn, V., Laing, A.C., Robinovitch, S.N., 2008. Automated postural responses are modified in a functional manner by instruction. Exp. Brain Res. 186, 571–580. Whelan, P.J., 1996. Control of locomotion in the decerebrate cat. Prog. Neurobiol. 49, 481–515. Wichmann, T., DeLong, M.R., 2013. Chapter 43: The basal ganglia. In: Kandel, E.R., Schwartz, J.H., Jessell, T.M., Siegelbaum, S.A., Hudspeth, A.J. (Eds.), Principles of Neural Science. , fifth ed. McGraw-Hill, New York, NY, pp. 982–998. Xu, Y., Chun, M.M., 2006. Dissociable neural mechanisms supporting visual short-term memory for objects. Nature 440, 91–95.