Elsevier

Gait & Posture

Volume 16, Issue 1, August 2002, Pages 1-14
Gait & Posture

Review Article
Attention and the control of posture and gait: a review of an emerging area of research

https://doi.org/10.1016/S0966-6362(01)00156-4Get rights and content

Abstract

Research on the relationship between attention and the control of posture and gait is a new and expanding area with studies on young adults revealing the role of cognitive factors in the control of balance during standing and walking. The use of dual task paradigms to examine the effect of age related changes in attentional requirements of balance control and age-related reductions in stability when performing a secondary task has shown that these are important contributors to instability in both healthy and balance-impaired older adults. The attentional demands of balance control vary depending on the complexity of the task and the type of secondary task being performed. New clinical assessment methods incorporating dual-task paradigms are helpful in revealing the effect of disease (e.g. Parkinson's disease) on the ability to allocate attention to postural tasks and appear to be sensitive measures in both predicting fall risk and in documenting recovery of stability.

Introduction

Postural control has been defined as the control of the body's position in space for the purposes of balance and orientation [1], [2]. It has traditionally been considered an automatic or reflex controlled task, suggesting that postural control systems use minimal attentional resources. However, recent research, has provided evidence against this assumption. These studies suggest that there are significant attentional requirements for postural control, and that these requirements vary depending on the postural task, the age of the individual and their balance abilities.

This paper will review some of the research related to four aspects of attention and the control of posture and gait. We will examine (1) research exploring the relationship between attention, posture and gait in young adults; (2) research investigating the effect of aging on this relationship; (3) research on attention and postural control in balance-impaired fall prone elders; and (4) clinical studies that have begun to apply attention and postural control research to understanding balance and gait problems in specific patient populations. We begin by defining attention and discussing some of the limitations and controversies regarding the dual task paradigm, the primary approach to studying the relationship between attention and postural control.

Attention will be defined here as the information processing capacity of an individual. An assumption regarding this information processing capacity is that it is limited for any individual and that performing any task requires a given portion of capacity. Thus, if two tasks are performed together and they require more than the total capacity, the performance on either or both deteriorates [1], [2], [3], [4].

Research for studying attention and posture control has used dual task paradigms in which postural control (considered the primary task) and a secondary task were performed together. The extent to which the performance on either task declined indicated the interference between the processes controlling the two tasks, and thus the extent to which the two tasks shared attentional resources [5].

The use of a dual task paradigm to study attention and postural control is not without controversy. Some researchers have stated that using a dual task paradigm to study attentional demands of postural control requires that changes in performance must be limited to the secondary task with no changes occurring in the primary (postural) task [6], [7]. Thus results focus on discussing changes in the secondary task, and in this way, attentional demands associated with changes in postural tasks are clearly identified.

In contrast, other authors have examined performance changes in both tasks during a dual task experiment [1], [2], [8], [9], [10]. In these experiments, the dual task design is used both to (1) examine the attentional demands of postural tasks (inferred by changes in the secondary cognitive task); and (2) to examine the effects of performing an attentionally demanding cognitive task on the control of posture. In the latter case, postural control in effect becomes the secondary task, subject to change during the performance of a concurrent task.

It has been suggested that studies showing interacting effects of primary and secondary tasks should not be used to infer attentional demands associated with postural control [11]. It is our belief that these studies are helpful in showing the attentional demands of postural control; however, they are limited in their ability to clarify the exact attentional cost of postural tasks, because of the interacting effects between the two tasks.

Kerr et al. [5] published the first article to demonstrate the attentional demands of stance postural control in young adults. In their experiment, Kerr et al. [5] hypothesized that a difficult balance task would interfere with a spatial (visual) memory but not a verbal memory task, since postural control required visual/spatial processing. Thus, they suggested, there would be visual interference between the postural task and the visual spatial task. They asked 24 young adults to stand with blindfolds on in a tandem Romberg position as their postural task. Changes in postural control were determined by using force platform measures to calculate center of pressure. The visual/spatial cognitive task was the Brooks spatial memory task that involved placing numbers in imagined matrices and then remembering the position of these numbers. The nonspatial verbal memory task involved remembering similar sentences. They found that performing the memory task with the concurrent balance task caused an increase in the number of errors in the spatial but not the nonspatial memory task. The difference in mean number of errors in spatial and nonspatial task in sitting versus when performing the balancing task is shown in Fig. 1. There was no significant difference in postural sway during the performance of either cognitive task. The authors concluded that in young adults, postural control is attentionally demanding; however not all cognitive tasks affect postural control in the same way.

Kerr et al.'s study was the first to show that stance postural control was attentionally demanding in young adults. The next study we review examines whether attentional demands vary as a function of the type of postural task being performed. In this study by Lajoie et al. [12], young adults were asked to perform an auditory reaction time task while sitting, standing with a normal versus reduced base of support and during walking (single vs double support phase). Results from this study indicated that reaction times were fastest for sitting, and slowed for the standing and walking tasks. Reaction times were slower when subjects stood with a narrow base of support compared to a normal base of support. In addition, reaction times were slowest in the single support phase compared to the double support phase of the step cycle.

The authors reported that there was no change in the gait cycle associated with performance of the secondary task. They concluded that postural control is attentionally demanding and demands increase with the complexity of the postural task being performed. The authors noted that a limitation of their study was the use of a simple relatively non-attentionally demanding secondary task. They suggested that use of a more complex secondary task might have resulted in interference with balance and gait.

Lajoie et al.'s study focused on examining the attentional demands associated gait, reporting no change in gait parameters associated with a secondary task in young adults. In contrast, Ebersbach et al. [6] specifically studied the effect of concurrent tasks on the control of gait. Ten young adults (age 25–42 years) walked on a 10 m conductive walkway that allowed the recording of foot contact during gait. Gait parameters (stride time, double support time) were measured under a single task condition (walking without a concurrent task) and four dual task conditions presented in random order: (1) memory-retention task (digit span recall); (2) fine motor task (opening and closing a coat button continuously during gait); (3) a combination task (digit recall and buttoning task); and (4) finger tapping at 5 Hz or faster. The only dual task condition that produced a significant decrease in stride time (increased stride frequency) was finger tapping. The other gait parameter measured, double support time was significantly affected when the fine motor and memory tasks were performed synchronously with the walking; no other dual task condition affected this parameter. Interestingly, the authors noted that performance of the gait task did affect the digit recall task. The mean digit span recall was 6.7 (range 6–8) during quiet stance, but reduced to 5.8 (range 4–8) during gait. In this study, even the significant changes in gait parameters are fairly small, again suggesting that performance of multiple tasks during a relatively simple task such as unperturbed gait does not present a significant threat to stability in healthy young adults.

One interesting question regarding the attentional requirements of postural control is the time course of attentional demands associated with the recovery of stability. McIlroy et al. [13] hypothesized that the processing requirements of postural control vary during the time course of recovery of stability; therefore the attentional demands would also vary. To test their hypothesis McIlroy and colleagues examined attentional demands continuously during what they refer to as a ‘seated balancing task’. Six young adults were tested under three task conditions: seated balancing task, visuomotor tracking task, and a dual task condition involving simultaneous performance of the visuomotor tracking and balance tasks (both perturbed and non perturbed). In the balance task the feet of seated subjects were strapped to a foot pedal that controlled the rotation of an inverted pendulum that was free to rotate about the ankle axis in the sagittal plane. Subjects were required to maintain the upright position of the pendulum. During some of the trials perturbations were given to the pendulum resulting in a forward pendulum rotation. EMG signals from the soleus and tibialis anterior were recorded. The secondary task in this dual task design was a visuomotor tracking task that required the subjects to track a moving target on a computer screen. The root mean square tracking error was recorded. The results found a disruption in the visuomotor tracking task (as indicated most often by a temporary pause in the tracking) that occurred in 86% of the perturbed trials. The onset of the pause in tracking occurred on average 325±93 ms after the onset of the perturbation. In contrast the balance reaction recorded in the tibialis anterior EMG occurred at an average latency of 90±32 ms. In addition, tracking error was greater during the restoration of the pendulum to an equilibrium position. The authors concluded that in the dual task condition attention was substantially diverted from the visuomotor task when balance was perturbed, presumably redirected to the control of the compensatory response required to restabilize the inverted pendulum. From these results they suggest that balance control involves three distinct phases, each with distinct attentional requirements. The initial phase is automatic with minimal attentional demands; in a second phase, occurring 200–300 ms after the perturbation, there is an attentional shift completely away from the secondary task reflecting attentionally demanding balance control; finally there is a period of divided attention control between both the balance and the secondary task and this persists until equilibrium is restored. This can take up to several seconds.

The idea that attentional demands vary during the time course of recovery of stability is very intriguing. However, it is not clear whether results of this study can generalize to the control of human stance posture. The task of maintaining the equilibrium of an inverted pendulum, which McIlroy and colleagues refer to as a ‘balance task’ may not pertain to the systems controlling stability in independent stance.

The research does raise the intriguing issue of whether all aspects of postural responses are equally attentionally demanding. It is possible that the initial phase of compensatory responses are more automatic than later portions of the response, hence not as susceptible to the influences of additional cognitive demands. Some support for this comes from Rankin et al. [10] study, discussed in the next section of this paper, which examined early and late changes in the organization of postural muscle activity under dual task conditions in young and healthy older adults.

The next two papers reviewed raise questions regarding the interpretation of results from dual task research on attentional demands on postural control. Maki and McIlroy [14] explored the influence of both attention and arousal on postural control during dual task conditions. Stance postural control (center of pressure-COP) was measured while subjects (39 healthy young adults) performed four different conditions: no secondary task, (2) white noise, (3) listening to a spoken word recording of a book excerpt, (4) counting backwards silently by 7's. The white noise task was used to increase arousal, while the listen task was intended to divert attention without increasing arousal, and the math task was intended to increase both arousal and attention. Arousal was monitored using measures of skin conductance; in addition subjects’ state anxiety was determined using questionnaires. Results showed that the nature of the secondary task had an effect on arousal, with skin conductance being highest on the math task. Task related changes in postural control were limited to those subjects who had higher than average anxiety scores. These subjects leaned further forward during the math task. When skin conduction scores were used as a covariate in the analysis of anterior posterior center of pressure, the dual task effect on postural control during the math task was substantially reduced. The authors conclude that physiological arousal may be a potential confounder when attempting to understand the influence of attention on postural control.

A limitation of this study is that the math task increased arousal in half of the subjects, and this raises questions as to the degree to which this study may be generalized to other dual task studies. In addition, the only change in postural control that was noted was a 2% lean in the forward direction from the vertical position. It is not clear whether this slight shift in posture has any functional significance.

Yardley et al. [15] also question assumptions regarding the role of attentional demands on postural control in dual task research. They investigated the possibility that changes in postural sway seen when subjects performed a spoken task were due to perturbations to posture associated with the task of articulation, rather than due to competing demands for attentional resources. Their subjects included 36 healthy young adults with no history of balance system dysfunction. The postural task involved standing on a force platform; center of pressure was monitored and sway path (total distance in meters traversed by the COP during each 30 s trial) was measured. Twelve subjects were tested on a static forceplate, while 24 subjects were tested on an unstable surface (air filled circular rubber tube, inflated and attached to a piece of board resting on the force plate). Subjects were tested under three visual conditions, no vision, static visual image, and a moving visual image. The secondary tasks used included: counting backwards out loud (attention and articulation), silent counting backwards (attention without demands for articulation), number repetition (articulation alone), and no concurrent task. Subjects were asked to perform the secondary task as accurately and rapidly as possible even when trying to balance. Results showed that the postural tasks did not have an effect on performance of the secondary tasks. The secondary tasks did however affect postural sway. When standing on a stable surface, (shown in Fig. 2), the articulation task significantly increased postural sway; in contrast silent counting had no effect. When standing on an unstable surface, sway was impacted by articulation and visual conditions, but not by attentional load (silent counting). The authors conclude that the increased instability produced by the spoken mental arithmetic task was due primarily to the effects of articulation rather than mental activity. They suggest that the effect of articulation is possibly mediated by respiratory activity involved in speech that directly perturbs posture. However, they note, that instability could also be the result of central interference since speech and balance may share common structures.

The above research suggests that in young adults postural control may be attentionally demanding. These effects however, appear to be small, until the postural system is quite stressed and subjects are required to perform fairly complex secondary tasks. This is an emerging area of research and thus has many unanswered questions. Understanding the role of potential confounders (for example arousal and articulation) to the relationship between attention and postural control is very important.

Section snippets

Age-related changes in attention and posture control

In contrast to young adults, many researchers have found significant attentional demands associated with postural control in older adults, even under relatively simple conditions. In order to determine the age related changes in the relationship between attention and postural control, researchers have (1) compared declines in secondary task performance during the simultaneous performance of a postural task in older versus young adults; and (2) examined the effect of increasing cognitive demands

Contributions of attentional factors to balance impairments in older adults

Historically, age-related deterioration in balance abilities has been attributed to decreases in sensory or motor system function. However, parallel research in the area of cognition suggests that there may be other intrinsic sources of instability, including attentional allocation deficits. The focus of recent research studies is whether the age-related reductions in stability in many older adults are increased significantly with added cognitive demands. For example, research suggests that

Clinical research related to attention and postural control

Several studies have used a dual task paradigm to study the effects of a concurrent task on gait in individuals with PD [23], [24], [25]. Camicioli et al. [23] examined the effects of a simultaneous verbal fluency task on gait in individuals with PD with and without freezing of gait. Freezing in gait is defined as an unexpected and uncontrollable sudden halting during gait, and is a problem for many persons with PD because it impairs mobility, causes falls, and is poorly controlled by

Summary

In summary, research examining the relationship between attention and the control of posture and gait is a relatively new, but rapidly expanding area. Studies involving young adults are increasing our understanding of the role of cognitive factors in the control of stability during activities such as standing and walking. Studies using dual task paradigms to examine the effect of age related changes in attentional requirements of balance control and age-related reductions in stability when

References (34)

  • J.M. Bond et al.

    Goal-directed secondary motor tasks: their effects on gait in subjects with Parkinson disease

    Arch. Phys. Med. Rehabil.

    (2000)
  • L. Lundin-Olsson et al.

    Stops walking when talking as a predictor of falls in elderly people

    Lancet

    (1997)
  • A. Shumway-Cook et al.

    Motor Control: Theory and Practical Applications

    (2000)
  • A. Shumway-Cook et al.

    Attentional demands and postural control: the effect of sensory context

    J. Gerontol.

    (2000)
  • O. Neumann

    Automatic processing: a review of recent findings and a plea for an old theory

  • C.D. Wickens

    Attention and skilled performance

  • B. Kerr et al.

    Cognitive spatial processing and the regulation of posture

    J. Exp. Psychol.

    (1985)
  • G. Ebersbach et al.

    Influence of concurrent tasks on gait: a dual-task approach

    Percept. Mot. Skills

    (1995)
  • N. Teasdale et al.

    On the cognitive penetrability of postural control

    Exp. Aging Res.

    (1993)
  • E.A. Maylor et al.

    Age differences in postural stability are increased by additional cognitive demands

    J. Gerontol.

    (1996)
  • A. Shumway-Cook et al.

    The effects of cognitive demands on postural sway in elderly fallers and non-fallers

    J. Gerontol.

    (1997)
  • J.K. Rankin et al.

    A neuromuscular analysis of the influence of a cognitive task on postural stability in young and older adults.

    J. Gerontol. A Biol. Sci. Med. Sci.

    (2000)
  • B. Abernethy

    Dual-task methodology and motor skills research: some applications and methodological constraints

    J. Hum. Mov. Stud.

    (1988)
  • Y. Lajoie et al.

    Attentional demands for static and dynamic equilibrium

    Exp. Brain Res.

    (1993)
  • W.E. McIlroy et al.

    Temporal properties of attention sharing consequent to disturbed balance

    NeuroReport

    (1999)
  • B.E. Maki et al.

    Influence of arousal and attention on the control of postural sway

    J. Vestib. Res.

    (1996)
  • L. Yardley et al.

    Effect of articulatory and mental tasks on postural control

    NeuroReport

    (1999)
  • Cited by (1918)

    View all citing articles on Scopus
    View full text