Eur J Appl Physiol (1991) 63:363-367
European Jouma, of
Applied
Physiology and Occupational Physiology © Springer-Verlag1991
The influence of stationary auditory fields on postural sway behaviour in man S. A. Raper and R. W. Soames Anatomy and Human Biology Group, Biomedical Sciences Division, King's College London, Strand, London WC2R 2LS, UK Accepted June 28, 1991
Summary. Postural sway behaviour was investigated in 30 young subjects (15 male and 15 female) during 60 s of erect standing, under various combinations of auditory and visual input. Sway was assessed using a standard biomechanical measuring platform, the output of which led directly to an online computer from which the following parameters were determined: mean lateral and antero-posterior sway, velocity and radius of sway, length of the sway path and area within the sway profile. A marked difference in sway behaviour between the sexes was observed, with women showing increased magnitudes of some sway parameters. Postural sway was significantly increased in conditions without visual feedback. The presence of an auditory field tends to have a destabilising influence on sway behaviour, with both the direction of the sound source and the type of auditory input being important variables. Nevertheless there appears to be no interaction between the visual and the auditory environment in the control of posture. Key words: Postural sway - Stationary auditory field Vision
Introduction Postural stabilisation is a dynamic process involving multisensory inputs and results from the continuous incorporation of motor commands formulated by the central nervous system (Mirka and Brookhart 1981). It is thought to be regulated by visual, proprioceptive and vestibular afferent inputs, which are organised in a dominantly hierarchical manner, with the vestibular system providing the fixed orientation reference (Paulus et al. 1984). Postural imbalance is then a function of the magnitude of the mismatch between the various inputs. The system of postural control once developed,
Offprint requests to: R. W. Soames
however, remains "plastic" and can be modified, for example with practice (Fearing 1924) and prolonged optical reversal of vision (Gonshor and Melvill Jones 1980). In the ontogenic development of postural control the lower-level processes mature before the higher-level ones, so that "automatic postural adjustment mechanisms" provide the majority of sway compensation, being mediated primarily by the motor-active, supportsurface inputs (Forssberg and Nashner 1982). Although the vestibular system provides the essential gravitational reference, the subsequent relative weighting between the proprioceptive, vestibular and visual inputs is context-dependent. Consequently in young children, where the vestibular system is "mature" and the visual system is still maturing, the increased infant instability observed is a function of their inability to weight the inputs correctly in accordance with their context. However, once the child starts walking, the higher levels of postural control within the hierarchy mature and instability decreases. Butterworth and Hicks (1977) and Vidal et al. (1982), however, contend that the visual system functions to maintain postural stability well before the infant learns to stand, because at this time mechanical proprioception is unable to maintain control. The fact that inappropriate visual information overrides all other sensory inputs supports this view (Lee and Aronson 1974; Bles et al. 1977). Children up to the age of 5 years continue to rely upon vision for postural stability, but from then on its importance is thought to diminish (Brandt et al. 1976). Nevertheless in adults loss of visual input, as in blindfolding, leads to increased postural sway (Lee and Aronson 1974), and a deterioration in posture (Hlavacka and Krizkova 1979). It has been suggested that the lack of visual information under such conditions leads to a reduction in the efficiency of regulating posture rather than a reduction in its precision (Seidel and Brauer 1979). However, for blindfolded subjects proprioception has been shown to acquire more importance within the control system (Odenrick and Sandstedt 1984). Information regarding head position ap-
364 pears to be an important input in the control of posture. The p r i m a r y vestibular afferents provide information about head position and movement, while proprioceptors from the trunk and neck give information about head position in relation to the body, with interaction between the vestibular and neck afferents providing a physiological b a c k g r o u n d for a freely moving head (Lund and Broberg 1983). There appears then to be constant cooperation between the visual, vestibular and proprioceptive inputs to provide for postural control, but how does the auditory system fit into this hierarchical system? Individuals exposed to noise at work have been found to have poorer postural control, i.e. greater postural sway magnitudes, than those without such exposure (Era and Heikkinen 1985). Earlier Njiokiktjien (1973) had observed that in individuals loaded with an auditory task there was a tendency for them to reduce the magnitude of their sway behaviour. The object of the present report is to determine the importance o f the static auditory environment on postural behaviour and its control. One property of h u m a n auditory perception is the ability to locate spatially a sound source using both binaural and non-binaural cues. Binaural cues allow localisation of the sound in the horizontal plane by using differences in the temporal and intensity characteristics of the sound. However, sounds in the median sagittal plane rely on non-binaural cues. According to Oldfield and Parker (1984) a "cone of confusion" exists as there are a n u m b e r of possible sound locations that give rise to the same pattern received. These same authors also report that large errors in sound localisation occur when the sound is placed anteriorly or posteriorly, but not when placed laterally and that there is no difference between the right and left sides.
from the subject at head height. Each auditory field was presented as the subjects stood with either eyes open or eyes closed. The order of presentation of the various combinations of auditory and visual fields was randomised in order to minimise carry-over effects. Prior to the recording session each subject was asked to assume a comfortable position on the platform. No restrictions were made regarding foot placement except that it was within the confines of the platform. Once comfortable, an outline of the foot placement was made so that the foot position remained constant for each subject in each of the 60 s of erect standing required for each test. Between test sessions subjects were allowed to relax, and after each set of three tests were allowed a 2-rain seated rest period. Three-way analysis of variance was conducted on the data for each sway parameter to determine the influence of vision and the direction of the sound source on its magnitude, as well as to investigate whether men and women responded differently under these conditions.
Results F r o m the three-way analysis of variance there was no interaction between the three main variables, i.e. vision, male or female groups, or direction o f the sound source, except for the path length p a r a m e t e r with the pure-tone auditory field. In this case there were significant interactions ( P < 0.01) between the direction of the sound source and vision, sex and vision and direction, sex and vision. This suggests that for this calculated parameter at least the loss of visual feedback has a profound effect. In the following presentation of results all significant differences quoted are at the 1% level unless otherwise stated.
Vision
Methods Recordings of postural sway behaviour were obtained from 30 subjects, 15 men and 15 women, aged between 19 and 30 years, under a variety of static auditory and visual field conditions. Each of the subjects who agreed to participate had no history of auditory or vestibular disease, and had their hearing checked, using standard audiometric techniques, to determine that it was within normal limits prior to their inclusion in the study. Postural sway was determined from recordings of the three orthogonal ground-reaction forces (two horizontal and one vertical) together with their associated moments about the horizontal and vertical axes, using an AMTI biomechanical-force-measuring platform (model OR6-5-1). The signals from the platform were led directly to an IBM PC, where the analogue data were digitised, stored and subsequently analysed using commercially available software (AMTI, Bedas system). From the force and moment data the following sway parameters were calculated: mean lateral and antero-posterior sway (ram), area within the sway path (mm2), mean radius of sway (mm), length of the sway path (mm) and mean velocity of sway (mm/s). The static auditory fields consisted of either silence, a pure tone (250 Hz) or general background conversation, with both the tone and background conversation being presented at an intensity of 65 dB. Four static auditory field conditions were employed by having the sound come from either in front, behind, from the right or from the left sides of the subject, via a speaker placed 0.5 m
As might be expected the absence of visual feedback significantly increased the magnitude of all calculated sway parameters irrespective of the direction of the auditory field. Table 1 shows the m e a n sway values, together with their associated standard deviations, for the combined male and female groups for both the pure tone and background conversation (voices) auditory field conditions. No differences were observed in these parameters between the pure tone and b a c k g r o u n d conversation auditory fields for either the eyes open or eyes closed conditions.
Sex N o t all sway parameters showed a significant difference between the sexes; however, where there is a difference the w o m e n exhibited a greater sway magnitude. These data are presented in Table 2 irrespective of either the direction of the auditory field or whether visual feedback was employed or not. Significant differences between the sexes were observed in lateral sway, sway radius and sway area for both the pure tone and
365 Table 1. Means and associated standard deviations (in parentheses) of the various sway parameters with (EO) and without (EC) visual feedback Sway parameter a
Auditory field Pure tone
Xm (mm) Ym (mm) Radius (mm) Length(mm) Area (mm 2) Velocity (mm s -1)
Voices
EO
EC
EO
EC
2.10 (0.83) 3.29 (1.11) 4.33 (1.33) 512.2 (120.0) 69.6 (30.2) 8.62 (2.07)
2.90 (1.19)* 4.88 (1.77)* 6.00 (t.88)* 641.6 (165.8)* 104.0 (42.5)* 10.85 (2.70)*
2.21 (0.91) 3.50 (1.43) 4.68 (1.77) 542.8 (130.6) 73.3 (33.2) 9.11 (2.25)
2.74 (1.04)* 5.08 (2.28)* 6.15 (2.08)* 670.0 (146.2)* 109.3 (45.4)* 11.26 (2.54)*
a Xm, Lateral sway; Ym, antero-posterior sway * Significant greater (P< 0.01) than with visual feedback under similar auditory fields (pure tone or voices)
Table 2. The means and standard deviations (in parentheses) of various sway parameters for the male and female groups Sway parameter a
Auditory field Pure tone
Xm (mm) Ym (mm) Radius (mm) Length(mm) Area (mm 2) Velocity (mm s -1)
Voices
Male
Female
Male
Female
2.23 (0.83) 3.97 (1.65) 4.89 (1.73) 580.9 (146.8) 79.6 (32.0) 9.77 (2.45)
2.77 (1.23)* 4.20 (1.76) 5.47 (1.99)* 572.9 (169.2) 94.0 (47.7)* 9.70 (2.84)
2.30 (0.94) 3.99 (1.54) 5.05 (1.77) 602.9 (149.9) 82.5 (34.8) 10.15 (2.55)
2.65 (1.03)* 4.59 (2.41)* 5.78 (2.28)* 609.9 (155.0) 100.1 (48.8)* 10.21 (2.70)
a Xm, Lateral sway; Ym, antero-posterior sway * Significantly greater (P
European Jouma, of
Applied
Physiology and Occupational Physiology © Springer-Verlag1991
The influence of stationary auditory fields on postural sway behaviour in man S. A. Raper and R. W. Soames Anatomy and Human Biology Group, Biomedical Sciences Division, King's College London, Strand, London WC2R 2LS, UK Accepted June 28, 1991
Summary. Postural sway behaviour was investigated in 30 young subjects (15 male and 15 female) during 60 s of erect standing, under various combinations of auditory and visual input. Sway was assessed using a standard biomechanical measuring platform, the output of which led directly to an online computer from which the following parameters were determined: mean lateral and antero-posterior sway, velocity and radius of sway, length of the sway path and area within the sway profile. A marked difference in sway behaviour between the sexes was observed, with women showing increased magnitudes of some sway parameters. Postural sway was significantly increased in conditions without visual feedback. The presence of an auditory field tends to have a destabilising influence on sway behaviour, with both the direction of the sound source and the type of auditory input being important variables. Nevertheless there appears to be no interaction between the visual and the auditory environment in the control of posture. Key words: Postural sway - Stationary auditory field Vision
Introduction Postural stabilisation is a dynamic process involving multisensory inputs and results from the continuous incorporation of motor commands formulated by the central nervous system (Mirka and Brookhart 1981). It is thought to be regulated by visual, proprioceptive and vestibular afferent inputs, which are organised in a dominantly hierarchical manner, with the vestibular system providing the fixed orientation reference (Paulus et al. 1984). Postural imbalance is then a function of the magnitude of the mismatch between the various inputs. The system of postural control once developed,
Offprint requests to: R. W. Soames
however, remains "plastic" and can be modified, for example with practice (Fearing 1924) and prolonged optical reversal of vision (Gonshor and Melvill Jones 1980). In the ontogenic development of postural control the lower-level processes mature before the higher-level ones, so that "automatic postural adjustment mechanisms" provide the majority of sway compensation, being mediated primarily by the motor-active, supportsurface inputs (Forssberg and Nashner 1982). Although the vestibular system provides the essential gravitational reference, the subsequent relative weighting between the proprioceptive, vestibular and visual inputs is context-dependent. Consequently in young children, where the vestibular system is "mature" and the visual system is still maturing, the increased infant instability observed is a function of their inability to weight the inputs correctly in accordance with their context. However, once the child starts walking, the higher levels of postural control within the hierarchy mature and instability decreases. Butterworth and Hicks (1977) and Vidal et al. (1982), however, contend that the visual system functions to maintain postural stability well before the infant learns to stand, because at this time mechanical proprioception is unable to maintain control. The fact that inappropriate visual information overrides all other sensory inputs supports this view (Lee and Aronson 1974; Bles et al. 1977). Children up to the age of 5 years continue to rely upon vision for postural stability, but from then on its importance is thought to diminish (Brandt et al. 1976). Nevertheless in adults loss of visual input, as in blindfolding, leads to increased postural sway (Lee and Aronson 1974), and a deterioration in posture (Hlavacka and Krizkova 1979). It has been suggested that the lack of visual information under such conditions leads to a reduction in the efficiency of regulating posture rather than a reduction in its precision (Seidel and Brauer 1979). However, for blindfolded subjects proprioception has been shown to acquire more importance within the control system (Odenrick and Sandstedt 1984). Information regarding head position ap-
364 pears to be an important input in the control of posture. The p r i m a r y vestibular afferents provide information about head position and movement, while proprioceptors from the trunk and neck give information about head position in relation to the body, with interaction between the vestibular and neck afferents providing a physiological b a c k g r o u n d for a freely moving head (Lund and Broberg 1983). There appears then to be constant cooperation between the visual, vestibular and proprioceptive inputs to provide for postural control, but how does the auditory system fit into this hierarchical system? Individuals exposed to noise at work have been found to have poorer postural control, i.e. greater postural sway magnitudes, than those without such exposure (Era and Heikkinen 1985). Earlier Njiokiktjien (1973) had observed that in individuals loaded with an auditory task there was a tendency for them to reduce the magnitude of their sway behaviour. The object of the present report is to determine the importance o f the static auditory environment on postural behaviour and its control. One property of h u m a n auditory perception is the ability to locate spatially a sound source using both binaural and non-binaural cues. Binaural cues allow localisation of the sound in the horizontal plane by using differences in the temporal and intensity characteristics of the sound. However, sounds in the median sagittal plane rely on non-binaural cues. According to Oldfield and Parker (1984) a "cone of confusion" exists as there are a n u m b e r of possible sound locations that give rise to the same pattern received. These same authors also report that large errors in sound localisation occur when the sound is placed anteriorly or posteriorly, but not when placed laterally and that there is no difference between the right and left sides.
from the subject at head height. Each auditory field was presented as the subjects stood with either eyes open or eyes closed. The order of presentation of the various combinations of auditory and visual fields was randomised in order to minimise carry-over effects. Prior to the recording session each subject was asked to assume a comfortable position on the platform. No restrictions were made regarding foot placement except that it was within the confines of the platform. Once comfortable, an outline of the foot placement was made so that the foot position remained constant for each subject in each of the 60 s of erect standing required for each test. Between test sessions subjects were allowed to relax, and after each set of three tests were allowed a 2-rain seated rest period. Three-way analysis of variance was conducted on the data for each sway parameter to determine the influence of vision and the direction of the sound source on its magnitude, as well as to investigate whether men and women responded differently under these conditions.
Results F r o m the three-way analysis of variance there was no interaction between the three main variables, i.e. vision, male or female groups, or direction o f the sound source, except for the path length p a r a m e t e r with the pure-tone auditory field. In this case there were significant interactions ( P < 0.01) between the direction of the sound source and vision, sex and vision and direction, sex and vision. This suggests that for this calculated parameter at least the loss of visual feedback has a profound effect. In the following presentation of results all significant differences quoted are at the 1% level unless otherwise stated.
Vision
Methods Recordings of postural sway behaviour were obtained from 30 subjects, 15 men and 15 women, aged between 19 and 30 years, under a variety of static auditory and visual field conditions. Each of the subjects who agreed to participate had no history of auditory or vestibular disease, and had their hearing checked, using standard audiometric techniques, to determine that it was within normal limits prior to their inclusion in the study. Postural sway was determined from recordings of the three orthogonal ground-reaction forces (two horizontal and one vertical) together with their associated moments about the horizontal and vertical axes, using an AMTI biomechanical-force-measuring platform (model OR6-5-1). The signals from the platform were led directly to an IBM PC, where the analogue data were digitised, stored and subsequently analysed using commercially available software (AMTI, Bedas system). From the force and moment data the following sway parameters were calculated: mean lateral and antero-posterior sway (ram), area within the sway path (mm2), mean radius of sway (mm), length of the sway path (mm) and mean velocity of sway (mm/s). The static auditory fields consisted of either silence, a pure tone (250 Hz) or general background conversation, with both the tone and background conversation being presented at an intensity of 65 dB. Four static auditory field conditions were employed by having the sound come from either in front, behind, from the right or from the left sides of the subject, via a speaker placed 0.5 m
As might be expected the absence of visual feedback significantly increased the magnitude of all calculated sway parameters irrespective of the direction of the auditory field. Table 1 shows the m e a n sway values, together with their associated standard deviations, for the combined male and female groups for both the pure tone and background conversation (voices) auditory field conditions. No differences were observed in these parameters between the pure tone and b a c k g r o u n d conversation auditory fields for either the eyes open or eyes closed conditions.
Sex N o t all sway parameters showed a significant difference between the sexes; however, where there is a difference the w o m e n exhibited a greater sway magnitude. These data are presented in Table 2 irrespective of either the direction of the auditory field or whether visual feedback was employed or not. Significant differences between the sexes were observed in lateral sway, sway radius and sway area for both the pure tone and
365 Table 1. Means and associated standard deviations (in parentheses) of the various sway parameters with (EO) and without (EC) visual feedback Sway parameter a
Auditory field Pure tone
Xm (mm) Ym (mm) Radius (mm) Length(mm) Area (mm 2) Velocity (mm s -1)
Voices
EO
EC
EO
EC
2.10 (0.83) 3.29 (1.11) 4.33 (1.33) 512.2 (120.0) 69.6 (30.2) 8.62 (2.07)
2.90 (1.19)* 4.88 (1.77)* 6.00 (t.88)* 641.6 (165.8)* 104.0 (42.5)* 10.85 (2.70)*
2.21 (0.91) 3.50 (1.43) 4.68 (1.77) 542.8 (130.6) 73.3 (33.2) 9.11 (2.25)
2.74 (1.04)* 5.08 (2.28)* 6.15 (2.08)* 670.0 (146.2)* 109.3 (45.4)* 11.26 (2.54)*
a Xm, Lateral sway; Ym, antero-posterior sway * Significant greater (P< 0.01) than with visual feedback under similar auditory fields (pure tone or voices)
Table 2. The means and standard deviations (in parentheses) of various sway parameters for the male and female groups Sway parameter a
Auditory field Pure tone
Xm (mm) Ym (mm) Radius (mm) Length(mm) Area (mm 2) Velocity (mm s -1)
Voices
Male
Female
Male
Female
2.23 (0.83) 3.97 (1.65) 4.89 (1.73) 580.9 (146.8) 79.6 (32.0) 9.77 (2.45)
2.77 (1.23)* 4.20 (1.76) 5.47 (1.99)* 572.9 (169.2) 94.0 (47.7)* 9.70 (2.84)
2.30 (0.94) 3.99 (1.54) 5.05 (1.77) 602.9 (149.9) 82.5 (34.8) 10.15 (2.55)
2.65 (1.03)* 4.59 (2.41)* 5.78 (2.28)* 609.9 (155.0) 100.1 (48.8)* 10.21 (2.70)
a Xm, Lateral sway; Ym, antero-posterior sway * Significantly greater (P
