AMERICAN SOCIETY OF BIOMECHANICS
Presented at the Twenty-First Annual Meeting
of the American Society of Biomechanics
Clemson University, South Carolina
September 24-27, 1997
|
| AMERICAN SOCIETY OF BIOMECHANICS
Presented at the Twenty-First Annual Meeting |
Proprioceptive and visual signals are very important in maintaining proper body positions used for balance and control of movement in daily activities as well as acrobatic skills. To determine whether gymnasts are able to align their body to external axes with greater accuracy than non-gymnasts we compared their abilities to align the long trunk axis to visually identified external axes when: 1) standing (ST), suspended-upright (SU) and suspended--inverted (SI); 2) the head is aligned with the long trunk axis and tilted; and 3) allocentric cues are available or not.
It has been reported that positioning of the trunk can be used as a very good indicator of orientation ability (Hondzinski and Darling, 1993, Jakobs et al, 1985, Ross et al., 1969). These studies examined trunk alignment to either gravitational vertical or to anatomical position while horizontal. To our knowledge no studies have tested the accuracy of trunk alignment to oblique axes. In previous work we observed that divers and non-divers were very accurate at the task of aligning their trunk to the vertical whether vision was permitted or not (Hondzinski and Darling 1993). In contrast to our findings, Ross et al. (1969) observed that inverted SCUBA diver's use of vision (underwater) resulted in more accurate indication of vertical trunk alignment. Because it has been shown that visual perception of vertical is influenced by head and body orientation (Parker et al., 1983), the first purpose of this experiment was to determine how accurately subjects could align their longitudinal trunk axis to various oblique axes in three body positions and two head positions. The second purpose was to compare gymnasts to control subjects under these conditions with and without use of allocentric cues. Gymnasts were studied because their awareness of the body position relative to the gravitational vector and visual stimuli are both needed and used while performing. We hypothesized that subjects' errors in aligning the trunk to visually specified axes would be higher: (1) in the SI condition than the SU and ST conditions; (2) when the head is misaligned to the trunk; (3) when external orientation cues were not available; and (4) in controls than in gymnasts in the SI condition.
Five college gymnasts and five control subjects participated in this experiment. Infrared emitting diodes (IREDs) were attached to goggles worn by subjects, to a device that fit firmly to the trunk and to an external rod for optoelectronic recording of their orientation.
Suspension of subjects was accomplished using a tandem skydiving harness and a stable overhead spotting system, similar to that used in gymnastics training. A simple hollow metal pole, to which the subjects' legs were attached using "anti-gravity boots", was stabilized by securing its ends to ropes of the system.
A straight visible rod was presented to the subject at varying oblique angles (in three dimensions) by the experimenter. Subjects were instructed to align their longitudinal trunk axis to the rod as accurately as possible in each trial. Trials in each of the three body positions were performed with the head position in alignment with the trunk and tilted (in three dimensions) and in both darkness and in normal room lighting. In darkness subjects could view the dimly illuminated rod, but not the surrounding area. Subjects were able to complete dark trials in each condition before dark adaptation occurred. Twenty trials were tested in each of the 12 conditions.
Errors in trunk alignment to the external axes were measured in three dimensions and computed as frontal and sagittal plane errors. Variable errors (VEs), computed as the standard deviation of the signed individual trial errors in each plane in each condition, were used as a measure of random error. These were compared across group, plane and experimental condition using a repeated measures analysis of variance (ANOVA).
Surprisingly, variable errors were similar for gymnasts and controls (F1,8=0.769, p=0.4) even in the SI condition (F2,16=1.38, p=0.3) and during normally lit and darkened background environments (F1,8=1.521, p=0.25). VE data was therefore combined across groups and light/dark conditions (figure 1). Clearly, the subjects could accurately align the trunk to an external rod as indicated by the low mean VE in each condition. These low VEs show that subjects accurately transform visually specified axes into trunk orientation using kinesthetic information (proprioception, vestibular afferents and tactile senses).
Fig. 1 Mean variable trunk alignment errors are shown in both frontal and sagittal planes. ST-standing, SU-suspended upright, SI-suspended-inverted, A-head aligned, T-head tilted.
Figure 1 also shows that subjects had higher errors in the sagittal plane than in the frontal plane (F1,8=43.14, p<0.001), when the head was tilted (F1,8=47.89, p<0.001) and when suspended and inverted (F2,16=10.58, p<0.01). The errors were lower in the frontal plane probably because of the position of the eyes in humans. Left and right tilt in the frontal plane of an external object can easily be identified. It is more difficult to determine the angle of tilt in the sagittal plane because the rod projects a vertical image on the retina due to their frontal plane orientation. Differences in rod dimension must be used for determining the sagittal plane angle to align the trunk appropriately.
Greater alignment errors when suspended suggest that ground contact by the feet is important in maintaining accurate perceptions of trunk orientation. This agrees with the results of Thomas and Lyons (1966) that ground contact is important in visual awareness of the vertical in pigeons. The increased errors may also be due to the novelty of the suspended positions.
When the head and trunk are aligned, the task is presumably easier because the vestibular otoliths can be used to specify trunk orientation. Tilting the head requires incorporation of the neck angle information to determine trunk orientation, which resulted in larger perceptual errors. This is consistent with other findings that tilting the head causes greater errors in trunk alignment to either the vertical or an anatomical position when horizontal (Hondzinski and Darling, 1993, Parker et al., 1983).
Hondzinski, J.M. and Darling W.G. ASB Proceedings, 17th annual meeting 113, 1993.
Jakobs, T. et al. Exp Brain Res, 90, 129-138, 1985.
Parker, D.E. et al. Percept & Psychophys, 33, 139-146, 1983.
Ross, H.E. et al. Aerospace Med, 40, 728-732, 1969.
Thomas, D.R. et al. Percept & Psychophys, 1, 93-95, 1966.
A special Thanks to the Iowa Space Grant Consortium, Bill Morrissey, Jay Speckeen and Jeff Davidson.