Sex differences in the association of postural control with indirect measures of body representations

Abstract

Besides anthropometric variables, high-order body representations have been hypothesised to influence postural control. However, this has not been directly tested before. Moreover, some studies indicate that sex moderates the relationship of anthropometry and postural control. Therefore, as a proof of concept we investigated the association of body representations with postural control as well as the influence of participants' sex/gender. Body image measures were assessed with a figural drawing task. Body schema was tested by a covert and an overt task. Body sway was measured during normal bipedal quiet standing with eyes closed (with/without neck extended). Statistical analysis consisted of hierarchical multiple linear regressions with the following regression steps: (1) sensory condition, (2) sex/gender, (3) age, (4) anthropometry, (5) body schema, (6) body image, (7) sex/gender-interactions. Across 36 subjects (19 females), body schema was significantly associated with body sway variability and open-loop control, in addition to commonly known influencing factors, such as sensory condition, gender, age and anthropometry. While in females, also body image dissatisfaction substantially was associated with postural control, this was not the case in males. Sex differences and possible causes why high-order body representations may influence concurrent sensorimotor control of body sway are discussed.

Figure 1

Regression results (entry method) for whole group (a) and for gender groups (b). $\mathrm{\Delta }$ R${}^2$ for each hierarchical regression step. SC = sensory condition (eyes closed (EC); neck extended, eyes closed (NE-EC)); Gender (male; female); Anthr. = anthropometry (height (H); weight (W)); BS = body schema (${\mathit{TaP}}_{\mathrm{v}}$; ${\mathit{LAT}}_{\mathrm{a}}$); BI = body image (${\text{BIDS}}_{\mathrm{abs}}$; ${\text{BID}}_{\mathrm{abs}}$); gender-interactions: Hg, Wg, TaPg, LATg, BIDSg, BIDg. Bold with star indicates significance ($P\le 0.05$).

Figure 2

Gender-interactions: partial regression plots of absolute body image dissatisfaction (a) and absolute body image distortion (b).

Figure 3

Best regression plots with minimal number of predictors. Sway variability (SD CoP): upper row (a–c); Short-term stochastic activity (${\text{D}}_{\mathrm{s}}$): lower row (d–f); whole group (a, d): gender-interactions (${}_{\mathrm{gen}}$): for males (m), e.g. ${\text{BIDS}}_{\mathrm{gen}}=0$; for females (f), e.g. ${\text{BID}}_{\mathrm{gen}}=0$; males (b, e); females (c, f).

Figure 4

Experimental procedure. Balance: bipedal quiet standing with eyes closed (EC) and neck extended (NE-EC).

Figure 5

Taking-a-pasture task: four poses of different complexity (change in numbers of limbs involved) with Vicon full-body plugin-gait marker set and goggles to restrict vision of own body. (a) One upper extremity (non-dominant); (b) one upper and one lower extremity (both non-dominant); (c) two upper extremities and one lower extremity (non-dominant: arm above head); (d) all extremities explicitly involved in posture (non-dominant: arm facing head, foot extended); (e) button placement.

Figure 6

Laterality task; (a) top: 8 orientations/stimuli presented, examplary for FLAT; a bottom: task procedure, N and M represent the keyboard keys to press for HLAT, which corresponded to left and right pedal for FLAT; (b) setup during FLAT; (c) setup during HLAT.