Abstract: PURPOSE: Walking individuals experience pressure and friction from the ground. These forces are determined by the physical properties of the supporting surfaces and modulate postural control, balance, and gait. Thus, walkers exhibit different kinematical patterns when walking on surfaces with different physical properties. On a separate note, optical input, which provides essential information about the structures and changes in the environment, also affects walking and a blindfolded walker typically shows increased postural sway, difficulties in step initiation and execution, and impaired balance control as compared to their normal-viewing counterparts. Researchers of perception then dive in to study whether these kinematical differences could be picked up by observers to differentiate, recognize or identify walkers or walking events and their studies have yielded multifarious results (e.g., Beardsworth & Buckner, 1981; Jokisch, Daum, & Troje, 2006; Troje, & Westhoff, 2006). However, it is always assumed but never empirically measured how much kinematical difference there is, before testing the perception of it. We think it is absolutely important and necessary to measure the kinematical differences of action performance under various optical and dynamical conditions, which validates the perception studies. Using natural walking as an example, the current work recorded and compared gait and limb swing with or without optical input and on compliant or non-compliant surfaces to demonstrate the optical and dynamical effects on movement production.
METHODS: Twenty participants’ (age between 19 and 26, 10 females) natural walking was recorded using a motion capture system (Nokov, Mars 2H) in two optical conditions: with or without vision (by blindfolding the walker), and two physical conditions: walking a foam-paded surface with high compliance or on a concrete surface with low compliance. In each condition, every participant walked 8 times and each time they walked for 3 cycles (a cycle was defined as the phase between two successive heel-strikes of the right leg). Using right leg kinematics, gait parameters (stride duration, stride length, walking speed, vertical oscillation, path straightness) were calculated.
RESULTS: First, the effect of optical input. When blindfolded, participants exhibited reduced stride length, longer stride duration, slower walking speed, and less straight walking path. They also demonstrated decreased vertical bouncing and smaller range of motion on the head and joints, including the elbows, knees, and shoulders. Second, the effect of dynamical constraints. Walking on foam-paded surface, compared to walking on concrete floor, walkers showed reduced vertical bouncing, lower walking speed and smaller range of motion on the head and joints, including the elbows, and shoulders. There was an interaction between optical input and dynamical constraints and the effects of optical input on stride duration and path curvature were observed only in the high compliance foam-padded floor condition, not in the concrete floor condition.
CONCLUSIONS: The kinematics of walking revealed the dynamical interaction between the walker and the environment and reflected the availibility of optical input. These findings contribute to our understanding of the complex interplay between kinematics, optics, and dynamics during natural walking. They also make scientific inquiries about whether observers could visually detect such differences to perceive the walker, the walking event and the walking environment reasonable and justified.

Key words: optical input, walking events, postural sway, kinematical differences, dynamical constraints