Abstract
Possibilities for action depend on the fit between physical properties of the body and physical features of the environment. However, across the lifespan, both body and environmental properties are in constant flux. To guide actions adaptively, motor decisions must take these changes into account. In four experiments, we investigated how infants, young adults, and elderly adults navigated through doorways of varying width.
In Experiment 1, elderly adults (65+ years) decided whether they could walk through doorway apertures varying from 0 to 74 cm in .2 cm increments. Participants slightly overestimated their abilities, occasionally erring by attempting to squeeze through doorways a few centimeters smaller than their dynamic body width. Elderly adults' navigation strategies were idiosyncratic, and dependent on their own body and balance characteristics. In Experiment 2, we observed college-aged adults walking through doorways to compare their motor decisions with those of the elderly adults. Like the elderly, young adults occasionally attempted doorways that were slightly too small, indicating no decrement in perceptual control of action due to aging.
Experiment 3 assessed motor decisions at the other end of the lifespan. In contrast to adults, 16-month-old walking infants grossly overestimated their abilities by attempting to squeeze through impossibly small doorways. However, we wondered whether infants' errors might reflect their willingness to risk entrapment, rather than failure to perceive changing possibilities for action. In Experiment 4, we tested this possibility by removing the side wall from the doorway, creating an opening bounded by a wall on one side and a risky drop-off on the other. When the penalty for errors was falling off the walkway, infants responded cautiously by refusing to walk through impossibly small openings. Like the elderly adults, infants exhibited creative and idiosyncratic strategies for navigating through openings, reflecting sensitivity to the different environmental constraints.