Successful interactions within the environment are contingent upon the perceiver's ability to perceive the maximum extent over which they can perform actions, commonly referred to as action boundaries. Individuals are extremely calibrated to their action boundaries, and the perceptual system can quickly and flexibly recalibrate to changes in the size of action boundaries in the event of physiological and/or environmental changes. However, because even the most basic motor activities are subject to variability over time, the information upon which action boundaries are based must also be subject to variability. In this set of studies, we examined the effect of random and systematic variability in reaching experience on the perception of action boundaries for reaching using virtual reality. Participants were asked to estimate their reachability following experience reaching with either a long virtual arm, short virtual arm, or a virtual arm that varied in size. Overall, we found that individuals tended towards liberal estimates of their reachability; however, individuals can be influenced to be slightly more conservative after a higher percentage of short reaches. Consequently, when anticipating our reaching capability in the event of perceptual motor variability, individuals employ a liberal approach as it would result in the highest number of successful attempts.

No abstract

To successfully interact within our environment, individuals need to learn the maximum extent (or minimum) over which they can perform actions, popularly referred to as action boundaries. Because people learn such boundaries over time from perceptual motor feedback across different contexts, both environmental and physiological, the information upon which action boundaries are based must inherently be characterised by variability. With respect to reaching, recent work suggests that regardless of the type of variability present in their perceptual-motor experience, individuals favoured a liberal action boundary for horizontal reaching. However, the ways in which action boundaries are determined following perceptual-motor variability could also vary depending on the environmental context as well as the type of reach employed. The present research aimed to established whether the perceptual system utilises the same strategy for all types of reaches over different contexts. Participants estimated their overhead reachability following experience reaching with either a long or a short virtual arm, or a virtual arm that varied in length – while standing on the edge of a rooftop or standing on the ground. Results indicated that while similar strategies were used to determine action boundaries in both height- and non-height-related context, participants were significantly more conservative with their reachability estimates in the height-related context. Participants were sensitive to the probabilistic information associated with different arm’s reach they have experienced during the calibration phase, and used a weighted average of reaching experience to determine their action boundary under conditions of uncertainty.

The ability to accurately perceive the extent over which one can act is requisite for the successful execution of visually guided actions. Yet, like other outcomes of perceptual-motor experience, our perceived action boundaries are not stagnant, but in constant flux. Hence, the perceptual systems must account for variability in one’s action capabilities in order for the perceiver to determine when they are capable of successfully performing an action. Recent work has found that, after reaching with a virtual arm that varied between short and long each time they reach, individuals determined their perceived action boundaries using the most liberal reaching experience. However, these studies were conducted in virtual reality, and the perceptual systems may handle variability differently in a real-world setting. To test this hypothesis, we created a modified orthopedic elbow brace that mimics injury in the upper limb by restricting elbow extension via remote control. Participants were asked to make reachability judgments after training in which the maximum extent of their reaching ability was either unconstricted, constricted or variable over several calibration trials. Findings from the current study did not conform to those in virtual reality; participants were more conservative with their reachability estimates after experiencing variability in a real-world setting.