No abstract

There is a long-standing proposal for the existence of two neuroanatomically and functionally separate visual systems; one supported by the dorsal pathway to control action and the second supported by the ventral pathway to handle explicit perceptual judgments. The dorsal pathway requires fast access to egocentric information, while the ventral pathway primarily requires allocentric information. Despite the evidence for functionally distinct systems, researchers have posited important interactions. This paper examines evidence to what degree the interaction becomes more important when target-identity, the perception of which is supported by the ventral stream, is verbalized during the execution of a target-directed far-aiming movement. In the experiment reported here participants hit balls toward distant targets while concurrently making explicit perceptual judgments of target properties. The endpoint of a shaft served as the target, with conditions including illusory arrow fins at the endpoint. Participants verbalized the location of the target by comparing it to a reference line and calling out “closer” or “further” while propelling the ball to the target. The impact velocity at ball contact was compared for hits toward three shafts of lengths, 94, 100, and 106 cm, with and without verbalizations and delays. It was observed that the meaning of the expressed words modulated movement execution when the verbalizations were consistent with the action characteristics. This effect of semantic content was evident regardless of target visibility during movement execution, demonstrating it was not restricted to movements that rely on visual memory. In addition to a direct effect of semantic content we anticipated an indirect effect of verbalization to result in action shifting toward the use of context-dependent allocentric information. This would result in an illusion bias on the impact velocity when the target is embedded in a Müller-Lyer configuration. We observed an ubiquitous effect of illusory context on movement execution, and not only when verbalizations were made. We suggest that the current experimental design with a far-aiming task where most conditions required reporting or retaining spatial characteristics of targets for action over time may have elicited a strong reliance on allocentric information to guide action.

By assessing the precision of gestural interactions with touchscreen targets, the authors investigate how the type of gesture, target location, and scene visibility impact movement endpoints. Participants made visually and memory-guided pointing and swiping gestures with a stylus to targets located in a semicircle. Specific differences in aiming errors were identified between swiping and pointing. In particular, participants overshot the target more when swiping than when pointing and swiping endpoints showed a stronger bias toward the oblique than pointing gestures. As expected, the authors also found specific differences between conditions with and without delays. Overall, the authors observed an influence on movement execution from each of the three parameters studied and uncovered that the information used to guide movement appears to be gesture specific.