Manipulating peripheral visual information in manual aiming: Exploring the notion of specificity of learning

The purpose of this study was to determine the effects of the addition and removal of peripheral vision in a simple manual aiming task. In Experiment 1, subjects (n = 5) practised a manual aiming task that emphasised the central visual information available during the final homing-in phase, for a moderate number of trials in either a target-only or normal-vision condition. Following the practice phase, subjects were transferred to the other condition. Spatial accuracy data indicated that the subjects' error increased following the removal and addition of the relevant central visual information. In Experiment 2, the task constraints were manipulated such that the relevant visual information was available through the peripheral retina only. Subjects (n = 5) practised the task for a moderate and extensive number of trials in either a target-only or normal-vision condition. Following each practice phase, subjects were transferred to the other condition. Spatial accuracy data indicated that while subjects' error increased following the removal of peripheral vision, there were no detrimental effects following its addition. Spatial error was reduced when transferring to the normal-vision condition. Kinematic data indicated that, on transferring to the normal-vision condition, subjects adapted their response to exploit the added visual information, as evidenced by an increased number of movement corrections in the deceleration phase. The positive effects of the addition of peripheral vision are not consistent with the notion of increasing specificity to the information available during practice. (C) 1998 Elsevier Science B.V. All rights reserved.