Antipointing: perception-based visual information renders an offline mode of control.

Recent work by our group reported that mirror-symmetrical reaching movements (i.e., antipointing) are supported by perception-based visual information (Heath et al. 2009a). The present study was designed to determine if the perception-based visual information supporting antipointing results in a primarily offline mode of control; that is, a mode of control wherein the response unfolds with few-if any-online limb corrections. Participants reached directly to (propointing) or mirror-symmetrical (antipointing) to targets presented in the left and right visual fields. To examine the extent reaches were controlled online versus offline, we computed the proportion of variance (R (2)) explained by the position of the limb at 75% of movement time relative to the response's ultimate movement endpoint. The endpoints for propointing in left and right visual fields demonstrated robust endpoint accuracy and stability and were tied to low R (2) values. In contrast, antipointing elicited a marked degree of endpoint variability and were characterized by a visual-field specific pattern of endpoint bias. Moreover, the R (2) values for antipointing were more robust than propointing counterparts. Therefore, we propose that the metrical visual information supporting stimulus-driven propointing results in a primarily online mode of control whereas the relative nature of the perception-based visual information supporting antipointing results in an offline mode of control. More directly, we propose that the offline control of antipointing is attributed to the "slow" cognitive processing of visuoperceptual networks.