Dorsal and ventral stream contributions to goal-directed actions

Abstract

Goal-directed actions are of vital importance for our everyday life. Yet, their underlying mechanisms and neuronal correlates are still under debate. Two anatomically informed models try to integrate a variety of neurophysiological and functional descriptions of frontal, parietal and temporal areas: the idea of distinct fronto-parietal channels of reach vs. grasp motor control and the two visual stream hypothesis associating occipito-parietal processing with visuomotor control and occipito-temporal processing with visual perception. We addressed three controversial topics in the context of these two models. We investigated the lateralization of online control of visually-guided reaching and grasping in humans using an fMRI paradigm. The two channel hypotheses would suggest that corrections in grasping should be anatomically distinguishable from corrections in reaching. Our main finding was an increased coupling between the hemispheres when fast movement corrections were required. A specific increase of functional connectivity within the ipsilateral hemisphere without corresponding contralateral activation increases during movement corrections, suggested that activations of the ipsilateral PPC are of functional importance for visually-guided actions. Furthermore, the connectivity analysis demonstrated changes in inter-regional coupling between the reaching and grasping networks during grip perturbations but no difference between reaching and grasping when those actions were matched in difficulty, arguing against an effector specificity of different cortical channels during online control. Lesions in the posterior parietal cortex can cause optic ataxia, which is defined as a reaching deficit to visual targets in the periphery. While such modality-specificity is essential for the definition of optic ataxia, comparisons of reaching accuracy across modalities have rarely been conducted. We investigated the potential multimodality of optic ataxia in two patients who both showed the typical misreaching in the periphery for the visual modality. Reaching to auditory targets differed significantly from reaching to visual targets for both patients, arguing against an effect of optic ataxia on auditory-guided reaching. Reaching to proprioceptive targets was unimpaired in one patient, but impaired towards nonfoveated targets in line with optic ataxia. However, this misreaching for proprioceptive targets was observed for the whole hemifield but did not increase with eccentricity as observed for visually-guided reaching. Thus, we propose that optic ataxia is unimodal but misreaching to targets in other modalities may co-occur resembling optic ataxia. Finally, we examined the role of occipito-temporal regions in memory-guided reaching in a stroke patient suffering from lateralized visual form agnosia. In agreement with the only previous examination of memory-guided reaching in a patient suffering from visual form agnosia (David Milner, Dijkerman, & Carey, 1999), reaching to visual targets was unimpaired. In contrast, the patient showed deficits when reaching to memorized targets in the contralesional hemifield. In contrast to existing studies, we excluded working memory or short-term memory deficits that may account for the observed misreaching. A second experiment using a delayed localization task suggested that the misreaching during memory-guided reaching is associated with visuomotor processing, but not with purely perceptual deficits.