How to grasp a ripe tomato

Abstract

Fortunately, we don’t have to think about this when we are standing in the supermarket after a busy day. We adjust our grip without effort, making sure we don’t squish an overripe tomato, while we firmly grasp a hard green one. This is actually a complex task in which humans are surprisingly talented. Modern computers beat grandmasters in chess, but any six year old moves the pieces with more effortless skill than the best robot. The fluent integration of abstract knowledge (“that tomato is ripe”) and spatial information (“it is there”) is fundamental to many actions that make us so characteristically human and versatile: such as using tools and making gestures. This integration is not trivial, because the processing of abstract knowledge and the control of our movements are strictly segregated in the brain. The research described in this thesis entitled “How to grasp a ripe tomato” investigates the fundamental mechanisms that underlie the interaction between abstract and spatial processing in the guidance of grasping movements. For these studies subjects were asked to grasp objects in the magnetic resonance imaging (MRI) scanner, while abstract and spatial information was manipulated. This paradigm identified the brain regions responsible for this integration and allowed the investigation of their activity and interactions. In another set of studies the neural processes in these regions were temporarily disrupted using transcranial magnetic stimulation (TMS), while concurrently the brain activity of the subjects was measured by means of electroencephalography (EEG). Previously, it was assumed that movements are primarily spatially planned (“the object is there”), and that abstract knowledge is just the icing on the cake (“but don’t squeeze too hard”). However, with this technologically innovating research Lennart Verhagen shows that abstract and spatial information are already integrated at the very beginning of the movement planning. This integration takes place in the dorsolateral parieto-frontal circuit. A brain network that is, probably not by accident, also strongly active during tool use and gesture production. In this network a coarse sketch of the intended movement is constructed (“this is again a ripe tomato that you should grasp like this”). Subsequently, this initial structure informs the accurate guidance of the movement by the dorsomedial parieto-frontal circuit. Robots don’t grasp a ripe tomato as fluently as humans do, and no other animal uses tools, let alone gestures, as easily. We do this without effort, and don’t need to think about it. This is not a matter of higher cognitive skills, but of integrating abstract knowledge in the control of our movements. This research shows that abstract information is incorporated in our behaviour by shaping the initial structure of the movement plan. In this way a framework for the intended movement is constructed that is subsequently developed in a detailed plan. Sensory, motoric, and conceptual processes are often considered either in separation, or as one and the same, but it is the interaction between these domains that forms the foundation of characteristically human movements