Doctor of Philosophy
Humans, unlike any other species, use tools to achieve complex goals. New Caledonian Crows, among the best of avian tool-makers, use twigs to retrieve food in crevices, and veined octopuses use coconut shells as shelters. Humans, however, go above and beyond these simple behaviours. Even when compared to orders that are evolutionarily closest to humans such as non-human primates, tool use is indisputably more advanced in humans. Conventionally, neuroimaging researchers who have studied complex tool use in humans do so by presenting pictures of tools and measuring the brain activity evoked by actions potentiated by the tools. This method has revealed tool-selective regions that activate in response to pictures of tools but, critically, also activate in response to real actions with real tools. Though there is overlap between regions that respond to both pictures of tools and to real tool use, it is unclear whether tool pictures are indeed an effective proxy for real tool use. In light of this, the overarching goals of this thesis were, 1) from a methodological perspective, to determine whether different proxies for studying tool use are more effective than using pictures but less technically challenging than using real actions on real tools; and 2) from a theoretical perspective, to determine what these proxies can reveal about tool-related processing, particularly in brain regions involved in visuomotor control. In sum, the results from this thesis revealed, 1) that presenting videos of familiar tool actions is an optimal proxy to study tool use, and 2) that tool-selective regions are areas selective for actions afforded by tools, for the characteristic motion associated with tools, and for familiar tools of which functional associations are well-established. Taken together, this thesis offers support to the notion that tool-selective regions process information with the purpose of predicting upcoming actions and reasoning possible ways to use a tool to interact with a target. In agreement with the affordance perspective, tool-selective regions do so even when there is no intent to act on a tool.
Macdonald, Scott N., "Characterizing Tool-Selective Areas With Human Neuroimaging" (2017). Electronic Thesis and Dissertation Repository. 4717.