| Sumario: | <p>Humans have a remarkable capacity for generalizing experiences to novel situations. In a brain network that has expanded the most during mammalian evolution, known as the default mode network, grid cells and place cells organize spatial representations into mental maps. The aim of this thesis is to use non-invasive functional magnetic resonance imaging and computational analyses to test how well-understood codes at the single-cell level in animals also generalize to complex human behaviours.</p> <p>We first asked how brain mechanisms used for spatial navigation help us understand non- spatial cognition. We designed an equivalent computer task to the one used for real spatial navigation, but in an abstract space. Next, we tested for hexagonal signals using the same analyses as those developed for navigation in physical space. We found hexagonal grid- like signals in the same brain network for navigating in conceptual space and in physical space. This suggests that the brain organizes concepts into a mental map, allowing conceptual relationships to be navigated in a manner similar to that of space.</p> <p>Next, we aimed to develop a new analysis method that is closer to electrophysiology, and thus could detect gradients in place and grid-cell coding of physical space with neuroimaging. We show that our analysis can detect such gradients in simulated data. Moreover, using a big data approach, we found behavioural evidence that humans form precise maps of our environment. We also show preliminary evidence of neural patterns at submillimeter resolution, suggestive of place and grid-cell coding. Such analyses open the possibility to investigate functional gradients in place and grid-cell coding during novel behaviours that are more easily accessible in humans, such as non-spatial memory, abstract reasoning and imagination.</p> <p>Together, these studies illustrate the potential power of functional magnetic resonance imaging techniques to investigate precise computations in humans.</p>
|
|---|