Functional connectivity among cortical regions is shaped by associative experiences

Abstract

The emergence of coherent cognition and behavior relies on the coordination of scattered mosaics of functionally-specialized brain regions. However, little is known about how the neural mechanisms that coordinate this distributed brain activity are formed. Hebb proposed that brain regions organize dynamically into functional groups by the temporal structure of their neural activity as co-activation of brain regions leads to the Hebbian strengthening of the functional connectivity between them. To test this hypothesis, we trained subjects to learn the association between photos of houses and Chinese characters, and then examined whether the coherence between BOLD signal fluctuation in the parahippocampal place area (PPA) and that in the visual word form area (VWFA) increased after the learning. We found the training not only improved the behavioral performance in pair-matching task but more importantly increased the functional connectivity between the PPA and VWFA. The increased coherence was not due to the mental imagery of the associated stimulus (e.g. words) when a trained stimulus (i.e. houses) was presented, because the activation level of the trained houses in the VWFA was unchanged before and after training. Together, these findings suggest that the functional connectivity between co-activated regions, but not their individual activation, may be the material basis of mental associations. Interestingly, when the PPA was activated by untrained house or scene stimuli, the neural activity in the VWFA fluctuated in a coherent fashion, suggesting that the propagation of the temporal structure in neural activation from one node to another is not stimulus specific. On the other hand, the increased coherence was limited for non-coactivated regions as the connectivity strength between the VWFA and the lateral occipital complex (LOC) and that between the PPA and the LOC were unchanged. In sum, our results provided an example of how large-scale integration is formed through Hebbian learning.

Footnotes

 This work was supported by the National Natural Science Foundation of China (30325025 to J.L.) and the National Key Basic Research Development Program of China (2007CB516703 to J.L.).