Fractionating the Neural Circuits of Human Attention

Abstract

Attention can be goal-directed (willfully allocated), or stimulus-driven (captured by salient events). These two forms of attention are supported by at least partially distinct neural networks, and further, they can be fractionated into subprocesses. Specifically, for stimulus-driven attention, we sought to dissociate the neural substrates associated with dynamic attentional (orienting and reorienting), static attentional (maintenance), and evaluative processes. Here, we present preliminary results from a high-field fMRI study utilizing an extended-oddball paradigm to fractionate these processes using eye gaze as a proxy for the focus of attention. Salient, distracting, task-irrelevant videos appear in the visual periphery while participants perform a tedious discrimination task. Our results confirm the role of the dorsal anterior cingulate cortex (dACC) and anterior insula (AI) in contextual updating during the orientation of attention. Further, instead of the temporoparietal junction (TPJ) playing a primary role in the orienting of attention, this area was primarily active during the evaluation of the oddball stimuli. Finally, we observed an intriguing dissociation between orienting and reorienting, with the former engaging more medial parietal cortex when compared to the latter. The present results provide a mechanistic blueprint of the neural network subserving stimulus-driven attention.