Neural Processes Underlying the Flexible Control and Learning of Attentional Selection

In every-day life we are usually surrounded by a plethora of stimuli, of which only some may be relevant to us at a given moment in time. The dynamic interaction between internal factors, such as our previous experience and current goals, and external factors, such as salient sensory stimulation, determine where, how and what we attend to in our environment. This dissertation investigated some of the neural mechanisms that underlie successful goal-directed behavior in two conditions 1. when attention was actively cued to a target stimulus, and 2. when the attentional target had to be actively and repeatedly learned, in macaque monkeys and in humans. In Chapter 2, I investigated inter-areal spiketrain correlations in neuron pairs across the fronto-cingulate cortex when macaque monkeys are cued to shift their attention to one of two target stimuli. I found that neuron pairs in anterior cingulate cortex (ACC) and dorsal prefrontal cortex (PFC) with similar spatial preferences correlate their spiketrains at the time when attention needs to be actively shifted, suggesting that the flexible interaction between these two areas may support successful covert attention shifts. In Chapter 3, I show that when the attentional target stimulus needs to be repeatedly learned and is defined by only one of several stimulus features, neurons in macaque frontal and striatal regions encode prediction error signals that carry specific information about the stimulus feature that was selected in the preceding choice. These signals may be involved in identifying those synapses that require updating to allow flexible adjustments in goal-directed behavior. In Chapter 4, I found that when humans must repeatedly learn the identity of an attentional target, a human event-related potential over visual cortex that is thought to index attentional target selection, selectively decreases after successful learning, in particular for the distracting stimulus, and selectively increases for the target stimulus following negative feedback during learning. Overall, this dissertation provides novel insights into some of the complex neural mechanisms that support flexible control and learning of attention across brain regions of the human and non-human primate brain.