Grant: Discovery Grants Program - Individual

Selective attention is a set of core operations in the brain that control the flow of behaviourally relevant information. Since the first demonstrations of a neural basis of attention, researchers have focused almost exclusively on how attention changes the firing rate of neurons-expressed by the mean of their activity over experimental trials. However, driven by sophisticated new probes of neural activity in non-human animal models, our understanding of attention is changing dramatically. From this recent work, a novel mechanistic basis of attention has emerged, which is invisible to traditional approaches of averaging neural activity over trials. In addition to increasing a neuron's average responses, attention can reduce the variance, or noise, of a neuron's trial-to-trial responses. Attention-driven reductions in noise can have large effects on population neural activity because an individual neuron's noise is often correlated with the noise of neighboring neurons. By reducing shared noise in a population, attention can dramatically influence how much information is represented at any given time. Intriguingly, one of the brain's core neuromodulatory chemicals-acetylcholine-induces rapid changes in the noise correlations among neurons which closely mimic the effects of directed attention. Together, these discoveries reveal a potent mechanism of attention expressed in the noise of population coding and its neurochemical basis. However, the invasive recording methods used to advance these discoveries are not feasible in humans. As a result, cross-species translation has slowed. This is a big problem. Current non-invasive probes of population activity, such as functional magnetic resonance imaging (fMRI), almost exclusively rely on trial-averaging methods. Large-scale effects of attention on population noise are therefore neglected in human research despite their fundamental role in cortical information processing. Moreover, because the most commonly used neuroimaging techniques, such as fMRI and EEG, are not sensitive to the brain's neurochemical environment, our understanding of the links between cognition and neuromodulation in humans remains very basic. My lab will examine the mechanistic basis of neuronal noise using human in vivo neuroimaging in order to translate and extend recent findings from non-human animal research. Are the mass neuronal responses indexed by fMRI sensitive to attention-driven changes in noise correlation? If so, how do attention-driven changes in noise correlation relate to the amount of information carried in the population code? Is cortical cholinergic modulation detectable with non-invasive imaging probes such as 1H magnetic resonance spectroscopy? If so, do attention-driven changes in noise correlation rely on cholinergic modulation? My program of research will develop novel human neuroimaging methods to triangulate the links among directed attention, cholinergic modulation and population coding.