Gain neuromodulation mediates adaptive perceptual switches: Evidence from pupillometry, fMRI, and RNN Modelling

Christopher Whyte (University of Sydney)

Please note that this event is at 10am-12pm Australian Eastern Daylight Time.

Zoom link: https://monash.zoom.us/j/84663747139?pwd=OWZiajBnZHRVMS9SR2I3NVk5QkpkZz09

The event will include Christopher's presentation and a presentation from Claire O'Callaghan (also from University of Sydney).

Abstract for Christopher's presentation:

Perceptual updating has been proposed to rely upon network reset events that occur when bursts of ascending neuromodulatory neurotransmitters, such as noradrenaline, abruptly modulate the brain’s susceptibility to changing sensory activity. To test this hypothesis at a macroscale, we leveraged a combination of pupillometry, fMRI and recurrent neural network modelling of an ambiguous figures task. Behaviourally, qualitative shifts in the perceptual interpretation of an ambiguous image were associated with peaks in pupil diameter, an indirect readout of phasic bursts in neuromodulatory tone. We further hypothesized that increases in neuromodulatory tone, driven by stimulus ambiguity, should lead to heighted neural gain, which in turn should hasten perceptual switches. To test this hypothesis computationally, we trained a recurrent neural network to perform an analogous perceptual categorisation task and following training, allowed gain to change dynamically as a function of the network’s classification uncertainty, mimicking the hypothesised role of neuromodulatory tone. As predicted, we observed an earlier perceptual switch as a function of heightened gain. Analyses based upon dynamical systems theory revealed that the speeding effect of gain was mediated by a transient destabilization of the network’s dynamical regime in periods of maximal uncertainty. We leveraged a low-dimensional readout of the RNN dynamics, to develop two novel predictions that could be tested at a macroscale: perceptual switches should co-occur with peaks in low-dimensional brain state velocity and with flattened ego-centric energy landscape dynamics. We used dimensionality-reduced summaries of whole-brain fMRI dynamics to independently confirm each of these predictions. These results support the role of the neuromodulatory system in the large-scale network reconfigurations to mediate abrupt adaptive changes in perception.

Title and abstract for Claire's presentation:

Windows into noradrenaline’s role in decision making and learning through single dose drug studies in Parkinson’s disease

Single dose drug studies provide a powerful framework for understanding neuromodulatory systems, and they offer one of the rare opportunities for a causal manipulation in humans. In conditions like Parkinson’s disease – which undergo profound changes in neuromodulatory function – these studies can provide important clinical insights, as well as being a test bed for current normative neuroscience theories. I will present some of our recent findings using the noradrenergic reuptake inhibitor, atomoxetine, in Parkinson’s disease. I will focus on oculomotor decision making and reinforcement learning, which are two areas that have typically been viewed through a dopaminergic lens. Our results uncover some new aspects of noradrenaline’s contribution to these functions, and in doing so, further substantiate a role for noradrenaline in complex adaptive behaviour.