Understanding decision-making from the dynamics of large neural populations in behaving zebrafish – ERASYSBIO (ANR-09-SYSB-0009)

One of the central goals in brain research is the elucidation of the dynamic interactions among neurons and circuits underlying cognitive processes. In this project we focus on decision making as an essential element of cognition. We propose a multidisciplinary systems biology approach towards the understanding of the neuronal network mechanisms underlying decision making in behaving zebrafish. To study how simple decisions are represented and processed in the nervous system, we shall present to naïve zebrafish larva ambiguous visual stimuli (stimuli that induce two possible behaviours with similar probabilities). This type of stimulus always evokes similar patterns among sensory circuits, but depending on the behavioural choice, very distinctive ones among networks involved in decision making. To that end, the activity of large neuronal networks and behaviour will be simultaneously monitored, using a two-photon Ca2+ imaging custom-built system. Both behaviour and circuit activities will be analysed using decision-making theory, information-theory tools, and biophysical circuit models. Using these tools, we shall identify ‘network functional states’ and ‘cell-specific states’. We shall also examine the role of spontaneous activity in the brain. Ongoing activity, once interpreted as irrelevant random noise, has been found to exhibit highly coherent spatiotemporal patterns suggesting a possible role in cognition. To examine the role of ongoing spontaneous activity on the decision, we shall test the idea that decisions result from the interaction between the internal state of the brain and the activity evoked by external sensory stimulation. Finally, we will perform experiments in zebrafish modelling neurological disorders and addiction (e.g. Parkinson and amphetamine addiction) and then, expecting to provide a quantitative understanding of the impact of these diseases on decision making at both behavioural and neuronal circuit levels.