Modeling the brain during pain processing

The discovery represents a significant advance in researchers' understanding of how our bodies and brains respond to pain. The underlying circuitry of the pain process involves a specific configuration of interneurons, each of which link specific pairs of regions, or 'nodes' within the brain. Crucially, a certain fraction of these neurons will be inhibitory; varying the strengths of the connections they provide. To create a biologically plausible model, Montani and colleagues would first need to consider all possible links between specific pairs of nodes, and determine their relative strengths. Within a structure as complex as the brain, however, it would be virtually impossible to do this by considering each configuration individually.

The researchers overcame this issue using 'graph theory,' which studies structures made up of sets of nodes, which influence each other's behaviours via links of variable strengths. Using a novel statistical approach, they estimated the signals produced by each region of a virtual brain in a given configuration, and how far they diverge from realistic values. From their initial estimates, Montani's team could then build up a realistic graph by strengthening and weakening the influences of certain links. Their analysis revealed that a configuration where 20% of all interneurons associated with the pain process are inhibitory to information transmission.