To understand and correct the brain, we need a method to explore these attracting states. By exploration, we simultaneously tackle both clinical problems and scientific inquiries — two birds with one stone. If a patient is stuck in an unwanted attractor, exploring can perturb the individual out of such a state, to a nearby attractor and remedy the neurological condition. Furthermore, the general discovery of these attractors will further our understanding about how the brain actually works, creating a path for future personalized neurological treatments. To be clear, the brain can be broken down in to many individual regions, where each region may be comprised of hundreds of thousands to tens of millions of neurons. Those countless numbers of cells are all communicating with one another via electrical signals, and we want to be able to look at one piece, and be able to say, ‘Okay, this set of neurons has, for example, three attractors: A, B and C.’


Now, while we don’t know much about these attractors, we do know that moving between them can be difficult. It’s like trying to escape a maze. If we want to successfully explore, we need a highly sophisticated method — A method that learns a map of the neural dynamics as it goes, and uses that map to efficiently perturb the state of the system, traveling about these multiple attractors with ease. Furthermore, that method has to be robust, as everyone’s brain is different, so thus are the means to change between states. If transitioning between states is like escaping a maze. Then, due to various lifestyle and genetic factors, each individual’s mazes are different.