Topic: low-dimensional manifolds

A fundamental problem in neuroscience is how neuronal activity in the brain generates organized and stable behaviors across multiple timescales. Recent research in many model organisms, e.g. in monkeys, led to the observations that the neuronal encoding of individual actions, such as preparation for and execution of arm movements, is achieved by large neuronal populations but often lies on low-dimensional manifolds. However, it is largely unclear if such manifolds also coordinate higher dimensional activity patterns that correspond to simultaneously executed fine-tuned movements. In order to a

Artificial recurrent neural networks (RNNs) have become a customary research tool in theoretical neuroscience, being used as models of biological neural networks performing cognitive tasks. While RNNs are not generally intended as faithful, biologically realistic models of neuronal activity, there is a broad expectation that RNN dynamics may provide good models at the level of neural population dynamics. According to several authors, in both cases population activity should organize into low-dimensional manifolds encoding the relevant task variables (1). However, strong tests of this hypothesi