For decades, the brain has been studied as a black-box, providing a stimulus and recording the response. This reductionist open-loop approach is increasingly being replaced by closed-loop neuroscience, which covers transient interactions not only within the neural system, but also with the outside world. Thanks to advances in computer processing power and electronics, researchers can now take advantage of the real-time processing of large streams of neural data.
However, the diversity of closed-loop experimental algorithms and conditions requires both computational speed and high flexibility. To meet this need, Davide Ciliberti and Fabian Kloosterman developed Falcon, a multi-threaded open-source software in which an arbitrary processing graph can be loaded and executed. Falcon is highly versatile and customizable and gives the user direct control over central processing unit resources. It is capable of sub-millisecond response latency over high-rate streaming data acquired from high-count multi-electrode arrays.
Davide Ciliberti: “We envisage Falcon to be a useful tool to the neuroscientific community for implementing a wide variety of closed-loop experiments, including those requiring use of complex data structures and computationally intensive algorithms, such as population neural decoding or encoding from large cell assemblies.”