Thunderstruck: The ACDC model of flexible sequences and rhythms in recurrent neural circuits

• Published in: PLoS computational biology Vol. 18; no. 2; p. e1009854
• Main Authors: Calderon, Cristian Buc, Verguts, Tom, Frank, Michael J
• Format: Journal Article
• Language: English
• Published: United States Public Library of Science 01.02.2022; Public Library of Science (PLoS)
• Subjects: Basal ganglia; Biology and Life Sciences; Chain dynamics; Cognition; Computational neuroscience; Computer and Information Sciences; Context; Decomposition; Flexibility; Ganglia; Learning; Mathematical models; Medicine and Health Sciences; Models, Theoretical; Neural circuitry; Neural networks; Neural Networks, Computer; Physiological aspects; Recurrent neural networks; Rescaling; Research and Analysis Methods; Rhythm; Scaling; Social Sciences; Thalamus; United States
• ISSN: 1553-7358; 1553-734X; 1553-7358
• DOI: 10.1371/journal.pcbi.1009854

Adaptive sequential behavior is a hallmark of human cognition. In particular, humans can learn to produce precise spatiotemporal sequences given a certain context. For instance, musicians can not only reproduce learned action sequences in a context-dependent manner, they can also quickly and flexibly reapply them in any desired tempo or rhythm without overwriting previous learning. Existing neural network models fail to account for these properties. We argue that this limitation emerges from the fact that sequence information (i.e., the position of the action) and timing (i.e., the moment of response execution) are typically stored in the same neural network weights. Here, we augment a biologically plausible recurrent neural network of cortical dynamics to include a basal ganglia-thalamic module which uses reinforcement learning to dynamically modulate action. This "associative cluster-dependent chain" (ACDC) model modularly stores sequence and timing information in distinct loci of the network. This feature increases computational power and allows ACDC to display a wide range of temporal properties (e.g., multiple sequences, temporal shifting, rescaling, and compositionality), while still accounting for several behavioral and neurophysiological empirical observations. Finally, we apply this ACDC network to show how it can learn the famous "Thunderstruck" song intro and then flexibly play it in a "bossa nova" rhythm without further training.