Building on simple unsupervised matrix factorization techniques, the seqNMF algorithm successfully recovers neural sequences in a wide range of simulated and real datasets.
Time-ordered and flexible motor sequences in C.elegans are generated by combining an excitatory feedforward coupling and mutual inhibitions between neurons in different functional modules.
Hippocampal sharp-wave sequences, which are considered the key phenomenon underlying consolidation of episodic memory, are preserved even after an animal’s memory is impaired.
Detailed analysis of fMRI data shows that sequences of movements are associated with individual patterns of neural activity that become more distinct with training.
Homology information implicit in regions of conserved synteny allows quantification of gene origination by complete sequence divergence, revealing a larger-than-expected role for other mechanisms of origin, including de novo origination.
Precisely sequenced patterns of neuronal activity associated with the production of a skilled behavior begin and end as part of orchestrated activity across functionally diverse populations of cortical premotor neurons.
Inspired by the sparse, sequential neural activity patterns observed in striatum, a new circuit model implements variable-speed activity, the encoding of multiple sequences, and a tutor/student relationship between cortex and striatum.
