Mesoscopic Patterns of Neural Activity Support Songbird Cortical Sequences

Abstract

Time-locked sequences of neural activity can be found throughout the vertebrate forebrain in various species and behavioral contexts. From “time cells” in the hippocampus of rodents to cortical activity controlling movement, temporal sequence generation is integral to many forms of learned behavior. However, the mechanisms underlying sequence generation are not well known. Here, we describe a spatial and temporal organization of the songbird premotor cortical microcircuit that supports sparse sequences of neural activity. Multi-channel electrophysiology and calcium imaging reveal that neural activity in premotor cortex is correlated with a length scale of 100 µm. Within this length scale, basal-ganglia–projecting excitatory neurons, on average, fire at a specific phase of a local 30 Hz network rhythm. These results show that premotor cortical activity is inhomogeneous in time and space, and that a mesoscopic dynamical pattern underlies the generation of the neural sequences controlling song.

Abstract

Cortical premotor activity during singing behavior in songbirds is correlated spatiotemporally over a length scale of 100 µm, with neurons and interneurons firing at opposite phases of a 30 Hz network rhythm. Read the Synopsis.

Author Summary

“Time cells” can be found throughout the vertebrate forebrain in various species and behavioral contexts. These neurons fire sparsely at precise times during a stereotyped behavior; however, how a neural circuit supports this remarkable property is not known. Here, we describe that the premotor neuronal circuit that is required in birds for singing—“HVC”—is organized in a spatial and temporal manner that supports sparse sequences of neural activity. During song, one class of principal neuron fires during a specific phase of a 30 Hz network rhythm. Fluorescent imaging using head-mounted microscopes reveals that calcium activity in nearby principal neurons is correlated in time with a length scale of 100 μm. Thus, variations in the phase or precise timing of the network rhythm across HVC can provide a mechanism for coordinating the activity of principal neurons. This observation suggests that the timing of principal cells is supported by a mesoscopic spatiotemporal pattern of neural activity.