Indexed in

External link Full Text

License and use

Citations

Altmetrics

Analysis of institutional authors

Berbel Fernandez, BlancaAuthorLatorre Camino, RobertoAuthorVarona, PabloAuthor

Share

Emergence of sequential dynamical invariants in central pattern generators from auto-organized constraints in their sequence time intervals

Publicated to:NEUROCOMPUTING. 578 127378- - 2024-04-14 578(), DOI: 10.1016/j.neucom.2024.127378

Authors: Berbel, Blanca; Latorre, Roberto; Varona, Pablo

Affiliations

Abstract

Motor coordination by the nervous system involves generating and coordinating muscle sequential activity. Such activity must be robust to produce effective motion that relies on patterned motor events ordered in a specific sequence and, at the same time, flexible in timing and duration to adapt to intrinsic and environmental circumstances. Motor neural circuits produce the signaling that muscles use for their sequential action. Even in rhythmic repetitive motion such as walking, breathing and chewing, variability in the timing of the events that build the sequence can be observed within the overall ordered rhythm of the motion. In this context, central pattern generators (CPGs) are key neural circuits to study the cycle-by-cycle balance between the robustness of the sequential order of neural events and the irregularity of the intracycle intervals that shape a rhythmic sequence. Recently, the presence of sequential dynamical invariants in the form of robust cycle-by-cycle relationships between specific sequence time intervals was unveiled in living CPGs. Variability of other sequence intervals remains unrelated and coexists with the presence of the invariants. Sequential dynamical invariants have been proposed to underlie the cycle-by-cycle CPG coordination. In this work, we used a minimal CPG circuit building block of two interacting inhibitory model neurons to assess the emergence of coordinated variability in sequential neural activity. We studied the conditions that produce sequential dynamical invariants in the circuit using intrinsically irregular neuron models and symmetric and asymmetric network topologies. We quantified the circuit irregularity and related it to the presence of invariants under different configurations. We found the conditions to propagate and sustain coordinated variability between time intervals, and thus our study illustrates the auto-organized constraint-based mechanisms for motor sequence coordination that shape sequential dynamical invariants. This is the first model study in which sequential dynamical invariants are observed considering only the intrinsic variability of CPG neurons without external stimuli. Finally, we discuss possible applications of this research in the context of autonomous robotics and artificial intelligence.

Keywords