Ex vivo cortical circuits learn to predict and spontaneously replay temporal patterns

Area of Science:

• Neuroscience
• Computational Neuroscience

Background:

• Prediction and timing are hypothesized as fundamental computational primitives of neocortical microcircuits.
• Neural mechanisms may enable autonomous learning of temporal structures and internal prediction generation.

Purpose of the Study:

• To investigate the capacity of neocortical microcircuits to autonomously learn temporal structures and generate predictions.
• To test the hypothesis that local cortical circuits possess intrinsic learning rules for temporal information.

Main Methods:

• Training of cortical organotypic slices using dual-optical stimulation on specific temporal patterns.
• Whole-cell recordings to analyze network dynamics after 24-hour training.
• Observation of spontaneous neural activity for learned structure replay.

Main Results:

• Network dynamics consistent with training-specific timed prediction were observed.
• Spontaneous activity exhibited replay of the learned temporal structure.
• Some neurons showed timed prediction errors, responding more strongly to omitted expected stimuli.

Conclusions:

• Local cortical microcircuits are intrinsically capable of learning temporal information and generating predictions.
• Learning rules for temporal learning and spontaneous replay may be intrinsic to local circuits, not requiring top-down interactions.
• Asymmetric connectivity between neuronal ensembles with temporally-ordered activation underlies this learning.