Neuron codes and learning
Sequence learning, or knowledge of sequences, may be associated with several types of learning including languages and motor skills. The exact mechanisms are unknown, but may involve statistical learning rules, which themselves may relate to various kinds of brain rhythms. The EU-funded PSLOAHMD (Predicting sequential learning from oscillatory activity in human MEG-Data) project investigated connections between sequence learning and neuronal oscillations of various frequencies. The study particularly examined how the phase of theta oscillations (5-7 Hz) contributes to recall of sequences. Researchers also considered the how the amplitude of beta (15-30 Hz) and gamma (30-100 Hz) oscillations correlate with prediction of expected events. Workers observed oscillations in adults using magnetoencephalography. The non-invasive technique allows tracking of rapid brain oscillations, while offering resolution comparable to functional magnetic resonance imaging. Using spectral analysis techniques, the team assessed whether neuronal activity carried sequential information. Hence researchers quantified the relationship between oscillatory neuronal activity associated with learning sequences, and the degree of familiarity with images. The team observed elevated activity at three frequency bands, correlated with familiarity of sequences. Such findings supported the hypothesis whereby rhythmic neuronal activity may provide a learning role. Also, representation of sequences in human memory may involve an oscillation code. Specific frequency patterns may also be involved in how the brain predicts expected events, while other patterns may counteract sequence learning. The results may help to reveal the role of brain rhythms in learning.
Keywords
Prediction, sequential, PSLOAHMD, learning, oscillatory, frequencies
