Neurons learn by predicting future activity

Without being an expert in the field, I would be a bit surprised if this approach was truly novel. Can anyone shed some light on previous work that is similar to the one presented? Thanks in advance.