Yes, I believe arguments like GP vastly underestimate the amount of data involved in children's learning. All the data coming into their sensorium, from heard language to physical movement feedback, contributes to training all their skills, from physical movement to language.