Charlie Steiner comments on Learning biases and rewards simultaneously