This is a program with just one input, scenario, and only one output, motion (or actions) a. There's neither a separate reinforcement enter nor an information enter with the ecosystem. The backpropagated benefit (secondary reinforcement) is definitely the emotion towards the consequence predicament.