Behaviour follows its consequences: what gets rewarded grows, what gets punished fades, and what gets rewarded irregularly sticks best

Voluntary behaviour is shaped by what follows it. When an action brings something pleasant, the odds that you repeat it go up. When it brings something unpleasant, they go down. You never have to decide any of this: behaviour adapts to its consequences whether you notice or not. Underneath that sits a second layer that is easy to miss. It is not only whether a reward arrives that counts, but when and how often. The same behaviour with the same reward can be learned fast or slowly, and can stick weakly or strongly, depending purely on the rhythm in which the reward is handed out. That rhythm is called a reinforcement schedule. The surprising part is that irregular rewards work better than constant ones. Behaviour that is rewarded only some of the time lasts longer, even after the reward disappears. The logic, as Scott Young lays it out, runs like this: a predictable reward that stops has probably run out. An unpredictable reward that stops may just be a pause, so carrying on pays. That…