One of the most plausible interpretations is that dopamine's primary role in cognition isn't to model reward, but to predict uncertainty. This seems to be roug

One of the most plausible interpretations is that dopamine's primary role in cognition isn't to model reward, but to predict uncertainty. This seems to be roughly in line with how subjects with excessively high or low dopamine behave. Those on heroic doses of amphetamines become convinced of the most harebrained ideas. ("I know the CIA is listening to me through the toaster, better take it apart.") Whereas schizophrenic catatonia, a lack of will to even move, is often caused by impaired dopamine receptors. This clarifies why dopamine drives addiction and compulsion, without a corresponding reward in terms of actual satisfaction. The evolutionary environment heavily incentives risk aversion. The downsides of any major risk almost always outweigh the upsides. Keep pursuing a more nourishing but unpredictable food source and you'll eventually starve, regardless of how well fed you are in good times. Keep fighting opponents that are even matches, and eventually you'll get your head bashed in. Our brains are heavily wired to seek out small but certain rewards to the exclusion of large but risky ones. This is confirmed in behavioral economics, where subjects will instinctually and irrationally prefer low-risk, low-reward bets even when the payoffs are measured in pennies. If you feel shitty and depressed, then eating ice cream, banging heroin, or scrolling social media won't fix all your problems. But you know for sure it will make you feel a little better right away. In contrast taking positive steps to improve your life, like exercising, eating better, developing relationships, or mastering marketable skills are ultimately more rewarding but risky. You can put in a lot of effort for a long time and still have no guarantee of seeing any tangible payoff.