The psychologists will tell you to look at the famous Skinner Box experiments, in which he showed that intermittent reward was the most powerful motivator; e.g.

The psychologists will tell you to look at the famous Skinner Box experiments, in which he showed that intermittent reward was the most powerful motivator; e.g., the rats will press the lever much more when they only get occasional random rewards after pressing the lever vs reliable rewards for pressing every N times. While there's definitely an element to that, I think you are more onto the right answer. From my experience (having tried a range but certainly not all), and those of many I know, there is no drug like adrenaline - not even close. I read about a drug-rehab program that showed extremely high success rates, and it was based around teaching former addicts rock climbing. Having experienced both the confidence that comes from climbing, and the adrenaline, this was no surprise. What also was, sadly, not surprising was that it seemed to go away, I'm guessing because it couldn't get consistent funding for what most people considered too high a risk (as if lower success 'normal' programs posed no risk...).