Below is an analogy Jon Haidt uses in his book The Happiness Hypothesis to describe the relationship between our higher order cognition (the rider in his analogy) and emotion (the horse). He goes on to discuss how we feel that we are in control of the horse (or our emotions), but the horse does many things automatically that we have very little control over (nor would could our cognitions be efficient enough at controlling.
I first rode a horse in 1991, in Great Smoky National Park, North Carolina. I’d been on rides as a child where some teenager led the horse by a short rope, but this was the first time it was just me and a horse, no rope. I wasn’t alone—there were eight other people on eight other horses, and one of the people was a park ranger—so the ride didn’t ask much of me. There was, however, one difficult moment. We were riding along a path on a steep hillside, two by two, and my horse was on the outside, walking about three feet from the edge. Then the path turned sharply to the left, and my horse was heading straight for the edge. I froze. I knew I had to steer left, but there was another horse to my left and I didn’t want to crash into it. I might have called out for help, or screamed, “Look out!”; but some part of me preferred the risk of going over the edge to the certainty of looking stupid. So I just froze. I did nothing at all during the critical five seconds in which my horse and the horse to my left calmly turned to the left by themselves.
As my panic subsided, I laughed at my ridiculous fear. The horse knew exactly what she was doing. She’d walked this path a hundred times, and she had no more interest in tumbling to her death than I had. She didn’t need me to tell her what to do, and, in fact, the few times I tried to tell her what to do she didn’t much seem to care. I had gotten it all so wrong because I had spent the previous ten years driving cars, not horses. Cars go over edges unless you tell them not to.