We have all experienced it; second guessing ourselves. You circle one answer on a multiple choice exam, and then change your mind as well as your answer. You choose one apartment over another, then change your mind and go with the second one. Do you do better when you go with your gut or when you take time to analyze your options?
Researchers have found that people who change their answers did better than those who stuck with their first responses. This turns out to be true even for experts, such as master chess players. First moves are worse, even for simple chess problems.
Does this mean we should ignore our hunches or intuitions? Not necessarily. The decisions are the output of two processes, a fast intuition or emotion-based process and a slower, deliberative one. The intuitive activities are very similar to perceptual activities, such as seeing and hearing. Ask yourself this: When you glimpse something out of corner of your eye, what do you normally do next? You probably direct your attention to the new stimulus, allowing your visual system to process it in more detail. As it turns out, that is probably the best way to think about the role of intuition in decision-making. Your gut reaction tells you this particular choice deserve further deliberation; that is all no more.
It also matters a good deal whether your intuitions spring from a well-organized knowledge base or ignorance. People with higher trust in their feelings are more likely to correctly predict the final outcome than those with lower trust in their feelings. The researchers call this phenomenon the emotional oracle effect.
But there is one caveat: The accuracy of these intuitions depended on the amount of domain-relevant knowledge possessed by the person making these predictions. For example, only people who had some background knowledge about the current football season benefited from trusting in their feelings when predicting the winner of the national college football championship. This means that we can safely rely on our feelings or intuitions only when we are making decisions or predictions in domains we know well.