Representativeness Heuristic
Often in this situation, people estimate pretty highly that John is indeed a lawyer; but in reality, there's a 30% likelihood. Why? Because from that particular study, there were 70 engineers and 30 lawyers; but because we know that information about John—that he's into rare book collecting, that he's articulate and argumentative—those details tend to make us think, "Well he's definitely a lawyer." And therefore, we might estimate higher probability that indeed that's who he is. This is a representativeness heuristic at play. So we're going to decide if an example belongs to a certain class or group on the basis of how similar it is to other items in that class or group. A few other examples that you might think of is a rich car buyer or a student who belongs to a fraternity or sorority. Immediately you're going to think about what sort of, again, that prototype or concept, that natural concept is; and then if you were to see somebody, you would sort of apply that to that person and say, "Ok, how much is this person likely to belong to this group? How likely is this person that fits my schema of someone in a fraternity or sorority?" And etc. So this is an example of representativeness heuristics.