The Halo Effect and Unconscious Accuracy

<a href="http://en.wikipedia.org/w/index.php?title=Halo_effect&oldid=628757205">The Halo Effect is a cognitive bias in which a person, A, is more likely to believe that another person, B, is good in some respect, D, if they are known to be good in some other respect, C, even in the absence of independent or observed evidence that B is in fact good with respect to D.

The fact that the Halo Effect is a real cognitive bias is usually employed to imply that some person is more likely to make a faulty conclusion because of the Halo Effect. After all, bias usually implies error. However, the Halo Effect is an example of positive bias. This is because, as a matter of fact, people are often actually marginally better with respect to some random quality B if they are known in the first place to be good with respect to some quality A.

This is not a necessary relation by any means. It is a marginal normative relation. This is an important point even if it is a bit complex. The easy part is understanding that just because some person is good at one thing doesn't mean they are necessarily good at anything else. They might just be good at one thing.

The more subtle point is that, while they are not necessarily good at anything else, they probably are. People who are proficient at one thing are, in reality, rarely proficient at only that one thing. Given only the information that some person is proficient at one thing, they are rationally expected to be more proficient in general than they would be if we had not been given that information. In other words, proficiency in one area is not proof of general proficiency, but it is evidence of general proficiency.

The point becomes even more powerful if the trait is randomly selected with respect to the person who's quality is in question, but it need not be randomly selected to have some importance.

Interestingly, the Halo Effect may or may not be an example of unconscious accuracy. Unconscious accuracy could lead to the existence of true knowledge without conscious grounding. In other words, knowing that something is true without necessarily being consciously aware of why it is true or how we can know it is true. Knowing something is true without such other knowledge is often regarded as properly basic knowledge, although the reality of such knowledge is hotly debated.

Knowledge is often defined in epistemology as any belief which is both justified and true. The debate is on whether these qualities may be independent, or if it is required that they are all mutually interdependent. It's uncontroversial to say that knowledge exists when the qualities all exist interdependently, but it is not clear whether knowledge exist when such qualities exist independently.

If A is true, A is justified, A is my belief, I believe that A is true, I believe that A is justified, I can justify that A is true, and I can justify that I believe A, then with respect to me, A interdependently believed, true, and justified. Consider that A, B, and C are all qualities, and the set of characteristics of D includes A, B, and C, as well as every possible combination of those characteristics. In this case, D is knowledge.

If A is true, justified, and I believe it, then A possesses these qualities independently, even if I do not believe that A is justified and so on. That is, I believe that A is true but not justified. Consider that A, B, and C are all qualities, and the set of characteristics of D includes A, B, and C, but it does not contain every combination of those characteristics. The debate is on whether D is still considered knowledge.

If something is true, and you believe it is true, but you don't know why you believe it is true, can you still call it knowledge? This is the question. It was previously not thought to even be worth asking because it seems so odd that anyone would think something is true when they can't explain why it should be considered true. Unconscious accuracy makes the question now worth further investigating.