The first thing you need to do is define the word "belief."

Logically, I can deductively prove that I am male. However, there is an end to what formal proofs can do. From another point of view, there is a possibility (though unfalsifiable at present) that I am deceived, dreaming, or insane. So, I must accept that what I perceive is reality before my deduction becomes "knowledge."

It's worth pointing out that the above argument is somewhat irrelevant because if I am deceived, dreaming, or insane, there is little practical difference. If this is my reality, then my deduction is true within my reality, and my inability to perceive any other reality renders the question moot.

I believe that most people who know me think I am a good person. I do not know this, for A) I have not asked most of the people who know me, and B) I cannot know with certainty that they would tell me the truth. I do not have deductive certainty, only the probabilistic belief that it is so.

So, properly speaking, we can know something when we have overwhelming and essentially irrefutable evidence. To believe something, all we need is more evidence that leans toward the belief than leans away.

Another point about knowledge vs. belief. Knowledge, strictly speaking, must be true. There is no such thing as knowing that there are real fairies. One can only believe in fairies, since they do not exist. Furthermore, suppose that one day, someone discovers a fairy. Those who believe in fairies today still do not have knowledge of them, even though they do, in fact, exist. It is still belief today, because to the best of our knowledge, fairies do not exist.

(It should be noted that there are some who are not completely satisfied with the above explanations. It has been suggested that the definition of knowledge should include additional parameters. For example, one solution suggests that no false beliefs may be present along the "path to knowledge." Perhaps Todangst or Deludedgod are more familiar with some of these ideas. I believe that the currently accepted idea is that knowledge is essentially "justified true belief." )

Now, as to whether or not you can choose belief, it is easy to understand why we cannot believe that we are the other gender. We have incontrovertible evidence -- our bodies are the proof. But what about something that is not certain? I believe that there are more humans on the earth than the earth can sustain indefinitely. I don't have proof of this, but I have strong evidence based on what I have read, and whose opinions I trust.

Can I choose to believe otherwise?

No, I can't, and neither can you. Try the exercise yourself. Think of something that you believe, but that you can't exactly prove. Maybe you believe that the Republican Party is evil, or that your ex girlfriend isn't having very good sex with her current boyfriend. It doesn't matter. Just pick something, and decide to believe the opposite.

Now, think about what it would take for you to believe otherwise. Evidence. Not necessarily hard evidence, or conclusive evidence. In the case of your ex girlfriend, it might be as simple as overhearing another girl at a bar saying that her new boyfriend has the stamina of a pro athlete. Hardly conclusive, but it might be just enough evidence to tip you over to believing the other side.

There are some problems with belief when it is represented as an "A or not A" situation. Most notably, everything does not fall cleanly within this dichotomy. For instance, I don't believe that it is clear and sunny in San Francisco today. Neither do I disbelieve it. While I believe that the majority of days in San Francisco are clear and sunny, I don't have enough evidence about today in particular, so I neither believe nor disbelieve, strictly speaking.

Suppose you ask me to believe that it is sunny in SF today. I will ask why, and you will respond that you just talked on the phone with your cousin in SF, and she said it's sunny. Ok. I believe it. Why? My knowledge of SF's history, combined with the fact that I have no reason to doubt you or your purported conversation with your cousin. These pieces of data are enough to sway me.

Now, suppose that you ask me to believe that it is NOT sunny in SF today. Could I immediately do it? Perhaps, but here's the important part: If I do change my mind, it's because I think of a piece of evidence that changes it!

Perhaps I will think to myself, "He is asking me to do this as part of an experiment. He doesn't have a cousin in SF, and he only told me that to trick me!" This thought will be enough evidence to change my belief.

Can I now immediately change it back? Not without more evidence. I must come up with a reason for disbelieving my new theory, that you are trying to trick me. It probably wouldn't be difficult, but I cannot control it. I can either come up with it, or not.

If you take this example and apply it to any other belief, you will see that belief involves different degrees of certainty, but a degree of certainty is actually just a measure of how much more contrary data you would need for your mind to change! In the case of SF's weather, my belief is highly uncertain, so almost any new data will change my mind. Nevertheless, once I have an opinion, my mind is on one side of the fence or another, and I cannot alter this without inventing a reason. In the case of a strongly held belief, it would take mountains of contrary data for my mind to change.

The bottom line, then, is that a belief is a mental representation that is not fully justified, but is supported by enough data to make it appear that "A" is true rather than "not A." Belief is the end result of subconscious deduction and induction which can be described in the language of logic, but are not under the conscious control of the individual.

Deluded,

In my post, I was specifically avoiding neurology and focusing on epistemology, primarily because I don't know a damn thing about neurology.

Is there such a thing as "neurology for philosophers" that I could read so that I would not be flying quite as blind when someone shifts approaches on me?  I know I can't handle full fledged neuroscience without some kind of primer, and a lot of study, but I don't know where to look.  I guess another way of asking, is there someone who has explained neurology the way Dawkins explains evolution for laymen?

Daniel Dennett

I guess then does this person truly believe in this? What does it mean to truly believe in something?  Would you be believeing in santa because the argument that those other people gave wasn't convincing enough? would you change your mind if the argument was?  

I guess it would be a good point to make a way we can falsify the idea that one can choose beliefs, A way that would show that you believe things based on experience/evidence not some random choice you make. 

also, i geusswhat i mean by belief is basically anything one accepts as true or false that doesn't necessarily have to be justified and need not be objectively true.  in other words things that would count as knowledge if they could be justified and proven to be true objectively.