The other day, I was walking down the street and I saw a guy take out something out of his garage. I wasn’t sure what it was, but it looked like one of those carpet things you put on the floor of your car. It looked a bit too big for a car and too small for a truck. And there I stopped thinking.
All of the sudden, it struck me: this is a big mistake. I shouldn’t be content with ending a reasoning line with “I don’t know”. I should try to understand what it is, to create a theory that suits data (i.e. that it was too big for a car and too small for a truck), not to just give up and say that I don’t know.
I concluded that it was probably one of those things you have in the trunk of your car, not on the floor. Of course, I could have just walked over to the guy and asked him, but that isn’t really the point here. The point is to always offer SOME solution to a question and to never say that you don’t know. It doesn’t matter if it’s the wrong solution – this depends largely on (perceived) data. What matters is that you always create a theory that fits the given data and that is probable. Never let your reasoning chain just “hang” there. Always finish it.
That being said, it is equally, if not more important to avoid confirmation bias that all humans are prone to. Confirmation bias means that you give more significance to your explanation than to other explanations, even though other explanations work better with given data. This phenomenon occurs mostly when you create your theory before the arrival of new data. Even though you would have created theory no. 2 if you had had this new data beforehand, you might stick to your original theory now just because it’s “yours”. There is a lot of psychological literature out there about confirmation bias and I encourage the reader to educate himself/herself in this field as I certainly will.
Always finish your explanations. Your explanations must be probable and have to fit the data. When acquainted with new data, rearrange, fix or discard your original explanation. Repeat.