Whenever there is some new juice / ice cream center opened nearby, I visit it almost regularly and try each and every juice / ice cream flavor. The way I do it is as follows. I look at the menu card and short list a few of them.. not based on my liking but based on price may be! Then based on my liking, I chose the one that looks least favorable for me. On my next visit, I chose the second least favored and so on. I was telling this to a friend of mine yesterday - only to realize later the logic which I put behind such "least_favored_first" sequence is going to help me in making any decision in general.
One has to decide on something - whether to got for it or not, let's say X. There is something called Y which one is not much willing to do. There is something called Z which one likes to do. X and Z can co-exist. Y and Z make a better combination and X and Y make the toughest one; And one has to make a series of decisions, like probably go for Y, then if possible think about X if it could be done, if not may be Z and so on. One has some priors as already told but they come from one's mind, not necessarily from experience. (Of course, if they were from experience, the entire decision process would have been just a trivial one!)
As Z can happily co-exist with either X or Y, one will consider it as bonus, so one has to choose finally between just X or Y. Prior against Y is little higher than that against X - does not mean though that X is favored (hmm.. it's statistically wrong, but that's how it is!) One might tend to go for X in this case. But one has to also consider that - just because prior is high against Y, that makes one overlook probable risks involved in X. Y seems risky - this kind of prior makes one even NOT analyze X properly. Such a hurried decision only based on prior (that to against!) is entirely risky - no matter whether one chose X or Y.
Probably, the other way sounds wise. Though Y has higher against_prior, one can take a chance of choosing Y - get oneself convinced that Y indeed has a higher risk and one does not want to continue with it. Once that belief is there, one can easily face the hidden risks in X. The bottom-line - decide which has higher against_prior, get oneself convinced that it indeed does not suit one (if so) and then of course, the other option is always there as one always favored it apriori - than going with the less against_prior option, realizing that it does not suit one the way one thought previously and then getting back to higher_against_prior will be difficult - because one never favored it anyway.
Can link this to choosing ice cream flavors? Well, assume that my preferences are (in decreasing order) - tender coconut, anjeer, mango. People say mango is awesome! But, I am not crazy about mango so I kind-of give it lowest preference, or highest_against_prior. Then that means I am favoring tender coconut always - this is based on some information I already have - say from people telling me that it tastes good or may be because I like tender coconut and so have a reason to believe that I will like its ice cream also or something other than this. Note that, we have already assumed that I don't have experience of any of these flavors - else the decision is just trivial. As per my logic of choosing "least_favored_first", I go with mango. If I like it, well, good! I always have one more flavor to go for. If I don't like it - well, yeah! of course I knew I was not going to like it anyways! In both cases, I have a win-win. I can always taste other flavors or not taste may be, given that I liked mango so much! Another case - if I start with tender coconut - I like it - well and good! If I don't - well, this was my first preference, if I haven't liked this, the chance that I'll like other low preferences is still less - I might stop going to the ice cream shop altogether!
Sounds stupid?! hmm..! may be.. but it's gonna help me I think.
3 टिप्पणी(ण्या):
kahipan ...
what if you have only one chance to take the decision? will least_favoured_first work for this?
@devendra, if there is just one decision (which is hardly the case) no one will even think this much.
@koustubh, I know, "kahipan" aahe. lihiley me tasa post madhyech.
Post a Comment