Imagine that we have a game show where the host asks you to open one of two doors. You don't know what rewards are behind each door, but if you're willing to part with $500 of your previous winnings, the host will tell you. Should you pay $500 or not?
Of course, it depends on your prior belief in the difference in value of the two doors. If the difference is at least $1000, then it's worth it to pay so you can be sure to get the better door. Otherwise, it's not worth it to pay. The best way to assign your prior belief is to buy DVDs of the old episodes of the game show and see what sort of rewards have been behind the doors in the past. But buying DVDs itself costs money, not to mention the value of all that time you'd spend watching the show. It's not clear how much you'd stand to gain from knowing whether to pay $500 or not.
The game show is a metaphor for our existence. To navigate life in the best way possible, we need to know stuff.. But learning costs time and money, and sometimes isn't very fun. So there is likely some knowledge which has costs outweighing the benefits. But the problem is compounded because we don't know which knowledge is worth knowing.
In my life, I've been influenced by modern skepticism, which values truth above all else. But there's not actually a contradiction here. Skepticism doesn't take the ridiculously dogmatic view that it is always worth spending the time to know everything. In fact, through their actions, skeptics implicitly acknowledge the cost of knowing:
1. Skepticism collects experts in many fields and topics and makes their knowledge accessible so that no one needs to know everything.
2. Skeptics teach and learn the general guidelines of rational thinking. These are easily understood, and have broad applicability.
3. Skeptics focus on pseudoscientific beliefs that are either a) especially harmful, or b) entertaining. Countering harmful beliefs may cost a lot, but the benefit is great. On the flip side, there's not much benefit to making fun of ridiculous beliefs, but that's a form of skepticism that is more accessible. It's like gateway skepticism.
As a skeptic, I think there are a few things that could improve a person as a rational thinker, but which are not necessarily worth the cost. For instance, Brute Reason recently discussed (and criticized) the idea that we should surround ourselves with friends that disagree with us. It seems possible that having friends who disagree with you could indeed help you change your mind towards more correct beliefs. But even if this were true, is it worth the cost of altering the whole way you make friends? Surely there are easier ways to keep your beliefs in check.
Another example is the idea that we should "keep our identity small". In the linked essay, Paul Graham argues that discussions over religion and politics get nowhere because everyone has their religion and politics tangled up in their identity. It seems true that people are less willing to relinquish beliefs that relate to their identity. But I contend that getting rid of identity labels may be too high a price to pay, if it's possible at all. Besides, it only mitigates bias if both sides give up their identity, and there's no way to force the other side to do it. There's got to be a better way to reduce identity-related bias.