It seems that out of all the options available, humans are wired to always make the ideal (satisfying/rewarding) subjective choice, and not go with the second best option or some other.

When a person has a list of options to go with, ordered by preference, if he/she would go for what seems to him/her as the second best or other option, that specific choice would actually be the -real- #1 option, because the criteria, that made him go for #2 are more important then those which sorted his options initially.

Generally, going on a vacation (#1), is perceived as a better choice then going to work (#2). Unless, finishing a project results in extra money, or a sense of accomplishment, which would subjectively make the latter choice a real number one. And so, if staying at work would then be perceived as more important/rewarding/satisfying, would that person go on a vacation after all? And if they did, would that mean, that vacationing was the real #1 priority?

So is it possible to settle for a second or later best choice, if we have the options in correct order, based on the most important criteria?

  • Just a comment... many people (e.g., in arguing against altruism) tend to argue that we always pick our #1 preference. I have a problem with the unfalsifiability of such a thing (one can always say what one does is #1; one could even define it this way, but IMO that makes the rank meaningless). Perhaps a better way to look at this is to look at a negotiation (e.g., where to eat with a loved one); in such cases it's hard to argue that you're going with your own #1 preference (one could argue negotiating is the choice, but there's the problem of why you go there).
    – H Walters
    Apr 27, 2019 at 11:15
  • @HWalters I think I see your point. When choosing the restaurant for yourself, your #1 choice would naturally be your #1 favourite restaurant. But on hers/his birthday the criteria shift so that number one choice simply isn't your #1 restaurant as usual, but his/hers . Your objective here is to make your loved one happy, and so you will choose a place, that meets it best. But not second best. Different factors, such as opening hours or prices, may come into play. This would shift your criteria and a different place would then be perceived as best, but again, would you be able to go with #2?
    – marko-36
    Apr 27, 2019 at 14:26
  • 3
    I take it you don't have kids :)
    – Richard
    Apr 27, 2019 at 23:18
  • 1
    @Marko36 well thats all very dawkins, but i know the decisions i make are not the best for me personally, but for them. Some other human beings.
    – Richard
    Apr 28, 2019 at 8:08
  • 1
    @Marko36 not really. Many people choose to leave their families and resume a life lived solely for their own pleasure. I don't enjoy standing by a cold soccer pitch watching my kids play badly. I don't enjoy cleaning up sick at 3AM. I do it in spite of myself, because I love my children. I take your point, but sacrifice of any kind is intellectual not visceral.And long term sacrifice, knowing you'll probably be put in a retirement home as a reward, is difficult to reconcile with your premise. It requires a particular kind of honour code
    – Richard
    Apr 28, 2019 at 9:21

4 Answers 4


There's a complicated issue that arises in the concept of "ordered by preference."

On one hand, one may define this phrase such that what you describe is tautological. This is the argument used by the phrasing "Well you must have really wanted X all along." This phrasing is indeed useful for developing formulae approximating one's preference.

The other is that we define a priority order for options before the choice is made, and then a human exerts "free will" to decide between them. In this case, the devil is in the details. Given some unbounded amount of decision making time, we like to believe that humans will pick their first preference. This is at the root of why we model humans as "rational actors." They try to maximize some internal goal.

However, in practice, humans do not have unbounded time. The time one spends thinking affects the outcome, as the world keeps moving. As such, humans rarely follow this simple rule precisely. Consider the example of police officers who make a split second decision as to whether to shoot someone or not. It is highly unlikely that they will make a decision based on the objective truth of the entire scenario because they have not had time to analyze everything. The Trolley Problem is an excellent way to explore this: there's always some nuance that you haven't considered which could matter when you find yourself holding the controls of a runaway trolley.

If we try to mix these two, we see that a rational human must fundamentally be aware that, in the moment, they will make a decision based on incomplete information. Given enough time, humans will try to optimize their lives to "win" at this decision. But at this point it gets sufficiently murky that it's hard to keep going with the model of a simple ordered priority list. The long standing philosophies surrounding death show that we've got some really complicated thoughts regarding what our most important decision might be.

So, as such, I would say that if one uses the tautological pattern for priority, then we always follow it. However, if one takes one step off of this extreme, towards the idea of having a pre-determined ordered priority list, humans begin to do things which are not modeled well with this priority list.

  • Furtermore, if one takes the tautological view that "most rewarding" means "the one chosen", then it is not something we are wired for, as the question puts it, it just is so by definition... WRT rational choice, I do not think it is just a matter of insufficient time or information, as individuals are quite capable of holding contradictory opinions, desires and ethical views.
    – sdenham
    Oct 27, 2019 at 22:14
  • 1
    @sdenham As a skeptic, I feel that if an individual wishes to act rationally, thus resolving conflicting opinions, they can dig into the language and definitions behind their rational thought until they are murky enough to permit a resolution of the conflict. This may, of course, be on the time scale of science fiction with people who live "forever" (such as The Doctor from Dr. Who), so practically I would have to agree with you that we can hold contradictory opinions that cannot be resolved even if there is a desire to do so.
    – Cort Ammon
    Oct 27, 2019 at 22:47

Humans are capable to make their choices subject to an algorithm, such as a coin toss. If you are willing to follow the coin toss, then if the coin toss decides for the second best, humans can choose second best, fully aware of what they would have chosen without the coin toss.

Socially this is important as an example for group cohesion, where group members will follow the decisions of a leader (or a democratic majority) over their own preferences.

  1. Economists and rational choice theorists recognise the phenomenon of 'satisficing' where we do not go for the best (by relevant criteria), but for what is good enough or merely acceptable or satisfactory rather than the best possible or optimal - or in your terms 'the most rewarding option' (again relevant criteria).

  2. There's also the point that we can be mistaken, objectively or just by our own standards, about what is 'the most rewarding option'. We can't 'choose the most rewarding option' if we are ignorant of or mistaken about what that object is.


If one assumes that human choices are the result of an optimization algorithm as one would find running in an automated logic machine where what is "most rewarding" is optimized, one would expect the machine and the human to make the same choices.

However, such machines run on completely formal, "explicit knowledge" rather than "tacit knowledge". Further if one assumes that for a human being (not a machine) inference is rooted in that human's tacit knowledge, as Michael Polanyi suggests, then there may be a discrepancy between what a machine optimizes and what a human chooses.

Here is the question:

So is it possible to settle for a second or later best choice, if we have the options in correct order, based on the most important criteria?

There may be no correct order based on the most important criteria for the human to choose from resulting in the human making a choice based upon tacit rather than explicit knowledge, or, in other words, on non-verbalizable concepts rather than formal concepts. That human may then settle for what another human might consider to be not the best choice.

Polanyi, M. (1966). The logic of tacit inference. Philosophy, 41(155), 1-18.

  • I think I see your point. (And thank you for the edit, btw.) But is it important, whether the knowledge is explicit or tacit? It would actually be the latter, and so - yes a "best" choice is diiferent for different people. But the point is, if a person has a subjective number one choice (I believe so, regardless of the process / knoledge it made it a number one), can he go ahead with the second?
    – marko-36
    Apr 27, 2019 at 16:26
  • @Marko36 If one is going to have a "correct" order then one would need explicit knowledge. A machine should be able to validate the order and best choice if it is possible to create the ordering. Tacit knowledge allows the human to alter that order and choose something the machine would not. In order to go ahead with the second option, the human has to have an ordering. We assume we have this ordering because we assume we are making rational choices, but our choices may not be rational in the sense of reasoning from propositions of external knowledge. Apr 27, 2019 at 16:34

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .