I think you're right about the expectation of a "yes" or a "no" leading to comfort/discomfort making a request. If we're certain of a "yes," there's no reason to fear asking; if we're certain of a "no," there's no point. So once again it's the gray area in the middle that we need to address. A perfect mental model of the other person will of course remove any ambiguity, but really that puts us back where we started: in the absence of that model, looking for heuristics.
Another reason we worry about making requests is that even if the request is reasonable, we might have been imposing too often, and the sum of the requests could become unreasonable. Requests also expose us somewhat; to say "I need something, will you help?" is to make yourself vulnerable to whomever you're asking. The more confident we are we can trust that person, the easier it is to make the request. That's another source of ambiguity. I think there's also an idea in our culture that we shouldn't be "needy," which pushes the apparent line of "unreasonableness" to cover some requests which would be healthy or even necessary to make. And there's a more specific cultural pressure not to display emotion or to scoff at emotional display--especially if you're male. Finally, sometimes making a request really does have the potential for consequences: the canonical example is that asking out a friend might damage the friendship. Asking the huge angry-looking drunk guy to get out of your way might earn you a fat lip.
That's just a brainstorm of answers to your question, "why do we sometimes feel uncomfortable requesting something?" I haven't thought much about potential solutions or workarounds for them yet.
I don't think people being "unable to say no" is as uncommon as you think. I have friends of whom I say that I "don't trust their yeses," i.e. I'm not confident that if they say yes to a request it will be because they actually mean it.
The most common situation I can think of where this comes up is when the question takes the form "Do you mind X?" or "Is it okay if Y?" The phrasing of the question clearly "wants" a positive answer, which applies a little bit of social pressure to give one. For people who aren't good at asserting their boundaries or their needs, it's difficult to overcome that pressure. I am pretty good at thinking about and asserting boundaries, and when someone asks me a question like that, I have to catch myself so as not to instinctively say "yes" or "I don't mind" before I've actually consciously considered whether I do mind.
It's sometimes possible to predict when this might be a problem. When I see someone expressing discomfort after acquiescing to something, or leaving the room after saying they didn't mind me playing music in it (or whatever), I'll make a mental note not to trust that person's yeses. This widens the buffer between "clearly okay" and "clearly not okay," i.e. the gray area in which I'm uncomfortable asking for something.
That situation is a tough one to remedy. It's not my responsibility to guess what someone else wants when they aren't expressing it, or to teach them to assert their boundaries. However, I am able to help create an environment in which it's safe to say "no." I can show respect and appreciation when someone else makes their needs clear, as well as make mine clear without placing responsibility on other people to meet them. Of course, that's how all of this started--my desire to make my own needs clear!
It bears noting that one of the people I've been discussing this with off-site thinks I bring way too much concern for other peoples' opinions into my choices about whether to make requests or assert needs. He might be right, but if he is, analyzing the reasons for it should only bring me closer to a rational threshold.
I'm starting to be tempted to write a post just about this question, since it's generated so much discussion. But I'd like to have something closer to an answer to it first.
I'm working on a list of question types which come up frequently in day-to-day life but which I haven't yet found a reliable, rational way to answer. Here are some examples, including summaries of any progress made in the comments.
Practice the ability to judge how important something is to change, making sure to examine your criteria of importance. Identify the reasons you want to change it, and try to normalize your emotional response to the facts. Learn about the difficulty of changing a thing by investigating other peoples' attempts to. Be aware that, the less one knows about a field, the less one is able to judge how difficult a task in that field is. Ask an expert if you need to. Another heuristic for difficulty of changing something is that the closer it is to one's own mind, the more control one has over it. When you know as much as you can, do a cost-benefit analysis.
Unreasonable requests are those which would only be fulfilled if the asker had power over the askee which they do not, which represent an unequal exchange between equals, or which are not actually possible. We don't want to make unreasonable requests because they are at best unfair social impositions and at worst rude and damaging to relationships. This is complicated because requests between friends aren't about direct exchange; it's expected that sometimes one person will need help, and sometimes the other will, and in the long run it'll even out. In a strong friendship where both people have treated each other well, it's more appropriate to ask for a large favor than it would be to ask a relative stranger the same thing.
The difficult requests to judge are those where the power balance or strength of the relationship, or the values of what's being exchanged, are unclear. That is, they require a more accurate judgment of either the relationship itself or of the other person's needs and abilities than one is confident of making. As in the previous question, specialized knowledge can help predict how much trouble a given request might be. In the specific example above, one would need to understand the cost of lending an ear, in terms that could be compared to the benefit of venting. Prior communication is the best way to achieve this; basing an estimate on similar past situations is also good. Knowing nothing else, use an assumption of equality and the basic responsibility for oneself as heuristics. Finally, how close you can get to the edge of what's acceptable may depend on how much you trust the other person to tell you if your request is not reasonable (rather than acquiescing resentfully).
It's possible to minimize the necessity of choosing between these two things by doing work which is enjoyable and taking breaks after earning them. When the two types of activity do conflict, one way to get around it is to use time that wasn't available for work anyway to do unproductive things. Another is to have (and frequently review) clearly defined medium- and long-term goals, and weigh short-term choices against them. Doing this regularly may make it easier to judge activity choices on the fly.
No comments yet, but here are some questions this raises for me: What does it mean to have reached your peak in a skill--is there actually a maximum amount you can usefully learn and practice, or just a (potentially variable) point of diminishing returns? Is it possible to know there's more to learn but not be able to learn it?
So far, the only mitigating factor I've found is my overall physical and mental state. Being hungry, tired, or stressed makes it easier to fall into the anger cycle and harder to get out. Therefore, taking care of myself in general helps to prevent it, but it's not always possible to remedy those problems after the cycle has already started. When circumstances permit, physical activity may provide an outlet for the energy that keeps this cycle going.
A conversation offsite led to the following: Manipulation involves both deliberate instigation of emotion and trying to persuade someone to do what you want, but isn't defined by either. (The first one describes gestures of affection, and the second includes ordinary debate.) The definition we settled on was "using emotion to bypass someone's normal decision-making process." Trying to get someone to do what you want is not inherently manipulative; trying to make them feel something so that they will do what you want is.
Naturally I'm looking for ideas about how to answer these questions, including links to earlier thoughts about them[2], but you get bonus points for supplying actually usable heuristics, rather than just opining on my examples. But I'd also like to hear it if you've got any questions of your own that fit this form. Consider it a sort of lowbrow subset of open problems--difficulties you're aware of having on a regular basis but haven't yet been able to solve.
(Tag suggestions are appreciated. I'm unaccustomed to using content tags, so I made some guesses based on the site's tagcloud and what's on the similar Open Problems post.)
[1] I actually quite like the Serenity Prayer, despite being entirely nontheistic, because it presents a set of traits to aspire to for specific purposes I can get behind.
[2] Until I've read the entire LW archive, I'm constantly paranoid that anything I post will be a second-rate rehash.