I do mutual escalation if I'm really into them, approve/disapprove when I'm not sure if I'm into them, or like if they'd be safe to flirt with. But even with mutual escalation, I usually match what they (the guy) is doing, maybe go a tiny bit further, depending on how I feel that day. This is likely confounded by how I can get attracted to someone if we spend a few hours talking, even if I wasn't attracted when I first met them. I think I was more likely to do mutual escalation when I was younger too

What I really want to know is:

• Conditional on mutual interest, what are the relative frequencies of these two kinds of flirtatious escalation?
• Are there particular settings/cultures/etc in which those frequencies are very different?

Preamble: The below assumes interactions between a heterosexual man and woman, can't comment usefully on other scenarios.

I have no idea about the answer to the first question, but would bet serious money that the answer to the second question is yes.

The reason why, in previous conversations, I had said I personally wouldn't escalate much without some positive response from the person I'm attempting to flirt with, is because of my personal ethics and expectations around social conventions, which I assume are culturally influenced. I am strongly averse to being pushy about sexual interest, and expect the most likely response if my interest is not reciprocated to be the equivalent of silence - not giving some definitive indication to me the equivalent of "I'm not interested in you, please stop" because that risks a very negative reaction (from a generic male, not from me specifically, but the person I'm flirting with in this hypothetical has no way of knowing what I'm like, and would understandably prioritize safety over clear communication). My way of analyzing a response is, if I get no response, that means stop, if I get a negative response that also means stop, only if I get a positive response of some kind is it OK to continue.

Obviously, this is not the only way the ambiguous signal of "no response" could be interpreted. In a culture where it is strongly expected that women do not express sexual interest of any kind because that is shameful, the dynamic would necessarily be different, because a non-response is the expected response even when the woman being flirted with would like to indicate interest. Similarly, if the social expectation was that if someone would prefer I stop flirting they will say so, I would feel more comfortable interpreting a non-response as "it is OK to continue". I imagine there might be a culture where that expectation exists, although I don't know of a specific example.

Also relevant: I expect (due to gender stereotypes confirmed by conversations with women around me, and comparing their responses to my own under the assumption that I'm not too different from most men) women to be more attuned to social subtleties and subtexts than men. Basically, their sensitivity to potential alternate interpretations of the meaning of words or actions that are not surface-level obvious is higher. As a result, I expect them to be able to pick up my indication of interest fairly easily, as long as I'm being not-particularly-subtle by my own estimation. What I'm doing will be plausibly deniable, but still likely clear. I also expect women to do things they think of as not particularly subtle and easy to pick up on (which another woman might notice) that are completely missed by the person they are attempting to flirt with. There are many instances where a woman interprets a behaviour as "this person likes/supports/does not like/does not support me" and my response is "that is just one possible interpretation, what you're reading something into might have been unintentional or not related to you. You could be right, but I would put a lower probability on it than you seem to be". This is particularly true in terms of sensitivity to subtle indications of dislike/opposition, but true more generally.

Speculating about cultural change over time, I expect that our (US/Canada) past culture, which put more pressure on women not to "be too forward", and gave much more latitude to men to pursue women both subtly and directly, would have been more permissive of a man fumbling his way through flirting, getting no response, and continuing regardless, than our current culture is.

Re: your third prior, "People who are both bad at it and know they’re bad at it are usually very hesitant to send escalatory signals (if they even know how to, which they might not)"

I think this falls into a binary distribution. The experience of women I know is that when in their early 20's, they would get a large number of very clumsy attempts to indicate sexual interest. At the same time, most men tried flirting only rarely. From the typical young woman's perspective, the experience seemed to be "I get hit on constantly by distasteful men", while from the typical man's perspective it was "I rarely try anything with anyone, I don't want to come across as one of the men that causes problems for women by being an ass". I reconciled this by modeling the situation as "among those who are low-skill at flirting, which is most people, there are a majority of people who flirt rarely and selectively, investing a lot of effort and emotional weight into each attempt, and a minority of people who take a shot at everyone who crosses their path, basically taking the same approach as a spammer where a 0.1% success rate still yields a sufficient number of successes to be worthwhile". The experience of the recipients of the "flirting" (although arguably that isn't what's happening, it doesn't have the subtlety or skill you're thinking of) is that ~all the attempts are terrible by men who aren't interested in them as an individual at all, and this influences their impression of men in general. In any case: There are at least three possible responses to "I know I'm bad at flirting" - one is to spam the world with low-effort attempts to get a positive response, another is to be very hesitant to attempt to flirt, a third is to attempt to skill up. I've seen all 3.

I think that looking at flirtation in isolation is a mistake. Flirtation is a behavior attached to some motivation. That motivation might be coercive ("I want to sleep with that hot person, let me flirt with them and maybe it will happen") or it might be a genuine express of interest ("I really like that person, I'm going to go talk to them and not hide how I feel", which comes across as flirting) or it might be something else ("The person in the lab coat told me I have to flirt with a person for 10 minutes as part of a study to get extra credit in Psych 101").

So I don't exactly know how to answer your question, but I read this feeling like it's leaving out the most important questions:

• Do you have a romantic interest in a person?
• Do you want to sleep with them?
• Would you be open to romance/sex with them?
• How safe do I feel letting the other person know how I feel?

That study doesn’t directly address my main question, but it is at least strong evidence that the large majority of undergrads are not very good at this whole flirting business

Indeed, the solution is to create masters programs for flirting!