Crossing the Line
There are lines I've always felt I had to toe
Some were blurry, some unseen
Some I've had to learn to read between
So many boundaries
Far more than you know—"Crossing the Line" (extended lyrics), Rapunzel's Tangled Adventure
Emily Cibelli, Yang Xu, Joseph L. Austerweil, Thomas L. Griffiths, and Terry Regier's "The Sapir–Whorf Hypothesis and Probabilistic Inference: Evidence From the Domain of Color" is a cool paper about how language affects how people remember colors! You would expect the design of the eye and its colorspace to be human-universal (modulo colorblindness and maybe some women with both kinds of green opsin gene), but not all languages have the same set of color words. There are some regularities: all languages have words for light and dark; if they have a third color word, then it's red; if there's a fourth, it'll cover green or yellow—but the details differ, as different languages stumbled onto different conventions. Do the color category conventions in one's native tongue affect how people think about color, in accordance with the famous Sapir–Whorf hypothesis? Maybe—but if so, how??
Cibelli, Xu, et al. discuss an experiment where people are briefly shown a color, and then try to match it on a color wheel, either at the same time, or after a short delay. People aren't just not-perfect at this, but—particularly in the delayed condition—show a non-monotonic pattern of directional bias: colors just on the "blue" side of the green–blue boundary are remembered as being relatively more bluish than they really were, but very similar colors on the "green" side of the boundary are remembered as being relatively more greenish than they really were. (Where what counts as "blue" and "green" was operationalized by asking the same subjects to rate colors on a "not at all" to "perfectly" blue/green scale.)
How to explain this curious pattern of observations? The answer is—Bayesian reasoning! (The answer is always Bayesian reasoning.) Our authors propose a model in which a stimulus is encoded in the brain as both a fine-grained representation of what was actually seen (this-and-such color perception, with some noise/measurement-error), and as a category ("green"). Then a reconstruction of the stimulus that uses both the fine-grained representation and the category, will be biased towards the center of the category, with more bias when the fine-grained representation is more uncertain (as in the delayed condition).
The model gains further support from a similar "two-alternative forced-choice" experiment, where people try to tell the difference between the originally-displayed color and a distractor (rather than picking from a color wheel). English speakers are better distinguishing between an original and distractor on opposite sides of the green–blue boundary. Speakers of Berinmo (spoken in Papua New Guinea) and Himba (spoken in Namibia) don't have the green–blue distinction, but the Berinmo wor–nol and Himba dumbu–burou boundaries fall between what English speakers would call yellow and green. And as the model predicts, Berinmo and Himba speakers respectively do better at distinguishing between original and distractor on opposite sides of the wor–nol and dumbu–burou boundaries!
In addition to superior cross-category discrimination, the model also successfully predicts a within-category bias. Suppose one stimulus, which we'll call A, is a more central example of its category than stimulus B. Then in the two-alternative forced-choice paradigm, it's easier to distinguish A as an original from distractor B than it is to distinguish B as an original from distractor A, because the exceptional case B regresses towards the mean in memory.
Regular readers of The Scintillating But Ultimately Untrue Thought know where I'm going with this! Why do we care about the further question of what "gender" someone is, if we already have fine-grained perceptions of how the person looks and behaves? Because our brains use category-membership as an input into predictions when our perceptions are uncertain.
If categories influence judgement on tasks as simple as remembering colors, then on theoretical grounds, I would expect the effect of gender on perception of people to be much worse (that is, larger), because people are much more complicated than colors. With colors, what you see is basically what there is: if your memories or perception of 500-nanometer wavelength light get rounded off slightly bluewards or greenwards depending on how many color words are in your native language, that's bad compared to what a well-designed AI with access to the pure, unmediated colorspace could perceive, but at least that bias is only acting on the one dimension of color. In contrast, your observations of a particular person are going to be much sparser than everything your brain might want to predict about that person. Under those circumstances, the dominant algorithm might end up eating bias in order to reduce variance by having your priors about what humans are like do a greater share of the work—work that relies on the ways (some blurry, some unseen) that female humans are different from male humans.
Transgender people are in a uniquely epistemically privileged position to observe this process, as the change from not-passing to passing is simultaneously a small one as far as the person themselves is concerned, and a large one as far as how the person is percieved by others. In a couple paragraphs that make me feel sad and jealous (I can't say dysphoric because I don't know what that word means), Julia Serano explains what it's like to cross that line (in Ch. 8, "Dismantling Cissexual Privilege", of Whipping Girl: A Transsexual Woman on Sexism and the Scapegoating of Femininity):
[W]hen I eventually did transition, I chose not to put on a performance—I simply acted, dressed, and spoke the way I always had, the way that felt most comfortable to me. After being on female hormones for a few months, I found that people began to consistently gender me as female despite the fact that I was "doing" my gender the same way I always had. What I found most striking was how other people interpreted my same actions and mannerisms differently based on whether they perceived me as female or male. For example, when ordering drinks at bars, I found that if I looked around the room while waiting for my drink (as I always unconsciously had prior to transitioning), men started hitting on me because they assumed I was signaling my availability (when I was male, the same action was likely to be interpreted simply as me scoping out the room). And in supermarket checkout lines, when the child in the cart ahead of me started smiling and talking to me, I found that I could interact with them without their mother becoming suspicious or fearful (which is what often happened in similar situations where I was perceived as male).
During the first year of my transition, I experienced hundreds of little moments like that, where other people interpreted my words and actions differently based solely on the change in my perceived sex. And it was not merely my behaviors that were interpreted differently, it was my body as well: the way people approached me, spoke to me, the assumptions they made about me, the lack of deference and respect I often received, the way others often sexualized my body. All of these changes occured without my having to say or do a thing.
Serano goes on to suggest that social gender exists, not in the way individuals perform gender, but in how others perceive it, and that therefore efforts to create a less oppressive world must involve dismantling cisnormative assumptions: "if we truly want to bring an end to all gender-based oppression, then we must begin by taking responsibility for our own perceptions and presumptions[; t]he most radical thing that any of us can do is to stop projecting our beliefs about gender onto other people's behaviors and bodies."
I can see how one might derive that lesson from the described experiences of transitioning, but I think it's ultimately a flawed generalization from a necessarily unrepresentative experience. The ways people treated Serano differently after she transitioned despite Serano being the same person the whole time, are not arbitrary: that happened because the fact that Serano looked like a woman, prompted people to use mental models trained against the distribution of adult human females. (There might be reasons going back hundreds of millions of years for primate mothers to become suspicious or fearful of males near their children.)
In the same chapter of Whipping Girl, Serano mentions that in her days of identifying as a male crossdresser, she found it easier to pass in suburban areas rather than cities, "where people were presumably more aware of the existence of gender-variant people." This also makes tragic Bayesian sense: transitioning to organically be perceived as the other sex is easier to pull off when it's unexpected, because the lower the prior, the less of a likelihood ratio you need in order to reach a given posterior probability.
The change in other agents' behaviors elicited by crossing the line into sending the signals of a different type is so dramatic specifically because it's a rare, off-equilibrium play. Lines between categories are placed in the no man's land between regions of unusually high probability-density in configuration space. If there were much more probability-mass just on either side of a line people are using to make predictions and decisions, then the line wouldn't be there.
