Optimal Distinctiveness Theory
In which we explore the wild world of self-concept
Optimal Distinctiveness Theory
This post is one I do periodically that focuses on a psychological theory I find interesting in itself, rather than how it has something to do with privacy. I do think this has some applications there, but honestly about 90% of the motivation for this post is I just think this is a cool theory I don’t see many people talking about.
Fair warning, this theory is surprisingly easy to mis-understand. I have, myself, had to pretty radically re-adjust my understanding of it at least twice over the course of reading about it and writing this post, so while I’m pretty sure I’ve got a decent handle on what it’s about, it’s possible I’m still misunderstanding it. It’s not that it’s complex as such, so much as it’s just kind of easy to assume what the theory is or what the writers mean, and gloss over the rest.
Before we get into this theory itself, we need to do some quick definitions and explanations of some ideas.
Identity
Identity is… complicated, and can be viewed in many, many ways. I’ve mentioned narrative identity before – where we define identity as (loosely speaking) the stories we tell ourselves about the experiences we’ve had – but here we’re taking a looser perspective. For our purposes today, identity can be thought as more-or-less synonymous with “self-concept”, or how you understand yourself to be.
I think this is a useful way to think about this: have you ever had a friend telling a story about a situation you were apparently there, but you either don’t remember at all or only the vaguest hints of? And they say something like “oh, and then X did such-and such!” and you think “I don’t remember doing that, but it does sound like something I would do”. Self-concept is the thing you’re drawing on when coming to that conclusion. Like, if Mel Gibson said something bigoted again, I wouldn’t be surprised because that’s pretty much the picture I have in my head of how Mel Gibson is. Same thing, only about ourselves.
Now, humans are social creatures [citation needed], and a lot of our identity tends to be tied up in the groups we’re a part of. In collective cultures this can be quite pronounced – rather than saying things like “I’m a psychologist”, people might say “I’m a member of this psychology department” – but it’s present in basically all people to some extent or another. The groups we’re a part of tend to make up part of our self-concept; religious community, sports team (or indeed group of supporters), what kind of smartphone you use, family, colleagues, whatever. This can be bad in extreme cases, of course – cults seek to over-ride people’s identities to the point where the cult is the only group they feel any connection to – but in general is perfectly good and healthy.
I want to draw a hard (if somewhat fictitious) line between what I’m going to call “belonging to” and “identifying with” a particular group. “Belonging to” I’m going to say is largely externally defined – people with some particular constellation of biological traits “belongs to” the group of “male” or “vision-impaired”, for example. “Identifying with” is more internal – for example, I’m technically a biological male, but it honestly doesn’t really affect my self-concept that much. If I woke up tomorrow as a biological female, I’d probably be somewhat annoyed because I enjoy the physical benefits of a biologically male physiology (greater strength, don’t need to worry about menstruation or pregnancy), but it wouldn’t radically upset my sense of self that much because that doesn’t play a large role in my self-concept. But if I were to be somehow prevented from, say, doing research about people or reading about psychology, that would seriously throw me, because they are a large part of my self-concept.
Now, obviously these two do affect each other in serious ways, and some concepts only make sense as identities rather than group membership (religious identity, for example). But I wanted to create that (somewhat false) distinction just to really clearly illustrate what I mean when I’m talking about identity.
Another example: I was born in the country I live in, I grew up here, I have only ever lived here, and I am a citizen of this country. I “belong to” the group of People of This Country. However, while that has absolutely affected my development and worldview – if I grew up in communist China, for example, I’d be a very different person – the fact of that belonging doesn’t really play a large role in my self-concept. If I somehow was magically otherwise identical but now a member of the People of Other Country group, I would still recognise that person as me.
Which factors come into people’s identities are deeply individual and variable. I have friends whose gender is really really important to their identity, while it’s barely a footnote in mine. Meanwhile, some of my hobbies make up huge parts of my self-concept, even though they objectively have a tiny impact on me in any material fashion. Where I work has a huge impact on my life, but makes up a tiny part of my self-concept.
I’m starting to belabour the point so I’m going to stop now, but I hope I’ve made the distinction very clear. Identity, especially social identity which is what we’re talking about today, is distinct from but related to what groups we happen to belong to.
Optimal distinctiveness theory
So, we’ve established that simply belonging to a group is very different to identifying with a group. We all belong to lots of different groups – we’re members of a family, we’re citizens of a country, we’re workers at such-and-such place (or own such-and-such place) – but not all of those groups have the same impact on our self-concept. So, the question is: why do people identify with the groups they do, rather than others?
Optimal distinctiveness theory (ODT) says that part of the reason is that there are two important elements in what groups we identify with: how inclusive the group is, and how distinct it is. I’m going to use “membership of a workplace” here as a sample, but these apply to all social groups. Also note that how inclusive or distinct a group is viewed as varies over time and over people – I might think a group is super-inclusive, while Bob thinks it’s really not. This can be because the group changes, because we change, or because of the context we’re judging in changes, or any number of things.
Inclusiveness can be basically thought of as how “safe” we feel as a member of that group. Am I constantly worried that I’m going to be fired at any moment, and that I need to constantly work to avoid that situation, or am I absolutely essential to the company and cannot be fired without basically the whole thing crumbling down? The more safe you feel in that membership, the more inclusive the group is for you. It makes sense that a group you feel more “safe” as a member is going to play a larger role in your self-concept – if I make “working at this job” a part of my self-concept, but that state is precarious, I’m not only dealing with the usual anxiety of being able to pay the bills, but also my sense of self is going to be precarious and uncertain, which causes lots of problems.
Distinctiveness is “how unusual the group is to have membership”. This does not necessarily mean it’s good or prestigious; it can be useful to think of this as “unusual”. To a point this stands in opposition to inclusiveness – working at a small start-up with three other people is more distinct than working at a multi-national corporation that employs thousands of people, so the former is more likely to form a part of my self-concept, simply because it’s a more unusual descriptor that helps describe me as a unique person. So the more inclusive a group is, the more likely it is to have high membership, which makes it less distinctive or unusual. However, the more distinctive it is, the more likely it is that membership is conditional in some way, which means I’m probably going to feel less “safe” in any identity incorporating that.
This is an important point to focus on – it’s not that “people want to belong to groups that are distinct but not too much”, but more “the groups people use to make up their self-concept need to be unusual enough to be useful descriptors, but also they need to feel secure in their membership of that group which tends to trade off against the groups distinctiveness”. Sometimes people will seek out groups to belong to to feed this need, but more likely they’ll go through the groups they already belong to and the ones that are more distinctive will tend to become a larger part of their self-concept. For example, I’m probably not going to go try to become a really unusual profession just to feel more unique as a person, I’m probably just going to think of an unusual hobby I have and use that.
As with all of these things, it’s entirely the person’s perspective that is important, rather than reality. If I think supporting my local sports team is distinct, I’m going to be more likely to factor that into my identity even if it’s super-common. If I think my job is insecure, it’s unlikely to form a part of my self-concept, even if I’m actually in no meaningful danger there. So things like anxiety etc can play a huge role here.
While this is a relatively newish theory – the main foundational work came out in 1991, and it’s never been the Sexy Cool Topic leading to a glut of papers – there’s some evidentiary basis for this idea. People from racial minority groups are more likely to describe themselves in terms of their race when prompted to describe themselves, rather than those from racial majority groups, and boys from (numerically) female-dominated homes are more likely to mention being boys.
Now, maybe it’s because people from racial or gender minorities have their racial or gender background as just more relevant, so it’s obviously going to be on their mind. I’m sure that’s part of it, but not all. People randomly assigned to groups and told that their group was in the minority (more distinct) were more likely to identify with that group than if they were told the group was in the statistical majority, and were more likely to actively think about it. This is even when their group status has absolutely zero impact on their lives.
But maybe this is because humans evolved to be concerned about being in the minority because people in the minority have less support and are less likely to do well. Again, that might be part of it, but not all of it. People who are (randomly) assigned a majority and a minority status tend to select the minority status when given a choice – one study (p. 74) found 78% of the time. And when given a choice between two groups to join (one majority and one minority) people tend to choose the minority group. If the evolutionary/anxiety argument were true, they’d join the majority group because it’s “safer”.
So what?
Well, this idea has some pretty significant implications. First, if people want to feel distinct, they may exaggerate the differences between various groups to make their group more distinct. For example, a woman might say men are very different to women in important ways (increasing distinctiveness) while saying women are more similar to each other, especially in ways that suit their self-concept (increasing inclusiveness). Experimental work suggests this happens with academic discipline as well – what the group is is less important than the needs being met.
This has further flow-on effects as well. People tend to estimate the size of their identity groups significantly differently depending on what need (inclusiveness or distinctiveness) is being threatened – if people’s status within the group is being challenged, they tend to over-estimate the size of the group. If the group’s uniqueness is being challenged, they under-estimate the size (in that study. This has serious political implications which I’m not getting into.
People whose identity needs are being challenged react in ways to counter that challenge. So if say a person who identifies as being Group X hears someone say that “Group X is, like, super-basic, it seems like everyone is Group X now” is probably going to feel like it’s not that distinctive. So maybe they act in ways that exaggerate that distinctiveness, either through highlighting statistically unusual characteristics of Group X members, or maybe through describing other groups as being very different.
How does this relate to privacy?
I mean, I think I’ve been pretty open about the fact that it doesn’t really. But one way I think it can is in relation to how we think about the privacy “community”. Although it is definitely kind of a garbage fire in many ways (speaking generally, this tendency is by no means evenly distributed!), there is a community around privacy. There’s subreddits, Discords, Matrix servers, Mastodon instances (I’m guessing, honestly I’m still pretty vague on what Mastodon even is or what purpose it serves), websites, forums, Substacks (clears throat), all kinds of ways. And I think it’s fair to say a large portion of the people on those gathering places view those groups as part of their identity – some to a dysfunctional extent, but many in a normal and healthy way. But the whole Thing about privacy is that, done sensibly, is that it pertains to specific pieces or kinds of information. I’m willing to bet that for many people, the kinds of information they want to keep private are tied to their identity and self-concept in some way. I’m less concerned with people knowing my biological sex than I am with knowing what community groups I’m a part of, partially because it’s less important to my self-concept, and partially because one is more distinct – that is, identifying – than the other; both to other people, and to myself.
A person who belongs to a religious or political minority is more likely to want to keep that private even if they have no real concerns about persecution than members of majority equivalents. I kind of gestured at this idea when I wrote about psychological ownership, which is basically just a way of saying “this information is important to me”, which means “it forms a part of my self-concept”.
This is, I think, very important when threat modelling. You can’t control all information about you, so you have to decide what’s important. In privacy there’s a lot of concern over whether a particular piece of information can be “identifying” to you, and in many (although by no means all) cases this comes down to some kind of group identification or membership. Membership to a particular religious community (or non-membership, even) seems a good example here, but I’m sure you can fill your own in.
It also suggests that certain types of “violation” are going to be reacted to differently. If my membership of a certain group is important to my identity and that becomes (unwillingly) public, then I’m probably going to react very differently than if, I don’t know, the fact that I’m a customer of Power Company X becomes public.
But mostly, I just think it’s an interesting theory that really helps shine a light on how people think about themselves in a range of contexts.