This week we have a special edition focusing on avatars, the Proteus Effect, avatar psychology, Perceptual Control Theory, the Avatar Manifesto, runaway AI, and more!
This issue is dedicated to Martin Taylor who passed away on Friday 13 March 2023. He was a brilliant man that helped me work through and build on Perceptual Control Theory. Even though I never had the privilege to meet him in person, he was generous enough to send me a brand-new copy of William Powers’ essential Behavior: The Control of Perception because I couldn’t get one in Finland. He’ll be missed. May his insights continue to live on. Find his magnum opus, Powers of Perceptual Control: An Inquiry into Language, Culture, Power, and Politics here.
In SN #048 I shared Heather Dodds’ article about being in AltspaceVR while the sun set on the platform. In it she had mentioned the “Proteus Effect”, which has interested me for a while now, so I wanted to use this edition to learn by sharing.
Introductory terms:
The word ‘avatar’ literally means ‘descent’ in Sanskrit and is a Hindu concept that refers to the physical appearance of a deity on Earth. It was brought into its current pop cultural form by Neal Stephenson in, you guessed it, Snow Crash. In more academic terms, researchers Nowak and Fox define an avatar as “a digital representation of a human user that facilitates interaction with other users, entities, or the environment.” So this could also include people’s social networking profiles, for instance.
The ‘Proteus Effect’ is a fancy shmancy way of saying that the way people’s avatars look and act in a virtual world can affect how they behave in real life. So, for example, people who choose more attractive-looking avatars report more confident and outgoing behavior. The effect is named after Proteus, the Greek god of "elusive sea change". Proteus rockin’ his kitten avatar (Image courtesy of Joh Orengo/DALL-E 2 and the scraped art of unknown artists)
Funny how we borrow from ancient myths and religions to make the unfamiliar familiar and reorient ourselves in this dizzy world of emerging technologies. Social Representation Theory has some interesting things to say about this social psychological process. (You can’t stop me now, I’m in full nerd mode. Make way!)
So if our virtual personas can effect us in the real world based on their appearances and what we ascribe to those appearances then avatar designers can influence us users through their avatar designs.
(I actually just started taking a free Coursera course called
that’s all about this. I recommend it if you’re interested. Dr. Rabindra Ratan makes a fun topic even more fun.)
‘Avatar psychology’ (the future discipline that I thought I invented but obviously didn’t) is all about the “interrelationship between avatars and human beings”, and it’s actually part of a bigger field of research looking at how people behave when they communicate online called Computer-Mediated Communication (CMC).
The Human Psychology behind the Proteus Effect
As per Wikipedia (because I’m being lazy), according to researchers, there are three psychological processes that lead to the development of this effect.
Behavioral confirmation which means that people tend to act in a way that proves what they already believe or expect about someone else. In this case, a person’s pre-existing beliefs and expectations about the characteristics associated with their avatar (or others’ avatars) influence their behavior in the virtual environment. The trickster is tricky. The heroine is heroic.
Self-perception theory which says that people figure out how they feel about things by looking at how they behave and the situation they're in. So if a person acts in a certain way due to their avatar's characteristics, they may come to believe that these attitudes, etc. are a part of their real-world self.
Deindividuation which happens when people lose track of who they are and what they believe in when they're part of a large group or crowd. It can make people do things they wouldn't normally do if they were by themselves because they feel anonymous and less accountable for their actions (like behaving more aggressively in a virtual environment, especially if their Murder Troll avatar is associated with violent aggression). “Deindividuation” (Image courtesy of Joh Orengo/DALL-E 2 and the scraped art of unknown artists)
Okay, I can dig it.
But, Joh, how would all of this look through the lens of Perceptual Control Theory (PCT), the “theory [that can] be applied to diverse fields”, that you first introduced back in SN #021?
Well, I’m glad you asked, dearest Spatialist.
But before we get into it you, here’s a
if you’re new to the game.
If you want to get to the nitty gritty, here is
The central premise of PCT
is that you act to keep something near a state you want (coffee sweet, feet warm, metaverse dream alive, and so on) despite opposition from the environment. This is the purpose of behavior.
In the process of controlling your perception, you stabilize some property of your environment (liquid’s sugariness, appendages’ temperature, hype level constancy, and so on) but “all your intentional actions are… only to control your own perceptions”.
I could stop there, but I’d like Martin to rock your world.
“Perceptual Control Theory (PCT) takes as a primary assumption that all you can know of the world is either embedded in your genes or has been gained through your senses. The world ‘out there’[beyond your nervous system and skin], assuming one even exists, might be very different from the world you perceive and might work very differently from the way you perceive it to work[…] If your actions influence your perceptions in a reasonably consistent fashion, then it matters little what is ‘really’ out there, because you can affect the way it seems to be, just as though your perceptions truly reflected reality.”
-Martin Taylor’s “Perceptual control in cooperative interaction” in the Interdisciplinary Handbook of Perceptual Control Theory
Let that marinate for a minute or two.
The Proteus Effect on PCT
From a PCT perspective, the Proteus Effect can be understood as a control problem. All avatar users (people) “have purposes, internal motivations, which are individual to themselves” that they want to reach or keep at some level. In an MMO, for example, where one person wants to gain more XP, the other wants to rack up gold, where one person wants to satisfy their bloodlust, the other wants to troll.
Each of these could be lower level goals related to how users view themselves in a particular virtual environment (maiden, miser, morlock, mage and more). These lower level goals could even be related to a user’s self-image in the real world, the maintenance of which could be the highest level goal in a human’s hierarchy of goals.
Just like the real world, the virtual environment (which includes other users) is dynamic, and, in one way or another, it disturbs (blocks, etc.) the achievement of these goals or the maintenance of these states. These disturbances can happen through the very nature of the game/virtual world (jump on moving platforms to get through a board, capture a flag, survive to the end, keep socializing with your friends, etc.), through other users breaking the rules, or through bugs and glitches in the system, etc.
Users have to act to make sure that what’s actually happening in the environment matches what they want to happen (get a badge, meet someone new, cause wanton destruction, stay the leader, etc.). This is what behavior is all about after all.
When a person ends a session in a virtual environment, those particular goal-seeking actions also end (for the most part). There are certain things you just can’t do in the real world due to physical, legal, and moral constraints (duh).
That doesn’t mean that the way one looked or the way one was treated or the way one acted to overcome certain obstacles don’t have an affect on one’s goal-seeking or self-image in the real world (there are people learning real skills and making real money but also getting real eating disorders from their time online).
From a PCT perspective, my hypothesis (though not super sophisticated) is that manipulating one’s avatar in a virtual environment boosts one’s goal-seeking, including the maintenance or enhancement of one’s self-image, despite how skewed it might be, in the real world, which is why the avatar’s ‘influence’ carries over via the Proteus effect.
<Monster Sentence achievement unlocked!>
Basically, your avatar will effect you in the real world if it helps you maintain a certain perception you have (or want), especially if that perception is of yourself.
In this way I’d say it’s similar to listening to upbeat music (“I want to stay happy”), drinking 5 cups of coffee (“I need my hit”), or doing jazzercise (“I want to feel healthy”) because one wants to perceive oneself as a happy, productive, and/or healthy person.
5 cups of joe listening to upbeat music and doing jazzercise (Image courtesy of Joh Orengo/DALL-E 2 and the scraped art of unknown artists)
Each of the psychological processes associated with the Proteus effect such as behavioral confirmation, self-perception theory, and deindividuation can be reframed from a PCT POV, too, but I’ve gone on long enough so I’ll wrap it up for now. <*Cough* Cop out! *Cough*>
When it comes to designing avatars, designers can use PCT to understand how people use avatars to achieve their goals. By allowing users to customize their avatars and giving them control over their avatar's actions (to represent their ideal self or allow them to fit in with a particular group, for ex.), designers can help users feel more connected to their virtual environment and more invested in their experiences. Cha-ching!
The day after I finished the bulk of this newsletter Kavya Pearlman shared
Authored by Dino Ignacio, a Senior Design Manager at Roblox, the manifesto has 3 principles
1️⃣ Equity over Exclusivity 2️⃣ Agency over Automation 3️⃣ Clarity over Efficiency with the following guidelines: ✨ Avatars should always be driven by a user's intent. ✨ Avatars should have a way of indicating when they are asynchronous. ✨ A user must always have a choice over using tracking features when using their Avatar. ✨ When a user starts to build an avatar, the default should not bias toward a specific gender or ethnicity. ✨ Features that allow a user to create authentic self-representation, like skin color or basic eye shapes, should always be equitable.
(Thanks for the emojis, Kavya.)
Let’s continue this fascinating discussion. (Perhaps, I’ll have to write a part 2 once the Avatar Psychology for Designers course is over.)
Andrew Ng, “one of the world's most famous and influential computer scientists”, posted
“There is no realistic way to implement a moratorium and stop all teams from scaling up LLMs, unless governments step in. Having governments pause emerging technologies they don’t understand is anti-competitive, sets a terrible precedent, and is awful innovation policy. […]Let's invest more in safety while we advance the technology, rather than stifle progress. A 6 month moratorium is not a practical proposal. To advance AI safety, regulations around transparency and auditing would be more practical and make a bigger difference.”
This.
Plus, Theo Priestley adds,
“the call for a 6-month moratorium on the development of AI quickly became meaningless in the face of a growing, global developer movement to experiment and release versions with zero ethical guardrails in place. It's already too late for this particular battle, in a way OpenAI has well and truly beaten Google because it has recruited a willing army on its side and I'm not sure what's worse - the blinkered humans rushing blindly to be a part of something or the AI sitting there taking it all in.”
Runaway AI aka Lil Terminator Looking for a Home (Image courtesy of Joh Orengo/DALL-E 2 and the scraped art of unknown artists)
Don’t fret though.
The Singularity
the point in time when AI surpasses human capabilities, can’t happen
“until we employ an approach that views behavior from the perspective of the system and as a means by which perceptual goals are controlled will we achieve systems that are truly autonomous and intelligent”.
- Rupert Young, “Robotics in the real world: the perceptual control theory approach” in the Interdisciplinary Handbook of Perceptual Control Theory
Oops… Did I just let the artificial cat out of the bag?
<Your Spatial AD Here>
This space for rent!
Thanks for taking another trip on the feedback loop with us, Spatialists!
“All people are like grass, and all their glory is like the flowers of the field; the grass withers and the flowers fall…”
-1 Peter 1:24 (New International Version)
Don’t wait to let someone know that you appreciate them or their work. Do it today.