Empathetic models for tulpas - the standard?

[waffles]

If you hadn't seen, Kaj Sotala, a Finnish academic, recently gave a talk on tulpas at a conference, outlining this own hypothesis:

I hypothesize that tulpas may arise from the combination of three factors. First, conscious thought acts as a "reality simulator", and imagining something is essentially the same process as perceiving it, with the sense data being generated from an internal model rather than from external input (Hesslow 2002, Metzinger 2004). Second, our brains have evolved to be capable of modeling other people and predicting their behavior, so as to facilitate social interaction. Third, according to the predictive coding model of the brain (Clark 2013), action and perception/prediction are closely linked: doing something involves us predicting that we will do it, after which the brain carries out backwards inference to find the actions that are needed to fulfill the prediction.

Abstract, slides, Reddit thread.

 

Here's another one, not quite as recent. A Samuel Veissière, cultural anthropologist, gives some thoughts:

In the first wave of CSR theorizing, the inference of supernatural agents from the world around us is explained as more or less inevitable features of cognition; namely a tendency to attribute anthropomorphic animacy and agency to living things and inanimate objects alike.

[...]

In this model, humans are said to reason about supernatural agents by expecting them to reason like humans, particularly in terms of goal-directedness, shared intentionality, intuitive physics, naïve psychology, and semantic and episodic memory.

[...]

Could it be, then, that [...] our agent detection and projection abilities enable us to recreate personhood attributes with more intuitive precision in the absence of physical designata?

Paper here.

 

Finally, there's this.

 

 

Look alike? I think they do. They share the basic ideas, that

- humans have the capability to model other people in great detail

- a tulpa is such a model

 

Isn't this the standard idea, now? The tulpa standard model? Sure, it's not too fleshed out, at least on average, but it's fairly compelling. I'm not sure it's still true that 'we have no idea what a tulpa is', anyway.

[Yakumo]

Interesting material, thanks!

It all sounds pretty reasonable, I agree that being able to mimick and predict other persons behavior appears to be a key funktion of our mind. I also like the analogy to religious practices and to people claiming to communicate with 'God'. Whitehouse’s mode of religiosity theory is also worth reading. Offers some interesting perspectives as well as ideas for imagistic bonding with your tulpa (page 11 ff).

Rarely performed and highly arousing rituals invariably trigger vivid and enduring episodic memories among the people who participate in them. It appears to be a combination of episodic distinctiveness, emotionality, and consequentiality that together result in lasting autobiographical memories.

These memories can be so vivid and detailed that they can take the form of (what some psychologists call) flashbulb memories.

It is almost as if a camera has gone off in one’s head, illuminating the scene, and preserving it forever in memory.

[...]

Examples might include traumatic and violent initiation rituals, ecstatic practices of various cults, experiences of collective possession and altered states of consciousness, and extreme rituals involving homicide or cannibalism...

Well, you don't have to take it that far but you get the idea. I highly recommend reading tha paper.

 

The problem however is that all these hypotheses and models are somewhat arbitrary as

a) I can't think of a suitable way to falsify them

b) the models do not make predictions that may be easily tested for validity

 

Now I know this is a soft science but I wish people would come up with something more verifyable. But that's just me seeing things from a natural science perspective. I guess it's better than nothing. So I'd say we do have some decent hypotheses on what tulpas are and how they are created but the question if everything we call tulpa really works the same way remains. The phenomenon is so diverse that I doubt we will ever find a single simple answer to it.

 

Tulpa: "I spent my lunchbreak at work banging my co-worker and now I'm hungry."

Seriously, that's what he chose as a typical tulpa comment?

 

Anyway, good to see there is research conducted on this subject.

[Stevie]

Cool stuff, thanks for compiling it.

 

I'd say that the whole concept of tulpas being models that are just reinforced over time makes sense, if you look at the other side of the coin, with established tulpas falling asleep/being less reactive when they haven't been interacted with much lately, that seems to support the theory in a backwards kinda way.

 

Sometimes, people would report, if they've not been doing much with their tulpae, that their tulpa would start acting 'lazy,' just lying around all the time, not doing too much.

But then if they'd start proxying for them (even if they have to /force/ them to start talking), they'd start being more active again.

I know a few tulpae who are "sleeping all the time" when not doing much else.

I think these are the first signs of said decay.

 

Something that I've noticed too is that since starting this, the hypothetical conversations I have and whatever random thinking I do, especially in relation to other people, seems a lot more instant and autonomous. I know personal anecdotes don't count for squat, but it makes sense for there to be some trickle down effect.

 

Question is though, how does changing how we view tulpas change potential forcing methods? What people are doing now works, of course, but there has to be some way to find out what's strictly necessary for forcing, and what the extra stuff really is. Refining the system based on the new definition could maybe be a way to test its validity, even if it's not exactly foolproof.

[Nyx]

Thanks for sharing these. I studied anthropology in college, so Veissière's article will be a treat to read.

[waffles]

Whitehouse’s mode of religiosity theory

 

It's kind of interesting. I guess theories like that have more to do with concept propagation than blunt mechanisms. You have new members carrying out rituals (creation) that lead to religious experiences (having a tulpa) which assures the concept being passed on. I guess it's slightly unfair to stop there, because, well, that's essentially true of any effective (mental) practice; I guess it says something for the generality of the theory, at least. Maybe it's a bit more relevant if you consider the body of beliefs and assumptions that surrounds the practice.

 

 

The problem however is that all these hypotheses and models are somewhat arbitrary as

a) I can't think of a suitable way to falsify them

b) the models do not make predictions that may be easily tested for validity

I think one of the major things to take out of the model is that making a tulpa - or getting to an early stage of vocality - is not hard at all. Because everyone already has the mental faculties needed to predict other people, and such, getting responses out of it directly should be, more or less, a matter of listening. And I think it's fair to say that that's the way creation tends to go, specifically with early responses. A model that relies on you building up some kind of neural framework from scratch can't explain day one responses as the same thing as a tulpa, because that facility just shouldn't be there.

 

It might be a little tasteless to continue theorycrafting in response to this complaint, but one of the reasons that it doesn't seem to make more useful predictions is because it isn't specific enough. For instance, if I were to refine it as follows:

 

- the faculty for simulating/predicting how other people behave is pre-existing in humans

- there is essentially nothing except a lack of intent to listen on the part of the host (or possibly a lack of context - how often do/did you have a simulated conversation with someone in your head, and think nothing of it; I wish I could remember my introspections from before I made a tulpa, because I suspect I'd find that I did this then, as I do now. Stevie, can you compare/contrast?) that prevents direct interaction between the host and this faculty

- tulpa creation - or, the part concerned with getting responses - is primarily a matter of breaking this down, and learning to listen for responses

- subsequent interactions concretify the tulpa into a more stable personality

 

Now, I could predict the following:

- seeded with expectation, getting responses could take any time at all above zero seconds

- but getting a stable personality necessarily takes longer

- other things that must take time: separate memory, etc.

- the same faculty could potentially be used (from t+0) for possession, and such

 

Okay, I'm cheating. Those are retrodictions. Here are a few that are a bit more out there:

- given the correct expectations, communication between host and tulpa could instantly break down (at least, in early stages)

- there is no practical difference between narration/parroting/other methods, but personality forcing could be distinct

 

And so on.

 

 

Question is though, how does changing how we view tulpas change potential forcing methods? What people are doing now works, of course, but there has to be some way to find out what's strictly necessary for forcing, and what the extra stuff really is. Refining the system based on the new definition could maybe be a way to test its validity, even if it's not exactly foolproof.

 

See above, I guess. But it's hard to control creation (in others), at least in the context of an internet site, because those people are free to talk to others and acquire the beliefs and expectations of the community. Since, I think, these models tend to increase the space of effective methods to essentially everything (methods are irrelevant, beliefs are what counts), it wouldn't give any particularly positive evidence - one weird trick makes tulpa creation 900% easier, guide writers hate him! But if you were to pick another, more rigid theory, pick a method that shouldn't work very well, and land yourself some one-day tulpa fishes, I guess that's something.

 

Having said that, I guess any model that focuses away from expectations would end up stillborn. So I don't know; I guess I don't really have much of an alternative model in my mind, apart from a more basic habitual model. Maybe it says something about the model that a lot of different people say things like it.

[Ido]

Linking certain disorders to the inability or difficulty to create tulpas may shed light on the underlying neuronal processes.

The empathetic model could be tested on people with autism spectrum disorder who have severe difficulties in predicting human behavior. According to the hypothesis such individuals should have greater problems or take more time to create a tulpa than an unaffected control group.

 

However, if we look at the complexity of the tulpa phenomenon, I doubt this would lead to conclusive results. I think we have quite some members who have been diagnosed with Asperger syndrome but don't seem to have greater troubles with their tulpas than others. But than again, autism spectrum disorders are just as diverse as tulpa experiences.

 

Anyway, I think we should put more focus on the reports of people who have failed in creating a tulpa and look for possible causes. Such people should be encouraged to share their experience instead of vanishing from the forums as I'd guess it is mostly the case.

[Stevie]

- there is essentially nothing except a lack of intent to listen on the part of the host (or possibly a lack of context - how often do/did you have a simulated conversation with someone in your head, and think nothing of it; I wish I could remember my introspections from before I made a tulpa, because I suspect I'd find that I did this then, as I do now. Stevie, can you compare/contrast?) that prevents direct interaction between the host and this faculty

 

I wouldn't say that I have more simulated conversations, at least that's kind of my impression. More or less, the ones I do have seem to rely a lot less on me in that they kind of start more out of the blue, and they have kind of become a weird little way for me to process information. I also find that I think I do it for different reasons now too, and instead of being a group of thoughts that kind of pick apart the conversation I'm gonna have (i.e. "Maybe I should say it this way so she doesn't get ripshit mad") they're a lot more full bodied, if that makes sense.

 

@Ido Good point about the autism spectrum mention, can anyone with asperger's weigh in?