Internal communication framework
Usually when we think about humans and agency, we focus on the layer where one human is one agent.
We can also shift focus onto other layers and treat those entities as agents instead—for example, corporations or nations.
Or, we can take different parts within a human mind as agenty—which is a lens we explore in this post.
In this post, we first introduce a general, pretty minimalist framework for thinking about the relationship between different parts of the mind (Internal Communication Framework, or ICF). Based on this framework, we build a basic technique that can be used to leverage ICF, and which we believe can be helpful for things like:
Resolving inner conflicts
Building internal alignment
Increasing introspective access
We have shared ICF with a small number of people (~60-80) over a few years, and are writing this post partly to create a written reference for ICF, and partly to make the framework accessible to more people.
The framework
Some background context:
ICF is rooted in thinking about hierarchical agency (i.e. how agency, and strategic dynamics, play out composite agents and their parts).
It is mostly an attempt to generalise from a class of therapy schools and introspection techniques working with parts of the mind, and to make the framework cleaner from a theoretical perspective. (More on the relationship between ICF and some particular techniques from that class here
ICF was developed in 2018 by Jan Kulveit with help from Nora Amman, in the context of the Epistea Summer Experiment and then CFAR instructorship training.
ICF makes two core assumptions:
The human mind is made of parts
Interactions between parts, and between parts and the whole, work best when the interactions are cooperative and kind
There is a possibly interesting discussion about the epistemic status of these assumptions, which is not the point of this post, so we invite you to interpret them in whatever spirit you like—as a model which is somewhat true, a model which is wrong but useful, a metaphor for processing your experience…
The human mind is made of parts
These parts have different goals and models of the world, and don’t always agree. They are still all part of your mind.
If we make this assumption that the mind is made of parts, it becomes less obvious what we mean by terms like ‘you’ or ‘the self’. One metaphor for thinking about this is to view the whole set of parts as something like a ‘council’, which collectively has control over something like a ‘speaker’, which is the voice creating your internal stream of words, the appearance of a person, your sense of ‘self’...
Often, the whole is the most intuitive and helpful level of abstraction, and ‘you’ will make perfect sense. Sometimes, for example when experiencing inner conflict, ‘you’ will be confusing, and it will be more productive to work at the level of the parts instead.
Interactions between parts, and between parts and the whole, work best when the interactions are cooperative and kind
By cooperation, we mean something like choosing ‘cooperate’ in various games, in a game theoretic sense.
While game theory has formalised the notion of cooperation between agents at the same level of abstraction, we don’t have a similar formal model for positive interactions between a whole and its parts. We will refer to these interactions as kindness.
You may disagree with this assumption in the form of a normative claim. Descriptively, we think that cooperative and kind relations between parts, and between parts and the whole, tend to lead to more constructive interactions, and longer-term make people happier, more aligned and more agentic.
A few overall notes:
Different layers of agency can compete for power—e.g. for space to speak and being heard. The point of ICF is not to make the parts more powerful and the whole less powerful. The point is to make the vertical interaction more kind, and the horizontal interactions more cooperative.
Theoretically, one could use techniques which build on ICF as tools to coerce some parts, or help parts to achieve their goals at the expense of others. We think that doing this is highly prone to bad consequences for you in aggregate, and strongly discourage practising ICF techniques if this is what you want to do with them.
Where do these assumptions leave us?
ICF is pretty general. It’s a way of thinking about the relationship between parts of the mind. There’s no need to ‘do’ anything in particular with ICF: it might just be a model that you sometimes remember, or something which informs the way you think about certain things.
That said, it is also possible to build techniques based on ICF which you can then use in a more intentional way, to resolve inner conflicts or have certain kinds of experience. In the next section, we describe one such technique in detail.
An ICF technique
This is a basic, verbal ICF technique which we’ve taught to small numbers of people in the past.
In a nutshell, this technique involves facilitating conversation between different parts of yourself (typically with someone else holding space for you).
The set up
This ICF technique tends to work best with a facilitator, who holds space for the person ‘doing’ ICF, and helps them to explore whatever they are exploring. Often doing this reciprocally works well, where one person holds space for the other to do ICF, and then they swap around.
Choose a person, a place and a time where you feel comfortable and won’t be interrupted. It’s often good to be explicit about how long you want to spend doing ICF (or that you want it to be an open-ended container).
Summary: the steps
Step 0: land in yourself
Step 1: find a part, and unblend from it
Step 2: check how you feel about that part, then repeat step 1 until you feel only curiosity and kindness
Step 3: facilitate conversation
Step 4: close the space
Step 0: land in yourself
This will look different for different people and in different contexts. You could:
Do certain forms of meditation, focus, notice what’s happening in your body, observe your breath, circle yourself…
Explicitly identify an inner conflict or life situation you want to work on
Just notice out loud what’s been going on in your day/week/month until you notice things that you want to explore
The important thing is to arrive at something that feels alive in you, that you’re in touch with in the present moment and want to explore and be with. It doesn’t really matter how you get there, or how explicit/legible the thing is.
Step 1: find a part, and unblend from it
Examples of what finding a part could look like:
“There’s a part of me who really wants to take on the project.”
“I feel guilt.”
“I have some tension in the back of my neck.”
Unblending means creating some separation between a part, and the whole/the rest of the ‘council’. The idea here is that usually all parts are blended into a whole which generates the speaking ‘I’. To converse between parts, you need them to unblend and take on separate existence. Then the different parts can each take turns at having a voice. Another way of putting this: “you can’t smell the soup if your face is dunked into it.”[1]
Unblending is a mental motion, and it’s easier to understand by doing it than by reading about it. Some things that may be helpful for unblending:
Giving the part a name, or an image, or a personification (or having the part come up with a name for itself)
Explicitly asking the part if it’s happy to unblend
Visualising the part as localised in a particular part of your body, or as outside of your body, or in a particular place in a landscape
Expanding compassion and curiosity towards the part (in a way that’s similar to what’s being practised in metta meditation)
Sometimes it’s very easy to unblend. Other times, it’s not. This might be because:
The part feels scared or low trust. (e.g. ‘But if I unblend, the rest of you will try and get rid of me, and I’m tracking really important stuff you need to pay attention to.’)
The part feels so deeply part of who “you” are that you can’t really imagine existing separately from it. (e.g. ‘But caring about x is just me; I can’t locate as anything smaller than all of me.’)
Other reasons.
There are a few different approaches you can try if unblending is difficult:
Unblending from a different part first. Often a part doesn’t want to unblend because it’s at war with a part that’s still identified with you, and it correctly feels vulnerable to attack if it unblends. For example, if I find anxiety, and then the anxiety doesn’t want to unblend because it knows that I have strong judgement about feeling anxious, I could try unblending from the judgement first, and then come back to the anxiety and see if it feels ok about unblending now.
Trying to reassure the part about the fairness of the process. This might look like that you’ll pay attention to certain things, or that you won’t use coercion against the part.
Approaching the part as you would approach a child, or a wild kitten, or a good friend. Maybe it needs something, maybe it’s frightened, maybe slow is good.
Just sitting with the part inside of you and seeing what happens. Sometimes unblending isn’t accessible. You can still gain understanding from just sitting with the part that won’t unblend, seeing if it has anything to say, seeing how it feels in you.
Step 2: check how you feel, then repeat step 1 until you feel only curiosity and kindness
Once you’ve unblended from a part, check how you feel. Continuing with the examples from step 1, you might feel:
“But there’s also a part of me that thinks the project is too hard.”
“Now I’ve unblended from the guilt, I notice that I also have some fear.”
“I can still feel the tension in the back of my neck, but I feel unblended from it. I can feel a lot of spaciousness and potential in my forehead.”
So now you go back to step 1, and unblend from the new part you’ve found.
The ideal is to keep repeating step 1, until when you check how you feel, you only feel curiosity and kindness. In this ideal state, you’ve found and unblended from all the parts that have a vested interest, and the ‘you’ that is left no longer has any axe to grind. You just feel kindness and curiosity towards your parts, and that is a good place from which to facilitate conversation.
Step 3: facilitate conversation
The heart of ICF is facilitating conversation between parts. This can take lots of different forms:
Sometimes parts will want to talk directly with each other
Sometimes the kindness-and-curiosity-feeling self will mediate between parts
Sometimes one part might just monologue, or dialogue with the self, for most of the time
Sometimes there won’t be much in the way of verbal communication, and there will just be an experience of being with the parts and interacting non-verbally
Some people narrate most of what’s going on for them out loud. Others might spend a lot of the time in silence.
What’s important is ‘you’ (or in other words, the ‘whole’ or ‘the rest of the council’) ensures the kindness and fairness of the overall process. For example, similarly to facilitating a conversation between friends, if you notice one part is trying to manipulate or blackmail another, you should intervene, and possibly switch to some other communication form.
Step 4: close the space
Once you’re nearing the end of your time container, or once you feel complete, it’s time to close the space.
Things you might want to do at this point:
Check whether your parts have anything they want to say or ask for before you close
Thank your parts
Make commitments to your parts (like coming back to the conversation at some future point, not being violent to the parts, honouring the things they care about…)
Invite your parts back in
Important exceptions to these idealised steps
In the idealised implementation of this ICF technique, you do step 1 and 2 until you’ve found all the parts and just feel kindness and curiosity, and then you do step 3, which is sort of the heart of the practice.
This is often not actually how things go:
Often you won’t get to step 3 at all. You’ll keep doing bits of step 1 and step 2, finding more parts, losing parts, getting confused, and still not feeling kindness and curiosity.
Often the steps will be much more entangled. You might get to step 3, then experience some kind of shift where a new part comes along or a bunch of the other parts just dissolve in your experience. Then at some point you might notice you’re not just feeling kindness and curiosity, and need to go back to step 1.
Neither of these scenarios is a failure, or the technique done badly: they are part of the process. The goal of the technique is something like ‘increase your understanding of yourself’, and this is compatible with not following the steps, not resolving the conflict, feeling discomfort and confusion and frustration… The idealised steps are there to help people to navigate their internal experience. With practice, people often become better at ‘winging it’, and tracking what’s going on in the moment without needing the explicit steps to guide them.
Tips for practising this technique
If you get stuck on step 1: try focusing
If you get stuck on steps 2 or 3: often there’s a secret unblended part which has a lot of power over the council, fearing it may loose this power
Try doing it with a facilitator; this might feel weird at first, but in our experience it often helps people to keep up a consistent focus and don’t get stuck for too long
In general: go slowly, don’t push your parts
Tips for facilitating this technique
The main purpose of facilitation in this technique is to uphold the kindness and fairness of the overall process.
This means that it’s important for the facilitator to get into the mode of kindness and curiosity too, and to periodically check that they’re still feeling this.
In particular, a facilitator is not there to help particular parts to win in internal struggles with other parts. It can be tempting to help particular parts, especially because sometimes your parts will have feelings towards parts in the other person. When this is going on, it’s good to notice that it’s happening. We suggest noticing, then asking the relevant part in you the facilitator to unblend, such that you can return to the person doing ICF with kindness and curiosity.
Here are a few slightly more concrete suggestions for things you might want to do as a facilitator:
Reflect back what the person is saying
Check in periodically that the person is still feeling only kindness and curiosity; if they are feeling other things, suggest that they could try unblending
Check that the person is feeling present, in the moment, grounded, that the things they are with feel alive
Check that the parts feel safe and listened to
Share the way you feel about what’s going on
The broader ICF space
There are several general axes along which ICF techniques can vary, depending on the person and the circumstance:
Structuredness
More: you follow formal idealised steps
Less: you just sit and stuff happens, with no reference to any steps
Goal-directedness
More: you intend to make a decision or resolve a conflict, and you do
Less: you just want to spend some time with yourself
Legibility
More: you make your experience explicit to yourself (and the facilitator if there is one), and have a clear story when you close about what happened
Less: you’re mostly in direct physical experience, or metaphor, or visualisation; you don’t make any of this very explicit to yourself (or a facilitator); you don’t have much of a story about what happened when you close
Stability of parts
More: you orient towards your parts as more or less fixed and permanent. You’re talking to them in this moment, and you could find the same part at any future date and do more talking[2]
Less: you orient towards your parts as shifting configurations in the sand. Maybe they shift while you’re talking to them in this very moment; there’s no assumption that you’ll ever have a part just like this one again
There are many, many different possible ICF techniques (most of which are presumably undiscovered and could be found via experimentation). To give some examples:
Different media: drawing, writing, speaking, movement, staying silent…
Giving roles or archetypes to parts: animals, avatars, body parts, real people, characters from books, colours, shapes…
Visualising different settings for the session: garden, mountain, home, dinner table, mystical place…
Different forms of cooperation between your parts: fellowship of the ring, parliament, a roundtable…
Different layers of aggregation between parts, or the formation and dissolution of (temporary) coalitions: a group of parts might merge into a coherent ‘super-agent’ in a given context, or what initially felt like a single agent might factor into several sub-agents with different perspectives, or…
Appendix: ICF and other techniques
As noted above, ICF is an attempt to generalise from a class of therapy schools and introspection techniques working with parts of mind.
In this community, the best known examples from this class are Internal Family Systems (IFS) and Internal Double Crux (IDC). There are many more.
Relationship to IFS
IFS does something close to “family therapy” on parts, and written descriptions take relatively strong priors on what the types of parts there are (firefighters, exiles, managers).
From an ICF perspective:
Therapy is one type of conversation it might be useful for parts to have, but not the only kind.
The firefighter-exile-manager pattern is a good fit for some people in some situations, but not for all people in all situations.
ICF borrows strongly from IFS. In particular:
The process of unblending in the technique above is borrowed from IFS.
IFS focuses on making the process non-judgemental and kind.
Relationship to IDC
From an ICF perspective, basic IDC prompts the parts to form two coalitions, and to have a structured dialogue between coalitions in roughly the form of “double crux”. Sometimes, the composition of the coalitions change, and the parts are prompted to find new labels. Sometimes, the parts update their models, and the conflict is actually resolved. This often works well if there are actually just two parts, which differ mostly in data, but not directly in goals.
In our experience, experienced IDC practitioners seem to often do something slightly broader—weakening the priors (e.g. “there are two parts”) and not forcing the conversation into the “double crux” form, if it doesn’t seem to fit. In this case, IDC is basically a particular application of ICF.
Relationship to other techniques
There are many other approaches which work with parts, like those given in this list (e.g. Schema Therapy, Compassion-focused therapy…) and many others (e.g. Voice Dialogue Therapy…). These approaches often take relatively strong priors on what sorts of parts there are, or what the origin of the parts is, based on various different sources ranging from evolutionary biology to Buddhism.
While we don’t have direct experience with most of these approaches, our guess is that the core of what is helpful about them is “facilitated kind interactions with parts, including the dark ones”, and that the detailed priors on what the parts are or where they come from are not essential.
Instrumentally, these stories about the parts and their origins probably provide useful ‘hooks’ for some people to engage with the various approaches—but there’s an epistemic worry that people might reify the stories, which in many cases seem somewhat fake.
If you wanted to, you could easily invent many other approaches in a similar spirit, like:
Internal circling
Internal dialogue with paraphrasing
Your parts as reinforcement learning models
Your parts as players in various game-theoretic interactions
Your parts as manifestations of ancient archetypes
…
ICF tries to move in the opposite direction: dropping assumptions and having broad priors.
ICF is compatible with most of the more specific stories about parts and their origins, but from an ICF perspective, these are just stories, and not what is going on fundamentally.
- ^
H/t Luke Raskopf.
- ^
We want to plant a gentle flag of caution here. While exploring any of these axes is fine, and often just a feature of personal preference, there is a related risk of taking the framework too seriously. For example, someone could practise an ICF technique and come away with an experience of having fixed, permanent parts beyond the container of practising the technique.This might be helpful and fine in lots of cases and we don’t have empirical evidence that it’s problematic, but we worry about reifying parts and believing stories too much in a way that is epistemically inaccurate or constraining for the future actions of the individual. We recommend paying attention to your direct experience and how true things feel for you, and holding stories you make in the context of ICF lightly.
ICF is the only such mental viz whizz technique that has ever worked for me, and I say that having done CFAR, a dedicated focussing retreat, a weekend vipassana retreat, and a dedicated circling retreat.
I wonder if there is a danger of someone whose personality is not super-cohesive practicing this “unblending” and then, like Humpty-Dumpty, never be able to put together again.
This feels to me more like a special case of IFS than a generalization of it, but since I am happy to see IFS spread and these look like good instructions for doing it, I’m also happy to see this posted. :)
Could you say a bit more about the way ICF is a special case of IFS? I think I disagree, but also think that it would be interesting to have this view spelled out.
Everything that I read in the post looked like pretty standard IFS, in fact I’m pretty sure that there are several IFS sessions that I’ve facilitated that followed these steps exactly.
When the post has this:
Then coming from IFS, my perspective is kind of the converse: “conversations other than therapy might be useful to have, but sometimes therapy is useful too”. In other words, this post reads to me the way you’d describe IFS if you described everything else but the explicitly therapeutic moves and only stayed on the level of facilitating a conversation between parts or between parts and Self. And sometimes IFS stays on that level too, if that’s enough for resolving whatever issue the client is having, or if there isn’t any particular goal other than just improved self-understanding.
So I see this as a special case in the sense that “IFS can be just the steps you’ve outlined, or IFS can be this + more explicitly therapeutic moves that aren’t well-described by just ‘facilitating conversation’ anymore” while ICF is described as always being just these steps.
The other difference to IFS that you mention is
And I agree, but I think that that pattern is more of a pedagogical simplification in IFS in any case. I would expect that any IFS practitioner with any significant amount of experience will unavoidably realize that there are vast differences in how different people’s internal systems are organized and that often this pattern will match only approximately at best. And it doesn’t matter that much anyway—off-hand, I don’t recall any IFS materials that would tell you to diagnose whether a part is a firefighter or a manager. Rather they just tell people to ask much more open-ended questions like “what is this part trying to do”, “what’s the part afraid would happen if it didn’t do what it’s doing”, or “how old is this part” that have you get to know each part as an individual rather than trying to force it into a category. (Jay Earley’s commonly-recommended book doesn’t even use those terms, and just groups both managers and firefighters under “protectors”.)
The protector-exile distinction is given more weight—because asking things like “am I getting access to the exile” can be a genuinely valuable move if the process feels stuck—but again I’d expect any IFS practitioner with any significant amount of experience to be very aware of the fact that often there are parts that are both at the same time, or that don’t clearly match either of those categories.
There might be a bit of a philosophical question of “what is the true IFS”. Is it the IFS that’s commonly described in the written materials and lectures, or the IFS that experienced practitioners are familiar with?
I do agree that if you define “what is IFS” through what’s been written about it, then there’s a stronger case for saying that ICF is the more general version. But if you think that any written materials in a field like therapy are always just a barebones-training wheel version of the real thing that’s learned through practice and expert supervision, and define IFS as “what do actual experienced IFS practitioners end up at”, then I think that ICF and IFS end up looking quite similar.
In my view, you are possibly conflating—
ICF as a framework
- described “basic ICF technique”
To me, ICF as a framework seem distinct from IFS in how it is built. As you say, introductory IFS materials take the stories about exiles and protectors as pretty real, and also often use parallels with family therapy. On the more theoretical side, my take on parts of your sequence on parts is you basically try to fit some theoretical models (e,g, RL, global workspace) to the “standard IFS prior” about types of parts.
ICF is build the opposite way:
1. assume layered agency (which is basically “put different layers of organization into the focus of intentional stance”)
2. ask: what sort of phenomenology would this lead to? how to interact with it?
In practice, I have a somewhat less positive take on the barebones-training wheel version of the real thing: I think it would be also fair to say that experienced IFS practitioners doing the real thing unlearn part of what’s in the books. And sometimes end up in similar place to e.g. experienced IDC practitioners who also end up not doing the protocol described in LW posts. From this perspective, it makes sense to have a label for the core of the approaches which works, distinct from IFS label.
I think there is a name for the core of the approaches which works, which is “parts work.”
The ICF framework seems to add some things on top of the basic parts work idea that make it similar to IFS. For instance, the process of unblending at the beginning is basically the same as what IFS calls “getting into self”. In contrast, there are many effective parts work frameworks that do the work from a blended state, such as voice dialogue. It imports the assumption from IFS that there is some “neutral self” that can be reached by continually unblending, and that this self can moderate between parts.
In addition, IFS and ICF both seem to emphasize “conversation” as a primary modality, whereas other parts work modalities (e.g. Somatic Experiencing) emphasize other modalities when working with parts, such as somatic, metaphorical, or primal. Again, there’s an assumption here about what parts are and how they should be worked with, around the primacy of particular ways of thinking and relating which is heavily (if unconsciously) influenced by the prevalance of IFS and its’ way of working.
It seems like while ICF is trying to describe a general framework, it is quite influenced by the assumptions of IFS/IDC and imports some of their quirks, even while getting rid of others.
What’s described as An ICF technique is just that, one technique among many.
ICF does not make the IFS assumption that there is some “neutral self”. It makes a prediction that when you unblend few parts from the whole, there is still a lot of power in “the whole”. It also makes the claim that in typical internal conflicts and tensions, there just a few parts which are really activated (and not, e.g., 20). Both seems experimentally verifiable (at least in phenomenological sense) - and true.
In my view there is a subtle difference between the “self” frame and “the whole”/”council” frame. The IFS way of talking about “self” seems to lead some IFS practitioners to assume that there is some “self” agent living basically at the same layer of agency as parts.
ICF also makes some normative claims, which make it different from “any type of parts work”: the normative claims are about kindness, cooperation and fairness. If you wish, I can easily describe/invent some partswork protocols which would be un-ICF and in my view risky / flawed from ICF perspective. For example, I’ve somewhat negative prior on techniques trying to do some sort of verbal dialogue in blended state, or techniques doing basically internal blackmail.
Not really—the post mentions basically all these directions of variance among ICF techniques:
Maybe the confusion is because just one technique was described explicitly in the post.
Ah yeah probably, I only know ICF from the description in this post. So when I said ICF, I basically meant “the technique described here”.
I see, that’s a definite difference then. I read the article’s claim that ICF tries to “generalise from a class of therapy schools and introspection techniques working with parts of the mind” as meaning that the model takes existing therapy schools as its starting point to derive its assumptions from them and their empirical observations, as opposed to deriving things from layered agency in a more first-principles manner.
I guess that makes sense—but then is ICF sufficiently general for that either? E.g. if we’re talking about IFS ideas that one might want to unlearn, I think that sometimes it’s useful to abandon the assumption of the parts being discrete units, and at least this article made it sound like ICF would still assume that. But maybe I’d need to know more about the general framework to know what assumptions it makes, in order to have this discussion.
What bothers me about this framing is that experienced practitioners for many different skills end up unlearning some of the advice that’s useful for beginners. It’s just the nature of much knowledge that you need to explain it in a simplified kind-of-false form first, before you have hope of conveying the complete form.
But we don’t say that we should have a separate label for the core of physics or chemistry that works; we just call it all “physics” and “chemistry”.
Or for a different domain where the understanding is built up more from the learner’s own experience and pattern-recognition ability than learning increasingly sophisticated scientific theories, here’s Josh Waitzkin on how chessmasters end up unlearning previous things they knew about the value of individual chess pieces:
I’ve personally found that intentionally feeding attention / strength to a part which can play the mediator / meta-part role extremely valuable, and have seen it spark changes in people with otherwise serious long standing treatment resistant internal tangles. Not everyone has a part with enough neural weight and trust of their system to navigate the most severe internal conflicts, and encouraging them to empower some part of themselves can be transformative.
Seems like a good high level structure for internal work.
Some editing notes:
This paragraph is repeated a few lines down.
Power over repeats.
Typo in the last word.
I’d suggest using the LessWrong team’s get feedback option to catch these before posting next time?
Thanks for spotting these; I’ve made the changes!