Research Note 002:The Distance Is Not the Weight

What Cognitive Transitions Might Actually Cost Us

In the last research update, I argued that before Cognitive Mode Theory can meaningfully measure transitions between cognitive modes, the modes themselves need defensible boundaries.

In simpler terms:

Before we measure the distance between rooms, we have to draw the walls.

That gives me a rough order of operations for where the research needs to go next:

Mode boundaries → Transition identification → Transition distance → Transition weight

I'm being stubborn about that order because distance and weight aren't the same thing, and the further I get into this work, the more I think collapsing them would cause us to miss some of the most interesting questions CMT might eventually be able to answer.

So let's go back and walk through this together.

First: What Is a Cognitive Transition?

CMT proposes that complex work can be decomposed into distinguishable modes of cognitive activity. The current taxonomy includes nine proposed modes:

Observation · Research · Synthesis · Decision · Planning · Creation · Refinement · Execution · Reflection

A transition occurs when the cognitive demand of the work changes enough that completing the task requires the person to move from one mode into another.

Using the newsletter that haunts my dreams as an example, imagine I'm researching papers for an article. I'm locating information, examining sources, and trying to understand what's already known. Then I stop gathering information and begin determining how several of those findings relate to one another.

CMT might classify that as:

Research → Synthesis

Later, I decide what argument the article will actually make:

Synthesis → Decision

Then I begin writing:

Decision → Creation

From the outside, I've been doing one thing this entire time: writing an article. Cognitively, however, that single task contained several substantially different kinds of work, which means the arrows between those modes are becoming just as interesting to me as the modes themselves.

Task Switching Doesn't Tell Me Enough

We already have extensive ways of talking about interruptions, context switching, attention, executive function, and the cost of moving between tasks. CMT is asking something slightly different because I'm not convinced the task itself is always the most useful unit of analysis.

What happens when I change tasks without meaningfully changing cognitive mode? Conversely, what happens when I remain inside the same task while repeatedly changing modes?

Opening another research paper may technically constitute a task switch while leaving me firmly in Research. Meanwhile, moving from gathering evidence to determining what that evidence means can happen inside the same document, on the same computer, within seconds, while still requiring substantial cognitive reorientation.

So the question isn't simply whether the task changed. I want to know whether the cognitive demand changed, and if it did, how much reorientation did that change require?

Which brings me to transition distance.

Transition Distance

I'm currently using transition distance to describe the degree of cognitive reorientation required to move between two modes.

I'm not assigning numerical distances to the nine modes yet because doing so would imply a level of precision the research hasn't earned. Right now, distance itself is still a research question.

Does Research → Synthesis consistently require less reorientation than Research → Creation? Is Creation → Refinement relatively proximal because the person remains engaged with an existing artifact? Is Observation → Decision more distant because the cognitive demand shifts from receiving information to committing to an outcome?

Maybe, but “that sounds right” isn't enough.

Every arrow I draw is something I eventually need to defend, which is part of why Recursive Cognitive Decomposition exists in the first place. I need to observe these transitions repeatedly, determine whether the mode classifications themselves hold up, document what appears to change between them, and resist the urge to turn an intuitively satisfying relationship into a structural claim before I have enough evidence to support it.

That process is slow. Painfully slow sometimes.

It's also beginning to make each cognitive mode feel like its own chapter of the research. If I want to understand the transition between Research and Synthesis, I need to understand Research well enough to know where it ends and Synthesis well enough to know where it begins. If those boundaries are unstable, any claim I make about the distance between them inherits that instability.

Which brings us to the question I'm considerably more excited about.

Then Comes Weight

Suppose we eventually gather enough evidence to characterize the relative distance between modes. We still haven't answered the question most people are probably more interested in:

How hard was the transition?

That's what I'm calling transition weight.

Distance describes something about the structure of the transition itself, while weight describes the experienced cost of making it. Those things may interact, and I suspect they probably do, but they shouldn't automatically be treated as interchangeable.

Imagine two people performing the same sequence:

Research → Decision → Creation

If the cognitive demands associated with those modes prove reasonably stable, the structural relationship between them shouldn't need to be completely redrawn for every individual. Research → Decision doesn't become an entirely different transition because a different person is doing it.

The map can remain relatively stable while the cost of traveling across it changes substantially.

And that's where the person enters the model.

What Are You Carrying?

This is the question I keep returning to because two people can make what appears to be the same cognitive transition and walk away from it with completely different levels of fatigue, friction, or recovery time.

Eventually, CMT needs to investigate what contributes to that difference. Working memory, attentional regulation, executive function, cognitive flexibility, sensory load, emotional state, sleep, stress, environmental interruption, and prior cognitive load are all potential variables worth examining. Neurodevelopmental, psychiatric, or other conditions associated with differences in those systems may be relevant as well.

But I want to be very careful about the order in which I ask those questions.

CMT currently does not establish that a particular diagnosis produces a particular transition cost, and I don't want to reverse-engineer a scientific explanation for an experience simply because it feels intuitively true.

For example, asking, “How much heavier is Research → Creation for someone with ADHD?” would require me to assume several things I haven't established yet: that I've adequately defined Research and Creation, that I can characterize the distance between them, that I can reliably measure the experienced cost of that transition, and that ADHD itself is the appropriate explanatory variable for whatever difference I observe.

We aren't there.

A better question is: Once a transition has been sufficiently characterized, which individual and contextual variables predict meaningful differences in its experienced cost?

Then we can investigate whether particular conditions explain some of that variation rather than beginning with the diagnosis and working backward.

Comorbidity Makes This Even More Complicated

Comorbidity is one of the reasons I think that distinction is relevant. People don't experience cognition as neatly separated diagnostic variables, and the same person can be navigating differences in attentional regulation alongside anxiety, executive-function demands alongside depression, or several of these at once.

Even within the same person, those variables aren't necessarily static. Someone may be well rested one day and sleep deprived the next or emotionally regulated during one transition and already carrying massive amounts of stress during another.

If CMT eventually tries to explain transition weight using diagnosis alone, it risks recreating the same kind of oversimplification I'm trying to avoid everywhere else. A diagnostic label may be relevant, but the mechanisms and state variables underneath that label may ultimately tell us considerably more about why a transition became expensive in a particular moment.

So instead of only asking, “What conditions does this person have?”, I think CMT will eventually need to ask something closer to:

“What variables are active right now that could alter the cost of this transition?”

That gives us a much more interesting way to think about personalization.

The Map Can Stay the Same While the Terrain Changes

I'm starting to think about CMT as containing two related layers.

The first is the cognitive architecture of the work: what modes are required, in what sequence they occur, where transitions happen, and eventually how cognitively distant those modes appear to be from one another.

The second is the state of the person moving through that architecture: what cognitive resources are available, what constraints are active, what has already happened during the work episode, what the environment is demanding, and what could make a particular transition heavier right now.

The first gives us something resembling a map. The second tells us about the conditions under which someone has to travel through it.

That framework gives me a possible explanation for something ordinary workload descriptions don't always capture very well: two days can contain nearly identical task lists and feel nothing alike. Two people can perform the same job and experience dramatically different levels of cognitive friction, and something that felt relatively easy yesterday can feel almost impossible today even when the task itself hasn't meaningfully changed.

If that's true, the work isn't the only variable changing. The conditions under which cognition has to move through that work are changing too.

This Is Also Why RCD Is Taking So Long

I occasionally have to remind myself that Recursive Cognitive Decomposition is deliberately asking me to slow down at exactly the point where I want to accelerate.

RCD asks me to observe, assess, and document individual cognitive episodes rather than looking at an entire work process and deciding afterward that a sequence “looks like” Research → Synthesis → Decision → Creation. Every classification needs justification, ambiguous transitions need to remain ambiguous rather than being forced into the taxonomy, and modes that consistently blur together may require better definitions or may eventually challenge whether they should be separate modes at all.

That's why each mode is increasingly becoming its own chapter. I can't study the relationship between two modes without doing enough work on each side of that relationship to know whether I'm actually observing what I think I'm observing.

And every time the research produces another interesting question (especially one about individual differences or transition weight) I have to resist immediately adding another layer to the model.

The constraint I've given myself is:

You cannot responsibly model transition weight until you have sufficiently characterized transition distance, and you cannot characterize transition distance until the modes on either side have defensible boundaries.

As much as I want to get to weight, I'm still drawing the map. And in some places, I'm still figuring out whether I've drawn the walls in the right place.

Eventually, I Want CMT to Ask a Different Question About Exhaustion

We tend to quantify work by volume: how many tasks someone completed, how many meetings they attended, how many hours they worked, or how many things they managed to finish. Those measurements can tell us something useful about workload, but I don't think they can tell us the whole story of cognitive demand.

Eventually, I want to be able to look at a work episode and ask what cognitive modes it required, how frequently the person transitioned between them, how distant those transitions were, whether certain sequences repeatedly produced friction, whether transition costs accumulated over time, which individual or environmental variables altered those costs, and whether the architecture of the work itself created unnecessary cognitive movement.

Because when someone says, “I barely did anything today. Why am I so tired?”, counting the number of things they accomplished may be answering the wrong question.

I want to see the route they traveled.

Someone may not have completed an extraordinary amount of work at all. They may have spent the day repeatedly reorganizing cognition to meet different demands, carrying the residue of one cognitive state into the next while also navigating whatever attentional, emotional, environmental, or physiological variables were already present.

Which brings me back to the distinction I'm trying to preserve now, long before I'm ready to quantify either side of it:

How far did you travel, and what were you carrying when you made the trip?

Those are two different questions.

CMT needs to learn how to answer the first before I can responsibly claim anything about the second.

Cognitive Mode Theory is an evolving theoretical framework under active development and validation. Concepts including mode taxonomy, transition distance, and transition weight are currently research constructs under investigation rather than established cognitive measures.