You cannot catch yourself in flow

Saturday afternoon, something ordinary. Peeling, sanding, playing the same eight bars again badly. You look up and it is quarter past four.

You did not decide to keep going. At no point did you think, this is going well.

Now put that beside the last time you set out deliberately to enjoy something. The walk booked in as a treat, the film you had been saving. You checked in on yourself the whole way through it. Am I enjoying this. Is this working. Is this the good bit.

One of those two was flow. It was not the one you were paying attention to.

The checklists want you to catch it in the act

Search for what a flow state feels like and you get a list to hold up against your own head. Total absorption. Time distorting. A sense of effortless control. The pages describe the thing accurately enough, and then they attach an instruction that quietly ruins it: notice when this is happening to you, and arrange for more of it.

You cannot run that check. Not because you are bad at noticing things, and not because you have not practised. Because of what flow is built out of.

The part that switches off is the part that would notice

Flow's characteristics sort into two piles, and the difference between the preconditions and the symptoms is the useful distinction in the whole field. Three of them you can arrange in advance. The rest are what it looks like when the first three line up.

Reduced self-consciousness sits in the second pile. So does the transformation of time. Gold and Ciorciari, reviewing the neuroscience of flow in Behavioral Sciences in 2020, describe it as "the removal of self-consciousness in which a person loses the awareness of themselves as well as thoughts of everyday problems".

Read that as a job description rather than as poetry. The thing being removed is the narrator. The one that keeps a running commentary on how you are doing, whether this counts, whether you should be doing something else.

That narrator is the only apparatus you have for checking whether you are in flow. It is also the one thing that is definitionally not running while you are.

Researchers hit this wall first

This is not a philosophical point. It is a practical problem that people who measure flow for a living have to design around, and the way they design around it is the useful part.

Gold and Ciorciari put it plainly. Of the questionnaires used to capture the state, they write that "the introspection necessary for these measuring techniques has the danger of inhibiting the flow experience as it requires resources to be allocated to a different cognitive set as these are retrospective by nature".

Asking somebody in flow whether they are in flow takes them out of it.

So the Flow State Scale, the questionnaire that came after real time experience sampling, "is typically given at the end of a task in order not to force the participant out of the state during the task". Questionnaires are the primary method of studying flow, and they run afterwards because running them during does not work.

Flow is not something you can notice. It is only ever something you can have noticed.

Two honest limits before this gets stretched. This is a review paper describing a methodological constraint on questionnaires, not an experiment on people trying to spot their own good afternoons. And the constraint is about a formal instrument rather than a passing thought. The reason it still transfers is that both run on the same equipment, and the equipment is the thing flow suspends.

So what does it actually feel like

While it is happening, mostly nothing. That is the honest answer and it is why so many people conclude they have never had it.

There is no glow to detect while it is running. The state needs a challenge matched to your skill, neither too easy nor too difficult, which means working at the edge of what you can do rather than comfortably inside it. From the inside, that is often hard to tell apart from ordinary effort. Csikszentmihalyi's list does call the activity intrinsically rewarding, and the honest version of that is a reward you read afterwards rather than one you feel yourself collecting at the time.

Afterwards there are two traces, and both are structural rather than emotional.

The first is an amount of time you cannot account for. Not time that passed pleasantly. Time that is simply missing from the ledger.

The second is the absence of a decision. You did not talk yourself into the next twenty minutes. There was no negotiation, because the part of you that negotiates was not at its desk.

One caution about reading it afterwards

The retrospective view is not clean either, and it is worth knowing how it fails before you rely on it.

The same review notes that end-of-task reports "measure the experience across the whole task rather than for a particular time period", and adds that this "may further be influenced by the recency effect in memory". The last ten minutes tint the memory of the whole hour.

Which means a good session that ended in a mess gets filed as a bad session. This is how people quietly retire the thing that was working for them, on the evidence of one bad ending. It is the same mechanism that decides, on one joyless Tuesday, that the thing you actually enjoy was never enjoyable.

Three things to do with this

Ask once, at the end, and then stop asking. One question after you put the thing down, not a running audit. The audit is the interference.

Look at the trace rather than the feeling. Where did the time go, and did you have to decide to continue. Those two questions survive a bad mood in a way that "did I enjoy that" does not.

Give an activity more than one reading before you drop it. Given the recency effect, one bad ending is not evidence about the activity. Three are.

This is the part that is harder than it sounds, and it is the reason Flowdayz asks whether something felt good after you have done it rather than before. A feel good app cannot tell you what gets you into flow, because you are the only one with the data, and even you can only read it afterwards. What it can do is keep asking at the moment the answer still exists, and remember better than you do.

The thing that turns out to be yours is almost always smaller and duller than you expected. Ten minutes of peeling. The same eight bars. Nobody arrives at that by introspecting harder in the moment.

They arrive at it by looking up at quarter past four, twice, and finally believing the clock.