Using Collaborative Problem Solving to enhance Shutdown Ritual

What each one is actually for

Collaborative Problem Solving

Collaborative Problem Solving — sometimes abbreviated CPS, the approach developed by clinical psychologist Ross Greene — exists to make a repeating behaviour workable by finding the specific unsolved problem sitting underneath it. Its founding claim is that a child who is exploding at homework time is not choosing to explode; something about that moment is beyond what they can currently do, and the explosion is what happens when the demand outruns the capacity. So the method refuses to start with the behaviour. It starts with an inventory of the specific, situational problems that reliably precede trouble — not 'is defiant' but 'difficulty starting maths homework after dinner on weeknights.' Then it runs a three-part conversation at a neutral time, never during the meltdown. First you gather the child's concern, and you keep asking until you have something you did not already know — 'I'm tired by then and I don't get the first question and I don't want to say so.' Second, you state your concern in one plain sentence — 'I need it done before you're too tired to think.' Third, you invite them to invent a solution with you, and you both have to be able to live with it. You can picture the whole thing: two people, a table, one specific problem, nobody being told what will happen to them.

Shutdown Ritual

A Shutdown Ritual — the term comes from computer scientist Cal Newport's writing on deep work — exists to make the end of a workday real rather than notional. Its founding observation is that for most people work does not end, it just stops being typed. You close the laptop and the day carries on inside you: the message you meant to send, the number that looked wrong, the conversation you are going to have to have. So the practice installs a fixed closing sequence. You go through your inbox, your open tabs, your task list and your calendar; anything unfinished gets a written next action rather than a vague intention; anything unresolved gets a specific time when you will look at it; and then you say a set phrase out loud — Newport's own is 'schedule shutdown, complete' — that marks the boundary. The phrase is not decoration. It is the part that gives the mind a repeatable signal it can learn to recognise, in the way a specific piece of music at a specific hour eventually means bedtime. The whole sequence is meant to take ten to fifteen minutes, run in the same order, at roughly the same time, every working day.

What you get out of pairing them

What you get, if it works, is your evening back — and that is worth saying without dressing it up. But the more interesting thing on offer is subtler. Both practices take a problem that is currently living inside a person, where it can only be acted out, and move it onto a surface where it can be examined. A child who cannot say 'I don't understand the first question' will show you that fact by throwing the book. You at 10pm, who has not written down what the actual next step on the Henderson thing is, will show yourself that fact by re-running the same forty seconds of worry for an hour. In both cases the problem is not being solved because it has never been fully stated. Once it is stated — out loud, in words, in front of someone or at least on paper — it becomes a thing with edges, and things with edges can be worked on, delegated, scheduled, dropped, or found to be smaller than they felt. The reason to bring these two together is that a parent already running these conversations has a well-practised skill for opening a problem honestly and almost no habit of closing one on a cadence, and a person already running a shutdown routine has a reliable closing habit that never once asks the intrusive thought what it is actually worried about.

What Collaborative Problem Solving rests on

• Externalizing makes pattern visible randomised trials
  Putting an internal pattern into an external, structured form makes it visible and available for examination instead of remaining implicit and merely enacted.
• Parts have positive intent clinical practice
  A behavior that looks like misbehavior is usually a mistaken strategy aimed at meeting a legitimate underlying need rather than an expression of badness itself.
• Collaborative solution generation increases buy in observational studies
  Jointly generating a solution with the person who has the problem produces more durable follow-through than an imposed solution.

What Shutdown Ritual rests on

• Consistent cues build conditioned response plausible mechanism
  Repeating the same phrase or action at the same transition point conditions an association between that cue and the state it is meant to trigger.
• Ritual closure signals task completion plausible mechanism
  Performing a fixed closing sequence tells the mind a work period is finished even when more could theoretically be done which reduces residual intrusive thoughts about it.
• Unfinished tasks intrude until marked closed observational studies
  Uncompleted tasks tend to intrude on attention more than completed ones so explicitly marking a task as closed for now reduces that intrusion even without finishing it.

Where they actually meet

Look at what each practice does to an unfinished thing and the same move shows up twice on different material. Greene's conversation takes a problem that only exists as a repeated blow-up and gets it said in words that both people can hold — 'starting maths after dinner is hard because I'm tired and I don't want to admit I'm lost.' Newport's routine takes a problem that only exists as a nagging half-thought and gets it written as a line on a page — 'Henderson: draft two sentences on pricing, 9:15 tomorrow.' Neither has solved anything at that moment. What has changed in both cases is that the problem is now outside the person, in a form that has edges. That is the whole reason the shutdown phrase works at all, on the mechanistic account: you are not saying the work is finished, you are saying it has been recorded, and the part of you that was keeping watch over it can stand down because something more reliable than vigilance is now holding it. Greene's version has a second party in the room, which is a real difference and not a cosmetic one. But the underlying move — get the thing out of the person and onto a surface — is the same in both.

The principle underneath both

An unresolved thing that has never been fully said keeps operating from behind you, and the way it operates is by making you do something — explode, check, re-run, avoid. Say it fully, in concrete words, in a form that persists outside your head, and it stops driving and starts being something you can look at and decide about. That pattern is not owned by either lineage. Greene did not invent it and neither did Newport; it is also what a written worry list, a supervision session, a confession, and an itemised bill all do. There is a second pattern here worth naming separately, because it changes how you run both practices. The behaviour you want to get rid of is usually loyal to something. The child's refusal is protecting them from being visibly lost in front of you. The 10pm replay is a part of you doing night-watch on something it does not trust has been captured. If you treat either one as simply bad and try to suppress it, you are arguing with a guard who has a genuine reason to be standing there. If you address the reason, the guard has somewhere to go.

What that makes the real goal

The goal, stated so it does not belong to parenting or to productivity writing: get the unspoken concern out of the body and onto a shared surface, in words specific enough to act on, and then mark it closed for now so it stops running as a background process. Notice what that objective does not specify. It does not say the problem must be solved — both practices work partly by separating 'stated and scheduled' from 'finished.' It does not say who the second party is; it can be a child, a colleague, a page, a phone note. It does not say when. Once the goal is separable from the tradition, you can ask a much better question than 'should I do Collaborative Problem Solving or a shutdown routine.' You can ask: for this particular open loop, what is the honest statement of the concern, who or what needs to hear it, and what marks it closed?

Other practices that serve the same goal

Several other practices aim at the same target, and some carry evidence neither of these two has. Scheduled worry time — from cognitive behavioural work on generalised anxiety, where you postpone worrying to a fixed half-hour slot and write the worry down when it arrives outside that slot — has been tested in controlled trials and does reduce how much worry intrudes across the rest of the day. That is the closest thing to a trial-backed version of what a shutdown phrase is trying to do. Expressive writing, the fifteen-to-twenty-minute unedited write-out associated with social psychologist James Pennebaker, has a large and mixed literature but repeatedly shows reductions in intrusive thinking about the thing written about. On the relational side, Nonviolent Communication — Marshall Rosenberg's format of stating a plain observation, the feeling it produces, the need underneath it, and a specific request — is essentially the same discipline as Greene's first two steps, with better-developed language for the 'say your own concern in one sentence' part, though its own trial base is modest and mostly in nursing and education settings. And Getting Things Done, David Allen's system, contributes a specific idea worth stealing: the reason a note fails to quiet the mind is usually that it names a topic rather than a physical next action. 'Henderson' does not close a loop. 'Draft two sentences on pricing' does.

A tradition that already does this

This is not a foreign graft onto Collaborative Problem Solving; the closing half is already in it and is the half people drop. Greene's own assessment tool is a written inventory of specific unsolved problems, meant to be revisited — problems get worked one at a time, solutions get tried, and the list is supposed to be reviewed so that a solution that stopped working comes back around rather than quietly failing. In practice, families run the empathy conversation, feel the relief of it, and never return to the list. So the shutdown routine is not adding a step Greene would not recognise. It is supplying the discipline his own structure asks for and rarely gets: a fixed time, a fixed order, and a spoken marker that says this round is closed.

Run the parenting conversation on the thought that keeps waking you

Here is the concrete move. Tonight, when you do your closing sequence, take the one item that you know will come back at you later — not the whole list, one item — and run Greene's three steps on it, with yourself as both parties. Step one, the empathy step, is the one nobody does properly. Write the question at the top of a page: what's up with this one? Then answer it in the worry's own voice, not in your competent daytime voice, and keep going past the first answer. 'I'm worried about the Henderson thing' is not an answer; it is a label. Keep asking. The answer you want sounds like: 'if I send it as it stands, Priya will read the pricing paragraph and think I've already agreed to the discount, and I won't get to correct that before Thursday.' Now you have something with edges. Step two: state your own concern in one plain sentence, and let it be genuinely selfish, because the method requires both concerns to be real. 'I have two hours with my kid tonight and I lose them to this every week.' Step three: invent something that answers both. Not a resolution to worry less — that answers only your concern and the night-watch part of you will correctly ignore it. Something like: '9:15 tomorrow, before anything else, rewrite the pricing paragraph so it reads as a range, and send it by 10.' Write that line down where the closing sequence can see it. Then say your phrase. What has changed is that the phrase is now honest — the thing genuinely is recorded and genuinely has a time — and the part of you that would have kept watch has been given the one thing it was actually asking for. When this works you will notice it as a specific absence: the thought arrives once around 9pm, you recognise it, and it does not re-enter. When it is not working, the thought comes back with a new detail attached, which almost always means the step-one answer stopped short of the real concern.

Give the family's list of unsolved problems a closing time

The failure mode in real households is not the conversation. Parents who try Greene's approach are often startled by how well the first conversation goes. The failure is everything after it: the solution gets tried for four days, it half-works, nobody returns to it, and within a fortnight you are back to consequences — not because the method failed but because nothing in your week ever asks how it went. Newport's structure fixes exactly that gap, and it transfers almost without modification. Pick a fixed fifteen minutes once a week — Sunday evening works for most people because the school week has not started arguing with you yet. Have the written list of specific unsolved problems in front of you, phrased situationally: 'difficulty leaving the park when asked,' 'difficulty starting maths after dinner,' not 'won't listen.' Go down it and mark each one with one of three things: in trial, needs another conversation, or retired. 'In trial' means the agreed solution is still being tested and you are not going to fiddle with it this week. 'Needs another conversation' means the solution is failing, which in this method is information rather than a verdict — it nearly always means one party's concern was not fully on the table, and the next conversation goes back to step one rather than to a stricter version of the same solution. 'Retired' means it stopped being a problem, and you should say that out loud, because parents almost never notice the ones that resolved. Then close the same way every week: a set phrase, said aloud, that means you are not carrying the list into the evening. If you have a partner, say it to each other. Fifteen minutes, same slot, same order. The point is not tidiness. It is that a solution nobody reviews is indistinguishable from a solution that quietly died, and the review is what keeps a hard conversation from becoming a one-off.

Nine forty on a Tuesday, the laptop shut an hour ago

You are on the sofa. The laptop has been closed since half eight and you did the sequence — inbox to zero-ish, tabs closed, tomorrow's three things listed, phrase said. It has not worked. What is running is not even the work; it is the twenty seconds where your daughter said 'fine, whatever' and went upstairs, and underneath that is the maths homework, which is now four evenings in a row. Notice that you have two open loops from completely different parts of your life and both are doing the same thing to you, which is running on a loop without ever getting stated. So state them. Not in your head. Get the phone note or a scrap of paper. The work one first, because it is easier: what is actually unfinished about it? You write 'send the Priya email' and stop, because that is not it — you know what to send. What is unfinished is that you have not decided whether to hold the price. Write that: 'decide whether to hold the price. 9:15, before email.' The loop goes quiet, noticeably, in about the time it takes to write the second line. Now the harder one. What is your daughter's concern, in her words, that you do not currently know? You genuinely do not know, and that is the honest answer to write down: 'don't know why maths specifically. Ask Thursday, not tonight, not during.' And your own concern, one sentence: 'I don't want the last thing between us every night to be a fight about homework.' That is not a solution and it does not need to be. What you have done is convert two things that were running you into two lines on a page with a time attached to each. Then say the phrase, and mean it this time, because now it is true. The test of whether this worked is not that you feel calm. It is whether at eleven you find yourself re-running the twenty seconds on the stairs. Some nights you still will, and when you do, the reliable tell is that the line you wrote was a topic rather than a next move.

When this is worth reaching for

There is a specific, repeating open loop that crosses the boundary between work and home — which is most of the ones that keep people awake. Three situations in particular call for it. First, when your evenings are being eaten by a small number of the same recurring worries rather than by genuine emergencies: recurring is the signal that the loop is unstated rather than unsolved. Second, when you are already running Collaborative Problem Solving conversations with a child and you can feel them fading — the tell is that you cannot remember what you agreed about the second problem on the list. Third, when your shutdown routine has become a hollow performance: you say the phrase and nothing happens, which usually means you are recording topics instead of concerns and the part of you keeping watch is not fooled. It is also worth trying if you have a partner and the two of you have never once had a fixed time to look at the family's difficulties together outside of a crisis, because the single largest gain from the weekly slot is usually that the conversation stops happening at eleven at night in a whisper.

Where this does not apply

Do not run the parenting conversation on yourself as a substitute for having it with the child. Greene's method gets much of its effect from the second person actually being in the room and actually being surprised by what they hear; a solo version of it is a useful journaling structure and it is not the same practice. Do not use the weekly review as a channel for a child who is in acute distress, at risk, or in the middle of something that needs help now — a scheduled slot is exactly the wrong container for that, and the method's own instruction is that these conversations happen at neutral times, not that hard things wait for a slot. If the 10pm intrusion is not a work loop but something closer to persistent, distressing worry across many domains that you cannot put down, a shutdown phrase is a thin tool for it and the thing that has actually been tested for that pattern is structured therapy, not a closing checklist; that is worth taking to a clinician rather than to a productivity routine. And if your evenings are disrupted because your workload is genuinely larger than the hours available, no closing sequence will fix that, and treating it as a ritual problem is a way of not looking at the real one. Finally, some people find that writing an open worry down at night makes it more vivid rather than less for the first week or two. If that is what happens to you, move the writing to earlier in the evening rather than dropping it, and if it persists, the practice is not for you and there is no virtue in forcing it.

What the evidence does and does not support

These two sit at very different places on the evidence map, and it matters which one you are leaning on. Collaborative Problem Solving has been through randomised comparisons in child mental-health settings, including head-to-head work against standard parent-training approaches, with results generally showing comparable or better reductions in oppositional behaviour and in parent-child conflict; the literature is not enormous, several trials come from groups connected to the method's developer, and effects in ordinary households outside of clinics are much less well characterised than effects in treated samples. The Shutdown Ritual has nothing like that. It has never been trialled as a named practice. What supports it is one step removed: laboratory and field work showing that unfinished goals occupy attention, and specifically a set of studies finding that writing a concrete plan for an unfinished task reduces how much that task intrudes afterwards. Read carefully, that supports the writing-a-next-action part of the routine and says nothing at all about the spoken phrase, which rests on a plain conditioning argument rather than on any test. So if you want the better-supported half, it is the written next action, not the ritual around it. As for the combination described on this page — running the empathy step on your own evening worry, or putting a weekly closing review on a family's problem list — there is no study of it. To test it you would need something like a household-level trial in which families already using the method were randomised to add a fixed weekly review or not, with follow-through on agreed solutions measured at three and six months. Nobody has run that. What is written here is a mechanism argument from two practices that share a move, offered as something to try and watch, not as a finding.