Using Collaborative Problem Solving to enhance Commitment Contracts

What each one is actually for

Collaborative Problem Solving

Ross Greene's Collaborative Problem Solving starts from a different premise than most parenting advice: a child who explodes over a demand is not choosing to misbehave, they are missing a specific cognitive skill — flexibility, frustration tolerance, problem-solving — needed to meet that particular demand, the same way a child who cannot read is not choosing illiteracy. Plan B is the structured conversation built to find and build that missing skill, run in three ordered steps, scheduled proactively during a calm window rather than mid-meltdown: first the adult empathizes, asking open questions until the child's actual concern about the expectation is genuinely understood; then the adult states their own concern about the same expectation; then the two generate a solution together that addresses both concerns, rather than the adult simply imposing one. Greene contrasts this with Plan A, unilaterally imposing the adult's solution, and Plan C, dropping the expectation for now to lighten the load — Plan B is specifically the middle path where both people's concerns get solved together.

Commitment Contracts

Ian Ayres, in Carrots and Sticks, formalized a much older idea into a repeatable tool: your future self, standing in the actual moment of temptation, is not the same person, psychologically, as the calmer self who set the goal, and a commitment contract is a way of binding that weaker future self to the stronger present one's intentions. Concretely, you attach a real cost to failure — money forfeited to a charity you actually dislike, a public bet, a monetary stake held and enforced by a referee, often a platform or a trusted third party who verifies whether you met the goal — arranged and signed while you are motivated and reasoning clearly, so the tempted future self inherits a structure it cannot simply talk its way out of.

What you get out of pairing them

Plan B buys something Plan A structurally cannot: a solution the child actually helped build, which means they have a reason to follow it beyond simply avoiding punishment, and it buys the adult accurate information about what is actually hard, rather than a guess. What it does not reliably buy is teeth — an agreement reached collaboratively can still quietly slide once the good feeling of the conversation fades, because nothing in Plan B itself attaches a real cost to drifting away from what was agreed. Commitment contracts buy exactly that teeth: a real, felt stake that makes backsliding costly enough to actually change behavior in the moment it would otherwise happen. What they do not reliably buy is buy-in — a contract imposed on someone, even by their own past self, can breed the same resentment and covert resistance Plan A produces in a child, if the terms were never actually theirs to help write. Put together, Plan B supplies the buy-in a contract needs and the contract supplies the teeth Plan B is missing.

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 Commitment Contracts rests on

• Accountability to others increases follow through observational studies
  Having another person witness or enforce progress raises the likelihood that a commitment is kept.
• Loss aversion strengthens commitment randomised trials
  Framing a stake as a loss to avoid rather than a gain to earn increases adherence because losses loom larger than equivalent gains.
• Precommitment constrains future choice randomised trials
  Binding oneself in advance to a costly consequence removes the option to yield to in-the-moment temptation.

Where they actually meet

Both practices refuse to let the difficult moment be the moment the deal gets negotiated. Greene is explicit that Plan B has to be run proactively, scheduled into a calm window, because a child mid-explosion and an adult mid-frustration are both operating without reliable access to the reasoning either of them needs — attempting Plan B during the outburst just adds a second frustrated party to the first. Ayres is making the identical bet about a single person split across time: the version of you standing in front of a snooze button or a second drink is not going to negotiate well either, so the contract gets written by the version of you that can. One splits the negotiation across two people and insists it happen before the storm; the other splits it across two moments in the same person and insists on the same thing. Different unit of analysis, same refusal to trust the hard moment with the terms.

The principle underneath both

In both models, the entity that looks like it is failing — a child having an outburst, a future self reaching for the thing they said they would not — is not treated as bad or willfully defiant. Greene's framing is explicit: the outburst reflects a missing skill, a mismatch between what is demanded and what the child can currently do, not a character flaw to be punished out of them. Ayres's framing carries the same structure quietly: the tempted future self is not treated as a weaker, worse version of you to be shamed into compliance; it is treated as a predictable mismatch between present-moment incentives and long-term goals, which is exactly why the fix is architectural, a real cost built in advance, rather than moral, trying harder next time. Both practices design around the actual, unglamorous mechanism of the failure instead of moralizing about it.

What that makes the real goal

Both practices are, underneath the different vocabulary, doing the same piece of engineering: getting the terms of an agreement settled by the calmest, most capable version of the people or selves involved, and then building a structure sturdy enough that the terms hold even once a much less capable version of those same people or selves shows up later. Once stated that way, the target is not specific to parenting or to behavioral economics — it is a general problem of protecting a good decision from a future moment likely to unmake it, and any technique that moves the real negotiation earlier and gives the later moment less room to renegotiate is answering the same question.

Other practices that serve the same goal

Contingency management, a well-evidenced clinical technique mainly used in addiction treatment, combines both halves of this objective directly: clinician and patient jointly set a specific, agreed behavioral goal together, then a real incentive, typically a voucher or small cash reward, is administered by the clinician acting as referee for hitting it, with a solid base of randomized trials behind it. Gollwitzer's implementation-intention research pursues the precommitment half without the collaborative half, showing that a specific if-then plan made in advance closes much of the gap between intending something and doing it, even without another person enforcing anything. Restorative justice conferencing, used in some schools in place of suspension, pursues the collaborative half without the financial stake, bringing the people affected by a conflict together to jointly agree on repair, on the finding that agreements reached this way hold up better than ones imposed unilaterally.

A tradition that already does this

Contingency management is not just a parallel case — it is close to a working precedent for this exact pairing already in clinical use. A patient and clinician sit down together, while the patient is motivated and clear-headed, and jointly negotiate the specific target behavior and the reward structure, functioning as Plan B's collaborative diagnosis; the clinician then holds the referee role a commitment contract depends on, verifying results and administering the agreed consequence. It was not built by combining Greene and Ayres, but it independently arrived at the same fusion: negotiate the terms together, then let a real, administered stake carry the follow-through.

Using Collaborative Problem Solving to enhance Commitment Contracts

Most commitment contracts fail quietly for a specific, avoidable reason: the goal and the stake were set alone, by one person guessing at what would motivate their own future self, without anyone testing whether the terms actually address what makes the behavior hard in the first place. Run a compressed version of Plan B on yourself before writing the contract — literally ask, out loud or on paper, what is actually hard about this for me, the way you would ask a child, rather than assuming the answer is simply insufficient willpower. If the honest answer is that you skip the gym because evening sessions collide with a partner's schedule, not because the stake isn't large enough, no contract penalty fixes that; the concern needs a different solution, a morning slot, built the same way Plan B builds one — jointly, between your present concerns and your goal's concerns, before the stake gets attached to terms that were never actually going to hold.

Using Commitment Contracts to enhance Collaborative Problem Solving

Plan B is genuinely good at producing a solution both people agree to in the room, and genuinely weak at making sure that solution survives the next two weeks once the good feeling of being heard has faded and old patterns start pulling back. Borrow the referee structure from a commitment contract and build it into the agreement itself: when the solution Plan B produces involves a specific, checkable commitment — a chore done by a certain time, a device put away at a certain hour — name, together, in the same conversation, what happens if it slips, and who checks. This is not Plan A's imposed consequence smuggled back in through the side door; the stake itself gets negotiated collaboratively, the same way the solution was, which keeps the follow-through mechanism from undoing the buy-in Plan B just spent the whole conversation building.

Tuesday afternoon, before the homework fight starts again

You sit down with your ten-year-old on a calm Tuesday afternoon, well before the usual six o'clock homework standoff, and you ask what's hard about starting homework right after school. He says, after some circling, that he needs to move his body first or he can't sit still enough to actually read the page — the demand itself, sit down immediately, is the actual mismatch, not laziness. You share your concern: homework left until eight o'clock turns into a fight neither of you wants and cuts into sleep. Together you land on a plan — twenty minutes outside first, then homework, timer set by him. You both notice, out loud, that plans like this have slipped back to the old pattern before, usually around day four, so you agree on a specific check: he sets the timer himself and shows you the started page by four-thirty, and if four-thirty comes without it, the screen time he cares about that evening is shortened, a term he helped set, not one you handed down. The agreement is his as much as yours, and it now has something to actually hold it in place once the novelty of a new plan wears off.

When this is worth reaching for

A Plan B conversation — the collaborative one where you and the other person work out a solution together instead of you announcing it — went well, and it landed on something that has to hold for months rather than for one afternoon, and a week later nobody can say whether it is still holding. Or the situation is the mirror image of that: the contract is written, the stakes have been raised twice, and it keeps breaking anyway, because the goal was set for the person rather than with them and nobody has yet asked what actually makes it hard to keep.

Where this does not apply

Do not attach a financial or social stake to an agreement with a young child; the model Ayres built is for adults negotiating with their own future self, and a real cost imposed on a child's failure collapses back into Plan A with extra steps, punishment dressed as collaboration, undermining the trust Plan B depends on. And skip Plan B's collaborative diagnosis step for a goal that is genuinely just a matter of willpower under full agreement about what's hard, adding a diagnostic conversation to a problem that was never actually about missing skills mostly wastes time both people could spend on the contract itself.

What the evidence does and does not support

Collaborative Problem Solving has real randomized-trial support in clinical settings, showing significant reductions in explosive behavior and restraint use, and one head-to-head trial found it comparable to established Parent Management Training for oppositional youth, though most everyday home-use evidence remains observational rather than trial-based. Commitment contracts have solid evidence from savings and financial-commitment programs specifically; effects on health behaviors and habit change are positive on average but considerably more variable, and contracts tend to work best for people who already want the change and need help holding the line, not for manufacturing motivation that was not there to begin with. Nobody has studied using Plan B's conversation structure to generate a commitment contract's terms, or building a contract's referee role into a Plan B agreement; contingency management is the closest existing evidence base, and even that was not designed as a deliberate test of this specific combination.