Using Regret Minimization Framework to enhance Collaborative Problem Solving

What each one is actually for

Regret Minimization Framework

What is distinctive about the regret minimization framework, next to almost every other structured way of deciding, is that it has no second voice in it at all. You are alone with the choice. You put yourself at eighty in as much detail as you can manage, you look back at this stretch of your life, and you ask whether that version of you would regret not having tried the thing you are hesitating over. There is no step where you gather anyone else's information, no step where the answer gets tested against someone who will also live with it, and no step where you find out you were wrong about what somebody wanted. This is deliberate and it is what makes the tool fast. It is also why a verdict from it lands the way it does. What you carry out of that exercise is a settled position, arrived at through a private process nobody watched, and a settled position is exactly what most conversations cannot metabolise, because the other person is joining at the end of a process they never saw the start of.

Collaborative Problem Solving

Collaborative Problem Solving starts from a claim that changes what a difficult behaviour even is. The founding line is that children do well if they can, which means challenging behaviour is read as a sign that a specific situation is demanding a skill the child does not currently have, rather than as evidence about their motivation or character. The behaviour is the noise a stuck problem makes. The main tool is called Plan B, and it has three steps that are done in a fixed order for a reason. First, the empathy step: you name one narrow, specific, recurring problem and ask what is going on, then you keep asking without arguing until the child's actual concern is on the table in their words, which usually takes longer than adults expect and is where most attempts collapse. Second, you state your own concern in one sentence, not five. Third, the invitation: you say something like I wonder if there is a way we can do both, and you generate solutions together, discarding any that either of you cannot realistically live with. The model names two alternatives for honesty's sake. Plan A is imposing your solution, which is sometimes necessary and always carries a cost. Plan C is deliberately setting the problem aside for now, which is a legitimate choice rather than a failure, and knowing you have it is what keeps you from running a bad Plan B.

What you get out of pairing them

If you already run Plan B well, and parents who do usually run it a great deal better than they run their own adult relationships, there is a gap you may recognise. You will spend forty patient minutes finding out what your nine-year-old is actually worried about regarding homework, and then decide about the move, the job, the school, the money, alone in your head over three weeks and present the conclusion at dinner. The skill exists. It has simply never been pointed at the decisions where the stakes are highest and where the other adult has as much information as you do. What is on offer is not a new technique, it is aiming an existing one at material you have been keeping it away from, probably because the stakes make the collaborative version feel dangerous. Coming at this from the other direction, if you use the age-eighty projection and find it clarifying, the thing you may be missing is what a settled private verdict does to a conversation. Your clarity is real and it was expensive to get. It also means you walk in having already finished, and the other person can hear that in the first sentence regardless of how carefully you phrase it. What CPS lends is a way to arrive with your steadiness intact and your conclusion still genuinely open, which is a different thing from pretending it is open. And there is a version that works on yourself, which is the more surprising use: running the empathy step on your own hesitation, before the projection, often turns up a concern you did not know you had and that no amount of imagining yourself at eighty would have surfaced.

What Regret Minimization Framework rests on

• Self distancing improves judgment randomised trials
  Evaluating a situation from a distanced, third-person vantage point instead of an immersed first-person one improves the quality of judgment about it.
• Long term framing surfaces values practitioner report
  Framing a decision against a long time horizon surfaces which option better matches a person's deeper values and cuts through short-term ambiguity.

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.

Where they actually meet

Both practices get their leverage from distance, and they obtain it by completely different means. The regret framework gets it temporally, jumping forward to eighty so that everything currently loud goes quiet. Collaborative Problem Solving gets it structurally, putting the unsolved problem out on the table between two people instead of leaving it as an implicit pattern one of them is simply enacting and the other is simply reacting to; researchers call that move externalizing, meaning you take something internal and put it into a form outside the person where it can be looked at. In both cases what was running the moment becomes something that can be examined rather than only reacted to: an urgent present-tense feeling in one case, a child's frustration being read as defiance in the other. There is a second shared assumption that matters more for the pairing than the distance does. CPS assumes that behaviour which looks like obstruction is usually a strategy aimed at meeting a real need, badly. Applied to a decision, that assumption reframes your own stalling, or your partner's resistance, as something protecting a concern nobody has named yet, rather than as an obstacle to argue past. The seam, then, is that each supplies the kind of distance the other cannot reach: imagined time in one direction, a second voice and an external structure in the other.

The principle underneath both

The loud version of a problem is almost never the real one, and both practices are built around a device for getting under it. In a decision, the loud version is present-tense pressure, and the device is to move decades away so it stops carrying weight. In a conflict, the loud version is the behaviour, the refusal, the shouting, the shutting down, and the device is to stop responding to it and go and find the concern it is a bad strategy for. Neither field owns this. What both are refusing to do is negotiate with the surface, and what they are both after is the same thing: a concern stated plainly enough that it can be worked with. The shift is the same one in both cases too. Something that was simply happening to you, or through you, becomes something on the table that you can look at, which is a different relationship to the thing rather than a solution to it.

What that makes the real goal

Get every concern that will actually govern the outcome into words before anyone decides anything. That is the goal, and it belongs to neither tradition. Note it is concerns, not positions. A position is I want to take the job, and two positions in a room can only fight or split the difference. A concern is what the position is protecting: I am afraid of spending thirty years resenting a choice I never made, or I am afraid of losing the only place my mother still recognises me. Concerns can often both be met by something neither person had thought of, and positions almost never can. Stated this way, the objective gives you a test that works on any method, including these two: after this conversation, or after this exercise, can I state the other person's concern in a sentence they would recognise as theirs, and can they state mine. If the answer is no, no amount of good procedure has done the job, and if the answer is yes, you are already most of the way there regardless of which method got you there.

Other practices that serve the same goal

Nonviolent Communication is the best-known route to the same thing and, awkwardly, one of the least well evidenced. It offers a four-part sequence, observation, feeling, need, request, meaning you state what happened without evaluation, what you feel, what underlying need is unmet, and one concrete doable request, and its entire design is to get from positions to needs quickly; the evidence base is thin, mostly small studies and practitioner report, which is worth knowing before treating it as established. Motivational interviewing is the strongest-evidenced route here, with a substantial trial literature in behaviour change, and its core move is directly relevant: the reasons for change have to be elicited from the person rather than supplied to them, because reasons you supply get argued with and reasons they say out loud tend not to be. Interest-based negotiation, from the Getting to Yes tradition, makes the same positions-versus-interests distinction for disputes between parties who do not have to like each other; it is a practitioner framework rather than a tested protocol and it is unusually well suited to decisions where money and logistics are the substance. And Internal Family Systems runs the identical move inward, treating a stalling or resisting part of yourself as something with a protective job to do that can be asked about rather than overridden; it comes out of a clinical tradition with a small but growing trial literature.

A tradition that already does this

Interest-based negotiation reached the same conclusion a decade earlier, from labour disputes and diplomacy rather than from children. The Harvard negotiation work that produced Getting to Yes in 1981 argued that parties arrive with positions, that positions can only be split or fought over, and that behind each one sits an interest which is often compatible with the other side's even when the positions are not, so the practitioner's first job is to stop bargaining and start asking what the position is protecting. Ross Greene's model, built in psychiatric inpatient settings where the prior default was straightforward imposition, arrives at the same architecture with different material: gather the concern before proposing anything, and never negotiate with the surface. Two fields that were not talking to each other, working on adversaries in one case and on distressed children in the other, converging on the same first move. That is precedent for the shape and not evidence that combining it with a decision heuristic does anything.

Using the Regret Minimization Framework to enhance Collaborative Problem Solving

When an adult sits down to work out an unsolved problem with a child, the first thing to collapse is usually the adult's own steadiness, back into the accumulated irritation of a behaviour that has happened every evening this week. So use the projection as preparation, before the conversation starts, not during it, because there is no version of running it mid-sentence that does not read as absence. The concrete move takes ninety seconds in the hallway. Ask what the eighty-year-old version of you would think about how you are about to handle these specific five minutes, and be honest about what comes back, which is usually that they would not remember the shoes and would remember the tone. That is the whole exercise. It buys you the patience the empathy step actually requires, which is more than most people budget for: staying with what is going on with the shoes until the child's real concern surfaces, often three or four questions in, without correcting anything they say along the way. There is a second use that matters more, and it is a decision rather than a technique. If you run the ninety seconds and cannot get anywhere near genuine curiosity about what is going on for them, do not proceed. Use Plan C instead, which means deliberately setting this problem aside for now and saying so out loud, and come back to it tomorrow. This is not giving up; it is the model's own third option, and it exists precisely for this. A Plan B conversation entered while irritated reliably turns into Plan A wearing Plan B's clothes: you ask the questions, you nod, and you arrive at the solution you had already decided on, and children detect this immediately and learn from it that being asked is a formality. One honest Plan C costs far less than that.

Using Collaborative Problem Solving to enhance the Regret Minimization Framework

There are two imports here and they are different sizes. The small one first, and it is the one you can do tonight, alone, at a kitchen table with a piece of paper. Run the empathy step on yourself before you run the projection. Write the narrowly defined problem the way you would say it to a child, without any accusation in it: I have noticed I keep not doing the thing I say I would regret not doing. What is up with that? Then answer it, in writing, and do not argue with the answer, which is the discipline that makes the step work. Then drill in, exactly as you would with a child, asking what else and what about it specifically until something concrete appears, and it usually does: not I am scared of failing but I do not want to tell my brother I was wrong about this. Then state your own concern in one sentence. Then the invitation, to yourself: I wonder if there is a way to do both. What this catches that the projection cannot is that your hesitation is protecting something real, and the age-eighty question is designed to override hesitation rather than to interview it. The larger import is for when someone else lives inside the decision. Do not open with your verdict. Run an actual Plan B: name the problem narrowly and in shared terms, ask what their concern is and stay there until you could say it back in their words and they would recognise it, then give yours in one sentence, then invite solutions neither of you brought. Keep the projection's answer, because it is what lets you stay steady while they say something hard, but hold it as your concern rather than as the conclusion. The practical test that you have done this rather than performed it: at the end, could you state their concern in a sentence they would nod at, and did the solution contain anything that was not in your head when you walked in. If both answers are no, that was an announcement with questions attached.

Telling your father you are not taking over the family business

You have rehearsed the eighty-year-old's verdict for weeks, the version of you who stayed out of guilt and resents it against the version who left and built something, and the projection is not ambiguous any more. So you go to see him with the decision made. And the conversation is not with your imagined future self, it is with a seventy-year-old man in his kitchen, and walking in with a settled verdict makes it an announcement with a preamble, which he will hear in your first sentence no matter how gently you build up to it. So you change what the conversation is for. You start with the narrow problem, said plainly and without a conclusion attached: we have both been assuming for about ten years that I would take this on, and I have not once actually asked you what you want out of that. Then you ask, and you stay there, and you do not correct anything. What comes back the first time is about the business, the accounts, the staff. Three questions in, it is something else: that he is afraid the thing he spent thirty-five years building will simply stop existing, and that nobody will remember it happened. That is a concern, and it is not the same as needing you specifically to run it. Only then do you give yours, in one sentence: I am afraid of spending three decades resenting a choice I did not really make. Two concerns on the table, and now there are options neither of you walked in with, a sale with his name kept on it, a written history, a role you take that is not the one assumed. The projection work was not wasted; it is exactly what let you sit through the difficult part without getting defensive. But what kept the conversation from becoming two people talking past each other was asking first.

When this is worth reaching for

The decision is yours to make, and someone else has to live inside it. That is the condition, and it is narrower than it sounds, because plenty of decisions only feel shared. The tell is whether the other person's daily life changes if you go one way rather than the other. A second condition is more specific and easier to spot from outside: you have already decided, you know you have already decided, and you are now planning how to tell them, which is the moment where a good deal of durable damage gets done at speed. A third belongs to parents who use Plan B and would recognise it instantly, when you notice you have been more curious about your child's objection to a bedtime than about your partner's objection to a house move. And there is a case where this is emphatically not the moment: a child is in the middle of an outburst, or an adult is furious, and nothing about either practice works in that state. Concerns cannot be gathered from someone who is flooded and cannot be surfaced by someone who is. Wait for calm, on both sides, and treat the waiting as part of the method rather than a delay in it.

Where this does not apply

The regret framework is built for one person weighing one path against another; Collaborative Problem Solving is built for two people who have different information about what is actually driving a behaviour or a disagreement. Where a decision genuinely involves only you, no second party whose life reshapes around the answer, importing the collaborative structure invents a collaborator who is not there and adds procedure without a real second perspective to meet, and the honest version of that page is that these two barely touch. In the other direction, a genuine crisis behaviour that needs an immediate limit is not well served by pausing for either an age-eighty projection or a full collaborative diagnosis; the model's own account is explicit that some situations call for Plan A, imposing the answer, and that pretending otherwise is not kindness. Two further limits are worth stating plainly. Do not run a collaborative conversation about a decision that is actually fixed. If the answer cannot change, an honest announcement respects the person more than a process that gathers their concern and then does what was going to happen anyway, and people remember the second one for years. And where a relationship involves coercion or fear, the whole structure fails in a way that matters: Plan B assumes both parties can state a concern safely and that neither can simply overrule the other, and in a relationship where one person is frightened of the other, running it hands the stronger party a well-organised way to extract information. That situation needs outside support, not a shared problem-solving conversation.

What the evidence does and does not support

The pieces on this page are graded very differently and the differences track something real. Looking at your own situation from outside it, as though it belonged to someone else, has randomized-trial support for improving judgment quality, drawn from work on psychological distance and mostly on shorter-horizon and emotionally charged decisions. The stronger claim attached to the framework, that a long horizon reliably reveals what a person actually values, is anecdotal and comes from the framework's popularisation rather than from measurement. On the other side, putting an implicit pattern into an external form so it can be examined has trial support from related therapeutic research. Reading challenging behaviour as a poor strategy for meeting a real need is graded clinical, meaning it rests on treatment experience and case series within the tradition rather than on controlled comparison of that interpretation against another. And the claim that jointly generated solutions stick better than imposed ones is observational, which is the weakest of the three and also the one most people find intuitively obvious, a combination worth noticing. Collaborative Problem Solving itself has been evaluated in real settings, including inpatient and school programmes, with reported reductions in restraint and in disciplinary incidents; that literature is about the method as delivered to children with challenging behaviour, and it says nothing whatsoever about adults borrowing Plan B for their own life decisions, which is what half this page proposes. Nobody has tested an age-projection technique combined with a collaborative structure. What is here reasons from each method's separately evidenced mechanism.