The Wicked Problem of Knowing What to Do . . . and not doing it.

There's a passage in my Lose Weight for Good (opens in a new tab) program that I've come close to deleting a hundred times. I ask people to stand naked in front of a mirror, close their eyes, and imagine their ideal body: healthy, vibrant, capable. Then open their eyes and look at what's actually there.

I hate this exercise. I genuinely hate it.

Not because it doesn't work. It does, for some people. But because I know what it might do to others. For every person who feels that jarring awareness that sparks change, there might be someone who closes the book, feels worse about themselves, and never comes back. And that haunts me. Because my job isn't to add to anyone's suffering. My job is to lessen it, to help them find a path forward.

This tension (between what might create breakthrough and what might create harm) sits at the heart of what complexity theorists call a wicked problem. You never know how people will respond. The same intervention that liberates one person might devastate another. And you can't fully know which until after the fact.

That's the swamp we're in with behavioral change. And after losing and regaining over 500 pounds across multiple attempts, reaching well over 400 pounds before my 36th birthday, I know the swamp.

My Encyclopedia of Failure

By the time I walked into my doctor's office that day over 20 years ago (standing because I couldn't fit in the chairs with arms), I was a walking encyclopedia of weight loss knowledge. I could recite calorie counts. I understood macronutrients. I'd tried Atkins, South Beach, Weight Watchers, low-fat, no-carb, meal replacements, and exercise programs. I'd lost 50 pounds, 100 pounds, even 125 pounds at one time.

And I'd gained it all back. Every single pound, plus interest.

The doctor told me I was pre-diabetic. As a physical therapist, I'd seen what diabetes does to people: blindness, amputations, unrelenting pain. The thought terrified me. So I lost 50 pounds in two months. And then, inexplicably, gained it back.

"What the hell is wrong with me?" I remember thinking.

That's when it hit me: I didn't have a knowledge problem. I had an implementation problem.

I knew exactly what to do. What I didn't understand was why I couldn't do it. Why the intelligent, rational part of my brain couldn't coordinate with the part that actually controlled my behavior. Why my evenings ended with my head in the cookie jar despite my mornings starting with perfect resolve.

This is the central paradox not just of weight loss, but of all behavioral change: we know what works, yet we consistently fail to do it. Can you imagine how the world would change if every human being just did the things they already knew would improve their health?

Two Brains, One Body

To understand why behavioral change is so difficult, I had to understand what I eventually called "the elephant and the rider" (a metaphor I borrowed from psychologist Jonathan Haidt).

Your conscious, rational mind is the rider, sitting atop the massive, emotional, habit-driven elephant of your subconscious. The rider knows the facts. It can calculate calories, understand nutrition labels, and articulate all the reasons exercise matters.

But the elephant runs on autopilot. It creates habits: trigger-routine-reward loops that bypass conscious thought. And crucially, the elephant never forgets but also never cleans house. Every behavior you've ever reinforced, accidentally or intentionally, lives somewhere in your neural pathways.

When the rider tries to impose a diet through sheer willpower, it's like trying to steer a six-ton pachyderm with dental floss. You might succeed for a while, but willpower is exhaustible. Research by Roy Baumeister at Case Western Reserve showed that using willpower depletes glucose in the bloodstream and literally makes you tired. By evening, after a day of resisting temptation, your rider is exhausted. The elephant, noticing no one's holding the reins, wanders straight to the cookie jar.

This is why I kept failing. I was trying to solve an elephant problem with rider solutions.

The Algorithm Trap

Most weight loss programs treat weight loss as a purely algorithmic problem. And I get why. The mechanics seem straightforward:

• Eat these foods, not those foods
• Consume X calories per day
• Exercise for Y minutes
• Follow steps 1, 2, 3

Create a caloric deficit through controlled inputs (food) and outputs (exercise), and weight loss follows as predictably as 2 + 2 = 4.

Except it doesn't.

I'm working on the Harvard BOK Higher Education Teaching Certificate and we were asked to contemplate the differences between behaviorism and constructivism. As I contemplated this, I remembered my work with Lose Weight for Good and penned the blog post The Right Tool for the Right Mind: Rethinking How We Teach, (opens in a new tab) which explores behaviorist vs. constructionist teaching philosophy for different kinds of cognitive tasks.

For algorithmic tasks, behaviorist approaches work beautifully: direct instruction, systematic practice, immediate feedback, rewards for compliance. Dan Pink's research showed that for these routine tasks, external rewards can actually enhance performance.

But for heuristic tasks (complex, open-ended challenges requiring creativity), those same external rewards backfire. They narrow focus and shut down the flexibility needed for insight.

Here's what I realized: while the mechanics of weight loss are algorithmic (calories in, calories out), the implementation is profoundly heuristic. It requires:

• Navigating social pressure at restaurants and family gatherings
• Managing stress without using food as comfort
• Building new identity structures ("I'm a person who..." versus "I'm on a diet")
• Restructuring decades of unconscious eating habits
• Maintaining motivation when progress plateaus
• Adapting strategies as life circumstances change

These aren't problems you solve by following a script. They're what complexity theorists call wicked problems: situations where the solution changes the problem, where every individual's solution looks different, where there's no stopping rule that tells you you're done.

This is why my program doesn't start with meal plans. It starts with something that initially seems absurd: spending weeks building a "lighthouse" in your mind.

Building the Lighthouse: The Constructivist Phase

With all this weight loss and gain, I realized I needed a completely different approach. Not another diet. A fundamental reconstruction of how I thought about myself and my relationship with food and my body.

I started studying success itself. Not just weight loss success, but success in business, relationships, sports, art: any domain where people accomplished difficult things and sustained those accomplishments.

A pattern emerged: successful people didn't just follow better algorithms. They had better internal operating systems. Different beliefs, clearer visions, deeper purpose.

So I borrowed from constructivist education theory: the idea that you can't just transmit knowledge to people. They have to actively construct understanding by fitting new experiences into their existing mental models.

When I ask clients to identify limiting beliefs ("I can't lose weight because it runs in my family"), then systematically deconstruct and rebuild them with evidence, I'm using constructivist principles. I'm not transmitting knowledge; I'm facilitating the construction of new mental models.

Consider my "laddering" exercise for discovering purpose. I ask "why" repeatedly (annoyingly, like a persistent three-year-old) until we reach bedrock:

"Why do you want to lose weight?"
"To look good."
"Why is that important?"
"To be healthy."
"Why?"
"To feel better."
"Why?"
"To do more with my family."
"Why?"
"Because I love them and want to be with them."
"Why?"
"Because I don't want to be alone."

Now the weight loss journey has connected to something deep and visceral. It's no longer about following a diet (algorithmic compliance). It's about becoming someone who creates and maintains relationships (heuristic identity construction).

This aligns with Pink's research: for heuristic tasks requiring sustained effort, what works is intrinsic motivation: autonomy, mastery, and purpose.

You can't drill someone into being intrinsically motivated. You can't reward them into having a compelling vision of their future self. These emerge through a constructivist process: messy, individual, requiring struggle and discovery.

But here's where I learned something crucial: constructivism alone doesn't finish the job.

The Implementation Gap

Imagine if my program consisted entirely of:

• Journal about your feelings around food
• Explore your authentic relationship with your body
• Discover what movement brings you joy
• Construct your own understanding of nutrition

Some people would flourish. But many would drown in ambiguity. They'd understand the why of change but lack the how. They'd have beautiful vision boards and no practical skills for declining dessert at a business dinner.

I learned this from my physical therapy practice. I'd see patients who understood conceptually that they needed to exercise, who had all the motivation in the world, but who simply didn't know how to begin. Or who began and immediately hurt themselves because they lacked basic form.

Pure constructivism can leave learners floundering without adequate scaffolding, reinventing wheels badly, missing crucial knowledge that experts have already systematized.

You don't need to rediscover through personal struggle that vegetables have lower calorie density than cheese (0.41 calories per gram versus 4.03). That's established knowledge worth transmitting directly.

So after the vision-building phase comes what I call "training the elephant." And here, strategic behaviorism enters the picture.

Training the Elephant: The Algorithmic Phase

My approach to habit change is ruthlessly behavioral because that's what works for the elephant:

Identify the trigger: What cues the unwanted behavior?
Map the routine: What's the automatic behavior?
Find the reward: What does the behavior actually provide? (And be honest: often it's not what you think.)

Then systematically intervene:

• Avoid triggers where possible
• Create new triggers for desired behaviors
• Substitute routines that provide the same reward
• Make rewards immediate

This is pure behaviorism: stimulus-response conditioning, environmental modification. And for the algorithmic components of weight loss, it works:

• Building the habit of weighing yourself frequently
• Automating breakfast choices
• Establishing an exercise time
• Tracking food intake

Let me give you a personal example. I wanted to start walking in the evening, but couldn't get started. Every night, I'd think "I should walk," then find an excuse. Too tired. Too cold. Too whatever.

So I set a commitment that required zero willpower: "I will walk to my front door and open it."

That's it. For two weeks, I walked to the door, opened it, stood there like an idiot, and went back inside. My wife thought I'd lost my mind.

But here's what happened: Eventually, standing at an open door felt so absurd that the barrier to actually walking evaporated. The habit wasn't "go for a walk" (that was too big). The habit was "walk to the door and open it." Once that became automatic, the next step was almost effortless.

This is the brilliance of a behaviorist strategy: make the algorithmic component so simple that compliance becomes automatic, then gradually shape toward the desired behavior.

The Integration: Why You Need Both

After a couple of decades of research and experience, I can now see why weight loss requires both approaches:

The constructivist phase builds the foundation:

• Compelling vision (the lighthouse)
• Deep purpose (the "why" that sustains effort)
• Identity shifts ("I'm a person who..." not "I'm on a diet")
• Self-knowledge (understanding your personal triggers)

The behaviorist phase enables implementation:

• Habit automation (reducing willpower demands)
• Environmental design (making good choices easier)
• Systematic practice (building skills to automaticity)
• Immediate feedback (learning what works)

The art (and this is where it becomes teaching, not just telling) is knowing when to use which tool.

Use Constructivism When:

• Exploring motivation and purpose
• Building belief in possibility
• Developing self-awareness
• Creating vision and identity
• Working through emotional relationships with food

Use Behaviorism When:

• Building automatic routines
• Establishing measurement systems
• Creating environmental supports
• Practicing specific skills
• Responding to specific triggers

The Complications: Why It's Never Clean

But just as the education essay warned, it's never this tidy. And this is where we circle back to that mirror exercise I hate.

Complication 1: What works for one person devastates another.

That mirror exercise? For some people, it creates powerful motivation. The anger or frustration becomes fuel. They channel it into immediate action, and it kickstarts their journey.

For others, it reinforces shame, triggers despair, or activates trauma. They close the book and never come back. I've added to their suffering instead of lessening it.

This is the wicked problem aspect: I can't know in advance which person you are. Neither can you, necessarily, until you try it. And by then, if it was the wrong intervention, the damage may be done.

I've considered removing it. I've rewritten the setup a dozen times, trying to make it more compassionate, more optional, more qualified. But here's the brutal truth: sometimes awareness requires discomfort. Sometimes you need to see the gap between where you are and where you want to be.

The question is: how do you create awareness without creating shame? How do you generate the discomfort that motivates without the discomfort that paralyzes?

I don't have a perfect answer. This is the art, not the science. It requires reading the individual, their history, and their readiness. And in a book or audio program, I can't do that. I can only offer the tool and trust people to use it wisely or skip it entirely.

Complication 2: What's heuristic becomes algorithmic over time.

When you first start tracking food, it's a voyage of discovery. You're shocked to learn that your "small" serving of nuts contains 400 calories. This requires constructivist exploration: examining assumptions, building new mental models.

But after months of tracking, reading labels becomes automatic. The heuristic task has compressed into an algorithm. Now the question becomes: do you still need to track?

My answer: it depends on whether you're gaining weight. If yes, return to tracking (back to the algorithm). If maintaining successfully, you can trust the elephant's training (though periodic check-ins remain wise).

Complication 3: Individual differences matter enormously.

Some people thrive with structure. They love tracking macros, following meal plans, having clear rules. The security of knowing exactly what's expected enables them.

Others feel suffocated by rigidity. They need flexibility, experimentation, permission to adapt. Handing them a prescriptive meal plan might trigger rebellion.

I handle this through my "PROGRESS" commitment system, a meta-framework that helps individuals design their own solutions:

• Purpose: What's the why?
• Ready: Am I prepared for this?
• Objective: Is it measurable?
• Gradual: Small enough to achieve?
• Realistic: Can I actually do this?
• Exact: Clearly defined?
• Schedule: When specifically?
• Sustainable: Can I maintain this long-term?

This framework is itself constructivist. It teaches people to design their own algorithmic solutions. It's teaching the skill of learning to learn about yourself.

The Wicked Problem Framework

Weight loss is a wicked problem in the technical sense. Horst Rittel and Melvin Webber defined wicked problems as having:

• No definitive formulation
• No stopping rule
• No test for a solution
• Every attempt changes the problem space
• Every case is unique

Traditional diets fail precisely because they treat a wicked problem as if it were a tame, algorithmic puzzle. They offer standardized solutions to irreducibly individual situations.

But pure constructivism also fails, because wicked problems still contain algorithmic components that benefit from systematization.

The solution is strategic pluralism:

1. Use constructivism to map the wicked problem space: Help individuals understand their unique situation, constraints, motivations, and possibilities.
2. Use behaviorism to master the algorithmic components: Directly teach skills, automate beneficial routines, design supportive environments.
3. Iterate between the two: As you progress, what was heuristic becomes algorithmic. As life changes, what was algorithmic becomes heuristic again.

The Meta-Skill: Learning to Learn About Yourself

The deepest insight from my decade of research is that the ultimate goal isn't weight loss. It's developing the meta-capacity for self-directed change.

I call this installing the "For Good Operating System," an antivirus program for your mind that continuously monitors for drift and enables course correction.

This meta-skill is inherently constructivist. You can't be taught "how to be self-aware" through instruction and practice alone. You have to build it through repeated cycles of:

• Questioning (What's happening? Why?)
• Experimentation (What if I try this?)
• Reflection (What did I learn?)
• Adaptation (How do I adjust?)

But the tools you use can be algorithmic:

• Daily weighing (measurement)
• Food logging (tracking)
• Weekly commitment reviews (systematic reflection)
• Success journaling (evidence collection)

The structure enables the exploration. The algorithmic practices create the feedback loops that make constructivist learning possible.

What This Means for You

If you're trying to lose weight (or change any entrenched behavior), here's what I've learned:

Start with construction, not compliance:

Before setting specific behavioral goals, do the lighthouse work:

• Excavate limiting beliefs and rebuild them
• Create compelling vision
• Connect to deep purpose
• Build identity bridges

Then strategically deploy behavioral tools:

• Start with the smallest possible habits
• Focus on automation, not willpower
• Use the PROGRESS framework
• Build one habit at a time until it's automatic

Recognize when you need which approach:

• Feeling stuck or losing sight of why? Return to constructivism. Rebuild the lighthouse.
• Making progress but feeling scattered? Return to behaviorism. Tighten up the systems.

Accept that maintenance requires different skills than loss:

Keeping weight off requires shifting from "loss" goals to "life" goals. The habits that worked during loss might need modification during maintenance. This requires ongoing constructivist work: continuously rebuilding identity and purpose as your body and life change.

And here's something crucial I learned: plateaus aren't failures.

When people hit plateaus (when weight loss stalls), they get upset. They want to keep losing. But research shows that people who lose about 10% of body weight at a time, taking "plateau breaks," tend to keep it off long-term.

Why? Because these breaks allow your body's physiology to recalibrate. And they allow your mind to adjust. A plateau is actually an opportunity to learn the most important skill: maintaining weight. Take notes. What are you doing during the plateau? Because those behaviors (the ones keeping your weight steady) are exactly what you'll need to do forever.

The Teacher's Dilemma

For those helping others (coaches, therapists, educators, leaders), this framework offers both guidance and humility.

You need to be both sage on the stage and guide on the side, but strategically, not randomly:

Provide structure when people need:

• Clear instruction
• Systematic practice
• Immediate feedback
• Environmental design

Provide exploration when people need:

• Powerful questions
• Space for struggle
• Support for experimentation
• Help constructing meaning

The art is diagnostic: reading what this person needs right now.

But here's the humbling part: you'll get it wrong sometimes. You'll offer structure to someone who needs freedom. You'll offer freedom to someone who needs structure. You'll use the mirror exercise on someone it harms instead of helps.

This is inherent to wicked problems. You can't know in advance. The solution changes the problem. Every case is unique.

What you can do is:

• Make your intentions transparent
• Offer multiple pathways
• Monitor for signs you're on the wrong track
• Hold your methods lightly
• Stay connected to compassion

That last one matters most to me. My goal isn't to have everyone succeed with my methods. My goal is to lessen suffering and increase wellbeing. If my method isn't working for someone, I need to be willing to change course, even if it means they need a different approach entirely.

The Deeper Truth

My journey from 400 pounds to sustained health wasn't about finding the right diet. It was about understanding that lasting change requires building a different kind of person, not just following a different set of rules.

This aligns with the deepest insight from the education framework: the behaviorism-constructivism debate isn't really about learning theories. It's about what we think human beings are.

Are we stimulus-response machines, shaped by our environment? Or are we active meaning-makers, constructing our own understanding?

The answer is: both.

We are simultaneously:

• Habit-driven elephants and rational riders
• Products of conditioning and agents of choice
• Shaped by environment and shapers of meaning
• Needing structure and needing freedom
• Requiring instruction and requiring discovery

The wisdom isn't in choosing one vision over the other. It's in recognizing when each is needed and having the full toolkit available.

And perhaps most importantly: recognizing that behavioral change is a wicked problem that will require both tools, used strategically, over time, with compassion for yourself when you get it wrong.

The Invitation

I've spent a lot of time in this swamp. In fact, I have recently returned from a stress-induced exile to the swamp, having gained almost 50 pounds since moving to Florida. I am now down 45 pounds from that peak. How, you ask? I read my own book again.

So here's my invitation to anyone struggling with behavioral change:

Stop looking for the one right method. Stop believing that either "just follow the plan" or "just find your why" will be sufficient.

Start asking: What does this moment need?

Sometimes you need someone to show you exactly how. Give you the structure, take away the decisions, let you focus on compliance until the habit forms.

Sometimes you need space to struggle, discover, construct your own path.

The art is knowing which is which. And having the compassion to forgive yourself when you choose wrong.

I end my program with this reminder: the "For Good" operating system isn't just about keeping weight off. It's about becoming a force for good: someone who has learned to direct their own development and can help others do the same.

That's not an algorithmic outcome you can drill someone into achieving. And it's not a heuristic insight someone can discover alone.

It's both. It's neither. It's the synthesis.

It's the right tool for the right mind for the right moment, repeated across a lifetime of moments, with as much compassion as you can muster for yourself and others along the way.

That's not a diet. That's a way of being.

And maybe that's the answer to the wicked problem we've been wrestling with all along: recognizing that we're in a swamp, that the footing is uncertain, that we'll sometimes make things worse instead of better, but that we keep trying anyway because the alternative (giving up on ourselves or others) is worse.

I've kept that mirror exercise in my program. Not because I'm certain it's right, but because I'm uncertain it's wrong for everyone. And because navigating that uncertainty, that's the work. That's teaching in the face of wicked problems.

That's all of us, trying to help each other across the swamp, one uncertain step at a time.