⚖️ Module 15 — Shame-Attacking and the Behavioral Exercises | REBT Course
Updated: Oct 16

Free Course by Everything IFS Academy | Therapeutic Modalities Series
Help us reach the people who may need this most by sharing this free course with friends, family, colleagues, online communities, social media or on your own website.
🎧 Audio – Listen on the go - (coming soon)
⚖️Module 15 — Shame-Attacking and the Behavioral Exercises
Every belief taught in this course has been caught, cross-examined, and rebuilt on paper and in imagination. Now REBT takes the work outside, because Ellis was adamant that people rarely think their way into new acting; mostly, they act their way into new thinking. The flagship of REBT's behavioral toolkit is the shame-attacking exercise: deliberately doing something harmless but mildly embarrassing, on purpose, in public, while refusing to put your worth on the line for even one second of it. It is the boldest, funniest, most beloved move in the whole system, the one people remember decades after learning it, and beneath the comedy the exercise turns dead serious: a field experiment designed to disprove shame's central claim with evidence no argument could ever supply.
What A Shame-Attacking Exercise Is
A shame-attacking exercise is a small public act, chosen precisely because it’s the kind of thing shame forbids: silly, undignified, attention-drawing, and completely harmless. The classic canon, built up over decades of REBT practice, has become legend:
Walking a banana on a leash down a busy street, pausing occasionally to let it rest.
Singing off-key in public, at full volume, a verse or two at the bus stop.
Asking for directions to the building you are standing directly in front of, and listening earnestly to the answer.
Calling out the stops on the subway or elevator like an old-fashioned conductor, announcing the floors to no one who asked.
Wearing something ridiculous, a loud hat or mismatched shoes, for an ordinary afternoon of errands.
The assignments sound like pranks, and the design underneath is precise. Shame, as a belief system, makes two claims: looking foolish would be catastrophic, and being seen that way diminishes what a person is. A shame-attacking exercise arranges a direct test of both claims under controlled conditions, and the results come back the same way every time, which is exactly why the exercise works:
The catastrophe never comes. Beforehand, the dread predicts something enormous: gasps, ridicule, some vague social annihilation. What actually happens is famously, almost insultingly, little.
Strangers barely notice. This is the great discovery of every first-time shame-attacker: other people are busy. They're in their own heads, running personal ABCs, and the banana gets a glance, half a smile, and the sidewalk moves on. The spotlight shame promised turns out to be a lamp nobody plugged in. And the few who do look don't affect anything that matters; a stranger's passing "that's odd" was never load-bearing for a life, however much the shame apparatus insisted it was.
Worth survives foolishness intact. The person walks home having been maximally silly in public, and notices they are precisely who they were an hour ago. Nothing structural happened. The deed was ridiculous; the doer came through unrated, which is the unconditional self-acceptance of Module 9, no longer held as a philosophy but demonstrated, in the field, with a banana.
The exercise’s spiritual ancestor is worth a nod: a terrified nineteen-year-old running his own crude version of it on a hundred park benches in the Bronx, a story that belongs to Module 2. Everything REBT later refined was already in that experiment’s DNA.
The Rules That Keep It Useful
A shame-attacking exercise is not random chaos; it’s an instrument, and instruments have specifications. The non-negotiables:
Harmless to self and others. Rule out danger of any kind, along with risks to a job, a relationship, or anyone's safety. The exercise attacks shame, not livelihoods.
Legal. Fully and boringly so, without exceptions or clever edges. The point is embarrassment, never trouble.
Never distressing to bystanders or targeting anyone. The exercise makes its performer the only spectacle. Nothing that frightens people, mocks anyone, imposes on someone's day, or drafts an unwilling audience into discomfort. Singing at the bus stop qualifies; serenading a stranger who can't leave does not.
Aimed at a specific belief. Generic "do something embarrassing" is weaker than an exercise built around the exact shame a person carries. Someone whose belief is "I must not look incompetent" gets more from asking an obviously naive question in a shop than from a loud hat; a person carrying "I must not take up space" needs an act that visibly asks for attention rather than a private oddity. A test for matching exercise to belief: does the anticipation produce the precise feeling of "people will think I'm foolish"? When the answer is yes, it's calibrated. Anticipation that stirs some other feeling instead, dread, moral discomfort, disgust, means it's testing something else.
Repeatable. One exercise is a demonstration; a sustained run of them is what installs the change. The best exercises can be done again in different settings so the evidence accumulates. A once-only stunt, the wedding-toast dare, doesn’t allow that. And let the anticipation run its course rather than rushing through to get it over with; the dread beforehand is part of what’s being worked with, and it’s where the belief states its predictions most clearly.
Paired with rational self-talk during the act. This rule is the one people skip, which voids the whole exercise: stripped of the inner work, it's just embarrassment with extra steps. While the banana walks, the walker works, silently rehearsing the honest sentences: "I'm doing a silly deed, and I'm a fallible human doing it. My worth is not on this sidewalk." The behavior supplies the evidence; the self-talk makes sure it gets filed under the right belief.
Sized to the person. The correct dose is mildly-to-moderately uncomfortable, the level where the dread is real and the act is doable. And here the calibration matters enough to say plainly: mild discomfort is the point, flooding is not. If an exercise feels like too much, that is a signal to shrink it, the loud hat before the banana, humming before singing, never a failure. A smaller exercise actually performed beats a heroic one avoided, every time, and the belief revises on completed evidence only. Anyone whose social fear makes even small versions feel overwhelming deserves a qualified professional as a companion for this work; that's proper use of the tool, not a limitation of the person.
Risk-Taking Exercises
The first behavioral sibling trades silliness for stakes. A risk-taking exercise is deliberately doing a valued thing that carries a real but small chance of failure or rejection: asking for the discount at the register, submitting the piece to the magazine, saying the unpopular opinion in the meeting, inviting the new acquaintance to coffee, pressing the landlord for the repair, negotiating the quote. Where shame-attacks target the fear of looking foolish, risk-taking exercises target the fear of hearing no.
The design follows the same field-experiment logic: the belief "rejection would be unbearable and must be avoided" gets tested against real refusals, deliberately collected, in small denominations. A familiar discovery arrives: the no comes, or it doesn't, and either way the person survives it standing, until the idea that no is a catastrophe runs out of evidence.
One distinction keeps this sibling honest: a risk-taking exercise is not recklessness, and the difference is exact. The risk taken is small, chosen, and in service of something genuinely wanted; the downside is a bruise, never a wound. Asking for the discount costs at most a shrug, while betting the rent does not qualify, and neither does the grand gesture that gambles a friendship. Recklessness wagers things that matter; risk-taking exercises spend only the currency of momentary discomfort, and buy back, purchase by purchase, a life no longer priced by the fear of no.
Staying-In-There Assignments
The second sibling targets a different belief entirely: the "I can’t stand this" of low frustration tolerance, taught in Module 6. A staying-in-there assignment is deliberately remaining in an uncomfortable, completely safe situation longer than the escape urge wants: staying in the boring meeting without the phone, riding out the awkward silence instead of filling it, standing in the slow line without switching, sitting with the itch to leave the gathering for another twenty minutes, finishing the tedious task in one sitting.
The assignment exists to let a person witness, live, the one fact LFT depends on them never observing: discomfort, stayed with, peaks and falls on its own. The urge to escape rises, crests, and, if nobody obeys it, subsides, usually within minutes, and what remains afterward is a person still in the room, demonstrably standing the thing they couldn't stand. Escape always interrupts the curve at its peak, which is why the belief "unbearable" never gets contradicted; remaining in place lets it run to completion for once, and that finished arc is the evidence. The same self-talk rule applies here as everywhere in this lesson: in the moment, the sentences run, "I hate this, and I'm standing it, visibly, right now", so the body's data reaches the right belief.
Three More Tools From The Behavioral Wing
Shame-attacking, risk-taking, and staying-in-there are the three exercises REBT is known for. A quieter trio travels with them, and each tool solves a problem the famous ones leave open.
Acting as if
Ellis's name for behaving as though the rational belief were already fully held, before it comes naturally. A person who has built the flexible preference in Module 12 but still feels the pull of the old must makes the phone call, applies for the job, shows up at the event, the way someone who genuinely held the preference would, rather than the way the demand's constraint would have produced. The action runs ahead of the belief and supplies exactly the experience it needs to settle. An obvious objection is that this is faking it. It would be, if the belief were "everything will be fine and I feel great," because that's false and the data would rest on denial. But the beliefs REBT builds are more accurate than the ones they replace, not less; acting as if a sounder conviction were already in place isn't pretending, it's practicing alignment with something true before it feels that way. Someone learning a language forms correct sentences long before they think in it; the sentences are real, they just come out assembled. Acting as if is the same: effortful while the belief is new, and less so with every use. One condition: it works only after the dispute and the E have been done. Acting rationally on a foundation that was never built is the hollow version.
Rehearsal
Some real situations carry stakes high enough that walking straight in feels like too much, or call for words a person has never actually said out loud. Rehearsal fills that gap. Behavioral rehearsal is talking the conversation through alone, with words, tone, and posture included, several times, until the new response is at least familiar: a raise request, the boundary with a relative, a no to the person who pushes. Role-play does the same with a trusted friend playing the other party, who can respond in different ways so more than one outcome gets practiced. And reverse role-play, an unusually useful variant, has the person argue for their own irrational belief out loud while the friend takes the rational side; a demand that sounded reasonable inside the head rarely survives being defended aloud against a calm opponent. None of this guarantees the real conversation will go the way the run-through did; real conversations never do. What rehearsal does is make the new response available at all; something never once tried tends not to show up under pressure, and rehearsing also reveals, in advance, exactly where the words run out or the belief wavers, so those spots can be worked before the moment itself instead of during it.
Rewards and penalties
The last tool tackles the stubborn problem of not doing the exercise at all. Ellis recommended a plain self-management arrangement, sometimes written as a contract: a genuine reward, contingent on completing the assignment, and a mild penalty for ducking it. The reward is something the person actually enjoys and gets only afterward: good coffee, an episode, time on the hobby. The penalty is where the rules get strict, and Ellis was adamant. It must be a task, the neglected chore, an hour of something dull but useful, a small donation to a cause the person dislikes, and it functions as a lever, never a lash. Self-attack is off the table entirely. "I'm so useless, I deserve this" isn't a penalty; it's the global self-rating of Module 7 wearing a productivity costume, and it undoes the acceptance every other part of the course is building. The test of a good one: does it make skipping less likely without leading the person to think worse of themselves when they fall short anyway? Task-based penalties pass. Verdicts fail. And the whole arrangement is scaffolding, not the engine; once an exercise starts producing real change, the doing of it becomes rewarding on its own and the contract can fall away.
Designing any behavioral assignment
Whichever tool is in use, four features separate an assignment that shifts a belief from one that just produces frustration. Make it specific: "call the contractor Tuesday morning and ask for the quote breakdown," not "work on being more assertive." It should be clearly done or not done, so the evidence is clean. Tie it to a particular belief and the exact prediction it makes, which is what separates REBT homework from a generic dare to grit through unpleasant things. And keep it uncomfortable but achievable: if the belief says "if I speak up in a meeting I'll be humiliated," the assignment is a single comment, in one meeting, on a familiar topic, not a keynote. Too easy and no prediction gets tested; too hard and the exercise doesn't get done, or goes badly enough to feed the belief instead of starving it.
Why Acting Against The Belief Changes The Belief
Step back and look at what all of these exercises share, because the common engine is the real teaching of this lesson: beliefs revise on evidence the body collects.
A belief like "looking foolish would be catastrophic" or "I can’t stand tedium" is, at bottom, a prediction, and predictions are exactly as strong as their evidence file. The trouble is that avoidance keeps the file sealed: the person who never looks foolish never collects a single data point about what actually happens, so the prediction stands unchallenged for decades, growing more authoritative every year it goes untested. Arguments alone can’t fully break this, and the reason is now familiar from Module 14’s distinction between knowing lightly and knowing in the bones: the deep layer that manufactures feelings doesn’t read transcripts of debates. It reads experience.
Acting against the belief hands it experience. One lived contradiction outweighs a hundred rehearsed sentences, whether it's a banana walk where the sky visibly failed to fall, hearing no without being wounded, or riding out an urge to flee until it crested and passed, because it's evidence in the only language the deep layer accepts. REBT has a name for this whole family of work, in vivo, which is just Latin for in real life, as opposed to on paper or in imagination. And the thing that makes it REBT rather than plain exposure is the self-talk rule from earlier: the belief gets disputed while the fear is live, so the calm that follows rests on a changed belief, not merely on familiarity. That is also why these exercises aren't just getting numb to embarrassment. A person could grow used to doing silly things and still privately believe each one lowered their worth; shame-attacking, done properly, is the discovery that worth was never on the line in the first place, and that discovery travels to every other situation where shame used to govern, which numbness never does.
One honest question belongs here: what if the outcome is actually bad? The request is refused, the comment lands poorly, the attempt fails in front of people. It happens, and REBT doesn't pretend otherwise. But notice what the belief was predicting: not a bad outcome, an annihilation. A refusal is disappointing, not the obliteration forecast, and a public stumble, though genuinely a failure, does not end the person. So even a rough result disconfirms the catastrophe, and a "failed" exercise is often worth more than a smooth one, because it delivers the one piece of evidence a good outcome can't: that the actual worst case is survivable and recoverable. The traffic then runs both ways: the person who acted against the belief now believes differently, making the next act easier, generating more evidence, and loosening the grip further. REBT's whole behavioral wing is built on this loop. The thinking work of Modules 11 and 12 softens the belief enough to make the act possible; behavior then does what thinking alone never quite could, and the two together are why the therapy has "Behavior" in its name.
The Banana Walk
Idris bought the banana on Tuesday. It has been sitting on his counter for two days next to the piece of string, and the whole arrangement has been staring at him. Idris chose this exercise precisely because it targets his oldest belief, installed sometime around seventh grade: people are watching, and looking ridiculous in front of them is not survivable. Saturday morning, he ties the string to the stem. Trace the whole arc.
The dread beforehand: it starts the night before, right on schedule. The belief runs its previews: someone will laugh openly, another person will film it, he'll see a coworker, the word "unhinged" will occur to strangers. By morning the banana has acquired the menace of a court date. This dread, note, is the exercise already working; it's the old conviction showing itself at full volume, making its predictions on the record, where the next hour can check them.
The act: he picks a moderately busy street, three blocks planned, and starts walking, banana trailing behind on its leash at a dignified pace. The first thirty seconds are genuinely awful, face hot, the certainty of a thousand eyes. And he does the part that makes it an exercise instead of a stunt: the self-talk starts. "This is a silly deed. I’m a fallible human doing a silly deed on purpose. My worth is not on this sidewalk."
The world's verdict, carefully observed: a woman glances down at the banana, then up at Idris, and gives a small nod, as if the banana were a beagle. Two teenagers grin; one says "nice banana, man" with what is unmistakably approval. A man on his phone steps over the leash without breaking his sentence. That's it. Three blocks, maybe forty people, and the complete inventory of consequences is a nod, one compliment, and near-total indifference. The thousand eyes turn out to be about four, all of them mild.
The belief, adjusting: somewhere in block two, the heat in Idris's face runs out of fuel, and something unfamiliar arrives in its place, a loose, almost giddy lightness, the specific feeling of a lifelong prediction failing in real time. He looks as absurd as a person can look, the worst case is fully deployed, and nothing is happening. By block three he's walking slower on purpose. At home, he does the filing deliberately: the scheduled catastrophe never attended, and the man who looked ridiculous for three blocks is intact in every way that counts. Seventh grade's belief doesn't vanish that afternoon, but its evidence file now contains a contradiction it can never fully argue with, and next Saturday, the hard part will be finding something that still feels embarrassing enough to count.
Below this lesson, you’ll find an REBT practice built around one of the skills you just learned, along with a few ways to begin noticing and practicing it in everyday life this week.
REBT Practice: Your First Shame-Attack
This practice happens out in the world, and today's session is the ten minutes of planning it. You'll need a piece of paper or a notes app, and, later this week, a few minutes in public.
Choose your exercise, sized to you. It must be harmless, legal, targeting no one, and mildly-to-moderately uncomfortable, dread you can feel, in an act you can do. The loud hat, asking for directions to the building you're in front of, humming audibly in a store, the mismatched shoes. If your first idea makes your stomach drop hard, shrink it; if it makes you shrug, grow it. The right size makes you laugh nervously.
Write down shame's predictions, on the record, before you go: what exactly does the dread say will happen? Be specific. This list is the experiment's hypothesis, and you'll be checking it.
Write your during-the-act sentences, two or three, in your own words: "Silly deed, unrated doer. My worth isn't on this sidewalk." Rehearse them once out loud. These run on a loop while you perform.
Do the exercise this week. Pick the day and write it down now. During the act, run your sentences, and, this is the scientist's part, actually observe the world's response. Count the reactions. Note their actual size.
Afterward, file the evidence in writing: what shame predicted, what actually happened, side by side. Finish with one sentence about what the doer's worth did during the deed. If, in the moment, it ever tips from uncomfortable into too much, shrink it on the spot or stop entirely; a smaller completed exercise is a success, and there is always next Saturday.
This Week
Alongside the shame-attack, collect one small no on purpose: ask for the discount, the better table, the extra sauce, and notice what the no actually weighs.
Once this week, when the urge to escape a safe discomfort spikes, the line, the silence, the boring stretch, stay three extra minutes and watch the urge crest and fall on its own.
Start a running list titled "predicted vs. happened." Every entry you add this year is one more page in the file the old beliefs can't argue with.
Disclaimer:
Everything IFS Academy is an independent educational platform and is not affiliated with, endorsed by, or connected to the IFS Institute. While we strive for accuracy, errors can occur, and users are encouraged to cross-reference critical information. These courses, lessons, skills, and practices are offered for educational and self-reflection purposes only. They do not constitute therapy, mental health treatment, clinical training, or crisis support, and they should not be used as a substitute for professional mental health care.
Crisis Support: 🚨
If you are experiencing a mental health crisis, feel unsafe, feel at risk of harming yourself or someone else, or feel too overwhelmed to safely use self-directed practices, please pause this material and reach out for immediate support. Contact a licensed mental health professional, call or text 988 in the U.S. or Canada, or use your local emergency or crisis resources.



Comments