Incongruity Theory and Cognitive Humor Processing Explained

5.1 What Mental Models Are: Components, Assumptions, and Inference Paths

A mental model is the working structure your mind uses to interpret what it sees and to predict what should happen next. In humor processing, mental models matter because incongruity is rarely random; it is a mismatch between what a model expects and what the stimulus actually provides.

Components of a Mental Model

A mental model usually includes four parts.

1. Entities: the things you think are in play. In a joke, entities might be “a waiter,” “a dog,” or “a contract.”
2. Relations: how entities connect. For example, “waiter serves food,” or “contracts bind parties.”
3. Rules: constraints that govern what is allowed. Rules can be physical (gravity), social (politeness), or logical (if A then B).
4. Goals and perspectives: what matters to the interpreter. A listener might adopt the speaker’s goal (“be helpful”) or their own (“stay safe”).

When a model is active, it doesn’t just describe the world; it generates expectations. Those expectations become the baseline that later gets violated.

Assumptions as Default Fillers

Assumptions are the model’s shortcuts. They let you interpret incomplete information without waiting for perfect evidence. Assumptions can be explicit (“the phrase ‘on the house’ usually means free”) or implicit (“a dog can’t sign paperwork”).

A useful way to think about assumptions is that they are versioned: you can hold a default assumption, then revise it when the stimulus provides a reason. Humor often works by forcing a revision at the last moment, so the listener experiences a quick “wait, that can’t be right” moment before a new interpretation becomes coherent.

Inference Paths as the Model’s Reasoning Route

An inference path is the chain of reasoning your mind uses to move from input to interpretation. It typically includes:

• Parsing: identifying words, roles, and likely structure.
• Model selection: choosing which model best fits the cues.
• Prediction: generating what should follow.
• Constraint checking: testing whether the stimulus violates key rules.
• Resolution: if violated, searching for a revised model that restores coherence.

Different inference paths can coexist. A listener might simultaneously consider a literal model and a figurative model, then commit to one when the punchline clarifies the intended route.

A Concrete Example of Components, Assumptions, and Inference

Consider this joke:

“I told my computer I needed a break. Now it won’t stop sending me KitKat ads.”

Entities: “computer,” “me,” “ads.”

Relations: “computer responds to my request,” “ads are targeted.”

Rules: “a request for a break should reduce work,” “ads are persuasive messages.”

Goals and perspective: the listener adopts the speaker’s goal of taking a break.

The key assumptions are that the computer understands “break” as rest, not as a brand association. The inference path starts with a straightforward interpretation: request → system action → relief. Then the punchline supplies evidence that “break” is being interpreted through a different rule set: marketing targeting based on keyword associations. The listener resolves the incongruity by switching models from “helpful system” to “ad-targeting system.”

Practical Practice: Spotting the Model in One Sentence

Try this micro-exercise with any short line. First, identify the entities and relations. Next, write down one rule you think is being assumed. Finally, state the inference path in one step: “Given X, I expect Y, but the text gives Z.”

For example, with “The meeting could have been an email,” the model includes entities (meeting, email), relations (meetings communicate information), and a rule (meetings are for complex coordination). The assumption is that the meeting’s purpose requires more than an email. The inference path predicts a justification for the meeting; the punchline denies it by reframing the meeting as unnecessary.

Mental models are not mysterious. They are the mind’s organized guess about how the world works, including what it assumes by default and how it reasons when the guess fails.

5.2 Model Selection: Choosing the Interpretation Frame That Fits the Evidence

Model selection is the step where your mind decides which “story” to run. The same words can fit multiple stories, but only one story matches the evidence well enough to keep processing smooth. When the chosen model predicts the next bit correctly, the joke feels coherent until the punchline forces a switch. When the model is a poor fit, you get confusion that never resolves into a single satisfying interpretation.

The Core Idea: Competing Frames

A mental model is a structured set of assumptions about what is happening, who is involved, what counts as normal, and what kinds of outcomes are likely. Model selection happens fast because the brain prefers frames that:

1. Explain the current input with minimal strain.
2. Predict upcoming content better than alternatives.
3. Stay consistent with context cues like setting, relationship, and genre.

A useful way to think about it is as a competition. Each candidate frame generates expectations. The frame that best reduces prediction error becomes the working model.

Evidence Types That Guide Selection

You rarely select a model from raw text alone. You use multiple evidence channels, each with different reliability.

• Linguistic evidence: word choice, tense, grammar, and ambiguity. Example: “I used to be indecisive. Now I’m not sure.” The second sentence uses a contrast pattern that nudges you toward a self-referential frame.
• Context evidence: who is speaking, where it happens, and what the audience already knows. Example: In a workplace meeting, “Let’s circle back” strongly suggests a business frame, not a literal one.
• Genre evidence: jokes, sarcasm, riddles, and puns follow different conventions. Example: A riddle primes you to search for an answer, not for a literal event.
• Pragmatic evidence: intent signals like politeness, exaggeration, or mismatch between literal meaning and social function. Example: “Great job” said after a mistake often triggers a critical-evaluation frame.

Model selection is strongest when these evidence types point in the same direction.

A Systematic Selection Procedure

Use this sequence to analyze how a reader chooses a frame.

1. List plausible frames: Start with 2–4 candidates that could explain the setup.
2. Score fit to evidence: For each frame, ask what it predicts about the next phrase.
3. Track where predictions succeed: If the text keeps matching one frame, that frame gains momentum.
4. Watch for the violation point: In incongruity theory, the punchline often breaks the working model’s predictions.
5. Check the resolution path: The best frame switch is one that makes the punchline interpretable without rewriting everything.

This procedure mirrors how people naturally recover from a surprise: they don’t start over from scratch; they revise the most economical parts.

Example: Two Frames, One Punchline

Consider: “My therapist told me to stop living in the past. I told her I can’t. It’s too expensive.”

• Early evidence: “therapist told me” primes a self-improvement frame where the speaker follows advice.
• Working model: emotional-literal frame about mental habits.
• Punchline violation: “too expensive” breaks the emotional frame because it introduces a cost structure.
• Resolution: a figurative-to-literal frame switch. “Past” becomes a place you can live in, making the joke coherent.

Notice what makes this work: the resolution frame doesn’t require ignoring the therapist setup. It reinterprets the key noun (“past”) while keeping the rest stable.

Example: When Model Selection Fails

Now compare: “I tried to be spontaneous today. It didn’t work.”

The setup lacks strong evidence about genre or target. A reader can choose many frames—self-help, complaint, literal failure, or irony—but none is clearly favored. Without a clear working model, the punchline has less to break, so there’s no clean switch. The result is a flat feeling: the mind can’t tell which interpretation path to commit to.

Advanced Detail: Choosing the Smallest Revision

A high-quality joke often uses a “nearby” frame rather than a totally unrelated one. The smallest revision principle helps explain why some punchlines feel instantly right.

• If the working model is too far from the resolution frame, the reader must rebuild too much.
• If the working model is too close (no meaningful violation), the punchline won’t feel surprising.

Good incongruity sits in the middle: enough mismatch to trigger a switch, but not so much that the switch becomes a rewrite.

Practical Takeaway

When you’re analyzing or crafting a joke, treat model selection as an evidence-driven choice. Make sure the setup strongly favors one frame, then design the punchline so the evidence supports a different frame with minimal reconstruction. That balance is what turns “I didn’t expect that” into “Now it makes sense.”

5.3 Model Conflict: When Two Frames Compete and How Humor Resolves the Conflict

When you hear a joke, your mind doesn’t wait passively for a punchline. It runs multiple interpretation frames—small “models” of what’s going on—then tries to make the incoming details fit. Model conflict happens when two frames both seem plausible early on, but only one can fully explain the final information. Humor often comes from forcing that competition to peak, then steering the listener toward a resolution that reassigns meaning.

The Two-Frame Setup

A frame is a bundle of expectations: roles (who does what), goals (what someone is trying to achieve), norms (what counts as normal), and causal assumptions (what leads to what). In many jokes, the setup gives cues that support Frame A. The punchline then introduces a detail that also supports Frame B, but Frame B clashes with Frame A’s earlier assumptions.

A simple way to see the conflict is to track what each frame would predict. Frame A predicts a certain continuation; Frame B predicts a different one. The listener’s comprehension system keeps both alive until the punchline makes one of them untenable.

How Conflict Feels in Real Time

Model conflict is not just “confusion.” It’s a specific kind of cognitive tension: the mind can’t settle on a single coherent story. You may notice this as a brief pause, a reread, or a “wait, what?” moment.

In incongruity terms, the punchline supplies a violation that is locally interpretable under Frame B. The trick is that Frame A would require the violation to mean something else, or would treat it as an error. The listener’s system prefers the frame that yields the simplest overall coherence.

Resolution Mechanisms

Humor resolves model conflict by changing which frame is treated as primary. Three common mechanisms do this.

1. Reframing the violated detail: The punchline makes the “wrong” detail suddenly fit Frame B, while Frame A would require extra, awkward assumptions.
2. Constraint relaxation: The punchline loosens a strict expectation from Frame A (for example, changing a literal rule into a figurative one).
3. Role reassignment: The punchline swaps who the relevant agent is, or what role the speaker plays, so the same words now describe a different situation.

The listener’s final appraisal often includes a quick sense of “oh, that’s what this was about,” which is the cognitive system committing to the winning frame.

Example: A Clean Frame Competition

Consider: “I told my computer I needed a break. Now it won’t stop sending me KitKat ads.”

• Frame A (human-like help): The speaker asks for a break; the computer responds like a considerate assistant.
• Frame B (marketing automation): The computer follows its programmed advertising pipeline.

The setup supports Frame A because “I needed a break” sounds like a request that should change behavior. The punchline introduces a detail—ads—that is more naturally explained by Frame B. Frame A could still be forced to fit, but it would require extra assumptions like “the computer interpreted the break as a marketing opportunity.” Frame B wins because it explains the punchline without stretching.

The humor lands when the listener’s mind stops trying to keep Frame A alive and reassigns the situation to Frame B. The same phrase “sending me” becomes funny because it now has a different causal story.

Example: Role Reassignment Without Changing the Words

Try: “The therapist said I should stop living in the past. I told her I can’t. It keeps asking me to update.”

• Frame A (therapy advice): The therapist gives a general mental-health guideline.
• Frame B (software versioning): “The past” is literally an outdated system that demands updates.

The punchline doesn’t merely add a new fact; it changes what the earlier phrase refers to. The listener resolves conflict by switching the role of “past” from a metaphor to a technical object. Once that switch happens, the therapist’s advice becomes a setup for a literal contradiction.

Practical Diagnostic: Where the Conflict Peaks

To analyze a joke, identify the moment when two frames are both locally plausible. That’s usually right before the punchline’s key detail. Then ask what would have to be true for Frame A to remain viable. If the required add-ons feel forced, the joke is likely designed to push the listener toward Frame B.

A good rule of thumb: the punchline should not just be surprising; it should be the specific piece of information that makes the winning frame feel inevitable and the losing frame feel overworked.

5.4 Updating Strategies: Replacing, Integrating, or Hierarchically Reorganizing Assumptions

When a joke lands, the mind usually performs an update rather than a simple “aha.” The update has to reconcile the violated expectation with a coherent interpretation. Three common strategies explain how that reconciliation happens: replacing assumptions, integrating new information, or hierarchically reorganizing the model.

Replacing Assumptions

Replacing is the cleanest update: the mind discards an earlier assumption and swaps in a new one that fits the punchline.

A practical rule: replacement works best when the setup strongly commits you to one interpretation frame, and the punchline makes that frame untenable.

Example: “I told my computer I needed a break. Now it won’t stop sending me KitKat ads.”

• Setup assumption: the computer will understand “break” as rest.
• Violation: the system treats “break” as a product category.
• Replacement: the mind switches from a human-like intent model to a keyword-to-content model.

How to spot replacement in analysis: ask what single earlier belief becomes false in the punchline’s world. If one belief collapses and everything else can stay the same, you’re likely looking at replacement.

Best practice for writers: make the setup commitment specific enough that the audience forms a stable assumption, then ensure the punchline directly contradicts it.

Integrating New Information

Integration keeps the old assumptions but adds a new constraint or sub-meaning. The mind expands the model so both the setup and punchline can be true.

A practical rule: integration works best when the punchline doesn’t fully negate the setup; it refines it.

Example: “My therapist says I have a fear of commitment. I told her I’m not ready for that kind of relationship.”

• Setup assumption: the therapist’s diagnosis is about the speaker’s emotional readiness.
• Violation: the speaker reframes “commitment” as a specific type of relationship step.
• Integration: the mind keeps the fear-of-commitment idea but adds a nuance about what counts as “commitment” for this person.

How to spot integration: look for a shared core interpretation that survives the punchline, with only the boundaries or definitions shifting.

Best practice for writers: use the punchline to add a precise qualifier, not to erase the setup’s core.

Hierarchically Reorganizing Assumptions

Hierarchical reorganization changes the structure of the model, not just its contents. The mind moves from one level of explanation to another, such as from “what is happening” to “what kind of system is doing it.”

A practical rule: hierarchical reorganization appears when the punchline reframes the entire explanatory level.

Example: “The sign said ‘Bridge Out.’ So I took the scenic route… through the bridge.”

• Setup assumption: “Bridge Out” is a literal warning about a broken bridge.
• Violation: the speaker treats “bridge” as a metaphorical or linguistic object.
• Hierarchical reorganization: the mind shifts from a physical-world model to a language-as-structure model, where “bridge” can be navigated in a different sense.

This strategy often produces a stronger “surprise” because it changes what counts as relevant evidence. The audience doesn’t just update a belief; it changes the category of reasoning.

Best practice for writers: ensure the punchline provides enough cues that the audience can justify the new level of explanation, even if it takes a beat.

A Systematic Diagnostic Workflow

1. Identify the setup’s main assumption(s) that guide prediction.
2. Locate the punchline’s conflict: what exactly becomes incompatible.
3. Choose the update type by checking which parts of the model remain usable.
   • If one assumption becomes false while the rest stays stable, it’s replacement.
   • If the core meaning remains true but gains a qualifier, it’s integration.
   • If the reasoning level changes, it’s hierarchical reorganization.
4. Confirm by tracing resolution: the audience should be able to explain the punchline without redoing every earlier step.

Quick Comparative Example Set

• Replacement: “I asked the librarian for a book on paranoia. She whispered, ‘They’re watching you.’”
  • The mind swaps from normal library help to a conspiratorial framing.
• Integration: “I’m great at multitasking. I can waste time and be productive at the same time.”
  • The mind keeps “multitasking” but integrates a redefinition of productivity.
• Hierarchical Reorganization: “I tried to catch fog yesterday. Mist.”
  • The mind shifts from physical catching to wordplay about “mist.”

These strategies are not mutually exclusive in a single joke, but one usually dominates. Naming the dominant update makes analysis more precise and helps you predict why a punchline feels coherent after the initial jolt.

5.5 Practical Example Walkthrough: Analyzing a Joke by Tracing Model Changes Step by Step

We’ll analyze a short joke by tracking how a listener’s mental models change as new information arrives.

Joke: “I told my computer I needed a break. Now it won’t stop sending me KitKat ads.”

Step 1: Establish the Initial Mental Model

Start with what the audience expects from the setup. Most listeners build a model like this:

• The speaker asks for a break.
• The computer responds helpfully or at least stops working.
• The humor will likely come from a mismatch between “break” as rest and “break” as something else.

At this stage, the listener’s model predicts a resolution where the computer either misunderstands the request in a harmless way or performs an unexpected but coherent action.

Step 2: Detect the First Violation

The punchline begins with “Now it won’t stop sending me KitKat ads.” That directly violates the predicted outcome. The listener updates the model because the computer’s “response” is not a pause; it’s marketing.

This is not just a surprise event. It forces a new inference: the computer is not acting like a person who grants breaks; it’s acting like an ad-targeting system.

Step 3: Reframe the Meaning of “Break”

The listener now searches for a resolution that makes the violation coherent. The key is the double meaning of “break.”

• In the initial model, “break” means rest.
• In the revised model, “break” also points to “KitKat,” a brand name that contains “KitKat” and is associated with “break” as in taking a break.

The joke becomes coherent when the listener maps the request (“I needed a break”) onto a marketing association (“KitKat ads” as the system’s interpretation of the phrase).

Step 4: Track Model Changes as a Sequence

To make the process systematic, treat the listener’s understanding as a chain of model states.

1. State A: The computer is a conversational agent.
2. State B: The computer is an ad-targeting system.
3. State C: “Break” is interpreted as a cue for KitKat-related content.
4. State D: The computer’s “refusal” to stop is reinterpreted as relentless targeting, not refusal to rest.

Each state change reduces uncertainty. The laughter often arrives when the final state feels like the simplest explanation that fits all details.

Step 5: Explain Surprise and Why It Lands

The surprise comes from the listener’s initial assumption about agency. People expect a computer to respond to a request in a way that matches the request’s goal. Instead, the system responds in a way that matches its own objective function: showing ads.

The resolution is satisfying because it doesn’t require extra facts. The listener can generate a coherent interpretation using only the phrase “a break” and the brand cue “KitKat.”

Step 6: Apply the Same Method to a Variant

Variant: “I told my phone I needed a break. It immediately started playing break-dance videos.”

• Initial model: the phone grants rest.
• Violation: it plays content.
• Reframe: “break” as “break-dance.”
• Resolution: the phone interprets the word literally in a different domain.

This variant shows the same model-tracing pattern with a different resolution mapping.

By tracing model changes, you can explain the joke without hand-waving: the humor comes from a specific expectation, a specific violation, and a specific reinterpretation that restores coherence.

6.1 Perceptual and Linguistic Processing: Parsing the Stimulus

Before a joke can violate expectations, the mind has to read it. Perceptual and linguistic processing turns raw input—sound waves, letters, gestures—into structured representations. Those representations determine what expectations are even possible, which is why parsing quality strongly affects whether incongruity is detected.

From Signal to Units

The first job is segmentation: the brain breaks continuous input into manageable units. In speech, this means finding word boundaries despite slurring and variable speed. In text, it means recognizing characters, then grouping them into words and phrases. Segmentation is not perfect; it relies on context. For example, the sentence “I saw the man with the telescope” can be segmented so that “with the telescope” attaches to “man” or to “saw.” The later ambiguity is not a failure of parsing; it is a consequence of how the system chooses likely attachments.

A practical implication for humor is simple: if the audience cannot reliably segment the setup, the punchline has less to work with. A joke that depends on a specific grammatical attachment should keep the setup unambiguous until the intended moment.

Lexical Access and Meaning Candidates

Once units are identified, the mind retrieves word meanings. Many words have multiple senses, and the system often activates more than one candidate early. The phrase “bank account” pushes toward financial senses, while “river bank” pushes toward geography. Humor can exploit this by steering the audience toward one sense, then switching at the punchline.

Example: “I used to be afraid of elevators, but I’m fine now. I just take it one floor at a time.” The word “floor” is initially tied to rooms or levels, but the joke’s rhythm encourages a more literal, step-by-step interpretation that fits the emotional framing.

Syntactic Parsing and Structure Building

Parsing also builds structure: who did what to whom, and how phrases relate. The mind uses grammar-like expectations to predict upcoming structure. Consider: “The chef served the guests with a smile.” The prepositional phrase “with a smile” can be interpreted as an instrument or accompaniment. If the punchline later makes that phrase literal in an unexpected way, the humor lands because the audience already built a structure that the punchline can subvert.

A useful best practice is to choose a structure that is easy to compute quickly. Complex nesting can cause the audience to spend processing time on form rather than on meaning, leaving less capacity for incongruity detection.

Pragmatic Interpretation and Speaker Intent

Linguistic parsing does not stop at grammar. The mind also infers intent: whether the speaker is stating, requesting, joking, or being ironic. This inference uses cues like tone, politeness markers, and typical conversational goals.

Example: “Nice job breaking the vase.” The words “nice job” are positive, but the context of “breaking the vase” makes a literal compliment implausible. The audience’s pragmatic system flags mismatch and treats the utterance as evaluative sarcasm. Incongruity here is not only semantic; it is about the mismatch between surface positivity and intended meaning.

For humor design, this suggests a control knob: you can delay the pragmatic cue until the punchline, or you can provide it early so the audience knows to search for a non-literal resolution.

Timing: When Parsing Meets Prediction

As the audience parses, it also predicts what comes next. Prediction is tightly coupled to parsing because the brain uses partial structure to anticipate upcoming words. This is why punchlines often work best when they arrive right after a stable parse has been formed.

Example: “I told my computer I needed a break, and it said: ‘No problem—I’ll go to sleep.’” The audience predicts a human-like response to “break.” The punchline forces a re-parse of “sleep” as a literal computer action rather than a metaphor for rest.

If the setup is too long or syntactically unstable, the audience may not commit to a single parse, and the punchline becomes merely surprising rather than meaningfully incongruous.

A Worked Micro-Example

Take: “He promised he would be on time. He was… on time.”

First, “on time” is parsed as an adverbial phrase meaning punctuality. The repetition at the end encourages a different parse: “on time” as a literal location or state, not just a schedule property. The humor depends on the audience having already built a grammatical expectation for “be on time,” then encountering a punchline that invites a second, more literal structure.

In short, parsing is the foundation layer that creates the candidate meanings and structures the joke can later contradict. When parsing is stable, incongruity becomes precise; when parsing is shaky, the audience may notice oddness but miss the intended resolution.

6.2 Comprehension and Prediction: Generating Candidate Interpretations

When people read or hear a joke, they do not wait for the punchline to decide what it means. They generate candidate interpretations as the input arrives, using expectations built from language patterns, context, and prior experience. In incongruity terms, comprehension is the stage where the mind tries to make the incoming information fit one of several plausible models. The punchline works when the best-fitting model changes, stalls, or gets replaced.

Prediction as a Running Set of Hypotheses

Start with a simple idea: the mind prefers interpretations that explain the most recent input with the least extra work. Early words narrow the field. For example, hearing “I told my phone…” strongly biases you toward a story about communication, not a physics lecture. As more words arrive, the mind updates the set of hypotheses rather than committing to a single answer too early.

A useful way to think about this is “multiple drafts.” At any moment, several interpretations can be active, each with a different level of fit. The mind keeps the top candidates because they are cheap to maintain and likely to pay off. When a new phrase arrives that contradicts the current best candidate, the mind either revises that candidate or shifts attention to the next one.

How Candidate Interpretations Are Generated

Candidate generation uses three main inputs.

1. Lexical cues: word meanings and typical collocations. “Bank” can mean a financial institution or a river edge, so the next word matters.
2. Syntactic cues: how words are arranged. “The chicken is ready to eat” invites a different parse than “The chicken is ready; to eat…”
3. Context cues: who is speaking, where the scene is set, and what topic the conversation has established.

A practical example shows the mechanics. Consider: “I tried to catch fog yesterday.” The phrase “catch fog” is odd, but it is not yet a punchline. Your mind tries candidates: (a) a literal attempt at capturing fog, (b) a metaphor for trying to grasp something elusive, (c) a setup for a wordplay twist. The next sentence decides which candidate survives.

Ranking Interpretations by Fit

Not all candidates are treated equally. The mind ranks them by coherence with what has already been processed.

• Local coherence: does the interpretation make sense for the current clause?
• Global coherence: does it fit the emerging story across sentences?
• Constraint satisfaction: does it avoid contradictions with basic knowledge?

This is why some jokes feel “obvious” only after the punchline. Before the punchline, the correct interpretation may be present but ranked lower because it violates a constraint that later gets relaxed.

From Candidates to Comprehension

Comprehension happens when the mind settles on an interpretation that can be used to predict what comes next. This is not a final decision; it is a working model that guides attention.

For instance, in a classic setup like “Why don’t scientists trust atoms?” you likely predict a punchline about atoms being unreliable or small. That prediction shapes how you interpret the next words. When the punchline arrives—“Because they make up everything”—the mind performs a reinterpretation: “make up” shifts from “compose” to “fabricate,” and the model changes.

Example Walkthrough with Stepwise Candidate Shifts

Take: “The librarian told me to have a good day.”

• Early stage: You predict a polite closing. Candidate A: the librarian is being friendly.
• Next cue: “So I checked out a book on how to have a good day.” Candidate A still works, but it now predicts a normal follow-up.
• Later cue: “It was due yesterday.” Candidate B emerges: the phrase “have a good day” is being treated as a literal task with a deadline. The mind shifts from social politeness to a procedural interpretation.

Notice what changed. The joke does not merely add surprise. It forces a new working model that explains the earlier words differently.

Practical Checks for Writers and Analysts

When you analyze a joke, ask what candidates were most likely active at each point. Then ask which candidate becomes untenable and why.

• If the punchline requires a new meaning for a familiar phrase, the earlier candidate was probably ranked high due to everyday usage.
• If the punchline depends on a different parse, the earlier candidate was probably ranked high due to grammatical expectations.
• If the punchline depends on context, the earlier candidate was probably ranked high because the scene suggested a certain role for the speaker.

This is the core of comprehension in incongruity theory: prediction generates candidates, ranking chooses a working model, and the punchline changes the model by making the previous fit collapse or by relaxing the constraints that kept the correct interpretation out of the top spot.

6.3 Incongruity Detection: Identifying Where the Expectation Breaks

Incongruity detection is the moment your mind notices that “what should happen” and “what did happen” don’t match. In incongruity theory terms, you are not just seeing a mismatch; you are locating it in the incoming signal relative to the expectation baseline you built earlier. This section explains how that location is found, how to describe it precisely, and how to avoid common analysis mistakes.

The Expectation Baseline You Compare Against

Your mind forms a baseline from cues like genre, prior sentences, typical roles, and likely meanings. Detection begins when the stimulus forces a prediction update. The key is that the baseline is not a single guess; it is a set of constraints. For example, in a “knock-knock” setup, you expect a call-and-response structure. If the speaker skips the response, the break is structural, not just semantic.

A practical way to track the baseline is to name three things before the punchline: the expected category (what kind of thing this is), the expected role (who does what), and the expected consequence (what follows if the role behaves normally). Detection then asks: which of these three constraints fails first?

Where the Break Can Live

Expectation breaks tend to cluster in a few locations. Detecting the location matters because each location suggests a different resolution strategy.

1. Meaning break: a word or phrase is interpreted in a way that conflicts with the intended category.
2. Structure break: the grammar, sequence, or framing violates the pattern you were using.
3. Role break: the speaker, listener, or object behaves unlike the role you assigned.
4. Causal break: the story implies a mechanism, but the outcome follows an incompatible mechanism.
5. Pragmatic break: the utterance’s intent conflicts with what the context supports.

A common mistake is to label everything as “meaning.” If the joke’s punchline works by changing who is responsible, then the role break is the real starting point.

A Systematic Detection Procedure

Use a short, repeatable checklist.

Step 1: Mark the first cue that commits you. Identify the earliest sentence or phrase that forces a baseline. In many jokes, the setup contains a “commitment line” that narrows interpretation.

Step 2: Predict the next move using the baseline. Write a plain-language prediction: “After this, the speaker will likely…” Keep it concrete.

Step 3: Identify the earliest mismatch. Scan forward for the first element that makes your prediction fail. The punchline may be late; the break may happen earlier.

Step 4: Classify the break by constraint type. Choose one primary type from the list above. If multiple constraints fail at once, pick the one that creates the strongest processing pause.

Step 5: Note the evidence that would have prevented the break. This is how you confirm you located the right spot. If a small change to the setup would remove the mismatch, you likely found the true trigger.

Example: Detecting the Break Before the Punchline

Consider: “I told my computer I needed a break, and it said, ‘No problem—I’ll go to sleep.’”

• Commitment cue: “I told my computer I needed a break.” This commits you to a human-like conversation where the computer responds helpfully.
• Prediction: “It will acknowledge the request and pause my work.”
• Earliest mismatch: “it said, ‘No problem’” is fine, but the mismatch arrives at “I’ll go to sleep.” Your baseline expects a break for you, not for the computer.
• Break type: role and consequence. The role of “who takes the break” flips.
• Validation: If the setup had said “I needed a break from the computer,” the mismatch would weaken, confirming the trigger.

Notice that the punchline is also the mismatch point. In many jokes, though, the break happens earlier and the punchline merely reveals it.

Example: Structural Break in a Short Exchange

“I asked for a map. The clerk handed me a compass and said, ‘Good luck.’”

• Commitment cue: “asked for a map” commits to a specific artifact.
• Prediction: “I will receive a map.”
• Earliest mismatch: “handed me a compass.” That is a category mismatch.
• Break type: meaning break with a pragmatic layer. “Good luck” signals refusal or indifference rather than service.
• Why it matters: If you only call it “meaning,” you miss that the clerk’s intent is part of the detection.

Advanced Detail: Handling Multiple Breaks

When more than one constraint fails, detection should still pick an order. Ask which failure forces the largest reinterpretation cost. If the first mismatch is small but the later one forces a full frame change, the later one is the primary break even if the earlier one is technically incorrect.

A useful test is to try two reconstructions: one that keeps the original baseline and one that updates it. The break location is where the “keep baseline” reconstruction stops being coherent.

By the end of detection, you should be able to point to a specific phrase or beat, name the constraint type that fails there, and explain what prediction was active right before that failure. That precision sets up the next step: resolution, where the mind repairs coherence by choosing a new interpretation.

6.4 Resolution and Appraisal: Evaluating the Reframed Meaning

Resolution is the moment the mind stops treating the stimulus as a puzzle with no exit. Appraisal is the evaluation step that decides whether the reframed meaning is coherent enough to accept and whether it is pleasant enough to count as funny rather than merely confusing.

From Incongruity to Reframe

A reframed meaning is not just a new interpretation; it is an interpretation that explains the earlier mismatch without breaking too many other expectations. The mind typically does this by relaxing constraints that were too strict.

Consider a simple setup: “I told my computer I needed more space, and it said…” The audience expects a direct technical response. If the punchline is “it needed more space too,” the mind can relax the expectation that “space” must be only storage. The reframed meaning treats “space” as both physical memory and conversational room.

What “Coherent Enough” Means

Coherence is the practical test: does the new frame connect the setup and punchline with minimal extra assumptions? Coherence is not perfection; it is a cost-benefit judgment.

A useful way to think about it is as three checks:

1. Coverage: Does the frame explain the key words and actions?
2. Consistency: Does it avoid contradicting the most salient parts of the stimulus?
3. Economy: Does it require fewer new assumptions than the competing frames?

If a punchline forces you to reinterpret the setup in a way that makes the setup feel arbitrary, coherence drops. The audience may still “get it,” but the appraisal may land closer to annoyance than amusement.

Appraisal: Why Some Resolutions Feel Funny

Once a reframed meaning is coherent, appraisal asks whether the resolution is safe and satisfying.

1. Positive appraisal signals

• The reframed meaning restores predictability: after the punchline, the earlier mismatch looks like a deliberate setup for a different category.
• The resolution feels earned: the punchline uses information already present, not hidden context.
• The emotional tone stays compatible with the social situation. A joke that violates politeness norms can reduce amusement even when the cognitive resolution works.

2. Negative appraisal signals

• The resolution is too late or too costly. If you need to work hard to justify the punchline, the mind may label it as “not worth it.”
• The reframed meaning makes the speaker look incompetent or cruel in a way that the audience cannot comfortably ignore.
• The stimulus remains ambiguous after the punchline. If multiple frames fit equally well, the mind may keep searching, delaying appraisal.

A Systematic Evaluation Procedure

Use this step-by-step method to evaluate any joke’s resolution.

1. Identify the candidate frames the audience could adopt at the punchline.
2. Test coverage: underline the elements the frame must explain.
3. Test consistency: note the parts that would become contradictions.
4. Compare economy: count how many extra assumptions the frame requires.
5. Check appraisal constraints: consider social tone, emotional safety, and whether the resolution feels satisfying rather than merely technical.

This procedure mirrors how people often explain jokes to each other: they don’t just say “it means X,” they justify why X is the best fit.

Example: Two Resolutions, Different Appraisals

Take two punchlines for the same setup: “I tried to catch fog yesterday.”

• Punchline A: “I mist.”

  • Resolution: “mist” shifts from weather to a verb meaning to fail.
  • Coverage: explains “catch” and “fog.”
  • Economy: relies on a common wordplay mapping.
  • Appraisal: the reframed meaning is quick and socially harmless, so it tends to feel funny.

• Punchline B: “It was a complicated physics problem.”

  • Resolution: tries to replace the joke’s wordplay frame with an explanatory frame.
  • Coverage: only partially connects to “catch fog.”
  • Economy: adds assumptions about the audience’s tolerance for technical detours.
  • Appraisal: even if the audience can rationalize it, the resolution feels like a mismatch in genre, so amusement often drops.

Example: When Resolution Works but Appraisal Fails

A joke can be cognitively resolved yet not funny if the appraisal constraints fail. For instance, if the reframed meaning depends on humiliating a target in a way the audience cannot comfortably endorse, the mind may still compute the interpretation but appraise it as mean. In that case, resolution succeeds and appraisal rejects.

Resolution and appraisal are therefore coupled but not identical. Resolution supplies the “how it fits,” while appraisal supplies the “so what.”

6.5 Practical Example Walkthrough: A Stage-by-Stage Breakdown of a Classic One-Liner

Consider the classic one-liner:

“I told my wife she was drawing her eyebrows too high. She looked surprised.”

This joke is short, but it still runs through the same stages as longer humor: perception, prediction, incongruity detection, resolution, and appraisal. The key is that the punchline doesn’t merely add a new fact; it changes which mental model best explains the same observed information.

Stage 1: Perception and Linguistic Parsing

The listener hears a first-person narrator, a target (“my wife”), and an action (“told”). The phrase “she was drawing her eyebrows too high” sets up a visual domain: eyebrows, height, and a corrective judgment. Parsing is straightforward because the grammar is conventional and the nouns are concrete.

Practice: When you write a one-liner, keep the first clause syntactically ordinary. If the listener must fight the sentence structure, they spend cognitive effort on parsing instead of humor.

Stage 2: Initial Prediction and Mental Model Selection

Before the punchline, the listener predicts a resolution path consistent with the setup. A common expectation is:

• The narrator gives feedback.
• The wife responds verbally or emotionally to the criticism.
• The joke will likely involve embarrassment, disagreement, or a misunderstanding.

Mental model A (the “conversation model”) treats the story as a social exchange.

Practice: Use a setup that strongly cues a single model. Here, “I told my wife” primes the conversation model more than any other.

Stage 3: Incongruity Detection

The punchline arrives: “She looked surprised.” The word “surprised” is compatible with the conversation model, but the listener also notices a mismatch in the causal chain. If the wife is drawing her eyebrows too high, why would she look surprised rather than adjust them or comment?

At this moment, the listener’s system flags a small tension: the emotional outcome is plausible, but the mechanism is unclear.

Practice: Incongruity doesn’t have to be huge. A one-liner often uses a mild violation: the expected cause is missing or replaced.

Stage 4: Resolution Through Reframing

Now the listener revises the mental model. The eyebrows are not just a topic; they are a visible cue that can create the expression “surprised.” The wife’s “surprise” is inferred from her facial appearance, not from a reaction to being criticized.

Mental model B (the “visual expression model”) explains the same words differently:

• “Drawing her eyebrows too high” changes her facial expression.
• That expression looks like surprise.
• Therefore, “she looked surprised” is a description of appearance.

Resolution happens because the listener can map “too high eyebrows” to a recognizable expression pattern.

Practice: Aim for a resolution that reuses earlier information rather than discarding it. The joke works because the setup’s eyebrow detail becomes the punchline’s explanation.

Stage 5: Appraisal and Coherence Restoration

The listener evaluates coherence: the story now fits together without extra assumptions. The narrator’s “told” still matters, but it shifts from being the cause of an emotional response to being the trigger for a visual state.

Appraisal also includes effort cost. The listener doesn’t need to imagine complex events; they only need to reinterpret “looked surprised” as an appearance-based inference.

Practice: Keep the resolution within the same scene. If the punchline requires a new location, new characters, or a long backstory, the coherence restoration becomes expensive.

Stage 6: Timing and Processing Load

Even in a single sentence, timing matters. The phrase “She looked surprised” arrives after the eyebrow detail, so the listener can immediately test the visual-expression interpretation. If the punchline were delayed or replaced with an unrelated action, the reinterpretation would lose traction.

Practice: Place the punchline so the listener’s attention naturally returns to the most recent concrete detail.

A Quick Diagnostic Checklist You Can Apply

• Does the setup strongly cue one mental model?
• Does the punchline force a reinterpretation of an earlier detail?
• Is the resolution based on the same scene, not a new one?
• Can a listener explain the punchline in one sentence after the fact?

If you can answer “yes” to these, you’ve built a one-liner that processes cleanly through the same stages—setup to prediction, violation to detection, and resolution to coherence.

7.1 Operationalizing Incongruity: What Can Be Measured and How

Operationalizing incongruity means turning an abstract idea—“expectation violation”—into observable signals. The goal is not to measure “funny” directly, but to measure the ingredients that typically produce humor: prediction, mismatch, and the path to resolution.

What Incongruity Looks Like in Data

Incongruity is usually treated as a mismatch between what a person expects and what they encounter. To measure it, you first define the expectation baseline and then define the violation relative to that baseline.

A practical baseline can be built from:

• Linguistic norms: typical word meanings, grammar patterns, and collocations.
• Contextual norms: what the setting and prior lines suggest.
• Genre conventions: what a joke format typically promises.
• Task-specific expectations: what participants are told to look for.

Then the violation can be defined as one of these measurable events:

• Interpretation shift: the participant initially commits to one meaning and later revises it.
• Prediction error: the stimulus is harder to process than expected (slower reading, more errors).
• Resolution requirement: participants can only make sense after a reframing step.

Core Measurement Targets

You can operationalize incongruity using three main targets that map cleanly onto cognitive processing.

1. Mismatch detection

   • Observable via reaction time spikes, increased reading time, or higher error rates at the violating segment.
   • Example: In a sentence like “He mailed the letter to the moon,” readers often slow at “moon” because it breaks the expected destination pattern.

2. Resolution success

   • Observable via comprehension accuracy, reduced confusion ratings after the punchline, or improved coherence judgments.
   • Example: If the next line clarifies that “moon” is a nickname for a coworker, participants’ coherence ratings rise after the clarification.

3. Incongruity magnitude

   • Observable via graded difficulty: stronger violations typically produce larger processing costs or larger shifts in interpretation.
   • Example: “He mailed the letter to the moon” is a mild category error; “He mailed the letter to the color of the moon” tends to be harder because it adds an extra layer of abstraction.

A Simple Operational Framework

Use a three-part coding scheme: Baseline, Violation, Resolution. Each part should have explicit criteria so different coders agree.

• Baseline: what interpretation the audience would likely generate.
• Violation: what element breaks that interpretation (semantic, syntactic, pragmatic, causal).
• Resolution: what reinterpretation restores coherence.

Example coding for a short joke:

• Baseline: “A doctor treats patients.”
• Violation: “The doctor treated the appointment.”
• Resolution: “The appointment is metaphorically treated as if it were a patient (rescheduling, fixing, or ‘healing’ it).”

Behavioral Measures That Work Well

Reaction time and reading-time measures are often the most direct way to capture mismatch detection. The key is to align the measurement window with the violating element.

• Segmented timing: define the “violation region” (e.g., the word or phrase that breaks expectations).
• Compare against matched controls: use non-violating versions that keep length and structure similar.
• Track regressions: extra rereading suggests the mind is searching for a workable interpretation.

Example design logic:

• Condition A (violation): “She filed the report under yesterday.”
• Condition B (control): “She filed the report under today.”
• If A produces longer reading times at “yesterday,” that supports the idea that the violation creates processing cost.

Self-Report Measures with Clear Interpretation

Self-report can complement behavior, but it needs careful wording.

• Surprise/confusion ratings should be collected immediately after the violating segment.
• Clarity/coherence ratings should be collected after the resolution.

Example prompts:

• After the violation: “How unexpected was that line?”
• After the resolution: “How well did the story make sense by the end?”

Reliability and Validity Checks

Operationalization fails when definitions are vague. Use a rubric and test it.

• Inter-rater agreement: coders should consistently identify baseline, violation, and resolution.
• Manipulation checks: confirm participants actually experienced the intended mismatch.
• Control for confounds: keep word frequency, sentence length, and grammatical complexity as similar as possible.

Putting It Together: A Measurable Incongruity Score

A straightforward derived metric is the difference between processing cost at the violation and improvement after resolution.

• Incongruity cost: RT increase or confusion rating increase at the violating segment.
• Resolution benefit: accuracy or coherence improvement after the punchline.

When both move in the expected directions, you have evidence that the measured effect reflects incongruity processing rather than unrelated difficulty.

7.2 Experimental Approaches: Controlled Stimuli, Timing, and Task Design

Incongruity theory predicts that humor comprehension depends on how expectations are formed, violated, and resolved. Experiments test those steps by controlling the stimulus and the participant’s processing opportunities, then measuring behavior and timing. The goal is not to “catch laughter,” but to separate comprehension stages: detection of mismatch, generation of candidate interpretations, and selection of a coherent resolution.

Controlled Stimuli

Controlled stimuli start with a baseline condition that is easy to interpret and a violation condition that changes exactly one factor. A good baseline is not “no joke”; it is a version where the same words and structure lead to a stable interpretation.

A practical approach is to use minimal pairs. Keep the setup identical and change only the critical element that creates the incongruity. For example, if the target manipulation is semantic category error, then all other cues—sentence length, syntactic form, and plausibility of the setup—should remain constant.

To avoid accidental confounds, pretest the materials for comprehension and perceived similarity. If participants already disagree about what the setup means, the later effects might reflect misunderstanding rather than humor processing.

Example:

• Baseline: “The chef served the soup in a bowl.”
• Incongruity: “The chef served the soup in a calendar.”

Both sentences are grammatical and vivid, but only the second violates the expected container category. If you also changed the verb or added extra context, you would blur whether the effect comes from category mismatch or from something else.

Timing and Presentation Control

Timing manipulations test whether the mind resolves incongruity immediately or after additional evidence. Two common strategies are self-paced reading and fixed-timing presentation.

In fixed-timing designs, you can present the setup, then pause before the punchline. The pause length matters because it changes whether participants commit to an interpretation before the violation arrives. In self-paced reading, you can examine where processing difficulty spikes, such as at the word that first signals the mismatch.

A simple timing logic is to align the measurement window with the cognitive event you care about. If you want to measure detection, focus on the earliest point where the mismatch becomes identifiable. If you want to measure resolution, measure after the punchline or after a short continuation that supplies the reframe.

Example:

• Condition A: punchline appears immediately after setup.
• Condition B: a brief delay occurs, then punchline appears.

If resolution requires additional integration, Condition B may show different reaction-time patterns or different comprehension accuracy.

Task Design

Task choice determines what participants do with the stimulus. Humor tasks that only ask for “funny or not” often mix multiple processes: detection, interpretation, and social appraisal. Better tasks map onto specific stages.

A stage-aligned set of tasks might include:

1. Comprehension verification to ensure participants build a coherent interpretation.
2. Interpretation selection where participants choose among candidate meanings, including the intended resolution.
3. Mismatch localization where participants indicate where the expectation broke.
4. Speeded judgments that capture processing time without requiring full explanation.

These tasks can be combined, but you should avoid overloading participants with too many demands. If the task requires explanation, it can slow responses and change what “resolution” means.

Example: Participants read a short story with a single incongruity. Then they answer:

• “Where did the story become strange?” (setup, critical word, or punchline)
• “What best explains the ending?” (three options)
• “How coherent does the ending feel?” (1–7 scale)

The first question targets detection, the second targets resolution selection, and the third targets coherence appraisal.

A Cohesive Example Workflow

Start with a set of joke-like sentences that share the same setup structure. Create baseline and incongruity versions using minimal pairs. Pretest to confirm that the baseline is consistently understood and that the incongruity is noticed at the intended location.

Next, run two timing conditions: immediate punchline versus delayed punchline. Use a task that separates detection from resolution: first ask where the mismatch occurs, then ask which interpretation best explains the ending, and finally collect a coherence rating. If the delayed condition increases mismatch localization accuracy but reduces resolution selection speed, that pattern suggests participants detect the violation but need more time to commit to the reframe.

This workflow keeps the stimulus controlled, the timing informative, and the task aligned with the cognitive steps implied by incongruity theory.

7.3 Behavioral Measures: Ratings, Reaction Times, and Comprehension Accuracy

Behavioral measures translate “how the mind processes humor” into observable signals. In incongruity theory terms, you want evidence for three things: (1) people notice a mismatch, (2) they spend time resolving it, and (3) they end up with a coherent interpretation.

Ratings: What People Think They Saw

Ratings capture subjective appraisal, which is useful when you care about amusement, perceived surprise, or perceived coherence. A practical approach is to collect multiple ratings rather than one global score.

Common rating dimensions include:

• Amusement: “How funny was that?”
• Surprise: “How unexpected was the punchline?”
• Comprehension: “How easy was it to understand?”
• Coherence: “How well did the ending make sense?”

Easy-to-understand example: show two short jokes.

• Joke A: clear setup, mild violation, quick resolution.
• Joke B: setup is similar, but the punchline forces a reframe that takes effort. If Joke B produces higher surprise ratings but lower comprehension ratings, you likely have a stronger violation that is not yet fully resolved.

Best practice: anchor ratings with brief definitions and use consistent scales (for example, 1–7). Also collect comprehension confidence, because a person can rate “funny” while still being unsure why.

Reaction Times: How Long Resolution Takes

Reaction times (RTs) are especially informative because they reflect processing effort. In humor tasks, RTs can be measured at multiple points.

Two common RT designs:

• Self-paced reading or listening: measure time to read the punchline or the final clause.
• Response RT: measure how quickly participants answer a question after the punchline.

Easy-to-understand example: after each joke, ask “Which interpretation best fits?” with multiple choices. If participants take longer on the items where the incongruity requires a model switch, RTs should increase for those items.

Interpreting RTs systematically:

• Faster RTs often indicate the audience can map the punchline to an existing mental model.
• Slower RTs often indicate extra steps: detecting the mismatch, inhibiting the first interpretation, and selecting a resolution.

Best practice: separate “detection” from “resolution” when possible. If you only measure RT to the punchline, you may miss cases where the violation is noticed immediately but resolution happens later.

Comprehension Accuracy: Whether the Resolution Landed

Comprehension accuracy tests whether participants actually reached the intended interpretation. This is the behavioral counterpart to “constraint relaxation” in cognitive terms: did the mind settle on a coherent story?

A simple accuracy method is a forced-choice question that targets the resolution.

Easy-to-understand example:

• Joke: “I told my computer I needed a break. Now it won’t stop sending me Kit-Kats.”
• After the punchline, ask: “What is the joke’s resolution?”
  • Option 1: The computer is literally eating candy.
  • Option 2: The computer is interpreting “break” as a request for a specific kind of snack.
  • Option 3: The speaker is joking about workplace policies.

If accuracy is high for Option 2, you have evidence that participants used the intended reframe.

Best practice: design distractors that correspond to plausible but wrong mental models. If distractors are obviously wrong, accuracy becomes a test of guessing rather than processing.

Putting Measures Together: A Coherent Reading of Results

A single measure can mislead. Ratings might reflect social expectations (“I’m supposed to laugh”), RTs might reflect confusion unrelated to the intended resolution, and accuracy might reflect test-taking skill.

Integrated interpretation example:

• Condition 1 (easy resolution): higher comprehension accuracy, moderate RTs, high coherence ratings.
• Condition 2 (hard resolution): higher surprise ratings, longer RTs, lower accuracy, and lower coherence ratings. This pattern supports the idea that the incongruity is stronger and that resolution requires more cognitive work.

Best practice: report all three measures together and analyze them at the item level. Humor is variable across stimuli, so averaging can hide the jokes that truly drive the effect.

7.4 Physiological and Neural Measures: What They Indicate About Processing Stages

Physiological and neural measures help test the “stage” idea behind incongruity processing: first the brain builds expectations, then it detects a mismatch, and finally it resolves the mismatch into a coherent interpretation. The key is to treat each measure as a timing clue, not a direct readout of “humor” as a single thing.

What Counts as a Stage Signal

Neural and physiological signals are usually interpreted in terms of when they rise relative to stimulus events. A stage-relevant signal should (1) appear at a consistent time window across trials, (2) change systematically with expectation violation or resolution difficulty, and (3) align with behavioral outcomes like comprehension accuracy or reaction time.

A practical way to organize evidence is to map measures onto the processing pipeline:

• Early sensory and parsing: how quickly the stimulus is encoded and segmented.
• Prediction and mismatch detection: how strongly the system flags an expectation violation.
• Resolution and reappraisal: how effortful it is to reframe the meaning.
• Appraisal and response readiness: how interpretation supports the decision to laugh, rate, or judge.

EEG and ERP Measures for Timing

Event-related potentials (ERPs) are especially useful because they provide millisecond-level timing. Different components are often linked to different operations, though exact interpretations depend on task design.

• N1 and P2 windows often track early perceptual processing and attention allocation. If a joke’s setup is visually or linguistically harder, these components can shift even before incongruity is encountered.
• Mismatch-related responses are commonly examined in windows where the brain reacts to violations. In humor tasks, stronger mismatch signatures are expected when the punchline violates a well-established expectation.
• Late positivity and related late components are frequently associated with integration and reanalysis. These should increase when resolution requires more reinterpretation, such as when the punchline forces a new mental model rather than a simple rewording.

Example: Present two versions of a one-liner. In the first, the punchline flips a single word meaning (“bank” as money vs. river edge). In the second, it changes the implied role structure (“the doctor” is actually a mechanic). The second version typically produces longer integration demands, so late ERP effects should be larger and behavioral comprehension may lag.

fMRI for Spatial Localization

Functional MRI measures blood-oxygen changes, which are slower than EEG but useful for identifying brain regions involved in interpretation and control. In incongruity research, you often see involvement of:

• Language and semantic networks for meaning construction.
• Control and integration regions when the system must resolve conflict between competing interpretations.
• Valuation and appraisal-related areas when the task requires an explicit judgment of “funny” or when laughter is measured.

Example: If you compare jokes that resolve quickly versus jokes that remain ambiguous, the “harder to resolve” condition often shows greater engagement of regions tied to selection among interpretations and maintaining multiple possibilities.

Pupillometry and Heart-Rate Measures for Effort

Pupil dilation and heart-rate variability provide indirect but practical measures of cognitive effort and arousal. They are not specific to humor, but they can track how much work resolution requires.

• Pupil dilation tends to increase with processing load. A punchline that forces a new model should produce larger dilation than one that merely completes an already prepared frame.
• Heart-rate changes can reflect arousal and task engagement. If a violation is surprising but quickly resolved, arousal may spike briefly and then settle.

Example: In a joke where the setup strongly predicts a literal outcome, the punchline that forces a pragmatic reinterpretation (e.g., “I’m on a diet” said while eating something clearly not a diet food) should elicit more effort signals than a punchline that stays within the same pragmatic frame.

A Mind Map of Measure-to-Stage Links

Designing Evidence That Actually Separates Stages

To avoid confusing “heavier processing” with “humor,” tasks should separate the moment of violation from the moment of resolution. A clean design uses time-locked events:

1. Setup presentation: measure baseline prediction building.
2. Punchline onset: measure mismatch detection.
3. Resolution window: measure integration and reframe selection.
4. Response window: measure appraisal and decision.

Example: Use self-paced reading or controlled timing so the punchline appears at a known time. Then compare jokes where the punchline contains the violation but resolution is delayed by an additional clause. If late ERP and pupil effects increase mainly during the delayed resolution clause, that supports the stage separation rather than attributing everything to early surprise.

Interpreting Results Without Overclaiming

A single measure rarely proves a specific cognitive operation. Strong conclusions come from converging patterns: EEG timing that tracks mismatch and integration, fMRI that localizes semantic and control demands, and physiological effort measures that scale with resolution difficulty. When these signals move together in the expected order—early parsing, then mismatch, then integration—you get a coherent picture of how expectation violations become humorous experiences.

7.5 Practical Example Walkthrough: Designing a Simple Study to Test Incongruity Resolution

A good study tests whether people can move from “I’m confused” to “Oh, that makes sense,” and whether that move depends on the kind of incongruity you used. The simplest design uses short stimuli, a comprehension task, and a resolution measure.

Core Idea and Testable Claims

Start with two claims you can actually measure:

1. Resolution improves comprehension. When the punchline provides a clear reframe, participants should answer comprehension questions more accurately.
2. Resolution depends on the violation type and placement. If you move the same violation earlier or change the violation so it can’t be coherently resolved, accuracy and/or resolution ratings should drop.

To keep the study small, use one joke family and vary only one factor at a time.

Materials: Stimuli Set and Conditions

Create 12–24 items total. Each item is a two-part stimulus: setup followed by punchline. Use three conditions:

• Resolved Incongruity: The punchline supplies the reframe.
• Unresolved Incongruity: The punchline keeps the mismatch but removes the coherence cue.
• Control Coherence: The story stays consistent and never creates a strong expectation violation.

Example item (same setup, different punchlines):

• Setup: “I tried to fix my printer by talking to it politely.”
• Resolved Incongruity punchline: “It worked—after I apologized for the paper jam’s ‘feelings.’”
• Unresolved Incongruity punchline: “It worked—after I apologized for the paper jam’s ‘feelings,’ but nothing changed.”
• Control Coherence punchline: “It worked—after I apologized and reloaded the paper correctly.”

The resolved version gives a coherent reframe (personification as a playful explanation). The unresolved version blocks coherence by adding a contradiction that prevents the reframe from restoring meaning.

Participants and Procedure

Recruit a modest sample (for example, 60–120 participants) and assign items across conditions so each person sees a mix. A clean procedure:

1. Read setup.
2. Rate expected outcome on a 1–7 scale.
3. Read punchline.
4. Rate surprise on a 1–7 scale.
5. Answer a comprehension question (multiple choice) that requires the resolution.
6. Rate resolution clarity on a 1–7 scale: “How clearly did the punchline explain what was going on?”

Keep the comprehension question tightly tied to the reframe. For the printer example, the correct option might be “The joke treats the jam as if it has feelings, as a playful explanation.”

Measures and Analysis Plan

Use three dependent variables:

• Comprehension accuracy (correct vs incorrect).
• Resolution clarity (1–7).
• Surprise (1–7), treated as a check that incongruity is actually present.

A simple analysis:

• Compare Resolved vs Unresolved on accuracy and resolution clarity.
• Compare Control vs both to confirm that coherence reduces the need for resolution.
• Check whether surprise is higher in both incongruity conditions than in control; if not, your stimuli may not be creating the intended expectation violation.

If you want one extra step, include a short open response: “In one sentence, explain the punchline.” Then code whether participants mention the reframe (personification, category shift, causal reinterpretation, etc.).

Example Walkthrough: One Item Through the Whole Study

Take the printer item.

• Expectation rating: Many participants will predict a practical fix (reloading paper, cleaning heads).
• Surprise rating: The resolved punchline should feel surprising because it shifts from mechanics to playful personification.
• Comprehension question: The correct answer should require recognizing the reframe: the apology is a humorous “fix” because the jam is treated as having feelings.
• Resolution clarity: Participants should rate the resolved version higher because the punchline provides a coherent interpretation path.

Now compare the unresolved punchline. It still contains personification, so surprise may remain high, but the added contradiction (“nothing changed”) blocks the reframe from restoring meaning. That should lower resolution clarity and comprehension accuracy.

Practical Best Practices Embedded in the Design

• Use the same setup across conditions to isolate the effect of resolution cues.
• Make comprehension questions reframe-dependent, not surface-dependent. If the question only asks what word appeared, you measure recall, not resolution.
• Avoid multiple violations per item. One strong mismatch is easier to interpret than several competing ones.
• Pilot with a small group. If many participants report “I don’t get it” for the resolved condition, your punchline may not actually provide a usable reframe.

This design gives you a direct test of whether incongruity becomes funny (or at least understandable) when the mind can rebuild coherence from the violation.

8.1 Contextual Constraints: Setting, Relationships, and Shared Background

Incongruity theory treats humor as a mismatch between what a listener expects and what actually happens. Context determines what “expected” means in the first place. Without context, the same violation can feel random; with context, it feels targeted, interpretable, and therefore funny.

Setting as the Expectation Baseline

Setting includes the physical scene, the activity type, and the immediate conversational goal. These cues tell the audience which mental model to activate.

A simple example: imagine two lines.

• “Nice job, Captain.”
• “Nice job, Captain… on the wrong ship.”

In a nautical setting, “Captain” primes a role-based model with clear authority and competence. The second line violates that model by shifting the competence domain. In a school setting, “Captain” might prime a sports team model instead, and the same punchline can land as a different kind of mismatch.

Good practice is to make the setting cues do real work early. A joke that begins with “At the airport…” or “During the team meeting…” gives the audience a stable baseline so the later violation has something specific to break.

Relationships as the Social Model

Relationships shape expectations about politeness, power, and shared tolerance for bluntness. Humor often depends on how far the speaker can bend those norms without triggering rejection.

Consider this contrast:

• A close friend says, “You’re late again. I’m starting to think you’re allergic to time.”
• A stranger says the same line to you in a queue.

The words are similar, but the relationship changes the appraisal. With a friend, the audience predicts playful teasing and interprets the violation as affectionate. With a stranger, the audience predicts social risk and interprets the violation as disrespect. Incongruity may still occur, but resolution fails because the social model cannot be safely repaired.

A practical rule: match the “violation style” to the relationship. If the joke relies on mild disrespect, it needs a relationship where mild disrespect is already licensed.

Shared Background as the Common Ground Budget

Shared background is the set of facts, experiences, and conventions that the audience can reasonably assume. It includes language norms, local references, and everyday routines.

Shared background works like a budget. The more the joke depends on niche knowledge, the more the audience must spend to resolve the incongruity. If the budget runs out, the audience may still notice the mismatch but cannot build a coherent resolution.

Example: a joke about “paperwork” can be universal if it uses everyday bureaucracy cues. A joke about a specific internal form name can be funny only for people who share that reference. The second version can still be written well, but it must provide enough context to reconstruct the missing model.

A best practice is to choose references that are either widely shared or quickly made legible through the setup. If you use a specialized detail, the setup should signal what it is doing in the mental model.

How Context Controls Resolution Paths

Context doesn’t just create expectations; it also constrains which resolutions are plausible. When the audience hears a violation, they search for a reframe that restores coherence. Setting, relationships, and shared background determine which reframes the audience will consider.

For instance, a line like “I followed the instructions exactly” can resolve as:

• Competent compliance in a workplace context.
• Absurd literalism in a fantasy context.
• Suspicious excuse-making in a conflict context.

The audience’s search space changes because the context changes the likely intent model. That’s why the same punchline can be funny in one room and flat in another.

Integrated Example: One Joke, Three Contexts

Take a punchline: “I brought the wrong thing, but I brought it confidently.”

• In a cooking class, “wrong thing” can resolve as harmless mistake plus competence theater.
• In a medical appointment, the same line can resolve as negligence, which is harder to appraise as safe humor.
• With a friend helping you move, it resolves as friendly incompetence, because the relationship already supports playful framing.

The joke’s structure stays the same, but context changes the audience’s mental model, the appraisal of intent, and the ease of resolution.

Practical Takeaway

When you want incongruity to land, treat context as part of the mechanism. Setting provides the baseline, relationships regulate the social risk, and shared background supplies the materials needed to rebuild coherence after the violation.

8.2 Cultural and Linguistic Conventions: How They Shape Expectation Baselines

Humor depends on a baseline: what an audience expects to be normal, relevant, and interpretable. Cultural and linguistic conventions supply much of that baseline, often without anyone noticing. When a joke violates an expectation that the audience never formed, the violation can feel random rather than funny. When it violates an expectation that the audience strongly holds, the surprise becomes legible.

Conventions as Default Assumptions

Conventions include shared norms about politeness, roles, taboo topics, and what counts as a reasonable explanation. They also include linguistic habits such as typical word order, common idioms, and how speakers signal intent. In incongruity terms, conventions determine which interpretation path is initially selected.

A simple example is the difference between direct and indirect requests. In many English-speaking contexts, “Can you pass the salt?” is usually interpreted as a request, not a literal question about ability. A joke that treats it literally—“I can, but I’m not sure I should”—creates incongruity because the audience’s convention-driven expectation is that the speaker is being indirect.

Language Structure and Parsing Expectations

Linguistic conventions shape how people parse sentences. Word order and grammatical markers influence which meaning is computed first. If a joke relies on a specific parsing route, it can fail when the language nudges parsing differently.

Consider ambiguity in pronouns. In English, “He told her she was late” can be read as either the speaker judging her lateness or someone else judging it, depending on context. Languages with clearer marking for perspective or evidentiality reduce that ambiguity. The same surface sentence can therefore produce different expectation baselines across languages.

Idioms and Fixed Expressions

Idioms are expectation engines. People store them as chunks with predictable meanings. When a joke breaks an idiom by treating it compositionally, the audience experiences a controlled mismatch.

Example: “I’m on cloud nine” is normally understood as “I’m very happy.” A joke that responds, “Which cloud? I can’t find the Wi‑Fi,” uses the convention that “cloud nine” is figurative. The humor comes from the audience briefly trying to keep the idiom’s figurative frame, then being forced into a literal frame.

Idioms also carry cultural baggage. Some idioms are tied to specific technologies, histories, or social institutions. A literal-minded twist that works for one audience may land as confusing for another because the idiom’s figurative meaning is not shared.

Politeness Norms and Social Intent

Politeness conventions affect how people infer intent, which is central to pragmatic incongruity. Many cultures treat certain forms of disagreement as face-saving, while others treat them as straightforward.

Example: A character says, “That’s an interesting idea,” after a proposal. In some contexts, this signals mild approval; in others, it signals skepticism. A joke can exploit that by having the listener respond as if approval were intended, then revealing the mismatch.

If the audience’s convention interprets the phrase as criticism, the “reveal” becomes unnecessary. The joke’s resolution depends on the audience initially selecting the wrong intent model.

Taboo, Register, and Acceptable Targets

Cultural conventions also govern which topics are safe to violate. Humor often uses incongruity to test boundaries, but the baseline includes what the audience expects will be off-limits.

Example: A workplace joke that mocks a manager’s “leadership style” may be acceptable in one setting and risky in another. The same structural violation—turning a compliment into a critique—can shift from playful to uncomfortable depending on shared norms about hierarchy and respect.

This matters for cognitive processing: if the audience appraises the situation as socially threatening, resolution may not feel rewarding. The expectation violation still occurs, but the appraisal system may block laughter.

Integrated Example: One Joke, Two Baselines

Imagine a line: “Nice work. You really nailed it.” The humor could come from a convention where “nice work” is sincere praise. The punchline might show the “nailing” was literal—someone pinned a document to a wall instead of submitting it.

In a culture or group where “nice work” is commonly used sarcastically, the baseline already assumes irony. The audience may jump straight to the sarcastic frame, reducing the surprise. The joke still contains incongruity, but the resolution path is no longer delayed, so the comedic effect weakens.

The key takeaway is systematic: conventions determine which meanings are pre-activated, which interpretations are considered plausible, and which appraisals keep the violation in the “safe to laugh” zone.

8.3 Common Ground and Audience Design: Tailoring Humor to What Is Known

Common ground is the shared set of assumptions that lets an audience predict what comes next. In incongruity terms, it sets the expectation baseline; in cognitive terms, it reduces the cost of building and revising mental models. When you tailor humor to what is known, you’re not just choosing topics—you’re choosing which inferences the audience can safely make without extra explanation.

Start with a simple principle: the audience should be able to locate the violation quickly. If the setup requires unfamiliar background, the mind spends its effort on comprehension rather than on noticing the mismatch. A good test is to ask what the audience can infer after hearing only the first sentence. If the answer is “not much,” the punchline will likely arrive before the audience has a stable model to break.

Common ground comes in layers. First is linguistic common ground: shared meanings of words, idioms, and grammar patterns. Second is situational common ground: what typically happens in a setting, like a workplace meeting or a restaurant order. Third is social common ground: norms about politeness, status, and intent. Fourth is experiential common ground: what people have personally encountered, such as waiting on hold or learning a new device.

A practical way to design audience fit is to map the setup to the audience’s likely model. Consider three audiences and the same joke premise.

Example: “I told my phone to stop autocorrecting. Now it just ignores me.”

• For most smartphone users, the situational and experiential common ground is strong: autocorrect exists, it changes words, and “ignores me” is a person-like consequence.
• For someone who has never used autocorrect, the violation becomes harder to resolve because the baseline expectation—what autocorrect does—has gaps.

Next, decide how much ambiguity you can afford. Humor often works by letting the audience generate candidate interpretations, then selecting the one that restores coherence. If common ground is thin, multiple interpretations compete and resolution slows. That’s when jokes feel confusing rather than funny.

Example: “The meeting was so productive. We all agreed on the agenda.”

• In a typical office context, the audience expects meetings to be chaotic or at least time-consuming. The punchline violates that expectation by treating agreement on the agenda as an accomplishment.
• If the audience comes from a culture or role where meetings are routinely structured and agreement is normal, the violation loses contrast and the resolution feels flat.

Design also depends on audience goals. Some audiences want quick recognition; others tolerate slower reasoning. You can tune this by controlling how early the setup signals the relevant frame. A frame signal is a cue that narrows the mental model search space.

Example: “I’m not saying the printer is haunted, but it only works when nobody’s watching.”

• The “printer” cue anchors the situational model.
• The “works when nobody’s watching” cue points toward a familiar pattern: superstition-like behavior applied to a mundane device.
• If you remove the printer cue and start with “It only works when nobody’s watching,” the audience must invent the object, increasing processing load.

Mind maps help you check whether your setup provides enough common ground to make the violation legible.

Use the mind map as a checklist while drafting. For each layer, ask: what does the audience already know, and what will they assume by default? Then decide where the joke will violate expectations. The violation should be in the part the audience can predict, not in the part they must first learn.

Finally, remember that common ground is not only about knowledge; it’s also about comfort with the kind of mismatch being presented. Social common ground includes what feels acceptable to mock and what feels like a personal attack. If the audience shares the baseline but not the norm, the resolution may restore meaning yet still fail appraisal.

Example: A joke that relies on workplace hierarchy might land well with peers who share the norm of teasing status differences, but it can feel hostile to someone who expects strict professional boundaries.

Tailoring humor to what is known is therefore a balancing act: provide enough shared structure for the audience to build a stable model, then place the incongruity where that model can be broken cleanly and resolved quickly.

8.4 Misalignment and Failure Modes: When Incongruity Cannot Be Resolved

Incongruity theory assumes that a violation can be reinterpreted so the mind regains coherence. Sometimes that recovery fails. The audience may still notice the mismatch, but the system cannot find a stable resolution that fits the evidence. The result is not “no humor,” but a different outcome: confusion, irritation, or indifference.

What “Unresolved” Looks Like

A failure mode typically shows up as one of three experiences. First, the audience cannot generate a plausible alternative frame, so the mind keeps searching and never settles. Second, multiple frames remain possible, but none explains enough details to win. Third, a candidate resolution exists, yet it conflicts with constraints the audience treats as non-negotiable, such as grammar, physical causality, or social norms.

Failure Mode 1: No Viable Resolution Frame

This happens when the violation is too far from the audience’s expectation space. The mind can detect “something is off,” but it cannot map the stimulus to any coherent interpretation.

Example: Setup: “I set my alarm for 7 a.m., so I woke up at exactly 7.” Punchline: “Because the alarm clock is a time machine.”

The first sentence invites a straightforward causal model: alarm triggers wakefulness. The punchline introduces a new mechanism (time travel) without any bridging cues. If the audience treats time travel as outside the joke’s world, there is no resolution that preserves the initial causal model.

Practice for writers: Keep the violation within the same “world rules” you establish. If you want absurd mechanisms, seed them early with signals that the world is flexible.

Failure Mode 2: Persistent Ambiguity Without Payoff

Sometimes the mind can form several candidate frames, but the text never commits to one. The audience keeps oscillating, and the cognitive cost rises.

Example: Setup: “He said he’d be there in five minutes.” Punchline: “He meant five minutes ago.”

This can work, but it can also fail if the setup lacks a temporal anchor. If “five minutes” could refer to multiple reference points (now, then, or a previous promise), the audience may not know which frame the joke intends. The resolution becomes a guessing game rather than a satisfying reinterpretation.

Practice for writers: Add one cue that narrows the reference frame, such as a clear “from now” marker or a concrete timeline in the setup.

Failure Mode 3: Constraint Conflict

A resolution may be available in principle, but it breaks constraints the audience expects to hold. Constraints can be linguistic (syntax), logical (basic inference), physical (causality), or social (politeness, roles, intent).

Example: Setup: “The doctor said the medicine would help in an hour.” Punchline: “So I took it and waited for the medicine to arrive.”

The punchline can be read as wordplay, but it conflicts with the physical constraint that medicine is already in hand. If the audience cannot reinterpret “took it” as something else, the resolution fails.

Practice for writers: When you violate a constraint, make the violated constraint the target of the joke. Otherwise, the audience experiences the mismatch as an error rather than a deliberate shift.

Failure Mode 4: Intent Misread and Negative Appraisal

Even if the mind finds a resolution, it may appraise the violation as threatening, disrespectful, or unfair. Incongruity then lands on the wrong evaluation axis.

Example: A joke that relies on mocking someone’s disability can be “resolved” as sarcasm, but the appraisal may remain negative because the audience treats the target as deserving protection.

Practice for writers: Choose targets that the audience can safely treat as non-protected, or shift the joke toward self-incongruity (the speaker’s mistake) rather than other-directed blame.

A Systematic Diagnostic Checklist

When a joke fails, treat it like a debugging problem. Ask four questions in order.

1. Can the audience generate any coherent frame? If not, the violation is too distant.
2. Does the text narrow the frame choice? If not, ambiguity persists.
3. Does the best resolution violate a constraint the audience treats as fixed? If yes, the mind rejects it.
4. Does the resolution change appraisal in the wrong direction? If yes, the audience may not want the joke to “work.”

These questions prevent the common mistake of blaming “taste” alone. Many failures are structural: the mind either cannot find a stable reinterpretation or finds one that the audience refuses to accept.

8.5 Practical Example Walkthrough: Comparing the Same Joke with Different Context Cues

Take this single joke text:

“I told my computer I needed a break. Now it won’t stop sending me ‘coffee’ ads.”

The punchline is stable: the computer interprets “break” as a cue for “coffee.” What changes is the audience’s expectation baseline before the punchline. Context cues steer which mental model gets selected, so the same words can feel either coherent and funny or confusing and flat.

Step 1: Establish the Baseline Without Context

Start with the bare text. Many readers initially predict a literal interpretation: “break” means rest, so the computer should pause or stop working. That baseline makes the punchline surprising, because the computer does not grant rest; it redirects to marketing.

A quick way to see the baseline is to ask: “What action would I expect after the request?” Here, the expected action is a pause or a scheduling change. The joke violates that expectation by choosing a different meaning for “break.”

Step 2: Add Context Cue a and Track Model Selection

Context cue A: the speaker is a non-technical person who recently installed a “productivity” app.

Expected baseline shifts: the audience now predicts that the computer will interpret requests through app-driven automation and ad targeting. The violation still happens, but it is less like “the computer is irrational” and more like “the system is optimizing for engagement.”

Resulting resolution: the audience reframes “break” as a trigger for a marketing category, not a rest command. Incongruity becomes semantic and pragmatic at once: the word “break” is ambiguous, and the system’s goal is not the one the speaker assumes.

Step 3: Add Context Cue B and Track the Same Violation Differently

Context cue B: the speaker is a data scientist who knows the system uses keyword extraction and ad personalization.

Now the baseline includes a specific mechanism. The audience expects the computer to map “break” to related consumer categories. The surprise moves from “why would it do that?” to “why did the speaker phrase it that way?”

Resolution tightens: the joke lands as a critique of how language cues get repurposed by pipelines. The incongruity is still present, but it is more controlled, because the audience already has the right model ingredients.

Step 4: Add Context Cue C and Watch Resolution Struggle

Context cue C: the speaker is in a workplace where “break” is a formal term for a scheduled outage, and the computer is an internal system with no ad features.

Baseline now predicts a technical, domain-specific meaning. The punchline conflicts with that model: “coffee ads” should not exist in this environment. The audience may try to resolve it by assuming the system is compromised or that the speaker is exaggerating, but those resolutions cost extra effort.

This is the key difference: when context supports a plausible resolution path, the audience can restore coherence quickly. When context blocks the mechanism, the audience spends time searching for an explanation that the joke does not provide.

Step 5: Compare the Three Outcomes Systematically

• Cue A (non-technical, app context): fast resolution via pragmatic reframing of “break.”
• Cue B (technical, mechanism known): fast resolution via semantic ambiguity plus system goals.
• Cue C (domain mismatch): resolution becomes expensive, so the incongruity may feel like a mistake rather than a joke.

Step 6: Practical Rewrite Examples That Preserve the Same Core Joke

To make the comparison concrete, here are three short setup variants that keep the punchline unchanged.

Example Setup A (Non-technical App Context): “I asked my new productivity app for a break. Now it keeps serving me ‘coffee’ ads.”

Example Setup B (Technical Mechanism Context): “I told the keyword system I needed a break. Now it won’t stop sending me ‘coffee’ ads.”

Example Setup C (Domain Mismatch Context): “I told the outage scheduler I needed a break. Now it won’t stop sending me ‘coffee’ ads.”

Notice how the punchline stays the same, but the audience’s mental model selection changes. In A and B, the violation is interpretable as a reasonable consequence of the system’s goals. In C, the violation clashes with the environment, so the audience must invent an extra bridge.

That is the practical lesson: context cues don’t change the joke’s words; they change which interpretation path the audience can take with minimal effort.

9.1 Puns and Wordplay: Incongruity Through Multiple Meanings

Puns work when a single word or phrase supports more than one meaning, but the listener initially commits to the “wrong” one. In incongruity terms, the setup triggers an expectation about which meaning will be used, and the punchline forces a switch to an alternative interpretation. The humor comes from the mental cost of revising that commitment, followed by the satisfaction of coherence: the new meaning fits the sentence after the fact.

A useful way to think about puns is as a controlled ambiguity. The text provides enough cues to make one interpretation feel likely, yet it leaves a second interpretation available. If the second meaning is too obscure, the audience never reaches the resolution. If it is too obvious, the switch feels like a trick rather than a meaningful reframe.

Core Mechanism: Expectation, Ambiguity, Resolution

1. Expectation formation: The audience predicts a meaning based on syntax, common usage, and context. For example, “I used to be a banker” primes a literal identity reading.
2. Incongruity trigger: The punchline introduces a wordplay-compatible meaning that conflicts with the initial prediction. In “I used to be a banker, but I lost interest,” “interest” shifts from finance to emotion.
3. Resolution: The listener reinterprets the earlier part so the whole utterance becomes consistent. The sentence stops being a mismatch and becomes a compact two-meaning structure.

This process is systematic: puns are not just “funny because they sound similar.” They are funny because the listener’s prediction is temporarily wrong, and the language provides a clean route to correct it.

Types of Multiple Meanings

Homonym puns use identical forms with unrelated meanings. “Bark” can be a dog’s sound or a tree’s outer layer. The setup makes one domain salient; the punchline flips domains.

Polysemy puns use related senses of the same word. “Light” can mean illumination or not heavy. Because the meanings are connected, polysemy puns often feel smoother and require less cognitive re-mapping.

Morphological puns rely on how word parts change meaning. “Unclear” and “in clear” can be treated as different segmentations, especially in written wordplay.

Syntactic puns exploit how grammar changes interpretation. A phrase can be parsed as a noun phrase or a verb phrase, leading to different roles for the same words.

Each type creates a different kind of expectation baseline. Homonyms often create a sharper category shift; polysemy often creates a subtler shift in sense.

Practical Example Walkthroughs

Example 1: Homonym with a clean domain flip

• Setup: “I’m reading a book about anti-gravity.”
• Punchline: “It’s impossible to put down.”

Expectation: “anti-gravity” implies physical impossibility. Incongruity: “put down” is reinterpreted from literal placement to metaphorical stopping. Resolution: the sentence becomes consistent once “put down” is treated as an idiom.

Example 2: Polysemy with emotional vs financial sense

• Setup: “I used to be a banker.”
• Punchline: “But I lost interest.”

Expectation: “interest” points to finance. Incongruity: “lost interest” is a common emotional phrase. Resolution: the banker identity makes the financial sense plausible, while the punchline supplies the emotional sense.

Example 3: Syntactic ambiguity

• Setup: “Time flies like an arrow.”
• Punchline: “Fruit flies like a banana.”

Expectation: “flies” is read as a verb describing time. Incongruity: “fruit flies” becomes a noun phrase, changing who “flies.” Resolution: the second sentence reassigns grammatical roles and makes the parallel structure meaningful.

Advanced Details: Controlling the Cost of Reinterpretation

A pun succeeds when the listener can reach the alternative meaning without guessing wildly. That means the text should provide at least one strong cue for the initial interpretation and at least one cue that keeps the alternative meaning within reach.

Consider two failure modes. First, if the setup is too vague, the audience may never form a stable expectation, so there is no meaningful “wrongness” to correct. Second, if the alternative meaning is too far, the listener may not find a coherent reparse, so the resolution never arrives.

A practical drafting habit is to test whether the punchline can be understood in two steps: first as a normal reading, then as a wordplay reading. If the second step requires extra background knowledge or multiple leaps, the incongruity becomes noise rather than a structured surprise.

Puns are therefore a disciplined form of incongruity: they engineer a temporary mismatch between predicted meaning and realized meaning, then guide the audience to a single coherent interpretation that explains the whole sentence.

9.2 Irony and Sarcasm: Incongruity Through Pragmatic Reversal

Irony and sarcasm create humor by flipping what the speaker’s words pragmatically mean. The audience first builds an expectation about intent—what the speaker is likely trying to accomplish. Then the punchline forces a pragmatic reversal: the literal meaning is not the intended meaning. The incongruity is not just “wrong words,” but a mismatch between (1) the audience’s model of the speaker’s communicative goal and (2) the model implied by the final interpretation.

The Core Mechanism of Pragmatic Reversal

Start with a simple prediction loop. When someone says something, you infer a goal such as informing, requesting, praising, warning, or joking. Irony and sarcasm keep the surface form consistent with one goal while steering the listener toward another goal at the end.

A useful way to think about it is as a two-layer message:

• Literal layer: what the sentence would mean if taken at face value.
• Pragmatic layer: what the speaker is actually doing with the sentence.

Humor appears when the audience initially commits to the literal layer as the pragmatic layer, and then the context cues force a switch.

Expectation Setup: How the Audience Picks a Model

Pragmatic reversal works best when the setup gives enough cues for a stable initial model. Those cues can be tone, prior events, shared norms, or the speaker’s history in the conversation.

Example: “Great job cleaning the kitchen.”

If the kitchen is still messy, the audience’s first model might be “the speaker is genuinely praising.” That model is plausible because the phrase “great job” is a common praise template. The reversal happens when the context makes praise pragmatically impossible.

The Violation: When Literal Meaning Fails as Intent

The violation is the moment the audience realizes the literal meaning cannot be the intended communicative act. This is where incongruity becomes cognitive work.

Common reversal patterns include:

1. Praise for failure: the words imply approval, but the situation implies the opposite.
2. Agreement with the wrong conclusion: the speaker appears to endorse something, but the context shows they are undermining it.
3. Overly formal or overly enthusiastic wording: the style suggests sincerity, while the situation signals contempt or mockery.

Example: “Sure, I love waiting in traffic for an hour.”

The literal layer suggests enjoyment. The pragmatic layer, given the obvious inconvenience, is “I’m annoyed.” The humor comes from the audience’s quick model switch.

Resolution: Reframing the Speaker’s Goal

Resolution is not just “figuring out the opposite.” It is selecting a pragmatic goal that makes the whole utterance coherent.

A listener typically resolves irony/sarcasm by adopting one of these pragmatic goals:

• Mocking: the speaker uses the literal form to highlight the mismatch.
• Critiquing: the speaker communicates disapproval while avoiding direct confrontation.
• Signaling shared stance: the speaker invites the audience to join the evaluation.

Example: “Wow, you’re really on time.”

If the person arrived late, the coherent pragmatic goal is likely critique or teasing. The audience’s final interpretation restores coherence by aligning the utterance with the speaker’s stance.

Systematic Example Walkthrough

Consider: “That was a brilliant plan.”

1. Setup: The speaker says this right after a plan goes badly.
2. Initial model: “brilliant plan” is praise, so the audience predicts approval.
3. Violation: the outcome contradicts praise. The audience updates: praise cannot be the pragmatic goal.
4. Resolution: the audience selects a pragmatic goal such as critique or mockery.
5. Result: the audience experiences incongruity as a quick flip from sincerity to stance.

If the plan failed due to an unforeseeable external factor, the reversal may weaken. The audience might then keep a more charitable model, making the sarcasm less sharp.

Practical Best Practices for Writers and Speakers

• Make the initial model easy to form: use familiar templates like “great job,” “nice work,” or “what a genius idea.”
  • Example: “Nice driving” works better when the audience can immediately see the near-miss.
• Ensure the mismatch is visible, not hidden: irony depends on the audience noticing the conflict between words and reality.
  • Example: If the audience cannot see the mess, “Great job cleaning” becomes confusing rather than funny.
• Control the reinterpretation cost: the audience should not need a long chain of inference.
  • Example: “Sure, I’ll totally remember that” is clearer when the forgetfulness is already established.
• Avoid over-ambiguity: if multiple pragmatic goals fit equally well, the resolution stalls.
  • Example: “Interesting choice” after a mistake can be either mild sarcasm or genuine curiosity unless context narrows it.

A Quick Contrast with Straight Compliment

Straight praise aligns literal and pragmatic layers. Irony and sarcasm deliberately separate them. The humor is the moment the audience realizes the speaker is not describing reality; they are judging it.

That is why pragmatic reversal is a cognitive event: it turns interpretation into a small, satisfying correction rather than a passive reception of meaning.

9.3 Absurdity and Nonsense: Incongruity Through Broken Physical or Logical Constraints

Absurdity and nonsense jokes work by breaking the rules that normally keep the world predictable. In incongruity terms, your mind starts with a baseline model of how things should behave—physics, logic, causality, or everyday conventions. The punchline then violates that baseline in a way that is too “wrong” to be merely surprising; it forces a different interpretation of what kind of world you’re in.

The key is not random weirdness. The violation must be specific enough that the audience can locate the broken constraint, and stable enough that a resolution can be formed. Often the resolution is not “this is true,” but “this is the kind of statement where truth-conditions are suspended.”

Foundational Constraint Types

1. Physical constraint breaks: The joke contradicts how bodies, motion, or cause-and-effect typically work.

• Example: “I tried to catch fog yesterday. I missed.”
• Why it lands: The mind expects a catchable substance with a clear target. Fog behaves like an atmosphere, not an object. The violation is concrete, so the audience quickly tags the mismatch.

2. Logical constraint breaks: The joke violates inference rules like transitivity, identity, or basic “if-then” reasoning.

• Example: “My therapist says I have issues with commitment. I told her I’m already committed to not committing.”
• Why it lands: The audience tries to apply standard logic to “commitment,” then hits a self-contradictory loop. The resolution becomes: the terms are being used as word-objects rather than consistent concepts.

3. Causal constraint breaks: The joke scrambles the expected chain from action to outcome.

• Example: “I followed the recipe exactly. Now I’m out of ingredients.”
• Why it lands: The baseline model expects consumption to be the result of cooking, not the result of following instructions. The punchline reframes “exactly” as “use everything.”

4. Category and convention breaks: The joke treats a thing as belonging to the wrong system of rules.

• Example: “This meeting could have been an email. Unfortunately, it was a meeting.”
• Why it lands: The audience expects a practical convention (meetings are sometimes unnecessary). The punchline treats the label “meeting” as if it determines necessity.

How Resolution Works When Rules Are Broken

Resolution in absurdity often comes in one of three forms:

• Constraint suspension: The audience decides the joke operates in a “nonstandard physics” or “nonstandard logic” world. Once that frame is adopted, the details stop fighting each other.
• Reframing the target: The audience shifts from literal meaning to a meta-level meaning, such as “this is about language,” “this is about bureaucracy,” or “this is about how we reason.”
• Accepting a new mapping: The audience maps familiar words to unfamiliar roles. For instance, “exactly” might mean “consume all resources,” not “follow steps precisely.”

A practical way to test whether the joke will resolve is to ask: after the punchline, can the audience state what rule was replaced? If the answer is “no,” the joke may feel like noise rather than controlled nonsense.

Systematic Construction Practices

1. Start with a normal model that is easy to activate. Everyday physics and everyday logic are reliable baselines. If the setup is already too abstract, the audience never forms a strong expectation to break.

• Example setup: “I tried to measure the room.”
• Baseline: measuring yields a number that corresponds to space.

2. Break one constraint clearly, not five at once. Multiple violations can overwhelm the audience’s ability to find a single resolution frame.

• Example: “I measured the room and it got smaller.”
• Single focus: measurement should not alter the room.

3. Make the violation stable enough to interpret. The punchline should behave consistently with the new “rules of the joke.”

• Example: “I measured the room and it got smaller. That’s how I know the ruler was honest.”
• Resolution: the ruler is treated as an agent with a strange rule.

4. Use language that signals the model switch. Small cues like “that’s how I know” or “unfortunately” can help the audience adopt the absurd frame.

• Example: “I asked the calculator for advice. It said I should round up my problems.”
• The phrasing invites a new mapping: calculators as counselors.

Example: From Broken Constraint to Coherent Resolution

Setup: “I bought a calendar to stay organized.”

Punchline: “It kept reminding me about events that hadn’t happened yet.”

Step-by-step reasoning: the baseline expects calendars to record past and planned events, not generate future reminders as if they were already known. The violation is causal and logical: reminders imply a timeline certainty the calendar shouldn’t have. The resolution is constraint suspension: the calendar is treated as a predictive device. Once that frame is adopted, the joke becomes coherent rather than random.

Absurdity works best when the audience can point to the broken rule and then quickly adopt a new interpretive frame. The humor is the moment the mind stops insisting on the original constraints and starts using the joke’s replacement rules.

9.4 Self-Deprecation and Role Humor: Incongruity Through Identity and Norms

Self-deprecating humor works by creating a mismatch between what a listener expects a “competent self” to claim and what the speaker actually claims. The speaker’s words violate a norm about identity presentation, then the listener resolves the violation by reinterpreting the role: the speaker is not performing incompetence as a literal report, but using it as a social signal. In incongruity terms, the expectation baseline is “people protect their image,” while the punchline supplies “people can safely undercut their image.”

Start with the identity baseline. Most audiences carry a default model of roles: the speaker is a person with agency, and the conversation has an implicit contract about status. Even when everyone knows the speaker is imperfect, the norm is that the speaker will manage that imperfection. Self-deprecation flips the management strategy. The violation is not merely “the speaker says something negative,” but “the speaker chooses a negative identity move in a context where positive management is expected.”

Next, specify the norm being violated. Common norms include competence signaling (don’t claim you’re bad at basic things), social reciprocity (don’t make others carry the emotional weight), and face-saving (don’t force the audience to decide whether to comfort you). A joke lands when the audience can treat the violation as low risk. That risk assessment is part of resolution: the listener’s mind checks whether the speaker’s statement is a playful role performance or a genuine self-attack that would require serious support.

Then explain the resolution mechanism. Resolution often comes from three cues working together.

First, exaggeration boundaries. The speaker typically uses a degree of overstatement that signals “this is a bit.” For example, “I’m so bad at directions I once asked my GPS for a refund” is unlikely to be a literal account, so the listener can reinterpret the identity claim as comic framing.

Second, audience alignment. The speaker may include a detail that invites shared recognition: “I tried to assemble the chair without reading the instructions. It turns out the instructions were the only part that survived.” The listener resolves the incongruity by mapping the speaker to a familiar role: the relatable person who makes ordinary mistakes.

Third, relational intent. Self-deprecation often functions as a social lubricant. “I’m not saying I’m late because I got distracted. I’m saying I’m late because my brain thinks ‘five minutes’ is a suggestion” signals that the speaker is cooperating with the group’s norms rather than rejecting them.

A useful way to analyze these jokes is to track three layers of meaning: the literal content (I’m incompetent), the role content (I’m performing a harmless identity move), and the relational content (We can laugh together because the risk is low). Incongruity appears at the boundary between layers. The listener’s job is to decide which layer should dominate.

Example: Competence Norm Reversal with Low Risk

Setup: “I tried to cook dinner.”
Punchline: “I made a sauce so thick it needed a forklift.”

The expectation baseline is that cooking attempts are at least plausibly competent. The violation is the extreme claim of failure. Resolution is easy because the imagery is clearly exaggerated and the audience can treat it as a playful identity move rather than a request for sympathy.

Example: Role Humor Through Shared Miscalibration

Setup: “I’m great at planning.”
Punchline: “I planned to start early. Then I started planning to start early.”

Here the norm is that planning should produce action. The incongruity is that the speaker’s identity as a planner collapses into a loop. The resolution comes from recognizability: many listeners have experienced similar mental procrastination, so the identity violation becomes a shared cognitive pattern.

Example: Norm Violation That Becomes Too Costly

Setup: “I can’t do anything right.”
Punchline: “I’m basically a lost cause.”

This often fails because the statement removes the exaggeration boundary and increases perceived threat. The listener may struggle to resolve the violation into a harmless role performance. When the audience model shifts toward “this needs care,” laughter becomes socially risky, and the incongruity stops being funny.

Self-deprecation therefore is not just “being negative.” It is a controlled identity mismatch where the speaker borrows the audience’s norm about image management, breaks it in a bounded way, and then supplies enough cues for the listener to restore coherence. When that restoration is quick and low-cost, the humor feels like an agreement: the speaker and audience share the same reality, even if the speaker’s self-description is intentionally off by a few degrees.

9.5 Practical Example Walkthrough: Matching Joke Types to Specific Incongruity Patterns

A useful way to analyze humor is to treat each joke as a controlled experiment: the setup builds expectations, the punchline violates one of them, and the resolution restores coherence. This section walks through that mapping for several common joke types, showing which incongruity pattern is doing the work.

Step 1: Identify the Joke Type by Its Surface Signals

Joke types are recognizable by their “surface mechanics.” Wordplay tends to advertise itself through ambiguous terms. Irony often uses praise that clearly conflicts with the speaker’s intent. Absurdity relies on physical or logical impossibility. Role humor depends on who is speaking and what that role usually permits.

Step 2: Match the Joke Type to the Incongruity Pattern

Below is a mind map that links joke types to the most typical expectation violations.

Step 3: Walk Through Examples with Explicit Expectation Tracking

Example 1: Pun That Switches Meanings

Joke: “I told my computer I needed a break. Now it won’t stop sending me Kit-Kats.”

• Expectation baseline: “A break” is rest for the user, and the computer follows the instruction literally.
• Violation: “Break” is interpreted as a snack brand cue, not a rest cue.
• Incongruity pattern: Semantic ambiguity.
• Resolution: The mind selects the alternate meaning of “break,” making the punchline coherent.

Practice tip: If a pun feels flat, the audience may not share the same dominant meaning for the ambiguous word. Adjust the setup so the intended meaning is the most accessible one.

Example 2: Irony That Reverses Intent

Joke: “Great. The meeting is canceled—right after I printed forty pages.”

• Expectation baseline: The speaker will treat the cancellation as good news.
• Violation: The speaker’s emotional stance signals the opposite.
• Incongruity pattern: Pragmatic reversal.
• Resolution: The listener infers that “great” is not literal approval but a cue for frustration.

Practice tip: Irony needs a reliable signal that the speaker’s intent differs from the literal wording. Without that signal, the listener may interpret it as genuine praise.

Example 3: Absurdity That Breaks Causal Constraints

Joke: “I tried to catch fog yesterday. I missed—then it followed me home.”

• Expectation baseline: Fog is a passive atmosphere; catching it should be impossible but not behavior-like.
• Violation: Fog is treated as an agent that can pursue.
• Incongruity pattern: Causal or structural impossibility.
• Resolution: The mind temporarily adopts a “world where fog behaves like an animal” model.

Practice tip: Absurdity lands better when the setup is ordinary enough that the audience commits to the normal model first. The punchline then has a clear job: replace the model.

Example 4: Role Humor That Violates Norms

Joke: “I asked the barista for ‘something strong.’ They handed me a glass of water and said, ‘That’s our strongest flavor: honesty.’”

• Expectation baseline: A barista will interpret “strong” as intensity (taste or alcohol-free strength cues).
• Violation: The barista reframes “strong” as truthfulness.
• Incongruity pattern: Agentive or norm violation with a pragmatic reframe.
• Resolution: The listener treats the barista’s response as role-play: the job is not just serving drinks but delivering a witty social correction.

Practice tip: Role humor depends on audience comfort with the persona. If the role is too rigid, the norm violation may feel random rather than playfully intentional.

Step 4: Diagnose Which Incongruity Is Actually Present

Sometimes a joke contains multiple violations. Use a quick checklist:

• If the punchline hinges on a word having two meanings, it’s usually semantic ambiguity.
• If the punchline hinges on “what the speaker really means,” it’s usually pragmatic reversal.
• If the punchline hinges on impossible cause-and-effect, it’s usually causal or structural impossibility.
• If the punchline hinges on who is acting and what that person is “supposed” to do, it’s usually agentive or norm violation.

Step 5: Convert the Mapping Into a Rewriting Rule

When revising, keep the joke type constant and swap only the incongruity mechanism. For instance, if you want to keep it ironic, don’t replace the pragmatic reversal with a pure pun; the resolution path will change, and the audience may feel the joke “switch gears.”

10.1 Excessive Incongruity: When Resolution Becomes Too Costly

Incongruity is funny when the mind can repair the broken expectation without spending too much effort. “Too costly” happens when the resolution requires heavy reinterpretation, too many missing assumptions, or a payoff that arrives late or never. The result is not just confusion; it’s a feeling that the joke asks for work the audience didn’t agree to do.

The Cost Model of Humor Resolution

Start with a simple idea: the audience runs a prediction, detects a mismatch, and then tries to restore coherence. Each repair step has a cost.

• Cognitive cost: how many new inferences must be made to reach a stable interpretation.
• Memory cost: whether earlier details must be held and recombined in a precise way.
• Time cost: whether the audience needs extra seconds to find the right frame.
• Social cost: whether the audience feels excluded by jargon, insider knowledge, or unclear intent.

A joke becomes “too costly” when the repair steps exceed what the audience can reasonably complete before the moment passes.

Failure Mode 1: Resolution Requires Too Many Assumptions

Some punchlines are funny because they add one clean twist. They are not funny when they require a chain of hidden premises.

Example:

• Setup: “I tried to be productive today.”
• Punchline: “So I scheduled my procrastination to optimize the entropy of my calendar.”

The mismatch is present, but the resolution demands specialized framing (entropy, optimization) and a precise mapping between “procrastination” and “calendar entropy.” Many audiences can’t complete that mapping quickly, so the joke lands as “overly technical,” not humorous.

Best practice: If you want a technical flavor, keep the resolution to one or two assumptions. For instance, replace the punchline with something that points to a concrete, everyday mechanism: “I scheduled my procrastination—now it’s on my calendar like an appointment.”

Failure Mode 2: Ambiguity Without a Clear Payoff

Incongruity often works by offering multiple candidate interpretations, then selecting one at the punchline. It fails when candidates remain equally plausible and no single resolution becomes dominant.

Example:

• Setup: “The meeting was going great until the chair started…”
• Punchline: “…thinking about its options.”

The audience may imagine many directions: literal thinking, metaphor, or a setup for a specific punchline that never arrives. The mind keeps searching, but the joke doesn’t provide the constraint needed to stop the search.

Best practice: Ensure the punchline commits to one resolution frame. Even if the frame is absurd, it should be internally consistent with the setup.

Failure Mode 3: Late Resolution That Arrives After Interest Drops

Timing matters because attention is limited. If the audience must wait too long for the “aha,” the repair process may be abandoned.

Example:

• Setup: “I told my phone a joke.”
• Middle beats: “It didn’t laugh. It just…”
• Punchline: “…updated its terms of service.”

The final twist can be funny, but if the middle beats are too vague, the audience spends effort guessing what kind of resolution is coming. By the time the terms-of-service angle appears, the audience may feel the search cost exceeded the payoff.

Best practice: Put the key constraint earlier. For example: “I told my phone a joke. It didn’t laugh; it asked me to accept new terms.”

Failure Mode 4: Resolution Conflicts with the Audience’s Social Model

Sometimes the mind can resolve the incongruity, but the resolution feels socially inappropriate or unfair. The cost becomes emotional rather than cognitive.

Example:

• Setup: “I love giving advice.”
• Punchline: “Especially when I’m wrong—then I can be confidently incorrect.”

The audience may understand the wordplay, but the resolution can read as dismissive of real mistakes. The mind stops laughing because the social appraisal turns negative.

Best practice: Align the target of the joke with the audience’s comfort. If the humor is about the speaker’s flaw, keep it self-contained rather than turning it into a judgment of others.

A Practical Diagnostic Checklist

When a joke doesn’t land, ask four questions in order.

1. Can a typical listener reach a single resolution within a few seconds? If not, reduce assumptions or add a constraint.
2. Does the punchline clearly select one frame? If not, tighten the commitment.
3. Does the setup contain enough information to justify the resolution? If not, adjust the setup so the audience’s predictions are reasonable.
4. Does the resolution feel socially safe for the audience? If not, revise the target or tone.

Excessive incongruity isn’t about being “too weird.” It’s about asking the audience to pay more than the joke returns. When the repair is cheap and the payoff is clear, incongruity becomes laughter instead of labor.

10.2 Ambiguity Without Payoff: When Multiple Interpretations Persist

Ambiguity without payoff happens when a joke creates more than one plausible interpretation, but none of them becomes clearly better once the punchline arrives. In incongruity terms, the mind detects a mismatch, then tries to restore coherence. If the resolution step fails to pick a single frame, processing stalls: the audience keeps juggling candidates instead of landing on one.

The Core Mechanism

Start with expectation formation. The setup nudges the audience toward a baseline mental model: who is speaking, what kind of statement this is, and what counts as a normal outcome. The punchline then violates that baseline. A healthy joke offers a resolution path that makes one interpretation dominate.

In ambiguity without payoff, the violation is real, but the evidence is underdetermined. Two or more frames remain equally consistent with the text, so the audience cannot justify a confident reframe. The result is not “confusion” in the sense of not understanding words; it is confusion about which meaning is intended.

Why Multiple Interpretations Persist

Several conditions encourage persistence.

1. The punchline is compatible with multiple categories. If the key word can be read literally or metaphorically, both readings may survive.
2. The punchline lacks a disambiguating constraint. A resolution usually adds a detail that rules out alternatives. Without it, the mind keeps searching.
3. The setup supports competing frames. If earlier lines already allow two plausible contexts, the punchline may simply shift emphasis without closing the loop.
4. The joke depends on shared assumptions the audience might not share. When common ground is missing, the audience cannot choose the intended frame.

A Concrete Example: The Missing Constraint

Consider this joke-like exchange:

Setup: “I tried to be spontaneous today.”

Punchline: “Now everyone thinks I’m planning something.”

The violation is that “spontaneous” should imply unpredictability, yet “planning” suggests intention. But the punchline can be read in at least two ways.

• Frame 1: Personality inference. People know the speaker’s habits, so they assume any action is strategic.
• Frame 2: Social interpretation. In that environment, spontaneity is rare, so any deviation is treated as a scheme.

Both frames fit. If the audience cannot decide which one the speaker means, the joke becomes a statement about “how people are,” not a specific incongruity with a clear resolution. The mind keeps trying to locate the intended target.

A payoff would add a constraint that selects one frame, such as a detail about the speaker’s reputation (“I’m the guy who always has a calendar invite”) or about the setting (“in this office, every surprise is a meeting”). Without that, the punchline feels like it gestures at a truth rather than resolving the mismatch.

Another Example: Wordplay That Doesn’t Close

Wordplay often risks ambiguity without payoff when the pun is technically possible in multiple directions.

Setup: “I asked the librarian for something inspirational.”

Punchline: “She said, ‘We have plenty—just not the kind you can quote.’”

This can resolve as:

• Frame A: Copyright or citation rules. The librarian enforces quoting limits.
• Frame B: Genre mismatch. Inspirational works exist, but not in quote-friendly form.

If the text never signals which frame is intended, the audience may understand the words but not the joke’s point. The incongruity is present, yet the resolution is not anchored.

How to Diagnose the Problem Quickly

Ask two questions.

1. After the punchline, can you state the intended meaning in one sentence? If you need two competing sentences, ambiguity is winning.
2. Does the punchline rule out at least one candidate frame? If it doesn’t, the audience has no reason to stop searching.

Fixing Without Over-Explaining

The goal is not to add a lecture. It is to add one small constraint that makes a single interpretation clearly superior.

• Add a detail that ties the violation to one domain (workplace, family, bureaucracy, romance).
• Replace a general noun with a specific one that narrows category membership.
• Ensure the setup does not accidentally license a second context that the punchline cannot eliminate.

When ambiguity persists without payoff, the audience’s mental model never fully updates. A good punchline forces a choice; a weak one leaves the choice open.

10.3 Negative Appraisal: When Violations Trigger Threat or Discomfort

Incongruity becomes funny when the mind can reframe the violation into something coherent. Negative appraisal happens when that reframing feels unsafe, unfair, or too costly. The key shift is from “I can solve this” to “I might be harmed, judged, or stuck.”

The Appraisal Switch from Play to Threat

Humor processing relies on a temporary permission structure: the audience treats the situation as non-literal and manageable. When cues suggest the violation is real, malicious, or socially risky, the permission structure collapses. The mind then allocates attention to protection rather than interpretation.

Three common triggers:

1. Perceived personal targeting: The violation seems aimed at someone’s status, competence, or identity.
2. Unresolved danger cues: The scenario resembles a genuine threat (physical harm, coercion, humiliation).
3. Norm violation without repair: The joke offers no plausible resolution that restores coherence.

A simple way to test this is to ask: “If the punchline were removed, would the setup still feel benign?” If the answer is no, negative appraisal is likely.

How Mental Models Turn Violations Into Discomfort

Mental models include not only facts but also expectations about intentions. When a joke violates expectations, the mind tries to infer intent. Negative appraisal occurs when intent inference flips from “misdirection” to “attack.”

Example: Someone says, “Nice job, you really nailed it,” while the context shows repeated mistakes and visible frustration. The words are incongruent with the evidence. If the audience infers the speaker is mocking rather than encouraging, the violation is appraised as social threat.

Another example: A joke about a coworker’s accent can be incongruent in wording, but the appraisal depends on whether the audience believes the speaker is challenging the person’s dignity. If the mental model becomes “the speaker is reducing them,” laughter competes with discomfort.

When Resolution Fails to Restore Coherence

Resolution is not just “finding the punchline meaning.” It also includes restoring a sense that the interaction is safe and fair. Some incongruities are hard to resolve because they require the audience to ignore too many constraints.

Consider a joke that depends on a stereotype and offers no alternative frame. Even if the wordplay is clear, the audience may not find a resolution that preserves respect. The mind then treats the violation as persistent rather than temporary.

A practical diagnostic: if the audience must supply a generous interpretation to make the joke work, many will not. Humor that relies on “you must assume good intent” is more vulnerable to negative appraisal.

The Role of Ambiguity and Timing

Ambiguity increases processing load. When the mind cannot quickly determine whether the violation is playful, it may default to caution. Timing matters because early cues guide appraisal.

Example: A sarcastic remark delivered with a flat tone and no contextual markers can read as criticism. If the speaker instead signals playfulness early—through shared context, prior banter, or a clearly non-literal framing—the same words are more likely to be reinterpreted as humor.

Incongruity that arrives abruptly can also feel threatening because the audience has less time to switch from literal to playful mode.

Concrete Examples with Appraisal Outcomes

1. Benign violation: “I told my computer I needed more space. Now it keeps sending me reminders.” The violation is about a harmless system behavior, and resolution is straightforward.
2. Discomfort risk: “You’re so good at that… for someone who struggles.” The structure implies a negative baseline. Even if the speaker intends encouragement, many audiences infer condescension.
3. Threat-like framing: “Relax, it’s fine—I’ll just do it the unsafe way.” If the setup resembles real risk, the audience may appraise it as endorsing harm rather than joking.

Practical Guidelines for Keeping Violations Safe

• Signal the frame early: Use context that makes the non-literal mode plausible.
• Choose violations that are reversible: Prefer incongruity that can be resolved without requiring the audience to ignore harm.
• Avoid “no-win” targets: If the joke depends on someone being diminished, negative appraisal is likely.
• Ensure a real resolution: The punchline should restore coherence and fairness, not just wordplay.

A final check: if a reasonable listener could interpret the setup as criticism or threat before the punchline arrives, the joke is fighting negative appraisal rather than riding it.

10.4 Social and Ethical Constraints When Humor Fails Due to Contextual Mismatch

Incongruity theory explains why a violation can feel funny, but it does not guarantee that the violation is acceptable. Humor also depends on social expectations: who is speaking, who is being targeted, what norms are active in the moment, and what kinds of harm the audience can reasonably infer. When contextual cues and ethical boundaries do not align, the mind still detects an incongruity, yet the resolution path is blocked by appraisal.

Contextual Mismatch as a Constraint on Resolution

A joke typically offers a resolution that restores coherence. Social and ethical constraints add extra “gates” before that resolution can be accepted. The audience may still find the punchline clever, but they may refuse the intended frame because it conflicts with norms such as fairness, respect, safety, and role-appropriate behavior.

Consider three common mismatch patterns:

1. Targeting mismatch: The setup implies a harmless target (a situation, a shared habit), but the punchline shifts the target to a person or group.
2. Role mismatch: The speaker’s role (teacher, manager, friend) implies certain boundaries, but the joke violates them.
3. Intent mismatch: The wording invites a benign interpretation, yet the punchline signals contempt, coercion, or exclusion.

Each pattern can break the resolution because the audience’s mental model updates from “this is playful” to “this is harmful or disrespectful,” and that update changes the appraisal stage.

A Practical Mind Map of Failure Points

Examples That Show Where the Gate Closes

Example 1: Targeting mismatch

• Setup: “I tried to be productive today, but my calendar keeps scheduling meetings with my future self.”
• Punchline: “Turns out my future self is the kind of person who never shows up.”

The first beat invites a light, self-referential frame. The punchline shifts from a situation (calendar behavior) to a judgment about a “person,” which can feel like a personal attack. If the audience reads “future self” as a stand-in for someone present, the ethical gate closes because the joke now depends on demeaning.

Example 2: Role mismatch

• Setup in a workplace meeting: “Let’s keep it simple—if the report had a personality, it would be… honest.”
• Punchline: “Because it would admit it’s missing half the data.”

This can be funny when peers are joking. It can fail when a manager uses it to pressure an employee. The incongruity is still there, but the audience’s mental model includes power dynamics. The resolution “this is just humor” becomes less plausible, because the role implies consequences.

Example 3: Intent mismatch through stereotyping

• Setup: “Some people are just bad at directions.”
• Punchline: “You can tell by how they act—like they’re from a different place.”

Even if the speaker intends it as general clumsiness, the punchline recruits group identity as the explanation. The audience’s appraisal shifts from “a quirky trait” to “a stereotype.” The ethical gate blocks the resolution because the joke asks the audience to accept an unfair causal story.

How to Design Context So the Resolution Stays Available

A useful best practice is to make the target and role legible early. You do not need to add disclaimers; you need to align cues.

• Keep the target stable: If the setup points to a situation, avoid a punchline that reassigns the blame to a person.
• Match the speaker’s role to the tone: In authority contexts, prefer jokes that critique systems or behaviors rather than individuals.
• Use ambiguity carefully: Pronouns like “they” can accidentally widen the target from a specific person to a group.
• Choose violations that do not require harm: Incongruity can come from logic, timing, or perspective without implying disrespect.

A Quick Diagnostic for Contextual Mismatch

If a joke fails, ask what the audience likely inferred at the appraisal stage:

• Did they infer a target that was not signaled?
• Did they infer a power imbalance that changes the meaning of teasing?
• Did they infer an intent that conflicts with respect or inclusion?

When the answer is yes, the problem is not only “the joke was surprising.” The problem is that the audience cannot accept the resolution frame without violating social and ethical expectations.

10.5 Practical Example Walkthrough: Diagnosing Why a Joke Falls Flat and How to Fix It

A joke falls flat when the audience cannot complete the expected path from setup to resolution. In incongruity terms, either the violation is too weak to notice, the resolution is too expensive to compute, or the context makes the wrong mental model feel “most likely.” The quickest diagnosis is to treat the joke like a small system: baseline prediction, violation location, resolution mechanism, and appraisal.

Step 1: Pick a Single Test Joke and Write Its Intended Path

Start with one short joke you want to work on. Then write four one-sentence notes:

• Baseline: What does the audience predict after the setup?
• Violation: What expectation is broken, and where?
• Resolution: What reinterpretation makes the broken expectation coherent?
• Appraisal: What emotional evaluation should follow (amused, impressed, relieved)?

Example draft:
“Why did the programmer quit his job? He didn’t get arrays.”
Baseline: The reason will be a job-related complaint.
Violation: “Arrays” is treated as a literal thing you can “get.”
Resolution: The audience realizes “arrays” sounds like “a raise.”
Appraisal: Light wordplay amusement.

If your notes don’t fit cleanly, the joke is already telling you where it’s failing.

Step 2: Diagnose the Failure Mode Using a Mind Map

Step 3: Apply the Diagnosis to a Concrete Example

Consider a revised version that often fails:
“Why did the programmer quit his job? He didn’t get arrays. That’s why.”

What goes wrong?

• Baseline: Still job-related.
• Violation: The audience may notice “arrays,” but the punchline’s sound-alike “a raise” is not forced.
• Resolution: “That’s why” adds a generic closure without supplying the missing link.
• Appraisal: The audience may feel they’re being asked to do extra guessing.

You can test this by asking: if someone reads only the punchline, can they infer the intended resolution? If not, the bridge is missing.

Step 4: Fix by Editing One Variable at a Time

Try three targeted fixes.

Fix A: Tighten the baseline cue “Why did the programmer quit his job? He didn’t get a raise.” This removes the need to compute the pun, so it’s less incongruous but more reliably funny for audiences who dislike wordplay.

Fix B: Increase salience of the violation “Why did the programmer quit his job? He didn’t get arrays—at least, that’s what HR said.” Now the audience has a reason to treat “arrays” as something said aloud, making the sound-alike more noticeable.

Fix C: Add a bridge that preserves surprise “Why did the programmer quit his job? He didn’t get arrays—he got ‘a raise,’ apparently.” The bridge line explicitly connects the reinterpretation while keeping the initial violation.

Each fix changes a different part of the system. If you change everything at once, you won’t know what solved the problem.

Step 5: Use a Quick Checklist Before You Finalize

• Does the setup narrow the audience to one dominant baseline frame?
• Is the violation located early enough to be noticed before the audience gives up?
• Does the resolution provide a clear reinterpretation path, not just a conclusion?
• Does the joke reward the audience for the effort it demands?
• Does the appraisal match the target (light, not hostile)?

Step 6: A Second Example with a Different Failure Mode

Draft:
“I told my friend I don’t trust stairs. They’re always up to something.”
Common failure: Some readers miss the wordplay because “stairs” and “up to something” are not linked quickly.

Fix:
“I don’t trust stairs. They’re always up to something—literally.”
The added “literally” signals the intended reinterpretation, reducing resolution cost while keeping the incongruity.

By diagnosing baseline clarity, violation salience, resolution bridge, and appraisal fit, you can turn “no laugh” into “late laugh” and then into “right on time,” without guessing blindly.

11.1 Step 1—Identify the Expectation Baseline: What the Audience Predicts

Expectation baseline is the audience’s best guess about what will happen next, built from what they already know and what the text or scene has signaled so far. In incongruity theory terms, you’re not hunting for the punchline yet; you’re mapping the “normal” interpretation that makes the later violation feel like a violation rather than random noise.

Start with a simple rule: the baseline is whatever the audience would predict if they had to bet on the next moment. That prediction can be about meaning (what a phrase refers to), structure (what kind of sentence comes next), causality (what causes what), or social norms (what people in this situation typically do).

What Counts as Baseline Signals

Baseline signals come from multiple channels, and you can usually find at least three in a short joke.

1. Linguistic cues: word choice, tense, and familiar constructions. If you say “I tried to be productive,” many listeners expect a self-improvement story.
2. Genre cues: the form itself. A riddle primes “answer expected.” A dialogue primes “two speakers with roles.”
3. Context cues: setting, relationship, and stakes. “At the airport” primes time pressure and rules.
4. Cultural and social cues: politeness norms, taboo boundaries, and common scripts. “Asking for a raise” primes negotiation dynamics.

A baseline is strongest when the signals converge. If the cues point in different directions, the audience’s predictions become unstable, and the later violation may land as confusion.

A Systematic Method to Build the Baseline

Use this sequence every time, even when the joke seems obvious.

1. Freeze the moment before the punchline: stop reading at the last “setup” word.
2. List the audience’s likely interpretations: write 2–4 candidate predictions, not just one.
3. Rank them by plausibility: which interpretation is most likely given the cues so far?
4. Name the mental model: what “world” is being assumed? For example, “workplace rules,” “family roles,” or “physics as usual.”
5. Identify the expected next action or meaning: what should happen next if the model stays intact?

This method prevents a common mistake: describing the baseline as “what the audience thinks is funny.” Humor is not the baseline; the baseline is the expected meaning or action.

Example: Baseline from a Short Setup

Consider the setup: “I told my computer I needed a break, and it said, ‘No problem—taking one now.’”

Before the punchline, the baseline prediction is that the computer will respond in a literal, human-like way: it will “pause” or “restart” something, or it will acknowledge the request. The mental model is “a computer follows commands and communicates status.” The expected next meaning is a benign, operational response.

Now imagine the later violation: it “takes a break” by shutting down the system. That violation works because the baseline already made “break” mean “pause my work,” not “stop everything.”

Example: Baseline from Structure

Setup: “Why did the scarecrow win an award? Because he was outstanding in his field.”

Here the baseline is structural. The audience expects a causal explanation pattern: “Because…” followed by a reason that matches the question’s framing. The mental model is “wordplay will reinterpret a literal phrase.” The prediction isn’t just the answer; it’s the kind of answer that fits the question’s grammar.

Common Failure Modes to Watch

• Baseline too vague: if the setup provides no stable model, the audience can’t tell what was expected.
• Baseline too many: if multiple models are equally plausible, the later violation may not feel like a single, targeted break.
• Baseline mismatch with audience: if the cues rely on knowledge the audience lacks, the baseline never forms.

Step 1 is complete when you can state, in one or two sentences, what the audience expects next and which mental model supports that expectation. Only then does it make sense to locate the violation in Step 2.

11.2 Step 2—Locate the Violation: Pinpoint the Incongruity Type and Location

Step 2 answers two questions: what kind of expectation breaks, and where in the joke the break becomes visible. If you can name the violation type and point to the exact phrase or beat that triggers it, you can explain the humor without guessing.

Identify the Incongruity Type

Use a quick classification pass. Read the joke once for meaning, then a second time for mismatch.

• Semantic Incongruity: the words point to one meaning or category, but the punchline forces another. The violation is about what things are.

  • Example: “I told my computer I needed a break. Now it won’t stop sending me Kit-Kats.” The expectation is that “break” means rest; the punchline treats it as a snack.

• Syntactic or Structural Incongruity: the sentence structure or expected grammar pattern doesn’t behave normally, causing a misparse or a reassembly.

  • Example: “The chef served the soup cold, but the customer was the one who was hot.” The structure sets up a temperature contrast, then swaps which entity is “hot,” forcing a structural reinterpretation.

• Pragmatic Incongruity: the speaker’s intent or conversational norms don’t match what the audience assumes. The violation is about what the speaker is doing.

  • Example: “Nice job on the report—especially the parts you didn’t write.” The surface praise conflicts with the pragmatic intent of criticism.

• Agentive or Causal Incongruity: actions or causes happen in a way that violates the expected mechanism. The violation is about how events connect.

  • Example: “I tried to meditate, but my thoughts filed a complaint and demanded a manager.” The expected mechanism of meditation (quieting thoughts) is replaced by a new causal chain.

A practical rule: if the mismatch is primarily about word meaning, choose semantic; if it’s about sentence form, choose structural; if it’s about speaker intent, choose pragmatic; if it’s about cause and agency, choose agentive/causal.

Pinpoint the Violation Location

Location is the smallest unit where the audience’s prediction stops working. That unit can be a single word, a clause boundary, or a timing beat.

Follow this method:

1. Mark the expectation boundary: identify the last moment before the audience feels confident about what comes next.
2. Mark the first mismatch token: find the earliest word or phrase that makes the prior prediction fail.
3. Mark the resolution pressure: note what the punchline forces you to reinterpret, even if you don’t fully understand it yet.

Mini Example Walkthrough

Joke: “I used to hate facial hair. Then it grew on me.”

• Expectation boundary: “I used to hate facial hair.” The audience predicts a complaint or a preference change.
• First mismatch token: “Then it grew on me.” The phrase “grew on me” is expected to mean “became acceptable,” but it also reads literally as hair growth.
• Resolution pressure: you reinterpret “grew on me” as a double meaning, restoring coherence.

Here, the violation type is semantic (double meaning), and the location is the phrase “grew on me.”

A Systematic Output Template

When you finish Step 2, write a compact statement that includes three parts: type, location, and mismatch.

• Type: semantic / structural / pragmatic / agentive-causal
• Location: quote the exact phrase or beat
• Mismatch: one sentence describing what prediction fails

Example output: “Semantic violation at ‘grew on me’: the phrase shifts from literal growth to an idiom for acceptance, forcing a new meaning assignment.”

This is the bridge to Step 3, because the resolution you describe later should directly address the mismatch you just located.

11.3 Step 3—Describe the Resolution: Specify the Reframe and Coherence Restoration

A joke’s resolution is the moment the mind stops treating the stimulus as “wrong” and starts treating it as “meaningful.” In incongruity theory terms, you explain how the audience replaces the initial expectation with a new interpretation that fits the evidence. This step is where you make the mechanism explicit: what gets reframed, what constraints are restored, and why the new frame feels coherent.

The Reframe: What Changes in the Audience’s Model

Start by naming the original frame the audience used. Then state the reframe in plain language: the new assumption that makes the punchline compatible with the setup.

• Reframe as category shift: the audience realizes the target is not the category they first assumed.
• Reframe as role shift: the audience realizes who is responsible for the action or who holds the relevant knowledge.
• Reframe as rule shift: the audience realizes the “rules” governing the situation are different than expected.
• Reframe as perspective shift: the audience realizes the statement is from a different viewpoint (literal vs figurative, speaker vs observer).

Example:

• Setup: “I told my computer I needed a break.”
• Punchline: “It said: ‘No problem. I’ll go to sleep.’”

Original frame: the computer is a tool that follows human instructions. Reframe: the computer follows its own operational logic; “break” maps to a system state.

Once you specify the reframe, you can describe the coherence restoration: the audience’s mental model now predicts the punchline rather than contradicting it.

Coherence Restoration: How Meaning Gets Put Back Together

Coherence restoration is the process of reducing conflict between the audience’s expectations and the incoming details. You can describe it as constraint satisfaction: the new frame must explain the punchline without forcing the audience to ignore key information.

Use three checks:

1. Coverage: Does the new frame account for the setup details, not just the punchline?
2. Consistency: Do the parts of the joke agree with each other under the new frame?
3. Economy: Is the reframe simple enough that the audience can adopt it quickly?

Example:

• Setup: “My friend is great at telling jokes.”
• Punchline: “He’s so good that I don’t even hear the punchline.”

Reframe: the “joke” is not primarily the spoken content; it’s the social effect of attention and timing. Coherence restoration: the audience can treat the missing punchline as a consequence of the friend’s delivery control, not as a random omission.

Mind Map: Resolution Mechanics

A Systematic Template You Can Apply to Any Joke

Use this sequence to keep your explanation tight and complete.

1. Name the initial expectation: what the audience predicts.
2. State the reframe: the new assumption that changes the interpretation.
3. Show the mapping: how the punchline’s elements fit the new assumption.
4. Confirm coherence: which constraints are restored (coverage, consistency, economy).
5. Mention the prediction flip: the audience now expects the punchline to follow.

Example:

• Setup: “I tried to catch fog yesterday.”
• Punchline: “Mist.”

Initial expectation: catching fog should involve a physical action with a tangible result. Reframe: “catch” is treated as a wordplay target; the outcome is a pun on “mist.” Mapping: fog → mist; catch → the sound-alike resolution. Coherence restoration: coverage is satisfied because the setup’s fog term is directly reused; consistency holds because the joke’s “result” is linguistic, not physical; economy holds because the audience can adopt the wordplay frame immediately.

Common Failure Modes to Avoid

A resolution description is incomplete if it only says “the mind changes its mind.” Make it concrete.

• Vague reframe: “They interpret it differently.” Instead, specify category, role, rule, or perspective.
• No mapping: the punchline must be shown to fit the new frame.
• No coherence check: if coverage or consistency is missing, the explanation doesn’t justify why the audience would accept the resolution.

When you do Step 3 this way, you’re not just labeling the punchline as funny. You’re explaining the exact mental move that turns contradiction into understanding.

11.4 Step 4—Account for Surprise: Explain Why the Violation Feels Notable

Surprise is the feeling that something important has changed. In incongruity-based humor, that feeling is not random; it comes from a mismatch between what the mind expects and what it actually receives. Step 4 asks you to explain why the mismatch is noticeable enough to be processed as a “violation,” rather than just a minor oddity.

Start with the Expectation Baseline

Before you can justify surprise, you need to state what the audience was predicting. Expectations come from prior language patterns, typical roles, causal stories, and social norms. When the setup is clear, the baseline becomes stable, and the punchline has something to break.

Example: “I told my computer I needed more space, and it said…”

• Baseline: computers respond with storage-related actions.
• Surprise target: the response will not be storage-related in the usual way.

Explain the Prediction Error Signal

The mind treats the incoming information as evidence. If the evidence strongly contradicts the baseline, the system registers a prediction error. Surprise rises when the contradiction is both strong and immediate.

Example: “…it said, ‘Windows.’”

• The word “Windows” is a category shift: from storage to an interface feature.
• The shift is fast because the listener already committed to a storage frame.

A useful way to describe this in plain terms is: the punchline forces a re-evaluation at the exact moment you thought the story would continue normally.

Tie Surprise to Salience and Timing

Not all violations feel equally surprising. Two factors matter most.

1. Salience: the violated element is prominent in the listener’s attention. If the punchline changes a key noun or verb, the mismatch is harder to ignore.
2. Timing: the violation arrives at a point where the listener expects resolution or continuation. If the punchline appears right after a complete setup, the listener has already formed a strong prediction.

Example: “My friend said he’s great at math. I asked him to help me with my budget—he said he only works in decimals.”

• Salience: “works in decimals” reframes the skill.
• Timing: the request for budgeting makes the listener expect practical help.

Distinguish Surprise from Confusion

Surprise is about a sharp mismatch; confusion is about insufficient information. Humor needs surprise that can be resolved. If the audience cannot find a coherent reframe, the mismatch stays unresolved and becomes irritation.

Example of low humor: “I tried to bake a joke, but the oven refused.”

• Surprise exists, but the listener lacks a clear resolution path.

Example of higher humor: “I tried to bake a joke, but the oven refused to preheat—apparently it only runs on punchlines.”

• The listener can map “preheat” to “setup” and “punchlines” to “heat,” restoring coherence.

Show How Resolution Makes Surprise Pay Off

Step 4 should connect surprise to the later resolution described in Step 3. Surprise feels “worth it” when the reframe explains why the violation occurred.

Example: “He said he’d be on time. Then he showed up yesterday.”

• Surprise: “on time” conflicts with “yesterday.”
• Resolution: the phrase “on time” is interpreted as “at the correct time in the story,” not in real time.

The explanation should make clear that the punchline is not merely different; it is different in a way that supports a new interpretation.

Use a Mind Map to Keep the Logic Tight

Mind Map: Why the Violation Feels Notable

A Compact Example Walkthrough

Consider: “I asked the librarian if they had books on paranoia. She said, ‘They’re right where you left them.’”

• Baseline: paranoia books will be categorized and retrieved normally.
• Violation: “right where you left them” implies the listener’s mental state controls the location.
• Surprise: the punchline shifts from physical retrieval to a psychological interpretation at the moment the listener expects a straightforward answer.
• Resolution: the listener reframes the librarian’s statement as a joke about the mind’s tendency to assume patterns and blame oneself for “misplacing” certainty.

In Step 4, your goal is to state these links explicitly: what was expected, what contradicted it, why the contradiction was noticeable, and why the later reframe makes that notice feel like humor rather than noise.

11.5 Practical Example Walkthrough: Applying the Toolkit to a Short Set of Jokes

Below is a systematic walkthrough using the toolkit from this chapter. The goal is not to “rate” jokes, but to explain how a listener’s expectations are built, where they break, and how resolution restores coherence.

Toolkit Recap in One Pass

1. Identify the expectation baseline: what the audience predicts will happen next.
2. Locate the violation: the exact place where the prediction fails.
3. Describe the resolution: the reframe that makes the new meaning coherent.
4. Account for surprise: why the violation stands out given the baseline.

Short Set of Jokes

Joke A: “I told my computer I needed a break, and it said: ‘No problem—I’ll go to sleep.’”

Step 1: Expectation Baseline

You expect a human-like response that either refuses your request or jokes back without changing the “system.” The baseline includes a mental model: computers don’t “choose” to rest.

Step 2: Locate the Violation

The violation is agentive and pragmatic: the computer behaves like a cooperative agent with intent (“I’ll go to sleep”).

Step 3: Describe the Resolution

Resolution comes from a dual-model shift. One model treats the computer as an inanimate tool; the other treats it as a character speaking in metaphor. The punchline reframes “sleep” as a literal system state, not a personal decision.

Step 4: Account for Surprise

Surprise is moderate because the language cue (“it said”) already invites personification. The novelty is that the personification is grounded in a real computer action, so the reframe feels stable rather than random.

Joke B: “Why don’t scientists trust atoms? Because they make up everything.”

Step 1: Expectation Baseline

You predict a straightforward reason involving reliability, measurement error, or scientific skepticism. The baseline includes a mental model of atoms as physical units.

Step 2: Locate the Violation

The violation is semantic and category-based: “make up” shifts from “compose” (physical composition) to “invent” (fabricate).

Step 3: Describe the Resolution

Resolution is a word-meaning switch that preserves coherence. Once “make up” is interpreted as “fabricate,” the joke’s causal logic works: scientists distrust what “makes up everything.”

Step 4: Account for Surprise

Surprise comes from constraint relaxation on the phrase “make up.” The listener briefly commits to one meaning, then revises when the second meaning fits the punchline’s structure.

Joke C: “I used to hate facial hair. Then it grew on me.”

Step 1: Expectation Baseline

You expect a conventional emotional change explanation: “I changed my mind” or “I learned to like it.”

Step 2: Locate the Violation

The violation is causal and semantic: the phrase “grew on me” is treated as both figurative and literal. The mind model shifts from psychology to biology.

Step 3: Describe the Resolution

Resolution occurs when the listener maps the emotional idiom onto a physical process. The new interpretation makes the causal chain coherent: the hair literally grows, and the preference metaphorically “grows” too.

Step 4: Account for Surprise

Surprise is low-to-medium because the idiom “grew on me” is familiar. The punchline’s work is to tighten the metaphor into a literal reading without adding extra steps.

Mind Map: Toolkit Trace Across the Set

## Applying the Incongruity Toolkit - Expectation Baseline - Joke A: computers as tools, no agency - Joke B: atoms as physical units - Joke C: emotions as non-literal change - Violation Location - Joke A: agentive speech and intent - Joke B: semantic shift in “make up” - Joke C: causal idiom treated literally - Resolution Mechanism - Joke A: dual-model reframing, “sleep” as system state - Joke B: word-meaning switch, “fabricate” fits logic - Joke C: metaphor-to-literal mapping, growth as both processes - Surprise Accounting - Joke A: personification grounded in real action - Joke B: constraint relaxation on phrase meaning - Joke C: familiar idiom, tightened interpretation - Output - Coherence restored - Listener updates mental model with minimal extra cost

Integrated Takeaway: What Makes These Examples Work

Across the set, each punchline performs a specific kind of model repair. Joke A repairs the “agency” assumption by grounding personification in a real system behavior. Joke B repairs a “meaning” assumption by switching the phrase’s sense while keeping the causal structure intact. Joke C repairs a “figurative vs literal” assumption by aligning the idiom with a plausible physical event.

If you want to apply the toolkit quickly to any new joke, start by naming the baseline model in one sentence, then point to the smallest unit that breaks it (a word, a role, a causal link). Finally, state the resolution as a single reframe that would let the listener say, “Oh, that’s what it meant.”

12. Applied Practice: Building and Revising Humor with Cognitive Constraints