A car accident is the same event. But describe it as "cars contacted each other" and witnesses estimate impact speed at 31.8 mph. Describe it as "cars collided" and the estimate jumps to 34.0 mph. Call it a "crash" and it's 38.1 mph. Call it "cars smashed into each other" and it's 40.5 mph.
Same accident. Different word. Different perception of the same event. This is the Loftus & Palmer (1974) effect, foundational research in memory and language.1 Descriptive language doesn't just describe reality—it actively shapes how the brain encodes and remembers that reality.
Words are lenses. The same event looks different through different linguistic lenses. "Smashed" loads the memory with violence, impact, force. "Contacted" loads it with gentleness, proximity, connection. The word itself becomes part of the memory, changing what the memory contains.
Häagen-Dazs understood this profoundly. They don't sell ice cream—a commodity. They sell "indulgence," "luxury," "a moment of pure pleasure." The descriptive language rewrites what the ice cream is. The ice cream tastes different when it's described as "indulgence" than when it's described as "frozen dairy." Same product, different language, different perception of taste itself.
Loftus & Palmer's research showed that language doesn't just describe events—it shapes the encoding of those events in memory. When you hear "smashed," your brain activates concepts of force, violence, impact. These concepts become part of the memory trace itself. The accident becomes, in memory, more violent than the actual accident was.
This isn't conscious distortion. It's not that people are lying. It's that language literally changes what gets encoded. The verb you use determines what neural patterns activate, and what activates gets remembered.
Shotton emphasizes this with descriptive language in branding. Häagen-Dazs doesn't describe the ice cream. They describe the experience of eating the ice cream: "indulgence," "luxury," "a moment for yourself." The language activates concepts that the actual ice cream doesn't contain. The neural activation becomes part of the experience memory, so the ice cream tastes like what the language promises.
This is why wine descriptions matter. "Earthy," "fruity," "bold"—the language activates taste-related neural regions in ways that the wine might not. The language doesn't change the wine's chemistry. It changes how the brain processes the wine, and that processing changes the experience of the wine.
Loftus extended this finding: the specific verb used during memory encoding affects what gets remembered about the event. Not just emotionally (the accident feels more violent), but factually.
Weeks later, when asked "Did you see broken glass?"—people who heard "smashed" were 32% more likely to falsely remember broken glass that didn't exist. The verb activated the memory of "breaking" so strongly that the brain filled in details that weren't there.
This shows the mechanism at full power: language doesn't just color experience. It actively shapes what gets encoded and what gets retained. The verb restructures the memory itself.
Step 1: Identify the neutral descriptor of your product What is it, objectively? Ice cream is "frozen dairy product." A car is "assembled metal and plastic." A newspaper is "printed paper with text." These are neutral. They activate minimal neural patterns.
Step 2: Identify the experience you want encoded Not what the product is, but what you want the customer to remember feeling after using it. Häagen-Dazs wants "indulgence," "luxury," "self-care." A luxury car wants "power," "control," "freedom."
Step 3: Select verbs and descriptors that activate those experience concepts Don't describe the ice cream. Describe the eating of the ice cream using language that activates the experience concepts. "Indulge," "savor," "treat yourself," "escape."
For a car: "command," "unleash," "dominate the road."
For a product's social function: don't say "helps you stay in touch." Say "deepens connection," "strengthens bonds," "keeps loved ones close."
Step 4: Test the language against competitors Loftus showed that the specific verb matters. 27% difference in impact estimate between "contacted" and "smashed." Test your descriptive language by measuring perception shifts.
Actually ask customers: "How would you describe using our product?" and "How would you describe using the competitor's product?" What verbs show up? Which set of verbs do you want to activate?
Step 5: Repeat the language consistently The linguistic encoding effect is strongest with repeated exposure. Use the same descriptors across marketing, packaging, customer service. Each repetition reinforces the neural activation. Häagen-Dazs uses "indulgence" repeatedly. That word gets wired into the memory of the product.
Language that doesn't match reality backfires. If you describe ice cream as "luxury" but it tastes cheap, the linguistic activation creates cognitive dissonance. The memory of the experience contradicts the language. Trust breaks.
The language has to be believable within the context of the product. You can use "indulgence" for premium ice cream. You can't use "indulgence" for budget ice cream without creating dissonance.
Descriptive language pairs with pricing psychology. Expensive products described with "indulgence" language feel like the price is justified. Cheap products described with "indulgence" language feel misleading. The language and price have to align.
Shotton shows this: Häagen-Dazs can use "indulgence" language because the high price signals "this is premium." A budget brand using identical language creates dissonance. The customer feels manipulated by language that doesn't match the price.
Behavioral-Mechanics → Expectation Assimilation: Descriptive language sets expectations. "Smashed" sets the expectation that the accident was severe. When the visual experience arrives, it gets assimilated to the expectation set by the language. Expectation Assimilation explains how language sets the expectation that shapes perception of reality.
Psychology → Present Bias: Descriptive language activates emotions in the present. "Indulgence" activates present desire and pleasure. "Luxury" activates present status and self-worth. The language makes the present emotional experience heavier than the future cost, supporting present-bias purchasing. Present Bias explains why emotional language in the present overrides rational future consideration.
Behavioral-Mechanics → Concreteness Bias: Concrete language ("smashed," "crisp," "bold") activates more neural regions than abstract language ("good," "quality," "nice"). Concrete language creates stronger memory encoding. Concreteness Bias explains why specific sensory verbs encode memories more strongly than generic abstract descriptors.
Sharpest Implication: Language doesn't just communicate about your product—it literally changes what your product becomes in customer memory. The specific verbs and descriptors you use are neurologically active in ways that reshape product experience. The implication: you can make the same product feel like different products just by changing the language. A "budget" car described as "nimble" and "efficient" feels different from the same car described as "cheap" and "basic." The language rewires the neural encoding, so the actual experience of using the product changes.
Generative Questions: