Framing Effect: A Two-Frame Test Before You Decide

Years ago, in a design studio, we spent an afternoon arguing about two lines of packaging copy for the same tub of yoghurt. One read "95 percent fat free". The other read "contains 5 percent fat". Same tub, same spoon. The room split down the middle, and I remember thinking the argument was about words. It wasn't. It was about which half of the truth we wanted the shopper to hold.

It took me years to notice that the shopper was not the only one being framed. So was I, every time someone described a decision to me. A contract is "guaranteed income" or "locked-up hours". The facts sit still. The preference moves.

Psychologists call this the framing effect. In their 1981 paper "The Framing of Decisions and the Psychology of Choice", Amos Tversky and Daniel Kahneman asked people to choose between two programmes for an outbreak of disease expected to kill 600 people. Described as gains, Programme A would save 200 people for certain, and Programme B had a one-third chance of saving all 600 and a two-thirds chance of saving no one. Seventy-two percent chose the certain rescue. A second group saw the same programmes as losses. Programme C meant 400 people would die for certain; Programme D carried a one-third chance that nobody would die and a two-thirds chance that all 600 would. Now 78 percent took the gamble. Only the words had changed, and the majority walked to the opposite side of the room.

You may ask: are we really that easily fooled? Here is where it gets interesting. In 2014 David Mandel published "Do framing effects reveal irrational choice?", and his answer was: not entirely. Most participants read "200 people will be saved" as "at least 200", so the certain option quietly carried a hidden upside in the gain frame that it did not carry in the loss frame. When he rewrote the options so the numbers meant exactly what they said, the famous gap closed in his experiments. Part of what looks like a broken mind is really an ambiguous sentence, and the cure starts with precise language rather than self-reproach.

But hold on. Mandel did not conclude that framing is harmless. Words smuggle in information, and we rarely stop to ask what a description rules in or out. In everyday decisions nobody hands us "exactly". They hand us "guaranteed", "risk", "opportunity", "cut". So the engine keeps running, and the engine is loss aversion: a loss of a given size feels heavier than a gain of the same size. Describe an outcome as something you might lose and you have raised its weight without touching its value.

Framing is easy to confuse with its cousin, the anchor. Anchoring is about a number that pulls your estimate toward itself. Framing is about a description that flips your preference between options whose numbers never moved. One bends how much. The other bends which.

So what do I do about it? I don't have a cure, and I'm wary of anyone who says they do. What I have is a two-frame test. It won't stop the frame from working on me, but it lets me catch it at work before I sign.

1. Write the decision neutrally. Options, outcomes, numbers, time horizon. No adjectives. If you can't strip the adjectives out, that is the first finding.
2. Write the gain frame. For each option, what does it secure, keep, or win? Use the same numbers and the same horizon as the neutral version.
3. Write the loss frame. For each option, what does it give up, cost, or risk? This is the version your gut has probably already written.
4. Notice whether your preference moved. If you prefer one option in the gain frame and the other in the loss frame, the words are deciding. You are not.
5. Ask which frame feels heavier, and why. Sometimes the heavy frame is pointing at something real: a loss you cannot recover. Sometimes it is only the word "lose" doing its work. Write down which you think it is.
6. Record both frames before you commit. Put them in a decision journal with your reasoning. Hindsight is very good at remembering only the frame that turned out to be right.

Last year I was offered a retainer with a single client. Neutral version: 40 percent of my hours for twelve months at a fixed rate. Gain frame: it guarantees 40 percent of next year's income. Loss frame: it locks up 40 percent of next year's hours. In the gain frame I wanted to sign that afternoon. In the loss frame I wanted to stall. Then came question five. Was the heaviness real? Partly, yes. The hours were the one thing I could not buy back, and the client's field was not where I wanted to grow. So the loss frame was carrying information, not only emotion. I took a shorter version of the retainer, and I knew why. The test had not told me the answer. It told me which of my two reactions deserved a hearing, and that is the most a bias check can honestly promise.

This is also where ClearMind fits, quietly. It will not declare the right option. It works as a mirror: it asks for the decision in your own words, reflects the gain and loss versions back at you, and points out where your preference shifted between them. The judgment stays with you. And since framing rarely arrives alone, the broader cognitive bias checklist is worth a pass before anything large.

The same mountain is shade on one face and sun on the other. The mountain has not changed. Only where we are standing has.

A frame is not the picture. It is the edge someone chose to draw around it.