Framing

What it is

Ask two groups of people to choose a medical treatment. Tell one group it has a ninety per cent survival rate. Tell the other it has a ten per cent mortality rate. The numbers are arithmetic complements of each other, but the choices are not: the survival framing draws more people toward the treatment than the mortality framing does. Nobody in either group has done the maths wrong. Each group answered the question it was actually asked, and the question changed shape even though the underlying fact did not.

That is framing: the finding, formalised by Daniel Kahneman and Amos Tversky inside prospect theory, that a decision depends on how the options are described as much as on what the options are. A price framed as a saving reads differently from the same price framed as a cost. A discount framed as a gain lands differently from the identical amount framed as an avoided loss. The mind is not running one calculation on raw information and then reporting the result. It is running the calculation the frame hands it, and then treating that result as though it came from the facts alone.

Framing works as an umbrella term for a family of more specific effects that recur across this site. The context effect is framing applied to an object rather than a statement, where price, setting or story changes how something tastes or looks rather than how a sentence reads; the wine-pricing research anchoring that page, Plassmann, O'Doherty, Shiv and Rangel's 2008 fMRI study, is arguably the single most rigorously documented case of framing in this vault, so its detail lives there rather than being repeated here. The white cube is framing applied to a physical room, where the gallery's blank walls constitute the frame. Status signaling depends on framing too, since a costly display only reads as a signal inside a context that has already taught its audience what to look for.

None of this makes framing a trick that works on any claim in any order. It works because the mind treats a frame as informative, and most of the time a frame genuinely is informative: a discount usually is a saving, a survival statistic usually does describe real odds. The cases worth paying attention to, and the ones this concept exists to name, are the ones where two frames describing the identical fact pull people toward opposite decisions.

In effect

Ninety per cent survival, ten per cent mortality

Daniel Kahneman discusses this framing effect at length in Thinking, Fast and Slow (2011), building on his and Amos Tversky's earlier work formalising prospect theory. Presented with a medical outcome stated as a survival rate, people lean toward the treatment; presented with the identical outcome stated as a mortality rate, the same people lean away from it, even when the two figures are mathematically forced to add to one hundred per cent. The frame, not the arithmetic, is doing the deciding.

The virtuoso in the subway

Nir Eyal, in Hooked (2014), tells the story of Joshua Bell, one of the world's best violinists, busking incognito in a Washington DC Metro station during rush hour on a multi-million-dollar Stradivarius. Commuters walked past a virtuoso performance for the price of loose change, because the frame surrounding him said "subway busker," not "world-class soloist" (pp.87–88). Nothing about the playing changed between the concert hall and the platform. The label attached to the person did.

Same wine, a different frame

The wine-pricing research discussed in depth on the context effect, where Hilke Plassmann and colleagues showed that a higher stated price raised both reported pleasantness and measured brain activity for an identical wine, is framing operating at the level of a single number on a label. It belongs here as an example of the pattern, but the citation and the neuroscience live on that page rather than being duplicated on this one.

What it does not say

Framing describes a robust and heavily replicated pattern in judgement and choice, but it is not a claim that any frame moves any audience by the same amount. Effect sizes vary with how familiar people are with the material, how high the stakes feel, and how much time they have to think, and some studies find the effect shrinks or disappears under repeated exposure to the same choice. It is also not a claim that people are simply irrational. Treat it instead as evidence about what information the fast, automatic layer of judgement actually uses, which is usually the frame in front of it rather than every fact that could in principle be recovered from it.

It is worth distinguishing framing's well-established core, the gain-loss and survival-mortality effects Kahneman and Tversky documented, from its more troubling extensions into labelling people rather than statements. Adam Alter's Drunk Tank Pink extends a version of this argument to faces labelled by race or pupils labelled by expected ability, and that broader territory mixes a robust finding, the misinformation effect, with social-priming claims that did not survive replication. Framing-of-statements is solid ground. Not everything filed under the same broad heading is equally solid.

Who developed it

Daniel Kahneman and Amos Tversky formalised framing as part of prospect theory, their joint research programme on decision-making under uncertainty. Neither has a dedicated page on this site; both are referenced here by name because the underlying research, not a biography, is what the concept rests on.

---