Behavioral Advertising Map

← Decoder · The Hook

The Hook Goal: data Documented finding — Supported by a specific experimental or empirical result.

Confirmshaming & trick questions

also: guilt opt-out · double negative · manipulative decline link

The decline option is worded to shame you (‘No thanks, I don’t want to save money’) or phrased as a confusing double negative, so the easy, dignified path is to consent. It exploits the friction of re-reading and the social sting of the framing.

Vulnerabilities it exploits

  • Loss aversion

    Robust evidence strength: Well-replicated or backed by strong primary evidence.

    Kahneman & Tversky, “Prospect Theory: An Analysis of Decision under Risk,” Econometrica 47 (1979)

    The value function is steeper for losses than gains. The popular “2:1” ratio is a later, more precise estimate (loss-aversion coefficient ≈ 2.25; Tversky & Kahneman, 1992) — it is not stated as an exact figure in the 1979 paper.

  • Social proof

    Mixed evidence strength: Cited, but of uneven or debated replication strength.

    Cialdini, Influence (1984); roots in Asch (1956) conformity studies

    People look to others’ behavior to decide their own, especially under uncertainty. Robust in some settings (conformity, descriptive norms) but effect sizes vary widely by context; “X people are viewing this” style cues blur genuine social proof with fabricated urgency.

The evidence

What we actually know

Luguri & Strahilevitz (2021) Robust evidence strength: Well-replicated or backed by strong primary evidence.

“Shining a Light on Dark Patterns”

Journal of Legal Analysis 13, no. 1

A double-negative ‘trick question’ raised acceptance of an unwanted program from 19.2% to 33.4%, and only 16.7% of those enrolled later realized they had signed up. Less-educated participants were significantly more vulnerable.

Gray, Kou, Battles, Hoggatt & Toombs (2018) Robust evidence strength: Well-replicated or backed by strong primary evidence.

“The Dark (Patterns) Side of UX Design”

Proceedings of the 2018 CHI Conference

Established the most-used qualitative taxonomy of five categories — nagging, obstruction, sneaking, interface interference, and forced action — under which confirmshaming falls.

In the wild

  • A newsletter pop-up whose only decline button reads ‘No, I prefer to stay uninformed.’

    Pattern coined by Harry Brignull; catalogued at deceptive.design.

  • An opt-out checkbox phrased so that un-checking it is required to NOT receive marketing — a deliberate double negative.

The antidotes

What helps

Evidence-backed

  • Neutral, symmetrical choice wording is the design fix; under CPRA, consent obtained via a dark pattern is not valid consent.

Practical / common-sense

  • Slow down and re-read any consent toggle with a double negative; the manipulation lives in the wording.

Where the law stands

Different rules in different places

The rules are not the same everywhere — and they move. We show each jurisdiction separately rather than implying one global rulebook.

  • EUIn force

    Digital Services Act, Article 25 (+ EDPB Guidelines 03/2022)

    Manipulative interface wording that impairs free, informed decisions is prohibited for online platforms.

  • USIn force

    California CPRA/CCPA

    Consent obtained through a dark pattern is explicitly not valid consent.