Persuasion, Ethics & Dark Patterns

Every psychological lever in this library can inform or exploit. This file is the line between the two. Read whenever designing for conversion, retention, notifications, or pricing.

Cialdini's principles of influence (1984; +1 in 2016)

Reciprocity, commitment/consistency, social proof, authority, liking, scarcity, unity. Legitimate when the underlying facts are true (real scarcity, real reviews, real expertise); deceptive when manufactured.

Dark patterns (Brignull, 2010 — now "deceptive patterns")

Named types (Brignull's taxonomy at deceptive.design; expanded by Gray et al. 2018 and the OECD/FTC literature):

• Sneaking — sneak-into-basket, hidden costs, drip pricing.
• Urgency/scarcity (false) — fake countdowns, fake stock levels.
• Misdirection — visual prominence tricks; trick questions; confirmshaming ("No thanks, I hate saving money").
• Obstruction — "roach motel": easy in, hard out (cancellation mazes). Regulators act on this even without a dedicated rule — see the regulatory landscape below.
• Forced action — forced enrollment, privacy Zuckering (coerced oversharing), forced continuity (silent trial→paid conversion).
• Nagging — repeated interruptions to extract consent previously denied.
• Interface interference — preselected harmful defaults, disguised ads, hard-to-see opt-outs, unequal button weights on consent dialogs.

Prevalence & harm: a crawl of ~11,000 shopping sites found dark patterns on ~11% of them, concentrated among the most popular sites (Mathur et al. 2019); an EU Commission study (2022) found most popular sites/apps deploy at least one. Luguri & Strahilevitz (2021) showed dark patterns substantially increase sign-ups for unwanted services, with stronger effects on less-educated users — an equity problem, not just an annoyance. Deceptive patterns also erode long-term trust and retention and create growing legal exposure.

Regulatory landscape (as of mid-2026 — verify before relying on it)

• US federal: the FTC's 2024 "click-to-cancel" (Negative Option) Rule was vacated by the 8th Circuit in July 2025 on procedural grounds, days before its compliance deadline; in January 2026 the FTC began a fresh rulemaking (ANPRM stage), which will take years. Meanwhile the FTC still enforces via FTC Act §5 and ROSCA — e.g., the record $2.5 B Amazon Prime settlement (Sept 2025) over deceptive enrollment and obstructed cancellation, and the Epic Games/Fortnite dark-patterns settlement (2023).
• US states: California's AB 2863 (effective July 2025) requires same-medium, click-to-cancel subscription cancellation and express consent for renewals/free-trial conversions; CPRA (CA) and the Colorado Privacy Act declare consent obtained through dark patterns invalid.
• EU: DSA Art. 25 (in force) bans deceptive/manipulative interface design on online platforms; GDPR + the Unfair Commercial Practices Directive cover consent and commerce; a dedicated Digital Fairness Act targeting dark patterns, addictive design, and unfair personalization is in preparation (Commission proposal expected 2026).
• India: CCPA "Guidelines for Prevention and Regulation of Dark Patterns, 2023" enumerate 13 prohibited patterns (false urgency, basket sneaking, confirmshaming, forced action, subscription traps, drip pricing, etc.); a June 2025 advisory ordered platform self-audits. Design takeaway: cancellation as easy as sign-up, symmetric consent choices, and honest scarcity are becoming legal baselines across major markets, not just ethics.

Ethical tests to apply before shipping a persuasive pattern

1. Transparency test (Thaler/Sunstein): would it still work if the user fully understood what you're doing?
2. Regret test (Eyal): will users act, then regret it once informed?
3. Symmetry test: is leaving as easy as joining? Is opting out as visible as opting in?
4. The mom test / front-page test: comfortable explaining it to someone you love, or seeing it reported publicly?
5. Beneficiary test: who wins if it works — the user's goals, or only the metric?

Designing FOR user goals (positive persuasion)

Commitment devices users choose (deposits, scheduled sends, focus modes); honest friction on destructive/expensive actions; defaults set to the user-serving option; progress and streaks with forgiveness mechanics (streak freezes) rather than loss-threats; notifications that are user-configurable, bundled, and valuable enough that users would miss them.

Attention ethics

Time-on-site is not a proxy for value. Prefer metrics like task success, retention driven by outcome achievement, and "time well spent" framings (Center for Humane Technology). Autoplay, infinite scroll, and variable- reward notification schedules should be opt-in-able and interruptible; consider natural stopping points (chapter breaks, "you're all caught up" — a pattern Instagram itself shipped).

What the wellbeing evidence actually shows: large-scale analyses find the association between overall screen time and adolescent wellbeing is small (Orben & Przybylski 2019), and moderate use may be harmless or mildly positive ("Goldilocks hypothesis," Przybylski & Weinstein 2017) — so don't design around raw screen-time guilt. The defensible targets are specific harms: displacement of sleep, compulsive checking loops, and use the user themselves regrets (the regret test). Support user self-regulation: usage dashboards, session limits, notification bundling/scheduling, and respecting OS-level focus modes.

Sources

• Cialdini, R. (1984; 2021). Influence; (2016) Pre-Suasion. Harper.
• Brignull, H. (2010–). deceptive.design (dark-patterns taxonomy); Brignull (2023). Deceptive Patterns.
• Gray, C. M. et al. (2018). "The Dark (Patterns) Side of UX Design." Proc. CHI '18.
• Luguri, J. & Strahilevitz, L. (2021). "Shining a Light on Dark Patterns." Journal of Legal Analysis, 13(1).
• Mathur, A. et al. (2019). "Dark Patterns at Scale: Findings from a Crawl of 11K Shopping Websites." Proc. CSCW.
• OECD (2022). "Dark Commercial Patterns" report; EU Digital Services Act (2022), Art. 25; European Commission (2022) behavioural study on dark patterns.
• U.S.: FTC v. Amazon Prime settlement (W.D. Wash., Sept 2025); 8th Cir. vacatur of the FTC Negative Option Rule (July 2025); California AB 2863 (2024); India CCPA Dark Patterns Guidelines (2023).
• Orben, A. & Przybylski, A. (2019). Nature Human Behaviour, 3; Przybylski & Weinstein (2017). Psychological Science, 28(2).
• Thaler & Sunstein (2008). Nudge — transparency standard.