One sentence: Almost every decision users make in an interface is not “rational calculation,” but is carried along by a set of predictable mental shortcuts (heuristics) and cognitive biases. Persuasive Design works with these psychological laws to lower the friction of user action and guide them toward outcomes that both you and they want.
There is only one key dividing line — the Golden Rule: Persuasion must be built on “behaviors the user is already willing to accept” (helping them do what they want to do more smoothly), rather than tricking them into doing what they do not want to do. Once you cross the line by relying on misdirection, coercion, or pre-checked boxes to drive conversions, it becomes a dark pattern— short-term numbers look good, but trust collapses in the long run. For details on ethical red lines, see NNGroup Interaction Patterns · Ethics · Globalization · Emerging Technologies (15 patterns · deceptive patterns/sneaking/dangerous UX · inclusive design · global cross-cultural/language switching/sizes/omnichannel · VR-AR/GenAI prompt controls · user autonomy · UX copy length/recommended content).
These 26 articles can be grouped into eight blocks:Six Principles of Persuasion (Cialdini) → Cognitive Biases and Decision-Making → Innate Limits of Attention → How Users Find Their Way (Foraging/Scent/Cost) → Mental Models and Learning Cost → How Trust Is Built Layer by Layer → Driving Motivation with Three Core Psychological Needs → Privacy and Ethical Boundaries.
I. The Underlying Principles of Persuasion (Cialdini’s Six Principles of Influence)
Among the principles of influence summarized by social psychologist Robert Cialdini, these notes cover four. What they have in common: they are all “judgment shortcuts” for the brain — when information is insufficient or there is no time to think carefully, a simple rule replaces effortful rational analysis.Used correctly, they save users effort; used crookedly, they become manipulation.
| Principle EN/CN | Core idea (one sentence) | How to implement | Key numbers/experiments | Counterexamples (don’t do this) |
|---|---|---|---|---|
| Reciprocity | Give first, then ask — people are innately driven to return kindness | Content first (release part of a white paper before asking for a form); delay permission requests (ask for microphone permission only when the microphone is needed); take on the initial effort so users get a taste of the benefit first | comScore “fill out the form before giving content” vs. Nielsen “give content for free first”: giving content first → users submit more truthful information and build a deeper relationship; asking for the form first → higher short-term submission rate | Clear to-do app shoves 7 pages of tutorials on the opening screen; Expedia asks for location permission before any search; Evernote asks for push permission before the user has used it |
| Social Proof | Everyone else is doing it, so I’ll do it too — use group behavior to reduce one’s own uncertainty | User reviews (Adagio Teas); “people who viewed this also viewed” (eBags); like counts + usernames (Etsy); subscriber counts | McDonald’s 1958 ad “Over 100 million burgers sold”; long lines outside Apple stores = this phone is worth buying | Too little data becomes a burden instead (only 2 shares → makes the article look bad); piling on too many reviews/likes → cluttered pages, slower loading |
| Authority | What experts say is more credible — obeying authority is an effort-saving decision shortcut | Authority photos (white-coated doctors / suited lawyers); authority symbols (scales of justice); well-known institution logos; expert or celebrity endorsements | Bickman 1974: when making a request in a guard uniform,92% complied; when the same person wore casual clothes, compliance dropped sharply to 42%42%. Milgram experiment: 65%65% administered the highest electric shock under authority’s instruction | Fake/irrelevant authority (singer Bocelli endorsing math software); displaying client logos without authorization (violating nondisclosure agreements) |
| Commitment & Consistency | People want to stay consistent with their past behavior — small commitments first, then larger commitments | Make the initial commitment low-threshold, low-risk; let users put in effort first (sunk cost psychology); make commitments public to elevate them to a social level | Yelp: write a review first (no registration required) → ask for registration only after writing (users don’t want to waste what they’ve written) → then invite them to write another; Fitbit set a goal → visualize progress → share publicly with friends; 23andMe questionnaire progress bar + ranking | Forcing registration before any use right from the start (even if Google login reduces cost, it is still perceived as high risk before trust is built) |
Why these four work: They all originate from evolution — reciprocity let early humans survive through cooperation, social proof let you avoid the risk of “there’s a reason the whole village isn’t eating that mushroom,” authority lets children believe “don’t touch the hot stove,” and consistency makes you appear predictable and accepted in society. Understanding that “they are effort-saving shortcuts” makes their boundaries clear:The less information there is, the lower the risk, and the tighter the time, the more shortcuts dominate; once users enter high-stakes scenarios where they have time to think carefully (large financial decisions), shortcuts give way to rational analysis.
II. Cognitive Biases and Decision-Making Mechanisms (How Users Make Choices)
The “irrationality” users show when facing options actually follows stable patterns. Once you understand these patterns, you can present options in a better way (note: this means changing the presentation frame, not changing the facts).
| Bias/Mechanism EN/CN | Core idea | How to use it in design | Key numbers/experiments |
|---|---|---|---|
| Prospect Theory & Loss Aversion | The pain of losing > the pleasure of gaining(for equal amounts) | ① Use certain small rewards(10% off coupon at checkout) over uncertain large rewards(raffle to win $1000); ② emphasize “what you lose by not acting” (insurance emphasizes potential high expenses); ③ framing: “95% satisfaction” is more persuasive than “5% complaint rate” | Kahneman & Tversky 1979. Certainty bias: most people choose a certain $900 over a 90% chance of $1000; but when facing losses, they instead take risks (isolation effect: the same equivalent choice, with different starting points → opposite choices) |
| Anchoring | The first number hijacks subsequent judgments, even if it is completely irrelevant | ① Provide good defaults/suggested values(mortgage calculators, gamma value 2.2); ② at the start of a process state the expected time clearly(be accurate or slightly conservative, otherwise it feels deceptive); ③ in e-commerce,crossed-out original price + discounted price raises perceived value | 5-second estimate of 1×2×…×8: median 512; reverse order 8×7×…×1: median 2250(true value is 40320 in both cases). Using the last two digits of SSN as an anchor: low-digit group bid $16.09, high-digit group $55.64. Oxfam A/B test: adding suggested donation amounts → form completion rate +23% |
| Compensatory vs Noncompensatory Decision-Making | When there are many options → first use 'veto rules' to eliminate; when options are few → then weigh pros and cons | Noncompensatory = filters / faceted navigation(set hard criteria to exclude unqualified options);Compensatory = comparison tables(strengths offset weaknesses; the option with the highest overall score wins); in reality the two stages are mixed | Compensatory decision-making is suitable for roughly 5–7 options; comparison tools work best when supporting fewer than 5 options. Wayfair used filters to quickly narrow down 20,903 sofa options |
| Satisficing | Users stop when they find something "good enough," rather than searching for the "best" option | Inverted-pyramid writing (key points first) + descriptive headings + key-point summaries (Investopedia);sorting features let users quickly lock in options by the dimension they care about most (price/rating/time) | satisfice = satisfy + suffice. Every action has an interaction cost; users subconsciously weigh the benefit of "looking for something better" against the cost, and usually choose to stop |
| Priming | Things seen first unconsciously influence subsequent behavior, even if they seem unrelated | Pay attention to the "cues" from interface elements: a "promo code" field on the checkout page can prompt users to leave the site to look for coupons → interrupt purchase → churn | Pottery Barn's "Promotion Code" field induced users to leave the checkout flow. In usability testing, the wording "keypad" made all users search for "keypad" instead of "keyboard," misleading the design team |
| Functional Fixedness | People habitually use objects according to their 'traditional purpose', which stifles innovation | Three ways to break it: ① Abstract the problem(strip away surface details to grasp the core, draw inspiration from distant fields such as jujitsu/gardening/woodworking); ② Borrow ideas across domains;③ Delay judgment(don't dismiss ideas too early in brainstorming) | Classic candle problem:5-year-old children were faster than adults/older children to think of using the thumbtack box as a stand — the more experience people have, the more they are trapped by the idea that "the box is for holding tacks." Positive example: InVision shows a download box directly when right-clicking the logo, breaking the convention that "right-click = properties menu" |
Why: These are all the brain's "energy-saving modes." Loss aversion originates from survival (losses can be fatal, gains are just icing on the cake); anchoring and satisficing are products of "bounded rationality" — the brain has limited computing power and can only use approximations; functional fixedness is a side effect of experience (the more skilled you are, the more fixed your patterns become).Design implications: You are changing the "frame of information presentation," not the facts themselves — rephrasing the same true information or changing its reference point changes users' choices. This is exactly where its power lies, and exactly where the ethical risk lies.
III. Innate Limits of Attention (Users Simply Don't See It)
Usability testing has repeatedly proven: users will look right at things clearly displayed on the screen and not see them. This is not because users are unintelligent; it is how the attention mechanism works. Design must work with it, rather than assuming "if I put it there, users will see it."
| Phenomenon EN/CN | Core principle | How to respond | Examples |
|---|---|---|---|
| Change Blindness | When a change occurs outside the focus of attention, users do not notice it — even if the change is obvious | ① Reduce competing simultaneous changes; ② Concentrate related changes in the same area; ③ Use (single, concise) animation to guide the eye; ④ Use high contrast/shadows to highlight the changed area; ⑤ Display new elements near the user's current operation area | Psychology experiment: when a blank screen is inserted between two images, many people need to view them 20–40 times before spotting the difference (the "flicker effect"). After clicking the hamburger menu, users only stare at the expanded area, so the replacement of the search icon in the top-right corner is ignored; the navigation bar that appears when scrolling up on NYTimes is too inconspicuous and gets ignored |
| Tunnel Vision / Selective Attention | Users only stare at areas they believe are relevant to the task and filter out everything else | ① Place key information and instructions together(within the user's primary line of sight); ② Verify with user testing (designers know what is important; users may not see it that way) | Westfield timeline: the year "2000" was displayed in large text in the top-left corner of a pop-up, but users completely ignored it and only found it after reading the entire paragraph of text.Banner Blindness: Users actively ignore elements that look like ads — so don't style critical information to look like ads |
| Chunking | Package information into meaningful small chunks so the brain can scan and remember it | Short paragraphs + white space; 50–75 characters per line; clear visual hierarchy to group related information; segment number strings (phone/card numbers); law of proximity (align related content close together) | George Miller, 1956: short-term memory holds roughly 7±2 chunks; but what matters is the number of chunks, not the volume of content(you can remember 7 letters, or 28 letters grouped into 7 four-letter words).Common misconception: don't rigidly limit navigation to ≤7 items because of this — menus rely on recognition, not recall, and 21 items (hp.com) work fine when the structure is clear. Subsequent research found short-term capacity actually fluctuates between 3–6 |
Why: Attention is a scarce resource; the brain must filter out the vast majority of stimuli through "selective attention" in order to focus on completing tasks — this is a feature, not a bug.Hard constraint for design: You cannot guarantee users will see something just by "putting it there" or "making it bigger"; you must reduce competition, align with focus, and verify visibility with real user testing. This section and NNGroup Usability Testing and the Ten Usability Heuristics (Complete Table of 10 Heuristics · Application to Complex Applications · User Pleasure Hierarchy · Testing Methods: 5-User Rule/Recruitment/Remote/Task Scenarios/Step-by-Step Tasks/Attitude vs Behavior/4-Step Analysis/Small-Sample Error/Competitive Evaluation) share the same underlying logic as "recognition over recall" and "aesthetic and minimalist design (signal-to-noise ratio)."
IV. How Users Find Their Way in Products (Foraging / Scent / Cost)
These three articles form a linked set of theories explaining why users click here instead of there, and when they give up — at the core, "users are subconsciously doing a calculation of benefit ÷ cost."
| Theory EN/CN | Core principle | Formula/key concepts | Design points |
|---|---|---|---|
| Information Foraging | Users are like foraging animals: they move between "information patches," staying where the yield is high and moving on when it is low | Information gain rate = information value ÷ cost of acquisition. Patch = information source, scent = attractiveness, enrichment = optimization strategy (F-pattern scanning/layer-cake scanning/keyword retrieval) | Optimize the page's main task rather than trying to satisfy everyone; clear headings/keywords/lists reduce cost; good design can make "enrichment" unnecessary (when search engines rank results well, users don't need to modify their queries) |
| Information Scent | Users judge whether a link is worth clicking based on the "scent they smell from afar" (cues) | Scent is determined by four parts:link label (most important) + text/thumbnails around the link + context of the page it is on + the user's own background knowledge | Labels should be concise and accurate, avoid jargon; summaries should supplement details; images should be descriptive rather than decorative; the same "Christmas" link has strong scent on Williams Sonoma (surrounded by tableware images) and weak scent on Harry & David (surrounded by fruit) |
| Interaction Cost | All the effort users expend to achieve a goal All the effort users expend to achieve a goal(read/scroll/search/think/click/type/wait/switch attention/remember) | Expected utility = expected benefit − expected interaction cost; the ideal is zero cost | Reduce operation steps, waiting, and attention switches; clear layout lowers the cost of "finding." If interaction cost is too high and the benefit is not particularly unique → users will simply switch to another site |
Why: This set of theories translates "usability" into a calculable ledger.The most practical point for PMs: You don't need to wait for launch to look at data; simply doing an "interaction cost analysis" (listing every step of effort to complete a task) lets you predict in advance where drop-off will happen — in the Dictionary.com etymology lookup example, "scanning irrelevant search suggestions" and "scrolling horizontally and not finding the 'etymology' tab" are both quantifiable, optimizable cost points.Cisco using NGFW/AnyConnect/Duo as link labels is a classic case of "scent failure" — for people who don't understand the terms, there is zero scent, and the links are skipped outright.
V. Mental Models and Learning Cost (Why "Innovation" Often Flops)
| Topic EN/CN | Core principle | Key points | Examples |
|---|---|---|---|
| Mental Models | Users hold a set of expectations about how a system works, based on past experience; when system behavior does not match → confusion, disorientation, blaming the product | Jakob's Law: Users expect your site to behave the same as other sites they have used. Models have inertia, and even outdated models still influence behavior. Designers' models are far deeper than users', so they overestimate what is "obvious" | Shopping carts should be under user control (Nomad Lane automatically added shipping insurance that had to be manually canceled → violated the model); the browser back button is understood as "undo/go back to previous page" (but on mobile it often means "navigate up" → confusion); multiple search boxes make users choose the wrong one (University of Michigan Library) |
| Skeuomorphism | Borrow elements from the physical world to help users transfer existing models to new interactions | Suitable for groups that have not yet built digital models | The children's app Duo ABC uses metaphors of doors and maps to help children understand operations |
| Power Law of Learning | Time to complete a task declines according to a power law with repetition, dropping quickly at first then reaching a plateau;a new interface must compete with old interfaces that users have already reached a plateau on | Response strategies: follow the user model (align with expectations, use card sorting to map navigation habits) or improve the user model (clear labels + explanations to educate users) | Learning curve comparison: Interface A is fast to learn but has low ultimate efficiency; Interface B is slow to learn but has the highest efficiency —choose B for high-frequency use (enterprise employee directory), choose A for low-frequency use (annual tax filing). "Repeating 1 time" on a new interface ≈ "repeating 5 times" on an old interface, so users are extremely reluctant to switch |
Is innovation worth doing? (Decision tree): ① Is the long-term benefit significantly higher than the old design? ② Do users have sufficient motivation to practice repeatedly until proficient? ③ Can you accelerate reaching the plateau through high-frequency use or easier learnability? Proceed only if all three hold; innovation is easily absorbed when there are loyal users (Apple can push mandatory learning), high-frequency use, or an industry-standard shift (iOS swipe-to-delete, hamburger menu).
Why: Consistency is not designers being lazy; it saves users the real cost of "climbing the learning curve from zero." This echoes "commitment and consistency (Section 1)" and "negativity bias (Section 7; moving the logo to the center made returning to the homepage 6 times harder)" —deviating from users' existing models = directly raising interaction cost + triggering negative memories.
VI. How Trust Is Built Layer by Layer (Credibility Engineering)
Trust is the prerequisite for conversion. These three articles piece together the full picture from three angles: "the ladder of trust," "sources of credibility," and "how users form first impressions."
① The Trust Pyramid: 5 Experience Levels (Hierarchy of Trust) — analogous to Maslow,don't ask for higher-level needs before lower-level needs are met:
| Level (from low to high) | What the user is asking | What design must provide |
|---|---|---|
| 1 Basic relevance and trust | Can this site help me achieve my goal? Is the content credible? | Clear descriptive labels, intuitive experience, external recommendations/reviews |
| 2 Interest and preference | Why should I choose this over others? | Differentiated value (unique features/offers) |
| 3 Trust with personal information | Is it worth registering and giving my information to this site? | Transparent privacy policy, don't ask for sensitive information too early |
| 4 Trust with sensitive/financial information | Do I dare give it my credit card/address? | Security indicators such as SSL, trusted payment methods |
| 5 Willingness to build an ongoing relationship | Am I willing to subscribe/commit long-term? | Clear service value, pressure-free opt-out options |
Counterexamples: Threadup asked users to register (Level 3) when they were still at Level 1; Nextdoor asked for a detailed address + email on first visit (Level 4).Fix: First satisfy Levels 1–2 with reviews/social proof, let users search the community by zip code to judge relevance, then guide them to join.Giving users a sense of control(free browsing/searching/trialing) itself enhances trust.
② Four Elements of Credibility (Trustworthiness, Nielsen 1999, still valid today):
| Element | Why it matters | Key practices |
|---|---|---|
| Design Quality | A professional, legitimate appearance is the first step of trust | Clear navigation labels (do not use "creative" categories); visuals should match brand tone (Singaporean research: cleaning companies appear more credible when using green colors and white space);Typos and broken links instantly destroy trust("They didn't even get this right") |
| Upfront Disclosure | Hiding information signals lack of good faith | Clearly state contact information, fees, return policies, and shipping costs; do not require cumbersome forms to get a quote;Avoid login walls(honestbee required users to enter an address before viewing content → triggered aversion) |
| Content should be comprehensive, accurate, and timely | Comprehensiveness signals professionalism; accuracy and timeliness signal commitment to customers | Provide relevant advice and tips; show the entire process(a cleaning company should show photos of cleaning in progress, not just finished results); cover all services (only showcasing commercial building cases makes users doubt whether you serve small residential properties) |
| Connect to the rest of the web | Isolated websites appear unpolished or as if they are hiding something | Maintain a presence on third-party review platforms and social media;External independent reviews are trusted more than self-promotional testimonials on your own site |
③ Prominence-Interpretation Theory —— Credibility judgment follows two steps: users first notice(prominence) a certain element, then make an interpretation(interpretation: positive or negative). Prominence is affected by engagement, website topic, task, experience, and individual differences; interpretation is affected by cultural assumptions, skills, and context.
Case study: MoneyStrands placed a "spy" image on its early homepage, which did not match expectations for a financial service → negative interpretation → reduced credibility. Although SaveTheChildren's photos of children convey anxiety (a negative emotion), they match donor expectations → they actually increase credibility.Testing method:Five-second test(show the page for a few seconds then ask users what they remember → what they remember is the prominent element; do not tell users in advance that the page will disappear) + combine with usability testing to explore gaps between "expectations vs. reality".
Why: Trust is a sequential, cumulative process, not an on-off switch. View it in connection with "commitment and consistency" and "reciprocity" from the first section ——First provide value (reciprocity) → satisfy lower-level trust → start with low-threshold commitments → raise requests only after trust grows, this forms a complete, ethical conversion design mainline. The authority principle (from the first section) is also one of the sources of credibility.
VII. Drive Motivation with Three Core Psychological Needs (Self-Determination Theory + Emotion)
The previous sections covered "how to guide a single behavior"; this section covers "how to get users to keep using the product sustainably and enjoyably" —— by satisfying deep psychological needs, not through manipulation.
Self-Determination Theory (Ryan & Deci): People are driven by three basic psychological needs; satisfying them → increased motivation and well-being → increased stickiness.
| Need EN/CN | Meaning | Good design | Bad design |
|---|---|---|---|
| Autonomy | Use the product according to one's own preferences, pace, and order, rather than being forced into a single path | Duolingo/Khan Academy let users choose their own learning path; Grammarly allows skipping personalization on first use | Justanswer forces input only in a chat box, with no categories/FAQ; Capella forces users to fill in personal information to view program details ("cattle chute" style) |
| Relatedness | Feeling understood and supported by others | Well-timed, contextually appropriate messages (Apple Watch workout encouragement); support for content sharing/collaboration | Peel repeatedly pushes promotions right after a user buys a phone case → damages trust |
| Competence | Feeling effective and efficient when completing tasks | Pull-based contextual tips are better than lengthy tutorials: Otter uses layered tips (best practice: teach the calendar feature only after the user first taps "My Schedule") | FileZilla provides no beginner guidance, forcing users to figure it out on their own |
Three specific methods to increase autonomy (covered in a separate article):① ① Customizations(Messenger chat theme colors / Drive list vs. icon view / keyboard shortcuts — but most people don't bother customizing, so invest cautiously); ② Scannability(Morning Brew uses bolded headlines + emoji to let users quickly judge relevance); ③ Timing & Sequences(let users choose when and in what order to interact, e.g., tutorials can be skipped). Two notes:Control the number of options(too many options overwhelm people; simple > complex),provide meaningful defaults(weather defaults to current location but allows changes). The three needs sometimes conflict (wizards improve competence but limit autonomy, e.g., TurboTax) → use user testing to set priorities.
The three needs are not isolated: Open-world games (free exploration = autonomy, multiplayer cooperation = relatedness, easy to learn but with depth = competence) and collaborative design tools are models that satisfy all three. Autonomy is also directly tied to the California privacy law definition of dark patterns — the law defines a dark pattern as "an interface that materially impairs user autonomy, decision-making, or choice" (see details in NNGroup Interaction Patterns · Ethics · Globalization · Emerging Technologies (15 Patterns · Deceptive Patterns/Stealth/Dangerous UX · Inclusive Design · Global Cross-Culture/Language Switching/Sizing/Omnichannel · VR-AR/GenAI Prompt Controls · User Autonomy · UX Copy Length/Recommended Content)).
Negativity Bias —— a hard constraint for emotional design:
| Key points | Content |
|---|---|
| What it is | People pay far more attention to and remember negative experiences far better than positive ones: the pain of losing $20 > the joy of finding $20; one criticism outweighs many compliments |
| In UX | One small mistake (typo, lag) is remembered more than multiple strengths; when the interface meets expectations users stay silent, when it falls short they become critical;Moving the logo from the conventional left position to the center makes returning to the homepage 6 times harder |
| Six response tactics | ① Follow design standards (reduce cognitive load); ② Match workflows to user expectations (it is better to have a longer process than to violate expectations); ③ Use microcopy to proactively soothe concerns (Gitman emphasizes custom quality to ease frustration with long waits); ④ Write friendly error messages(counterexample: Four Seasons error prompts are stiff and lack empathy); ⑤ Sprinkle in small delightful surprises(MailChimp's post-send celebration feedback, TurboTax's low-pressure wording) — but these can only enhance an already good experience, they cannot fix fundamental usability flaws; ⑥ Test continuously |
Why: Negativity bias originates from survival instinct (being sensitive to danger signals helps you survive longer). Its implication for design is uncompromising —"Usable" is only the passing grade to avoid complaints; to satisfy users you must achieve "excellence"; and one unfixed small flaw can drag down the overall impression. That is exactly why Section V emphasizes consistency and this section emphasizes removing pain points before adding delights. (For the four-level pyramid of delight, see NNGroup Usability Testing and the Ten Usability Heuristics (Full Table of 10 Heuristics · Application to Complex Applications · User Delight Hierarchy · Testing Methods 5-User Rule/Recruitment/Remote/Task Scenarios/Step-by-Step Tasks/Attitude vs. Behavior/4-Step Analysis/Small Sample Error/Competitive Evaluation). )
VIII. Privacy and Ethical Boundaries: Creepiness–Convenience Tradeoff + Dark Pattern Red Lines
There is a line between persuasion and manipulation; this section directly addresses "how much privacy users are willing to trade for convenience".
Creepiness–Convenience Tradeoff: When users accept privacy-intrusive technology, they are conducting a cost-benefit analysis of "privacy loss vs. convenience gain".
| Dimension | Key points |
|---|---|
| Changes over time | Gradual acceptance: initial rejection (turn off Google Maps location) → discover convenience → try it → accept and become dependent. Exposure and familiarity reduce discomfort |
| Acceptance spectrum (bimodal distribution) | Digital Voyagers: want convenience, willing to share data, tolerate risk;Digital Pragmatists: low tolerance for fraud, more cautious; there is also a group that will never accept certain types of technology (e.g., smart home devices) |
| Cultural differences | Collectivist cultures (e.g., China) have higher privacy tolerance and lean toward convenience (they accept smart home devices after choosing a "trusted big brand"); individualist cultures (e.g., US/Canada) are more wary of "listening" devices |
| Design applications | Progressive data requests(do not ask for a bunch of information at the start; obtain it gradually as the relationship builds); for pragmatists,communicate transparently about data use and protection measures; make the convenience large enough to overcome privacy concerns |
Dark pattern red lines (ethical wrap-up running through the whole piece): The authority principle article directly states the golden rule —"The persuasiveness of design should be built on behaviors users are willing to accept, not on coercive misdirection." The California Privacy Rights Act (CPRA) defines a dark pattern as an interface that "materially impairs user autonomy, decision-making ability, or choice." Therefore, every technique in this article carries an implicit self-check:Am I helping the user do what they want to do more smoothly, or tricking them into doing what they don't want to do? The former is persuasion, the latter is manipulation — for a full discussion of ethical boundaries, globalization, and emerging patterns see NNGroup Interaction Patterns · Ethics · Globalization · Emerging Technologies (15 Patterns · Deceptive Patterns/Stealth/Dangerous UX · Inclusive Design · Global Cross-Culture/Language Switching/Sizing/Omnichannel · VR-AR/GenAI Prompt Controls · User Autonomy · UX Copy Length/Recommended Content).
Sourced from NNGroup Topics / Persuasive Design — Comprehensible Design (26 articles). Note: Although the original folder name contains "Comprehensible Design", the actual content is Persuasive Design/Behavioral Psychology, reorganized according to the persuasion and behavior framework. Credibility, delight hierarchy, recognition vs. recall, etc. overlap with NNGroup Usability Testing and the Ten Usability Heuristics (Full Table of 10 Heuristics · Application to Complex Applications · User Delight Hierarchy · Testing Methods 5-User Rule/Recruitment/Remote/Task Scenarios/Step-by-Step Tasks/Attitude vs. Behavior/4-Step Analysis/Small Sample Error/Competitive Evaluation); microcopy/error message wording is covered in NNGroup Content Strategy and Web Writing (4 Stages of Content Strategy · Content Inventory and Audit · Content Standards/Design Systems · 3 Content Management Models · How People Read Online 5 Scanning Patterns · Inverted Pyramid · 7 Concise Writing Methods · 5 Long-Form Layouts · Microcopy 3I/3C · 4 Tone Dimensions · Content Testing · Eye Tracking); dark pattern ethical red lines and globalization are covered in NNGroup Interaction Patterns · Ethics · Globalization · Emerging Technologies (15 Patterns · Deceptive Patterns/Stealth/Dangerous UX · Inclusive Design · Global Cross-Culture/Language Switching/Sizing/Omnichannel · VR-AR/GenAI Prompt Controls · User Autonomy · UX Copy Length/Recommended Content). Images/videos/original files are retained in sources.
Sources and related materials
- NNGroup Interaction Patterns · Ethics · Globalization · Emerging Technologies (15 Patterns · Deceptive Patterns/Stealth/Dangerous UX · Inclusive Design · Global Cross-Culture/Language Switching/Sizing/Omnichannel · VR-AR/GenAI Prompt Controls · User Autonomy · UX Copy Length/Recommended Content)
- NNGroup Content Strategy and Web Writing (4 Stages of Content Strategy · Content Inventory and Audit · Content Standards/Design Systems · 3 Content Management Models · How People Read Online 5 Scanning Patterns · Inverted Pyramid · 7 Concise Writing Methods · 5 Long-Form Layouts · Microcopy 3I/3C · 4 Tone Dimensions · Content Testing · Eye Tracking)
- NNGroup Usability Testing and the Ten Usability Heuristics (Full Table of 10 Heuristics · Application to Complex Applications · User Delight Hierarchy · Testing Methods 5-User Rule/Recruitment/Remote/Task Scenarios/Step-by-Step Tasks/Attitude vs. Behavior/4-Step Analysis/Small Sample Error/Competitive Evaluation)