The Dark (Patterns) Side of UX Design
8 likes35,634 views
The document discusses the ethical concerns surrounding dark patterns in user experience (UX) design, highlighting the need for increased awareness and engagement with ethics in HCI. It outlines a typology of dark patterns, categorized into five strategies, and stresses the importance of addressing user interests, designer intent, and potential manipulations in technology. The authors advocate for enhanced ethical practices in UX pedagogy and the need for a better understanding of how design impacts user agency.
The Dark (Patterns) Side of UX Design
- 1. COLIN M. GRAY, YUBO KOU, BRYAN BATTLES, JOSEPH HOGGATT, & AUSTIN L. TOOMBS PURDUE UNIVERSITY The Dark (Patterns) Side OF UX DESIGN
- 2. Increasing interest in critical/ ethical aspects of HCI & UX
- 3. Our Concern Ethics-focused methods are frequently bound to academia, making practitioner access to these conversations difficult, and activation of their implications problematic.
- 5. Harry Brignull UX practitioner “A user interface that has been carefully crafted to trick users into doing things…they are not mistakes, they are carefully crafted with a solid understanding of human psychology, and they do not have the user’s interests in mind”
- 7. EXPANDINGUPONBRIGNULL Which user or stakeholder interests are or should be kept in mind? What is the user being “tricked” into doing, and with what motivation? Are there instances where being tricked into doing something is desired by the user? Can interactions not designed to trick the user later become dark patterns through tech infrastructure changes?
- 8. EXPANDINGUPONBRIGNULL Which user or stakeholder interests are or should be kept in mind? What is the user being “tricked” into doing, and with what motivation? Are there instances where being tricked into doing something is desired by the user? Can interactions not designed to trick the user later become dark patterns through tech infrastructure changes?
- 9. EXPANDINGUPONBRIGNULL Which user or stakeholder interests are or should be kept in mind? What is the user being “tricked” into doing, and with what motivation? Are there instances where being tricked into doing something is desired by the user? Can interactions not designed to trick the user later become dark patterns through tech infrastructure changes?
- 10. EXPANDINGUPONBRIGNULL Which user or stakeholder interests are or should be kept in mind? What is the user being “tricked” into doing, and with what motivation? Are there instances where being tricked into doing something is desired by the user? Can interactions not designed to trick the user later become dark patterns through tech infrastructure changes?
- 11. Aims Connect Brignull’s typology more strongly to existing literature on ethics and values Create more tractable dark pattern categories for practitioner use & interrogation 1 2
- 12. Aims Connect Brignull’s typology more strongly to existing literature on ethics and values Create more tractable dark pattern categories for practitioner use & interrogation 1 2
- 13. Corpus Generation Collecting dark patterns exemplars.
- 14. CORPUS GENERATION Search Process Researcher 2 CS Background API Bot Crawling #darkpatterns Researcher 1 UX Background Commercial Sites BlogsSearch Engines Social Media Corpus of “dark patterns” related artifacts
- 15. 118 artifacts collected from social media outlets CORPUS GENERATION 45 from practitioner blogs40 from news outlets19 from personal product interactions10 from darkpatterns.org4
- 16. Corpus Analysis Coding and categorizing the collected artifacts.
- 17. CORPUS ANALYSIS Open Coding Approach Brignull’s Categories + Context Interaction Quality Intended User Group = Hierarchical Themes Reflecting Designer Strateg(ies)
- 18. Five Dark Strategies Appearing to serve as strategic motivators in practice. NAGGING OBSTRUCTION SNEAKING INTERFACE INTERFERENCE FORCED ACTION
- 19. 1 Nagging A minor redirection of expected functionality that may persist over one or more interactions. DARK STRATEGIES
- 20. 1 Apple: iCloud Storage NAGGING EXAMPLE
- 21. 2 Obstruction Impeding a task flow, making an interaction more difficult than it needs to be, with intent to dissuade an action. Brignull’s “Roach Motel”SUBTYPES Brignull’s “Price Comparison Prevention” Intermediate Currency DARK STRATEGIES
- 22. 2 Zynga: Unsubscribe OBSTRUCTION EXAMPLE
- 23. 3 Sneaking Hiding, disguising, or delaying the divulging of relevant information. Often intended to force uninformed decisions. Brignull’s “Bait and Switch” Brignull’s “Forced Continuity”SUBTYPES Brignull’s “Hidden Costs” Brignull’s “Sneak into Basket” DARK STRATEGIES
- 24. 3 Sleep Cycle: “Free Trial” SNEAKING EXAMPLE
- 25. 4 Interface Interference Privileging specific actions over others, thereby confusing the user or limiting discoverability of important actions. Brignull’s “Trick Questions” Toying with EmotionSUBTYPES False Hierarchy Brignull’s “Disguised Ad” Preselection Aesthetic Manipulation Hidden Information DARK STRATEGIES
- 26. 4 Two Dots: Muscle Memory INTERFACE INTERFERENCE EXAMPLE
- 27. 5 Forced Action Requiring users to perform a specific action to access (or continue to access) specific functionality. Social PyramidSUBTYPES Brignull’s “Privacy Zuckering” Gamification DARK STRATEGIES
- 28. 5 Instagram: No Option for “No” FORCED ACTION EXAMPLE
- 30. Qualities & Breadth of Potential “Dark”ness Reduction in user agency or action possibilities Persuasion or dark patterns? How to balance user and shareholder value?
- 31. Design Responsibility & Design Character When does a pattern become dark? Ethics codes generally conflate designer intent and eventual use Opportunities for ethical design leadership
- 32. Implications & Future Work Moving forward.
- 33. UX Pedagogy & Practice Increased ethical awareness in design and UX education Documentation of practice-based ethical dilemmas
- 34. Criticality & HCI Further integration and expansion of value-centered methods in UX Focus on applied, pragmatist views of ethics in HCI research and practice
- 35. THANK YOU COLIN M. GRAY colingray.me | uxp2.com gray42@purdue.edu This research was funded in part by National Science Foundation Grant No. #1657310