Research

Influencing User Decisions: Dilemmas in Designing Online Interfaces

Marie-Sophie Simon, Hanna Schraffenberger, Raphaël Gellert
NordiCHI’24
Abstract: The choices designers make about user interfaces can influence the choices end-users make when using those interfaces. These steering qualities of design have received much attention in the context of nudging and manipulative design (e.g., so-called ‘dark patterns’). This paper discusses ethical design practices and reviews the differences between nudging and manipulative design strategies, focusing on aspects that make them (un)acceptable. We show that existing distinctions are inconsistent and sometimes contradictory, often offering useful but partial or hard-to-follow advice. We illustrate these points with three concrete dilemmas that well-meaning designers face. Our review reveals that nudging and manipulative design both treat users as irrational and plagued by cognitive biases, which can be mitigated or exploited. We question this view and propose developing new interface design principles that treat users as rational and help them make their own deliberate autonomous decisions.

If Deceptive Patterns are the problem, are Fair Patterns the solution?

Tim de Jonge, Hanna Schraffenberger, Jorrit Geels, Jaap-Henk Hoepman, Marie-Sophie Simon, and Frederik Zuiderveen Borgesius
FAccT’25
Abstract: Researchers and legislators increasingly worry about deceptive patterns: common tricks on websites and in apps that make users do things they did not intend to do (previously: dark patterns). If these deceptive patterns are a problem, could “fair patterns” be the solution? We highlight several caveats to this approach. First, it is not obvious what it means for a design pattern to be fair. What is fair depends on the context and even within the same context, people disagree on what fairness means. Moreover, one fair design element does not guarantee a fair overall design. Combining these objections, it may be inappropriate to call a design pattern fair. Second, not all problems are adequately addressed by interventions at the design level. If all possible choices are unfair, design alone cannot make the situation fair. Societal problems must be solved at a societal scale, although design can contribute through incremental improvements. Progress in interface design does not need the concept of fairness: empirically informed solutions for specific problems appear more practical.

Judging a Tweet’s credibility — The effect of signature labels on perceived Tweet credibility

Radboud University 2022
Abstract: Since social media became, popular the world has been getting more connected and new kinds of communities have developed. However, by now, the negative aspects have also become more apparent than ever. Whether it’s cyberbullying, social media addiction or misinformation — social media’s negative aspects can not be overlooked. Specifically, the problem of misinformation has recently been given a lot of attention and the societal impact has become visible. Both the 2016 US presidential election and the current COVID-19 outbreak have been heavily influenced by misinformation. In this research, we look into the Twid-project as one of the options that are currently being developed to address the problem of misinformation. The goal of the Twid-project is to help Twitter users judge the credibility of Tweets. With Twid, a user posting a Tweet would have the possibility to sign their Tweets with verified attributes. Signing a Tweet with a relevant label would give the user the chance to provide reliable background information about themselves. This research tries to answer two questions. The first question is whether people perceive Tweets with a relevant attribute-based verification label as more credible in terms of message and account credibility than regular Tweets. Our second question is whether an active indication of a missing signature reduces the perceived message and account credibility compared to both a signed and a normal Tweet. Our results are promising and show that a relevant attribute-based signature indeed does increase the credibility of a Tweet with true information. We did not find any evidence that the active indication of a missing signature decreases credibility compared to a Tweet without any label. This indicates that users understand the meaning of these signature labels. In an exploratory analysis, we further find an indication that credibility and sharing behaviour are related to each other. With these results, we show that the Twid-project has the potential to be part of a solution in the fight against misinformation. We suggest directions for future work to further investigate the potential and understand the effects that are at play.