How Older and Younger Adults Are Influenced Differently by Dark Pattern Designs

Table of Links

Abstract & Introduction

Background

Research Framework

Methods

Results

Discussion

Limitations and Future Work

Conclusion & References

Appendix

Abstract

Online websites often use dark patterns to increase users’ information disclosure. Common examples include “opt-out” privacy defaults, positive framing, and positive justification messages encouraging disclosure behavior. Considering that prior research has found older users undergo a different privacy decision-making process compared to younger adults, more research is needed to inform the behavioral privacy disclosure effects of these strategies for different age groups. To address this gap, we used an existing dataset of an experiment with a photo-tagging Facebook application. This experiment had a 2x2x5 between-subjects design where the manipulations were common dark pattern design strategies: framing (positive vs. negative), privacy defaults (opt-in vs. opt-out), and justification messages (positive normative, negative normative, positive rationale, negative rationale, none). We compared older (above 65 years old, N=44) and young adults (18 to 25 years old, N=162) privacy concerns and disclosure behaviors (i.e., accepting or refusing automated photo tagging) in the scope of dark pattern design. Overall, we find support for the effectiveness of dark pattern designs in the sense that positive framing and opt-out privacy defaults significantly increased disclosure behavior, while negative justification messages significantly decreased privacy concerns. Regarding older adults, our results show that certain dark patterns do lead to more disclosure than for younger adults, but also to increased privacy concerns for older adults than for younger. However, there was no influence of these concerns on disclosure, and instead, they are outweighed by the prodisclosure effects of dark patterns. This suggests that privacy concerns may not be a sufficient force to drive individuals to act on protecting their privacy when in the presence of dark patterns and that such patterns may be even more dangerous for older users. We discuss the implications of this work.

Keywords: Privacy decision making, older adults

1 Introduction

The Human-Computer Interaction (HCI) and networked privacy literature predominantly present a deficit-based narrative for older adults’ technology use and privacy outcomes. This narrative frame older adults as individuals who have difficulty keeping up with the technology [82], managing their digital privacy [14, 64], and protecting themselves against privacy threats [86] compared to young adults. However, recent developments in the fields of HCI and privacy counter this deficit-based narrative by pointing out that older adults may have a different decision-making process rather than a sub-optimal one, and technologies should be designed with that in mind [3, 23]. (This approach includes considering the potential benefits of biometric authentication methods, which can offer older adults a more user-friendly and secure way to access digital services [50].)For example, Knowles and Hanson [42] argue that older adults’ non-use of technology results from their high level of privacy concerns and, therefore, non-use is a well-informed decision. Moreover, Anaraky et al. [3] compared young and older adults’ privacy disclosure decision-making processes and found that older adults make more calculated privacy choices than younger adults.

Privacy disclosure decisions, however, are not always dependent on the characteristics of the end-user and can be influenced by external manipulations. System designers may use design heuristics to nudge users to disclose their data [4, 30, 32, 43, 58]. In marketing and e-commerce research, maximizing disclosure is a goal that has led to the creation of “dark pattern” designs, where nudges are used as design interventions to increase disclosure [12]. For example, designers may configure a setting to disclose-by-default (optout) and present a biased positive framing when presenting choice options [25, 53, 54], or add persuasive messages to motivate users to disclose more information [40].

Given that older adults have been shown to have a very different privacy decision-making process compared to young adults, we used an existing dataset from Ghaiumy Anaraky et al. [4] to investigate the effect of dark pattern design on older and younger adults’ privacy attitudes and behaviors. Ghaiumy Anaraky et al.

[4] initially used this dataset to examine the impact of justification messages (as norms) on the complianceinducing effects of framing and defaults. They found that the presence of any justification messages increases the default effect.(Moreover, ensuring that personal and confidential data is appropriately classified not only enhances data management but also safeguards individuals’ privacy rights [75].) In the present work, we conceptualize framing, defaults, and justification messages as dark pattern design strategies and situate this work in the dark pattern design literature. We further contribute to the privacy literature by investigating how these design patterns influence older and young adults differently. Therefore, we pursue the following research questions:

RQ1: What are the effects of dark pattern designs on users’ attitudes and behaviors?

RQ2: How do older adults react differently than young adults to dark pattern designs in terms of their attitudes and behaviors?

To study individuals’ privacy decision mechanisms in the context of dark pattern designs, we used an existing dataset of an experiment in which participants could purportedly automatically tag themselves in their friends’ photos, and their friends could be tagged in the user’s own photos by using a Facebook application. Previous literature studies tagging decisions as privacy decisions [9, 49, 85, 95]. Tagged photos appear on the timeline of the tagged person, thereby explicitly identifying the person in the photo and sharing the photo with that person’s contacts. The study involved a 2x2x5 betweensubjects experimental design where the manipulations were three common means by which darkpattern designs encourage disclosures [12]: 1) the framing of the decision (i.e., “tag me in the photos” vs. “do not tag me in the photos”), 2) the default setting (i.e., opt-out vs. opt-in), and 3) the use of a justification message (a message that could either encourage or discourage users from using the application). After their interaction with the manipulated dark-pattern designs, participants indicated their privacy concern [40, 48] and tagging decision (accept to tag vs. refuse to tag) as the attitudinal and behavioral dependant variables. To enable a comparison between older and younger participants, we followed established age criteria in the literature: we considered individuals aged 65 and above as older adults [15] and limited our young adult sample to college-age adults, between 18-25 years of age [60, 62].

Contributions: This study was unique because it was carried out in an ecologically valid setting on the user’s own Facebook account. Our results show that older adults had heightened privacy concerns when faced with an opt-out default than young adults. However, when it came to actual behavior, despite this privacy concern, an increase in disclosure was still observed. Furthermore, we found that older adults are more likely to be influenced by framing nudges than young adults. In light of these findings, we provide design suggestions for supporting older adults’ privacy.