Privacy notices and consent forms are the means of conveying privacy policy information to users. In Europe, a valid consent needs to be confirmed by a clear affirmative action. Despite previous research, it is not yet clear whether user engagement with consent forms via different types of interactions for confirming consent may play a significant role in effectively drawing user attention to the content, even after repeated exposure. We investigate, in a laboratory study, how different types of interactions that engage users with consent forms differ in terms of their effectiveness, efficiency, and user satisfaction. In addition, we examine if and how habituation affects user attention and satisfaction, and the time they spend on giving their consent. We conducted a controlled experiment with 80 participants in four different groups where people either were engaged actively with the policy content via Drag and Drop (DAD), Swipe, or Checkboxes, or were not actively engaged with the content (as the control condition) in a first-exposure phase and in a habituation phase. We measured user attention to consent forms along multiple dimensions, including direct, objective measurements and indirect, self-reported measures. Our results show that the different types of interactions may affect user attention to certain parts of policy information. In particular, the DAD action results in significantly more user attention to the data items compared to other groups. However, with repeated exposure to consent forms, the difference disappears. We conclude that user engagement with policy content needs to be designed with care, so that attention to substantial policy information is increased and not negatively affected.

中文翻译：

---

用户参与隐私声明的困境

隐私声明和同意书是向用户传达隐私政策信息的手段。在欧洲，有效的同意需要通过明确的平权行动来确认。尽管先前进行了研究，但尚不清楚用户通过不同类型的交互来确认同意是否可以在有效吸引用户对内容的关注方面发挥重要作用，即使在反复曝光之后也是如此。我们在一项实验室研究中调查了通过同意书吸引用户的不同类型的交互在其有效性、效率和用户满意度方面有何不同。此外，我们还研究了习惯化是否以及如何影响用户的注意力和满意度，以及他们花费在表示同意上的时间。我们对四个不同组的 80 名参与者进行了一项对照实验，其中人们要么通过拖放 (DAD)、滑动或复选框积极参与政策内容，要么不积极参与政策内容（作为控制条件）首次暴露阶段和习惯阶段。我们从多个维度测量了用户对同意书的关注度，包括直接、客观的测量和间接的、自我报告的测量。我们的结果表明，不同类型的交互可能会影响用户对策略信息某些部分的关注。特别是，与其他组相比，DAD 操作导致用户对数据项的关注显着增加。然而，随着反复接触同意书，差异就消失了。