The Impact of a Chatbot's Ephemerality-Framing on Self-Disclosure Perceptions

TL;DR

This study investigates how a chatbot's ephemerality-framing—whether it presents as a Stranger who forgets conversations or a Familiar companion who remembers past interactions—affects users' self-disclosure. Using a 2 × 2 mixed design across two sessions, with Emotional versus Factual disclosure, the authors measure honesty, comfort, enjoyment, closeness, and continuation intentions, complemented by qualitative insights into memory, privacy, and user beliefs. Quantitatively, there are few global main effects, but memory framing interacts with disclosure type and order to influence comfort and enjoyment, while perceived intelligence and beliefs about chatbot emotions modulate engagement and closeness. Qualitatively, Stranger framing enhances anonymity and reduces judgment for Emotional disclosures, whereas Familiar framing supports rapport when initial disclosures are low-risk; these findings highlight design trade-offs between ephemerality and memory in shaping self-disclosure and have practical implications for tailoring chatbot interactions to context and user sensitivity.

Abstract

Self-disclosure, the sharing of one's thoughts and feelings, is affected by the perceived relationship between individuals. While chatbots are increasingly used for self-disclosure, the impact of a chatbot's framing on users' self-disclosure remains under-explored. We investigated how a chatbot's description of its relationship with users, particularly in terms of ephemerality, affects self-disclosure. Specifically, we compared a Familiar chatbot, presenting itself as a companion remembering past interactions, with a Stranger chatbot, presenting itself as a new, unacquainted entity in each conversation. In a mixed factorial design, participants engaged with either the Familiar or Stranger chatbot in two sessions across two days, with one conversation focusing on Emotional- and another Factual-disclosure. When Emotional-disclosure was sought in the first chatting session, Stranger-condition participants felt more comfortable self-disclosing. However, when Factual-disclosure was sought first, these differences were replaced by more enjoyment among Familiar-condition participants. Qualitative findings showed Stranger afforded anonymity and reduced judgement, whereas Familiar sometimes felt intrusive unless rapport was built via low-risk Factual-disclosure.

Figures (5)

• Figure 1: The experiment flow for user studies. Participants talked to either the Familiar or Stranger chatbot across two separate chatting sessions, with one session eliciting Emotional-disclosure and one Factual-disclosure.
• Figure 2: The chatbot interface as seen by participants in the first chatting session for the Factual-disclosure condition. The chatbot initiates conversation by framing its relationship with ephemerality and the user.
• Figure 3: Chatbot interface as seen by Familiar participants in the second session. Chatting history (in this case from the Emotional-disclosure condition) is displayed above the current session's conversation.
• Figure 4: Plots of Comfort of Self-Disclosure. Error bars represent standard error, and significant differences ($p < 0.05$) indicated by square brackets. Participants felt more comfortable during Emotional-disclosure when talking to the Stranger chatbot. This effect was particularly pronounced when Emotional-disclosure was elicited in the first chatting session.
• Figure 5: Plots of (a) Response Length (in both orderings), (b) Desire to Continue (Factual first), and (c) Enjoyment (Factual first). Significant differences ($p < 0.05$) indicated by square brackets. (b) and (c): Familiar received more positive engagement (desire to continue and enjoyment) than Stranger when ordering of disclosure type was Factual- followed by Emotional-disclosure, as would be seen in literature altman1973socialpenetrationtheory.