Building Long-Term Human–Robot Relationships: Examining Disclosure, Perception and Well-Being Across Time

Abstract

While interactions with social robots are novel and exciting for many people, one concern is the extent to which people’s behavioural and emotional engagement might be sustained across time, since during initial interactions with a robot, its novelty is especially salient. This challenge is particularly noteworthy when considering interactions designed to support people’s well-being, with limited evidence (or empirical exploration) of social robots’ capacity to support people’s emotional health over time. Accordingly, our aim here was to examine how long-term repeated interactions with a social robot affect people’s self-disclosure behaviour toward the robot, their perceptions of the robot, and how such sustained interactions influence factors related to well-being. We conducted a mediated long-term online experiment with participants conversing with the social robot Pepper 10 times over 5 weeks. We found that people self-disclose increasingly more to a social robot over time, and report the robot to be more social and competent over time. Participants’ moods also improved after talking to the robot, and across sessions, they found the robot’s responses increasingly comforting as well as reported feeling less lonely. Finally, our results emphasize that when the discussion frame was supposedly more emotional (in this case, framing questions in the context of the COVID-19 pandemic), participants reported feeling lonelier and more stressed. These results set the stage for situating social robots as conversational partners and provide crucial evidence for their potential inclusion in interventions supporting people’s emotional health through encouraging self-disclosure.

Studying and Eliciting Self-Disclosure: Interdisciplinary Review of Research Methodologies and Behavioural Paradigms

“…Employing artificial agents like robots (e.g., Laban, George, Morrison, & Cross, 2021;Laban, Kappas, Morrison, & Cross, 2023a), chatbots (e.g., Croes & Antheunis, 2021;Ho, Hancock, & Miner, 2018;Laban, 2021;Laban & Araujo, 2020, 2022Y.-C. Lee, Yamashita, Huang, & Fu, 2020), or virtual agents (e.g., Lucas, Gratch, King, & Morency, 2014;Lucas et al, 2017;Pickard, Roster, & Chen, 2016) as tools for eliciting self disclosure presents an innovative approach. It has been argued before that these artificial agents can be useful "stimuli" in psychological experiments to examine whether similar mechanisms of social cognition would be activated in interaction with an artificial agent, as would be elicited in interaction with another human (Wykowska, 2020;Wykowska, Chaminade, & Cheng, 2016).…”

“…This method can be particularly useful for eliciting sensitive or stigmatized information, potentially leading to more authentic disclosures (see Laban, programmed for consistency in responses, ensuring standardization across multiple sessions. It's noteworthy that this approach has been validated in engaging both adults (e.g., Laban et al, 2021;Laban, Kappas, et al, 2023a) and children (e.g., Abbasi et al, 2022;Neerincx et al, 2023), as well as people with clinically diagnosed conditions like ASD (e.g., Kumazaki et al, 2022) or mental health symptoms (e.g., Nomura, Kanda, Suzuki, & Yamada, 2020), not only in promoting self-disclosure but also in enhancing their well-being, highlighting artificial agents' versatility and effectiveness across different age groups and populations.…”

“…Computational automated content analysis uses algorithms to analyse large amounts of data, including self-disclosure behaviour and lexical features, providing a more efficient analysis of self-disclosure. Automated techniques can provide values for disclosure quantities (i.e., the volume of shared content; e.g., disclosure duration in seconds, length in number of words, and number or adjectives or specific words in the disclosure) (e.g., Laban et al, 2021;Laban, Kappas, et al, 2023a;Mejova & Hommadova Lu, 2023). Self-disclosure has been linked to the total number of words a person produces during an interaction or within a single turn in the interaction.…”

“…Another approachable computational method is sentiment analysis (e.g., VADER, see Hutto & Gilbert, 2014), a natural language processing method which extracts and analyses information classifying and rating positive and negative keywords (e.g., Baroutsou et al, 2023;Keijsers, Bartneck, & Kazmi, 2019). These measures provide valuable insights into different aspects of self-disclosure in interpersonal interactions (e.g., Laban et al, 2021;Laban, Kappas, et al, 2023a;Laban, Morrison, et al, 2022). However, automated methods may not capture the full context and emotional aspects of self-disclosure behaviour and may be limited by the specific algorithms used.…”

Knowledge cues to human origins facilitate self-disclosure during interactions with chatbots

“…Self-disclosure with a chatbot opens new interaction-related parameters, removing barriers to self-disclosure linked to increased vulnerability and fear of judgement (21; 24; 25; 26). Prior studies have shown that opening up to artificial agents can enhance individuals' mood and diminish sensations of loneliness and stress (27). Additionally, it has been observed that individuals experiencing negative emotional states (such as lower mood, increased feelings of loneliness, and higher stress levels), particularly those who identify as more introverted, tend to disclose more to artificial agents (28).…”