The growing popularity of fitness influencers on Instagram has created a competitive space where human and AI-generated influencers vie for user engagement. This study examines how agent type (human vs. AI), gender, and age influence perceptions of interpersonal attraction, credibility, and social presence. Using a 2 × 2 × 2 experimental design, 161 participants rated mock Instagram posts from influencers across these categories. Results showed that human influencers were rated significantly higher in social attraction, credibility, and social presence than AI influencers. Younger influencers were perceived as more physically attractive, while older influencers scored higher in character and social presence. Regardless of agent type, female influencers were rated higher in physical attraction. These findings highlight persistent demographic biases and a strong preference for human connection in digital fitness spaces, offering key insights for brands, marketers, and influencer strategy.

The purpose of this experiment was to examine the impact of computer-mediated word-of-mouth communication (WOM) on perceptions of a social robot instructing students. We hypothesized that participants receiving positive WOM about the social robot instructor would rate the robot higher for perceived credibility, attraction (social and task), social presence, and report higher levels of affective learning and state motivation to learn versus those participants receiving negative WOM (or no WOM) about the same performance. Results demonstrated that there are some factors in evaluating robot performance that are beyond the actual abilities of the robot. Results are examined using the Heuristic-Systematic Processing model, and the impact of WOM on perceptions of social robots in the environment. As a result of these findings, designers and practitioners need to be conscious of WOM.

This experiment was conducted to determine interpersonal impressions of two types of compliance-gaining strategies (warning and obligation) used to ask for the consideration of robot rights. Findings demonstrated that the strategy of obligation produced higher levels of petitioner perceived caring (a factor of source credibility) and higher perceptions of task attraction. Social attraction, competence, and character did not significantly differ by type of message strategy utilized.

Although both HRI and Communication Science often trace their origins to the transdisciplinary cybernetics of the 20th century, they have since developed in relative isolation, with scant scholarly exchange concerning the similarities and differences in their assumptions, insights, and approaches. The purpose of this half-day workshop is to explore the ways in which traditional communication theory and Human-Machine Communication theory from the discipline of Communication Science can be applied and be utilized in HRI studies.

The current study examines the pratfall effect and interpersonal impressions of how humans perceive robots that appear forgetful and apologetic regarding information given to them by a human. Results demonstrated that the non-forgetful robot was perceived higher for interpersonal impressions when compared to the forgetful and forgetful apologizing robots.

The purpose of this study was to test the strength of the machine heuristic; specifically when suspicion is primed in a message receiver concerning the veracity of information from a robot delivering news. Results demonstrate that low suspicion led to higher credibility evaluations, which consequently increased behavioral intentions. Findings are discussed in light of the MAIN model.

From keeping robots as in-home helpers to banning their presence or functions, a person’s willingness to engage in variably intimate interactions are signals of social distance: the degree of felt understanding of and intimacy with an individual or group that characterizes pre-social and social connections. To date, social distance has been examined through surrogate metrics not actually representing the construct (e.g., self-disclosure or physical proximity). To address this gap between operations and measurement, this project details a four-stage social distance scale development project, inclusive of systematic item pool-generation, candidate item ratings for laypersons thinking about social distance, testing of candidate items via scalogram and initial validity analyses, and final testing for cumulative structure and predictive validity. The final metric yields a 15-item (18, counting applications with a ‘none’ option), three-dimension scale for physical distance, relational distance, and conversational distance.

Extending previous research on the human-to-human interaction script, the MAIN model, and the Computers Are Social Actors (CASA) paradigm, this study examines the influence of weather-related communication sources with varying levels of expertise and agency cues (i.e. human versus bot) on impressions of communication quality. Specifically, an experimental design was used to measure people’s impressions of source credibility, task and social attraction, computer-mediated communication competence, and intent to interact in the future for three information sources: a professional meteorologist, an amateur meteorologist, and a weather Twitterbot. Results demonstrated similar perceptions of communication quality for the three Twitter agents. However, the Twitterbot was rated as significantly less socially attractive than the professional meteorologist and as significantly more task attractive than the amateur meteorologist delivering the same messages. Agency-related cues appear to trigger heuristics that account for these differences in perception. Implications for the use of Twitterbots in weather-related and information-sharing contexts are discussed.