Are People Polite to Smartphones? How Evaluations of Smartphones Depend on Who Is Asking

摘要

Studies following the CASA paradigm showed that computers sending social cues elicited social reactions in their human users who e.g. adopted social norms (Reeves and Nass 1996). As these reactions were originally exclusive for human-human interactions, the derived paradigm stated that "computers are social actors" (CASA; Nass et al. 1994) referring to the theoretical concept of media equation, basically saying that "media equals real life" (Reeves and Nass 1996, p. 5). Nass et al. (1999) focused on the norm of politeness. In their experiment they showed that the evaluation of a computer was more positive if the computer itself asked for it compared to another computer asking. Our study adopts this experimental approach. However, as technology has evolved since the 1990's we replaced desktop PCs with smartphones transferring the CASA paradigm to modern devices. In a laboratory experiment, participants (n = 108) interacted with a smartphone which they evaluated afterwards. There were three different settings with (1) the target phone itself (2) another given smartphone or (3) the participants' own smartphone asking for the evaluation. Analysis of variance revealed significant differences between the first and third setting (F[2,105] = 3.35, p = .04, η~2 = .06) with evaluations being significantly better if the target phone itself asked for them. Homogeneous answers were interpreted as an indicator of dishonesty. Results revealed that evaluations on one's own smartphone were significantly less homogeneous than on the target phone (F[2,105] = 3.20, p = .05, η~2 = .06). Moreover, within experimental group 3, the participants' closeness to their own phone was shown to be significantly negatively associated with the evaluation of the target phone. In sum, results are interpreted as indicators of smartphones eliciting social norms of politeness. Hence, both the CASA paradigm and the integration of a psychological perspective constitute a heuristically fruitful approach for the analysis of users interacting with (modern) devices.