Page 117 - Demo

115Can third-party observers detect attraction?6Figure 1. Experimental setup of Prochazkova et al. (2022). Reprintedwith permission.(Current) experimental taskThe experimental task was controlled by a script written in E-Prime (Version 2; Psychology Software Tools, Pittsburgh, PA). Figure 2 illustrates theprogression of a typical trial in the task. Each trial started with a screenshowing the trial number (1.5 s), followed by the presentation of the videoseither in a side-by-side (woman left) or one-by-one fashion (3 s). Participants were instructed to attend to the video presented, with no specificinstructions regarding which video they should attend to specifically. Another screen followed on which participants were asked to indicate whetherthe person(s) they viewed would like to go on another date with their partner (separately for the male and female couple member displayed: “Doeshe want to go on another date with her?” “Does she want to go on anotherdate with him?”) and remained on the screen until an answer was provided(in Experiment 1, responses were provided using a pen and paper questionnaire). Finally, a screen appeared on which participants were required toindicate their degree of certainty regarding the previous response, which alsoremained on the screen until an answer was provided.Statistical analysesWe computed accuracy scores by comparing the participants’ responses(0 = no; 1 = yes) with the actual responses of the members of the datingcou-ples. These accuracy scores (0 = incorrect; 1 = correct) were analyzedusing Bayesian logistic multilevel modeling (MLM). Bayesian MLM’s useallowed us to both account for the dependence in our data andquantify the support for either the null or alternative hypothesispresent in our data (Jeffreys, 1961; M. D. Lee & Wagenmakers, 2013;van Doorn et al., 2019). All our analyses consisted of three models,each testing a separate hypothesis. We conductedIliana Samara 17x24.indd 115 09-04-2024 12:19