Page 58 - Demo

56Chapter 3to exclude trials: if the RT was slower than the subject’s median RT + 2.5 ⇥MAD, we excluded the trial. These outlier criteria resulted in the exclusionof 299 trials (5.83%). Hereafter, we centralized the RTs by subject. Thiswas done to make it easier to set a prior for the Intercept. All factorialpredictors were sum coded, and pre-date attractiveness ratings were centeredat 4 because this was the middle option.We followed a similar procedure for the analysis that investigated theassociation between date outcome (i.e., willingness to go on another datewith dating partner) and post-date attractiveness rating on immediate attention. Two participants dropped out before the speed-date part of theexperiment. Therefore, we had to exclude their data, leaving us with datafrom 56 participants. Some participants did not go on a speed date withevery opposite-sex person they saw on the stimuli, either due to dropoutsor unequal group size. After excluding the cases where date outcomes weremissing for either the probe or the distractor stimulus, we ended up with3460 data points out of the original 5220. Hereafter, we again excluded outliers by subject (see above), resulting in the exclusion of 209 trials (6.04%).Eye-trackingIn total, 36 participants completed the eye-tracking task. One participantdid not complete the pre-date attractiveness rating task and did not participate in the speed-dates. Therefore, we excluded their data (45 trials) fromthe analysis. Furthermore, we excluded 6 trials because participants werenot looking at the stimuli, leaving us with 1569 trials from 35 participantsto investigate the effect of attractiveness on voluntary attention. For theanalysis that investigated the effect of date outcome and post-date attractiveness rating on voluntary attention, we had a smaller number of trials dueto the fact that not all people that were rated for attractiveness participatedin the speed-dates (either due to dropout or due to unequal group sizes). Intotal, we could include 1009 trials from 35 participants.Eye-tracking data were recorded continuously throughout the task witha sampling rate of 60 Hz. Here, only data during the stimulus presentationwere analyzed. Fixations on either area of interest (AOI) were logged usinga custom E-prime script. We excluded practice trials (6.25%) and gaze samples where either the left or right pupil was not recorded (3.50%). Followingthese criteria, we were left with 90.25% of the data intact.Statistical analysesAll analyses were performed in R statistics Version 4.1.3 (R Core Team,2022). First, we calculated the Intra-Class Correlations (ICC) for the individual pre-date attractiveness ratings. We used the R package irrNA(Brueckl & Heuer, 2022), because it properly deals with missing values inIliana Samara 17x24.indd 56 08-04-2024 16:35