Page 165 - Demo

Eye-tracking reveals bias to flanges in orang-utans1637Thus, the Looking time bias-score represents the proportion of the total fixation time on the stimuli directed towards the flanged male stimulus. In Experiment 2, we also explored how the orang-utans divided their attention when presented with two stimuli of males of the same morph compared to when presented with males of different morphs. Because we could not calculate a Looking time bias-score in the same way as described above, because in some trials there were either two flanged male- or no flanged male-stimuli, we calculated the score differently for this analysis. We used the location of the photos as a reference point to calculate the looking time bias by calculating the bias toward the left picture (Roth et al., 2023). Hereafter, we tested whether this bias is affected by the stimulus category on the left side and right side on the screen.To study whether orang-utans had an attentional bias toward flanged males, we modelled Looking time bias-score as a function of the Intercept and the Location of flanged male stimulus (left/right on the screen) to control for potential side biases. To investigate the effect of different combinations of male morphs on orang-utan attention, we modelled the Looking time bias-score as a function of the interaction between Left stimulus morph (flanged/unflanged) and Right stimulus morph (flanged/unflanged). The same model formulas were specified for all four model components (mu, phi, zoi, and coi). In all analyses, we allowed Intercepts to vary by Subject, and Session nested in Subject. Furthermore, we weighed each trial by the total fixation duration on the stimuli combined in that trial relative to the subject’s average. Thus, trials in which the orang-utans paid more attention to the screen had a larger weight in the analysis. In this manner, we avoided that trials where the orang-utans were relatively distracted or disinterested would have a large influence on the outcome of our analysis.To explore whether we could find attentional biases towards specific males in Experiment 2, we calculated a Looking time bias-score for each stimulus in each trial, reflecting the proportion of total fixation duration that the orang-utan fixated on stimuli depicting a specific individual. We constructed three separate models: one for each female. Due to computational issues, we could not construe an individual preference-model for the male participant, Amos. Within each model, we modelled Looking time bias as a function of Individual of interest (Bako, Sibu, Jingga or Wousan) and Other individual presented (Bako, Sibu, Jingga or Wousan), to control for the effect of the second stimulus. We allowed the Intercept to vary by session.For the zero-one inflated beta models, we used a Gaussian prior with M= 0 and SD= 1 for the Intercept of the beta component of the model and for all independent variables. This also applied to the independent variables in the Tom Roth.indd 163 08-01-2024 10:41