Effects of repeated stimulus presentations on Oppel-Kundt and Müller-Lyer illusions

We present a post-hoc analysis examining whether repeated trials and multiple sessions affect the measured strength of two length illusions, Oppel-Kundt (O-K) and Müller-Lyer (M-L), as well as a non-illusory control stimulus. Data were taken from earlier studies in which participants adjusted a variable segment to match a reference segment. Short sessions featured five trials per figure, whereas extended sessions contained 24 or 26 trials for O-K and M-L. Linear mixed-effects models tested how trial number, session number, initial interval length, and figure subtype influenced illusion magnitude. O-K illusion magnitude tended to decline across sessions in the extended sessions, although individual observers displayed varying trends-some increased, others decreased. M-L illusions did not show the pronounced adaptation reported in previous work. These findings highlight the need to manage repeated presentations: controlling viewing durations, inter-session intervals, and participant strategies can help limit adaptation. Methodologically, mixing different stimuli, randomizing key parameters, and scheduling breaks appear to yield more stable measurements of illusion magnitudes.