The movie fragment in the neutral condition contained two crossroads in Amsterdam showing normal traffic. In the positive affect condition, participants could choose between different movie fragments: a fragment from a Disney movie (“The Little Mermaid” or “Lion King”) or a sketch from a Dutch comedy program (“De Lama’s”). In contrast

to the movie fragment in the neutral condition, the fragment in the positive affect condition was hypothesized to induce positive affect. Participants filled out the same questionnaire as discussed above. The eye movement task as outlined above consisted of 200 trials. For the analysis of the questionnaire we used paired t-tests with the within-subjects variable time (before vs after movie). Saccades were automatically detected using software

developed by SR Research. Thresholds for detecting the onset of saccadic GSI-IX movements were accelerations of 8000 (deg/s_squared), velocities of 30.0 (deg/s), and distances of 0.5 (deg) of visual angle. Movement offset was detected when velocity fell below 30.0 (deg/s) and remained at the level for 10 consecutive samples. Saccade latency was defined as the interval selleck screening library between target onset and the initiation of a saccadic eye movement. Trials were excluded when the latency of the saccade was lower than 80 ms or higher than 600 ms (see e.g., Nijboer, Vree, Dijkerman, & Van der Stigchel, 2010), or further than two and a half standard deviations away from the subject’s mean latency. Moreover, trials were excluded from analysis in which no saccade, too early or a too small first saccade (<3°) was

made. The endpoint of the first saccade had to have an angular deviation of less than 22.5° from the center of the target or the mirrored target location. In the first case, the saccade was classified as a prosaccade; in the second case the saccade was classified as an antisaccade. In other situations, the saccade was classified as an error and not analyzed. An Analysis of Variance (ANOVA) with Condition (positive affect vs neutral) and Task (prosaccade vs antisaccade) as within-subjects factors was used to analyze effects on saccade latency. Only trials in which oxyclozanide the first eye movement was initiated correctly (either a prosaccade or antisaccade, depending on the task) were included in the saccade latency analysis. To investigate the effect of induced positive affect on errors, a paired t-test was run on antisaccade trials with Condition (positive affect vs neutral) as the factor. Additional comparisons were made between the positive affect and neutral conditions for the percentage erroneous eye movements with express (80–130 ms) and regular (>130 ms) latencies. In the neutral condition, none of the questions was responded to differently before or after participants saw the movie (p’s > 0.05). In the positive affect condition, participants were more amused (t(11) = 5.00; p < 0.001) and more positive (t(11) = 2.35; p < 0.