Figure 4

Average bias towards a specific stimulus variant per participant and experimental condition (music valence and stimulus type), as measured by difference from an unbiased (i.e., 50:50) proportion of both stimulus variants. Black bars represent group averages as well as 95% confidence intervals for within subject designs. The face stimulus, compared to the gabors, induces variant specific biases, since upright faces are preferentially perceived when competing with tilted faces. This effect is however not moderated by induced affective valence.