Stimulus Traits and Inter-subject Variability in Provoking Harmonic Expectancy Violation Juslin and Västfjäll's BRECVEM framework (2008, 2010) proposes Musical expectancy as one of seven “psychological mechanisms” for evoking emotions. Indeed, a violation of expectation is frequently pointed to as a trigger for music-induced frisson. However, the exact combination of personal, inter-personal, and musical contexts that may create the perfect circumstances for positively-valenced expectancy violations remains an open research question. Yet, before we can model such complex relationships with listening experiments, there are numerous variables relating to stimulus selection that, despite their potential impact on a study’s results, have largely remained unexamined. For example: what role do stimuli properties play in “setting up” an expectancy violation? what is the general propensity or likelihood of the “average” piece of music to contain an expectancy violation? and what is the general agreement across a random sample of the population of the location of such a violation? This poster reports empirical results from a between-subjects empirical experiment with 100 participants listening to 30 different popular music clips under different conditions, finding that: the lead-in time to the target can be effective at a mere 15 seconds; that agreement on the location of “harmonically surprising” events overall was significant, but contained a large amount of variation; that ecologically-valid stimuli appeared slightly less likely to be heard as surprising (compared with ‘contrived’ stimuli); as well as an interesting (but challenging to interpret) effect for the role of familiarity. Implications for future music expectancy research as well as more general music emotion work will be discussed.