Since results from our previous experiment suggested that the effect of pronounceability on harmfulness ratings may be specific to the stimuli used in the original study finding the effect (Song & Schwarz, 2009), we conducted another experiment to replicate the original study using both original and newly constructed stimuli. In our previous experiment, participants judged harmfulness of poisons and medicines, while in the study by Song & Schwarz (2009), participants judged harmfulness of food additives. The present experiment explored generalizability of the effect on new stimuli using the original food additives scenario. <br><br> Based on our previous experiment, we suspect that the original positive relationship between pronounceability and safeness may hold only for the subset of stimuli used by Song and Schwarz (2009) and that there might be no relationship for the similar, but new stimuli. <br> Song, H., & Schwarz, N. (2009). If It's Difficult to Pronounce, It Must Be Risky Fluency, Familiarity, and Risk Perception. *Psychological Science, 20*, 135-138.