This study replicates an experiment described by Patsouras et al. [5] or Rader et al. [8] in which subjects had to evaluate the loudness of a passing-by train noise while watching pictures of the train, the colour of which being varied. In contrast to these studies, no influence of train colour on loudness evaluation could be found. This discrepancy might be due to cultural differences (French subjects vs. German or Japanese ones). But a more realistic reason can be a jury sampling effect, as this influence can exist for very few subjects (one among twenty-two in this experiment).