This article describes an empirical study designed to contribute to research into interpreting quality by moving towards a definition of the criterion of ‘fluency’ in the evaluation of simultaneous interpreting. Based on an interdisciplinary review of the literature and preliminary tests of the production and perception of pauses and their effect on speech comprehension, ‘silent pauses’ are investigated as a subparameter of fluency. Following the line of research initiated by Collados (1998), the study analyses the quality expectations of 43 expert users as well as their evaluation of German-Spanish interpretations manipulated by additional silent pauses. Although the differences found in the evaluation study cannot be shown to be statistically significant, there are trends indicating that the principal hypothesis, whereby silent pauses as a subparameter of fluency have a negative effect on fluency evaluation, has received some empirical support. Subjects appear to have detected and responded to the experimental stimulus (two levels of additional silent pauses between 2 and 6 seconds in duration) by giving lower mean ratings for ‘fluency’ to the experimental videos than to the control.