In recent years, Web surveys have become a standard survey mode. So far, online questionnaires resemble their paper counterparts to a great extend: online measurement instruments rely mostly on visually presented written questions with associated response categories. However, compared to paper and pencil questionnaires Web surveys allow for a richer communication with the respondent: graphical elements, pictures, and animated GIFs are used to enhance the appearance of web pages. As a result some Web surveys make use of a humanized interface, e.g. a picture of the investigator, pictures of people performing activities that the respondents are supposed to report on and the like. Earlier studies could demonstrate that those low-intensity humanizing elements have no or only marginal effects on the responses obtained. However, with the wide spread availability of broadband Internet connections video and/or audio are being used in Web surveys as well. Accordingly, in this study we will assess the impact of pre-recorded video clips of interviewers reading the questions to the respondents. We will focus on the effects of this new way of transmitting the questions on question understanding and response errors. We assume that the audio-visual channel changes the cognitive processing of the questions und thus the responses obtained. The paper reports results from a field-experimental study on the impact of video support in Web surveys on question understanding. Within a Web survey among university students a standard interactive online questionnaire was used. A random sub-sample answered a version of the questionnaire that consists not only of written questions but of corresponding video files of interviewers reading the questions to the respondent. 800 respondents were randomly assigned to either of two versions: (1) a traditional text-based version of the questionnaire, (2) a video-enhanced version using a human interviewer. The survey covered a variety of questions on relationships and sexual behaviors including several highly sensitive items on sexual practices, sexually transmitted diseases and the like. Using a series of split experiments on question order and question understanding we will assess to what extent the processing of the questions is affected by the presence or absence of the interviewer. Based on the findings we will develop a heuristic model of the impact of various communicative channels on the cognitive processing of survey questions.