This study is the first to show that the choice for a positive or a negative question wording systematically affects survey answers: respondents are more likely to answer no or disagree to negative attitude questions (e.g. X is bad. Yes/No) than to answer yes or agree to equivalent positive ones (e.g. X is good. Yes/No). To unravel the causes for this positive/negative asymmetry, we related
wording effects to the cognitive processes underlying question answering. Two eye-tracking studies were conducted to obtain information about the temporal aspects of the question-answering process for positive versus negative questions. The observed time differences were subsequently related to a theoretical model for the question-answering process. This was done using a combination of insights about processing survey questions, experimental designs in which that knowledge was used, and newly designed eye-tracking measures that are relevant for these specific purposes. With this combination of strategies we were able to demonstrate that contrastive questions cause similar attitudes to be retrieved from memory. Therefore, they measure the same underlying attitude, and hence, are equally valid. Wording effects arise when respondents translate their own opinion into the response options. During this stage, the answering options are assigned a meaning that is not absolute, but relative to the polarity of the evaluative term in the question. The conclusions drawn from this study have implications for theories about text processing in general, and for theories about the cognitive processes of question-answering in particular. These implications, as well as the implications for survey practice, are discussed in this dissertation.