Most research on images in web surveys has examined how images affect reports to questions (i.e., visual context effects; Couper, et al. 2004). Research gaps remain, however, especially with respect to how respondents process images in surveys. Some research has examined how the placement of images in a survey affects whether respondents fixate on images and how fixation affects reports (e.g., Tourangeau, et al. 2005). What remains unanswered is how respondents process an image when it is the first, second, third, or fourth image to appear in the survey and whether subgroups of respondents process images differently (e.g., gender, age, literacy level). We answer these questions using data from an eye-tracking study that included images placed next to four questions throughout the questionnaire. We will: 1) examine if the percent of respondents who fixate on images and the amount of time spent looking at them differs by when the images appear in the survey, 2) examine whether these vary by subgroups, and 3) test if visual context effects require fixation on images and/or longer fixation or if they can occur even when respondents only peripherally process the images. This paper examines image processing and visual context effects on adult respondents from the general population with data collected between October 2013 and February 2014. Preliminary analysis from a pilot study with college students (N=41) indicates that the percentage of respondents who fixated on images declined after the second image in the survey (p=0.015). We also found that respondents who fixated on images reported spending more money on vacation than those who did not fixate on the images, suggesting that visual context effects may differ by whether respondents fixate on images. Our paper ends with recommendations about using images in web surveys.