Intergroup contact is regarded as one of the most effective ways to reduce prejudice. However, recent longitudinal studies using contemporary statistical techniques (e.g., random intercept cross-lagged panel models [RI-CLPMs]) have failed to find evidence of within-person changes in prejudice following contact fluctuations. We propose that past time-lags may have been too long to capture change and conducted three studies with shorter time-lags of single days, weeks, or months. We also considered effects of positive versus negative contact frequency. We consistently found that people who are less prejudiced have more contact (i.e., between-person effects); however, fluctuations in naturally occurring contact were not followed by corresponding within-person changes in prejudice, suggesting shorter-term contact fluctuations are detached from prejudice. With abundant support for contact in the field, we argue that prejudice may be impacted by major contact events, or through gradually acquired cumulative experiences, but effects are not apparent when examining “thin-slices” of time.