Combining the literature on humor, affect, the parasocial contact hypothesis, and stigma, this dissertation examines the relationship between humor and mental illness stigma in entertainment programming. The model predicted that cognitive and affective humor would predict positive affect and approach of the character with a mental illness. Positive affect and approach were hypothesized to predict parasocial interaction. Parasocial interaction was hypothesized to predict less social distance and less stigma. Study 1 examined this relationship using the television program, Enlightened. Two conditions (one with more humor and one with less humor) were presented to participants (N = 106). Results showed that more perceived cognitive (surprise) and affective (dark) humor predicted positive affect. Those who perceived less disparagement humor were more likely to want to approach the character. Approach predicted PSI and PSI predicted less stigma towards those with mental illness and predicted less social distance. As participants perceived more affective humor, they also felt more stigma towards those with mental illnesses. Study 2 (N = 82) replicated much of these results with different stimuli from the program, Girls. Specifically, affective and cognitive humor predicted positive affect (note: cognitive humor approached significance). Disparagement humor predicted approach tendency. Approach tendency predicted PSI and PSI predicted less social distance between the viewer and a person with a specific mental illness. There was a direct relationship between perceiving more affective humor and reporting more stigma towards those with mental illnesses. Implications for entertainment, theory, and health communication are discussed in addition to limitations and directions for future research. Advisors/Committee Members: Stevens, Elise, Dillman Carpentier, Francesca, Gibson, Rhonda, Comello, Nori, Hoffner, Cynthia, Raney, Art.