While assigning student roles is a popular technique in online discussions, roles and the responsibilities allocated to them have not been consistently assigned. This makes it difficult to compare implementations and generate principled guidance for role design. This study critically examined frequently assigned student roles and identified a set of seven common functions they ask learners to perform (motivate others to contribute, give direction to the conversation, provide new ideas, use theory to ground the discussion, bring in (relevant external) sources, respond to previous comments, and summarize existing contributions). The latter six functions were used to generate a targeted set of role descriptions (Traffic Director, Starter, Inventor, Importer, Mini-me, Elaborator, Questioner, Devil's Advocate, Synthesizer, Wrapper) and refine a content analysis scheme to assess function fulfillment in online discussions. The roles were tested in a semester-long mixed-level blended educational technology course of 21 students; student feedback was solicited via survey. Roles designed to elicit the functions give direction, use theory, bring in source, respond, and summarize showed a greater degree of function fulfillment than roles which were not. Students found the Starter role very valuable for giving direction to the discussion, and saw moderate value in the Synthesizer and Wrapper role's summaries. The Devil's Advocate role did not fulfill its respond function but was valued by students, suggesting a new possible critique function. All roles were found to use theory, possibly due to task structure. Implications for future research and practice in the design of online discussions are discussed.