Given the rapid growth of the older adult population in the US and the number of adult children providing informal care to their older adult parents, there is a need for a better understanding of how communication can be used to negotiate the tension between dependence and autonomy in these caregiving relationships (Egbert, 2014; Morgan & Hummert, 2000). Guided by psychological reactance theory (Brehm, 1966; Brehm & Brehm, 1981), which posits that individuals value their autonomy and are motivated to restore it when they perceive that it is being threatened, the purpose of this dissertation was twofold: (a) to explore the role of reactance in communication between adult child caregivers and their older adult parents and (b) to create messages that can be used by caregivers to simultaneously encourage older adult parents' healthcare compliance and optimize their perceived autonomy. These two purposes were addressed across three studies. In Study One, results of focus groups with older adults who receive care from at least one adult child (N = 19) revealed three types of caregiving messages used by adult children that older adults perceive as freedom-threatening (i.e., offering directives, expressing doubt, loss-framing), as well as themes related to the types of health behaviors usually at the crux of these messages and older adults' responses to these messages. In Study Two, results of interviews with adult child caregivers of at least one older adult parent (N = 14) revealed that adult children use a variety of messages in attempts to gain compliance from older adult parents, including types that overlap with what older adults perceive as freedom-threatening. In Study Three, older adults who receive care from at least one adult child (N = 288) were randomly assigned to either an autonomy-supporting caregiving message or one of multiple freedom-threatening messages, each created based on results of the first two studies. Results revealed that relative to the autonomysupporting message, messages in which caregivers were offering directives or expressing doubt triggered higher perceptions of freedom threat among older adults, which was related to greater experiences of reactance, which in turn was linked to greater intentions to engage in a variety of freedom restoration behaviors. With one exception, trait reactance and paternalism beliefs did not moderate older adults' experience of the reactance process following exposure to these caregiving messages. Taken together, these results provide evidence that the reactance process is applicable to communication between adult child caregivers and their older adult parents and yield suggestions for how to circumvent reactance in this context. iii