This dissertation examined the role of cues and the moderating impact of causal attributions (Weiner, 1985) driven by the Media Equation Hypothesis (Reeves & Nass, 1996) within Tomlinson and Mayer’s (2009) Causal Attribution Model of Trust Repair on dynamic learned trust in automation (TiA). Relatedly, this research investigated the moderating impact of the Perfect Automation Schema (Dzindolet et al., 2002) on dynamic learned TiA in addition to exploring how dynamic learned TiA changes as a function of trust in oneself (Trust vs. Confidence Hypothesis; Lee & Moray, 1994), and as a function of past experience. 80 participants completed a pilot study during which they alternated between performing and observing an automated system perform a visual search task. Periodically, participants rated their trust in themselves and their trust in the automated system. Findings from the pilot study suggested lower response variability for participants prompted after each trial for a trust assessment – relative to those prompted after every five trials. This informed the design for the dissertation study, which prompted participants for a trust assessment after every trial. 60 participants completed the dissertation study in which they alternated between performing and observing an imperfect automated system perform an identical task. Findings provided support to the Trust vs. Confidence Hypothesis (Lee & Moray, 1994) – participants’ self-reported trust in themselves shared an inverse relationship with their self-reported TiA. While this dissertation did not find support for Tomlinson and Mayer’s (2009) model, causal attributions for oneself, relative to the automated system provided support to the Media Equation Hypothesis (Reeves & Nass, 1996) and the Unique Agent Hypothesis (de Visser et al., 2016). Findings from an exploratory analysis suggested that Propensity to Trust Machines (Merritt & Ilgen, 2008) rather than the Perfect Automation Schema (Dzindolet et al., 2002) moderated average levels of TiA.