Understanding the cognitive heuristics people use to make decisions and the biases in judgment resulting from their use has been a focus of risk communication research for nearly forty years. During that period, research has demonstrated consistent effects of these heuristics and biases across multiple contexts using fairly standardized approaches to measurement. My previous research, however, has found that people may make decisions inconsistent with the expectations of known cognitive biases when those decisions are situated within narratives. This dissertation expands on that work and examines the role narratives play in decision making, specifically in the reduction of cognitive biases caused by the use of mental heuristics when reading stories. Three experiments empirically examine three known biases: prospect theory framing effects, the effect to which people make different decisions depending on whether a choice is presented as a loss or as a gain (Chapter 2); anchoring effects, the effect to which people judge a decision based on an initial unrelated piece of information given prior to the decision (Chapter 3); and the gambler's fallacy, the faulty notion that the result of an independent event is related to the result of previous independent events (Chapter 4). Chapter two tests whether the influence of narratives over cognitive biases may result from goals held by readers. It found further evidence that narratives reduce the biases related to gain-loss framing effects and no evidence that this reduction was due to goals held while reading. Chapter three explores the role of perceived realism in bias reduction and also tests the anchoring and adjustment heuristic in a narrative. It found that narratives reduce biases related to the anchoring and adjustment heuristic and no evidence that perceived realism of the narratives influenced the reduction. Chapter four examines whether the presentation of a story as either first person or third person influences a reader's decision and also tests the gambler's fallacy in a narrative. The experiment yielded inconclusive results, specifically a difference between the narrative and non-narrative conditions that did not meet the standard and acceptable level of statistical significance. This is notably different from the results presented in chapters 2 and 3, and is possibly due to the fact that unlike the non-narrative versions of the framing and anchoring experiments, narrative information is inherently contained in the non-narrative condition of the gamblers fallacy as it pertains to an ordered sequence of events. Overall, the findings expand our understanding of narrative stories in risk communication as they relate to cognitive heuristics and biases. Specifically, the three experiments suggest that narratives reduce the biases caused by cognitive heuristics in risk decision-making.