The idea that science—and human knowledge more broadly—aims to approach the truth about the world is widely shared among scientists, philosophers, and laypeople. The theory of truthlikeness (verisimilitude, truth approximation) makes this idea precise. Over the last 50 years, truthlikeness has garnered attention primarily within a narrow circle of logicians and philosophers of science, remaining largely overlooked by psychologists and cognitive scientists. This paper proposes truthlikeness theory, and more broadly cognitive decision theory, as a valuable framework for future empirical research in cognitive sciences. We argue that truthlikeness offers a normative perspective that can complement and enhance existing approaches in the study of human reasoning and decision-making. By examining selected case studies, we demonstrate how truthlikeness theory can serve as a benchmark for designing experiments, interpreting empirical results, and enriching debates on human rationality. Specifically, we show how many so-called biases or fallacies, documented by psychologists and purportedly demonstrating human irrationality, may be construed as rational strategies to balance informativeness and truth in belief, judgment, and decision-making.