Constructivism has long been touted as the end-all solution for having K-12 students learn science. At the core of this didactic method is the assumption that given the chance, children will naturally be able to act and think like scientists. In this paper, I review the recent evidence from the cognitive neuroscience and neuroimaging communities that suggests that this core assumption is more than likely incorrect. I also review the recent literature on the philosophies of learning and cognition, which suggests that learning can be made more meaningful through emphasizing the human interrelationships that are central to our thinking processes. I propose an intersubjective approach to science education -- an approach in which learning is underpinned with rich intersubjective interactions between the teacher and the students, and the emphasis of what is to be learned is not a factual content, but rather a human activity that undergirds the process of science. By teaching our students about the human interactions that comprise the scientific endeavor, the larger societal community may come to have a better understanding of science. Further, I argue that narratives, which can provide a context for students to understand what scientists do and why they do it, may even serve to motivate more of our students to become scientists. Perhaps by using these more intersubjective approaches to science education, we can begin to improve public understanding of science.