Researchers Marina Dubova from the University of California, Berkeley, and Qiawen E. Liu from Princeton University have sought to clarify the complex concept of science and its boundaries. Their recent study addresses the challenging question of what distinguishes scientific practices from non-science and pseudoscience.
Dubova and Liu conducted a survey involving 91 scientists and 99 members of the public, examining the accuracy of 18 metaphors traditionally used to describe science. Their findings were published in Royal Society Open Science.
In their exploration, Dubova expressed that defining science poses significant challenges due to its multifaceted nature, drawing parallels to how human concepts, like “bird,” resist precise definitions. “All human concepts are fuzzy and versatile,” she noted. “Defining anything and putting strict boundaries around concepts becomes problematic.” This is particularly relevant in science, as varied interpretations can complicate the distinction between scientific and non-scientific practices.
Dubova emphasized the need to differentiate between scientific and non-scientific advice in order to help the public evaluate information sources effectively. “We want to help people trust scientific advice compared to other kinds of information we might encounter,” she said.
The study revealed that scientists exhibit a greater disagreement regarding the definition of science compared to the public. Dubova suggested that this variability reflects the diverse ways scientists operate within their field. “Scientists engage in science in versatile ways,” she remarked, highlighting a lack of focus on discussing the nature of science itself during scientists’ training.
Interestingly, the findings also indicated that the public holds a more idealized view of science than scientists do. Dubova noted that while many people express a strong belief in the value of science, this idealism could complicate their expectations of scientific outcomes. “If the public is expecting a definitive finalized conclusion, they might reject advice that includes a lot of uncertainty,” she explained.
Additionally, Dubova discussed the use of metaphors in their research, highlighting her interest in the metaphor of science as “haptic engagement with reality,” which emphasizes the active interaction of scientists with the phenomena they study rather than mere observation. “We’re just trying to explore it in this tactile way,” she said, stressing the importance of viewing science as a dialogue with nature.
Looking ahead, Dubova noted the potential influence of artificial intelligence (AI) on scientific inquiry. While she expressed excitement about the advancements AI could bring, she acknowledged concerns about its impact on exploration within science, citing the risk of developing a scientific monoculture due to reliance on biased AI models. “We need to keep in mind how we scale up science with these models,” she cautioned.
The researchers plan to expand their work by investigating how public perceptions of science can affect communication and understanding between scientists and the public. Their study marks the beginning of an exploration into the ever-evolving nature of scientific understanding.


