Scientists spend their careers trying to understand how the world works. Yet even in their own specialties, complete understanding may remain elusive.
That idea lies at the center of a new article by Richard Shiffrin , Distinguished and Luther Dana Waterman Professor of Psychological and Brain Sciences in the College of Arts and Sciences at Indiana University Bloomington, Stephen Stigler of the University of Chicago, and Frank Keil of Yale University. Published in Computational Brain & Behavior, the article, " Illusions of Understanding in the Sciences ," explores the perhaps surprising claim that scientists often believe they understand more than they actually do.
The authors posit that this gap between confidence and understanding is a fundamental feature of how science operates in a world of nearly limitless complexity.
"The universe we inhabit being infinitely complex, scientists never understand completely, even in their narrow specialties," Professor Shiffrin said. "Often they have an illusion of understanding far exceeding their reality."
While Shiffrin is best known for decades of influential research on cognition and memory, this project reflects a broader line of inquiry into how scientists themselves practice science. Stigler is a historian of statistics and emeritus professor in the University of Chicago's Department of Statistics, and Keil is a developmental psychologist and professor at Yale University whose work has examined how both children and adults explain the world around them.
Together, the researchers asked a deceptively simple question, how deep is scientific understanding?
Their answer begins with an example many scientists encounter throughout their careers, linear regression, a statistical technique used to estimate how one factor changes as another changes, such as how crop yields respond to rainfall or how test scores relate to study time.
Researchers across disciplines routinely use the method to identify relationships among variables and analyze data. Yet the authors argue that even many experienced users do not fully understand all of the assumptions, limitations, and causal implications involved.
The example serves as a gateway to a larger argument.
"Scientists seek to understand the causes of observed phenomena," the authors write in the study. "Beliefs that they have succeeded are based on understanding that is rarely or possibly never complete."
According to the article, understanding exists on a continuum rather than as an all-or-nothing achievement. Scientists may possess enough knowledge to make accurate predictions, develop useful theories, design successful experiments, or create technologies that benefit society without fully understanding every mechanism at work.
The authors illustrate the concept with examples from psychology and other disciplines. A researcher might explain forgetting as the suppression of unwanted memories, a theory that can inspire productive experiments and new lines of inquiry. Yet deeper questions remain about the underlying cognitive and neurological mechanisms involved. According to the authors, scientists often work successfully with explanations that are useful without being fully complete.
That distinction becomes especially important when mathematical models and computer simulations enter the picture.
"The illusion that a scientist has a valid and complete explanation tends to be magnified when the data is well described by mathematical and computer simulation models," the authors write. They note that predictive success can sometimes create the impression of deeper understanding than actually exists because "prediction does not imply causality, but gives the illusion that it does."
This assertion may appear counterintuitive given science's extraordinary achievements. Modern medicine, communications technology, transportation systems, and countless other advances have transformed daily life through scientific discovery.
However, the authors contend that incomplete understanding and scientific progress often coexist. "This limited understanding is nonetheless sufficient to explain how science has produced amazing progress with enormous benefits to society," Shiffrin said.
The authors depict science as a process of gradually improving explanations, rather than as a collection of absolute truths. Researchers build models, test hypotheses, gather new evidence, revise earlier assumptions, and move incrementally toward deeper understanding.
The consequences of overestimating understanding extend beyond academic debates. According to the authors, assumptions about what scientists know can influence nearly every stage of the research process, from experimental design and theory development, to teaching and communication.
"It makes a case that the many incomplete levels of understanding affect, often unwittingly, the ways scientists design experiments, test theories, comprehend, communicate, and teach," the authors observe.
The topic resonated widely within the scientific community. The essay attracted 32 published commentaries from researchers across multiple fields, an unusually robust response that prompted the three authors to prepare a collective reply.
The discussion reflects a longstanding interest in the strengths and limitations of human cognition. Recognizing those limitations, the authors assert, can ultimately strengthen science rather than weaken it.
Acknowledging gaps in understanding encourages greater caution when drawing conclusions, greater humility when communicating results, and greater openness to alternative explanations. It also reflects the reality that scientific knowledge is continually evolving.
That is, science advances through a continual process of refining, testing, and deepening humanity's understanding of how the world works. And, paradoxically, recognizing how much remains unknown may be one of the clearest signs of scientific progress.
