Inquiry into the history of science shows an early “inherence” bias

(Press-News.org) Early scientific theories—such as those explaining basic phenomena like gravity, burning, and the movement of molecules in water—centered on presumed inherent properties rather than external factors, thereby misleading famous philosophers and scientists, from Aristotle to Scottish botanist Robert Brown, in their theorizing. 

A new study by a team of psychology researchers has now found that this tendency is in fact common in the history of science. Moreover, through a series of experiments and surveys, the paper’s authors conclude these misfires were likely driven by cognitive constraints, among scientists and non-scientists alike, that have acted as a bottleneck to discovery and shaped the trajectory of scientific theories over millennia.

The study, which appears in the journal Proceedings of the National Academy of Sciences, was conducted by researchers at the University of Edinburgh and New York University.

“Early scientific theories across multiple fields share a common pattern, in that they focus too much on built‑in features and too little on interactions with surroundings,” explains Zachary Horne, a lecturer in psychology at the University of Edinburgh and the paper’s lead author. “This bias appears throughout the history of science, and its ‘fingerprints’ can even be seen among scientists today.”

Horne and the paper’s other authors, Andrei Cimpian, a professor of psychology at NYU, and Mert Kobas, an NYU doctoral student, point to early theories of gravity as evidence of this systematic “inherence bias” in scientists’ initial attempts to explain a phenomenon.

Medieval scholars proposed that throwing an object gave a projectile an internal “impetus”—an assumed substance inside the object—that keeps it moving until that impetus runs out. When the substance is exhausted, they theorized at time, the object simply falls to the ground. In the 17th century, Galileo Galilei and Isaac Newton replaced this theory by accurately demonstrating that unless an external force acts on the object, it will keep moving in a straight line at steady speed. 

“Some of the most significant achievements have come about as a result of scientific ingenuity, but our cognitive processes, which favor explaining phenomena in terms of their inherent properties rather than external factors, seem to have historically slowed scientific discovery,” notes Cimpian.  

In their PNAS study, the authors surveyed historians of science in the US, Canada, and the UK and asked them to cite examples of major transitions in the history of science. More specifically, they were asked to draw from the historic record to list an initial explanation for an observation, as well as a subsequent explanation for the same observation—as in this example concerning tides, provided by one of the historians:

Observation: “Tidal motions of Earth’s large bodies of water”

Initial explanation: “Sloshing due to motion of earth” (Galileo; coded by the researchers as “inherent”)

Subsequent (and accurate) explanation: “Gravitational influence of the moon” (Kepler, Newton; coded by the researchers as “extrinsic”)

These responses were coded as “inherent” or “extrinsic” by doctoral students with training in philosophy of science. The results showed that the vast majority of the nearly 80 examples the historians listed were focused on inherent properties in their initial explanations. In contrast, subsequent explanations of the same phenomena showed less of this bias. 

An example of molecular behavior is illustrative. In 1827, Scottish botanist Robert Brown studied pollen grains’ structure. Under a microscope, the grains appeared to move around rapidly when they were suspended in water. Brown and other biologists at the time hypothesized that the motion of these grains was due to a “vital force” present within living matter. However, as Brown pursued this line of reasoning further, he noticed inconsistencies between explanation and the data, indicating the phenomenon was not due to an inherent factor. Decades later, others correctly identified an unseen—and environmental—factor to explain this movement: fast-moving molecules in the surrounding water, an external force, colliding with the pollen grains, which together form the centerpiece of what’s now known as “Brownian motion.” 

These findings led the PNAS authors to another consideration: Is this bias, found among history’s most prominent scientists, also found today among scientists and non-scientists alike? In other words, is it possible that scientists and non-scientists in the 21st century are also affected by these same cognitive bottlenecks when it comes to their theorizing about scientific phenomena? 

To explore this, the researchers conducted a series of experiments with practicing scientists, adult non-scientists, and children (aged 5-9). They provided novices and scientists with real scientific observations—with which both the scientists and non-scientists were unfamiliar—and asked them to explain why these phenomena occurred. For instance, children were asked to explain “why a hammer fell at the same speed as a feather on the moon” while adult non-scientists were asked to explain the presence of sediment in distilled water after it was boiled. By contrast, scientists were asked to explain more complex phenomena as diverse as why the tadpoles of the poisonous Dart frog are not poisonous or why an unfamiliar planet loses mass over time or has a magnetosphere of a certain size. These explanations were then coded for whether they were focused on inherent properties or extrinsic interactions. 

As with scientists from the past, the initial theories of both today’s scientists and non-scientists overweighted internal properties in their explanations. For example, when asked to explain the presence of sediment in distilled water after it was boiled, a non-scientist participant said that “heat caused the water to begin to turn into soil”—an idea reminiscent of Aristotle. Even trained scientists showed similar tendencies, albeit with more sophistication in the mechanisms they hypothesized. When asked to explain why some planets lose mass over time, some scientists cited “ejecta from volcanoes” and “transformation of matter on the planet into gas that dissipates away from the planet,” both of which neglect complex interactions between planets and their stars.

These tendencies demonstrate the difficulty of scientific inquiry—our most successful epistemic enterprise—rather than the incapability of scientists, the paper’s authors observe. This is a difficulty that may have impacted our greatest thinkers, both past and present. 

“The path from initial explanatory intuitions to mature scientific understanding is rarely straightforward,” the authors write in their conclusion. “This work suggests that one systematic source of detours may lie in our cognitive architecture itself—in the basic information processing constraints that guide how we first attempt to make sense of unfamiliar phenomena. Understanding these constraints is crucial not just for advancing cognitive science, but also for improving how we train future generations of scientists.”

# # #

END