Ghosts of Theories Past: The Ever-Present Influence of Long-Discarded Theories

Ghosts of Theories Past: The Ever-Present Influence of Long-Discarded Theories Andrew Shtulman (shtulman@oxy.edu) Joshua Valcarcel (jvalcarcel@oxy.edu) Department of Psychology, Occidental College 1600 Campus Road, Los Angeles, CA 90041 (Johnson & Carey, 1998; Slaughter & Lyons, 2003); the transition from a behavioral theory of illness to a germ- based theory (Au et al., 2008; Solomon & Cassimatis, 1999); and the transition from an essentialist theory of evolution to a selection-based theory (Shtulman, 2006; Shtulman & Schulz, 2008). Different scholars have characterized these transitions in different ways. Carey (2009) and Smith (2007) have characterized them as a series of conceptual differentiations, in which new conceptual boundaries are established, and conceptual coalescences, in which old conceptual boundaries are collapsed. Thagard (1992) and Chi (2008) have characterized them as the reassignment of a key concept or system of concepts from one branch of an ontological hierarchy to another. And Vosniadou (1994) and Wellman and Gelman (1992) have characterized them as a revision of the core presuppositions of a framework theory, or a theory that defines a domain’s ontological categories and causal mechanisms. Common to all characterizations is a commitment to knowledge restructuring, or the conversion of one conceptual system into another by radically altering the structure (and not just the content) of that system. Implicit in the idea of knowledge restructuring is the idea that early modes of thought, once restructured, should no longer be accessible, for the basic constituents of that system are no longer represented. A number of recent findings have challenged this idea, however, by showing that early modes of thought do sometimes reemerge later in life. Lombrozo, Kelemen, and Zaitchik (2007), for example, found that adults with Alzheimer’s disease endorse teleological explanations for natural phenomena that young children also endorse but that age-matched adults without Alzheimer’s disease do not – explanations like “trees exist so that plants and animals have shade” and “rain exists so that plants and animals have water to drink.” Kelemen and Rosset (2009) found that, under speeded conditions, college undergraduates are also inclined to endorse the same kinds of explanations. The fact that individuals who explicitly disavow such explanations under normal circumstances will still endorse them under abnormal circumstances (i.e., cognitive load or biological impairment) implies that teleological thought is suppressed with additional education but never completely overwritten. In a different line of experiments, Goldberg and Thompson-Schill (2009) found that adults are slower and less accurate at categorizing plants as living things than at categorizing animals as living things, despite knowing full well that both plants and animals are alive. While young children explicitly deny that plants are alive, they typically Abstract When individuals replace their naive theories of natural phenomena with more accurate, scientific ones, what happens to the original theories? Are they overwritten or merely suppressed? We investigated this issue by asking college undergraduates to verify two types of statements as quickly as possible: statements whose truth value was the same across naive and scientific theories of the phenomena at hand (e.g., “The moon revolves around the Earth”) and statements whose truth value differed across those theories (e.g., “The sun revolves around the Earth”). Participants verified the former more quickly and more accurately than the latter, though this difference was smaller for early-developing domains (fractions, germs, inheritance, matter, physiology) than late- developing domains (astronomy, evolution, mechanics, thermodynamics, waves). These findings suggest that intuitive theories survive the acquisition of a mutually incompatible scientific theory, coexisting with, or even competing with, that theory for many years to come. Keywords: intuitive theories, conceptual change, knowledge acquisition, science education Introduction Knowledge acquisition can take two very different forms. One form, known as “knowledge enrichment,” involves the accretion of beliefs expressible in terms of preexisting concepts, like learning the traits of an unfamiliar animal or learning the history of an unfamiliar country. Another form, known as “conceptual change,” involves revising the very concepts that articulate those beliefs, like learning that the earth is a sphere, not a plane, or learning that weight is a relational property of objects, not an intrinsic property. The result is a capacity to represent information incommensurate with our prior beliefs and experiences. Numerous instances of conceptual change have been documented in the cognitive development and science education literatures. They include the transition from an impetus-based theory of motion to an inertial theory (Clement, 1982; McCloskey, 1983); the transition from a substance-based theory of energy to a process-based theory (Reiner, Slotta, Chi, & Resnick, 2000; Wiser & Amin, 2001); the transition from a geocentric model of the solar system to a heliocentric model (Siegal, Butterworth, & Newcombe, 2004; Vosniadou & Brewer, 1994); the transition from a tactile theory of matter to a particulate theory (Nakhleh, Samarapungavan, & Saglam, 2005; Smith, 2007); the transition from an integer-based understanding of fractions to a division-based understanding (Hartnett & Gelman, 1998; Moss & Case, 1999); the transition from a psychological theory of bodily functions to a vitalist theory