- Top of page
Science education studies have revealed that students often have misconceptions about how nature works, but what happens to misconceptions after a conceptual change remains poorly understood. Are misconceptions rejected and replaced by scientific conceptions, or are they still present in students' minds, coexisting with newly acquired scientific conceptions? In this study, we use functional magnetic resonance imaging (fMRI) to compare brain activation between novices and experts in science when they evaluate the correctness of simple electric circuits. Results show that experts, more than novices, activate brain areas involved in inhibition when they evaluate electric circuits in which a bulb lights up, even though there is only one wire connecting it to the battery. These findings suggest that experts may still have a misconception encoded in the neural networks of their brains that must be inhibited in order to answer scientifically.
For at least 30 years, researchers in science education have studied people's spontaneous conceptions about how nature works (Duit & Treagust, 2012). These studies have shown that these intuitive conceptions are often opposed to the scientific knowledge taught in schools (Liu, 2001). For example, many people believe that heavier objects fall faster (even in the absence of air resistance, which is false), or that it is warmer in summer because the Sun is closer to the Earth (which is also false). If these misconceptions were not so difficult to change, they would not be a problem. However, one of the most robust findings of science education research about misconceptions is that they are particularly hard to change (Duit & Treagust, 2012; Periago & Bohigas, 2005; diSessa, 2006; Vosniadou, 2012; Wandersee, Mintzes, & Novak, 1994), which poses a serious challenge for science teachers who try to change their students' misconceptions into scientifically valid knowledge.
The problem of the persistence of nonscientific conceptions during science education has led to a field of research called “conceptual change” (for a review, see Duit & Treagust, 2012; diSessa, 2006; Vosniadou, 2008, 2012). This field tries to understand why students' misconceptions are hard to change, what changes during conceptual change, and how to facilitate the learning of unintuitive scientific concepts. Over the years, researchers in this field have proposed several theoretical models to answer these questions (Carey, 2009; Chi, 1994; Giordan & DeVecchi, 1987; Mortimer, 1995; Nussbaum & Novick, 1982; Posner, Strike, Hewson, & Gertzog, 1982; diSessa, 1993; Smith, 2007; Stavy et al., 2006; Vosniadou, 1994).
Most of these models (Carey, 2009; Chi, 1994; Duit & Treagust, 2003; Giordan & DeVecchi, 1987; Nussbaum & Novick, 1982; Posner et al., 1982; Smith, 2007; Vosniadou, 1994) share a common postulate according to which conceptual change is hard to achieve not only because students must abandon their initial misconceptions, but also because they must radically restructure their knowledge structure in order to accommodate new scientific concepts and theories. For example, according to Duit and Treagust (2003), misconceptions are tangled in a broad conceptual system and, consequently, the “conceptual structures of the learners have to be fundamentally restructured in order to allow understanding of the intended knowledge, that is, the acquisition of science concepts” (p. 673). According to another researcher, Vosniadou (2012), students' conceptions are caused and supported by epistemological and ontological presuppositions that must be replaced during conceptual change. Finally, for Chi (1994), conceptual change necessitates an important reorganization of learners' knowledge structures in which concepts must change their meaning and be reassigned to a different ontological category.
As argued by Shtulman and Valcarcel (2012), implicit to the idea of knowledge restructuring is the idea that learners' initial knowledge systems, which are assumed to have been considerably altered during conceptual change, no longer exist after a conceptual change, since they have been replaced by something else. Indeed, in most conceptual change models, the coexistence of scientific and nonscientific conceptions in learners' minds is either implicitly rejected or presented as an intermediate step, suggesting that conceptual change is occurring, but has not been achieved yet (see Posner et al., 1982).
Recently, a number of findings have challenged the idea that misconceptions disappear definitively after a conceptual change. For instance, two studies have pointed out that naive modes of thought about how nature works, usually only common during childhood, can re-emerge later in life. Indeed, seniors with a decreasing inhibition capacity due to Alzheimer's disease return to teleological explanations (Lombrozo, Kelemen, & Zaitchik, 2007) and animist thinking (Zaitchik & Solomon, 2008) to explain how nature works. Moreover, other studies have shown that healthy adolescents (Babai & Amsterdamer, 2008), adults (Shtulman & Valcarcel, 2012), and even professional scientists (Goldberg & Thompson-Schill, 2009; Kelemen, Rottman, & Seston, 2012) need more time to correctly answer questions related to misconceptions, as if they needed to inhibit (i.e., control, deactivate, or suppress) a spontaneous and appealing, albeit wrong, answer. All of these studies suggest that misconceptions and naive thinking about nature have possibly never disappeared from the brains of adolescents, adults, seniors, and professional scientists and, therefore, need to be inhibited.
Although the idea that learning certain scientific concepts requires fighting against our natural tendencies and intuitions is not new (Bachelard, 1938), the concept of inhibition has rarely been used in conceptual change research. There are, however, a few notable exceptions. For instance, Kwon and Lawson (2000) have shown that, among different students' characteristics (such as planning ability, age, disembedding ability, and mental capacity), the ability to inhibit is the single best predictor of the capacity to improve their understanding of scientific concepts related to air pressure. There is also a pilot functional magnetic resonance imaging (fMRI) study, conducted by Dunbar, Fugelsang, and Stein (2007), that shows differences in brain activation between novices and scientific experts when they evaluate the correctness of videos showing balls of different sizes falling at different speeds. According to the authors' interpretation, the results support the idea that experts still need to inhibit the misconception that “heavier balls fall faster,” even if they most likely overcame it several years earlier.
The concept of inhibition refers to the cognitive ability to resist a habit or a spontaneous and tempting response or strategy. At a neural level, it refers to the capacity of a neural network to deactivate another neural network that would otherwise be activated. A number of neuroimaging studies have used cognitive tasks (Stroop, counting Stroop, go/no-go, etc.) where inhibition is needed to overcome a prepotent but inappropriate response. These studies have shown that the anterior cingulate cortex (ACC), the ventrolateral prefrontal cortex (VLPC), and the dorsolateral prefrontal cortex (DLPC) are more activated when inhibition is required (Bush et al., 1998; Menon, Adleman, White, Glover, & Reiss, 2001; Monchi, Petrides, Petre, Worsley, & Dagher, 2001; Nathaniel-James, Fletcher, & Frith, 1997). According to Botvinick (2007), the ACC is associated with error detection and decision making. It detects that a particular situation or task requires higher cognitive control. Consequently, the ACC may be the brain region that triggers the inhibition process, while the DLPC and the VLPC may be more directly responsible for the inhibition of spontaneous answers or strategies (Aron, Robbins, & Poldrack, 2004; De Neys, Vartanian, & Goel, 2008; Liddle, Kiehl, & Smith, 2001; MacDonald, 2000).
This article aims to contribute to the debate about what happens to misconceptions after a conceptual change (Shtulman & Valcarcel, 2012). Are they rejected and replaced by scientific conceptions, or are they still present in the minds of students, coexisting with newly acquired scientific conceptions? In order to answer this question, we propose to use fMRI to observe the differences in brain activation between novices (with misconceptions) and experts in science (seemingly without misconceptions because they answered correctly) when they respond to questions involving a common misconception. On the basis of studies discussed previously, we hypothesize that scientific experts will show more activation than novices in brain areas involved in inhibition such as the ACC, the DLPC, and the VLPC (Bush et al., 1998; Menon et al., 2001; Nathaniel-James et al., 1997) because they need to inhibit a misconception that is still encoded in their brains' neural networks.
Since students' misconceptions in electricity are among the most frequent and persistent (Wandersee et al., 1994), and because brain mechanisms of expertise in that domain have never been studied, we have chosen to identify the differences in brain activation between novices and experts during a task based on a common misconception in electricity. This misconception, especially frequent at the beginning of conceptual change, states (wrongly) that only one wire is sufficient to light a bulb (Çepni & Keleş, 2006; Periago & Bohigas, 2005). The task will be described in the next section.