In Physics Education, Perception Matters

Student difficulties in science learning are frequently attributed tomisconceptions about scientific concepts. We argue that domain-general perceptual processes may also influence students’ ability to learn and demonstrate mastery of difficult science concepts. Using the concept of center of gravity (CoG), we show how student difficulty in applying CoG to an object such as a baseball bat can be accounted for, at least in part, by general principles of perception (i.e., not exclusively physics-based) that make perceiving the CoG of some objects more difficult than others. In particular, it is perceptually difficult to locate the CoG of objects with asymmetric-extended properties. The basic perceptual features of objects must be taken into account when assessing students’ classroom performance and developing effective science, technology, engineering, and mathematics (STEM) teaching methods. High school and college students often encounter difficulties grasping and applying concepts in introductory physics classes. Physics educators and researchers have responded to this challenge with the development of the field of physics education research (PER; McDermott, 1991). Within PER, students’ difficulties are frequently attributed to misconceptions about physics concepts and addressed through teaching methods including tutorials (McDermott, Shaffer, & the Physics EducationGroup at theUniversity ofWashington, 2002) and interactive lecture demonstrations (Sokoloff & Thornton, 2004) designed to explain difficult physics concepts in new or varied ways. Rarely, however, are basic and domain-general (i.e., not exclusively physics-based) mechanisms of perception thought about as a hindrance to physics learning. The field of cognitive science (including psychology and neuroscience) has identified several general 1Department of Psychology, The University of Chicago 2Department of Physics, DePaul University Address correspondence to Sian L. Beilock, Department of Psychology and Committee on Education, The University of Chicago, 5848 S. University Avenue, Chicago, IL 60637; e-mail: [email protected] principles of perception. Could domain-general perceptual processes also influence physics learning? If so, this would suggest rethinking certain physics teaching methods to include consideration not only of physics concepts but also of basic learning strategies allowing for more accurate perceptual understanding. Here, we use the concept of center of gravity (CoG), often referred to as center ofmass,1 to explore the role of basic perceptual features in student understanding. Several classroom studies involving CoG have been published (Brose & Kautz, 2011; Liby, Friedenberg, & Yancopoulos, 2009; Ortiz, Heron, & Shaffer, 2005). We focus on research by Ortiz et al. (2005), who used college physics and engineering classroom data to show that student performance on CoG questions and related topics involving extended objects (e.g., a balanced baseball bat) with an unequal or “asymmetric” mass distribution was much poorer than on similar questions involving discrete-asymmetric objects (e.g., two nonidentical crates on opposite sides of a seesaw). Originally interpreted as evidence that students have an incomplete understanding of CoG, it is also possible that there is something inherently different in the perception of discrete objects versus extended objects that contributes to poorer performance on the latter.That is, differences in students’ performance on the CoG problems mentioned above may—at least in part—be due to general perceptual features that make it harder to determine the CoG of some objects relative to others. In support of this idea, Redish (2004) suggested that students may perceive an extended baseball bat differently than the discrete crates and ignore their new, formal physics knowledge about CoG. In other words, difficulties solving the baseball bat CoG problems (and similar asymmetric, extended objects akin to the baseball bat) may be driven by general perceptual features of items, not simply subpar physics knowledge. Previous research has shown that individuals’ ability to locate CoG changes based upon the perceptual features of objects. For instance, symmetry plays a large role in the ability to accurately locate CoG. For extended shapes, such as triangles, rectangles, and polygons (Davi, Doyle, & Proffitt, 1992), it is perceptually easier to determine the CoG of 164 © 2015 International Mind, Brain, and Education Society and Wiley Periodicals, Inc. Volume 9—Number 3 Jason R. Sattizahn et al. symmetric relative to asymmetric objects. For discrete systems of multiple dots, greater reflective and rotational symmetry results in increased accuracy in locating the system’s CoG (Liby & Friedenberg, 2010). Orientation and size of objects also impact the difficulty of locating an object’s CoG for extended shapes (Bingham&Muchisky, 1993), as do size ratio, separation, and orientation of systems of discrete dots (Friedenberg&Liby, 2002, 2008).TheOrtiz et al. (2005) findings described above for the baseball bat and crates suggest that the extension of an object or system (i.e., whether it is discrete or extended) may also contribute to students’ ability to accurately locate the CoG of an object. In the current work, participants were recruited to take part in a CoG finding task outside of a physics classroom. By providing participants with a simple definition of CoG, and by varying the perceptual features contained by objects on the CoG finding task, we effectively removed our CoG finding task from a physics context to explore how and if general perceptual features impact CoG finding.We crossed the perceptual features of symmetry (i.e., symmetric vs. asymmetric) and extension (extended vs. discrete) in our CoG finding task (see Figure 1). We hypothesize that problems requiring determination of the CoG for an object that is both asymmetric and extended may be most perceptually demanding for students because solutions require simultaneously taking into account two perceptual features (i.e., symmetry and extension). Our goal was to demonstrate how basic perceptual features of objects might explain poor performance observed when students work on CoG questions and related topics in the physics classroom. While our CoG finding task was intentionally performed outside of a physics context, we did gather information on the extent of our participants’ previous physics exposure (i.e., number of physics classes taken in college). This information allowed us to explore if participants’ previous physics education related to performance on the CoG finding task. By controlling for participants’ physics experience, we provide additional evidence that it is the perceptual features manipulated within the experiment (rather than conceptual physics knowledge per se) that are responsible for our results.