Do perceptually salient elements in physics problems influence students' eye movements and answer choices?

Abstract

Several reasons have been proposed to explain students' incorrect answers to conceptual physics problems. Heckler [3] proposed with a perceptual basis: plausible and salient "eye catching" features in a problem capture students' attention. Once students attend to these perceptually salient features, less salient albeit thematically relevant features are not considered and students answer the problem incorrectly based on the salient features. To test this hypothesis we recorded eye movements of introductory physics students on 15 conceptual problems with diagrams. Each diagram contained areas consistent with documented novice-like answers and other areas consistent with the scientifically correct answer. We manipulated the luminance contrast of the diagrams to produce three versions of each diagram, which differed by the area with the highest level of perceptual salience. We found no effect of the salience on the correctness of students' answers. We also discuss how the salience manipulations influence eye movements.

Linking attentional processes and conceptual problem solving: visual cues facilitate the automaticity of extracting relevant information from diagrams

“…Nevertheless, there is a critically important difference between the results of our studies and those of Grant and Spivey (2003) and Thomas and Lleras (2007 , 2009 ), who proposed the provocative idea that simply having the viewer’s low-level attentional movements embody a problem’s solution is sufficient to facilitate finding the correct solution. Specifically, our research, including both the current and previous studies ( Madsen et al, 2013a , b ) has shown that while attending to relevant information in a problem is a necessary condition for correctly solving the problem, it is generally not sufficient to correctly solving it. The current study has specifically shown that cues, which both draw attention to solution-relevant information, and facilitate organizing and integrating it, facilitate both immediate problem solving and generalization of that ability to new problems.…”

“…Prior research on eye movements and problem solving has shown that overt attention can illuminate the cognitive processes involved in problem solving ( Epelboim and Suppes, 2001 ; Knoblich et al, 2001 , 2005 ; Grant and Spivey, 2003 ; Jones, 2003 ; Thomas and Lleras, 2007 , 2009 ; Bilalić et al, 2008 ; Eivazi and Bednarik, 2010 , 2011 ; Madsen et al, 2012 , 2013a , b ; Lin and Lin, 2014 ; Susac et al, 2014 ). However, we are particularly interested in two directions of causal relationships between overt attentional selection and the higher-level cognitive processes involved in problem solving: (1) the causal relationship starting from higher-level cognitive processes involved in problem solving and ending with attentional selection; and (2) the reverse causal relationship starting from attentional selection and ending with the higher-level cognitive processes involved in problem solving.…”

“…This raised the question of whether guiding solvers’ overt attention to the relevant information would facilitate their correctly solving those or similar problems. In one study, we modified the bottom-up visual saliency (as measured by a computational model) of the relevant vs. irrelevant regions in physics problems (by increasing or decreasing the luminance contrast of the lines in the problem diagrams; Madsen et al, 2013b ). Interestingly, we found that solvers’ mandatory top-down processes (prior knowledge) guided their overt attention, overwhelming any potential effects of stimulus saliency ( Madsen et al, 2013b ).…”

“…In one study, we modified the bottom-up visual saliency (as measured by a computational model) of the relevant vs. irrelevant regions in physics problems (by increasing or decreasing the luminance contrast of the lines in the problem diagrams; Madsen et al, 2013b ). Interestingly, we found that solvers’ mandatory top-down processes (prior knowledge) guided their overt attention, overwhelming any potential effects of stimulus saliency ( Madsen et al, 2013b ). Nevertheless, as before, those who attended more to relevant information were more likely to correctly solve the problems ( Madsen et al, 2012 ).…”

Comparing retrieval-based practice and peer instruction in physics learning

“…Initial tasks used in this study were from the Force Concept Inventory (FCI) validated by Hestenes and collaborators [31]. Near and far transfer tasks used in this study had been validated in previous studies [32][33][34][35].…”