You are here:

The effect of computer visualizations on students' mental models of dynamic nature of physical equilibrium

, University of Northern Colorado, United States

University of Northern Colorado . Awarded


Dynamic computer visuals have been developed to improve the understanding of chemistry. Mental models are small scale representations of reality that are constructed from perceptions or experiences (Craik, 1943). Novice learners possess mental models that may differ from those of experts, which better represent scientific reality (Osborne, 1982). What experts learn from a simulation may not be the same as for novices, thus novices may need additional supplementary support for more effective use of simulations.

In this study, first, the mental models of physical equilibrium held by experts and novices were determined and compared. Second, the effects of computer lessons that included a computer visualization of physical equilibrium and a supplementary worksheet, either more or less guided, on mental models of novices were investigated. The study was composed of two major sections: a study of solubility equilibria (the SE-Study) and a study of liquid-vapor equilibrium (the LV-Study). Ten high school students and 32 second-semester college general chemistry students, four university chemistry instructors and two high school chemistry teachers participated in the SE study. One hundred ninety one first-semester general chemistry students, one university chemistry instructor and one high school chemistry teacher participated in the LV-study.

In the first stage of the SE-study the mental models of expert and novices were identified and the key features they emphasized in the open-ended content questionnaire and the individual interviews were compared. The results of the analysis showed that the macroscopic features of novice mental models were similar to those of experts but the molecular features were different; the novices were less likely to emphasize the dynamic nature of equilibrium and had specific misconceptions. In the second stage of the SE study, mental models of the dynamic nature of solubility equilibria held by novices were investigated before and after using a simulation or a screen-captured animation of the same simulation. Students were randomly assigned to work with one of two types of supplementary worksheets, more or less guided, during the lesson. The key features extracted from the mental models of novices before and after the implementation were coded and analyzed by Chi-square analysis. Novices tended to use molecular features significantly more frequently (p<.01) in their final mental models than in their prior mental models. Based on the responses of novices to the open-ended questionnaire, the Mental Models of Equilibrium Scale (MMES) was developed to measure conceptual understanding of equilibrium at the molecular level. On the MMES, 1 was defined as poor understanding and 5 was defined as good or expert understanding of physical equilibrium. The comparison of mental models of novices before and after the implementation showed that their mean score on MMES after the implementation was significantly higher (p=.002) than the mean score earned prior to the implementation.

In the LV-study students worked with a simulation of liquid-vapor equilibrium supported by either a more guided or less guided worksheet. Novice mental models of dynamic physical equilibrium, as assessed by the MMES, improved significantly (p=.000) after working with the computer lesson.

The changes in students' mental models of equilibrium were assessed not only by the MMES, but also by the conceptual pre- and post-tests of solubility and liquid-vapor equilibria (CSPrT, CSPoT, CLVPrT, and CLVPoT) in each study. Even though students' scores on the mental models of equilibrium scale were improved significantly after the implementation, no significant difference was found between mean scores on the conceptual pre- and post-tests, which consisted of multiple choice and true/false types of questions. These results suggest that when students are asked to express themselves in their own words or drawings, changes in their mental models are better elicited than they are asked to select the correct answer for a particular situation.

In both of the studies, some students corrected their misconceptions and some gained new misconceptions from the visualizations. In the SE-study the interactivity of the solubility simulation was not found to affect the mental models of novices differently from the animation of solubility equilibria. In addition, no significant difference in mental models of the dynamic nature of physical equilibrium was found between the groups using worksheets of different guidance levels in either the SE- or LV-study. However, in the LV-study comments about the worksheets on the evaluation questionnaire were more positive for students who had used the open-ended version of the worksheet. The cognitive loads of the visualization and the worksheet perceived by the novices were negatively correlated with their prior knowledge, suggesting that students who had higher prior knowledge found the visualization and the worksheet easier to understand. These students also achieved significantly higher scores on the MMES after the implementation. (Abstract shortened by UMI.)


Akaygun, S. The effect of computer visualizations on students' mental models of dynamic nature of physical equilibrium. Ph.D. thesis, University of Northern Colorado. Retrieved March 22, 2019 from .

This record was imported from ProQuest on October 23, 2013. [Original Record]

Citation reproduced with permission of ProQuest LLC.

For copies of dissertations and theses: (800) 521-0600/(734) 761-4700 or