Bound to Fail: Challenges Faced in the Design of Molecular Level Visualizations

Resa Kelly, San Jose State University, Department of Chemistry

What compromises should the people creating chemistry-related animations make?

The nature of matter at the particulate level is certainly more complex than we can show in a computer animation and probably more complex than we are capable of imagining. For the people creating computer animations that illustrate the particulate nature of matter, this conclusion can be liberating, but also frustrating. If we are bound to fail in our animation of reality, we can relax and just focus on trying to illustrate key ideas that we would like to illustrate for our chemistry students. However, the impact of the animations, if done properly or improperly, can have a lasting impact on the students' conceptions. Animators need to try to answer a few important questions. If a simplified depiction of a complex molecular level event is portrayed in the visualization, how will it affect students' understanding? Will the animation lead to enhanced understanding or will it lead to misconceptions? If we animate the complexity of the molecular level event as accurately as possible, will the students be able to make sense of it? The purpose of this paper is to start a dialogue among chemical educators about the necessary compromises animators must make.

We (Mark and Resa) have separately created animations that illustrate precipitation reactions, so we will use our animations as examples of the decisions that we made.

Complex or Simple: How much complex and dynamic character should we try to capture?

Mark: My first thought is that we should try to illustrate as much of the dynamic nature of particle behavior as possible within the constraints of computer processing power and animator time and patience, but when I think about this more realistically, I recognize that it is possible to do too much. The closer we get to what we think is reality, the more cluttered and confusing the animation. The most important thing is to focus on the key ideas we want to present and ask ourselves how we can best illustrate these ideas. The trick is to find the best balance of cluttered chaos on one hand and simple, clear illustrations of key concepts on the other. Because I created my animations to accompany my textbook (An Introduction to Chemistry), I've decided to keep more chaos than I would have otherwise. Readers of my text will have seen still images in the book before they view the animations. You can see the still images for precipitation reactions at
http://preparatorychemistry.com/Bishop_Book_atoms_7.pdf

Despite this, it's on my to-do list to have back-to-back animations for each step in my presentation, one as cluttered and chaotic as possible and a simplified version of the same thing that shows the key idea or ideas more clearly.

Resa: In order to determine how much complexity to incorporate into animations, I feel that one has to carefully examine the perspectives of both experts and novices. Recently, I studied the key features that thirteen instructors incorporated in their drawn explanations of precipitation reactions, as well as the kinds of features they orally described but were unable to illustrate. In addition, I examined the kinds of misconceptions twenty-one first year chemistry students commonly had about the same reactions so that these could be addressed by the visualization. In this research, instructors’ and students’ drawings of the following three chemical equations were analyzed to determine the key features instructors depicted in terms of structure and function and the misconceptions that students depicted.

AgNO₃(aq) + NaCl(aq) -> AgCl(s) + NaNO₃(aq)
KNO₃(aq) + NaCl(aq) -> No Rxn
MnCl₂(aq) + 2AgNO₃(aq) -> 2AgCl(s) + Mn(NO₃)₂(aq)


Ultimately, it was determined that instructors’ explanations involved the following events: (1) The nature of the aqueous reactant solutions prior to mixing, (2) a view of the reactant species immediately after mixing, but prior to product formation, (3) the dynamics of the reacting species and non-reacting species colliding and forming aggregates, and (4) the formation of a precipitate. It was also noticed that instructors included water molecules in hydration spheres surrounding ions, but many did not incorporate the solvent water molecules.

As mentioned previously, first year chemistry students were asked to draw and explain their atomic level understanding of the same three equations to learn the kinds of misconceptions they had so that these wrong ideas could be pointedly addressed in the animations. The findings (in press) indicated the following misconceptions: Aqueous reactants exist as molecular pairs prior to mixing (12 of 21), upon mixing reactant molecules break apart (10 of 21), the precipitate (21 of 21) and aqueous product (8 of 21) were represented as molecular pairs. Ultimately, six students believed that the reactant molecules separated into ions then formed molecules for products.

The results of theses studies were used to develop the following molecular level animations:

Electronic Learning Tool

Mark: On the particle level, things move very quickly, and if we were able to accurately show this movement, our animations would be frantic. I tell my students that if they don't get at least slightly tense when they try to visualize the motion of particles, they just aren't thinking hard enough. I don't want to cause seizures in folks trying to follow my fast moving images on the computer screen, but I do try to give the viewers a sense that particles are moving very quickly, colliding with other particles and changing their velocity and direction many times per second. I think that within the limits of computer processer speed, computer memory, and the viewers' abilities to follow fast moving particles, at least portions of our animations should be as fast and frantic as possible. Although many students surely prefer slow and simple, I think it's OK to make it slightly challenging to see what's happening. All of the segments of my animations either loop or can be repeated to allow the viewer to take enough time to see the key points.

Resa: I prefer to depict things at a rate that is slower than authentic conditions, and I’ve noticed that students prefer to see slower motion as well. It allows them to focus on the events. In the past, I recall that some students would find the fast pace of moving atoms to be distracting, suggesting it just looked like a bunch of balls moving around. With a slower animation, students can focus more readily on the processes without feeling overwhelmed, and nearly all of the students that I have interviewed have indicated that they are aware when an animation is showing slower moving atoms. They usually state that they do not mind the slower pace and actually find it helps them follow what is happening. In summary, I believe that slower movement of the microscopic event is helpful for learning.

Mark: I think consistency is very important. Because I've written my own text, made the textbook illustrations, and created the computer animations, they are consistent in the way particles are shown and the language used to describe these particles. See

http://preparatorychemistry.com/Bishop_Book_atoms_7.pdf
http://preparatorychemistry.com/precipitation_flash.htm
http://preparatorychemistry.com/Bishop_Section_7_3_atoms.htm

Resa: I believe that consistency is important, but I also think that we can teach multiple perspectives. In my animations, I provide both two dimensional and three dimensional views. With the two dimensional view, it is easier to show interactions between ions in aqueous solution, but with a three dimensional view the students are able to see how water completely surrounds ions. Both perspectives are important and with buttons, students can see these two perspectives separate from each other.

Mark: Some people prefer to be passive viewers, but I think that most people like to play. Many of my tools have a series of questions coupled with the animations to get some participation.

http://preparatorychemistry.com/element_properties_flash.htm

Resa: According to Hegarty (2004), interaction can be distracting and ultimately shift the students’ focus from making sense of the illustrated events. The student may muscle through the motions to get to the end or to the answer they seek without comprehending what they are viewing. Thoughtful interaction is important, but extremely challenging to construct. In summary, the user needs to be involved enough that they remain attentive and aware of what the animation is trying to teach them. Of course this is easy to state, but very difficult to construct.

Particles shown in textbooks and in animations are usually described as solid spheres with highlights to create the illusion of three dimensions, but these don't give a true representation of the electron probability distribution.

Maybe we should consider changes to images that represent the electron probability cloud more accurately and to try to show the diffuse boundaries of the atom.

Mark: I've been moving to "fuzzies". I'm a little worried that because people are most comfortable with the familiar solid-looking spheres, when they see something that is not described in the traditional way, they will think it looks odd, but I'm OK with some oddity

http://preparatorychemistry.com/element_properties_flash.htm

Resa: I am inclined to agree that showing representations that more accurately convey the nature of the electron cloud would be very helpful. Unfortunately, this seems to be a limitation of the technology that isn’t always design friendly. I find that many students are not certain of how to draw atoms, thus the solid looking spheres that most instructors are familiar with may be an issue for the students. Some students have a hard time seeing a colored sphere because they have not yet learned to recognize that colors can be used as symbols of elements. In recent interviews with 17 chemistry students viewing an animation of precipitation reactions, one student requested to have the symbols placed on the spheres as labels to help her more quickly recognize the element being depicted. The question is, does the student need a label or is she more comfortable with a letter symbol or structural formula representation? Many expert instructors, myself included, draw their understanding of molecular events using symbols and structural formula representations to illustrate concepts. I prefer to use structural formulas only because my artistic skills are lacking and the structures are still useful for conveying the molecular processes I wish to address.

It's important that our animations fit on one screen for most computers, but there's a compromise here too. If we program for fitting our animations on one screen for 640 x 480 screen resolution, the animations are very small at higher resolutions. If you go for best fit at higher resolutions, the users with lower resolution screens won't see all of the animation without scrolling up and down.

Mark: The screen resolution 640 x 480 is pretty much history, but 4-5% of computers are still using 800 x 600. I've decided to be sure that my animations fit on one screen for 800 x 600 screen resolution. I'm pretty close to going to 1024 x 780 for the low end resolution for which my animations fit in one screen. Resa: I am not a fan of super small animation screens, yet I don’t like the looks of an enlarged screen. The resolution looks grainy and it feels overwhelming, kind of like being in the front row at a movie theatre. A Goldilocks compromise seems necessary.

In closing, we hope that this paper leads to further discussion on design principles for the development of successful, instructional visualizations so that our next paper can be titled: Bound to Succeed!

References: Kelly, R.M., Barrera, J. H., & Mohamed, S. C. (in press) An Analysis of Undergraduate General Chemistry Students' Explanations of the Submicroscopic Level of Precipitation Reactions. J. of Chem. Ed.
Hegarty, M. (2004) Dynamic visualizations and learning: getting to the difficult questions..Learning and Instruction (14) 343-351.

Return to Fall 2009 CCCE Newsletter