VanLehn gives a very holist view of what models can look like, how students can use them, how these models can be assessed, and what difficulties might arise in the process of creating these models. The first and most important distinction he makes is the different connotations of "modeling" in science and math versus in the science education community, "the term 'modeling' is often used more broadly to encompass any educational activity that involves a model" (373). However, for models to be truly useful, we must focus on actually creating models like we are for the Zika virus.
One aspect of VanLehn's that I found particularly insightful was his distinction in activity types based on the type of model being made and the way the students learn about the system. This second category reminds me of Hestenes' point that often times nature may conduct experiments for us, but useful models can still be made from these examples. It seems as if a student could incorporate several presentations of systems when modeling Zika. First, while VanLehn's "resources" give the student the full information, creating the Zika virus model requires a student to know about the modes of transmission, standing water, ways to reduce risk, etc. and all of that information could be provided as a resource. The second useful strategy is virtual field studies- this is inherently what our models are all about. Furthermore, over time students could be able to test their models against real field studies as they become available and thus their models can be enhanced and corrected if necessary.
In regards to the actual types of models students could construct, VanLehn offers three classes: constraint systems, system-dynamics, and behavior models. Here also various categories could sufficiently guide the modeling of the Zika virus. A constraint model test the outcome if one aspect of a system is disrupted. Here is where students could test the effectiveness of risk reduction methods or allow the students to isolate certain variables such as standing water, or daylight hours to see which aspects have a large role in the spread of the virus. Secondly, system-dynamics is only slightly different than a constraint model because system-dynamics can predict behavior over time- the essence of our interest into the Zika virus. VanLehn acknowledges that these models can become very complex as diagrams, which is why they can also be expressed in text. Finally, agent-based models allow the most freedom to control for the greatest number of factors and works quite easily with StarLogo Nova (our preferred modeling site).
One thing I have struggled a lot with is the issue of assessing models, which especially for a case like Zika there may not be one purely correct answer because we simply do not know the full affects yet. That is what I especially liked about VanLehn's paper- his attention to modes of assessment and what to look at. The identification of recognizing faulty behavior, troubleshooting, and extensions as the main aspects to base assessment off of really makes sense to me. Personally, I think it is the extension aspect that truly emphasizes the importance of models, through modeling students can learn about all the pieces of science and nature that interact within a system and how these systems can affect our lives.