I liked the last 3 sections before the conclusions on assessment, student difficulties, and scaffolding. Given how important it is to consider these things ahead of time when planning model construction activities, such as the one we are doing for Zika, I appreciated Van Lehn's presentation of these.
I thought that the assessment section was very interesting. The question of how we can assess our students' modeling skills and understanding of the scientific concepts seemed to be rather difficult prior to reading this passage. It is good to have examples of possibilities aside from in class and written discussions/defenses of their models. In modeling, like in everything we do as teachers, I think it is important to stress that we need to define our learning objectives first before deciding how our modeling activities and assessments of them will look, and I appreciate Van Lehn's discussion of some objectives for which modeling can be a good activity. For the Zika modeling activity, we would need to consider how we will assess both our students' understanding of the scientific concepts we are discussing as well as their understanding and use of modeling.
I think that the misconceptions and scaffolding possibilities presented offer some great insights into our students' minds when it comes to how we can actually go about doing this. Presumably, when we start teaching, our first classes will be our students' first classes or among their first classes to focus on modeling. Therefore, having an idea as to the types of misconceptions they will likely bring to the activity of modeling itself on top of their misconceptions about the content will be useful. In the same vein, the wide variety of scaffolds presented in the article will help us help our students learn how to model. I would especially like for us to discuss meta-tutoring further in class, perhaps see it in action, because of the empirical evidence for its success.