In the same way that a book does not read itself, multimedia representations of teaching are only the beginning of a cognitive and educational chain that can lead to improved teaching and learning. I believe that considerations of how viewers (particularly current and prospective teachers) make sense of MRTs and apply them to improving their own practice can help us to design effective MRTs.

MRTs have enormous potential for capturing the dynamics of teaching and learning in classroom settings in ways that will allow us to focus on processes and to create vivid materials that can be useful to practitioners. They can also create a priceless historical record that can serve as a benchmark for understanding educational change over time. I’d like to say a bit about the use of these materials in teacher development.

Research in athletics and in a few complex domains (such as the training of pilots) has shown that experience viewing situations that might arise in the course of performing that skill can help speed the acquisition of skill. So what obstacles stand in the way of our doing likewise? Here are three:

1. 1.The illusion of iconicity. There is a strong tendency to believe that we see what’s there to be seen, something easily disproved by the “inattentional blindness” research of Simons and colleagues (e.g., Simons & Chabris, 1999), showing that viewers reliably miss major aspects of complex events.
   
   

2. 2.Cultural biases. Americans show a strong bias to assume that we do what we do because of the kind of person we are (termed the “fundamental attribution error” by Ross, 1977). In our work (Miller & Zhou, 2007), American viewers watching classroom video tend to comment on teacher characteristics, whereas Chinese viewers are much more likely to notice features of the lesson being taught. This is a problem for learning from classroom video, because the kinds of things Americans notice (e.g., whether the teacher is warm, confident, etc.) are not the kinds of things you can probably improve about yourself by watching MRTs.
   
   

3. 3.Point of view matters. Two studies (Farrow, Chivers, Hardingham, & Sachse, 1998; Williams, Ward, Knowles, & Smeeton, 2002) has shown that watching video representations of tennis serves can improve your ability to return serves when playing that game. But what they show their budding tennis players differs notably from what we show prospective teachers. The video shown prospective teachers is typically shot from the student’s point of view, something that is comfortable and familiar to most viewers. But it doesn’t show what teachers see, which presents a huge potential stumbling block to applying what they see in MRTs to classroom contexts. We are currently collecting teacher-perspective video, both to understand how teachers manage to attend to a classroom of students, and also to produce materials that may be more effective in helping prospective teachers to learn what to see in a classroom in a context similar to what they will see as they teach.
   

Multimedia Representations of Teaching hold enormous promise in teacher education. Cashing out that promise will require us to take the problem of viewing seriously as a complex cognitive task in its own right, and producing MRTs and viewing tasks that will facilitate a focus on aspects of teaching that are amenable to improvement (e.g., questioning or explaining strategies as opposed to global personality features), and presented in a way that will maximize transfer to classroom contexts.