Teaching students how programmers think.

Wednesday, February 20, 2019 - 11:00
TH 434
Dastyni Loksa

I believe all students who want to learn computer science can. However, many students, particularly those from underrepresented populations, lack valuable experiences that contribute to adopting a computer science mindset. We can address this by explicitly teaching how to think like a computer scientist which can lead to more productive, confident, and independent students. The question, then, is how do programmers think when solving programming problems, and how can we teach students to do the same? In this talk I will present my research on understanding the metacognitive skills and behaviors that programmers engage in while solving programming problems. I will also present my work on explicitly scaffolding metacognition and self-regulation skill development for novice programmers.


Dastyni Loksa is a Ph.D. candidate from the University of Washington Information school. He has worked for several technology companies including VeriSign, Bandai Namco, and Signio and brings a perspective of computer science primarily as an opportunity for innovation and beneficial cultural impact. He believes computer science is an inherently creative discipline and that the field would benefit greatly by facilitating the inclusion of creative thinkers who might otherwise pursue fields like art or design. To this end, his research is focused on understanding how programmers think while solving programming problems and how we might explicitly teach problem solving for programming. He has investigated how programmers’ metacognition (thinking about their own problem solving process) develops and how it impacts their programming success. He has also found that teaching a problem solving framework for how to think about programming problems and scaffolding practice using it creates more productive, independent, and confident students. He has published research at the International Computing Education Research (ICER), Computer Science Education (SIGCSE), and Conference on Human Factors in Computing (CHI) conferences.