null Experienced programmers watch different

OW_EyeTracking_15002_head_large.jpg
Digital Learning
Experienced programmers watch different
When removing errors from a code, the eye movements of an experienced computer programmer differ from those of a beginner. If the expert gets the assignment to transfer his knowledge of the wrong code to the beginner via an instruction video, he adjusts his viewing behaviour. These are two preliminary results from the research project 'Expertise-related differences in gaze during code debugging'.

Educational eye movements

PhD candidate Selina Emhardt of the Open University has investigated in close cooperation with postdocs from Utrecht University whether, when watching a wrong code, an experienced programmer’s eye movements are different from a beginner’s ones. And if so, how the expert's gaze changes when he or she performs this task with an educational goal. Both hypotheses appeared to be largely true within the test group of eighteen experts (two women) and 23 beginners (seven women).

Indirect cues

Follow-up research should now show how this knowledge about didactic eye movements of experts can best be used in instructional videos. Selina Emhardt: 'What works best in the learning process? We want to investigate the practical implications for learning when using this kind of indirect cues.’ The researchers use 'eye tracking' technology to follow the eye movements of an expert programmer. The movements are converted into visible points, lines and circles. Thus, beginners can view an instructional video with the gaze of an experienced programmer. A direct cue would be for example pointing out something.

Instructional videos

Instructional videos are increasingly used in learning complex, problem-solving tasks, such as programming. With their research, scientists from the Open University and Utrecht University want to improve education via instructional videos using Eye-Movement Modeling Examples (EMMEs). The research is part of the project You see? Effects of establishing joint attention in pre-recorded and real-time modeling examples on learning, which is part of the NRO research program for fundamental education research (PROO).