neural_nets_chapter has 2 private notes and 0 public notes for this article.
If you are neural_nets_chapter then you can log in to see the private notes.
In this commentary, we focus on two aspects of the target article. The first is the decision to separate the gaze following model into two trainable components, a ‘When’ component to determine when to shift gaze and a ‘Where’ component to determine where gaze should be shifted to. The second is the lesson that the authors draw regarding multiple causality in developmental disorders, that is, from the finding that very different computational causes can lead to similar deficits in the emergence of gaze following. In that context, we assess a version of the gaze following model given Attention Deficit/ Hyperactivity Disorder (ADHD).