SHIFT

An Interdisciplinary Framework for Scaffolding Human Attention and Understanding in Explanatory Tasks

SHIFT is an adaptive computational framework designed to improve human understanding in interactive tasks through adaptive verbal scaffolding. By analysing human cognitive state in real time - incorporating task awareness, processing capacity and interaction history - SHIFT selects personalised verbal strategies such as affirmations, negations and hesitations to support more natural, effective human-robot communication. SHIFT is implemented in Python and has a ROS interface that allows real-time data exchange for monitoring and querying the model. It also includes a graphical user interface designed to make the model's decision-making process transparent and interpretable.