Include Neurodiversity in Foundational and Applied Computational Thinking

• In Event: Emerging Technologies for Supporting Neurodiverse Learners

Abstract

Funded by the US Department of Education’s Education Innovation and Research Program, the INFACT project aims to Include Neurodiversity in Foundational and Applied Computational Thinking (CT). An objective of this work is to promote productive persistence for neurodiverse learners by embedding supports for executive function into online and offline CT learning activities for grades 3-8.

A consortium of university and non-profit partners are working on INFACT for the design, development, implementation and research of a comprehensive set of teaching and learning materials for grades 3-8. INFACT introduces foundational CT practices such as problem decomposition, pattern recognition, abstraction, and algorithm design (Shute, Chen, and Asbell-Clarke, 2017) through learning games, robotics, and introductory coding activities. Activities focus on computing structures such as conditionals, loops, variables, and functions through daily life examples and application in robotics, coding, and other CT domains.

INFACT pays particular attention to the inclusion of a wide range of learners, especially neurodiverse learners. INFACT recognizes that each learner brings a unique set of learning assets as well as cognitive, social, and emotional learning needs (Immordino-Yang, Darling-Hammond, and Krone, 2018). To allow neurodiverse learners to demonstrate their individual strengths in problem solving, INFACT provides supports for attention and working memory so that they can thrive in areas where they may be strong such as systematic thinking and detailed pattern recognition (Baron-Cohen, 2020; Shmulsky, Gobbo, & Bower, 2019).
In the design research phase of INFACT, we have embedded a set of scaffolds for executive function within the CT learning game called Zoombinis. Zoombinis is a series of logic puzzles that require learners to decompose a problem and find solutions using conditional and algorithmic thinking. The scaffolds we’ve designed include graphical organizers that help learners keep track of the information they gather during the puzzle solving as well as a flashlight tool that highlights salient information on the screen. These tools do not help with the logic to solve the puzzle, just with the attention and working memory needed to persist with the puzzles productively.
In previous research, we were able to build and validate Educational Data Mining (EDM) detectors of CT practices within learners’ gameplay behaviors and show that the more they played and used these CT strategies in the game, the more learners showed improvement on CT measures outside the game (BLINDED). Building from this work, we are developing automated detectors of productive persistence so that we can measure the impact of the scaffolds for neurodiverse learners. An efficacy study, conducted by the project’s evaluation partner Knology, will examine the impact of scaffolds on the productive persistence of neurodiverse learners in an implementation study next year. Our team is hoping that in future work these detectors can be used in adaptivity models that customize the delivery of the experience based upon the detected state of the learner.

Authors

• Mia Almeda, Education Development Center, Inc.

• Jodi Asbell-Clarke, TERC

• Tara L. Robillard, TERC

• Erin Bardar

• Teon Edwards, TERC

• Ibrahim Dahlstrom-Hakki, TERC