Human-AI Coordination to Induce Flow in Adaptive Learning Systems

Document Type

Article

Publication Date

1-1-2024

Identifier/URL

40960262 (Pure)

Find this in a Library

Catalog Record

Abstract

The coordination between humans and artificial intelligence (AI) systems has the potential to achieve outcomes that neither humans nor AI could achieve alone. AI can process large amounts of data rapidly while humans are able to make use of the AI capabilities to achieve desirable outcomes. In this chapter, we focus on the use of AI in improving user experience in adaptive learning systems. Particularly, we are concerned with whether and how AI can assist in inducing a state of flow in human users. First, we review the literature on flow and adaptive instruction. Then, we describe an experimental study aiming to test an AI agent designed to coordinate with the human user and induce a state of flow. For the experiment, we developed an interactive version of the game Tetris based on the Meta-T software. In this version, we created a balancing feedback loop intended to keep the human player in a continuous state of flow. The human plays the standard Tetris game while an AI algorithm attempts to determine the player’s skill and dynamically alters the game difficulty to match it. The experimental study pitching this adaptive condition against easy and hard conditions shows that the adaptive condition has a positive effect on a composite criterion made of 60% performance and 40% flow. Arguably, this is a realistic criterion for many human performance domains. The adaptive condition, powered by the AI algorithm, does well on this composite criterion because it avoids the pitfalls of the easy and hard conditions: the easy condition hurts performance while the hard condition hurts flow.

Comments

Publisher Copyright: © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024.

DOI

10.1007/978-3-031-53957-2_7

Link to Full Text

Catalog Record

Share

COinS