Use ANN-based adaptive feedback to provide scalable, process-oriented real-time feedback at scale in higher education¶

Strategy · All strategies
Evidence · 5 claims (4 for, 1 mixed) · 4 studies (2 theoretical, 1 causal, 1 review), q2–q3 · 1 of 4 report an effect size · 4 claims rest on one study

Description¶

The article recommends automatic adaptive feedback in simulations as a way to give large student cohorts elaborate feedback that would otherwise require instructor effort. It states that such feedback offers "scalable, elaborate, process-oriented feedback in real-time to high numbers of students in higher education". The strategy is implementable in simulation-based learning environments where learner responses can feed a neural network that generates adaptive feedback.

Design Implications¶

Context¶

Requirements¶

• Simulation-based learning environment; artificial neural network infrastructure for automatic feedback generation

Constraints¶

• In the reported study the benefit appeared for justification quality, not diagnostic accuracy

Target Learners¶

• Pre-service teachers in higher education

Target Learning Goals¶

• Diagnostic reasoning and justification quality

Claims¶

• Automatic adaptive feedback based on artificial neural networks facilitates pre-service teachers' quality of justifications in written assignments in simulation-based learning [+W]
• Adaptive feedback does not improve pre-service teachers' diagnostic accuracy in simulation-based learning [~W]
• Static expert-solution feedback had detrimental effects on the learning process when pre-service teachers worked in dyads [+W]
• Real-time adaptive scaffolding based on note-coded proof indicators elicited more elaborate diagnostic reasoning [+W]
• Automated real-time feedback in an adaptive diagnostic ITS produced greater progression on a SOLO-based mathematics assessment than the same system without real-time feedback [+W]

Related Strategies¶

• Use synthetic LLM-generated error datasets for AI tutors, teachable agents, and teacher professional development
• Use simulated student sections as a scalable supplement to teacher training and lesson preparation
• Adaptive scenario generation: introduce training scenarios targeting weaknesses identified from real-time performance metrics

Examples¶

-

Key Sources¶

• Sailer, Michael; Bauer, Elisabeth; Hofmann, Riikka; Kiesewetter, Jan; Glas, Julia. (2022). Adaptive Feedback from Artificial Neural Networks Facilitates Pre-Service Teachers' Diagnostic Reasoning in Simulation-based Learning. Learning and Instruction. https://doi.org/10.1016/j.learninstruc.2022.101620