An exploration of how television, algorithms, and learning machines quietly reshape what it means to be awake
Preserving productive struggle in adaptive learning requires designing AI systems that support the learner’s thinking process without taking it over. The goal is to keep students operating within their optimal zone of growth, where content is challenging enough to stretch their capabilities, but adequately supported so it does not cause helplessness. Below following a scenario that is comparing how AI systems can work in a passivating respectively activating way.
Scenario: Solving a 2D Vector Physics Problem
A high school student, Maya, is working through a physics module in a Learning Management System (LMS). She encounters a multi-step problem calculating post-collision velocities. Maya makes a fundamental conceptual mistake: she forgets that velocity is a vector quantity and treats directional values ($+v$ and $-v$) as positive scalars in her equation $m_1 v_{1i} + m_2 v_{2i} = (m_1 + m_2) v_f$.
| Interaction Stage | Passive AI Scaffolding (Friction-Eliminating) | Active, Struggle-Preserving AI (Friction-Balancing) |
|
Error Trigger |
System detects an incorrect numerical entry after 45 seconds of input. |
System detects vector sign omission by comparing intermediate variable calculations and input timing. |
|
AI System Response |
Automatically lowers difficulty level, presents a simplified 1D single-direction problem, and displays the corrected equation with numbers pre-populated: $10(5) + 5(-2) = 15 v_f$. |
Keeps problem difficulty constant. Prompts with a Socratic hint: “Notice the arrows on your diagram indicating motion toward the left versus right. How does direction impact $v_{2i}$ in vector equations?” |
|
Student Action |
Clicks “Continue,” copies the pre-populated values into a calculator, and submits the answer without re-evaluating vector logic. |
Pauses, re-reads the problem diagram, realizes that leftward velocity requires a negative sign, and manually modifies the equation. |
|
Cognitive State |
Passive Compliance: Low cognitive load, high surface engagement, no metacognitive effort or reflection on why the previous attempt failed. |
Productive Struggle: Moderate cognitive load, high reflection; student actively restructures her mental model of vector quantities. |
|
Feedback Loop |
System praises rapid completion: “Great job! You mastered this problem.” Advances student to next unit. |
System highlights strategy: “Spotting the sign error on directional velocity was key. How will you check vector signs in 2D components next?” |
|
Retention Impact |
When faced with a similar 2D problem on an unassisted exam 3 weeks later, Maya repeats the vector sign error. |
Maya independently checks direction signs before executing calculations on future assessments. |
From this background we are now creating a intecative book of best practice for adaptive AI developers as well as course developers.
Written by
LarsGoran Bostrom
We help organisations and humans to interactivating themselves by turning screens into active development and safeguard your personal integrity and organisation’s information security. Find out more on eLearningworld Europe AB and B-InteraQtive Publishing