Abstract:

Problem Statement:
Adaptive Learning (AL) is a cutting-edge approach to learning in contemporary education that provides personalized learning experiences. However, the method's effectiveness is limited, primarily because an AL-based course cannot adapt cognitive load to students' cognitive functions in real time during the lesson. The novel Cognitive-Adaptive Learning (CAL) method applies knowledge from cognitive science and an artificial intelligence system that comprehensively analyzes biomarker dynamics of students’ cognitive load, comparing them with their performance during instruction. It enables advanced teaching to continuously monitor students' cognitive functions during the study process, tailoring the study course to their content comprehension and memorization during the lesson. A growing body of research demonstrates an association between simple reflex dynamics and cognitive functions. However, it remains unclear whether subjects' simple reflexes reflect the dynamics of their cognitive functions (such as perception, attention, and memory) during real-time learning to such an extent that they would be relevant markers of learning performance.

Objective:
The goal of the current research is to test the relationship between instructions provided by the CAL system and subjects' learning outcomes. Does the CAL teaching method have a greater impact on memory and learning than ordinary teaching methods?

Significance:
The outcome allows us to design a Cognitive-adaptive learning (CAL) system based on Intelligent Systems that can be used as a template to produce specific CAL courses in different disciplines. This learning tool can enhance the effectiveness of e-learning programs. These data pave the way for further research on Cognitive-adaptive learning (CAL) relying on continuously screening cognitive functions in students during the study process.

Methodology:
The study examines students’ performance (N=15) using a Randomized Control Trials design that includes random assignment of participants to different conditions. While two control groups study the basic or advanced courses, an intelligent system presents the experimental group with a page, allowing it to choose from the basic or advanced course options page by page based on the individual student's cognitive load assessment. Thus, the algorithm enables the adaptation of the curriculum (choosing between a basic and an advanced course) to students' cognitive load during learning. This transforms ordinary learning into the Cognitive-adaptive learning that modifies cognitive load based on students' comprehension and retention of the material.

Results:
The outcome shows increased students’ performance in CAL design; the difference in performance between the three groups (two control and one experimental group) is significant, with a p-value below 0.05.

Key Findings:
The most significant finding of the study is the relationship between instructions provided by the CAL system and subjects' learning outcomes. The CAL teaching method shows a greater impact on memory and learning than ordinary teaching methods.

Keywords:

Adaptive Learning, Intelligent Systems, e-learning.