The Effectiveness of Aural Instructions with Visualisations in E-Learning Environments

Abstract

Based on Mayer’s (2001) model for more effective learning by exploiting the brain’s dual sensory channels for information processing, this research investigates the effectiveness of using aural instructions together with visualisation in teaching the difficult concepts of data structures to novice computer science students. A small number of previous studies have examined the use of audio and visualisation in teaching and learning environments but none has explored the integration of both technologies in teaching data structures programming to reduce the cognitive load on learners’ working memory.
A prototype learning tool, known as the Data Structure Learning (DSL) tool, was developed and used first in a short mini study that showed that, used together with visualisations of algorithms, aural instructions produced faster student response times than did textual instructions. This result suggested that the additional use of the auditory sensory channel did indeed reduce the cognitive load.
The tool was then used in a second, longitudinal, study over two academic terms in which students studying the Data Structures module were offered the opportunity to use the DSL approach with either aural or textual instructions. Their use of the approach was recorded by the DSL system and feedback was invited at the end of every visualisation task.
The collected data showed that the tool was used extensively by the students. A comparison of the students’ DSL use with their end-of-year assessment marks revealed that academically weaker students had tended to use the tool most. This suggests that less able students are keen to use any useful and available instrument to aid their understanding, especially of difficult concepts.
Both the quantitative data provided by the automatic recording of DSL use and an end-of-study questionnaire showed appreciation by students of the help the tool had provided and enthusiasm for its future use and development. These findings were supported by qualitative data provided by student written feedback at the end of each task, by interviews at the end of the experiment and by interest from the lecturer in integrating use of the tool with the teaching of the module. A variety of suggestions are made for further work and development of the DSL tool. Further research using a control group and/or pre and post tests would be particularly useful.