Adaptive Learning and Aviation

Author: Mary Watt

For anyone involved in aviation training, the term adaptive learning is quickly becoming a familiar buzzword. In industry publications, trade shows, panel discussions, and online forums it turns up with increasing frequency. But what does it really mean?

Adaptive learning is training that is tailored to an individual learner’s needs, automatically delivering the training needed for a learner with a particular profile. It could be adaptive in terms of content, responding to the safety record of the individual, the observations of a training manager, or the outcome of testing. It could be adaptive in terms of delivery, changing to fit the preferences of the learner or the best delivery mode for a particular training need. It could also be adaptive in terms of the speed and number of repetitions of a particular topic, responding to the learner’s errors in a testing environment. How it adapts depends on an algorithm, a set of rules developed to respond to a particular learner’s profile and actions. These rules are based on data and experience, accumulated from many sources.

Associated with adaptive learning are a number of terms that are often used interchangeably.

Individualization:

Refers to the pacing of learning. Some learners require more time to master a topic, or more repetitions in order to reach mastery; others are able to skip sections or move faster.

Differentiation:

Refers to changes in presentation that reflect the preferences of the learner. For instance, some may prefer watching videos to reading text. While the concept of ‘learning styles’ has long been debunked, making allowances for learner preferences can help with motivation.

Personalization:

Refers to a comprehensive adjustment of all facets of a learning experience to individual needs. It encompasses both individualization and differentiation.

The end result of all this adaptation is a training program designed to be more time-efficient, more cost-efficient, and targeted to the specific needs of a particular learner. However, there are some caveats. Although the potential is there, adaptive learning is only as good as the algorithms that drive it. Good algorithms require large quantities of data and expert input into the rules created. There are privacy issues around collecting data, and the ‘rules’ of an algorithm need to be rigorously evaluated for cultural and other biases. The quality of the training will always be the most important factor in successful training interventions. Nevertheless, the potential for improving training and training outcomes is enormous and well worth pursuing.

_____________

Mary Watt is an instructional designer and educational technology strategist with more than 20 years experience in developing technology-enhanced learning projects. She works with TrainingPort to explore new and emerging technologies in online learning.