Augmented Multiphase Learning
In this project, we investigate the effectiveness of augmented feedback modalities and modulation of task conditions to support multiphase learning of real-life, complex tasks. Effective methods for the motor learning of novice and skilled humans are studied for various classes of tasks. For this project, different robots are used to render manifold task characteristics and conditions. When appropriate feedback modalities and designs are found for the successful learning for distinct skill levels and task categories, trainees can be supported from novice to an expert level along the multiphase learning process of a complex movement.
Involved people
Dr. Peter Wolf
When learning a new skill in sports or rehabilitation, receiving extra source of information becomes crucial. In general, availability of information accelerates motor learning and improves the quality performance. In addition to receiving the information as a form of an augmented feedback, information can also be implicitly received by being exposed to different task conditions. In sports, athletes can receive the information either as a feedback from their coaches or by performing the task under various conditions.
In order to support successful learning, external information should be competent for the skill level of learner for a certain task. As a general approach, the movement structure should be introduced to the novice to understand the task in the beginning of the learning phase, and then transition should be realized for fine tuning of the movement.
Until now, rather simple and laboratory tasks have been studied to assess the effectiveness of external information methods. Additionally, results of the application of feedback modalities have usually been generalized to learning of many other tasks without considering the relationship between spatial and temporal aspects of the movement. However, tasks that people interact in their daily lives are more complex and have different spatio-temporal characteristics; thus, a caution should be taken when inferring results. For example, reaching and grabbing an object is classified as a discrete task (representation of spatial and temporal characteristics are distinct), while stirring an egg in a bowl is a continuous task. Therefore, effectiveness of various feedback modalities and designs should be determined for different complex tasks and movement characteristics.
Furthermore, external information methods have mostly been used to study the learning of novice subjects. Learning of skilled subjects has rarely been considered in complex tasks. However, a complete learning process begins when a novice subject is in cognitive stage and then progresses to more skilled (associative) level and finally reaches to autonomous stage. When appropriate external information methods can be supplied at a specific learning phase and they can be changed to more suitable methods in later phases, learning process can be supported in all phases of learning.
Due the fact that impact of information depends on the task characteristics and conditions as well as the skill level, there is a necessity to extend the knowledge on the effectiveness of information designs for various skill levels for different task categories. Therefore, this project aims to find appropriate methods to provide information that can help naïve or skilled people to learn discrete and/or continuous complex tasks.