Humans have a big capacity to learn new skills. Skill learning is a change, usually an improvement, in perceptual, cognitive, or motor performance. This has to be the result of training, and persist for several weeks or months (as to distinguish it from effects related to adaptation or other short-lived effects). With the right training, humans can improve in basically any task.
Researches make a distinction between two types of learning. Early, fast learning happens within minutes, while the participant is becoming familiar with the task and stimulus set. Slow learning arises through practice and requires many hours or days to become effective.
General learning refers to learning effects that at the time of retention testing not only had high savings on the trained task, but also transfer to new tasks and contexts.
Problems with learning
Many times the participants only improve on the trained task, and there is no transfer to other tasks. Specificity was found in perceptual learning, in the motor domain and in cognitive trainings. Also the tasks are often boring and don’t stimulate the best results. Finally, task improvement is not always because of the training. Motivation, mood and wanting to please the researcher influence the task score.
Because of these problems, the researchers have formulated a few questions. They want to identify training regimens that lead to performance improvements and that generalize beyond the training context and persists over time. Also they want to find out which factors contribute to a more general learning outcome.
In some training regimens learning seems to be more general. These trainings are usually more complex than the simple laboratory experiments, and they are closer related to real life events (think of videogame training or athletic training). However, it is important to evaluate how they established the causal link between the training and the improved performance. They determine it through a training study in which non-game players are trained on an action videogame. The skill is assessed before and after the training. Then their results are compared with the performance of a control group that played a non-action game for the same amount of time. However, the causal relationship might be caused by population bias or test-retest effects.
Natural training and brain training
Forms of natural training are playing videogames or doing sports. Brain training needs to be distinguished from natural training, because it is specifically designed to train certain skills. Natural training taps many different aspects of cognitive control. But in brain training, the parts are separated. The training is usually broken down into subdomains and the different subdomains are being trained individually. However, research indicates that this type of learning does lead to faster learning, but it can be detrimental during the retention phase. In the end it often leads to less robust retention and lesser transfer across tasks.
Depending on the cognitive domain, the learning mechanisms may vary. There are however some mechanisms of learning that seem to be shared across domains.
The reverse hierarchy theory
This theory states that information flows in a feed-forward manner through hierarchically organized structures. Information at the lower levels of processing decays as information flows through. However, just to have the information at the higher level is not enough to maintain task performance. Feedback searches need to be done, which go down into the hierarchical structure to find the most informative level of representation. Learning is a top-to-down process, and only tasks that are being handled at the high levels of the hierarchy will show transfer of learning. Tasks that are being handled on lower level processing will show less generalization of learning than those tasks that are being handled by higher level processing.
There are some characteristics that different complex trainings contain that are responsible for the improvement in learning and that make transfer possible. The following characteristics have been thus far identified:
Task difficulty. This is about manipulating the task difficulty in the appropriate manner. Participants will learn new skills and techniques by going through different levels. This way they will have learnt something at the end, which they could not have done at the beginning. Learning rate is at a maximum when the task is challenging, but doable.
Arousal. Trainings with extremely low or extremely high levels of arousal tend to lead to low amounts of learning. Between these extremes there is a level of arousal which leads to a maximum amount of learning.
Variability. Low input variability leads to learning at levels of representation that are specific to the items that are being learned, but they are too rigid to generalize new stimuli. A high variability ensures that the newly learned information are at flexible levels of representation.
- Crossroads lead you through the JoHo web of knowledge, inspiration & association
- Use the crossroads to follow a connected direction
JoHo & Partnernieuws
Ben je JoHo abonnee dan kun je je eigen notities maken, die vervolgens in het notitieveld worden getoond. Deze notities zijn en blijven alleen zichtbaar voor jouzelf. Je kunt dus aantekeningen maken of bijvoorbeeld je eigen antwoorden geven op vragen