Characteristics of language-learning tasks

The table below draws on the research findings to date to suggest how to devise tasks that promote fluency, accuracy, or complexity in your students’ language.

Characteristics of task	Promotes fluency	Promotes accuracy	Promotes complexity
Input
Contextual support (e.g. pictures, diagrams)	There is contextual support	The task has no contextual support	The task has no contextual support
Number of elements	There are only a few elements in the task input		The task has many elements
Topic	The topic is familiar or the topic generates debate
Conditions
Shared versus split information			The information is shared
Task demands	The task poses a single demand		The task poses a single demand
Opportunity for advance planning	The task allows advance planning	If the task allows advance planning this may promote accuracy
Monitoring their own performance as they speak or write	Tasks that involve monitoring while they are done impair fluency	Tasks that involve monitoring while they are done promote accuracy
Task outcomes
Closed versus open outcome	The task outcome is closed	The task outcome is open	The task is open and has divergent goals
Inherent structure of the outcome	The task outcome has a clear inherent structure	The task outcome has a clear inherent structure and there is opportunity for planning
Discourse mode			Narrative tasks promote more complexity than argument tasks. Discussion tasks produce the least complexity of the three.

Return to top

Example: Promoting fluency

Read down the columns to see the task features that are required to promote particular aspects of language. For example, if you wish to help your students develop fluency in using the language they already know, you will design a task that has the features listed in column 2. This means you will provide contextual support, there will not be a large number of elements to the input, and the topic will be familiar or one that leads to debate. The task will have a single demand and the students will be able to plan in advance. The language-learning outcome will be closed, with a clear structure.

Terms used to describe tasks

Contextual support

Contextual support refers to material, such as pictures and diagrams, that supports the language of a learning task by providing some of the meaning. These reduce the learner’s reliance on the language itself to communicate the meaning.

Shared and split information

If the information is shared, all the learners have access to the same information. If it is split, the learners have different items of information and are not allowed to share it. Split information tasks force all learners to participate, but shared information promotes more complexity in the learners’ language, possibly because they each have to deal with a larger amount and diversity of information.

Closed and open outcomes

The outcomes of a task can be closed or open. Closed outcomes to tasks are when there are only certain correct outcomes or solutions. Problem-solving tasks often have only one or two solutions and are tasks with closed outcomes. Role-play and interviews, on the other hand, are almost completely open.

Inherent structure of the outcome

The degree of a language-learning outcome’s openness varies according to its inherent structure; for example, explaining to someone how to get to the ATM has an inherent structure in the form of a (mental) map of the route. Although an interview has an open outcome in the form of the answers received, the outcome may have an inherent structure if the information gained is to be reported in a number of fixed categories. If the students report back in any way they choose, then there is no inherent structure. They will probably report back in relation to the questions they asked.

Task demands

Some tasks ask learners to do several things at the same time. A single-task demand has been found to promote both fluency and complexity. It is probably better, then, to break complex tasks down into several tasks. These single-demand tasks can then be completed one after the other. However, it will be important to make sure learners integrate what they learn from the separate tasks and do not treat them as isolated events.