Mikey Does

Task Complexity

Task complexity refers to the cognitive and structural demands inherent in a communicative task — the amount of information that must be processed, the number of elements that must be held in working memory, and the degree to which the task pushes beyond a learner’s current automatised resources. It is distinct from task difficulty, which refers to learner-specific factors (motivation, confidence, prior knowledge) that affect performance independently of task design. Task complexity is a property of the task itself; difficulty is a property of the interaction between the task and the learner. The leading theoretical framework is Peter Robinson’s SSARC model (Cognition Hypothesis). Related to cognitive load and task-based language teaching.

Also known as: cognitive complexity (in Robinson’s framework), task demand

In-Depth Explanation

Task complexity matters for SLA because it predicts where learners direct their limited attentional resources. Skill acquisition theory and processing instruction both assume that learners have a constrained processing capacity: when demands exceed that capacity, something suffers — typically either accuracy (more errors) or fluency (more pausing, reformulation, more time). Understanding what makes tasks more or less complex allows teachers to design sequences that stretch learners productively without overwhelming them.

Peter Robinson’s SSARC model (Sequencing and Reviewing Complex Tasks, 2001; 2005) is the dominant framework. Robinson distinguishes task complexity along two main dimensions. The first is resource-directing features: factors that direct learner attention to specific linguistic forms, such as whether the task requires reference to future time (demanding morphology), spatial reasoning (demanding locative expressions), or intentional reasoning (demanding mental state language). More complex resource-directing tasks are claimed to promote acquisition by pushing learners to notice and produce specific forms.

The second dimension is resource-dispersing features: factors that split attention across multiple demands simultaneously — a task with few planning conditions, a task performed under time pressure, or a multi-step narrative. Resource-dispersing complexity tends to impair accuracy and fluency equally, because working memory is spread thin.

Robinson’s central claim (the Cognition Hypothesis) is that incrementally increasing resource-directing complexity — while managing resource-dispersing demands — leads to better L2 development than keeping tasks simple. This contrasts with Peter Skehan’s alternative model, which predicts that increased complexity will consistently trade off against either accuracy or fluency, without a net gain.

The debate between Robinson and Skehan has generated substantial research. Broadly, the evidence supports Robinson’s claim that complex tasks promote syntactic complexity in learner output, though effects on accuracy are less consistent. Pre-task planning time (a resource-dispersing variable) is one of the most replicable findings in the literature: even a few minutes of pre-task planning improves both fluency and complexity of learner production, with smaller effects on accuracy.

History

Task complexity as a formal construct emerged from the task-based language teaching research programme of the 1980s and 1990s. Early researchers (Long, Prabhu, Nunan) focused on how to design tasks effectively, but the question of what makes one task harder than another was not systematically theorised until Robinson’s cognition hypothesis (2001).

Robinson’s framework was directly influenced by Anderson’s cognitive theory of skill learning and by working memory models in cognitive psychology — particularly Baddeley’s multicomponent model. The attempt to ground SLA task research in cognitive architecture was a significant step toward making task complexity measurable rather than impressionistic.

Peter Skehan‘s Trade-Off Hypothesis, developed around the same period, proposed a rival model in which a learner’s attention is finite: any increase in task complexity forces trade-offs among accuracy, fluency, and lexicogrammatical complexity. The Robinson/Skehan debate became one of the defining theoretical arguments in TBLT research through the 2000s.

Common Misconceptions

  • “Task complexity = task difficulty” — these are explicitly distinct. A highly complex task (many elements, future reference, novel reasoning demands) may feel easy to an expert learner because they have automatised the relevant resources. Difficulty is a subjective, learner-relative experience; complexity is a design-level property.
  • “More complex tasks are always better for acquisition” — Robinson’s claim is that incrementally increasing resource-directing complexity promotes acquisition. Tasks too far beyond current resources produce communication breakdown, not acquisition. The key is appropriate staging.
  • “Task complexity only matters in TBLT classrooms” — any communicative task has a complexity profile. Conversation, reading, listening, and writing tasks all have features that make them more or less demanding, whether or not they occur in a formal TBLT context.

Criticisms

Robinson’s SSARC model has been criticised for being difficult to operationalise: measuring cognitive load on a task-by-task basis requires indirect methods, and different tasks that appear to have similar complexity profiles may produce very different learning outcomes. The prediction that increased complexity promotes acquisition has not always been replicated cleanly.

Skehan and colleagues have argued that Robinson’s model underestimates trade-off effects — that real-time communication pressure consistently impairs accuracy regardless of task type, and that complex tasks do not reliably produce the pushed-output effects Robinson predicts.

There is also a broader critique that task complexity research remains too narrowly cognitive, treating language learning as an information-processing problem while underweighting social, affective, and contextual factors.

Social Media Sentiment

Task complexity rarely appears by name in general language learning communities. It is primarily a research and teacher training term. In TESOL and applied linguistics forums and graduate school discussions, it figures in debates about task sequencing and TBLT implementation. Where it does appear in learner communities, it’s often reframed as a practical question: “what makes a comprehensible input source too hard?” or “should I make my study tasks easier or harder?” — both of which are folk versions of the task complexity debate.

Last updated: 2026-04

Practical Application

For learners, the practical implication is to manage complexity deliberately. A speaking task that requires simultaneous vocabulary search, grammar monitoring, and real-time memory retrieval puts maximum load on all systems at once. Adding planning time — even two minutes — dramatically reduces the resource-dispersing burden and lets you focus on producing more accurate or more complex sentences.

When choosing input and tasks, consider which features add resource-directing complexity (topics requiring future or hypothetical reasoning, unfamiliar vocabulary, multi-step events) versus resource-dispersing complexity (time pressure, no script, many simultaneous demands). You can control the latter through task design — planning time, note-taking, a familiar topic — so you can stretch on the former without overwhelming your working memory.

This is a principled way to think about i+1: the “plus one” is a resource-directing complexity increase, not simply harder vocabulary.

Related Terms

See Also

Sources

mikeydoes.com