Trade-Off Hypothesis
The Trade-Off Hypothesis is Peter Skehan's account of how second-language learners allocate limited attentional resources during task performance, and what this allocation predicts about the shape of their output. Developed in A Cognitive Approach to Language Learning (1998) and refined across the next two decades, the hypothesis frames task performance as a competition among three measurable dimensions of language use: complexity, accuracy, and fluency. When one rises, another typically falls.
The CAF Triad
The hypothesis rests on a triad of performance measures that have become the standard outcome variables in task-based SLA research:
| Dimension | What it captures |
|---|---|
| Complexity | Range and sophistication of syntactic structures and lexis used |
| Accuracy | Correctness of forms produced relative to target norms |
| Fluency | Speed, smoothness, and lack of dysfluency in production |
Skehan's claim is that learners' attentional resources are bounded and that pushing more attention onto one dimension tends to leave less for the others. The cognitive architecture cannot do everything at once.
The Central Mechanism
Attention and working memory are treated as a single bottleneck, in line with Skehan's broader Limited Attentional Capacity (LAC) model. Two coping strategies follow:
- Default to formulaic processing. Under load, learners rely on memorised chunks and routinised phrases that can be retrieved holistically. This protects fluency at the cost of restructuring.
- Default to analytic processing. Under different conditions, learners parse and assemble structures element by element. This supports accuracy or complexity but slows production.
Task design influences which strategy dominates. Pre-task planning, task structure, familiarity, and time pressure each tip the balance.
The Debate with Robinson
The Trade-Off Hypothesis is the principal counterweight to Robinson's Cognition Hypothesis, and the two positions structure most cognitive research in task-based language teaching. The disagreement is sharp.
| Question | Skehan (Trade-Off) | Robinson (Cognition Hypothesis) |
|---|---|---|
| What does increased task complexity do? | Forces trade-offs across CAF | Pushes accuracy and complexity up together along resource-directing dimensions |
| Is attention pooled or specialised? | Single limited pool | Multiple resources that can be recruited in parallel |
| What sequencing principle follows? | Manage cognitive load to avoid overload | Sequence simple to complex along resource-directing dimensions |
| What happens under high load? | Fluency or accuracy collapses | Attention is mobilised toward new functional demands |
Both positions have generated substantial empirical work. The picture is genuinely mixed: trade-offs appear in many studies, but so do simultaneous gains under particular complexity manipulations. Differences in target language, proficiency level, task type, and outcome measure all moderate the result.
Pedagogical Implications
If the trade-off picture is correct, classroom decisions involve managing rather than eliminating competing demands.
- Pre-task planning frees attention from content generation so learners can attend to form during performance, typically raising complexity and accuracy at some cost to fluency.
- Task repetition lets formulation become routinised across trials, freeing attention for accuracy or complexity by the second or third performance.
- Online planning (allowing learners time during performance) typically supports accuracy at some cost to fluency.
- Task structure (clear narrative or sequential structure) reduces conceptualisation load and can improve fluency without harming accuracy.
The general advice is to know which dimension a task is targeting and not expect simultaneous gains across all three.
Empirical Picture
Reviews of task-based research find consistent evidence that planning supports complexity and accuracy at some cost to fluency, that task repetition benefits later performances, and that complexity manipulations produce interpretable but mixed CAF profiles. The hypothesis predicts the direction but not always the magnitude of trade-offs, and a substantial number of studies report results that are consistent with neither Skehan nor Robinson in their pure forms.
Criticisms
- Single-pool assumption. Cognitive psychology since the 1990s has tended to favour multiple-resource models, weakening the architectural premise of LAC.
- Operationalisation problems. CAF measures vary across studies, making comparison and synthesis difficult.
- Underestimates individual differences. Aptitude, working memory, and motivation can swamp task effects in any given study.
- Risk of post-hoc fitting. Because trade-offs can occur on any pair of dimensions, almost any pattern of results can be reconciled with the hypothesis after the fact.
References
- Skehan, P. (1998). A Cognitive Approach to Language Learning. Oxford University Press.
- Skehan, P., & Foster, P. (2001). Cognition and tasks. In P. Robinson (Ed.), Cognition and Second Language Instruction (pp. 183–205). Cambridge University Press.
- Skehan, P. (2009). Modelling second language performance: Integrating complexity, accuracy, fluency, and lexis. Applied Linguistics, 30(4), 510–532.
- Skehan, P. (Ed.). (2014). Processing Perspectives on Task Performance. John Benjamins.