Multiple resource theory (MRT), developed by Christopher Wickens (1984, 2002), challenges the idea that attention is a single, undifferentiated resource. Instead, Wickens proposed that the human information-processing system contains multiple pools of resources, organized along several dimensions. Two tasks interfere more when they draw on the same resource pools and less when they use different pools — explaining why some task combinations are much harder than others.
Resource Dimensions
Wickens identified several key dimensions along which resources are structured. Processing stage: perceptual/cognitive processing vs. response processing use separate resources. Perceptual modality: visual processing and auditory processing draw on separate resources. Processing code: spatial processing and verbal/linguistic processing use different resources. Visual channel: focal (foveal) and ambient (peripheral) visual processing are partially independent.
Modality: Visual ←→ Auditory
Code: Spatial ←→ Verbal
Visual Channel: Focal ←→ Ambient
Interference ∝ resource overlap between tasks
Predictions for Dual-Task Performance
MRT predicts that two tasks will interfere less when they are separated along resource dimensions. A visual-manual task and an auditory-vocal task should produce less dual-task interference than two visual-manual tasks. A spatial task and a verbal task should interfere less than two spatial tasks. These predictions have been extensively tested and generally supported, making MRT one of the most practically useful theories of divided attention.
MRT has been enormously influential in the design of complex systems such as aircraft cockpits, vehicle dashboards, and military interfaces. The theory provides concrete design guidelines: distribute information across modalities (visual and auditory displays), use compatible stimulus-response mappings, and avoid overloading any single resource pool. For example, MRT predicts that adding auditory alerts to a visually demanding flying task will produce less interference than adding visual alerts — a prediction confirmed by aviation research and now standard in cockpit design.
Empirical Support and Limitations
The modality dimension has received the strongest support: visual-auditory task combinations consistently produce less interference than within-modality combinations. The code dimension (spatial vs. verbal) is also well supported. However, the specific number and nature of resource pools remains debated, and the theory does not specify the mechanism by which resources are allocated or the capacity of each pool. Some critics argue that apparent resource effects can be explained by structural interference (tasks competing for the same processing hardware) without invoking the metaphor of limited resources.
Comparison with Other Theories
MRT stands in contrast to Kahneman's (1973) single-resource theory, which proposed one undifferentiated capacity pool, and to bottleneck theories, which propose that interference arises from serial processing at specific stages rather than from resource sharing. In practice, all three types of limitation may operate: structural bottlenecks at response selection, modality-specific resource limits during encoding, and a general capacity limit related to executive control. MRT captures the resource component of this complex picture.