In this article, we develop as framework for understanding how cerebral specialization of function contributes to the flexibility of human information processing. We propose that the left and right hemispheres together form a system of two mutually inaccessible and finite pools of resources. Further, we propose that these two types of resources cannot be made available in different amounts at any given time. This framework is essentially a special case of a multiple-resources model of limited-capacity information processing. It accounts for a broad range of data from experiments involving perceptual and cognitive information processing, control of motor performance, and changes in electrical activity of the brain. It also provides insights into why the cerebral specialization literature has been plagued with problems that have made theorizing so difficult. In addition, the theory provides insights into mechanisms that might be responsible for patterns of task interference that are not easily handled by an information-processing model in which processes compete for supplies from a single pool of undifferentiated resources. Thus, the framework we are proposing has important theoretical and methodological implications for researchers in both divided attention and cerebral specialization.

Two experiments tested the limiting case of a multiple resources approach to resource allocation in information processing. In this framework, the left and right hemispheres are assumed to have separate, limited-capacity pools of undifferentiated resources that are not mutually accessible, so that tasks can overlap in their demand for these resources either completely, partially, or not at all. We tested all three degrees of overlap in demand for left hemisphere supplies, using dual-task methodology in which subjects were induced to pay different amounts of attention to each task. Experiment 1 compared complete and partial overlap by combining a verbal memory load with a task in which subjects named nonsense syllables briefly presented to either the left or right visual field (LVF and RVF, respectively). Experiment 2 compared complete versus no overlap by using the same verbal memory load combined with a laterally presented same-different judgment task that did not require a spoken response. Decrements from single-task performance were always more severe when the visual field task stimulus was presented to the RVF. Further, subjects in Experiment 1 were able to trade performance between tasks on both LVF and RVF trials because there was always at least some overlap in left hemisphere demand. In Experiment 2, performance trade-offs were observed on RVF (complete overlap) trials, but not on LVF trials, where no overlap in demand existed. These results contradict a single-capacity model, but they support the idea that the hemispheres' resource supplies are independent and have implications for both cerebral specialization and divided attention issues.

No abstract

No abstract

whether this task combination would reveal a specialized resource responsible for coordinating motor outputs when at least one of those outputs involves speech. A second, related purpose was to demonstrate the importance of manipulating task priority when attempting to determine the resource requirements of a task.The rationale underlying our approach and its methodology derives from the multiple resources model of Navon and Gopher (1979), which builds on earlier constructs about divided attention (Norman & Bobrow, 1976). The general idea is that there are different types of processing resources. Two tasks can compete for the same type of resource without showing any performance decrement, relative to when they are performed in isolation, until the aggregate demand exceeds the available supply. After this point, there should be interference on one or both tasks. However, performance decrements can ensue from causes unrelated to competition for the same scarce resource (e.g., concurrence costs in Navon & Gopher's, 1979, terminology; scheduling conflicts in Norman & Shallice 's, 1985). Consequently, just demonstrating that two tasks interfere with each other is insufficient evidence from which to conclude that the tasks demand a common resource. To draw this conclusion, performance trade-offs have to be observed when task priority is manipulated. In particular, when two tasks demand the same type of resource, shifting attention from one to the other should cause an increase in performance on the emphasized task at the expense of performance on the other task. When two tasks have nonoverlapping demands, performance should not trade off. Therefore, failing to ma-This document is copyrighted by the American Psychological Association or one of its allied publishers.This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.