2003, 3 (4), 255-274 | TOR D. WAGER and EDWARD E. SMITH
The meta-analysis by Wager and Smith (2003) examines the neural basis of working memory (WM) across 60 neuroimaging studies using PET and fMRI. The study considers three types of storage material (spatial, verbal, and object) and three types of executive function (continuous updating, memory for temporal order, and manipulation of information). Key findings include:
1. **Material Type Dissociation**: In the posterior cortex, there is a clear dissociation between spatial and object storage, with the superior parietal cortex preferentially activated for spatial tasks and the inferior temporal cortex for object tasks. However, this dissociation is not observed in the frontal cortex.
2. **Verbal WM and Frontal Lateralization**: Some support is found for left frontal lateralization in verbal WM tasks, particularly for tasks with low executive demand. This suggests that Brodmann's areas (BAs) 44, 45, and 46 in the left frontal cortex are involved in verbal WM, though the activation is spread across multiple BAs.
3. **Executive Function and Frontal Cortex**: Executive demand increases right lateralization in the frontal cortex for spatial WM tasks. Tasks requiring executive processing generally produce more dorsal frontal activations, but not all executive processes show this pattern. Specific areas like BA 7 in the superior parietal cortex are involved in all types of executive function.
4. **Cluster Analysis**: The cluster analysis provides a data-driven classification of brain regions, showing clear specialization for spatial and object storage in the posterior cortex and for executive functions in the frontal and parietal regions.
5. **Interactions Between Material and Executive Demand**: The study also examines interactions between storage material and executive demand, finding that certain regions show increased activation with executive demand, particularly in the right frontal and parietal cortices.
6. **Selective Attention**: Selective attention, which involves storing one attribute of a multidimensional stimulus while ignoring others, produced more frequent activation in BA 32 in the medial prefrontal cortex.
Overall, the meta-analysis provides a comprehensive overview of the neural mechanisms underlying WM, highlighting the dorsal-ventral dissociation in spatial and object storage and the lateralization of frontal regions in executive functions.The meta-analysis by Wager and Smith (2003) examines the neural basis of working memory (WM) across 60 neuroimaging studies using PET and fMRI. The study considers three types of storage material (spatial, verbal, and object) and three types of executive function (continuous updating, memory for temporal order, and manipulation of information). Key findings include:
1. **Material Type Dissociation**: In the posterior cortex, there is a clear dissociation between spatial and object storage, with the superior parietal cortex preferentially activated for spatial tasks and the inferior temporal cortex for object tasks. However, this dissociation is not observed in the frontal cortex.
2. **Verbal WM and Frontal Lateralization**: Some support is found for left frontal lateralization in verbal WM tasks, particularly for tasks with low executive demand. This suggests that Brodmann's areas (BAs) 44, 45, and 46 in the left frontal cortex are involved in verbal WM, though the activation is spread across multiple BAs.
3. **Executive Function and Frontal Cortex**: Executive demand increases right lateralization in the frontal cortex for spatial WM tasks. Tasks requiring executive processing generally produce more dorsal frontal activations, but not all executive processes show this pattern. Specific areas like BA 7 in the superior parietal cortex are involved in all types of executive function.
4. **Cluster Analysis**: The cluster analysis provides a data-driven classification of brain regions, showing clear specialization for spatial and object storage in the posterior cortex and for executive functions in the frontal and parietal regions.
5. **Interactions Between Material and Executive Demand**: The study also examines interactions between storage material and executive demand, finding that certain regions show increased activation with executive demand, particularly in the right frontal and parietal cortices.
6. **Selective Attention**: Selective attention, which involves storing one attribute of a multidimensional stimulus while ignoring others, produced more frequent activation in BA 32 in the medial prefrontal cortex.
Overall, the meta-analysis provides a comprehensive overview of the neural mechanisms underlying WM, highlighting the dorsal-ventral dissociation in spatial and object storage and the lateralization of frontal regions in executive functions.