Functional anatomy of listening and reading comprehension during development

Abstract

Listening and reading comprehension of paragraph-length material are considered higher-order language skills fundamental to social and academic functioning. Using ecologically relevant language stimuli that were matched for difficulty according to developmental level, we analyze the effects of task, age, neuropsychological skills, and post-task performance on fMRI activation and hemispheric laterality. Areas of supramodal language processing are identified, with the most robust region being left-lateralized activation along the superior temporal sulcus. Functionally, this conjunction has a role in semantic and syntactic processing, leading us to refer to this conjunction as "comprehension cortex." Different from adults, supramodal areas for children include less extensive inferior frontal gyrus but more extensive right cerebellum and right temporal pole. Broader neuroanatomical pathways are recruited for reading, reflecting the more active processing and larger set of cognitive demands needed for reading compared to listening to stories. ROI analyses reveal that reading is a less lateralized language task than listening in inferior frontal and superior temporal areas, which likely reflects the difficulty of the task as children in this study are still developing their reading skills. For listening to stories, temporal activation is stable by age four with no correlations with age, neuropsychological skills or post-task performance. In contrast, frontal activation during listening to stories occurs more often in older children, and frontal activation is positively correlated with better performance on comprehension questions, suggesting that the activation of frontal networks may reflect greater integration and depth of story processing.

Figure 1

3-D Rendering of whole brain activation for Listening Comprehension and Reading Comprehension tasks for children ages 4–12 years old (FWE Corrected, p<.05, >20 voxels in cluster). Right sagittal view (on left), Left sagittal view (on right); and axial slices in neurological convention (left is left hemisphere).

Figure 2

Boxplots of Absolute Mean LI value for Age Group for Wernicke's Area (WA), Inferior Frontal Gyrus (IFG), and Middle Frontal Gyrus (MFG). (A) Listening Comprehension Task. (B) Reading Comprehension Task.

Figure 3

Task-Dependent Activation for subjects who completed both Listening and Reading Comprehension tasks; yellow activation shows areas recruited by the Reading Comprehension task; blue activation shows areas recruited by the Listening Comprehension task; green areas represent activation overlap between the two tasks (FWE Corrected, p<.05, >20 voxels in cluster).

Figure 4

Main effect of task and ROI on mean laterality for 36 subjects who completed both Listening and Reading tasks. Regions of Interest (ROI): Wernickes' Area (WA), Inferior Frontal Gyrus (IFG), and Middle Frontal Gyrus (MFG). Post-hoc, MFG is less lateralized than IFG and WA (*p<.05).