doi:10.1016/j.neuropsychologia.2007.12.021 
Copyright © 2007 Elsevier Ltd All rights reserved.
Sex differences in neural processing of language among children
Douglas D. Burmana, b, 
, 
, Tali Bitanc and James R. Bootha, b
aDepartment of Communication Sciences and Disorders, Northwestern University, Evanston, IL, USA
bDepartment of Radiology, Evanston Northwestern Healthcare, Evanston, IL, USA
cDepartment of Communication Disorders, Haifa University, Mt. Carmel, Haifa, Israel
Received 19 April 2007;
revised 17 October 2007;
accepted 14 December 2007.
Available online 4 January 2008.
References and further reading may be available for this article. To view references and further reading you must
purchase this article.
Abstract
Why females generally perform better on language tasks than males is unknown. Sex differences were here identified in children (ages 9–15) across two linguistic tasks for words presented in two modalities. Bilateral activation in the inferior frontal and superior temporal gyri and activation in the left fusiform gyrus of girls was greater than in boys. Activation in the left inferior frontal and fusiform regions of girls was also correlated with linguistic accuracy irregardless of stimulus modality, whereas correlation with performance accuracy in boys depended on the modality of word presentation (either in visual or auditory association cortex). This pattern suggests that girls rely on a supramodal language network, whereas boys process visual and auditory words differently. Activation in the left fusiform region was additionally correlated with performance on standardized language tests in which girls performed better, additional evidence of its role in early sex differences for language.
Keywords: fMRI; Gender; Development; Reading; Skill
Fig. 1. Activation and sex differences during language tasks. Activation across both language judgment tasks and sensory modalities was elicited across all age groups irrespective of sex (yellow in brain images), but girls (pink) showed significantly greater activation than boys (blue) in bilateral regions of IFG and STG as well as left FG. Task, modality, age, and sex were entered into an ANCOVA model with accuracy as a covariate. Graph data were derived from ROI analysis of five regions showing significant sex effects (p < 0.005 with a Bonferroni correction); the BOLD signal represents the estimated partial means derived from the mean activity of each region-of-interest after removing variance attributable to age and accuracy.
Fig. 2. Sex differences in laterality reflect a threshold effect. Using a more stringent statistical threshold (p = 1.0 × 10−7 with a FDR correction), the left hemisphere showed similar patterns of activation by boys (blue) and girls (pink), including substantial overlap (cyan), but only girls showed bilateral activation in IFG and STG. IFG = inferior frontal gyrus; STG = superior temporal gyrus; FG = fusiform gyrus; ITG = inferior temporal gyrus; Cun = cuneus; Ling = lingual gyrus.
Fig. 3. Activation during nonlinguistic sensory tasks and sex effects. (A) Activation by nonlinguistic visual stimuli showed greater activation by girls (pink) in fusiform gyrus (FG); boys did not show greater activation anywhere. (B) Activation by nonlinguistic auditory stimuli showed greater activation by girls in superior temporal gyrus. STG = superior temporal gyrus; FG = fusiform gyrus.
 |
Fig. 4. Sex differences in correlation of activation with performance accuracy grouped by judgment task and by modality. (A) After accounting for modality effects, activation in MTG/FG and IFG was correlated with performance accuracy on both rhyming and spelling tasks among girls (pink) but not boys. (B) After accounting for task effects (rhyming vs. spelling), activation by auditory word stimuli among boys (blue) was correlated with performance accuracy in STG and IFG, whereas activation by visual word stimuli was correlated with performance accuracy in SPL/PreCun. Among girls (pink), correlation of accuracy with activation for auditory word stimuli was generally distinct from regions correlated with accuracy among boys, although overlap (cyan) was evident in IFG and (slightly) in posterior MTG. Activation by visual word stimuli was not correlated with performance accuracy irrespective of language judgment among girls. IFG = inferior frontal gyrus; MTG/FG = middle temporal gyrus extending into fusiform gyrus; SPL/PreCun = superior parietal lobule extending into precuneus.
Table 1.
Standardized scores of subjects by age and sex
Mean standardized scores and standard deviation (in parentheses) are provided for each sex and age group. Bold headings identify tests that had sex differences across age groups using an ANOVA with sex and age as factors; bold scores in a cell identify sex differences significant within an age group. PPVT = Peabody Picture Vocabulary Test-III; WRAT = Wide Ranging Achievement Test-III (spelling subtest); WJ-III = Woodcock-Johnson III Tests of Cognitive Abilities (wordID = word identification, read = reading fluency); CTOPP = Comprehensive Test of Phonological Processing (PA = phonological awareness; PM = phonological memory; RN = rapid naming); TOWRE = Test of Word Reading Efficiency (PDE = pseudoword decoding efficiency).
* ANOVA, main effect of sex,
p < 0.10.
** ANOVA, main effect of sex,
p < 0.05.
† Student
t-test (girls vs. boys of same age),
p < 0.05.
Table 2.
Accuracy performance on perceptual and lexical tasks
For each task and modality, accuracy is listed by age and sex. Numbers represent mean percent accuracy.
Table 3.
Main effects of language tasks after partialing out effects of accuracy
The statistical threshold was p = 0.05 with a family-wise error correction for multiple comparisons. Brain activation maxima are listed in MNI coordinates, with Brodmann areas estimated from the PickAtlas and aal toolboxes for SPM2. IFG = inferior frontal gyrus; MFG = middle frontal gyrus; STG = superior temporal gyrus; MTG = middle temporal gyrus; IT = inferior temporal gyrus; fusiform = fusiform gyrus; MeFG = medial frontal gyrus; AC = anterior cingulate gyrus; IPL = inferior parietal lobule.
Table 4.
Location of sex effects (female > male) during language tasks after partialing out effects of age and accuracy
Sex differences in the magnitude of activation were identified in an ANCOVA model that partialed out effects of age and accuracy, using the main effects map of activation by language tasks as a mask. The threshold was p < 0.05, using a FDR correction for multiple comparisons and an extent threshold of 25 voxels. Regional abbreviations as in Table 3.
Table 5.
Activation maps with a stringent threshold show laterality differences between sexes
Sex differences in laterality were seen after increasing the threshold for activation (p = 1.0 × 10−7 using a FWE correction and extent threshold of 25 voxels for the ANCOVA model described for Table 4). The sites of activation for girls and for boys are listed. Regional abbreviations as in Table 3.
Table 6.
Location of sex effects during perceptual tasks
Sex differences in activation magnitude during nonlinguistic perceptual tasks identified in an ANCOVA model with age and accuracy as covariates; the area of activation during language tasks was used as a mask. The threshold for these sex effects was p < 0.05, using a FDR correction for multiple comparisons and an extent threshold of 25 voxels. Regional abbreviations as in Table 3.
Table 7.
Sex correlations of activation with accuracy on rhyming and spelling tasks after partialing out effects of sensory modality and age
Brain activation was correlated with performance accuracy separately for each sex within an ANCOVA model that partialed out effects of sensory modality and age; the conjunction map of activation by both rhyming tasks at p < 0.05 (uncorrected) was used as a mask for accuracy on rhyming judgments, whereas the conjunction map of activation by both spelling tasks at p < 0.05 (uncorrected) was used as a mask for accuracy on spelling judgments. The activation threshold was p = 0.05, using a FDR correction for multiple comparisons and an extent threshold of 15 voxels. Regional abbreviations as in Table 3.
Table 8.
Sex correlations of activation with accuracy on auditory and visual tasks after partialing out effects of language judgment task and age
Brain activation was correlated with performance accuracy separately for each sex and sensory modality within an ANCOVA model that partialed out effects of language judgment task and age. The conjunction map of activation by both auditory language tasks at p < 0.05 (uncorrected) was used as a mask for auditory accuracy judgments, whereas activation by both visual language tasks at p < 0.05 (uncorrected) was used as a mask for visual accuracy judgments. The activation threshold was p = 0.05, using a FDR correction for multiple comparisons and an extent threshold of 15 voxels. Regional abbreviations as in Table 3.
Table 9.
Brain areas where activation on visual tasks was correlated with standardized test scores
The activation threshold was p < 0.05, using a FDR correction for multiple comparisons and an extent threshold of 15 voxels. Test abbreviations as in Table 1.
Abstract
Purchase PDF (965 K)
E-mail Article
Add to my Quick Links
Cited By in Scopus (0)
Purchase PDF (111 K)
