Child and adolescent males with ADHD have been found to show smaller caudate volumes (Filipek et al., 1997; Semrud-Clikeman & Pliszka, 2006). Studies in ADHD have found smaller volumes specific to the right caudate, in addition to smaller volumes of the right frontal lobe, right globus pallidus, and left cerebellum in males but not females (Castellanos et al., 1996). These consistent gender-based differences have been found in the cerebellum and in total brain volume for both genders with ADHD. Additionally, it has been suggested that females and males use different behaviorally relevant neurocognitive strategies. Several studies have found greater right prefrontal cortex activation in females. However, the spatiotemporal dynamics of neural events associated with these sex differences remains unclear. One event-related potential (ERP) study examined the neural substrates of human cognitive sex differences elicited by a visual attention challenge (an area of deficit often associated with ADHD) (Neuhaus et al., 2009). Equal numbers of male and female participants (22) matched for age, education, and nicotine use were studied with 29-channel-electroencephalography recorded under a visual selective attention paradigm, the Attention Network Test. Visual ERPs were topographically analyzed and neuroelectric sources were estimated. This analysis revealed a novel frontal-occipital second peak of the visual N100 that was significantly increased in females relative to males, as well as a corresponding central ERP component, that was recorded in females exclusively, at around 220 ms. Thus, this study found a strong sex-based difference in the relationship between stimulus salience and central ERP component amplitudes. Subsequent source analysis revealed increased cortical current densities in right rostral prefrontal (BA 10) and occipital cortex (BA 19) in female subjects that was not found in males. This study was the first to report a three-part association between sex differences in ERPs, visual stimulus salience, and right prefrontal cortex activation during attentional processing and suggests that anatomical differences exist between males and females with ADHD.

One complication in understanding ADHD is that it is considered a developmental disorder that, by current definition, has onset prior to 7 years of age. Imaging studies have provided some insight into brain differences, but thus far have almost exclusively focused on children 7 years of age and older. Recently, Mahone et al. (2011) completed a study using high-resolution anatomical (MPRAGE) images to better understand the neurobiological development of ADHD in younger children presenting with symptoms of the disorder, while controlling for language disorders. Cortical regions were delineated and measured in the study using automated methods in Freesurfer, while basal ganglia structures were manually delineated. Results indicate that even younger children with ADHD show significantly reduced caudate volumes bilaterally. However, there were no significant group differences in cortical volume or thickness in this age range. Furthermore, results indicated that, after controlling for age and total cerebral volume, left caudate volume was a significant predictor of hyperactive/impulsive behavior, but not inattentive symptom severity. Although results are not gender-specific, they suggest that irregularities in basal ganglia, particularly caudate development, appear to play an important role in children presenting with early onset symptoms of ADHD (Mahone et al., 2011).

The mechanisms underlying the increasingly recognized motivation deficits that contribute to ADHD are being examined through neuroimaging. Tomasi and Volkow (2011) compared the functional connectivity density between 247 ADHD and 304 typically developing control children from a public MRI database. Results of their analysis indicated enhanced connectivity within reward-motivation regions and decreased connectivity with regions from the default-mode and dorsal attention networks. Again, while not sex specific, this study suggests impaired interactions between control and reward pathways in ADHD that may contribute to both attention and motivation deficits.

Increasingly sophisticated methods examining cortical topography, cortical thickness, or voxel-based morphometry, as well as neuroimaging techniques that provide a means to uncover the neurobiological mechanisms by which gene variants impact the brain, are beginning to better demonstrate more clearly delimited anatomic abnormalities. The overall pattern of results from the more sophisticated imaging techniques is not clearly convergent, potentially suggesting heterogeneity in ADHD (Castellanos et al., 2009). Models focusing on intra-subject variability, intrinsic brain activity, and reward-related processing are progressing rapidly and should provide clearer understanding of the multiple types of ADHD, as well as a better understanding of variations of connectivity contributing to differences between ADHD processing anomalies in males and females. Additional studies focusing on how genetic changes can affect brain structure, chemistry, and function are also promising (Durston, 2010).

In conclusion, findings suggest that lower glucose metabolism is more characteristic of girls with ADHD. However this has not been tested with samples that control for sexual maturation, cognitive ability, and genetic relatives with ADHD. Additionally, studies have shown that both males and females with ADHD show stable reductions in brain volume as compared to controls, with males with ADHD found to show smaller caudate volumes. Emerging imaging techniques, particularly those based on the approach of resting-state functional connectivity, and genetic neuroimaging, appear to be promising in better understanding gender differences for the disorder, as well as offering a possible means of addressing the complex heterogeneity of ADHD as testable models of pathophysiology are developed.

While there is now an informative body of research on the use of neuroimaging techniques in revealing brain morphometric differences in people with learning disabilities (Altarelli et al., in press; Hynd & Semrud-Clikeman, 1989), most differences have generally been studied with either a mixed gender sample or with males only. These studies have typically been focused on the language regions of the brain, and language laterality and degree of difference has varied among the studies due to methodological issues. Most studies have not controlled for sex and have thus provided little information about possible sex-specific variations in brain structure and function in children with learning disabilities. However, with less invasive techniques becoming available, younger children are now beginning to be included in study samples.

Language acquisition, strongly tied to learning ability, is inherent in changes that occur in the rapidly developing young brain. During the years of language acquisition, the brain not only stores linguistic information but also adapts to the grammatical regularities of language. Advances in functional neuroimaging have substantially contributed to systems-level analyses of brain development, including overall language acquisition, including reading language, and understanding of the contributions of cortical plasticity for second language acquisition (Sakai, 2005). There is about a 2.5 to 3:1 gender difference in dyslexia (a very broad term defining a learning disability that impairs a person’s fluency or comprehension accuracy for reading) for boys to girls, with studies showing females have more cell density in the planum temporale than males (Witelson, Glezer, & Kigar, 1995; Witelson & Kigar, 1988a, 1988b, 2004; Witelson & McCulloch, 1991). Given the lower incidence of dyslexia in girls, it may be that this cell density is protective for reading and written language problems in females.

More recent neuroimaging studies investigating the neural correlates of verbal fluency (VF) focused on sex differences. It has been hypothesized that reading disabled children have a domain-general deficit in processing rapidly occurring auditory stimuli that degrades speech perception, thereby limiting phonologic awareness and so, reading acquisition. This was not proved to be the case in a study with 100 7–11-year-old children with learning disability and 243 non-learning disabled children who were evaluated on a two-tone auditory discrimination paradigm (Waber et al., 2001). Those with a learning disability committed more errors, but effects of timing were comparable to their non-disabled peers. The same result was obtained for a subsample of good and poor readers. The authors concluded that task performance did predict reading, spelling, and calculation ability. However, while neural processes underlying perception of speech and other auditory stimuli may be less effective in poor readers, rate appears to not be specifically affected regardless of sex.

Another study focusing on fluency rate found activation in the inferior frontal gyrus (IFG), insula, anterior cingulate cortex (ACC), medial frontal gyrus (mFG), superior (SPL) and inferior parietal lobules (IPL), inferior visual areas, cerebellum, thalamus, and basal ganglia (Gauthier et. al., 2009). This study was completed with right-handed, French-speaking, male (172) and female (159), college-age subjects performing a phonological verbal fluency task. Results indicated males, more than females, activate several regions involved in mental imagery strategies linked to word generation, including the lingual gyrus, inferior temporal gyrus (ITG) and posterior cingulate, in addition to one region linked to performance monitoring (the ACC), and other areas associated with the implementation of top-down control processes (right superior frontal gyrus (SFG) and right, dorsolateral prefrontal cortex (DLPFC)), as well as the cerebellum. Thus, findings indicated activity in three discrete subregions of the ACC related to sex, performance and their interaction, respectively. In addition to these differences, results show males seem to require greater use of visual mental strategies, subserved specifically by the precuneus, to reach a verbal fluency level equivalent with that of high-scoring females. High-scoring men also were found to solicit more areas in the left hemisphere implementing top-down control processes (DLPFC). Regardless of sex, a low level in verbal fluency scores related to greater activity in the dorsocaudal ACC. In females only, the low performance level is associated with a stronger activation in a more rostral part of the dorsocaudal ACC, which reinforces the role of the ACC in general performance monitoring. Findings also indicated that the cerebellum seems to contribute to the high verbal fluency performance level independently from sex.

Another investigation examined the impact of subject age, language task, and cortical region on the occurrence of sex differences in functional MRI with 205 (104 male, 101 female) right-handed, monolingual, English-speaking children between the ages of 5 and 18 (Plante et al., 2006). The study used fMRI at 3T to evaluate BOLD signal variation associated with sex, age, and their interaction. Brain activation in classical, left hemisphere language areas of the brain and their right homologues were assessed for sex differences. For this study, children completed up to four language tasks, which involved listening to stories, prosody processing, single word vocabulary identification, and verb generation. Sex difference for behavioral performance was found for the prosodic processing task only. Left lateralization was present for both frontal and temporal regions for all but the prosody task, although no significant sex differences were found for the degree of lateralization.

Although sex differences in visuospatial processing, also thought to contribute to learning difficulties, are a consistently robust finding in adults, few neuroimaging studies have examined this issue in younger populations. Clements-Stephens, Rimrodt, and Cutting (2009) used functional neuroimaging (MRI) to examine whether sex-based differences in visuospatial processing exist in children, and/or if they develop during childhood. Thirty-two participants, between 7 and 15 years of age (16 males, 16 females) matched on performance, participated in this study. Overall, both groups showed overlapping activation in the superior parietal lobes bilaterally, extrastriate cortex, and cerebellum. Males had significantly greater activation in the cerebellum and right lingual gyrus. Formal comparisons between age groups indicated older males show engagement of left hemisphere regions, while females show greater engagement of right hemisphere regions traditionally associated with visuospatial processing. Results also suggest older males, as compared to younger males, may engage regions that are associated with a visuomotor network, whereas females may tend to use areas indicated in spatial attention and working memory to complete visuospatial tasks. Furthermore, results suggest that differential engagement of networks associated with visuospatial processing may be due to differences in strategy use that are evident early on and continue to develop over time.

Connectivity neuroimaging studies are also beginning to be conducted with younger children, and appear to support the view that a left-lateralized brain network is crucial for language development in children. Kikuchi et al. (2011) used a custom-sized magnetoencephalography (MEG) system, a noninvasive brain imaging technique that is a practical neuroimaging method for use in young children, to investigate brain networks of 78 right-handed preschool children (32–64 months) while they listened to stories with moving images. MEG produces a reference-free signal, and is therefore a useful tool to compute coherence between two distant cortical rhythms. Results indicated that left dominance of parietotemporal coherence in theta band activity (6–8 Hz) was specifically correlated with higher performance of language related tasks. This laterality was not correlated with nonverbal cognitive performance, chronological age, or head circumference. These results suggest that it is not the left dominance in theta oscillation per se, but the left-dominant phase-locked connectivity via theta oscillation that contributes to the development of language ability in young children.

The growing body of evidence that male and female brains develop and mature at different rates (Holland et al., in press) suggests that early differences in brain development likely contribute to learning differences that may also be sex-specific. Much larger numbers of subjects would be needed to confirm findings from neuroimaging studies involving learning disabilities. However, it does appear that over time, males develop a more integrated visuomotor/visuospatial network while females appear to develop a more spatial attention/working memory system for learning. The differences in activation patterns in females may be mediated by a strong verbal strategy used early in development. Additionally, differences seen in males are consistent with visually based strategy use in which it appears that males may rely on imagery and are more hands-on when completing tasks. Males’ reliance on a visuomotor network may also account for the traditional advantage reported on visuospatial tasks. The differential engagement of different networks associated with visuospatial processing between males and females seen with more complex tasks (such as mental rotation) seems to be supported, which could reflect a true sex-based difference in visuospatial processing that may affect learning in general. Also, emerging research suggests that the in utero experience of testosterone may affect functional asymmetries and could be linked to some differences in learning. These areas of research, as well as the relationship of handedness and/or age effects in differences of area and volume for key structures, are variables that require additional study across sexes. Additionally, there are very few studies that include sufficient numbers of minorities, and the possible effects of ethnicity on brain development have not been explored sufficiently. As these students are often over-identified for special education remediation, there may be even more reason to develop gender-specific research in this area.

Characteristics that are attributed to autistic children, including impairments in social interaction, deviations in language usage, and restricted and stereotyped patterns of behavior have been found regardless of age, IQ, and gender. However, brain changes due to age, IQ, and gender may pose potential confounds in autism neuroimaging analyses (Kurth et al., 2011). Causes and contributing factors for autism continue to be poorly understood. The prevalence of children identified with autism is rising with the most recent data from the Center for Disease Control (2006) indicating an average of 1 in 110 children in the United States being diagnosed with autism. The most current information suggests that genetic and environmental factors contribute etiologically to this disorder. Data gained from twin, family, and genetic studies support a role for an inherited predisposition; with, clinical, neuroanatomic, neurophysiologic, and epidemiologic studies suggesting that gene penetrance and expression is likely influenced by the prenatal and early postnatal environment (Hertz-Picciotto et al., 2006). Some studies link autism to xenobiotic chemicals and/or viruses, including the CHARGE (Childhood Autism Risks from Genetics and Environment) project, begun in 2003 to investigate underlying environmental and genetic causes for autism and triggers of regression and to systematically study a wide spectrum of chemical and biologic exposures, susceptibility factors, and their interactions (Hertz-Picciotto et al., 2006). To date, CHARGE is the largest and most comprehensive assessment of children with autism ever undertaken. It aims to distinguish subgroups, or phenotypes, of autism based on thorough biomedical and behavioral analyses of affected children. A wide variety of studies have resulted from the ongoing CHARGE project, including the launch of another comprehensive autism-focused venture, the Autism Phenome Project (APP). However, little of the derived information is sex-specific.

Others have begun to research the relatives of people with autism who show milder expression of traits for the disorder, referred to as the Broader Autism Phenotype (BAP). Several neurofunctional and neuroanatomical studies of autistic individuals and their relatives using neuroimaging techniques such as fMRI, MRI, EEG, MEG, and DTI were compiled to examine important differences in brain structure, activity and connectivity in and between regions of the brain (Sucksmith, Roth, & Hoekstra, 2011). This extensive examination of the neural substrates of the BAP contributes to our overall understanding of the disorder, and helps to delineate better the possible heritable endophenotypes so that autism susceptibility can be more reliably indexed. The compiled research, although again not sex-specific, also adds to our understanding of the neural correlates of the cognitive aspects of autism (e.g., sensory perception, social cognition, and visual attention), and contributes toward a solution that will bridge the gap between genes and clinical autism diagnosis.

One hallmark symptom for children with autism includes deficits in the perception of social stimuli that may contribute to the characteristic impairments in social interaction. The cortical processing of voice is abnormal in even high-functioning autistic children (Groen et al., 2008). Significant differences in response time to the perception of a voice are noted with children with autism responding much more slowly compared to control children. No gender differences were found for either the autistic or control samples.

Another area of perception of social stimuli that may contribute to the characteristic impairments in social interaction for autistic children is the ability to understand and respond to emotions. Most neuroimaging studies that have reported gender differences in response to human emotions have used face photographs. One exception is a study completed by Fine, Semrud-Clikeman, and Zhu (2009) that used human face photographs of positive and negative emotions, along with video vignettes of positive and negative social human interactions, in an attempt to provide a more ecologically appropriate stimuli paradigm. Although this study involved healthy, non-autistic male and female young adults (ten each), findings contribute to the overall understanding of sex-specific differences in processing emotion. Using conservative ROI (region of interest) analysis, the authors found greater male than female activation to positive affective photographs in the anterior cingulate, mFG, SFG and superior temporal gyrus, all in the right hemisphere, with no significant ROI gender differences observed for the negative affective photos. More activation was noted in ROIs of the left posterior cingulate and the right ITG to positive social videos, and in the left middle temporal ROI for negative social videos for males as compared to females. Additionally, males were more lateralized than females. Furthermore, although more activation was observed overall to video compared to photograph conditions, males and females appear to process social video stimuli more similarly to one another than they do for still photos. As males are four times more likely to be diagnosed with autism, it may be that further research can expand on this information concerning differences in processing human emotion that may be useful in understanding sex-specific differences in emotional processing between autistic males and females.

Connective imaging has also been employed with autistic subjects. As autism has been hypothesized to reflect neuronal disconnection, several recent studies have examined key thalamic relay nuclei and cortico-thalamic connectivity in the pathophysiology of the disorder. One study, using DTI with 17 boys with ASD and 17 typically developing controls, produced results (using whole-brain voxel-wise analyses) that evidenced disturbances in the thalamo-frontal connections. These findings, although found in a sample that includes only males, so cannot be noted as sex-specific, emphasize the role of hypoconnectivity between the frontal cortex and thalamus in autistic children (Hong et al., 2011).

Imaging studies have also examined genetic variation in ASDs, particularly the oxytocinergic system that may modulate sociality and indicate risk for social dysfunction (Tost et al., 2010). Oxytocin, centrally released, facilitates offspring survival by initiating mother–infant bonding and the onset of maternal behavior. The neural architecture of the oxytocinergic system targets a variety of brain areas, but for children with autism, the most salient are those critical for emotion regulation (e.g., amygdala, lateral septum, and brainstem) (Lee et al., 2009). Given the known heritability of social behavior in humans, it is possible that the genetic variation in the oxytocinergic system may modulate sociality and contribute to risk for social dysfunction. Consequently, common variants in the oxytocin receptor gene (OXTR) have been examined in the context of risk for ASD. Toth et al. (2007) used a multimodal imaging intermediate phenotype approach to show that a common genetic variant in OXTR linked to social function predicts individual differences in brain structure, brain function, and personality in healthy humans. The same study provided evidence for a sex-dependent impact of OXTR genotype on limbic structures related to prosocial temperament. Together, these findings indicate a neural mechanism for genetically increased risk of social impairment, predominantly in males, that seems to be of potential relevance for autistic (and possibly other psychiatric) disorders. Other connectivity imaging studies using DTI and diffusion tensor tractography have been conducted with findings implicating the corpus callosum (Hong et al., 2011) and the cerebellar peduncles (Brito et al., 2009). Additionally, morphometric measures of the basal ganglia and thalami in 3–4-year-old children have been studied which indicate that rigid social behavior (repetitive behavior), common early in the clinical course of ASD, may be associated with decreased volumes of the basal ganglia and thalamus. Although these studies indicate reduced connectivity in corpus callosum, internal capsule, and superior and middle cerebellar peduncles and reduced volumes in the basal ganglia and thalamus as compared to the findings in age-matched healthy children, they still do not control for sex, and so it is unknown whether these findings manifest differently in males and females with autism.