**2. Magnetic Resonance Imaging (MRI) studies**

#### **2.1. Head circumference and total brain volume**

Atypical head circumference growth curve in the first two years of life is a phenotypical risk indicator for autism [5, 6]. In the case of autism, head circumference that is normal or near normal size at birth follows an accelerated growth pattern at about four months [7, 8]. It has been shown that 37% of autistic children between the ages of two and four meet the criteria for developmental macrocephaly [9].

© 2015 The Author(s). Licensee InTech. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

In a follow-up study by Dawson et al. [2007], involving 28 children with autism from birth up to 36 months, has been shown accelerated growth of head circumference and previ‐ ous studies have indicated that repeatedly [7]. This head circumference growth pattern occurs independently from autistic regression [10]. In autism, head circumference growth rate has been compared to the period from birth to 12 months; after the 12th month, these findings shows that unusually rapid head growth is limited to the 1st year of life [11]. Many studies have shown that the behavioural symptoms of autism become easily understanda‐ ble during the first eight- to 12 months. [12, 13]. Therefore, the onset of accelerated head circumference growth between four- and 12 months partially leads to significant behaviou‐ ral symptoms and an overlap between these symptoms. Interestingly, Dawson et al. reported a slowdown in the rate of head circumference growth in the next 12 months to be associated with a loss of or deceleration in the acquisition of new skills. Another study [7] reported that rapid growth in head circumference between birth and the 12-month period, followed by a slowdown in growth after 12 months to be a risk indicator for the development of autism symptoms at 24. months [14].

Magnetic resonance imaging (MRI) is used in order to measure the size and shape of brain structures. The results of research conducted using the MRI method has been consistent with results yielded by head circumference studies in terms of autism. Sparks et al. (2002) reported that children aged three to four with autism had a significantly larger total brain volume compared to normally developing peers or developmental delayed peers [15]. Another study [16] showed that 90% of children with autism spectrum disorders aged between two and four had larger brain volumes than usual. Using the MRI method, the size of the brain among autistic children aged between one-and-a-half and four years old was shown to be abnormally increased (about 5-10%) [8, 15, 16]. Courchesne et al. [16] proposed that an increase in brain volume among children with autism in younger age groups tended to decrease front-to-back (maximum at the frontal lobe, with the occipital lobe growth showing the least). However, it has not been clearly determined whether the growth pattern of young children (four years old) is permanent or not among older children and adolescents [16, 17]. It has been found that brain size in autistic children at birth was 13% smaller than the control group, reached a 10% larger size at the age of one and was at the onset of puberty only 2% larger than the control group; these results were obtained by Redcay and Courchesne [6] following the evaluation of head circumference and brain weight using MRI brain volume and autopsy studies. It has been reported that in a large proportion of individuals with autism in adulthood, the brain volume did not differ from healthy controls [6].

The hypothesis has been put forward that unusual brain growth curves lead to an abnormal pattern of changes in cortico-cortical connections. Changes in the brain during development adversely affect the development and persistence of growth curves among short-and longhaul connections. It has been suggested that the growth rate of brain size is less than the normal rate where developmental disorders are concerned, which leads to an increase in long-haul connections. Moreover, if brain size is larger than normal in developmental disorders such as autism, leads to a reduction in long-distance structural and functional connections [18].

#### **2.2. White and grey matter changes**

In a follow-up study by Dawson et al. [2007], involving 28 children with autism from birth up to 36 months, has been shown accelerated growth of head circumference and previ‐ ous studies have indicated that repeatedly [7]. This head circumference growth pattern occurs independently from autistic regression [10]. In autism, head circumference growth rate has been compared to the period from birth to 12 months; after the 12th month, these findings shows that unusually rapid head growth is limited to the 1st year of life [11]. Many studies have shown that the behavioural symptoms of autism become easily understanda‐ ble during the first eight- to 12 months. [12, 13]. Therefore, the onset of accelerated head circumference growth between four- and 12 months partially leads to significant behaviou‐ ral symptoms and an overlap between these symptoms. Interestingly, Dawson et al. reported a slowdown in the rate of head circumference growth in the next 12 months to be associated with a loss of or deceleration in the acquisition of new skills. Another study [7] reported that rapid growth in head circumference between birth and the 12-month period, followed by a slowdown in growth after 12 months to be a risk indicator for the

Magnetic resonance imaging (MRI) is used in order to measure the size and shape of brain structures. The results of research conducted using the MRI method has been consistent with results yielded by head circumference studies in terms of autism. Sparks et al. (2002) reported that children aged three to four with autism had a significantly larger total brain volume compared to normally developing peers or developmental delayed peers [15]. Another study [16] showed that 90% of children with autism spectrum disorders aged between two and four had larger brain volumes than usual. Using the MRI method, the size of the brain among autistic children aged between one-and-a-half and four years old was shown to be abnormally increased (about 5-10%) [8, 15, 16]. Courchesne et al. [16] proposed that an increase in brain volume among children with autism in younger age groups tended to decrease front-to-back (maximum at the frontal lobe, with the occipital lobe growth showing the least). However, it has not been clearly determined whether the growth pattern of young children (four years old) is permanent or not among older children and adolescents [16, 17]. It has been found that brain size in autistic children at birth was 13% smaller than the control group, reached a 10% larger size at the age of one and was at the onset of puberty only 2% larger than the control group; these results were obtained by Redcay and Courchesne [6] following the evaluation of head circumference and brain weight using MRI brain volume and autopsy studies. It has been reported that in a large proportion of individuals with autism in adulthood, the brain volume

The hypothesis has been put forward that unusual brain growth curves lead to an abnormal pattern of changes in cortico-cortical connections. Changes in the brain during development adversely affect the development and persistence of growth curves among short-and longhaul connections. It has been suggested that the growth rate of brain size is less than the normal rate where developmental disorders are concerned, which leads to an increase in long-haul connections. Moreover, if brain size is larger than normal in developmental disorders such as autism, leads to a reduction in long-distance structural and functional connections [18].

development of autism symptoms at 24. months [14].

32 Autism Spectrum Disorder - Recent Advances

did not differ from healthy controls [6].

Abnormal brain growth in autistic children primarily stems from cerebral white and grey matter. However, Herbert and colleagues asserted that this growth originates from the disproportionate increase of white matter, not grey matter [19]. Abnormalities in white matter volume can be linked to differences in axonal density and organization, myelination abnor‐ malities or the abnormal proliferation of glial cells [20]. In two different studies with autistic children (between one-and-a-half and four years old) has been shown to significantly increase white matter rather than grey matter [8, 16]. However, it is not clear whether this increase in older children and adolescents is permanent or not [21, 22]. Even though the growth rate of grey matter has been shown to be smaller than that of white matter in early life, it is reported to be persistent in adulthood [16, 21].

Diffusion tensor imaging (DTI), also called diffusion tensor magnetic resonance imaging (DT-MRI) is a method used to assess the integrity of white brain matter. One of the parameters determined by this method, which identifies the movement of water molecules in the brain, is fractional anisotropy, which reflects asymmetry in fluid movement [23]. High fractional anisotropy values reflect a more intense or more proper structure of the brain. Two separate studies in children and adults with autism have shown a decrease in cerebral white matter using fractional anisotropy. A decrease in the temporal cortex among adults with autism has also been indicated using fractional anisotropy [24], as well as on the ventromedial prefrontal cortex, anterior cingulate, temporal lobe, amygdala and cortical and along subcortical regions containing the corpus callosum in autistic children and adolescents [25]. The most consistent findings concerning decreases in the brain using fractional anisotropy has been found for the corpus callosum [25, 26]. In another study, a 14% reduction in the size of the corpus callosum was shown, which is associated with decreased fractional anisotropy on genu and splenium [27]. In contrast to these findings, Ben-Bashat et al. [28] observed an association between increased fractional anisotropy values and white matter maturity in autistic children at younger ages. These findings have been associated with abnormal increases in brain volume during the early ages of children with autism [28].

A reduction in fractional anisotropy alongside an increase in white matter volume may reflect abnormal connections in the form of increased non-myelinated white matter connectivity. Extremism in the weak links due to the activity of immature myelination may adversely affect information processing. A decrease in white matter integrity reduces the brain's functional integration,afactorthathasservedasabasisofcurrenttheoriesaboutabnormalconnections[20].

Recently, corpus callosum abnormalities have been associated with the theory of insufficient common functional connectivity (underconnectivity) in adults with autism [29]. Functional brain imaging studies have shown a decrease in the activation of the synchronization of many brain regions concerning different functions like social content interpretation [30], working memory [31], executive functions [29] and visual imagery [32]. These findings have led to the hypothesis that insufficient cortical connections may be associated with autistic disorders [33]. The reduction of white matter structural integrity in the context of autism may cause differ‐ ences in functional connectivity, while theory of mind deficits [34] within the autism context can potentially be responsible for weak central coherence as well as social and cognitive symptoms [26].

#### **2.3. Cerebral cortex**

The largest and most consistent increase has been reported for the frontal lobe, despite a grey and white matter increase having been in the frontal, temporal and parietal lobes in several studies [21, 35, 36]. In autism, there exists an opposite growth rate from the norm following the period of accelerated growth in cerebral and cerebellar regions. For example, it has been shown that frontal lobe grey and white matter volume increase by 19% at two- and four years of age and 46% between the ages of nine and 12 in normal children, while in autistic children, these rates are 1% and 14%, respectively [36, 37].

In autism, abnormal asymmetry patterns in the frontal and temporal regions associated with language have been observed. Herbert et al. has shown that lateral inferior frontal cortex language (Broca's area is associated with pars opercularis) has reverse asymmetry in children with autism. It has also been shown that autistic men have a 27% larger volume in the right side of the frontal language region, compared to controls' 17% larger volume on the left side of the frontal language region. In addition to these findings, planum temporal asymmetry has been reported to be fairly different between the two groups; autistic males showed a 25% left dominance, while this rate was only 5% for the control group. Differences in terms of rightsided symmetry at the supramarginal posterior gyrus have been identified between autistic (39% greater) and control groups (greater than 2%). Another asymmetry region is the posterior superior temporal cortex (greater at the right in controls) associated with Wernicke's area, although this is not statistically significant. Structural abnormalities like abnormal asymmetry observed in autistic males in language regions may be associated with abnormalities in language skills [38].

#### **2.4. Cerebellum**

The cerebellum plays a role on coordination between voluntary movements and complex movements. Data obtained from animal and human studies have shown that the cerebellum may play a role in cognitive processes, in language use and emotion [39, 40]. Several MRI studies have determined that when autistic patients in different age groups are compared with a control group, there is a significant increase in cerebellar volume. This increase in cerebellar volume function is usually proportional to the total brain volume [41]. Different from other studies, in a study with autistic children below the age of three (18-35 months), there was no difference in terms of cerebellum size [8]. In contrast to the increase in the total volume of the cerebellum, some autistic children showed a relatively small volume vermis [42, 43]. Specifi‐ cally cerebellum region is vermian lobules VI-VII that reported about volume reduction in autism [16]. Vermis hypoplasia in autism is associated with deficits in automatically attention directing and research behaviour. [37]. Courchesne et al. [44] suggest that autistic disorders have two subtypes associated with cerebellum pathology: [1] vermis hypoplasia and [2] vermis hyperplasia [44]. However, no differences have been reported concerning IQ levels being affected by vermal volume in different study groups matched for intelligence quotient (IQ) [45]. In addition, it has been suggested that increases in the volume of the frontal lobe is associated with a reduction in cerebellar vermis volume. It has been shown that patients with normal vermis volume have normal frontal cortex volume, while patients with vermis hypoplasia have greater frontal cortex volume. Abnormal neuronal signals from subcortical structures can affect the development of the cerebral cortex and increased neuronal activity can in this way lead to growth among neuronal elements. Therefore, it has been claimed that abnormal neuronal activity in the cerebelloretinal-thalamocortical projections (possibly associated with a reduction in inhibitor signals as a result of the premature reduction in the number of cerebellar Purkinje cells) can lead to developmental failure in the frontal lobe and in the other input regions [37]. These volume changes are not specific to autism and are also commonly found in various developmental and psychiatric disorders [43, 46]. In autism, hypoplasia at the structure of the brain stem has been identified by Hashimoto et al. [47] in a study that evaluated cerebellum and brain stem structures.

#### **2.5. Amygdala**

**2.3. Cerebral cortex**

34 Autism Spectrum Disorder - Recent Advances

language skills [38].

**2.4. Cerebellum**

these rates are 1% and 14%, respectively [36, 37].

The largest and most consistent increase has been reported for the frontal lobe, despite a grey and white matter increase having been in the frontal, temporal and parietal lobes in several studies [21, 35, 36]. In autism, there exists an opposite growth rate from the norm following the period of accelerated growth in cerebral and cerebellar regions. For example, it has been shown that frontal lobe grey and white matter volume increase by 19% at two- and four years of age and 46% between the ages of nine and 12 in normal children, while in autistic children,

In autism, abnormal asymmetry patterns in the frontal and temporal regions associated with language have been observed. Herbert et al. has shown that lateral inferior frontal cortex language (Broca's area is associated with pars opercularis) has reverse asymmetry in children with autism. It has also been shown that autistic men have a 27% larger volume in the right side of the frontal language region, compared to controls' 17% larger volume on the left side of the frontal language region. In addition to these findings, planum temporal asymmetry has been reported to be fairly different between the two groups; autistic males showed a 25% left dominance, while this rate was only 5% for the control group. Differences in terms of rightsided symmetry at the supramarginal posterior gyrus have been identified between autistic (39% greater) and control groups (greater than 2%). Another asymmetry region is the posterior superior temporal cortex (greater at the right in controls) associated with Wernicke's area, although this is not statistically significant. Structural abnormalities like abnormal asymmetry observed in autistic males in language regions may be associated with abnormalities in

The cerebellum plays a role on coordination between voluntary movements and complex movements. Data obtained from animal and human studies have shown that the cerebellum may play a role in cognitive processes, in language use and emotion [39, 40]. Several MRI studies have determined that when autistic patients in different age groups are compared with a control group, there is a significant increase in cerebellar volume. This increase in cerebellar volume function is usually proportional to the total brain volume [41]. Different from other studies, in a study with autistic children below the age of three (18-35 months), there was no difference in terms of cerebellum size [8]. In contrast to the increase in the total volume of the cerebellum, some autistic children showed a relatively small volume vermis [42, 43]. Specifi‐ cally cerebellum region is vermian lobules VI-VII that reported about volume reduction in autism [16]. Vermis hypoplasia in autism is associated with deficits in automatically attention directing and research behaviour. [37]. Courchesne et al. [44] suggest that autistic disorders have two subtypes associated with cerebellum pathology: [1] vermis hypoplasia and [2] vermis hyperplasia [44]. However, no differences have been reported concerning IQ levels being affected by vermal volume in different study groups matched for intelligence quotient (IQ) [45]. In addition, it has been suggested that increases in the volume of the frontal lobe is associated with a reduction in cerebellar vermis volume. It has been shown that patients with normal vermis volume have normal frontal cortex volume, while patients with vermis Amygdala volume shows an increase in proportion to the total cerebral volume in children with autism. Sparks et al. [15] found that autistic children (aged between 36-56 months) had an abnormal amygdala growth rate (13-16%). It has been reported that an increased amygdala volume (without increasing total cerebral volume) at three years of age is associated with severe progress in children aged between three and six years old [48]. An increase in amygdala volume has been associated with more severe anxiety [49] and with poor communication skills and social skills [48]. In a study with autistic males aged between the eight and 18 compared with a healthy control group, , it was found that amygdala volume had grown up to 15% between 8-12 age period, but in the period between the ages of 13-18 were found to be any difference at the amygdale volume. Amygdala volume among men with healthy development increased by about 40% between the ages of eight and 18; however, this was not the same for males with autism. These findings are important for the realization of amygdala volume initially being larger than normal in autistic children, and also important for indicate that autistic children have not age-related increase at amydala volume in the preadolescence period like healthy controls [50]. These findings are supported by some MRI studies, including those pertaining to autistic adolescents and adults where results showed an amygdala volume not significantly different or smaller [51] than those in control groups. It has been suggested that amygdala abnormalities in autism spectrum disorders play a central role in social symptoms [52].

#### **2.6. Hippocampus**

In autism, findings are contradictory regarding hippocampal volume. A MRI study by Schumann et al [50] showed that autistic children had an increase of hippocampal volume and this increase persisted during adolescence. A study that included autistic adolescents and young adults reported a decrease in hippocampal volume [51]. Where autism is concerned, various studies have shown no significant differences in hippocampal volume [53, 54].

#### **2.7. Corpus callosum**

The corpus callosum is responsible for transferring cortical and subcortical information between homologous regions of the cerebral hemisphere. It is associated with bilateral sensory and motor integration such as bimanual motor coordination, visual attention scrolling and procedural memory processes. In autism, especially in the posterior region of the corpus callosum, a volume decline was noted in [55]. These findings are associated with interhemi‐ spheric weakness in autism [20].

#### **2.8. Caudate nucleus**

It has been shown that the caudate nucleus volume is increased in autism. This increase may be associated with observed repetitive and ritualistic behaviour in adolescents and adults with autism [56, 57].
