Autism spectrum disorder (ASD) is a neurodevelopmental disorder characterized by qualitative impairments in reciprocal interactions and communication as well as repetitive and stereotyped behaviors. As to neural substrate, subcortical structures such as the amygdala–hippocampus region (Endo et al., 2007), the thalamus (Tamura et al., 2010) and other cortical regions including the mirror neuron system (MNS) have been examined as potential foci of interest for autism spectrum disorder (ASD). However, the rostral prefrontal cortex (RoPFC) is also involved in higher cognitive processing, and willed action may also be involved (Dumontheil et al., 2008). Postmortem studies in ASD patients have shown minicolumn abnormalities in the RoPFC (Brodmann area (BA) 10) and the adjacent prefrontal cortex (PFC) (BA 9 (Casanova et al., 2006)). These findings form the basis for investigating these regions in patients with ASD (Casanova et al., 2006). Moreover, the RoPFC may be susceptible to developmental abnormalities because of its prolonged maturation period (Dumontheil et al., 2008).
In the context of ASD, the RoPFC has been investigated predominantly through the use of word fluency tasks (WFT) (Kawakubo et al., 2009). These tasks act as a common probe for executive function, which is thought to activate the RoPFC. However, WFTs are not specifically related to the psychopathology of ASD, and the RoPFC is involved in a wide range of other functions including willed action, executive control (Burgess et al., 2007) and intentional imitation (Chiavarino et al., 2007). Impairments in imitation and the associated involvement in the MNS have been implicated in ASD (Iacoboni and Dapretto, 2006, Rizzolatti et al., 2006, Williams et al., 2006 and Verhoeven et al., 2010); however, there are ongoing debates regarding these findings (Southgate and Hamilton, 2008), especially on the other brain regions beyond narrowly defined MNS. Therefore, within the present study we sought to measure hemodynamic responses in the RoPFC during two kinds of intentional imitation tasks of hand motion, anatomical imitation (AI) and mirror-image imitation (MI), with functional near-infrared spectroscopy (fNIRS), a methodology that enables the subjects to freely move their hands in a natural position. To our knowledge, this is the first fNIRS study to examine the issue in detail. We assume that a remaining component by subtracting MI from AI would represent an executive process of looking from the other person's viewpoint.
It is hypothesized that functional hemispheric difference in the RoPFC may be revealed when ASD patients and typically developing individuals are examined with tasks emphasizing the above component. The rationale for investigating the hemispheric differences in the RoPFC is derived from two major fields. One is from the findings that brain regions closely connecting with the RoPFC show significant differences in structural and functional asymmetry between ASD and control subjects. Of the three brain regions densely connecting with the RoPFC (the superior temporal gyrus (STG) through the extreme capsule, the amygdala through the uncinate fasciculus, and the anterior cingulate gyrus (ACG) through the cingulate fasciculus (Petrides and Pandya, 2007)), asymmetric activation of the STG have been well documented in ASD.
The other rationale derives from findings about the size and the white matter integrity of the corpus callosum (CC). A meta-analysis found a significantly small CC in autism, with the greatest reduction of the rostral body containing motor neurons from the PFC (Frazier and Hardan, 2009). Diffusion tensor imaging studies also revealed microstructural abnormalities of inter-hemispheric white matter, even in high-functioning autistic children: reduced fractional anisotropy (FA) and increased mean diffusivity of the genu of the CC (Alexander et al., 2007), and higher apparent diffusion coefficient and lower white matter density of the anterior third transcallosal fiber tracts (Hong et al., 2011). Reduced FA of the CC was found concurrently in white matter adjacent to the ventromedial PFC as well as in ACG, the STG and the temporal lobes approaching the amygdala (Barnea-Goraly et al., 2004), which are known to connect with the RoPFC as mentioned before. These studies suggest that functional lateralization may be present in the RoPFC and may play a crucial role in controlling hierarchically lower brain regions such as the ACG, the STG, and the amygdala.
