Brain Struct Funct DOI 10.1007/s00429-017-1492-1 ORIGINAL ARTICLE Computational neuroanatomy of human stratum proprium of interparietal sulcus 1,2,3 2,4,5 1 Maiko Uesaki • Hiromasa Takemura • Hiroshi Ashida Received: 24 March 2017 / Accepted: 4 August 2017 Ó The Author(s) 2017. This article is an open access publication Abstract Recent advances in diffusion-weighted MRI areas associated with optic-flow processing using fMRI. (dMRI) and tractography have enabled identification of SIPS was identified bilaterally in all subjects, and its major long-range white matter tracts in the human brain. anatomical position and trajectory are consistent with Yet, our understanding of shorter tracts, such as those previous post-mortem studies. Subsequent evaluation of within the parietal lobe, remains limited. Over a century the tractography results using the linear fascicle evaluation ago, a tract connecting the superior and inferior parts of the and virtual lesion analysis yielded strong statistical evi- parietal cortex was identified in a post-mortem study: dence for SIPS. We also found that the SIPS endpoints are stratum proprium of interparietal sulcus (SIPS; Sachs, Das adjacent to the optic-flow selective areas. In sum, we show hemispha¨renmark des menschlichen grosshirns. Verlag von that SIPS is a short-range tract connecting the superior and georg thieme, Leipzig, 1892). The tract has since been inferior parts of the parietal cortex, wrapping around the replicated in another fibre dissection study (Vergani et al., intraparietal sulcus, and that it may be a crucial anatomy Cortex 56:145–156, 2014), however, it has not been fully underlying optic-flow processing. In vivo identification and investigated in the living human brain and its precise characterisation of SIPS will facilitate further research on anatomical properties are yet to be described. We used SIPS in relation to cortical functions, their development, dMRI and tractography to identify and characterise SIPS and diseases that affect them. in vivo, and explored its spatial proximity to the cortical Keywords Stratum proprium of interparietal sulcus Á Diffusion-weighted MRI Á Tractography Á fMRI Á Optic Electronic supplementary material The online version of this flow Visuo-vestibular integration article (doi:10.1007/s00429-017-1492-1) contains supplementary Á material, which is available to authorised users. & Maiko Uesaki Introduction [email protected] & Hiromasa Takemura Anatomical connections through the white matter axon [email protected] bundles (i.e. fascicles, tracts) establish fundamental fea- 1 Department of Psychology, Graduate School of Letters, tures of the brain’s information processing (Catani and Kyoto University, Kyoto, Japan Ffytche 2005; Catani and Thiebaut de Schotten 2012; 2 Japan Society for the Promotion of Science, Tokyo, Japan Bullock et al. 2005; Fields 2008, 2015; Wandell and Yeatman 2013; Wandell 2016). Diffusion-weighted mag- 3 Open Innovation and Collaboration Research Organization, Ritsumeikan University, Osaka, Japan netic resonance imaging (dMRI) and tractography provide a unique opportunity to identify and characterise the white 4 Center for Information and Neural Networks (CiNet), National Institute of Information and Communications matter tracts in the living human brain (Catani et al. 2002; Technology, and Osaka University, Suita, Japan Wakana et al. 2004; Mori and Zhang 2006; Catani and 5 Graduate School of Frontier Biosciences, Osaka University, Thiebaut de Schotten 2012; Craddock et al. 2013; Wandell Suita, Japan 2016; Rokem et al. 2017). 123 Brain Struct Funct A body of dMRI research has successfully identified convergence of visual and vestibular information regarding several major long-range white matter tracts, such as the self-motion, involving those optic-flow selective areas superior longitudinal fasciculus and the inferior longitudi- (Kleinschmidt et al. 2002; Wiest et al. 2004; Kova´cs et al. nal fasciculus, in a consistent manner with the known post- 2008; Fetsch et al. 2009; Butler et al. 2010; Prsa et al. mortem anatomy (Catani et al. 2002; Wakana et al. 2004; 2015; Uesaki and Ashida 2015). To fully understand the Schmahmann et al. 2007), and ergo opened new avenues to underlying mechanism of optic-flow processing, it is research on the properties of major human white matter essential to investigate how communication between the tracts in relation to development and diseases (Lebel et al. superior and inferior parts of the parietal cortex is sup- 2012; Yeatman et al. 2012a, 2014a; Ogawa et al. 2014; ported by the white matter anatomy. Malania et al. 2017). More recent dMRI studies have We used dMRI and tractography to identify the identified shorter white matter tracts including frontal aslant anatomical location and trajectory of SIPS in the living tract and vertical occipital fasciculus, which, partially for human brain. The ensemble tractography (Takemura et al. their relatively short trajectories, had previously received 2016a) yielded bilateral identification of SIPS in all sub- very little attention in the neuroscience literature (Thiebaut jects. Evidence for SIPS was evaluated based on the con- de Schotten et al. 2012; Catani and Thiebaut de Schotten sistency across datasets, comparison with post-mortem 2012; Catani et al. 2013; Yeatman et al. 2013, 2014b; fibre dissection studies (Sachs 1892; Vergani et al. 2014), Takemura et al. 2016b, 2017). Some of those studies have and the virtual lesion analysis (Pestilli et al. 2014; Leong suggested the importance of the shorter tracts in relation to et al. 2016; Gomez et al. 2015; Takemura et al. 2016b). We cognitive functions and diseases (Kinoshita et al. 2015; also explored the functional relevance of SIPS by per- Kemerdere et al. 2016; Kronfeld-Duenias et al. 2016; Duan forming fMRI experiments on the same subjects and et al. 2015; Takemura et al. 2016b; Lee Masson et al. 2017). examining the spatial proximity between the SIPS end- Here, we focus on a short white matter tract connecting the points and functionally defined cortical regions previously superior and inferior parts of the parietal cortex, wrapping associated with optic-flow processing (Cardin and Smith, around the inferior parietal sulcus. This tract was initially 2010, 2011; Greenlee et al. 2016). Results showed that the described by the German neurologist Heinrich Sachs (1892) dorso-lateral SIPS endpoints are near VIP, PcM and p2v, as the stratum proprium of interparietal sulcus (hereafter, we whereas the ventro-medial SIPS endpoints are near PIC?; refer to this tract as SIPS). Except for one recent fibre dis- placing SIPS in a good position to channel sensory signals section study replicating Sachs’s work (Vergani et al. 2014), between the distant cortical areas underlying visuo- this tract has been largely overlooked in the literature. Given vestibular integration necessary for optic-flow processing the functional MRI (fMRI) evidence indicating the involve- and perception of self-motion. Finally, we also demonstrate ment of the superior and inferior parts of the parietal cortex in evidence of SIPS in additional 90 subjects from publically crucial cognitive functions (Corbetta and Shulman 2002; available Human Connectome Project datasets. Culham et al. 2006; Uncapher and Wagner 2009;Cardinand For the first time in the living human brain, we describe Smith 2010; Blanke 2012;Greenleeetal.2016), SIPS is human SPIS in a manner consistent with the previous post- likely a necessary and important tract supporting those mortem dissection studies. Our findings confirm that SIPS functions. Yet, the characteristics of SIPS are poorly under- is a short-range tract connecting the superior and inferior stood due to the lack of studies replicating SIPS in the living parts of the parietal cortex, wrapping around the intra- human brain, using three-dimensional digital anatomical data parietal sulcus. The findings also highlight that in vivo such as dMRI and reproducible computational analyses. identification and characterisation of SIPS open new ave- One of the cortical functions that involve the parietal nues to studying this tract in relation to cortical functions, cortex is optic-flow processing. Optic flow is the pattern of their development, and diseases that affect them. visual motion signals elicited by self-motion (Gibson To facilitate future research, we make the code to 1950, 1954), and is an important cue to accurate perception identify human SIPS publicly available in Github reposi- of self-motion. A network of sensory areas in the parietal tory [https://github.com/htakemur/SIPS] and also as part of cortex has been shown to be involved in optic-flow pro- AFQ toolbox [Yeatman et al. 2012b; https://github.com/ cessing (Cardin and Smith 2010, 2011). Those optic-flow yeatmanlab/AFQ]. selective areas include the ventral intraparietal area (VIP), the precuneus motion area (PcM) and the putative area 2v (p2v) located in the superior part of the parietal cortex, and Materials and methods the posterior-insular complex (PIC?: Deutschla¨nder et al. 2004; Cardin and Smith 2010, 2011; Biagi et al. 2015; We analysed dMRI data of 100 human subjects, from three Uesaki and Ashida 2015; Wada et al. 2016) in the inferior independent datasets. One set of dMRI data was acquired at part of the parietal cortex. Several studies have described a Kokoro Research Center, Kyoto University (KU dataset), 123 Brain Struct Funct along with fMRI measurements to identify cortical regions for tractography based on a model that is capable of activated by optic-flow