Bilateral dorsal and ventral ber pathways for the processing of affective prosody identied by probabilistic ber tracking Sascha Frühholz a,b, , Markus Gschwind c,d , Didier Grandjean a,b a Neuroscience of Emotion and Affective Dynamics Laboratory, Department of Psychology, University of Geneva, Geneva, Switzerland b Swiss Center for Affective Sciences, University of Geneva, Geneva, Switzerland c Laboratory of Neurology and Imaging of Cognition, Department of Neurosciences, University of Geneva, Geneva, Switzerland d Neurology Service, Department of Clinical Neurosciences, Centre Hospitalier Universitaire Vaudois, Lausanne, Switzerland abstract article info Article history: Accepted 2 January 2015 Available online 9 January 2015 Keywords: Emotion Prosody Voice DTI Probabilistic ber tracking Auditory pathways Dorsal and ventral pathways for syntacto-semantic speech processing in the left hemisphere are represented in the dual-stream model of auditory processing. Here we report new ndings for the right dorsal and ventral temporo-frontal pathway during processing of affectively intonated speech (i.e. affective prosody) in humans, to- gether with several left hemispheric structural connections, partly resembling those for syntacto-semantic speech processing. We investigated white matter ber connectivity between regions responding to affective prosody in several subregions of the bilateral superior temporal cortex (secondary and higher-level auditory cor- tex) and of the inferior frontal cortex (anterior and posterior inferior frontal gyrus). The ber connectivity was investigated by using probabilistic diffusion tensor based tractography. The results underscore several so far underestimated auditory pathway connections, especially for the processing of affective prosody, such as a right ventral auditory pathway. The results also suggest the existence of a dual-stream processing in the right hemisphere, and a general predominance of the dorsal pathways in both hemispheres underlying the neural pro- cessing of affective prosody in an extended temporo-frontal network. © 2015 Elsevier Inc. All rights reserved. Introduction Cortical auditory processing involves several perisylvian regions, which are interconnected by different ber pathways. Recent studies (Friederici et al., 2006; Rauschecker and Scott, 2009; Saur et al., 2008) have predominantly identied left hemispheric processing pathways within a dual-stream model of auditory processing (Hickok and Poeppel, 2007). They include ventral pathways from anterior superior temporal gyrus (STG) to the anterior inferior frontal gyrus (IFG) and dorsal pathways, which project to the posterior IFG via the posterior STG (Hickok and Poeppel, 2007; Rauschecker and Scott, 2009). Especial- ly the dorsal pathway seems strongly left lateralized (Hickok and Poeppel, 2007). The ventral pathways convey sound-invariant meaning (Belin and Zatorre, 2000b; Rauschecker and Scott, 2009), such as speech semantics (Hagoort, 2005). The dorsal pathways serve sound-to-motor mapping (Saur et al., 2008) and the processing of temporal auditory se- quences (Belin and Zatorre, 2000b; Rauschecker and Scott, 2009), which are also necessary for the understanding of speech syntax (Friederici et al., 2006). Compared to a predominant role of the left brain for syntacto-semantic processing (Specht, 2014), the emotional intonation in speech, that is the affective prosody, strongly, but not ex- clusively, activates regions in right STG and IFG (e.g. Alba-Ferrara et al., 2011; Beaucousin et al., 2007; Ethofer et al., 2006; Fruhholz et al., 2012). Thus, investigating the neural basis of affective prosody processing pro- vides an ideal paradigm to investigate right hemispheric auditory path- ways using diffusion-weighted imaging techniques together with functional magnetic resonance imaging. These temporo-frontal pathways for affective prosody processing have been rarely studied (Ethofer et al., 2012; Glasser and Rilling, 2008), mainly pointing to a right dorsal pathway (Gharabaghi et al., 2009; Glasser and Rilling, 2008), but also providing evidence for the possibility of a right ventral pathway (Ethofer et al., 2012). However, these studies explored pathways, rst, only for circumscribed temporal regions, second, without specifying frontal target regions, third, without quantifying the architecture of these pathways, and therefore without dissociating the different possible functional roles of these pathways (Ethofer et al., 2012; Glasser and Rilling, 2008). Taken together, these critical points might have led to a considerable underestimation of the importance and complexity of the temporo-frontal white matter path- way connectivity. Ventral and dorsal pathways, for example, are sup- posed to originate in multiple STG seed regions (Friederici, 2011; Fruhholz et al., 2012). Furthermore, these pathways probably terminate in the anterior as well as in the posterior IFG (Fruhholz and Grandjean, 2013b), serving to evaluate (Schirmer and Kotz, 2006) and to categorize NeuroImage 109 (2015) 2734 Corresponding author at: Swiss Center for Affective Sciences University of Geneva, 9 Chemin des Mines, CH-1202 Geneva, Switzerland. E-mail address: sascha.fruehholz@unige.ch (S. Frühholz). http://dx.doi.org/10.1016/j.neuroimage.2015.01.016 1053-8119/© 2015 Elsevier Inc. All rights reserved. Contents lists available at ScienceDirect NeuroImage journal homepage: www.elsevier.com/locate/ynimg