Brain activity for peripheral biological motion in the posterior superior temporal gyrus and the fusiform gyrus: Dependence on visual hemifield and view orientation

Abstract

Biological motion, the movement of the human body presented by a small number of point lights, activates among other regions lining the posterior superior temporal sulcus (pSTS) and gyrus (pSTG) and of the fusiform gyrus. In previous studies with foveal stimuli the activity in the pSTS/pSTG was often confined to the right hemisphere and bilateral in fusiform gyrus. We presented biological motion stimuli in peripheral vision and measured the BOLD responses with functional MRI to test whether the right dominance in pSTS/pSTG also occurred with peripheral stimuli. We found activation exclusively in the right pSTG for both visual hemifields. In the fusiform gyrus activation was found in both hemispheres and for peripheral stimuli strongest for contralateral stimulation. However, in both fusiform gyri leftward-facing stimuli activated different subfields than rightward-facing stimuli, indicating a clustering of the selectivity for the orientation of the human body form. No such clustering was observed in the pSTG. The results indicate for the fusiform gyrus an organization with respect to the view orientation of the stimulus.

Introduction

The human visual system is equipped with mechanisms sensitive to activities performed by other individuals. For example, humans can easily recognize actions, such as walking, from moving point lights attached to the major joints of an otherwise invisible body (Johansson, 1973). The recognition of such point-light walkers is known as biological motion perception. Many brain imaging studies investigated the neuronal networks underlying biological motion perception. Among others, they identified regions in the posterior bank of the human superior temporal sulcus (pSTS) (Beauchamp et al., 2003, Bonda et al., 1996, Grèzes et al., 2001; Grossman and Blake, Grossman and Blake, 2001, Grossman and Blake, 2002, Grossman and Blake, 2004, Grossman et al., 2005, Michels et al., 2005, Pelphrey et al., 2003, Peuskens et al., 2005, Puce et al., 1998, Santi et al., 2003, Saygin et al., 2004, Thompson et al., 2005) and gyrus (pSTG) (Grèzes et al., 1998, Howard et al., 1996, Santi et al., 2003, Servos et al., 2002, Vaina et al., 2001) and the fusiform gyrus (Beauchamp et al., 2003, Bonda et al., 1996, Grossman and Blake, 2002, Grossman and Blake, 2004, Michels et al., 2005, Peelen and Downing, 2005, Pelphrey et al., 2005, Ptito et al., 2003, Santi et al., 2003, Vaina et al., 2001). Most of these studies reported stronger activation in the right pSTS/pSTG than in the left pSTS/pSTG (Beauchamp et al., 2003, Bonda et al., 1996, Grèzes et al., 1998, Grèzes et al., 2001, Grossman and Blake, 2001, Grossman et al., 2000, Grossman et al., 2005, Pelphrey et al., 2003, Peuskens et al., 2005, Puce et al., 1998, Santi et al., 2003, Wheaton et al., 2004). A possible explanation for this asymmetric activation pattern is a functional lateralization. However, previous imaging studies have used only parafoveal stimuli. Therefore, it is unknown how well the right hemisphere dominance holds up for peripheral stimuli.

The perception of biological motion differs somewhat between foveal and peripheral viewing. Detection of biological motion in random dot noise is more difficult in the periphery than in the parafovea (Ikeda et al., 2005), presumably because of differences in visual grouping processes that are required to join the individual light points into a coherent body structure. Indeed, peripheral discrimination of point-light walkers is good if stimuli are not embedded in noise (Thompson et al., 2007). We have recently observed an asymmetry of the recognition ability of biological motion in the visual periphery in which a walker facing away from fixation is better recognized than a walker facing towards fixation (de Lussanet et al., 2008). This behavioral observation could be traced back to asymmetrical BOLD activation by the walker stimulus in areas of the mirror-neuron system (which also responds to biological motion; Santi et al., 2003, Tai et al., 2004).

Here we use the BOLD activations in posterior temporal cortex to investigate the organization of pSTS/pSTG for peripheral biological motion stimulation. The processing of central and peripheral visual stimuli is organized retinotopically in lower and mid-level visual areas (Engel et al., 1997, Huk et al., 2002, Sereno et al., 1995) and the strongest activations occur usually in the hemisphere contralateral to the stimulus. Higher visual areas, such as pSTS/pSTG, are thought to lack such retinotopy. In monkeys, cells in STPa (presumably homologous to human pSTS) possess large receptive fields that extend to the ipsilateral visual field without any retinotopic organization (Bruce et al., 1981). Cells in STPa respond to peripherally presented biological motion (Oram and Perrett, 1994). Electrophysiological studies in STPa furthermore showed that biological motion sensitive cells often show a preference for a particular orientation (i.e. facing direction) of the walker stimulus, or for a combination of orientation and motion direction of the walker (e.g. facing right and walking forward) (Jellema et al., 2004, Oram and Perrett, 1994, Oram and Perrett, 1996). Other neurons in STPa respond to static views of bodies or faces (Perrett et al., 1991, Perrett et al., 1994). Since biological motion perception may be achieved by the analysis of templates (Lange and Lappe, 2006, Lange et al., 2006) or snapshots (Giese, 2004) of human body configuration it is interesting to investigate any functional specialization within pSTS/pSTG for different orientations of the walker. In the monkey, cells recorded during presentation of walking stimuli did not seem to cluster by their function. Cells with different sensitivities (form, motion, and location) were found within a range of < 1 mm (Jellema et al., 2004). However, the STS region contains a functional organization for objects of different visual categories (Logothetis et al., 1999, Pinsk et al., 2005, Tsao et al., 2003). For instance, Pinsk et al. (2005) reported distinct face and body-selective regions in the posterior and anterior STS.

In contrast to STPa, cells in the inferotemporal cortex (ITC) of the monkey, a possible homologue of the human fusiform gyrus, are anatomically clustered by their function for stimuli of the same object category (Tanaka, 1996, Wang et al., 1998). Wang et al. (1998) recorded responses of ITC cells to different facing directions of profile and front views of faces. The critical features for the activation of single cells were first determined in unit recordings with electrodes. In subsequent optical imaging, Wang et al. (1998) looked for the representation of the critical features and showed that the critical features activated different patchy regions, covering the site of the electrode penetration at which the critical feature had been determined. Because signals in optical imaging reflect average neuronal activities in the examined regions, the optical imaging result indicates a regional clustering of cells in the ITC by their feature selectivity. Some functional clustering is also seen in human fusiform gyrus. For instance, pictures of entire human bodies activate a different region of the fusiform gyrus than images of faces (Peelen and Downing, 2005, Peelen et al., 2006). With respect to template- or snapshot-based models of biological motion perception it is interesting to study responses to body actions with different facing directions in the fusiform gyrus, since the fusiform gyrus may provide shape information about body orientation for the analysis of body motion (Lange and Lappe, 2006).

In the present study we ask whether there are functional sub-regions in the fusiform gyrus that are specific for the facing direction of biological motion. Furthermore, we test earlier findings that the right STS is activated more strongly than the left STS with peripheral biological motion. Third, we investigate the differences in the activation in the fusiform gyrus for ipsilateral and contralateral peripheral stimulation.