Abstract
Correlated sensory inputs coursing along the individual sensory processing hierarchies arrive at multisensory convergence zones in cortex where inputs are processed in an integrative manner. The exact hierarchical level of multisensory convergence zones and the timing of their inputs are still under debate, although increasingly, evidence points to multisensory integration (MSI) at very early sensory processing levels. While MSI is said to be governed by stimulus properties including space, time, and magnitude, violations of these rules have been documented. The objective of the current study was to determine, both psychophysically and electrophysiologically, whether differential visual-somatosensory (VS) integration patterns exist for stimuli presented to the same versus opposite hemifields. Using high-density electrical mapping and complementary psychophysical data, we examined multisensory integrative processing for combinations of visual and somatosensory inputs presented to both left and right spatial locations. We assessed how early during sensory processing VS interactions were seen in the event-related potential and whether spatial alignment of the visual and somatosensory elements resulted in differential integration effects. Reaction times to all VS pairings were significantly faster than those to the unisensory conditions, regardless of spatial alignment, pointing to engagement of integrative multisensory processing in all conditions. In support, electrophysiological results revealed significant differences between multisensory simultaneous VS and summed V + S responses, regardless of the spatial alignment of the constituent inputs. Nonetheless, multisensory effects were earlier in the aligned conditions, and were found to be particularly robust in the case of right-sided inputs (beginning at just 55 ms). In contrast to previous work on audio-visual and audio-somatosensory inputs, the current work suggests a degree of spatial specificity to the earliest detectable multisensory integrative effects in response to VS pairings.
Highlights
Human event-related potentials (ERP) studies have shown that when information from various sensory modalities is presented concurrently, multisensory interactions often occur within the first 100 ms post-stimulation (e.g., Schroger and Widmann, 1998; Giard and Peronnet, 1999; Foxe et al, 2000; Fort et al, 2002; Molholm et al, 2002, 2004; Schürmann et al, 2002; Foxe and Schroeder, 2005; Murray et al, 2005)
They showed that integration in superior colliculus (SC) neurons was greatest for inputs presented simultaneously or in close temporal coincidence, that the magnitude of the multisensory effect was inversely related to the effectiveness of the constituent unisensory inputs, and of particular importance to the current study, that multisensory integration (MSI) was greatest for stimuli presented to the same spatial location
In Murray et al (2005), we found that the earliest AS interactions were localized to auditory cortical regions, and yet the laterality of this effect was tied to the side of somatosensory stimulation rather than auditory stimulation; suggesting that the more precise spatial information available to the somatosensory system dominated during early sensorycortical integration
Summary
Human ERP studies have shown that when information from various sensory modalities is presented concurrently, multisensory interactions often occur within the first 100 ms post-stimulation (e.g., Schroger and Widmann, 1998; Giard and Peronnet, 1999; Foxe et al, 2000; Fort et al, 2002; Molholm et al, 2002, 2004; Schürmann et al, 2002; Foxe and Schroeder, 2005; Murray et al, 2005). Foxe et al (2000) showed auditory-somatosensory (AS) interactions at just 50 ms, a finding corroborated and extended by Gonzalez Andino et al (2005) and Murray et al (2005) In all of these studies, in addition to the consistent finding of early multisensory interactions, spatio-temporal mapping has revealed a family of subsequent multisensory processing stages across a widely distributed network of sensory and higherorder regions. Much of the current work has been guided by the seminal work of Stein, Meredith, Wallace and colleagues who, in a series of studies using single-unit recordings in the superior colliculus (SC) of cats and monkeys, detailed a basic set of principles for MSI (Stein et al, 1975, 1993; Meredith and Stein, 1986; Meredith et al, 1987; Stein and Wallace, 1996; Wallace et al, 1996). In the ongoing attempt by ERP and neuroimaging researchers to detail the functional significance of the aforementioned cortical integration effects, these principles have provided a solid launching point (see Foxe, 2008)
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
