Role of Supramarginal Gyrus in the Integration of Visual Features Across Eye Movements

点击翻译按钮获取中文摘要

Vision is constantly augmented by rapid eye movements called saccades, resulting in frequent spatiotemporal interruptions between brief periods of stable fixation. Trans-saccadic integration of visual features (TSI) is thus fundamental to visual perception, and forms an ideal model system for understanding interactions between different signals in the visual system. In our previous NSERC-funded work we identified a network of cortical areas that contribute to TSI, including the frontal and parietal eye fields, dorsolateral prefrontal cortex, and early visual cortex. But until now, the key site(s) where trans-saccadic feature storage and integration occurs has remained elusive. Recently, using an fMRIa (adaptation) paradigm we identified an area in the supramarginal gyrus (SMG), immediately adjacent to the human parietal eye fields, that appears to be directly involved in the integration of object orientation across saccades. The SMG -located in the inferior parietal lobule and couched between the superior parietal and temporal cortices— is intriguing because it is associated with complex visual behaviors and has already been implicated in the integration of ‘dorsal stream’ (spatial/sensorimotor) and ‘ventral stream’ (feature/object) visual signals. Motivated by this exciting finding and its broader implications, we aim to focus on the key role of SMG in the TSI network through a trans-disciplinary approach. We will use 1) state-space models of spatial updating and feature integration to generate simulated predictions for our psychophysical and physiological experiments, 2) fMRIa adaptation protocols to confirm SMG’s role in integration for different types of feature integration, frames of reference, and types of eye movement, 3) diffusion tensor imaging to understand its connections with other areas of the brain involved in trans-saccadic integration, 4) fMRI-guided Trans-cranial magnetic stimulation (TMS) to localize the causal role of SMG signals for visuospatial integration, 5) magnetoencephalography (MEG) to investigate the timing and synchrony of SMG with other TSI sites, and 6) animal neurophysiology to explore the detailed cellular responses of SMG in monkeys trained to perform similar TSI tasks. Establishing the specific mechanism and role of SMG in TSI will provide an important advance for understanding active vision. But further, our strategic focus on area SMG could answer a fundamental question in neuroscience: demonstration of a key role of the inferior parietal lobule in the integration of dorsal and ventral stream signals for cognition and behavior. This topic (TSI) has numerous applications for human-computer interaction and computer vision, in areas ranging from transportation to robotics. Finally, our uniquely diverse combination of techniques will provide a powerful training ground for students and post-doctoral fellows.