Spatiotopic updating across saccades revealed by spatially-specific fMRI adaptation
Introduction
We perceive the visual world to be stable despite the fact we make frequent saccades that completely alter visual input. Compounding this issue, topographic visual representations in the brain are in eye-centered (retinotopic) rather than world-centered (allocentric/spatiotopic) coordinates. The issue of how the brain creates continuity between retinotopic and spatiotopic references frames over successive eye movements is a long-standing question (for review, see Melcher, 2011).
A critical aspect of visual stability is the spatial updating of salient objects across saccadic eye movements. This process would allow the brain to keep track of the most important items in the scene, either by continuously updating retinotopic maps based on a copy of the eye-movement command (“remapping”: Duhamel et al., 1992; for review, see Wurtz et al., 2011) or by explicitly representing spatiotopic coordinates (for review, see Burr and Morrone, 2011). If this spatial updating process includes information about the object, then it could allow the visual system to integrate information about that object over time, rather than starting afresh with each new fixation (Melcher and Morrone, 2003, Melcher and Colby, 2008). A number of studies have reported behavioral correlates of such trans-saccadic updating processes (Prime et al., 2006, Van Eccelpoel et al., 2008, Wittenberg et al., 2008, Demeyer et al., 2009, Ong et al., 2009, Fracasso et al., 2010, Fabius et al., 2016), although the neural mechanisms that underlie these processes remain a matter of debate.
At present, there are two main lines of neuroimaging evidence for spatial updating in humans. The first set of studies have presented a single stimulus for a time period prior to a horizontal saccadic eye movement that would bring that stimulus into the opposite hemifield (Merriam et al., 2003, Medendorp et al., 2005). They compared trials in which a pre-saccadic stimulus was present to those in which a saccade was made with a blank screen. The logic behind these studies is that a greater response in the hemisphere where the stimulus would have been after the saccade reflects an active remapping of the stimulus representation. One limit to these studies is that they investigate only the pre-saccadic stimulus and not whether the updating of this stimulus influences processing of the stimulus after the saccade. A second set of studies has measured processing of a post-saccadic stimulus based on the presence of a pre-saccadic stimulus. In particular, these studies have focused on object processing in inferior temporal cortex (McKyton and Zohary, 2007), motion processing in MT/MST (d'Avossa et al., 2007) and, most recently, on memory for an oriented grating (Dunkley et al., 2016). However, given that there was only a single stimulus on the screen that was presented in the same spatial location before and after the saccade, it leaves open the possibility that the changes in fMRI signal reported in such studies reflect a more general and high-level effect of repetition rather than a spatially-specific adaptation.
The question of whether saccadic updating in the brain is spatially specific, or not, remains a key issue for theories of visual stability (Bays and Husain, 2007). We investigated this question using a variant of previous studies that have measured repetition suppression (RS, or ‘fMRI-adaptation’: Grill-Spector et al., 2006) across saccades (McKyton and Zohary, 2007, Golomb et al., 2011). In general, a repeated stimulus should evoke a weaker fMRI response than a novel stimulus (Grill-Spector et al., 1999). Here, we introduced a manipulation in which the post-saccadic stimulus was either in the same or different location (upper/lower hemifield) from the pre-saccadic stimulus. This allowed us to distinguish retinotopic from spatiotopic representations while discounting more general effects of simply repeating the same stimulus. This was accomplished by comparing RS for trials in which the two stimuli were shown in the same versus different locations. Based on previous neurophysiological evidence and the theory of object-based remapping (Melcher and Colby, 2008), we hypothesized that spatial congruency effects would be present in frontal and parietal areas implicated in spatial maps and saccades, as well as in early visual areas involved in processing the stimulus.
Section snippets
Participants
Nine individuals participated in this study (4 female, mean age: 31.6). All participants were highly trained in running eye movement studies. Participants gave informed consent and all procedures were approved by the University of Trento Human Research Ethics Committee.
Stimuli and procedure
Stimuli were presented via a coil-mounted mirror and a rear-projected screen using ASF (A Simple Framework; Schwarzbach, 2011) based on the Psychophysics Toolbox (Pelli, 1997). Two fixation crosses were presented in the left and
Spatially-specific repetition suppression in frontal-parietal regions
The main focus of our analysis was the presence of spatially-specific adaptation effects when S1 and S2 were shown in the same spatial location on the screen (both upper or both lower) compared to when they were shown in different spatial locations. First, we confirmed that the presence or absence of a saccade had little effect on the processing of S2 once the effect of the saccade itself had been linearly subtracted (contrast: [S2noSacc > fixation-only] and [S2Sacc > saccade-only], panels b &
Discussion
In this study, we found spatially-specific repetition suppression for trials in which a stimulus was presented in the same world-centered location but different retinotopic locations before and after a saccade. We can divide the regions involved into two functional clusters. The first set of areas, the frontal eye fields (FEF) and posterior parietal cortex (PPC), have previously been implicated in the representation of spatial saliency maps that keep track of important items and update these
Acknowledgements
The research leading to these results has received funding from the European Research Council under the European Union's Seventh Framework Programme (FP/2007–2013: ERC Grant Agreement n. 313658). This research was supported by the Provincia Autonoma di Trento and the Fondazione Cassa di Risparmio di Trento e Rovereto.
References (48)
- et al.
Visual stability based on remapping of attention pointers
Trends Cogn. Sci.
(2010) - et al.
The frontal eye field as a prediction map
Progress. Brain Res.
(2008) - et al.
Parametric integration of visual form across saccades
Vision. Res.
(2010) - et al.
Trans-saccadic interactions in human parietal and occipital cortex during the retention and comparison of object orientation
Cortex
(2016) From thought to action: the parietal cortex as a bridge between perception, action, and cognition
Neuron
(2007)- et al.
Repetition and the brain: neural models of stimulus-specific effects
Trends Cogn. Sci.
(2006) - et al.
Differential processing of objects under various viewing conditions in the human lateral occipital complex
Neuron
(1999) Spatiotopic transfer of visual-form adaptation across saccadic eye movements
Curr. Biol.: CB
(2005)- et al.
Trans-saccadic perception
Trends Cogn. Sci.
(2008) - et al.
Spatial updating in human parietal cortex
Neuron
(2003)