Canonical retinotopic shifts under an inverse force field explain predictive remapping

ABSTRACTSaccadic eye-movements allow us to bring visual objects of interest to high-acuity central vision. Although saccades cause large displacements of retinal images, our percept of the visual world remains stable. Predictive remapping — the ability of cells in retinotopic brain areas to transiently exhibit spatio-temporal retinotopic shifts beyond the spatial extent of their classical receptive fields — has been proposed as a primary mechanism that mediates this seamless visual percept. Despite the well documented effects of predictive remapping, no study to date has been able to provide a mechanistic account of the neural computations and architecture that actively mediate this ubiquitous phenomenon. Borne out by the spatio-temporal dynamics of peri-saccadic sensitivity to probe stimuli in human subjects, we propose a novel neurobiologically inspired phenomenological model in which the underlying peri-saccadic attentional and oculomotor signals manifest as three temporally overlapping forces that act on retinotopic brain areas. These three forces – a compressive one toward the center of gaze, a convergent one toward the saccade target and a translational one parallel to the saccade trajectory – act in an inverse force field and specify the spatio-temporal window of predictive remapping of population receptive fields.

Attention Trade-Off for Localization and Saccadic Remapping

Predictive remapping may be the principal mechanism of maintaining visual stability, and attention is crucial for this process. We aimed to investigate the role of attention in predictive remapping in a dual task paradigm with two conditions, with and without saccadic remapping. The first task was to remember the clock hand position either after a saccade to the clock face (saccade condition requiring remapping) or after the clock being displaced to the fixation point (fixation condition with no saccade). The second task was to report the remembered location of a dot shown peripherally in the upper screen for 1 s. We predicted that performance in the two tasks would interfere in the saccade condition, but not in the fixation condition, because of the attentional demands needed for remapping with the saccade. For the clock estimation task, answers in the saccadic trials tended to underestimate the actual position by approximately 37 ms while responses in the fixation trials were closer to veridical. As predicted, the findings also revealed significant interaction between the two tasks showing decreased predicted accuracy in the clock task for increased error in the localization task, but only for the saccadic condition. Taken together, these results point at the key role of attention in predictive remapping.

Spatial updating of attention across eye movements: A neuro-computational approach

While scanning our environment, the retinal image changes with every saccade. Nevertheless, the visual system anticipates where an attended target will be next and attention is updated to the new location. Recently, two different types of perisaccadic attentional updates were discovered: Predictive remapping of attention before saccade onset (Rolfs, Jonikaitis, Deubel, & Cavanagh, 2011) as well as lingering of attention after saccade (Golomb, Chun, & Mazer, 2008; Golomb, Pulido, Albrecht, Chun, & Mazer, 2010). We here propose a neuro-computational model located in LIP based on a previous model of perisaccadic space perception (Ziesche & Hamker, 2011, 2014). Our model can account for both types of updating of attention at a neural systems level. The lingering effect originates from the late updating of the proprioceptive eye position signal and the remapping from the early corollary discharge signal. We put these results in relationship to predictive remapping of receptive fields and show that both phenomena arise from the same simple, recurrent neural circuit. Thus, together with the previously published results, the model provides a comprehensive framework to discuss multiple experimental observations that occur around saccades.

Predictive remapping of visual features beyond saccadic targets

Visual stability is thought to be mediated by predictive remapping of the relevant object information from its current, pre-saccadic locations to its future, post-saccadic location on the retina. However, it is heavily debated whether and what feature information is predictively remapped during the pre-saccadic interval. Using an orientation adaptation paradigm, we investigated whether predictive remapping occurs for stimulus features and whether adaptation itself is remapped. We found strong evidence for predictive remapping of a stimulus presented shortly before saccade onset, but no remapping of adaptation. Furthermore, we establish that predictive remapping also occurs for stimuli that are not saccade targets, pointing toward a ‘forward remapping’ process operating across the whole visual field. Together, our findings suggest that predictive feature remapping of object information plays an important role in mediating visual stability.