Gardner Lab

Psychology Department

Neurosciences Institute

Stanford University


日本語   

Differences

This shows you the differences between the selected revision and the current version of the page.

shared:vision_lunch 2018/05/17 00:01 shared:vision_lunch 2018/05/22 23:51 current
Line 10: Line 10:
===== Next meeting: 5/29/18 Guillaume Riesen ===== ===== Next meeting: 5/29/18 Guillaume Riesen =====
-=== TBA ===+=== Rivalry and fusion can coexist in a tristable dynamic state ===
== Abstract == == Abstract ==
-TBA+Dichoptic inputs are either ‘fused’ together or engage in binocular rivalry - outcomes considered separately in most of the literature. Experiments and models have typically treated them as mutually exclusive, with fusion not reportable in most rivalry paradigms. We find a dynamic “tristable” state in which fused, left- or right-eye dominant percepts can result from static stimuli. Statistical analysis indicates that this is not simply the result of a parallel fusional process intermittently halting or interrupting rivalry; rather both should be considered part of a single process. 
 + 
 +We parametrically varied the fusability of dichoptic stimuli and had observers report their appearance over time. 19 participants were shown dichoptic Gabor patches (carrier spatial frequency: 5 cpd, SD=0.7) with orientation disparities of [0 10 20 25 30 35 40 50 90] degs about the vertical using a pellicle beam splitter. Observers reported the apparent orientations of the patches (clockwise, counter-clockwise, vertical or unsure) relative to reference points for two one-minute periods each. For non-zero disparities, rotated appearances signified left- or right-eye views while verticality indicated fusion. To elicit fusion without stereopsis, a second set of stimuli were shown with everything rotated 90 degs to produce vertical disparities. 
 + 
 +Perception was tristable between fusion and rivalry for patches ~30 degs apart. 7 participants were excluded due to slow rivalry (mean durations >25s at 90 degs) or non-fused percepts at zero disparity. The rest reported exclusive rivalry for disparities >~35 degs and exclusive fusion at <~20 degs. At ~20-35 degs, subjects reported periods of both. The same was observed for vertical disparities, but with exclusive rivalry from ~30 degs on.  
 + 
 +We reject two basic models where a separate fusion process interrupts rivalry to produce tristability. The ‘storage’ model has rivalry paused during fusion, as seen when attention is removed from rivalry. The dominant eye before and after fusion would then usually be the same. The ‘interruption’ model has rivalry continue underneath fusion. This would cause differences in the durations of rivalry periods flanking fusion (which are ‘cut short’ by it). Bootstrapped simulations of these models do not match our data. 
 + 
 +We have demonstrated that certain static stimuli can be seen as fused and rivalrous over time, revealing a novel temporal aspect of fusion as it alternates with rivalry. Statistical analysis rules out simple combinations of rivalry and fusion processes and presents the need for a dynamic model capable of generating all three outcomes. 
====== Schedule ====== ====== Schedule ======
Line 31: Line 40:
^Date^Speaker/JC^Title^ ^Date^Speaker/JC^Title^
|05/23/18|No Vision Breakfast (VSS)|-| |05/23/18|No Vision Breakfast (VSS)|-|
-|05/29/18|Talk: Guillaume Riesen (PhD Student, Gardner Lab, Stanford)|TBA|+|05/29/18|Talk: Guillaume Riesen (PhD Student, Gardner Lab, Stanford)|Rivalry and fusion can coexist in a tristable dynamic state|
|06/06/18|Journal Club: Elias Wang|Image reconstruction by domain-transform manifold learning (https://www.nature.com/articles/nature25988)| |06/06/18|Journal Club: Elias Wang|Image reconstruction by domain-transform manifold learning (https://www.nature.com/articles/nature25988)|
|06/13/18|Talk: Hsin-hung Li|TBA| |06/13/18|Talk: Hsin-hung Li|TBA|