research
Projects
Current Projects
(1) The influence of visual references on saccadic adaptation.
(2) The localization bias before saccades.
Past Projects
(1) Attention modulation on visuomotor transformation.
(2) Long-term saccadic adaptation induced in dark environment.
(3) A modeling study on perisaccadic mislocalization. (undergraduate thesis)
Now I will introduce these projects:
.
The influence of visual references on saccadic adaptation?
In our previous study, we have observed a long-term saccadic adaptation induced and tested in completely dark environment, which is very different from the previous reports of short-term saccadic adaptation. This phenomenon suggests that in this special condition, people tend to use a different strategy or visuospatial representation that is little recalibrated by daily experience. In other words, different systems are involved in visuospatial processing, and the brain will integrate the information from both systems. In dark environment without any references, for example, people can only depend on the head-centered position of the visual stimulus (egocentric system); and in light condition, people will refer to the its location relative to the screen (allocentric system). The discrete systems are just my conjecture, but they can explain the long-term adaptation in darkness.
To test this idea, I investigate on saccadic adaptation in different conditions. In this study, human subjects were asked to perform a double-step target task (that is, when they begin to move their eyes, the target point will jump to a new place) in three different conditions: 1) dark, 2)dim light on the screen, and 3) grid lines on the screen. Some of the current data from one subjects:
As shown in the figure:
1. The long-term adaptation in darkness is consistent with our previous observation.
2. The training in darkness seemed to have some effect on the saccadic gain (SG, saccade-amplitude divided by target-eccentricity) in dim light conditions, but is much smaller than the effect shown in darkness.
3. In the grid condition, the first testing session showed very little effect by previous adaptation training in dim and dark. One day after a training session in this grid condition, the subjects showed different SG levels when tested in three conditions: the SG in grid condition most recovered, in dim less recovered, and in dark almost no recovery. If there are two discrete systems processing ego and allo information respectively, the three different performances may suggest different weights of these two information. Grid: much more allo. Dim: allo+ego. Dark: almost ego. In this case the results can be well interpreted.
I am trying to confirm this phenomenon in more subjects and do some computational simulations to interpret the results.
The localization bias before saccades?
You may heard about the perisaccadic mislocalization. With references on the screen, people will misperceive the location of the rapid flash appearing around the time when they make saccade. Many studies reported a mislocalization toward saccadic target in ~100ms around saccade. But do you know? Earlier before saccade, there is a systematic localization bias toward the orininal fixation point (FP)! I first observed this phenomenon when I did experiments for my undergraduate thesis in the topic of perisaccadic mislocalization. Now it has been confirmed by the results from several subjects.
To explain the results, I assume that this localization bias is caused by the influence of some kind of activity before
saccade on the spatial working memory. It seems that there is a critical period for working memory formation. In this period, the location of the flash in mind is very sensitive to disturbance (saccadic-related activity). But I am not sure what's the source of
this activity. Can it be some enhanced attention to the original
fixation point before it shifts to the target? Or other
saccade-related representation such as remapping? I am still considering about the possible mechanism underlying this phenomenon, and hope to get any evidence from neurophysiological studies.
Attention modulation on visuomotor transformation?
This work has been accepted by Journal of Neurophysiology.
Covert Attention Regulates Saccadic Reaction Time by Routing between Different
Visual-oculomotor Pathways
Shaobo Guan, Yu Liu, Ruobing Xia and Mingsha Zhang
[Abstract]
Covert attention could implicitly modulate the behavioral performance such as saccadic reaction time (SRT): comparing saccades to the cued with uncued locations, SRT is shorter soon after the onset of a task irrelevant cue but becomes longer at ~200ms later, known as Inhibition of Return (IOR). But the mechanisms underlying attention modulation are not very clearly known. Here, we propose two possible mechanisms: either by selective routing visuomotor signal through different pathways (routing hypothesis), or by generally modulating the speed of visuomotor transformation (shifting hypothesis). To test these two hypotheses, we designed a Cue Gap Paradigm by introducing a gap interval to the conventional IOR paradigm. The cue manipulated the location of covert attention, and the gap interval resulted in a bimodal distribution of SRT, with an early mode (express saccade) and a late mode (regular saccade), due to the coexistence of two distinct visuomotor pathways. The routing hypothesis predicts changes in the proportion of express saccades versus regular saccades, while the shifting hypothesis predicts a shift of SRT distribution. Our results showed IOR at mean SRT level consistent with previous reports. Moreover, the cue-induced attention modulated the relative proportion of two modes, rather than shifted the mean reaction time of either mode. These results demonstrate that the covert attention modification of the mean SRT is largely attributed to selective routing between visuomotor pathways rather than generally modulating the speed of visuomotor transformation within visuomotor pathways.
[Keywords] covert attention, cue, express saccade, IOR, routing
Long-term saccadic adaptation induced in dark environment?
This work was recently submitted to Journal of Neurophysiology.
Visual impact but not decay of motor memory diminishing the retention of saccadic adaptation
Jing Wang, Ruobing Xia, Mingsha Zhang, Yujun Pan
[Abstract]
Saccadic adaptation has been considered as an ideal behavioral model for studying the mechanisms of motor learning and memory. Although accurate motor memory is essential in motor control, previous studies reported that the effect of saccadic adaptation on oculomotor system was rather short. For instance, even after 19 days of successively training, the saccadic gain (SG, saccadic-amplitude / target-eccentricity) could recover to pre-adaptation level within 7-10 days. Is it because the decay of motor memory is too fast to retain the learned motor performance, or because of the impact from other signals? Since vision in primates powerfully affects behavior, it is possible that the visual input interferes with motor memory and affects behavior. To test such possibility, we asked human subjects to perform a backward double-step target task in a completely dark booth in order to exclude the visual influence on behavioral performance. Through analyzing the characteristics of saccades, our data showed two distinct results comparing with previous studies. First, after 24-72 hours of a training session, the SG remained at lower or similar level as that in the end of the training session, which indicated that there was no significant recovery of SG in day(s) after training. Second, the effect of saccadic adaptation lasted more than 2 months, which was much longer than previous reports. We conclude that the previously observed short-term effect of saccadic adaptation is mainly caused by the impact of visual information, but not by the poor retention of motor memory.
[Keywords] saccadic adaptation, visual impact, motor learning, motor memory, darkness
A modeling study on perisaccadic mislocalization
This is my undergraduate project. I would like to share my PPT for my thesis defence. This is my first formal study in neuroscience, maybe very naive. But I enjoy the whole process very much.
The perisaccadic flash is a good probe to uncover the visuospatial neural process around the time of saccade, and also relates to other functions, such as working memory (how saccade influences the spatial memory of flash), attention (how perisaccadic attention changes and affects the spatial perception) and decision making (how people choose different strategies in dark and light environments).
