Eye Tracking Transsaccadic Perception

The human visual system constantly integrates information across saccades, to create a stable and continuous perception of the world. However, visual processing differs significantly between the fovea (central vision) and the periphery. In their recent paper, “A bias in transsaccadic perception of spatial frequency changes,” Sharvashidze, Hübner, and Schütz (2024) addressed two key questions. Firstly, whether spatial frequence (SF) appearance differs before and after a saccade, and secondly, how this potential discrepancy impacts transsaccadic change discrimination. Previous research suggested that SF appears higher in the periphery compared to the fovea, leading to a hypothesis that the visual system might predict SF decreases after a saccade.

Transsaccadic Change Discrimination Methodology

The researchers conducted two experiments. Experiment 1 focused on both appearance discrimination (comparing SF before and after a saccade) and change discrimination (identifying SF increases or decreases during a saccade). Experiment 2 aimed to rule out response bias and further examine the underlying mechanisms by comparing transsaccadic change discrimination with presaccadic SF discrimination, where participants only had to discriminate SF based on presaccadic information 

Eye movements were recorded by an SR Research EyeLink 1000 Plus, sampling at 1000 Hz, which allowed the researchers to:

• Accurately detect saccade onset and offset: This was crucial for controlling the timing of stimulus presentation (presaccadic and postsaccadic) and ensuring that changes occurred precisely during the saccade.
• Monitor gaze position: Eye tracking ensured participants were fixating correctly on the central stimulus and making the intended saccades to peripheral targets. This was vital for distinguishing between foveal and peripheral processing.
• Identify and exclude problematic trials: Trials with blinks, saccades landing outside the target region, or abnormal saccade latencies were identified and excluded, maintaining the integrity and reliability of the data.
• Analyze saccade latencies: The researchers analyzed saccade latencies as a function of presaccadic SF, replicating previous findings that longer saccade latencies are associated with higher presaccadic SFs. This provided additional insights into the dynamics of visual processing before a saccade.

Without the high temporal and spatial precision offered by the EyeLink system, it would have been impossible to accurately time the stimulus changes relative to saccade execution, or to control for the subtle but significant variations in eye movements that could influence perception. 

Results Suggest Masking or Overwriting of Presaccadic Information

该研究得出了几个重要发现：

• No consistent difference in SF appearance across saccades: Contrary to some previous findings, Experiment 1 found no significant difference in the perceived SF before and after a saccade.
• A strong bias to report SF increases: Despite the lack of appearance differences, participants showed a clear and persistent bias to perceive SF increases across saccades. This bias was evident in both blank and no-blank conditions of Experiment 1 and replicated in Experiment 2, ruling out a simple response preference.
• Improved precision with postsaccadic blanking: A 200-ms postsaccadic blank improved the precision of SF change discrimination in Experiment 1, but it did not eliminate the directional bias. Interestingly, this improvement was not observed in Experiment 2, suggesting that blanking may be less effective when stimuli are viewed parafoveally rather than foveally.
• Presaccadic discrimination is more precise than transsaccadic change discrimination: Experiment 2 revealed that participants could discriminate SFs with much higher precision in a presaccadic-only task compared to the transsaccadic change discrimination task. This suggests that transsaccadic change detection is impaired, possibly due to masking or overwriting of presaccadic information by postsaccadic input.
• Evidence for masking/overwriting: The authors propose that the observed bias might stem from more effective masking or overwriting of the presaccadic stimulus by postsaccadic low SF stimuli, especially during a transition from high to low SFs. This is further supported by differences in processing speeds between magnocellular (low SF) and parvocellular (high SF) pathways.

The research provides valuable insights into the complexities of transsaccadic perception, particularly concerning spatial frequency changes. The persistent bias towards perceiving SF increases across saccades, coupled with evidence suggesting masking or overwriting of presaccadic information, opens new avenues for understanding how the visual system integrates dynamic visual input. This study illustrates how high-precision eye tracking is not just a tool, but an integral component of experimental design in vision research, enabling researchers to meticulously control for and measure the subtle yet profound interactions between eye movements and perception.

有关眼动追踪如何帮助您的研究的信息，请查看我们的解决方案和产品页面或联系我们。我们很乐意为您提供帮助！