Interference between smooth pursuit and color working memory

Shulin Yue, Zhenlan Jin, Chenggui Fan, Qian Zhang, Ling LiJournal of Eye Movement Research 2017 · DOI

Abstract

Spatial working memory (WM) and spatial attention are closely related, but the relationship between non-spatial WM and spatial attention still remains unclear. The present study aimed to investigate the interaction between color WM and smooth pursuit eye movements. A modified delayed-match-to-sample paradigm (DMS) was applied with 2 or 4 items presented in each visual field. Subjects memorized the colors of items in the cued visual field and smoothly moved eyes towards or away from memorized items during retention interval despite that the colored items were no longer visible. The WM performance decreased with higher load in general. More importantly, the WM performance was better when subjects pursued towards rather than away from the cued visual field. Meanwhile, the pursuit gain decreased with higher load and demonstrated a higher result when pursuing away from the cued visual field. These results indicated that spatial attention, guiding attention to the memorized items, benefits color WM. Therefore, we propose that a competition for attention resources exists between color WM and smooth pursuit eye movements.

Introduction

Cognitive psychologists have long established that spatial working memory (remembering where something is) and spatial attention are inextricably linked. However, the relationship between non-spatial working memory (remembering what something is, like its color) and eye movements remains a subject of debate.

Some theories suggest that visual working memory is modular, meaning color memory should not be affected by the motor task of tracking a moving target. Others suggest a "central executive" model where all complex tasks draw from a single pool of resources. This study sought to resolve this by testing whether smooth pursuit—a specific type of eye movement used to track moving objects—interferes with the ability to maintain color information in mind.

Methods and Materials

The researchers recruited 16 healthy college students (11 male, 5 female; mean age 22.8) for a behavioral study using an EyeLink 1000 Plus tracker. The experiment employed a modified delayed-match-to-sample (DMS) paradigm across three sessions: WM-only, pursuit-only, and a dual-task (WM + pursuit) session.

Figure 1: Experimental paradigm showing consistent (pursuing toward the remembered location) and inconsistent (pursuing away) conditions.
  1. Encoding: Subjects were shown 2 or 4 colored squares in both the left and right visual fields. A cue indicated which side to memorize.
  2. Retention: During the 1500ms retention interval, the squares vanished. In the dual-task condition, a central cross moved toward or away from the cued side, which subjects tracked smoothly.
  3. Consistency: This created two conditions: Consistent (pursuing toward the remembered location) and Inconsistent (pursuing away from it).
  4. Test: A test display appeared, and subjects indicated if any colors on the cued side had changed.

Key Insights and Results

Mutual Interference

The interference between tasks was found to be bidirectional. Adding a pursuit task significantly reduced color memory accuracy ($p=0.008$). Conversely, increasing the memory load from 2 to 4 items degraded the quality of the smooth pursuit, evidenced by lower pursuit gain and a higher frequency of "catch-up" saccades.

Spatial Scaffolding for Color

A striking finding was that pursuit direction significantly impacted performance. Even though the task required remembering colors, memory was significantly better when the eyes moved toward the location where the colors had previously been (the consistent condition).

Figure 4: Comparison of (A) accuracy and (B) reaction time between consistent and inconsistent conditions across memory loads.

As shown in Figure 4, accuracy was higher in the consistent condition than the inconsistent condition ($p=0.046$). Furthermore, higher memory loads and the presence of the pursuit task both significantly lengthened reaction times ($p < 0.05$). This suggests that spatial attention acts as a functional "rehearsal" mechanism even for non-spatial features like color.

Amodal Resource Competition

The authors propose that color WM and smooth pursuit share common attentional resources of a "central, amodal origin." This aligns with models of a "central executive" that manages different types of information. Interestingly, pursuit performance in the "inconsistent" condition was actually better than in the consistent condition. The authors liken this to an "anti-saccade" task: the brain works harder to suppress the urge to look at the remembered location, leading to higher cognitive control and better pursuit tracking.

Limitations and Future Work

The authors acknowledge they cannot entirely rule out the effect of retinal eccentricity; in the consistent condition, the eyes end up closer to where the test stimuli appear, which might provide a slight visual advantage. Additionally, subjects might automatically encode the location of the squares despite only needing to remember colors. Future research could explore whether other non-spatial features, such as shape or texture, show similar interference patterns with different types of eye movements.

Summary updated

Abstract

Spatial working memory (WM) and spatial attention are closely related, but the relationship between non-spatial WM and spatial attention still remains unclear. The present study aimed to investigate the interaction between color WM and smooth pursuit eye movements. A modified delayed-match-to-sample paradigm (DMS) was applied with 2 or 4 items presented in each visual field. Subjects memorized the colors of items in the cued visual field and smoothly moved eyes towards or away from memorized items during retention interval despite that the colored items were no longer visible. The WM performance decreased with higher load in general. More importantly, the WM performance was better when subjects pursued towards rather than away from the cued visual field. Meanwhile, the pursuit gain decreased with higher load and demonstrated a higher result when pursuing away from the cued visual field. These results indicated that spatial attention, guiding attention to the memorized items, benefits color WM. Therefore, we propose that a competition for attention resources exists between color WM and smooth pursuit eye movements.
BibTeX
@article{yue2017interference,
  author = {Shulin Yue and Zhenlan Jin and Chenggui Fan and Qian Zhang and Ling Li},
  title = {{Interference between smooth pursuit and color working memory}},
  journal = {Journal of Eye Movement Research},
  year = {2017},
  doi = {10.16910/jemr.10.3.6},
}
Figure