Novel n-back spatial working memory task using eye movement response

Cameron B. Jeter, Saumil S. Patel, Anne B. SerenoBehavior 2011 · DOI

Introduction

Working memory (WM)—the ability to temporarily hold and manipulate information—is a critical biomarker for neurological and psychiatric diseases such as Parkinson’s and schizophrenia. Traditional assessments typically rely on manual responses, such as key presses. However, this introduces a significant "noise" factor: patients with these conditions often experience psychomotor slowing. Consequently, it becomes difficult to determine whether a delayed response stems from a cognitive memory deficit or a physical motor impairment.

Furthermore, standard n-back tasks often involve long, continuous streams of stimuli, which can be taxing for children or patients prone to fatigue. To address these challenges, researchers developed a simplified, discrete-trial task utilizing eye movements (saccades), which are faster and less influenced by the motor deficits common in clinical populations.

Methods

The researchers adapted the spatial n-back task into a saccade-based format using infrared eye-tracking. Unlike continuous streams, this task is structured into individual trials. Participants fixate on a central point while a sequence of peripheral stimuli (boxes) appears.

Experimental design showing the sequence of events for visible and not-visible conditions across 0-back and 1-back trials.

Upon a "go" signal (the disappearance of the fixation point), participants must look at the location where a stimulus appeared n items ago:

  • 0-back: Look at the most recent location shown (simple target acquisition).
  • 1-back: Look at the second-to-last location shown (requires memory maintenance and updating).

The study also investigated the "Visual Anchor" effect by comparing a Visible condition (where the last stimulus remained on screen) to a Not Visible condition (where it vanished). The task was validated across three experiments involving healthy adults and children to assess its robustness and developmental sensitivity.

Key Results

The primary finding was a clear "WM effect"—a significant performance gap between 0-back and 1-back trials—particularly in the "Not Visible" condition.

Error rates for 0-back and 1-back trials in both Visible and Not Visible conditions, demonstrating the increased cognitive load of the 1-back task.

The Impact of Visibility

In the "Visible" condition, the presence of the final stimulus acted as an external cue (a "where not to look" signal), artificially lowering the difficulty of the 1-back task. Removing this stimulus (the "Not Visible" condition) forced participants to rely entirely on internal working memory. In this condition, healthy adults showed a sharp increase in error rates from 0.90% (0-back) to 8.73% (1-back).

Developmental and Structural Robustness

  • Children: While children successfully completed the task, they exhibited a much larger WM effect than adults (error rates of ~4.25% for 0-back vs. ~25.81% for 1-back), reflecting the ongoing development of the prefrontal cortex.
  • Simplification: Even when the number of possible locations and sequence lengths were reduced, the WM effect remained statistically significant, proving the task is scalable for different cognitive ability levels.

Discussion and Future Directions

The use of saccades provides a "purer" measure of working memory by bypassing the neural pathways associated with manual limb movements. Additionally, the discrete-trial structure makes this task highly compatible with event-related fMRI designs, allowing for the isolation of brain activity during specific memory phases (e.g., encoding vs. retrieval).

While this study validated the task in healthy populations, its true potential lies in clinical applications. Future research should:

  1. Validate in Clinical Groups: Directly compare patients with Parkinson's or schizophrenia to healthy controls to confirm that saccadic responses effectively bypass psychomotor slowing.
  2. Increase Cognitive Load: Test 2-back or 3-back versions to identify the limits of spatial WM using eye movements.
  3. Neural Mapping: Integrate the task with fMRI to map the specific circuits involved in spatial WM updating.

By providing a more direct window into cognitive function, this saccade-based n-back task offers a promising tool for both basic research and clinical diagnostics.

Summary updated

BibTeX
@inproceedings{jeter2011novel,
  author = {Cameron B. Jeter and Saumil S. Patel and Anne B. Sereno},
  title = {{Novel n-back spatial working memory task using eye movement response}},
  booktitle = {Behavior},
  year = {2011},
  doi = {10.3758/s13428-011-0093-9},
}
Figure