A scalable touchscreen-based spatial working memory task for cross-species research

Area of Science:

• Neuroscience
• Cognitive Science
• Primate Models

Background:

• The common marmoset is a valuable nonhuman primate model in neuroscience.
• Development of cognitive tasks for marmosets has not kept pace with neural and genetic tools.
• Scalable and mechanistically informative cognitive paradigms are needed.

Purpose of the Study:

• Introduce and validate a touchscreen-based spatial working memory task.
• Enable direct cross-species translation between marmosets and humans.
• Provide parametric control over memory maintenance and interference demands.

Main Methods:

• A Delayed Non-Match-to-Position task was administered on touchscreens.
• Twelve marmosets and seventy-one humans participated.
• Retention delay (1, 5, 10s) and spatial separation were systematically varied.

Main Results:

• Accuracy decreased with increased delay and decreased spatial separation in both species.
• Delay and spatial separation showed interacting, additive effects on performance.
• Choice latency analyses supported performance interpretations, with slower responses on incorrect trials.

Conclusions:

• The developed task is scalable and translationally aligned for spatial working memory research.
• This paradigm effectively captures interacting maintenance and interference processes.
• It serves as a platform for circuit-level investigations and cognitive assays in marmosets.