Related Experiment Video
Updated: Oct 26, 2025

06:03
Author Spotlight: Investigating the Impact of Aging on Hippocampal-Dependent Spatial Learning
Published on: February 16, 2024
2.5K
Mice learn multi-step routes by memorizing subgoal locations
Philip Shamash1, Sarah F Olesen1, Panagiota Iordanidou1
1UCL Sainsbury Wellcome Centre for Neural Circuits and Behaviour, London, UK.
Nature Neuroscience
|July 30, 2021
Summary
Mice learn complex routes by remembering key locations, not just by repeating movements or building maps. This "subgoal memorization" helps them navigate efficiently, even in new environments.
Area of Science:
- Neuroscience
- Animal Behavior
- Cognitive Science
Background:
- Mammalian navigation strategies for multi-step routes remain largely unknown.
- Understanding how animals learn complex paths is crucial for cognitive science.
Purpose of the Study:
- To investigate how mice learn to navigate multi-step routes, specifically to shelter under threat.
- To determine the underlying behavioral strategies and memory mechanisms involved in route learning.
Main Methods:
- Observing mouse behavior in response to threats with blocked direct paths to shelter.
- Analyzing navigation patterns, obstacle negotiation, and the adoption of specific strategies over time.
- Testing the role of spatial memory and comparing findings with existing models of spatial learning.
Main Results:
- Mice initially used sensory cues to navigate around obstacles.
- Within 20 minutes, mice adopted a 'subgoal strategy,' running to obstacle edges.
- This strategy persisted even after obstacle removal, indicating spatial memory use.
- A hybrid approach involving memorizing salient locations from 'practice runs' explained memory formation, not standard models.
Conclusions:
- Mice utilize a novel 'subgoal memorization' strategy to learn efficient multi-step routes.
- This strategy involves remembering salient locations encountered during exploratory runs.
- Subgoal memorization appears to be a fundamental mechanism for rodents learning new environments and tasks.

