Related Experiment Video
Updated: Jun 29, 2026

07:21
Automated Interactive Video Playback for Studies of Animal Communication
Published on: February 10, 2011
Symbol-value association and discrimination in the archerfish
Naomi Karoubi1, Tali Leibovich2, Ronen Segev1
1Life Sciences Department and Zlotowski Center for Neuroscience, Ben Gurion University of the Negev, Beer Sheva, Israel.
Plos One
|April 6, 2017
Summary
Archerfish can associate quantities with symbols, demonstrating value-based decision-making and numerical ordering abilities previously unseen in lower vertebrates. This suggests an approximate magnitude system may be widespread in the animal kingdom.
Area of Science:
- Comparative Cognition
- Ethology
- Neuroscience
Background:
- Symbolic representation of quantities is crucial for mathematical cognition in humans.
- Evidence suggests this ability extends to primates, birds, and dolphins, but its prevalence in lower vertebrates is unknown.
Purpose of the Study:
- To investigate whether lower vertebrates, specifically archerfish, can associate symbolic representations with quantities.
- To determine if archerfish can make decisions to maximize rewards based on symbolic quantity information.
- To explore the extent of numerical understanding and ordering in fish.
Main Methods:
- Training archerfish to link different geometric shapes with varying amounts of rewards.
- Observing decision-making processes related to food intake based on learned associations.
- Assessing the fish's ability to mentally organize and understand numerical sequences.
Main Results:
- Archerfish successfully associated specific quantities with distinct geometric symbols.
- Fish demonstrated decision-making to maximize food intake, indicating value-based choices.
- Archerfish showed the capacity to understand and order up to four quantities.
Conclusions:
- Lower vertebrates like archerfish possess the ability to associate values with symbols.
- These findings suggest the existence of an approximate magnitude system in fish.
- The study indicates a broader evolutionary basis for numerical cognition across diverse animal groups.

