Shared system for ordering small and large numbers in monkeys and humanspost by Bruce W. Lee (bruce-lee) · 2024-02-09T04:45:52.957Z · LW · GW · 0 comments
This is a link post for https://pubmed.ncbi.nlm.nih.gov/16683927/
Cantlon, J. F., and Brannon, E. M. (2006) conducted a study to explore the numerical understanding in monkeys and its comparison with human adults, providing substantial evidence for the existence of a shared, nonverbal system for numerical representation across species. Their research investigated the ability of rhesus monkeys to apply learned numerical rules beyond their training range and directly compared this capability to that of humans in a numerical ordering task.
The authors initiated their study by training two adult female rhesus macaques to order numerical values from 1 through 9. The training was followed by testing the monkeys with novel numerical values (10, 15, 20, and 30) to assess their ability to extend the learned rule to larger, untrained values. The results showed that monkeys could indeed apply the learned ordinal rule to these larger values, suggesting that their numerical representation does not have an upper limit and is influenced by the ratio of the values compared rather than their absolute magnitude.
Further, Cantlon and Brannon (2006) conducted a second experiment to compare the performance of monkeys and humans on the same ordinal comparison task, involving the numerical values 2 through 30. Both species demonstrated a qualitative and quantitative similarity in performance, reinforcing the notion of a single, shared, nonverbal mechanism for numerical representation. This finding is supported by the systematic control of accuracy and latency by the ratio of the values compared, in line with Weber’s law, for both monkeys and humans).
The study also highlighted that while humans and monkeys share a similar mechanism for numerical comparison, the sensitivity to numerical differences, as estimated by the internal Weber fraction, indicated a slight difference in numerical acuity between the species. However, the authors noted that the performance difference might be attributable to a speed-accuracy trade-off, as monkeys responded significantly faster than humans in their tasks.
The fact that monkeys easily extended a rule learned with the values 1 through 9 to the values 10, 15, 20, and 30 suggests that monkeys do not treat values under and over 10 qualitatively differently. ... the strongest evidence to date in favor of a nonverbal system, shared by animals and humans, that allows both representation and comparison of numerosities and that is controlled by Weber’s law.
This study provides strong evidence for a shared, evolutionarily primitive system for representing and comparing numerical values between monkeys and humans. Their work contributes to our understanding of the cognitive foundations of numerical representation and suggests that the basic mechanisms of numerical cognition are evolutionarily conserved across species.
Comments sorted by top scores.