Number-Space Association in Macaques

This study demonstrates that while rhesus and crab-eating macaques do not exhibit a systematic spatial bias based on absolute numerical magnitude, they do display directional number-space associations (leftward for decreases, rightward for increases) when responding to relative numerical changes and expectancy violations in comparative contexts.

The Gist

The Big Question: Do Monkeys Have a "Mental Number Line"?

Imagine you are asked to think about the number 2 and the number 8. If you are like most humans, you probably picture the 2 on the left side of your mind and the 8 on the right. We humans have a "Mental Number Line" (MNL) where small numbers live on the left and big numbers live on the right.

Scientists have debated for a long time: Is this just something we learn from reading books (left-to-right), or is it something we are born with?

To find out, researchers studied monkeys. They wanted to see if monkeys naturally link small numbers to the left side and big numbers to the right side, or if they need to be taught.

The Experiment: A Monkey Buffet

The researchers set up two experiments using food (raisins) because monkeys love food, and it's a natural way to test their brains without them feeling like they are in a classroom.

Experiment 1: The "Equal Choice" Test

The Setup: Imagine a monkey sitting in front of a table. On the left side, there is a tray with 4 raisins. On the right side, there is another tray with also 4 raisins. The monkey can choose either side to get a snack.
The Goal: The researchers did this with different numbers (1 raisin vs. 1, 8 vs. 8, 24 vs. 24). They wanted to see if the monkey would naturally reach for the "small" number with its left hand and the "big" number with its right hand, even though the amounts were the same on both sides.

The Result: Nope.
The monkeys didn't care about the left or right side based on the number. If there were 24 raisins on both sides, they picked randomly. They didn't show a "left for small, right for big" pattern.
However, there was a tiny clue: When the numbers got very big, the monkeys were slightly more likely to use their right hand to grab the food. It was like their muscles got a little more "right-brained" when the task felt bigger, even if they didn't move their body to the right.

Experiment 2: The "Surprise Party" Test

The Setup: This time, the researchers changed the rules. They used a "habituation" trick.

1. The Routine: They showed the monkey a tray with 16 raisins (a big amount) over and over again. The monkey got used to this big number.
2. The Surprise: Suddenly, they swapped the tray for one with only 4 raisins (a small amount).
3. The Reverse: They did the opposite with other monkeys: get used to 4 raisins, then surprise them with 16.

The Result: Bingo!
When the monkey was used to a big number (16) and suddenly saw a small number (4), they tended to reach to the left.
When the monkey was used to a small number (4) and suddenly saw a big number (16), they tended to reach to the right.

The Analogy: The Elevator and the Floor

Think of the monkey's brain like an elevator.

• Experiment 1 was like standing in the elevator with the doors closed. The monkey didn't know which floor it was on, so it just stood still. It didn't know if it needed to go "Left" (down) or "Right" (up).
• Experiment 2 was like the elevator moving.
  • If the elevator was on the 16th floor (Big) and suddenly dropped to the 4th floor (Small), the monkey felt a "drop" and instinctively leaned Left.
  • If the elevator was on the 4th floor and suddenly shot up to the 16th floor, the monkey felt a "lift" and instinctively leaned Right.

What Does This Mean?

The study tells us three important things:

1. It's About Change, Not Just Numbers: Monkeys don't have a fixed map where "4 is always Left." Instead, their brains are like a balance scale. They only show a left-or-right bias when something changes compared to what they just saw.
2. The Body Knows Before the Mind: Even when the monkeys didn't move their whole body to the left or right, their hands showed the pattern. When the number got bigger, they used their right hand more. When it got smaller, they used their left. It's as if their muscles knew the direction before their eyes did.
3. It's Biological, Not Just Cultural: Since monkeys don't read books or write from left to right, this suggests that linking numbers to space is a natural, biological instinct. It's built into our brains (and monkey brains) to help us understand the world, long before we learn to read.

The Bottom Line

Monkeys don't have a permanent "number line" in their heads like a ruler. Instead, they have a dynamic radar. When a number changes—getting bigger or smaller—their brains instantly map that change to space: Bigger goes Right, Smaller goes Left. It's a survival skill that helps them quickly judge if they are getting more food or less, and their bodies react to that change automatically.

Technical Summary

1. Problem Statement

Humans organize numerical magnitudes along a spatial continuum known as the Mental Number Line (MNL), typically mapping smaller numbers to the left and larger numbers to the right (the SNARC effect). While cultural factors (e.g., reading direction) influence this mapping, evidence from infants and non-human animals suggests a biological basis. However, findings in non-human primates regarding a directional MNL have been inconsistent and often contradictory. Previous studies in macaques have suffered from small sample sizes, methodological heterogeneity, and a reliance on trained tasks or artificial stimuli (e.g., dots on screens) rather than ecologically relevant stimuli. The authors aimed to resolve these inconsistencies by investigating whether macaques spontaneously map numerical magnitudes onto space using food-based tasks, distinguishing between absolute magnitude processing and relative magnitude (comparative) processing.

2. Methodology

The study utilized two experiments involving 17 macaques (7 Macaca mulatta and 10 Macaca fascicularis) housed at the Institut des Sciences Cognitives "Marc Jeannerod" in Lyon, France. The experiments employed spontaneous, food-related tasks to minimize top-down cognitive control and maximize ecological validity.

Experiment 1: Absolute Magnitude and Spontaneous Choice

• Design: An equal-comparison task where monkeys chose between two identical food quantities presented simultaneously on the left and right.
• Stimuli: Six pairwise combinations of food items (raisins): 1×1, 2×2, 4×4, 8×8, 16×16, and 24×24.
• Procedure: Monkeys were presented with the trays without prior habituation to a reference number. They were required to gaze at both containers and then reach for one item.
• Variables Measured:
  • Spatial Choice: Left vs. Right selection.
  • Hand Preference: Left vs. Right hand used for grasping.
• Analysis: Generalized Linear Mixed-Effects Models (GLMMs) were used to analyze choice and hand preference as a function of numerical pair identity, numerical magnitude (continuous log-transformed), and categorical grouping (Small: 1–8 vs. Large: 16–24).

Experiment 2: Relative Magnitude and Habituation-Dishabituation

• Design: A habituation-dishabituation paradigm to test sensitivity to changes in numerical magnitude relative to a baseline.
• Stimuli: Two specific quantities: Small (4×4) and Large (16×16).
• Procedure:
  • Habituation Phase: Subjects were repeatedly exposed to a reference quantity (either 4×4 or 16×16) for 10 trials.
  • Dishabituation Phase: Subjects were presented with a "novel" trial containing the alternative quantity (e.g., if habituated to 16, they saw 4).
  • Sequences: Each subject completed two sequences (Habituation Large $\rightarrow$ Novel Small; Habituation Small $\rightarrow$ Novel Large).
• Variables Measured: Spatial choice and hand preference during the novel (dishabituation) trials compared to habituation trials.
• Analysis: GLMMs with fixed effects for Pair (4 vs. 16), Trial Type (Habituation vs. Dishabituation), and their interaction.

3. Key Results

Experiment 1: No Absolute Spatial Bias

• Spatial Choice: No systematic left-to-right (or right-to-left) bias was found based on absolute numerical magnitude. Choice behavior was significantly influenced by the monkey's physical position relative to the apparatus but not by the quantity of food (1–24 items).
• Hand Preference: While spatial choice remained unbiased, hand use showed a magnitude-dependent modulation.
  • As numerical magnitude increased (logarithmically), the probability of using the right hand increased significantly.
  • Monkeys were approximately twice as likely to use their right hand for large quantities (16–24) compared to small quantities (1–8).
• Conclusion: Numerical magnitude modulates motor response systems (hand use) but does not drive allocentric spatial choice in isolation.

Experiment 2: Directional Bias via Relative Change

• Spatial Choice: A clear directional bias emerged only during dishabituation (expectancy violation).
  • Numerical Increase (4 $\rightarrow$ 16): Monkeys showed a significantly higher probability of choosing the right side.
  • Numerical Decrease (16 $\rightarrow$ 4): Monkeys showed a significantly lower probability of choosing the right side (i.e., a bias toward the left).
• Hand Preference: A similar direction-dependent modulation was observed in hand use.
  • Increases in quantity elicited increased right-hand use.
  • Decreases in quantity elicited reduced right-hand use.
• Conclusion: Spatial-numerical associations in macaques are not fixed mappings of absolute numbers but are dynamically generated by relative numerical changes (expectancy violations).

4. Key Contributions

1. Resolution of Inconsistencies: The study clarifies why previous macaque studies yielded mixed results, suggesting that a rigid MNL may not exist for absolute quantities in this species, but rather a comparative mechanism.
2. Dissociation of Motor vs. Spatial Processing: The authors demonstrate a dissociation where numerical magnitude influences lateralized motor output (hand use) even when it does not influence allocentric spatial choice (left/right selection).
3. Ecological Validity: By using real food stimuli and spontaneous behaviors rather than trained touchscreen tasks, the study provides evidence for more naturalistic numerical processing.
4. Relative vs. Absolute Processing: The findings support the hypothesis that number-space associations in non-human primates are primarily driven by relative magnitude comparisons and expectancy violations, similar to findings in human newborns and chicks, rather than static absolute mappings.

5. Significance

• Evolutionary Basis: The results support the existence of an evolutionarily conserved, biologically grounded mechanism for number-space associations that predates cultural influences like reading direction.
• Cognitive Mechanism: The study suggests that the "Mental Number Line" in macaques is not a static internal map but a dynamic process activated during comparative evaluation and prediction error (when a stimulus violates an established expectation).
• Methodological Implication: Future research on non-human numerical cognition should prioritize comparative paradigms and ecologically relevant stimuli over absolute quantity tasks to detect spatial-numerical biases.
• Neuroscientific Relevance: The findings align with neurophysiological data showing overlapping neural substrates for numerical and spatial processing in the intraparietal sulcus (IPS), suggesting that these associations may be mediated by attentional and motor planning mechanisms triggered by magnitude deviations.

In summary, the paper concludes that macaques possess a left-to-right number-space association, but it is context-dependent, emerging robustly only when numerical magnitude is evaluated relative to a reference point and involves a violation of expectation.