An emergentist perspective on the origin of number sense

  title={An emergentist perspective on the origin of number sense},
  author={Marco Zorzi and Alberto Testolin},
  journal={Philosophical Transactions of the Royal Society B: Biological Sciences},
  • M. Zorzi, Alberto Testolin
  • Published 1 January 2018
  • Biology
  • Philosophical Transactions of the Royal Society B: Biological Sciences
The finding that human infants and many other animal species are sensitive to numerical quantity has been widely interpreted as evidence for evolved, biologically determined numerical capacities across unrelated species, thereby supporting a ‘nativist’ stance on the origin of number sense. Here, we tackle this issue within the ‘emergentist’ perspective provided by artificial neural network models, and we build on computer simulations to discuss two different approaches to think about the… 

Figures from this paper

Numerosity discrimination in deep neural networks: Initial competence, developmental refinement and experience statistics.

The findings suggest that it may not be necessary to assume that animals are endowed with a dedicated system for processing numerosity, since domain-general learning mechanisms can capture key characteristics others have attributed to an evolutionarily specialized number system.

Visual sense of number vs. sense of magnitude in humans and machines

Deep neural networks are tested on the same numerosity comparison task that was administered to human participants, using a stimulus space that allows the precise measurement of the contribution of non-numerical features, suggesting that numerosity is a major, salient property of the authors' visual environment.

Numerosity representation in a deep convolutional neural network

This work examined whether the numerosity underestimation effect, a phenomenon indicating that numerosity perception acts upon the perceptual number rather than the physical number, can be observed in deep convolutional neural networks (DCNNs), and found that number-selective units at late layers operated on the perceived number, like humans do.

Visual number sense in untrained deep neural networks

It is shown that number-selective neurons can arise spontaneously in untrained deep neural networks and that the responses of these neurons can induce the abstract number sense, the ability to discriminate numerosity independent of low-level visual cues, which provides insight into the origin of innate cognitive functions.

Perception of visual numerosity in humans and machines

This model accurately simulated the psychophysics of numerosity perception and the associated developmental changes: discrimination was driven by numerosity information, but non-numerical features had a significant impact, especially early during development.

Numerosities and Other Magnitudes in the Brains: A Comparative View

Evidence for magnitude representation, both discrete (countable) and continuous, following the sensory input path from primary sensory systems to associative pallial territories in the vertebrate brains is reviewed.

Numerosity Representation in InfoGAN: An Empirical Study

This work trains in unsupervised and semi-supervised fashion a latent-space generative model that has been shown capable of disentangling relevant semantic features in a variety of complex datasets, and test its generative performance under different conditions.

Non-numerical strategies used by bees to solve numerical cognition tasks

We examined how bees solve a visual discrimination task with stimuli commonly used in numerical cognition studies. Bees performed well on the task, but additional tests showed that they had learned

Modelling the development of counting with memory-augmented neural networks

This work found that a memory-augmented modular network architecture based on the recently proposed Emergent Symbol Binding Network (ESBN) exhibited an inflection during learning that resembled human development and could shed light on the emergence of systematicity in humans.

Representational and Embodied Accounts of Abstract Concepts: A Framework for Prelinguistic Acquisition of Numerical Cognition

It is proposed that a renewed theoretical framework that extends beyond the representational-embodied dichotomy is needed, if a detailed understanding of numerical cognition is desired.



Development of Elementary Numerical Abilities: A Neuronal Model

The computer simulations account for several phenomena in the numerical domain, including the distance effect and Fechner's law for numbers, and demonstrate that infants' numerosity detection abilities may be explained without assuming that infants can count.

Is There Really an Evolved Capacity for Number?

Representation of Number in Animals and Humans: A Neural Model

This article addresses the representation of numerical information conveyed by nonsymbolic and symbolic stimuli and presents a concrete proposal on the linkage between higher order numerical cognition and more primitive numerical abilities and generates specific predictions on the neural substrate of number processing.

Ontogeny of Numerical Abilities in Fish

The results suggest the ability of guppies to discriminate small numbers is innate and is displayed immediately at birth while discrimination of large numbers emerges later as a result of both maturation and social experience.

Individual differences in non-verbal number acuity correlate with maths achievement

There are large individual differences in the non-verbal approximation abilities of 14-year-old children, and that these individual Differences in the present correlate with children’s past scores on standardized maths achievement tests, extending all the way back to kindergarten.

Monotonic Coding of Numerosity in Macaque Lateral Intraparietal Area

It is shown, for the first time, that a population of neurons in the lateral intraparietal area of monkeys encodes the total number of elements within their classical receptive fields in a graded fashion, across a wide range of numerical values.

Core systems of number

From basic network principles to neural architecture: emergence of spatial-opponent cells.

  • R. Linsker
  • Biology
    Proceedings of the National Academy of Sciences of the United States of America
  • 1986
This paper is the first of three that address the origin and organization of feature-analyzing cells in simple systems governed by biologically plausible development rules, and introduces the theory of "modular self-adaptive networks," of which this system is an example, and explicitly demonstrates the emergence of a layer of spatial-opponent cells.