Skip to main content
SpringerLink
Log in

A robot that counts like a child: a developmental model of counting and pointing

  • Original Article
  • Published:
Psychological Research Aims and scope Submit manuscript

Abstract

In this paper, a novel neuro-robotics model capable of counting real items is introduced. The model allows us to investigate the interaction between embodiment and numerical cognition. This is composed of a deep neural network capable of image processing and sequential tasks performance, and a robotic platform providing the embodiment—the iCub humanoid robot. The network is trained using images from the robot’s cameras and proprioceptive signals from its joints. The trained model is able to count a set of items and at the same time points to them. We investigate the influence of pointing on the counting process and compare our results with those from studies with children. Several training approaches are presented in this paper, all of them use pre-training routine allowing the network to gain the ability of pointing and number recitation (from 1 to 10) prior to counting training. The impact of the counted set size and distance to the objects are investigated. The obtained results on counting performance show similarities with those from human studies.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8
Fig. 9
Fig. 10
Fig. 11
Fig. 12
Fig. 13

Similar content being viewed by others

Availability of data and material

The datasets generated during the current study are available from the corresponding author on reasonable request.

Code availability

The code created in the current study is available from the corresponding author on reasonable request.

Notes

  1. We also tested the model without that feature (when gestures are produced but the hand is not visible). That case is similar to what was presented by Pecyna and Cangelosi (2018). We found better performance with hand visible but these analyses are not in the scope of this paper.

  2. Having this information in the visual input is useful for the model because the Convolutional Part (image processing part) can learn to react differently whenever the image has to be processed or not. We did not include other trigger data in the visual input because of the pre-training method we initially (and as a subsidiary study) used, where only part of the network was trained and it was not necessarily containing the visual processing units.

  3. These seven angles composing the gesture output are: shoulder pitch, shoulder roll, shoulder yaw, elbow, wrist pronation/supination (roll), wrist pitch and wrist yaw.

  4. Also a few subsidiary tests were performed using the presented model and they showed that the gesture pre-training increases the final training speed significantly (where the recitation pre-training had positive influence only if used together with the gesture pre-training).

  5. At the beginning of our research, we were training the model to recite in a similar manner as in the case of gesture pre-training—only part of the network was trained. We found, however, that such a pre-trained network when trained to count, immediately loses the ability to recite (even if it is included in the final training set of simulated skills). This is likely because the model was never trained to produce any type of gesture output (in the case of recitation in the final training and current recitation pre-training the network is producing gestures—base position) and through backpropagation, gesture output error modifies the weights responsible for number recitation.

  6. Those values represent the situation when children were counting objects from a smaller set: 7-10 (as it is closer to the one we use). We considered only pointing conditions and not the touch one which is analysed later. The values were expressed as the percentage of correct answers

  7. We only considered the results of small sets from Alibali and DiRusso (1999) and we used the mean value and standard deviation presented in the article for the t test (2 conditions were compared: touch and point, conducted by 20 participants). Alibali and DiRusso (1999), however, found a statistical difference as they considered both ranges (small and large sets) together (having a larger number of samples). They also used a different test—repeated measures ANOVA.

  8. In the case of the experiment with children, the small sets were covering 7 to 10 objects and big ones 13–17.

References

  • Alibali, M. W., & DiRusso, A. A. (1999). The function of gesture in learning to count: More than keeping track. Cognitive Development, 14(1), 37–56.

    Article  Google Scholar 

  • Amit, D. J. (1988). Neural networks counting chimes. Proceedings of the National Academy of Sciences, 85(7), 2141–2145.

    Article  Google Scholar 

  • Andres, M., Di Luca, S., & Pesenti, M. (2008). Finger counting: The missing tool? Behavioral and Brain Sciences, 31(6), 642–643.

    Article  Google Scholar 

  • Bertenthal, B., & Von Hofsten, C. (1998). Eye, head and trunk control: The foundation for manual development1. Neuroscience and Biobehavioral Reviews, 22(4), 515–520.

    Article  PubMed  Google Scholar 

  • De La Cruz, V. M., Di Nuovo, A., Di Nuovo, S., & Cangelosi, A. (2014). Making fingers and words count in a cognitive robot. Frontiers in Behavioral Neuroscience, 8, 13.

    Google Scholar 

  • Dehaene, S. (1992). Varieties of numerical abilities. Cognition, 44(1–2), 1–42.

    PubMed  Google Scholar 

  • Dehaene, S., Dupoux, E., & Mehler, J. (1990). Is numerical comparison digital? Analogical and symbolic effects in two-digit number comparison. Journal of Experimental Psychology: Human Perception and Performance, 16(3), 626.

    PubMed  Google Scholar 

  • Di Nuovo, A. (2017). An embodied model for handwritten digits recognition in a cognitive robot. In 2017 IEEE Symposium Series on Computational Intelligence (SSCI) (pp. 1–6). IEEE.

  • Di Nuovo, A. (2018). Long-short term memory networks for modelling embodied mathematical cognition in robots. In 2018 International Joint Conference on Neural Networks (IJCNN) (pp. 1–7). IEEE.

  • Di Nuovo, A. (2020). A developmental neuro-robotics approach for boosting the recognition of handwritten digits. In 2020 International Joint Conference on Neural Networks (IJCNN) (pp. 1–8). IEEE.

  • Di Nuovo, A., De La Cruz, V. M., & Cangelosi, A. (2014a). Grounding fingers, words and numbers in a cognitive developmental robot. In 2014 IEEE Symposium on Computational Intelligence, Cognitive Algorithms, Mind, and Brain (CCMB) (pp. 9–15). IEEE.

  • Di Nuovo, A., De La Cruz, V. M., & Cangelosi, A. (2015). A deep learning neural network for number cognition: A bi-cultural study with the iCub. In 2015 Joint IEEE International Conference on Development and Learning and Epigenetic Robotics (ICDL-EpiRob) (pp. 320–325). IEEE.

  • Di Nuovo, A., De La Cruz, V. M., Cangelosi, A., & Di Nuovo, S. (2014b). The iCub learns numbers: An embodied cognition study. In 2014 International Joint Conference on Neural Networks (IJCNN) (pp. 692–699). IEEE.

  • Di Nuovo, A., & Jay, T. (2019). The development of numerical cognition in children and artificial systems: A review of the current knowledge and proposals for multi-disciplinary research. Cognitive Computation and Systems.

  • Di Nuovo, A., & McClelland, J. L. (2019). Developing the knowledge of number digits in a child-like robot. Nature Machine Intelligence, 1(12), 594–605.

    Article  Google Scholar 

  • Elman, J. L. (1990). Finding structure in time. Cognitive Science, 14(2), 179–211.

    Article  Google Scholar 

  • Fischer, M. H., Kaufmann, L., & Domahs, F. (2012). Finger counting and numerical cognition. In F. Domahs, L. Kaufmann, & M. H. Fischer (Eds.), Handy numbers: Finger counting and numerical cognition (p. 6). Frontiers E-books.

  • Fuson, K. C. (1988). Children’s counting and concepts of number. New York: Springer.

    Book  Google Scholar 

  • Fuson, K. C., Richards, J., & Briars, D. J. (1982). The acquisition and elaboration of the number word sequence. In C. J. Brainerd (Ed.), Children’s logical and mathematical cognition (pp. 33–92). New York: Springer.

    Chapter  Google Scholar 

  • Gallistel, C. R., & Gelman, R. (1992). Preverbal and verbal counting and computation. Cognition, 44(1–2), 43–74.

    Article  PubMed  Google Scholar 

  • Gallistel, C. R., & Gelman, R. (2000). Non-verbal numerical cognition: From reals to integers. Trends in Cognitive Sciences, 4(2), 59–65.

    Article  PubMed  Google Scholar 

  • Gelman, R. (1980). What young children know about numbers. Educational Psychologist, 15(1), 54–68.

    Article  Google Scholar 

  • Gelman, R., & Gallistel, C. R. (1978). The child’s understanding of number. Cambridge: Harvard University Press.

    Google Scholar 

  • Gigliotta, O., Ponticorvo, M., Doricchi, F., & Miglino, O. (2019). Midpoint: A tool to build artificial models of numerical cognition. In International Work-Conference on the Interplay Between Natural and Artificial Computation (pp. 88–96). Springer.

  • Goldin-Meadow, S., Levin, S. C., & Jacobs, S. (2014). Gestures role in learning arithmetic. In L. D. Edwards, D. Moore-Russo, & F. Ferrara (Eds.), Emerging perspectives on gesture and embodiment in mathematics (pp. 51–72). Charlotte: IAP.

    Google Scholar 

  • Graham, T. A. (1999). The role of gesture in children’s learning to count. Journal of Experimental Child Psychology, 74(4), 333–355.

    Article  PubMed  Google Scholar 

  • Hochreiter, S., & Schmidhuber, J. (1997). Long short-term memory. Neural Computation, 9(8), 1735–1780.

    Article  PubMed  Google Scholar 

  • Le Corre, M., & Carey, S. (2007). One, two, three, four, nothing more: An investigation of the conceptual sources of the verbal counting principles. Cognition, 105(2), 395–438.

    Article  PubMed  Google Scholar 

  • Le Corre, M., Van de Walle, G., Brannon, E. M., & Carey, S. (2006). Re-visiting the competence/performance debate in the acquisition of the counting principles. Cognitive Psychology, 52(2), 130–169.

    Article  PubMed  Google Scholar 

  • LeCun, Y., Bengio, Y., & Hinton, G. (2015). Deep learning. Nature, 521(7553), 436–444.

    Article  PubMed  Google Scholar 

  • Metta, G., Natale, L., Nori, F., Sandini, G., Vernon, D., Fadiga, L., et al. (2010). The iCub humanoid robot: An open-systems platform for research in cognitive development. Neural Networks, 23(8), 1125–1134.

    Article  PubMed  Google Scholar 

  • Pecyna, L., & Cangelosi, A. (2018). Influence of pointing on learning to count: A neuro-robotics model. In 2018 IEEE Symposium Series on Computational Intelligence (SSCI) (pp. 358–365). IEEE.

  • Pecyna, L., Cangelosi, A., & Di Nuovo, A. (2019). A deep neural network for finger counting and numerosity estimation. In 2019 IEEE Symposium Series on Computational Intelligence (SSCI) (pp. 358–365). IEEE.

  • Rodriguez, P., Wiles, J., & Elman, J. L. (1999). A recurrent neural network that learns to count. Connection Science, 11(1), 5–40.

    Article  Google Scholar 

  • Rotondaro, F., Ponticorvo, M., Gigliotta, O., Pinto, M., Pellegrino, M., Gazzellini, S., et al. (2019). The Number Interval Position Effect (NIPE) in the mental bisection of numerical intervals might reflect the influence of the decimal-number system on the gaussian representations of numerosities: A combined developmental and computational-modeling study. Cortex, 114, 164–175.

    Article  PubMed  Google Scholar 

  • Ruciński, M., Cangelosi, A., & Belpaeme, T. (2012). Robotic model of the contribution of gesture to learning to count. In 2012 IEEE International Conference on Development and Learning and Epigenetic Robotics (ICDL) (pp. 1–6). IEEE.

  • Saxe, G. B., & Kaplan, R. (1981). Gesture in early counting: A developmental analysis. Perceptual and Motor skills, 53(3), 851–854.

    Article  Google Scholar 

  • Schaeffer, B., Eggleston, V. H., & Scott, J. L. (1974). Number development in young children. Cognitive Psychology, 6(3), 357–379.

    Article  Google Scholar 

  • Scherer, D., Müller, A., & Behnke, S. (2010). Evaluation of pooling operations in convolutional architectures for object recognition. In International conference on artificial neural networks (pp. 92–101). Springer.

  • Testolin, A. (2020). The challenge of modeling the acquisition of mathematical concepts. Frontiers in Human Neuroscience, 14, 100.

    Article  PubMed  PubMed Central  Google Scholar 

  • Wynn, K. (1990). Children’s understanding of counting. Cognition, 36(2), 155–193.

    Article  PubMed  Google Scholar 

  • Wynn, K. (1992). Children’s acquisition of the number words and the counting system. Cognitive Psychology, 24(2), 220–251.

    Article  Google Scholar 

Download references

Funding

The authors acknowledge the UK EPSRC support through the project grant EP/P030033/1 (NUMBERS) and the EU support via the H2020 Marie Skłodowska-Curie ITN APRIL project (Grant Agreement No 674868).

Author information

Authors and Affiliations

  1. Centre for Robotics and Neural Systems, University of Plymouth, Plymouth, UK

    Leszek Pecyna

  2. University of Manchester, Manchester, UK

    Angelo Cangelosi

  3. AIST-AIRC, Tokyo, Japan

    Angelo Cangelosi

  4. Sheffield Robotics, Sheffield Hallam University, Sheffield, UK

    Alessandro Di Nuovo

Authors
  1. Leszek Pecyna
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Angelo Cangelosi
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Alessandro Di Nuovo
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Leszek Pecyna.

Ethics declarations

Conflict of interest

The authors declare that they have no conflict of interest.

Ethical approval

This article does not contain any studies with human participants or animals performed by any of the authors.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Pecyna, L., Cangelosi, A. & Di Nuovo, A. A robot that counts like a child: a developmental model of counting and pointing. Psychological Research 86, 2495–2511 (2022). https://doi.org/10.1007/s00426-020-01428-8

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s00426-020-01428-8

Search

Navigation