Abstract
For millions of years, biological creatures have dealt with the world without being able to see it; however, the change in the atmospheric condition during the Cambrian period and the subsequent increase of light, triggered the sudden evolution of vision and the consequent evolutionary benefits. Nevertheless, how from simple organisms to more complex animals have been able to generate meaning from the light who fell in their eyes and successfully engage the visual world remains unknown. As shown by many psychophysical experiments, biological visual systems cannot measure the physical properties of the world. The light projected onto the retina is, in fact, unable to specify the physical properties of the world in which humans and other visually ‘intelligent’ animals behave; however, visual behaviours are habitually successful. Through psychophysical evidence, examples of the functioning of Artificial Neural Networks (ANNs) and a reflection upon visual appreciation in the cultural and artistic context, this paper shows (a) how vision emerged by random trial and error during evolution and lifetime learning; (b) how the functioning of ANNs may provide evidence and insights on how machine and human vision works; and (c) how rethinking vision theory in terms of trial and error may offer a new approach to better understand vision—biological and artificial—and reveal new insights into why we like what we like.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Explore related subjects
Discover the latest articles, news and stories from top researchers in related subjects.Notes
A neuron is a specialised cell for the transmission of electro-chemical signals.
Glia cells are non-neural cells in the nervous system. Their main role is to provide structural support to neurons. They are not directly involved in the transmission of signals.
Synapses are connections between neurons and a target cell. The role of synapses is to allow communication by receiving or transmitting chemical signals.
In biology an event, in the form of energy, that provokes and activates a receptor cell.
Biological creatures gifted with photoreceptors able to transmit information about the outside world.
Images should be intended both as the light pattern detected by the retina, as well as the visual result of the processed light-pattern, meaning its visual representation—e.g., the image of a cat.
It is important to highlight that a light pattern does not resemble the physical world, nor any ‘image’ of the world produced by the human brain. A light pattern should be intended as the light configuration that is perceived, determined by its frequency of occurrence and its consequent usefulness (for more, see Purves et al. 2015., Yang and Purves 2004 , Rao et al. 2002).
On 11 of May 1997, after a 6-match game, Deep Blue beat the world-best chess player—Garry Kasparov. The Deep Blue’s win had an important symbolic significance in the advancement of ‘intelligent’ machine–artificial intelligence.
Reinforcement learning, in machine learning’s context, refers to the ability of an agent to learn by trial and error without supervision. Reinforcement learning, unlike supervised learning, does not require any labelled data. In the context of machine vision, a labelled data, for instance, might be a photograph in which the objects represented are specified—e.g., a cat, an apple or a house. Labels are usually obtained by asking annotators—humans—to make judgments about the content of a given image.
Visual appreciation refers not only to the aesthetic appreciation of a work of art but also to the ability to analyse, describe, interpret and make connections between works of art and their cultural context.
References
Baxandall M (1988) Painting and social experience in 15th century Italy. Oxford University Press, Oxford
Berger J (2008) Ways of seeing. Penguin Classics, London
Creanza N, Kolodny O, Feldman MW (2017) Cultural evolutionary theory: how culture evolves and why it matters. Proc Natl Acad Sci 114(30):7782–7789. https://doi.org/10.1073/pnas.1620732114
Churchland PS (1989) Neurophilosophy toward a unified science of the mind-brain. A Bradford Book, Cambridge (MA)
Downing KL (2015) Intelligence emerging. Adaptivity and searching in evolving neural systems. Intelligence emerging. MIT Press, Cambridge
Greenwood A, Bartusiak MF (1992) Neural networks: computational neuroscience: a window to understanding science at the frontier takes you on a journey how the brain works?. The National Academies Press, Washington, Science at the Frontier
Hassabis D, Hubert T, Schrittwieser J, Silver D, et al (2018) AlphaZero: shedding new light on chess, shogi, and Go. DeepMind Blog. https://deepmind.com/blog/article/alphazero-shedding-new-light-grand-games-chess-shogi-andgo?fbclid=IwAR0j8BhGOwaLBPRu0opVBEX0g4TgztWwTxA4_J3ozYsBeymnG5_QQWoqMXg. Accessed 28 January 2020
Hebb DO (2002) The organization of behavior: a neuropsychological theory. Psychology Press, London
Isbell LA (2009) The Fruit, the tree, and the serpent: why we see so well. Harvard University Press, Cambridge
Kanizsa G (1997) Grammatica del vedere Saggi su percezione e Gestalt. Il Mulino, Bologna
Lotto B (2017) Deviate: the science of seeing differently. Hachette Books, New York
McCulloch WS, Pitts WH (1943) A logical calculus of the ideas immanent in nervous activity. Bull Math Biophys 5:115–133
Parker A (2004) In the blink of an eye. Basic Books, New York
Pinker S (2013) Stylish Academic writing. https://www.youtube.com/watch?v=IE-TTz13P7w. Accesed 18 January 2020
Purves D, Monson BB, Sundararajan J, Wojtach WT (2014) How biological vision succeeds in the physical world. Proc Natl Acad Sci USA 111(13):4750–4755
Purves D, Morgenstern Y, Wojtach WT (2015) Perception and reality: why a wholly empirical paradigm is needed to understand vision. Front Syst Neurosci 9(November):1–10. https://doi.org/10.3389/fnsys.2015.00156
Purves D, Lotto B (2003) Why we see what we do: an empirical theory of vision. Sinauer Associates, Sunderland MA
Purves D (2019) Brains as engines of association: an operating principle for nervous systems. Oxford University Press, Oxford
Rao RPN, Olshausen BA, Lewicki MS (2002) Probabilistic models of the brain: perception and neural function. MIT Press, Cambridge MA
Robson D (2020) A brief history of the brain. NewScientist. https://www.newscientist.com/article/mg21128311-800-a-brief-history-of-the-brain/. Accessed 26 December 2019
Senatore A, Reese TS, Smith CL (2017) Neuropeptidergic integration of behavior in trichoplax adhaerens, an animal without synapses. J Exp Biol 220(18):3381–3390. https://doi.org/10.1242/jeb.162396
Silver D, Hubert T et al (2018) A general reinforcement learning algorithm that masters chess, shogi, and Go through self-play. Science 362(6419):1140–1144. https://doi.org/10.1126/science.aar6404
Treccani C (2018) How machines see the world: understanding image annotation. NECSUS. Eur J Media Stud 7(1): 235–254. https://doi.org/10.25969/mediarep/3425
Van Le Q, Isbell, et al (2013) Pulvinar neurons reveal neurobiological evidence of past selection for rapid detection of snakes. Proc Natl Acad Sci 110(47):19000–19005. https://doi.org/10.1073/pnas.1312648110
Yang J, Purves D (2004) The statistical structure of natural light patterns determines perceived light intensity. Proc Natl Acad Sci 101(23):8745–8750
Zarkadakis G (2015) In Our Own Image. Rider Books, London
Acknowledgements
Many thanks to Olli Tapio Leino for his continuous support, to Patricia Armati for her comments and to Dale Purves for his guidance in writing this paper.
Funding
The writing of this paper was possible thanks to the financial support of the School of Creative Media-City University of Hong Kong and the University Grants Committee, Hong Kong, SAR, China.
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Treccani, C. The brain, the artificial neural network and the snake: why we see what we see. AI & Soc 36, 1167–1175 (2021). https://doi.org/10.1007/s00146-020-01065-0
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00146-020-01065-0