The Brain & AI

New Biography of Mike Gaze

Michael Gaze (1927-2012) was a contemporary of Roger Sperry who went in search of the rules of brain wiring. Inspired by Sperry, Gaze tested the core ideas of the field using electrophysiology to map connectivity – and described just how relative positioning, as opposed to Sperry’s rigid code, can lead to synaptic specificity in the visual system. Never as famous as Sperry, Gaze was very careful and ultimately correct about the rules that lead to brain wiring. As a former Fellow of the Royal Society, a new biograhy has now been published by Royal Society Publishing, authored by his former graduate student Tony Hope.

Excerpt from the biography:

Mike has contributed more than any other scientist to establishing the role of plasticity in the formation of nerve connections. Schmidt (2020), in his masterly account of the current state of the field, cites Mike’s papers more times than those of any other scientist. Mike trained many young scientists who have gone on to have excellent careers. He influenced these scientists through collaboration and example. The approach to science that he exemplified combined intellectual and moral qualities. He carried out his experiments with great care. He followed the evidence, inferring only what logically followed from his results and accepting the conclusions, come what may. These conclusions sometimes led to controversy. He did not avoid controversy, but he did try to prevent controversy from becoming personal: ‘Let’s see what the data say’ was his approach.

In 2021 (Peter) Robin Hiesinger published The self-assembling brain, in which he recounts some of the history and context of Sperry’s and Mike’s work (Hiesinger 2021). This book examines how the genetically encoded development of neuronal connections requires the type of plasticity investigated by Mike. As Hiesinger notes, developmental plasticity represents an extension of learning into development that is missing in current artificial neural network design; yet, it is exactly the developmental growth process that unfolds the specificity, robustness and sheer beauty of the brain connectivity so much admired by Mike. ‘From the perspective of an information theorist, the distinction [between strict chemoaffinity and a systems matching approach] can barely be overstated’ (Hiesinger 2021, p. 196).

The recent molecular studies have shown that the first two elements of Sperry’s theory are correct: there are orthogonal chemical gradients across both retina and tectum relevant to the generation of the retinotectal map. The third aspect of Sperry’s theory, the mapping rule, however, is wrong. The chemical gradients influence, but do not rigidly determine, termination sites. Mike’s insistence that the mapping rules ensure relative but not absolute positions—systems matching—have been strongly endorsed by molecular studies (Schmidt 2020, p. 419; Hiesinger 2021, p. 192), and Mike was correct that the connections between retina and tectum are ‘sliding’ (continually changing) during normal development in non-mammalian vertebrates (Schmidt 2020, pp. 429–430).

The surge in molecular studies that occurred in the mid 1990s used dye tracing in post-mortem mouse material. These were limited to studying small numbers of nerve fibres and only at a specific stage of development (Schmidt 2020, ch. 8). This meant that the physiological approach, backed up by anatomical studies, as championed by Mike, was lost for a decade or more. When combinations of mutations were generated, affecting molecules involved in both rostro-caudal tectal polarity and early post-natal activity, that physiological analysis was re-established. This enabled an understanding of how the entire set of connections functions, and also how the connections change over time (Schmidt 2020, p. 329).