The problem with prediction


Article by Joseph Fridman: “…At precisely the same moment in which the idea of predictive control has risen to dominance within the corporate sphere, it’s also gained a remarkable following within cognitive science. According to an increasingly influential school of neuroscientists, who orient themselves around the idea of the ‘predictive brain’, the essential activity of our most important organ is to produce a constant stream of predictions: predictions about the noises we’ll hear, the sensations we’ll feel, the objects we’ll perceive, the actions we’ll perform and the consequences that will follow. Taken together, these expectations weave the tapestry of our reality – in other words, our guesses about what we’ll see in the world become the world we see. Almost 400 years ago, with the dictum ‘I think, therefore I am,’ René Descartes claimed that cognition was the foundation of the human condition. Today, prediction has taken its place. As the cognitive scientist Anil Seth put it: ‘I predict (myself) therefore I am.’

Somehow, the logic we find animating our bodies is the same one transforming our body politic. The prediction engine – the conceptual tool used by today’s leading brain scientists to understand the deepest essence of our humanity – is also the one wielded by today’s most powerful corporations and governments. How did this happen and what does it mean?

One explanation for this odd convergence emerges from a wider historical tendency: humans have often understood the nervous system in terms of the flourishing technologies of their era, as the scientist and historian Matthew Cobb explained in The Idea of the Brain (2020). Thomas Hobbes, in his book Leviathan (1651), likened human bodies to ‘automata’, ‘[e]ngines that move themselves by springs and wheeles as doth a watch’. What is the heart, Hobbes asked, if not ‘a Spring; and the Nerves, but so many Strings …?’ Similarly, Descartes described animal spirits moving through the nerves according to the same physical properties that animated the hydraulic machines he witnessed on display in the French royal gardens.

The rise of electronic communications systems accelerated this trend. In the middle of the 19th century, the surgeon and chemist Alfred Smee said the brain was made up of batteries and photovoltaic circuits, allowing the nervous system to conduct ‘electro-telegraphic communication’ with the body. Towards the turn of the 20th century, the neuroscientist Santiago Ramón y Cajal described the positioning of different neural structures ‘somewhat as a telegraph pole supports the conducting wire’. And, during the First World War, the British Royal Institution Christmas lectures featured the anatomist and anthropologist Arthur Keith, who compared brain cells to operators in a telephonic exchange.

The technologies that have come to dominate many of our lives today are not primarily hydraulic or photovoltaic, or even telephonic or electro-telegraphic. They’re not even computational in any simplistic sense. They are predictive, and their infrastructures construct and constrain behaviour in all spheres of life. The old layers remain – electrical wiring innervates homes and workplaces, and water flows into sinks and showers through plumbing hidden from view. But these infrastructures are now governed by predictive technologies, and they don’t just guide the delivery of materials, but of information. Predictive models construct the feeds we scroll; they autocomplete our texts and emails, prompt us to leave for work on time, and pick out the playlists we listen to on the commute that they’ve plotted out for us. Consequential decisions in law enforcement, military and financial contexts are increasingly influenced by automated assessments spat out by proprietary predictive engines….(More)”.