janet grau

Learning Machines

Learning Machines (Everything We’ll Ever Think is Potentially There from the Start)

Lecture performance with video, 2022

Drawing on recent findings in philosophy, cognitive science and developmental psychology, the lecture performance Learning Machines (Everything We’ll Ever Think is Potentially There from the Start) explores the ways in which knowledge, imagination and play are intertwined. For example, ‘pretending’ involves more than just the power of imagination – it is a cultural process that can actually implement new realities.

For this performance, I created scenes for the video camera with children that reference well-known research experiments. These scenes on the screen are a visual parallel to what is being discussed in the lecture.

The Reverend Thomas Bayes, an 18th-century philosopher, mathematician, and theologian, developed models of human learning that are increasingly influential in developmental psychology and machine learning (artificial intelligence). 

"At least since Noam Chomsky, there have been debates about whether we have innate knowledge. The Bayesian picture characterizes knowledge in terms of a set of potential hypotheses about the world. We initially believe that some hypotheses are less probable and other are more so. As we collect new evidence, we can rationally update the probability of these hypotheses. We can discard what initially looked likely and eventually accept ideas that started out as longshots. If this picture is right (and I think it is), there’s some sense in which everything we’ll ever think is potentially there from the start."

The quotes in this performance are from the writings and research of Alison Gopnik (professor for developmental psychology and philosophy at UC Berkeley) and are used with the author’s permission.

The scenes in the video quote Albert Bandura’s Bobo Doll experiments, which he conducted to test his theory of social learning. Many thanks to Mohamad Alraghban, as well as N.B., P.S., and S.L. for helping me reimagine Bandura’s experiments for the camera, and to Christian Buck for allowing us to use his studio.

Photos: Irina Yakimova
Video stills: Janet Grau

Learning Machines
WordPress Cookie Plugin by Real Cookie Banner