New Algorithm Illuminates Free Will

Findings from experiments at Stanford have recorded the brain signals when a primate exercises free will by changing its mind. The experiments were led by electrical engineering Professor Krishna Shenoy and neuroscientist  Matthew Kaufman while he was a graduate student in Shenoy’s lab and published this week in eLife.

In the reported experiments, Kaufman taught laboratory monkeys to perform a decision-making task. He then developed a novel technique to track the brain signals, an improvement on the “single trial decoder” algorithm.

“We are seeing many cognitive phenomena in the brain for the first time,” said Kaufman, who is currently a postdoc at Cold Spring Harbor Laboratory.

“The most critical result of our work here is that we can track a single decision and see how the monkey arrived there: whether he decided quickly, slowly, or changed his mind halfway through.”

In Kauffman’s experiments, monkeys were trained to reach for either of two targets within a maze displayed on a computer screen.



While it was possible for the monkey to reach either target, sometimes, one or the other target was blocked, resulting in a forced choice of one target. However, in other trials, the researchers would switch between the configurations while the monkey was already in the process of choosing. During the experiments 192 electrodes in the monkey’s motor and premotor cortex recorded the activity from the moment that the targets appeared on screen. These measurements continued until the monkey began to move. The interval between the targets’ appearance and the beginning of movement was recorded and marked the time of decision or alternatively signaled the monkey’s hesitation.

The Stanford team’s findings shed new light on a debate about human consciousness dating back to the 1980s. At the University of California, San Francisco,  neuroscientist Benjamin Libet conducted a now famous experiment to assess the nature of free will. Libet’s experiments showed that brain activity began several seconds before subjects became aware that they planned to take an action; in Libet’s experiments this entailed pushing a button. Libet concluded that the desire to move arose unconsciously prior to any awareness of the decision to move and therefore any apparent “free will” could only be a conscious veto of a prior existing unconscious decision. 

But based on these new observations at Stanford, Kaufman claims that the brain activity in Libet ‘s experiments does not demonstrate a lack of free will. You can plan to make a particular movement unconsciously, but then change your mind a second later. This suggests a conscious intention at work. If Kaufman is right it would settle this long standing philosophical and scientific debate that has significant bearing on transhumanist projects such as mind uploading and Artificial General Intelligence.

This research was supported by a Director’s Pioneer Award from the National Institutes of Health and by a REPAIR grant from DARPA.