Seminar: Algorithm-augmented cumulative cultural evolution

A balance of exploration and exploitation, or individual and social learning, shapes cumulative cultural evolution and collective innovation. Machines can increasinglyexplore and combine information from vast informational landscapes in a manner that far exceeds human capacities. The impact of such exploration on human innovation and cultural evolution is unknown and should be influenced by how humans explore and learn from machines. In two online experiments, participants were in either human-only or hybrid human-algorithm teams to solve an innovation task, with the transparency of the algorithm’s random exploration strategy and the availability of payoff information manipulated. Results suggest algorithm augmented innovation is conditional on the extent of algorithm exploration and strategies used by participants. Unlike the algorithm, participants explored in a directed rather than random manner and showed a bias against copying the algorithm, which limited social transmission. Simulations suggest that greater social learning, algorithmic appreciation and random exploration by humans could enhance innovation. Thisstudy suggests that the extent of algorithm search and the complementarities between human and machine strategies are critical to how machines may influence cultural evolution and collective innovation.

Maria Pykälä is a doctoral student at the Department of Organizational Behavior at the University of Lausanne. Her background is in evolutionary anthropology and sheuses experiments and simulations to study cultural evolution and social learning. Maria’s research focuses on social learning among humans and in hybrid human-machine systems, with a particular interest in how machines can complement and augment human collective intelligence.

Alternatively, join online:

Microsoft Teams

Meeting ID: 339 754 724 564 8
Passcode: gy3is3zj