View by Day

Prof. Jaromir Kovarik

University of the Basque Country, Spain

Authors: Dan Friedman, Jaromír Kovářík, Friederike Mengel

Abstract: In a series of laboratory and online experiments we investigate the perceived and actual influence, direct and indirect, that early movers have over later movers in a sequential move, binary action coordination game played on a network. Each subject is assigned to a particular node in a specified 5 or 10 node network, and chooses In or Out knowing which previous movers who are neighbors have chosen In. We measure influence by comparing the number of subsequent In choices following an initial In versus Out choice by a robot. Influence is highly heterogeneous across different network positions. People are aware of their influence and act on it, but tend to overestimate influence on average. A simple adaptive learning model explains entry and perceptions of influence reasonably well, especially for well connected nodes.