Simulating paradigm shifts using a lock-in model

In this paper, we extend Brian Arthur's original lock-in model of increasing returns to adoption with a changing rate of technical change within competing technologies. This is done by including an S-curve in the pay-off function of alternatives reflecting the stage of the life-cycle of each "paradigm" in terms of its problem-solving capacity. Furthermore, we divide agents in generations that succeed one another in time. A new generation differs from an old generation in that the agents of a new generation do not take into account the previous decisions of agents of the old generation, but are only influenced by agents of the same generation. Simulations show that an escape of paradigm lock-in is possible when the end of the life-cycle of the existing dominant paradigm coincides with the advent of a new generation of agents. The model provides an analytical understanding of the interplay of cognitive and social dynamics that explain paradigm shifts.