Pay Attention: Object Consideration as a Mechanism of Network Diffusion

Abstract

Theories of diffusion in networks rely on two broad classes of mechanisms: social influence directs the flow of information, and influence affects object valuation. This paper proposes an integrated model of decision-making for the adoption process. This model reveals a neglected middle step between information and valuation: social influence affects whether agents consider a particular object as relevant to the adoption decision. I identify this attention-driving mechanism using data on traders in an online foreign exchange platform. Features of the setting exclude the possibility of information- or valuation-driven diffusion, but traders still adopt one another’s trading behaviors, and they do so most for those rare behaviors that lack external drivers of attention. I discuss the importance of attention- driven diffusion for future work and the value of an integrated decision model in delineating conditions under which existing theories of influence apply.