Article Preview
Top1. Introduction
Knowledge spillovers are studied widely in literature and linked to innovation measures and outputs (Jaffe, 2008; Feldman & Audretsch, 1996). Spillovers are defined as the migration of knowledge beyond the domain borders (Fallah & Ibrahim, 2004). In our study, spillovers are not only expressed in terms of knowledge transfer but also mobility of scientists. Hence, spillovers result in formation of new domains as a consequence of knowledge and skill transfer.
A research question of interest is to examine the connection between individual rationality and aggregate efficiency. Axtell and Epstein (2006) discuss empirical data, which demonstrate that all individuals should not necessarily be rational to produce efficiency in macro level outcomes of a system. Given that individual rationality is bounded, authors explore how much rationality should exist in a system to generate desirable macro-level patterns. In this work, we do not propose to discern minimum level of rationality, but rather aim to address how rationality affects the spread of knowledge spillovers as well as growth and development of domains.
Scientists join or leave a problem domain on the basis of problems to be solved and tasks to be accomplished, and their position in the scientific landscape depends upon their knowledge, levels of interest, personal learning objectives, resources, and commitments (Hollingshead et al., 2002). Motivation of scientists as a personal interest is one of the main indicators for willingness of contribution. So, individual motivation is taken as a main force driving the commitment of a scientist to a problem domain.
An analogy can be built between predators and scientists. Predators are expected to abandon their current patch (e.g., domain) when local capture rate (e.g., problem solving success) is lower than estimated capture rate in the overall environment (Bernstein et al., 1988). Information foraging theory, that is developed by Pirolli and Card (1999) assumes that people, if they have opportunity, would adjust their strategies or the topology of their environment to maximize their rate of information gain. Humans are stated as informavores, who seek different information sources (Pirolli, 2007). In our study, scientists join or abandon problem domains based on perceived cues about their performance in attaining the desired outcome. The cues are represented by the instrumentality component of an individual’s motivation, which is described in detail in Section 3. Also, in this model, motivated and successful scientists recruit new scientists just as genetically more adapted predators are more likely to have an offspring in their natural environment.
Interaction landscape between scientists and information repositories in real life determines the time costs, resource costs, and opportunity costs of different information foraging strategies (Pirolli & Card, 1999). People are also selfish in applying their cost-benefit analyses. In accordance with these observations, three different characteristic of the scientists are defined in relation to cost-benefit decisions of the scientists.
David (1998) defines the force of open science’ s universalist pattern as providing entry into scientific artifacts and open discussion by all participants, while promoting openness in regard to new findings. Carayol and Dalle (2007) explain open-science phenomenon as significant freedom of scientists to choose what they want to do and how they want to do.
In light of these observations, we present an agent based model, called “KnowledgeSpill” to create a virtual environment where scientists have limited omniscience. In the model, opportunities in a particular problem domain deplete over time. We visualized the impacts of individual rationality and openness on the growth of scientific domains. In Section 2, we present the conceptual basis for our model. In Section 3 and Section 4, we describe the model structure and mechanisms in detail by using the Overview, Design concepts, and Details (ODD) protocol and descriptive visualizations. Section 5 discusses preliminary results and the qualitative observations derived from the created scenarios. In Section 6, we conclude by summarizing our findings in relation to reviewed literature and suggest potential avenues of future work.