Search Results

With the growth of the information economy, the proportion of knowledge-intensive goods to total goods is constantly increasing. Lawrence Lessig has argued that IPRs have now become too favourable to existing producers and that their ‘winner-take-all’ characteristics are constraining the creators of tomorrow. This chapter looks at how variations in IPRs regimes might affect the creation and social cost of new knowledge in economic systems. Drawing on a conceptual framework, the Information Space or I-Space, to explore how the uncontrollable diffusibility of knowledge relates to its degree of structure, this chapter deploys an agent-based modelling approach to the issue of IPRs. It takes the ability to control the diffusibility of knowledge as a proxy measure for an ability to establish property rights in such knowledge. Second, it takes the rate of obsolescence of knowledge as a proxy measure for the degree of turbulence induced by different regimes of technical change. Then, it simulates the quantity and cost to society of new knowledge under different property right regimes.

A definition of computer simulations is given, emphasizing the dynamic aspects of simulations. The important role of visualization in many simulations is explained. Computational issues in simulations are connected to observational issues with instruments via the microscope metaphor. Digital simulations are compared with analog simulations, and the advantages and disadvantages of simulations are explored. Agent-based simulations are compared to equation-based simulations.

In this chapter the methodological choice common to the contributions included in this volume is justified and discussed. Different approaches to cognitive modeling and social simulation are briefly reviewed, followed by a discussion of the reasons why modeling socio-cognitive phenomena is so hard, including: the Social Intelligence Hypothesis, Social Embedding, and the complexity of social Micro-Macro causation. Due to these fundamental difficulties, it is argued that, of all the approaches currently available, only agent-based simulation holds the prospect of adequately representing and understanding phenomena such as social norms. However, agent-based simulation is not without difficulties, so the balance between relevance and generality, as well as the twin processes of emergence and immergence in social simulation, are discussed.

Agents can shape their environment through the decisions they make, and when they do this in a coordinated manner, profound structure can emerge. Previously, the authors of this chapter have shown via computer simulations that agents using a simple recognition-based decision heuristic to choose among options in their environment could make some options much more popular than most others, if the agents communicated their knowledge directly to each other. This chapter shows that such environment shaping effects can arise even among non-communicating agents. All the agents need do is note the presence or absence of others making the same choices, akin to social conformity effects in human and social enhancement in foraging animals. Thus, social cognition mechanisms that are much simpler than full-blown communication can still enable strong environment structure to emerge in a population of interacting agents.

In a feedback-driven model of language change, conflict between patterns at distinct levels of linguistic organization can be understood as a form of multi-level selection. Simulations of conflicts between phonological and morphological patterns show that positive feedback can promote pattern consolidation at one level at the expense of the other. In this way, analogical generalizations over subparts of the lexicon can be shown to be emergent properties of complex linguistic systems.

This chapter—which still grapples with the third question posed at the beginning of the book; i.e., how to characterize agents that are able to recognize, adopt, and comply with norms—presents and discusses an agent architecture, EMIL-A, implementing the model of norms introduced in the preceding chapter. The simulations presented are intended to check the roles of norm-recognition and normative beliefs in favoring norm emergence and innovation in a highly fragmented social environment. In particular, on one hand, the chapter aims to examine the performance of our normative agents compared to other, cognitively less complex, agents, like imitators following the rule of the majority. On the other hand, it aims to check whether normative systems can reach convergence even when cultural or material artifacts are obstacles to the spreading of beliefs and behaviors. Will norm-detectives succeed in finding out (new) norms? Or will convergence collapse?

This chapter argues for the role of probabilistic enhancement in phonologization through computational simulation of an ongoing sound change in Seoul Korean. Two challenges faced by a phonologization model of sound change are addressed: explaining which cues are selected for phonologization, and explaining why phonologization is often accompanied by dephonologization. It is proposed that cues are targeted for enhancement as a probabilistic function of their statistical reliability in signaling a contrast. Simulation results using empirically derived cue values are taken to support the idea that loss of contrast precision may drive the phonologization process.

This chapter focuses on three questions: what norms are, how they emerge, and how much and what type of mental complexity they need; and the chapter presents a dynamic model of norms and the corresponding agent architecture (EMIL-A), and shows the results of its application to a stylized environment (a social multi-setting world). The chapter then illustrates a simulator, EMIL-S, on which EMIL-A has fully been implemented, showes its effects on the emergence of a new norm in a more complex artificial context (artificial Wikipedia), and compares the results with data from a survey on the real-world domain of reference (Wikipedia). This chapter describes EMIL-I-A (EMIL Internalizer Agent), an extension of EMIL-A designed to account for a deeper form of norm immergence than addressed so far; i.e., norm internalization. Then it presents simulation results aimed at testing how EMIL-I-A performs in dynamic, unpredictable scenarios.

We compare the predictions of our theory with empirical income data from a dozen different countries. We define a new measure of inequality, called the non-ideal inequality coefficient. We show that Norway is close to ideal inequality for the bottom 99% of the population while the U.S. is the most non-ideal at the other extreme. The other countries are in between these two. We find it remarkable that the Scandinavian societies have discovered the near-ideal share by themselves in practice without any prior knowledge of even its existence.