Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

To Stand on the Shoulders of Giants: Should We Protect Initial Discoveries in Multi-Agent Exploration?

Hodaya Lampert, Reshef Meir, Kinneret Teodorescu

TL;DR

This study analyzes whether protecting initial discoveries (e.g., patents) promotes or hinders exploration in competitive, multi-agent settings. It combines a theoretical model of exploration with a concrete treasure-hunt game and large-lab experiments under Protection and No Protection, plus a Singleton control. Key findings show that protection enhances coordination and overall search efficiency by reducing duplicated efforts, but tends to suppress sequential, follow-on discoveries and yields only limited gains for the initial discoveries, likely due to underweighting of rare events in learning environments. The results have policy implications for open science, licensing, and data-sharing mechanisms, suggesting that information exchange and failure-revealing markets may capture the efficiency benefits of protection without sacrificing subsequent innovation.

Abstract

Exploring new ideas is a fundamental aspect of research and development (R\&D), which often occurs in competitive environments. Most ideas are subsequent, i.e. one idea today leads to more ideas tomorrow. According to one approach, the best way to encourage exploration is by granting protection on discoveries to the first innovator. Correspondingly, only the one who made the first discovery can use the new knowledge and benefit from subsequent discoveries, which in turn should increase the initial motivation to explore. An alternative approach to promote exploration favors the \emph{sharing of knowledge} from discoveries among researchers allowing explorers to use each others' discoveries to develop further knowledge, as in the open-source community. With no protection, all explorers have access to all existing discoveries and new directions are explored faster. We present a game theoretic analysis of an abstract research-and-application game which clarifies the expected advantages and disadvantages of the two approaches under full information. We then compare the theoretical predictions with the observed behavior of actual players in the lab who operate under partial information conditions in both worlds. Our main experimental finding is that the no protection approach leads to \emph{more} investment efforts overall, in contrast to theoretical prediction and common economic wisdom, but in line with a familiar cognitive bias known as `underweighting of rare events'.

To Stand on the Shoulders of Giants: Should We Protect Initial Discoveries in Multi-Agent Exploration?

TL;DR

This study analyzes whether protecting initial discoveries (e.g., patents) promotes or hinders exploration in competitive, multi-agent settings. It combines a theoretical model of exploration with a concrete treasure-hunt game and large-lab experiments under Protection and No Protection, plus a Singleton control. Key findings show that protection enhances coordination and overall search efficiency by reducing duplicated efforts, but tends to suppress sequential, follow-on discoveries and yields only limited gains for the initial discoveries, likely due to underweighting of rare events in learning environments. The results have policy implications for open science, licensing, and data-sharing mechanisms, suggesting that information exchange and failure-revealing markets may capture the efficiency benefits of protection without sacrificing subsequent innovation.

Abstract

Exploring new ideas is a fundamental aspect of research and development (R\&D), which often occurs in competitive environments. Most ideas are subsequent, i.e. one idea today leads to more ideas tomorrow. According to one approach, the best way to encourage exploration is by granting protection on discoveries to the first innovator. Correspondingly, only the one who made the first discovery can use the new knowledge and benefit from subsequent discoveries, which in turn should increase the initial motivation to explore. An alternative approach to promote exploration favors the \emph{sharing of knowledge} from discoveries among researchers allowing explorers to use each others' discoveries to develop further knowledge, as in the open-source community. With no protection, all explorers have access to all existing discoveries and new directions are explored faster. We present a game theoretic analysis of an abstract research-and-application game which clarifies the expected advantages and disadvantages of the two approaches under full information. We then compare the theoretical predictions with the observed behavior of actual players in the lab who operate under partial information conditions in both worlds. Our main experimental finding is that the no protection approach leads to \emph{more} investment efforts overall, in contrast to theoretical prediction and common economic wisdom, but in line with a familiar cognitive bias known as `underweighting of rare events'.

Paper Structure

This paper contains 37 sections, 6 theorems, 12 equations, 21 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Previous experimental studies
  3. The current framework
  4. A Theoretical Model for Competitive Exploration
  5. Costs and utilities
  6. Protected research
  7. No protection
  8. The Competitive Treasure Hunt Game
  9. Protection condition
  10. No Protection condition
  11. Simulation Results
  12. Theoretical Predictions
  13. Behavioral predictions
  14. Experiment Design
  15. Participants
  16. ...and 22 more sections

Key Result

proposition 1

The optimal strategy in the protected condition is to play $x^*_0 = 1$, and $r^*_0$ is the unique $r$ s.t. $c'(r)=\frac{R_r+R_a}{2}$.

Figures (21)

  • Figure 1: A screenshot of the game, the Protection condition. Black hexagons represent failed searches, red hexagons are treasures that were found by other players and yellow hexagons are hexagons that were found by the player himself.
  • Figure 2: Left: Part of a screenshot in the No Protection condition. Notice that in this condition, there are no protected areas thus each mine can be discovered by more than one player. Right: Part of a screenshot in the Singleton condition.
  • Figure 3: Exploration rates for first treasures.
  • Figure 4: Exploration rates for subsequent treasures.
  • Figure 5: A screenshot of a typical end game under the protection condition.
  • ...and 16 more figures

Theorems & Definitions (10)

  • proposition 1
  • proof
  • proposition 2
  • corollary 1
  • proof
  • proposition 3
  • proposition 4
  • proof
  • proposition 5
  • proof