The HoneyComb Paradigm for Research on Collective Human Behavior
January 19th, 2019
Here, we present the computer-based, multi-agent game HoneyComb, which enables experimental investigations of collective human movement behavior via black-dot-avatars on a virtual 2D hexagonal playfield. Different experimental conditions, like variable incentives on goal fields or vision radius, can be set, and their effects on human movement behavior can be investigated.
The computer-based multi-agent game HoneyComb, provides an experimental paradigm for investigating human movement coordination, leadership and other types of collective behavior. Using the HoneyComb paradigm, spatial temporal data of individual movement behavior can be collected and the development of behavioral patterns and collective structures can be analyzed. Our vision is that HoneyComb would provide new strategies for protecting crowds, for example it can have fast respondence identify affective approaches to leading a crowd to safety.
As long as a stable LAN and Java run-time environments are installed for each work station, there shouldn't be any problems operating software all the work station computers. A visual demonstration of the HoneyComb game who will make it possible for potential users to survey the whole process of setting up and running experiment to completion. Demonstrating the procedure will be Marie Ritter, a PHD student from my laboratory.
Select a location away from a high traffic area, such as in the computer lab or other specified area with individual work stations that can be configured as a local area network. Arrange for a 10 to 12 notebook computers of the same type to be used for the experiment on individual work station tables with chairs. With one centrally localized computer to function as the server.
Connect the notebooks to the server computer via ethernet cables and a network switch to create a local area network as illustrated. The server program and the client programs need a JAVA run-time environment which is available on all common operating systems. Install partition walls between the individual work stations to prevent visual sensory communication between neighboring participants.
Acquire one time-use ear plugs to be distributed to all the participants to prevent audible communication between participants. On the day of the procedure, assign the participants to their work stations and give the final instructions before the participants insert their ear plugs. It is crucial to the experiment that the participants not talk or otherwise communicate with each other while waiting to enter the work station room or while playing the HoneyComb game.
Then click okay, to start the session. The experiment will now be controlled only by the instructions on the screens visible to the participants. When all of the participants have clicked to indicate they have finished reading the instructions, observe whether the participants are mouse controlling their avatar dots on their HoneyComb virtual 97 hexagon play fields.
Have the participants start the testing phase in the central of the field and then left-click into the adjacent small hexagon of their choice to move their avatar dot onto the HoneyComb virtual play field according to the previously provided on-screen instructions. Only adjacent fields can be selected. After each move a small tail will appear for 4000 milliseconds indicating the last direction from which each participant tailed.
Allow each participant to partake in the experiment only one time to avoid possible biases. Once the game has been completed, have the participants fill out questionnaires assessing their demographic status. Big five personalty factors, perceived levels of stress or calmness and pay satisfaction.
While the participants are filling out their questionnaires, prepare anonymous envelopes with the appropriate amount of money earned in HoneyComb game according to the amounts computed on the server screen. Then distribute the earned payments to the participants as they exit the testing area and close the server program. When the server program has finished closing, transfer the data into two text files, marked by the day and time stamp of the experiment onto a USB flash drive.
In the representative experiment, two minority group members were informed about the location of one, two euro prize hexagon and five, one euro prize hexagons. The eight members of the majority group were not informed about the two euro prize hexagon and instead were shown the locations of six equally rewarded one euro gold hexagons. No participants were told that there were two different subgroups.
In the course of this experiment based solely on their movements without visual or audible communication. The two person minority group successfully led the other 80%of the participants to the two euro prize even though participant perception was restricted to only adjacent avatars on the play field. Surprisingly neither personalty variables nor computer literacy among the minority participants played a crucial role in their success.
Here are the main distances between group members overtime for the whole group, compared to both players who were informed about the location of the high reward two euro gold field and eight uninformed players are shown. By the end of the game, one uninformed player had lost the group and arrived on a single euro gold field. It is important to find a location with enough space for the 10 to 12 participants and their work stations and let the participants don't communicate while waiting for the game or during the game.
The game could be modified with environmental information, such as different cover stories, distinct participant avatars or communication via signaling to analyze collective behaviors in different settings. Our novel approach for conducting leadership research can be used to assess how leaders in a group emerge. And depend on the followership of others for successful leadership.
Overview
The multi-agent game HoneyComb serves as an experimental paradigm for studying collective human movement behavior. Participants control avatars on a virtual hexagonal playfield, allowing researchers to analyze coordination and leadership dynamics in group settings.
Key Study Components
Area of Science
- Collective behavior
- Human movement coordination
- Leadership dynamics
Background
- HoneyComb enables the collection of spatial-temporal data on individual movements.
- The game is designed to minimize communication between participants to ensure unbiased results.
- It aims to develop strategies for crowd management and safety.
- Participants engage in a controlled environment with specific instructions and restrictions.
Purpose of Study
- To investigate how individuals coordinate movements in a group.
- To analyze the emergence of leadership within collective behaviors.
- To explore the effects of incentives on group dynamics.
Methods Used
- Participants control avatars in a virtual game environment.
- Data on movement patterns are collected and analyzed.
- Participants are isolated to prevent communication.
- Questionnaires assess demographic data and psychological factors.
Main Results
- Minority group members successfully led the majority to higher rewards.
- Communication restrictions did not hinder the effectiveness of leadership.
- Personality traits did not significantly impact the outcomes.
- Spatial dynamics among participants revealed key insights into collective behavior.
Conclusions
- HoneyComb provides a novel approach to studying leadership and coordination.
- The findings suggest that effective leadership can emerge without direct communication.
- The game can be adapted for various experimental conditions to further explore collective behaviors.
Frequently Asked Questions
Executive Industry Relevance
Understanding collective human behavior provides strategic value for biopharma R&D by informing models of group decision-making, leadership emergence, and coordinated responses in preclinical and clinical trial settings. The HoneyComb paradigm enables controlled investigation of how individual behaviors scale to group-level patterns, supporting mechanistic de-risking in target validation and assay development. This approach enhances predictive confidence in translational research by linking individual phenotypic responses to emergent system behaviors.
Strategic Applications in Biopharma R&D
Early Discovery & Target Validation
- Scientific Value: Enables interrogation of therapeutic hypotheses through observation of how localized individual actions produce collective outcomes.
- Operational Value: Supports biological de-risking by clarifying how simple local parameters drive complex group behaviors relevant to pathway analysis.
- Predictive Value: Enhances portfolio triage by modeling how minority subgroups can influence majority responses, informing target confidence assessments.
Screening & Assay Development
- Scientific Value: Prepares validated biological systems for downstream workflows by measuring spatial-temporal movement parameters as quantitative phenotypic readouts.
- Operational Value: Addresses assay standardization and reproducibility through hexagonal coordinate tracking with 50 ms accuracy and individual ID resolution.
- Screening Readiness: Highlights scalability and platform reuse via computer-based multi-user adaptability to various research questions and situational conditions.
Translational & Preclinical Research
- Translational Continuity: Discusses disease relevance by connecting individual movement dynamics to group-level behavioral patterns observable in preclinical models.
- Risk-Adjusted Advancement: Describes how replication requirements and independent variable isolation support cross-functional collaboration in lead identification.
- Mechanistic De-risking: Focuses on predictive value by demonstrating how leadership emerges without direct communication, informing biomarker alignment strategies.
Pipeline & Workflow Integration
The HoneyComb method positions within the discovery continuum from hypothesis testing through lead identification to preclinical validation by enabling quantitative measurement of collective behavioral outputs.
- Discovery Biology: Explains how the method supports hypothesis testing, pathway clarification, and biological de-risking through controlled manipulation of cost-benefit ratios, incentives, and uncertainty.
- Screening: Describes assay readiness, reproducibility, and quantitative outputs via spatial-temporal data collection on individual movements over time.
- Analytics: Highlights measurements such as distances, velocity, and clustering that enable teams to compare conditions and analyze behavioral pattern development.
- Translational Research: Connects the method to preclinical continuity by combining movement data with questionnaire-derived non-computerized data for holistic behavioral assessment.
- Enterprise Reuse: Frames the method as a reusable capability rather than a single-use technique due to its adaptability to various experimental conditions and research questions.
Operational & Enterprise Impact
- Scientific Value: Predictive confidence, target validation, and reduction of mechanistic ambiguity in collective behavior modeling.
- Operational Value: Standardization, reproducibility, and scalability of human movement data collection in controlled environments.
- Strategic Value: Better go/no-go decisions, capital efficiency, and reduced late-stage biological risk through early-stage behavioral de-risking.
- Portfolio Impact: Risk-adjusted prioritization and advancement decisions based on emergent leadership and coordination dynamics.
Implementation Considerations
- Required scientific expertise in experimental design, behavioral science, and multi-agent system configuration.
- Instrumentation and analytical infrastructure needs including LAN-capable workstations, Java run-time environments, network switches, and partition walls for sensory isolation.
- Cross-team standardization requirements for consistent setup of individual work stations, ear plug distribution, and protocol adherence to prevent communication.
- Adaptation considerations across model systems via modification of environmental information, cover stories, participant avatars, and signaling mechanisms.
- Practical limitations include the need for dedicated space for 10–12 workstations, strict enforcement of non-communication protocols, and dependency on stable network and software environments.
