A Bayesian Theory of Conformity in Collective Decision Making
Authors: Koosha Khalvati, Saghar Mirbagheri, Seongmin A. Park, Jean-Claude Dreher, Rajesh PN Rao
NeurIPS 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | The viability of our framework is demonstrated on two different experiments, a consensus task with 120 subjects and a volunteer s dilemma task with 29 subjects, each with multiple conditions. We tested our framework on two different collective decision making experiments involving human subjects: a consensus task and a volunteer s dilemma task. Our normative Bayesian framework explained and predicted human behavior well on both of these tasks. |
| Researcher Affiliation | Academia | Koosha Khalvati Paul G. Allen School of CSE University of Washington koosha@cs.washington.edu Saghar Mirbagheri Department of Psychology New York University sm7369@nyu.edu Seongmin A. Park Center for Mind and Brain University of California, Davis park@isc.cnrs.fr Jean-Claude Dreher Neuroeconomics Lab Institut des Sciences Cognitives Marc Jeannerod dreher@isc.cnrs.fr Rajesh P. N. Rao Paul G. Allen School of CSE & Center for Neurotechnology University of Washington rao@cs.washington.edu |
| Pseudocode | No | The paper describes models and equations but does not contain any pseudocode or algorithm blocks. |
| Open Source Code | No | The paper does not provide any links to open-source code or explicitly state that code is made available. |
| Open Datasets | Yes | We analyzed the behavioral data of [23] with 120 subjects performing consensus decision making in groups of N = 6 or N = 4 members. We analyzed the behavioral data from a Volunteer s Dilemma task [24] where 29 subjects played 12 games of a multi-round thresholded Public Goods Game (PGG). |
| Dataset Splits | Yes | In addition to fitting accuracy, we calculated Leave-One-Out Cross Validation (LOOCV) accuracy where at each iteration, the left-out data point was one whole game. |
| Hardware Specification | No | The paper does not specify the hardware used for running the experiments. |
| Software Dependencies | No | The paper does not provide specific version numbers for software dependencies. |
| Experiment Setup | Yes | For the conformity (level-0) model, there was only one free parameter, the decay rate λ. For the POMDP model (level-1)... Overall, the level-1 model had two free parameters (decay rate and winning value) for each subject. For the level-2 To M model... Therefore, the free parameters of the level-2 (and higher) models are the same as the level-1 model. For the models for all levels of To M, there are 3 free parameters in total, i.e., λ, α0 and β0. The RL model had 5 parameters in total. The first parameter was a reward for generating public good... The next two parameters determined the chance of producing public good for k = 2 versus k = 4, and were used to define the initial Q-value of each action. The final two free parameters determined the learning rate... |