Raw data and additional plots for the paper Lying, more or less: A computer simulation study of graded lies and trust dynamics
收藏NIAID Data Ecosystem2026-03-11 收录
下载链接:
https://data.mendeley.com/datasets/8bpzprrp4y
下载链接
链接失效反馈官方服务:
资源简介:
May 3, 2020:
These are the raw data and additional plots for the simulations described in the paper Lying, more or less: A computer simulation study of graded lies and trust dynamics
It is contained in a zip archive.
The folder name describes the data:
XXstepsYYthresh: XX = the duration of the truth-default phase, YY = the distrust threshold (after the truth-default phase)
Filename prefixes denote the various lying styles:
v00: control run
v01: simple lying
v02: gambler's lying
v03: clairvoyant lying
Note: only control run and no-truth-default runs include data for variations that were omitted in the truth-default variations (inclinations to lie below 0.5 and data for biases below 0.5).
Plots like the one in Figure 5 for all variations of parameters are in the folder Plots -- there is also a script to produce them.
Raw data also include average credences and average trust for each step (row 1 = start, row 2: after round 1, etc).
Note 2: Average credences are in multiple columns: each column denotes the average credence about the corresponding coin in the following way:
If there are 11 columns, then the credences are for the whole hypothesis space (Col. 1: coin bias = 0, Col. 2: coin bias = 0.1, etc. to Col. 11: coin bias = 1).
If there are 6 columns (in the majority of the cases) we only include the credences for the relevant cases, i.e.
Col. 1: coin bias = 0.5, Col. 2: coin bias = 0.6, etc to Col. 6: coin bias = 1.
创建时间:
2020-06-09



