getUpdatedBias: Return vector of updated biases for agents

Description Usage Arguments Details

View source: R/simulation.R

Description

Return vector of updated biases for agents

Usage

1
getUpdatedBias(agents, slope, limit = 0.05)

Arguments

agents

tbl with a snapshot of agents at a given time

slope

slope of the sigmoid function run on influence

limit

clamp resulting bias to between limit and 1-limit

Details

Bias is calculated by feeding influence (difference between final and initial judgement) into a sigmoid function. This gives an amount of bias influence between -.5 and .5, for which we take the absolute value and subtract .5, giving a value between 0 and .5 which is around 0 for low levels of influence and rises rapidly to around .5 for higher influence levels. This bias influence is then scaled to between 0 and the agent's bias_volatility parameter (when bias influence is .5), and this is used as the amount the bias updates towards the final decision direction. The final value of the bias is clamped to avoid extreme values.


oxacclab/adviseR documentation built on Oct. 7, 2021, 8:05 p.m.