Erratum to: A moderate role for cognitive models in agent-based modeling of cultural change

Correction

After publication of (Abrams 2013), I discovered errors in two formulas in the settle-nets paragraphs of the POPCO main loop section in Methods (page 13 in the PDF file). I correct these errors here. In the definition of the weighted sum n _i of input activations a _j from nodes j to node i across links with weights w _ij , the direction of the inequality was reversed. The definition of n _i should read:

$$n_i=\sum _j{w}_{\mathit{\text{ij}}}max(0,a_j)\text{for}{w}_{\mathit{\text{ij}}}<0.$$

There were also errors in the definition of s _i , the weighted sum of a node’s current activation a _i and its summed inputs p _i and n _i across positively and negatively weighted links, respectively. The definition of s _i should read as follows:

$$s_i=(1-.1)a_i+(.99-a_i)p_i+(a_i-(-.99\left)\right)n_i$$

I regret these errors and hope that they do not cause confusion. Full source code for the version of POPCO used in Abrams (2013) can be found at https://github.com/mars0i/popco.