Simulation of linguistic change

Defining the social model

Society parameters

ALL_EQUAL constant ID for a FSID: social influence is equal for all speakers.
DIST_IMPACT [0;4] exponent of distance in the formula.
HYPER_INFLUENTIAL constant ID for a FSID. Like POISSON_CURVE but with some (approx. 2.5%) speakers who are hyper-influential.
MAX_AGE > 1 speaker dies when reaching this age.
MOVE_IN [0;1] chance of moving into the society.
MOVE_OUT [0;1] chance of moving away from the society.
MUT_RATE [0;1] mutation rate; chance of acquiring another E-pattern than that determined by the adaption algorithm
NORMALLY_DISTRIB constant ID for a FSID that distributes social influence normally.
NUM_IMPACT [0;4] exponent of the number of speakers using a certain E-pattern. a value of 0.8 for example accounts to the fact that an increase in the number of people using a certain E-pattern doesn't increase the over all impact of the E-pattern in a linear way.
POISSON_CURVE constant ID for a FSID that distributes social influence normally. after distribution social influence is squared
SOC_LEVELS 20 levels of social influences. they don't have to be equidistant. this depends on the used FSID (function for social influence distribution)
SUDDEN_DEATH [0;1] chance of dying before reaching the maximum age.

[Back] | [HOME]


Speaker parameters

age [0;MAX_AGE] age of speaker measured in generations.
a_v > 0 ID for the E-pattern the speaker uses at the moment.
inf [1;max{Society.getSocInf(SID)}] social influence of the speaker. the maximum possible influence and its distribution among the speakers is determined by one FSID (function for social influence distribution).

[Back] | [HOME]