Success-biased interaction to be paired with success_bias_teacher_selection that sets the learner's next behavior and fitness to be the teacher's current behavior and fitness.

success_bias_interact(learner, teacher, model)

net <- make_regular_lattice(10, 4)
model <- AgentBasedModel$new(network = net)
for (agent in model$agents) {
  # Agents need non-zero fitness for success bias to work
  agent$curr_fitness <- 1
}
output <- run(model, 10, 
              interact = success_bias_interact, 
              partner_selection = success_bias_select_teacher,
              iterate_model = iterate_learning_model)