The code from Mannering & Jones (2020) Catastrophic Interference in Predictive Neural Network Models of Distributional Semantics. Published in Computational Brain and Behavior.
Link to the paper Here.
Experiment 1 shows catastrophic interference effects in a neural network in both artificial and natural language corpora. Experiment 2 shows catastrophic interference effects in a neural network, a neural network with elastic weight consolidation (EWC), and a random vector accumulation (RVA) model.
We found that EWC was unable to prevent catastrophic interference in neural networks and that the RVA model was unaffected by catastrophic interference.