Loss of plasticity is different from catastrophic forgetting, which concerns poor performance on old examples even if they are not presented again, whereas loss of plasticity is about the inability to adapt to new tasks of the same difficulty.
In other words:
Maintaining stability is concerned with memorizing useful information.
Maintaining plasticity is concerned with finding new useful information when the data distribution changes.
Two types of loss of plasticity.
A) Unable to optimize new objectives.
B) Able to optimize new objectives, but unable to generalize.
They might be fundamentally diffeernt or there might be a common mechanism that can explain both.
References
Loss of plasticity in deep continual learning
Addressing Loss of Plasticity and Catastrophic Forgetting in Continual Learning