Proceedings of the Annual Meeting of the Cognitive Science Society

Humans have a remarkable capacity to reason about abstract relational structures, an ability that may support some ofthe most impressive, human-unique cognitive feats. Equality (or identity) reasoning has been a key case study of abstractrelational reasoning. This paper revisits the question of whether equality can be learned in non-symbolic neural networks.We find that simple neural networks are able to learn basic equality with relatively little training data. In a second casestudy, we show that sequential equality problems (learning ABA sequences) can be solved with only positive traininginstances. Finally, we consider a more complex, hierarchical equality problem, and find that this task can be solved witheither avast amount of training data or pre-training on basic equality. Overall, these findings indicate that neural modelsare able to solve equality-based reasoning tasks, suggesting that essential aspects of symbolic reasoning can emerge fromdata-driven,non-symbolic learning processes.