Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
This paper models System 1 and System 2 thinking as distinct semantic memory networks in both humans and LLMs, constructed from comparable datasets. They then use network-based evaluation metrics to investigate the relationship between these structures and implicit gender bias. The key finding is that semantic memory structures are irreducible only in humans, and that semantic memory structure relates consistently to implicit bias only in humans, suggesting LLMs lack the conceptual knowledge needed for bias regulation.
LLMs may never overcome bias in the same way humans do, as they lack the irreducible semantic memory structures that allow humans to regulate bias.
Implicit biases in both humans and large language models (LLMs) pose significant societal risks. Dual process theories propose that biases arise primarily from associative System 1 thinking, while deliberative System 2 thinking mitigates bias, but the cognitive mechanisms that give rise to this phenomenon remain poorly understood. To better understand what underlies this duality in humans, and possibly in LLMs, we model System 1 and System 2 thinking as semantic memory networks with distinct structures, built from comparable datasets generated by both humans and LLMs. We then investigate how these distinct semantic memory structures relate to implicit gender bias using network-based evaluation metrics. We find that semantic memory structures are irreducible only in humans, suggesting that LLMs lack certain types of human-like conceptual knowledge. Moreover, semantic memory structure relates consistently to implicit bias only in humans, with lower levels of bias in System~2 structures. These findings suggest that certain types of conceptual knowledge contribute to bias regulation in humans, but not in LLMs, highlighting fundamental differences between human and machine cognition.