# The role of System 1 and System 2 semantic memory structure in human and LLM biases

arXiv: https://arxiv.org/abs/2604.12816 · HTML: https://mlsanity.com/paper/2604.12816/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 20 of its 83 references that are in the map.
- First posted on arXiv 14 Apr 2026.

## Builds on (most cited first, 20 of 20)

- [Chain-of-Thought Prompting Elicits Reasoning in Large Language Models](https://mlsanity.com/paper/2201.11903/) (2022)
- [Emergent Abilities of Large Language Models](https://mlsanity.com/paper/2206.07682/) (2022)
- [From System 1 to System 2: A Survey of Reasoning Large Language Models](https://mlsanity.com/paper/2502.17419/) (2025)
- [Thinking Fast and Slow in Large Language Models](https://mlsanity.com/paper/2212.05206/) (2022)
- [Towards System 2 Reasoning in LLMs: Learning How to Think With Meta Chain-of-Thought](https://mlsanity.com/paper/2501.04682/) (2025)
- [Can Knowledge Graphs Reduce Hallucinations in LLMs? : A Survey](https://mlsanity.com/paper/2311.07914/) (2023)
- [The structure and dynamics of multilayer networks](https://mlsanity.com/paper/1407.0742/) (2014)
- [Prompting Techniques for Reducing Social Bias in LLMs through System 1 and System 2 Cognitive Processes](https://mlsanity.com/paper/2404.17218/) (2024)
- [Layer aggregation and reducibility of multilayer interconnected networks](https://mlsanity.com/paper/1405.0425/) (2014)
- [Dictionary-based Debiasing of Pre-trained Word Embeddings](https://mlsanity.com/paper/2101.09525/) (2021)
- [Reasoning on a Spectrum: Aligning LLMs to System 1 and System 2 Thinking](https://mlsanity.com/paper/2502.12470/) (2025)
- [Studying and improving reasoning in humans and machines](https://mlsanity.com/paper/2309.12485/) (2023)
- [Rethinking Language Models as Symbolic Knowledge Graphs](https://mlsanity.com/paper/2308.13676/) (2023)
- [LLM2: Let Large Language Models Harness System 2 Reasoning](https://mlsanity.com/paper/2412.20372/) (2024)
- [System 2 thinking in OpenAI's o1-preview model: Near-perfect performance on a mathematics exam](https://mlsanity.com/paper/2410.07114/) (2024)
- [The Latent Structure of Dictionaries](https://mlsanity.com/paper/1411.0129/) (2014)
- [Feature-rich multiplex lexical networks reveal mental strategies of early language learning](https://mlsanity.com/paper/2201.05061/) (2022)
- [The "LLM World of Words" English free association norms generated by large language models](https://mlsanity.com/paper/2412.01330/) (2024)
- [Detecting and Mitigating Bias in LLMs through Knowledge Graph-Augmented Training](https://mlsanity.com/paper/2504.00310/) (2025)
- [A word association network methodology for evaluating implicit biases in LLMs compared to humans](https://mlsanity.com/paper/2510.24488/) (2025)

## Built on by (most cited first, 1 of 1)

- [Evaluating Stochastic Collapse and Implicit Bias in Multimodal Large Language Models](https://mlsanity.com/paper/2606.05874/) (2026)

## Abstract (arXiv)

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.