Analyzing Memory Effects in Large Language Models through the Lens of Cognitive Psychology

Abstract

Large Language Models (LLMs) exhibit memory-like behaviors that resemble aspects of human cognition. In this work, we examine such behaviors through the lens of cognitive psychology, drawing parallels between LLM outputs and classical findings on human memory including primacy/recency effects, interference, and forgetting curves. By bridging machine learning and cognitive science, we offer a framework for interpreting LLM behavior and discuss implications for evaluation, reliability, and the design of more human-aligned AI systems.

BibTeX Copy

@misc{cao2025analyzingmemory,
  title = {Analyzing Memory Effects in Large Language Models through the Lens of Cognitive Psychology},
  author = {Zhaoyang Cao and Lael Schooler and Reza Zafarani},
  year = {2025},
  keywords = {preprint},
  eprint = {2509.17138},
  archiveprefix = {arXiv},
  abstract = {Large Language Models (LLMs) exhibit memory-like behaviors that resemble aspects of human cognition. In this work, we examine such behaviors through the lens of cognitive psychology, drawing parallels between LLM outputs and classical findings on human memory including primacy/recency effects, interference, and forgetting curves. By bridging machine learning and cognitive science, we offer a framework for interpreting LLM behavior and discuss implications for evaluation, reliability, and the design of more human-aligned AI systems.},
  url = {https://arxiv.org/abs/2509.17138},
  pdf = {https://arxiv.org/pdf/2509.17138},
}