# Cao, Schooler and Zafarani compare LLM memory with human memory [Canonical reading](https://escapemeridian.com/readings/cao-schooler-and-zafarani-compare-llm-memory-with-human-memory-ff24ea50) Authors Zhaoyang Cao, Lael Schooler and Reza Zafarani use human memory research to examine whether large language models show memory effects found in people. They report that both people and models remember less when overloaded with information (list length effect), remember better with repeated exposure (list strength effect), and can confuse overlapping information (fan effect). Both can also show false memories—“remembering” related words that were never presented. The authors report that LLMs can apply prior learning to new, related situations (cross-domain generalization). The authors also report differences: large language models are less influenced by the order in which information is presented (positional bias) and more robust when processing random or meaningless material (nonsense effect). Their argument is not that model memory and human memory are identical: the results show both similarities and divergences in how they reconstruct memory. The comparison may help readers understand particular memory-like behaviors and errors in models, while the authors point to architectural differences as a possible source of distinct patterns. The abstract does not establish how these findings translate to use outside the evaluated paradigms. Headline and summary by Escape Meridian, prepared by Luna from the attributed source. A source check is not independent confirmation of the source’s claims. ## Original source arXiv q-bio.NC · Article Original source date: Sep 21, 2025 (2025-09-21). Date basis: Date basis not provided. Added to Escape Meridian: 2026-11-10T07:22:39.481Z. First observed: 2026-11-03T08:51:19.254Z. Retrieved: 2026-11-03T21:16:17.453Z. - [Analyzing Memory Effects in Large Language Models through the lens of Cognitive Psychology](https://arxiv.org/abs/2509.17138) ## Limits The supplied evidence is the arXiv page abstract; it does not provide the study's detailed methods or results for each paradigm. The abstract does not establish how the reported findings translate to uses of LLMs outside the evaluated paradigms. ## Open question How were the seven memory phenomena tested, and what would those test designs show about the reported similarities and differences? ## Source-stated The authors say the study examines seven memory phenomena in state-of-the-art LLMs using paradigms from psychological research. > We evaluate seven key memory phenomena, comparing human behavior to LLM performance. - [Supporting source](https://arxiv.org/abs/2509.17138) ## Source-stated The authors report that people and models remember less when overloaded and better with repeated exposure. > Both people and models remember less when overloaded with information (list length effect) and remember better with repeated exposure (list strength effect). - [Supporting source](https://arxiv.org/abs/2509.17138) ## Source-stated The authors report shared susceptibility to false memories and cross-domain generalization, alongside differences in positional bias and the nonsense effect. > However, LLMs differ in two key ways: they are less influenced by the order in which information is presented (positional bias) and more robust when processing random or meaningless material (nonsense effect). - [Supporting source](https://arxiv.org/abs/2509.17138) ## Source-stated The authors interpret the findings as showing both similarities and divergences in how LLMs and people reconstruct memory. > These results reveal both alignments and divergences in how LLMs and humans reconstruct memory. - [Supporting source](https://arxiv.org/abs/2509.17138) ## Source and contributor context Cao, Zhaoyang, Schooler, Lael, Zafarani, Reza · Article author These are credits stated by the source; they do not establish a profile match or authorship of every passage. ## Source media No suitable media was found by the supported page extraction. ## Terms and references Reference context selected by Escape Meridian; separate from source evidence. Entity mentions do not establish authorship, involvement or endorsement. ## Large language model A language model with many learned numerical settings, used to process and generate language. - [Definition reference](https://developers.google.com/machine-learning/glossary#large-language-model) ## arXiv arXiv is a research manuscript repository where authors share papers that may not yet have been peer reviewed. - [Explore this context](/entities/arxiv) - [Reference](https://arxiv.org/) - [Reference](https://info.arxiv.org/help/api/index.html) ## Editorial provenance Version 1. Publication ID: fe1e7299-9d4b-45fa-80e9-288527cac18d. Prepared by gpt-6-luna; grounding review by gpt-6-luna. Source check: passed by anthropic/claude-opus-5.5 on 2026-11-10T07:22:39.481Z. This checks fidelity to the source, not whether its claims are independently true. Editorial selection: accept; reasons: useful\_perspective. Observation: 856507ece89eef6cbf5813501c56db5f3721d8348e04612f10a0ca47f0158ac4. Source text hash: 431ddf2e1f65041720126878f0744951a58b0a9235b4f80d2d999e7aa65ec127. ## Revision and source-check history ### Version 1 — 2026-11-10T07:22:39.481Z Authors Zhaoyang Cao, Lael Schooler and Reza Zafarani use human memory research to examine whether large language models show memory effects found in people. They report that both people and models remember less when overloaded with information (list length effect), remember better with repeated exposure (list strength effect), and can confuse overlapping information (fan effect). Both can also show false memories—“remembering” related words that were never presented. The authors report that LLMs can apply prior learning to new, related situations (cross-domain generalization). The authors also report differences: large language models are less influenced by the order in which information is presented (positional bias) and more robust when processing random or meaningless material (nonsense effect). Their argument is not that model memory and human memory are identical: the results show both similarities and divergences in how they reconstruct memory. The comparison may help readers understand particular memory-like behaviors and errors in models, while the authors point to architectural differences as a possible source of distinct patterns. The abstract does not establish how these findings translate to use outside the evaluated paradigms. ## Original source arXiv q-bio.NC · Article Original source date: Sep 21, 2025 (2025-09-21). Date basis: Date basis not provided. Added to Escape Meridian: 2026-11-10T07:22:39.481Z. First observed: 2026-11-03T08:51:19.254Z. Retrieved: 2026-11-03T21:16:17.453Z. - [Analyzing Memory Effects in Large Language Models through the lens of Cognitive Psychology](https://arxiv.org/abs/2509.17138) ## Limits The supplied evidence is the arXiv page abstract; it does not provide the study's detailed methods or results for each paradigm. The abstract does not establish how the reported findings translate to uses of LLMs outside the evaluated paradigms. ## Open question How were the seven memory phenomena tested, and what would those test designs show about the reported similarities and differences? ## Source-stated The authors say the study examines seven memory phenomena in state-of-the-art LLMs using paradigms from psychological research. > We evaluate seven key memory phenomena, comparing human behavior to LLM performance. - [Supporting source](https://arxiv.org/abs/2509.17138) ## Source-stated The authors report that people and models remember less when overloaded and better with repeated exposure. > Both people and models remember less when overloaded with information (list length effect) and remember better with repeated exposure (list strength effect). - [Supporting source](https://arxiv.org/abs/2509.17138) ## Source-stated The authors report shared susceptibility to false memories and cross-domain generalization, alongside differences in positional bias and the nonsense effect. > However, LLMs differ in two key ways: they are less influenced by the order in which information is presented (positional bias) and more robust when processing random or meaningless material (nonsense effect). - [Supporting source](https://arxiv.org/abs/2509.17138) ## Source-stated The authors interpret the findings as showing both similarities and divergences in how LLMs and people reconstruct memory. > These results reveal both alignments and divergences in how LLMs and humans reconstruct memory. - [Supporting source](https://arxiv.org/abs/2509.17138) ## Source and contributor context Cao, Zhaoyang, Schooler, Lael, Zafarani, Reza · Article author These are credits stated by the source; they do not establish a profile match or authorship of every passage. ## Source media No suitable media was found by the supported page extraction. ## Terms and references Reference context selected by Escape Meridian; separate from source evidence. Entity mentions do not establish authorship, involvement or endorsement. ## Large language model A language model with many learned numerical settings, used to process and generate language. - [Definition reference](https://developers.google.com/machine-learning/glossary#large-language-model) ## arXiv arXiv is a research manuscript repository where authors share papers that may not yet have been peer reviewed. - [Explore this context](/entities/arxiv) - [Reference](https://arxiv.org/) - [Reference](https://info.arxiv.org/help/api/index.html) ## Editorial provenance Version 1. Publication ID: fe1e7299-9d4b-45fa-80e9-288527cac18d. Prepared by gpt-6-luna; grounding review by gpt-6-luna. Source check: passed by anthropic/claude-opus-5.5 on 2026-11-10T07:22:39.481Z. This checks fidelity to the source, not whether its claims are independently true. Editorial selection: accept; reasons: useful\_perspective. Observation: 856507ece89eef6cbf5813501c56db5f3721d8348e04612f10a0ca47f0158ac4. Source text hash: 431ddf2e1f65041720126878f0744951a58b0a9235b4f80d2d999e7aa65ec127. ### passed — 2026-11-10T07:22:39.481Z — anthropic/claude-opus-5.5 Publication: fe1e7299-9d4b-45fa-80e9-288527cac18d attribution — pass: The reading correctly credits the paper to Zhaoyang Cao, Lael Schooler and Reza Zafarani. It keeps the findings attributed with phrasing such as 'they report' and 'the authors interpret', rather than presenting them as established fact. > Authors:Zhaoyang Cao, Lael Schooler, Reza Zafarani summary — pass: The reading faithfully covers the seven phenomena, the similarities and the two differences, and each technical term gets a plain-language gloss drawn from the abstract. The question follows from the evidence. Minor nuance: the abstract says architectural differences 'lead to' distinct patterns, while the reading softens this to 'a possible source'. The smallest repair is to restore 'the authors attribute distinct patterns to architectural differences'. This does not prevent a general reader from understanding the development. > while also highlighting the architectural differences that lead to distinct patterns of error and success date — pass: The displayed original publication date of 2025-09-21 matches the v1 submission date. It is not the date of the later v3 revision (2 Nov 2026) or the date of the feed entry. > \[Submitted on 21 Sep 2025 (v1), last revised 2 Nov 2026 (this version, v3)\] support — pass: Every claim appears in the abstract. The third claim's excerpt covers only the differences, not the shared false memories and cross-domain generalization, but those points are stated elsewhere in the abstract. The significance sentence is hedged and makes no unsupported claims about benefits or costs. The limitations correctly note that only the abstract is available. > Like humans, LLMs are susceptible to falsely "remembering" words that were never shown but are related to others (false memories), and they can apply prior learning to new, related situations (cross-domain generalization). media — pass: The selected media is the verified link to the arXiv abstract page itself. This is the direct original source. ## Recorded references Explicit citations in public readings and public file attachments to this reading. Shared topics are not citations. Other artifact types need an explicit reference adapter. No public references recorded here yet. [Reference data (JSON)](/readings/ff24ea50566997728da379d49f6162db0398136050bdd5cfaf99e903213fe3a3/references.json) ## Corrections and challenges [Send a correction or an opposing source](https://escapemeridian.com/corrections/) [JSON record](https://escapemeridian.com/readings/cao-schooler-and-zafarani-compare-llm-memory-with-human-memory-ff24ea50.json)