Expectation management in humans and LLMs
- 1The Gonda Multidisciplinary Brain Research Center, Bar-Ilan University
Abstract
Mirative markers, such as “surprisingly”, explicitly encode a violation of expectations. Such markers are used for expectation management during communication. Sensitivity to mirative markers relies on two abilities: (i) updating expectations upon recognizing a mirative marker, and (ii) identifying expectation violations warranting the use of a mirative marker. In this study, we compared sensitivity to mirative markers in humans and large language models (LLMs). In part 1, we used a sentence-completion task, where humans and LLMs were presented with sentence fragments and asked to continue them. Results show that for both humans and LLMs, the presence of a mirative marker significantly increased response entropy and decreased top-response probability, in line with theoretical accounts of mirative processing. In part 2, we created a novel task, mirative polarity selection, where humans and LLMs are presented with a sentence pair and asked to select whether it was connected by a mirative marker (“surprisingly”) or an anti-mirative marker (“unsurprisingly”). Results show that some LLMs perform at a human-equivalent level. We conclude that both humans and LLMs use mirative markers as cues for calibrating their subsequent expectations during sentence processing.
Related
- NMR-Challenge for LLMs: Evaluating Chemical Reasoning in Humans and AI
- Harnessing the Potential of LLMs in Hospital Management: Insights Into Medical Data Inquiry
- Development and Validation of the Food Expectation Questionnaire (FEX‐Q): An Image‐Based Instrument to Assess Food‐Related Perceptions and Symptom Expectations
- Hybrid retrieval–verification framework for hallucination reduction in medical LLMs
- Can LLMs Succeed in National Dental Examinations? A Comprehensive Analysis of Performance, Strengths, and Limitations
BibTeX
@article{menashe2026expectation,
title = {{Expectation management in humans and LLMs}},
author = {Benjamin Menashe and Austin Drake and Michal Ben-Shachar},
journal = {Scientific Reports},
year = {2026},
month = jul,
doi = {10.1038/s41598-026-62526-y},
url = {https://doi.org/10.1038/s41598-026-62526-y},
note = {CC BY},
}
RIS
TY - JOUR
TI - Expectation management in humans and LLMs
AU - Benjamin Menashe
AU - Austin Drake
AU - Michal Ben-Shachar
PY - 2026
DA - 2026/07/01
JO - Scientific Reports
DO - 10.1038/s41598-026-62526-y
UR - https://doi.org/10.1038/s41598-026-62526-y
AB - Mirative markers, such as “surprisingly”, explicitly encode a violation of expectations. Such markers are used for expectation management during communication. Sensitivity to mirative markers relies on two abilities: (i) updating expectations upon recognizing a mirative marker, and (ii) identifying expectation violations warranting the use of a mirative marker. In this study, we compared sensitivity to mirative markers in humans and large language models (LLMs). In part 1, we used a sentence-completion task, where humans and LLMs were presented with sentence fragments and asked to continue them. Results show that for both humans and LLMs, the presence of a mirative marker significantly increased response entropy and decreased top-response probability, in line with theoretical accounts of mirative processing. In part 2, we created a novel task, mirative polarity selection, where humans and LLMs are presented with a sentence pair and asked to select whether it was connected by a mirative marker (“surprisingly”) or an anti-mirative marker (“unsurprisingly”). Results show that some LLMs perform at a human-equivalent level. We conclude that both humans and LLMs use mirative markers as cues for calibrating their subsequent expectations during sentence processing.
SN - 2045-2322
ER -