{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,16]],"date-time":"2026-05-16T08:14:31Z","timestamp":1778919271447,"version":"3.51.4"},"reference-count":39,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2026,5,16]],"date-time":"2026-05-16T00:00:00Z","timestamp":1778889600000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2026,5,16]],"date-time":"2026-05-16T00:00:00Z","timestamp":1778889600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"name":"National Education Science Planning Project, Youth Projects of the Ministry of Education by Office of the Leading group for National Education Science Planning Project","award":["EBA240402"],"award-info":[{"award-number":["EBA240402"]}]}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["Cogn Comput"],"published-print":{"date-parts":[[2026,12]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>Large language models (LLMs) have emerged as efficient tools for generating psycholinguistic norm data. However, their capacity to capture embodied cognition, particularly metaphorical associations grounded in sensorimotor experience, remains insufficiently understood. The present study provides the first systematic investigation of taste-emotion metaphorical mappings in GPT-4o by comparing model-generated responses with established human norms across four experiments. In Experiments 1 and 2, we used free-association tasks, prompting GPT-4o via the OpenAI Application Programming Interface (API) to generate the taste word most strongly associated with emotion-related stimuli (Experiment 1) and the emotion word most strongly associated with taste stimuli (Experiment 2). In Experiments 3 and 4, we used Likert-scale rating tasks to quantify the associative strength of taste-emotion pairs (Experiment 3) and concept-taste pairs (Experiment 4). We systematically varied temperature settings (T\u2009=\u20090, 0.7, 1.0) and iteration conditions to assess response stability. Across tasks, GPT-4o showed weak-to-moderate correlations with human norms, indicating meaningful but incomplete alignment in metaphorical mappings. The model aligned more closely with human responses for basic emotion words (e.g., love) than for emotion-laden concepts (e.g., wedding), and performed better in structured rating tasks than in open-ended free-association tasks. GPT-4o also exhibited high internal consistency across temperature settings and iterations. These findings suggest that current LLMs can recover a substantial portion of conventional taste-emotion mappings from language alone, but remain limited in approximating fully embodied metaphorical knowledge. The results have important implications for the use of LLMs in psycholinguistic research, especially in domains where sensorimotor grounding is theoretically central.<\/jats:p>","DOI":"10.1007\/s12559-026-10594-7","type":"journal-article","created":{"date-parts":[[2026,5,16]],"date-time":"2026-05-16T07:54:23Z","timestamp":1778918063000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Tasting Emotions with GPT-4o: Comparing AI and Human Norms in Metaphoric Associations"],"prefix":"10.1007","volume":"18","author":[{"given":"Yanyun","family":"Zhou","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Wei","family":"Li","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Chi-Shing","family":"Tse","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jiushu","family":"Xie","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yanli","family":"Huang","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"297","published-online":{"date-parts":[[2026,5,16]]},"reference":[{"key":"10594_CR1","doi-asserted-by":"publisher","first-page":"7574","DOI":"10.3758\/s13428-024-02437-w","volume":"56","author":"YL Chan","year":"2024","unstructured":"Chan YL, Tse CS. Decoding the essence of two-character Chinese words: Unveiling valence, arousal, concreteness, familiarity, and imageability through word norming. Behav Res Methods. 2024;56:7574\u2013601.","journal-title":"Behav Res Methods"},{"key":"10594_CR2","doi-asserted-by":"publisher","first-page":"686","DOI":"10.3758\/s13428-015-0598-8","volume":"48","author":"TM Sutton","year":"2016","unstructured":"Sutton TM, Altarriba J. Color associations to emotion and emotion-laden words: A collection of norms for stimulus construction and selection. Behav Res Methods. 2016;48:686\u2013728.","journal-title":"Behav Res Methods"},{"key":"10594_CR3","doi-asserted-by":"publisher","first-page":"6082","DOI":"10.3758\/s13428-024-02337-z","volume":"56","author":"S Trott","year":"2024","unstructured":"Trott S. Can large language models help augment English psycholinguistic datasets? Behav Res Methods. 2024;56:6082\u2013100.","journal-title":"Behav Res Methods"},{"key":"10594_CR4","doi-asserted-by":"publisher","first-page":"1","DOI":"10.3390\/e24020280","volume":"24","author":"N Levshina","year":"2022","unstructured":"Levshina N. Frequency, informativity and word length: Insights from typologically diverse corpora. Entropy. 2022;24(2):280.","journal-title":"Entropy"},{"issue":"1","key":"10594_CR5","first-page":"1","volume":"9","author":"J Conde","year":"2026","unstructured":"Conde J, Mart\u00ednez G, Grandury M, Arriaga C, Haro J, Schroeder S, et al. Updating the German Psycholinguistic Word Toolbox with AI-Generated Estimates of Concreteness, Valence, Arousal, Age of Acquisition, and Familiarity. J Cogn. 2026;9(1):9.","journal-title":"J Cogn"},{"issue":"2","key":"10594_CR6","doi-asserted-by":"publisher","first-page":"1","DOI":"10.3758\/s13428-025-02938-2","volume":"58","author":"H Hagihara","year":"2026","unstructured":"Hagihara H, Miyazawa K. How well do large language models mirror human cognition of word concepts? A comparison of psychological ratings for early-acquired English words. Behav Res Methods. 2026;58:58 .","journal-title":"Behav Res Methods"},{"key":"10594_CR7","doi-asserted-by":"publisher","first-page":"597","DOI":"10.1016\/j.tics.2023.04.008","volume":"27","author":"D Dillion","year":"2023","unstructured":"Dillion D, Tandon N, Gu Y, Gray K. Can AI language models replace human participants? Trends Cogn Sci. 2023;27:597\u2013600.","journal-title":"Trends Cogn Sci"},{"key":"10594_CR8","doi-asserted-by":"crossref","unstructured":"Mart\u00ednez G, Molero JD, Gonzalez S, Conde J, Brysbaert M, Reviriego P. Using large language models to estimate features of multi-word expressions: Concreteness, valence, arousal. Behav Res Methods. 2025;57:5.","DOI":"10.3758\/s13428-024-02515-z"},{"key":"10594_CR9","first-page":"1","volume":"57","author":"M Brysbaert","year":"2025","unstructured":"Brysbaert M, Mart\u00ednez G, Reviriego P. Moving beyond word frequency based on tally counting: AI-generated familiarity estimates of words and phrases are an interesting additional index of language knowledge. Behav Res Methods. 2025;57:28.","journal-title":"Behav Res Methods"},{"key":"10594_CR10","doi-asserted-by":"crossref","unstructured":"He Q, Wang Y, Rao G. Generation and validation of teaching examples based on large language models. In: 2024 International Conference on Asian Language Processing (IALP). 2024;389\u2009\u2013\u200995.","DOI":"10.1109\/IALP63756.2024.10661177"},{"key":"10594_CR11","unstructured":"Zhu Y, Zhang P, Haq EU, Hui P, Tyson G. Can ChatGPT reproduce human-generated labels? A study of social computing tasks. arXiv preprint arXiv:2304.10145. 2023."},{"issue":"3","key":"10594_CR12","doi-asserted-by":"publisher","first-page":"181","DOI":"10.1177\/09637214241242746","volume":"33","author":"L Connell","year":"2024","unstructured":"Connell L, Lynott D. What can language models tell us about human cognition? Curr Dir Psychol Sci. 2024;33(3):181\u20139.","journal-title":"Curr Dir Psychol Sci"},{"issue":"10","key":"10594_CR13","doi-asserted-by":"publisher","first-page":"1220","DOI":"10.1080\/23273798.2022.2069278","volume":"37","author":"C Wingfield","year":"2022","unstructured":"Wingfield C, Connell L. Understanding the role of linguistic distributional knowledge in cognition. Lang Cogn Neurosci. 2022;37(10):1220\u201370.","journal-title":"Lang Cogn Neurosci"},{"issue":"2","key":"10594_CR14","doi-asserted-by":"publisher","first-page":"273","DOI":"10.1111\/j.1756-8765.2010.01106.x","volume":"3","author":"MM Louwerse","year":"2011","unstructured":"Louwerse MM. Symbol interdependency in symbolic and embodied cognition. Top Cogn Sci. 2011;3(2):273\u2013302.","journal-title":"Top Cogn Sci"},{"issue":"5","key":"10594_CR15","doi-asserted-by":"publisher","first-page":"e1555","DOI":"10.1002\/wcs.1555","volume":"12","author":"CP Davis","year":"2021","unstructured":"Davis CP, Yee E. Building semantic memory from embodied and distributional language experience. Wiley Interdiscip Rev Cogn Sci. 2021;12(5):e1555.","journal-title":"Wiley Interdiscip Rev Cogn Sci"},{"issue":"2","key":"10594_CR16","doi-asserted-by":"publisher","first-page":"381","DOI":"10.1111\/j.1551-6709.2010.01157.x","volume":"35","author":"M Louwerse","year":"2011","unstructured":"Louwerse M, Connell L. A taste of words: Linguistic context and perceptual simulation predict the modality of words. Cogn Sci. 2011;35(2):381\u201398.","journal-title":"Cogn Sci"},{"key":"10594_CR17","doi-asserted-by":"publisher","first-page":"91","DOI":"10.1016\/S0004-3702(03)00054-7","volume":"149","author":"ML Anderson","year":"2003","unstructured":"Anderson ML. Embodied cognition: A field guide. Artif Intell. 2003;149:91\u2013130.","journal-title":"Artif Intell"},{"key":"10594_CR18","volume-title":"Philosophy in the flesh: The embodied mind and its challenge to western thought","author":"G Lakoff","year":"1999","unstructured":"Lakoff G, Johnson M. Philosophy in the flesh: The embodied mind and its challenge to western thought. Chicago: University of Chicago Press; 1999."},{"key":"10594_CR19","doi-asserted-by":"publisher","first-page":"1045","DOI":"10.1037\/a0020970","volume":"136","author":"MJ Landau","year":"2010","unstructured":"Landau MJ, Meier BP, Keefer LA. A metaphor-enriched social cognition. Psychol Bull. 2010;136:1045\u201367.","journal-title":"Psychol Bull"},{"key":"10594_CR20","doi-asserted-by":"publisher","first-page":"106127","DOI":"10.1016\/j.cognition.2025.106127","volume":"260","author":"A Dymarska","year":"2025","unstructured":"Dymarska A, Connell L. Free word association is driven by local response chaining of linguistic and sensorimotor relationships. Cognition. 2025;260:106127.","journal-title":"Cognition"},{"issue":"1","key":"10594_CR21","doi-asserted-by":"publisher","first-page":"96","DOI":"10.1016\/j.cognition.2009.09.002","volume":"114","author":"MM Louwerse","year":"2010","unstructured":"Louwerse MM, Jeuniaux P. The linguistic and embodied nature of conceptual processing. Cognition. 2010;114(1):96\u2013104.","journal-title":"Cognition"},{"key":"10594_CR22","doi-asserted-by":"publisher","first-page":"1142","DOI":"10.1037\/a0033758","volume":"13","author":"KQ Chan","year":"2013","unstructured":"Chan KQ, Tong EM, Tan DH, Koh AH. What do love and jealousy taste like? Emotion. 2013;13:1142\u20139.","journal-title":"Emotion"},{"key":"10594_CR23","doi-asserted-by":"publisher","first-page":"339","DOI":"10.1080\/20445911.2021.2020797","volume":"34","author":"Y Zhou","year":"2022","unstructured":"Zhou Y, Tse CS. Sweet taste brings happiness, but happiness does not taste sweet: The unidirectionality of taste-emotion metaphoric association. J Cogn Psychol. 2022;34:339\u201361.","journal-title":"J Cogn Psychol"},{"key":"10594_CR24","doi-asserted-by":"publisher","first-page":"163","DOI":"10.1037\/a0025253","volume":"102","author":"BP Meier","year":"2012","unstructured":"Meier BP, Moeller SK, Riemer-Peltz M, Robinson MD. Sweet taste preferences and experiences predict prosocial inferences, personalities, and behaviors. J Pers Soc Psychol. 2012;102:163\u201374.","journal-title":"J Pers Soc Psychol"},{"key":"10594_CR25","doi-asserted-by":"publisher","first-page":"1222","DOI":"10.1126\/science.1165565","volume":"323","author":"HA Chapman","year":"2009","unstructured":"Chapman HA, Kim DA, Susskind JM, Anderson AK. In bad taste: Evidence for the oral origins of moral disgust. Science. 2009;323:1222\u20136.","journal-title":"Science"},{"issue":"4","key":"10594_CR26","doi-asserted-by":"publisher","first-page":"561","DOI":"10.1016\/j.jcps.2015.02.006","volume":"25","author":"AE Schlosser","year":"2015","unstructured":"Schlosser AE. The sweet taste of gratitude: Feeling grateful increases choice and consumption of sweets. J Consum Psychol. 2015;25(4):561\u201376.","journal-title":"J Consum Psychol"},{"issue":"3","key":"10594_CR27","doi-asserted-by":"publisher","first-page":"295","DOI":"10.1177\/0956797611398497","volume":"22","author":"KJ Eskine","year":"2011","unstructured":"Eskine KJ, Kacinik NA, Prinz JJ. A bad taste in the mouth: Gustatory disgust influences moral judgment. Psychol Sci. 2011;22(3):295\u20139.","journal-title":"Psychol Sci"},{"key":"10594_CR28","doi-asserted-by":"publisher","first-page":"986","DOI":"10.3389\/fpsyg.2020.00986","volume":"11","author":"Y Zhou","year":"2020","unstructured":"Zhou Y, Tse CS. The taste of emotion: Metaphoric association between taste words and emotion\/emotion-laden words. Front Psychol. 2020;11:986.","journal-title":"Front Psychol"},{"key":"10594_CR29","doi-asserted-by":"crossref","unstructured":"Zhu Y, Li J, Li G et al. Hot or Cold? Adaptive Temperature Sampling for Code Generation with Large Language Models. In: Proceedings of the AAAI Conference on Artificial Intelligence. 2024;38:437\u2009\u2013\u200945.","DOI":"10.1609\/aaai.v38i1.27798"},{"key":"10594_CR30","doi-asserted-by":"crossref","unstructured":"Renze M, Guven E. The effect of sampling temperature on problem solving in large language models. In: Findings of the association for computational linguistics: EMNLP 2024. 2024;7346-56.","DOI":"10.18653\/v1\/2024.findings-emnlp.432"},{"issue":"1","key":"10594_CR31","doi-asserted-by":"publisher","first-page":"1506","DOI":"10.1038\/s41598-024-85003-w","volume":"15","author":"I Azimi","year":"2025","unstructured":"Azimi I, Qi M, Wang L, Rahmani AM, Li Y. Evaluation of LLMs accuracy and consistency in the registered dietitian exam through prompt engineering and knowledge retrieval. Sci Rep. 2025;15(1):1506.","journal-title":"Sci Rep"},{"key":"10594_CR32","unstructured":"Guizeni S. Exploring the default temperature setting of GPT-4. Seifeur Guizeni - AI\/ML Engineer & SEO Consultant. 2024. https:\/\/seifeur.com\/gpt-4-default-temperature\/. Accessed 15 Jul 2025."},{"key":"10594_CR33","unstructured":"Bradley MM, Lang PJ. Affective norms for English words (ANEW): Instruction manual and affective ratings. Technical Report C-1, University of Florida, Gainesville. 1999."},{"issue":"1","key":"10594_CR34","doi-asserted-by":"publisher","first-page":"e53559","DOI":"10.2196\/53559","volume":"11","author":"J Davis","year":"2024","unstructured":"Davis J, Van Bulck L, Durieux BN, Lindvall C. The temperature feature of ChatGPT: modifying creativity for clinical research. JMIR Hum Factors. 2024;11(1):e53559.","journal-title":"JMIR Hum Factors"},{"key":"10594_CR35","unstructured":"Stevenson C, Smal I, Baas M, Grasman R, van der Maas H. Putting GPT-3\u2019s creativity to the (alternative uses) test. arXiv preprint arXiv:2206.08932. 2022."},{"key":"10594_CR36","unstructured":"Peeperkorn M, Kouwenhoven T, Brown D, Jordanous A. Is temperature the creativity parameter of large language models? arXiv preprint arXiv:2405.00492. 2024."},{"key":"10594_CR37","doi-asserted-by":"crossref","unstructured":"Li L, Sleem L, Gentile N, Nichil G, State R. Exploring the Impact of Temperature on Large Language Models: Hot or Cold? arXiv preprint arXiv:2506.07295. 2025.","DOI":"10.1016\/j.procs.2025.07.135"},{"key":"10594_CR38","doi-asserted-by":"crossref","unstructured":"Bender EM, Koller A. Climbing towards NLU: On meaning, form, and understanding in the age of data. In: Proceedings of the 58th annual meeting of the association for computational linguistics. 2020;5185-98.","DOI":"10.18653\/v1\/2020.acl-main.463"},{"key":"10594_CR39","doi-asserted-by":"publisher","first-page":"35","DOI":"10.1007\/s12559-024-10393-y","volume":"17","author":"R Mao","year":"2025","unstructured":"Mao R, Chen G, Li X, Ge M, Cambria E. A comparative analysis of metaphorical cognition in ChatGPT and human minds. Cogn Comput. 2025;17:35.","journal-title":"Cogn Comput"}],"container-title":["Cognitive Computation"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s12559-026-10594-7.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/article\/10.1007\/s12559-026-10594-7","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s12559-026-10594-7.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,5,16]],"date-time":"2026-05-16T07:54:28Z","timestamp":1778918068000},"score":1,"resource":{"primary":{"URL":"https:\/\/link.springer.com\/10.1007\/s12559-026-10594-7"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,5,16]]},"references-count":39,"journal-issue":{"issue":"1","published-print":{"date-parts":[[2026,12]]}},"alternative-id":["10594"],"URL":"https:\/\/doi.org\/10.1007\/s12559-026-10594-7","relation":{},"ISSN":["1866-9956","1866-9964"],"issn-type":[{"value":"1866-9956","type":"print"},{"value":"1866-9964","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,5,16]]},"assertion":[{"value":"5 August 2025","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"4 May 2026","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"16 May 2026","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"The authors declare no competing interests.","order":2,"name":"Ethics","group":{"name":"EthicsHeading","label":"Competing interests"}}],"article-number":"47"}}