{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,3]],"date-time":"2026-09-03T07:16:09Z","timestamp":1788419769597,"version":"build-2803163510"},"reference-count":45,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,9,3]],"date-time":"2026-09-03T00:00:00Z","timestamp":1788393600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Artif. Intell."],"abstract":"<jats:p>\n                    As large language models become primary communication partners, the stylistic and communicative character of their output\u2014specifically the degree to which it relies on intellectual, affective, or action-oriented language\u2014shapes how users interpret, and act on what they read. Yet this property is rarely measured or controlled directly. We introduce the Intellect-Emotion-Action Profile (IEAP), a purpose-built lexical framework featuring an inductively constructed dictionary from AI-generated text. IEAP decomposes any response into the proportional usage of intellectual, affective, and action words. Using a single automated harvester, we elicited and scored responses from four contemporary architectures (Claude, ChatGPT, Grok, and Gemini) across five question domains under four instructional-mode directives that span a cold-to-fire register (\n                    <jats:italic>n<\/jats:italic>\n                    \u202f=\u202f30 per architecture per domain). Three primary findings emerge. First, when no directive is applied, native communicative profiles are dominated by the underlying question, with architecture playing a secondary, architecture-dependent role; thus, the prompt establishes the baseline for the response. Second, an explicit mode directive displaces this profile substantially from its baseline. Specifically, the directives produced large, monotonic shifts in affective word usage, exceeding 50 percentage points in the most responsive domain with no reversals. Third, response depth alters length and conceptual breadth while leaving lexical composition essentially unchanged, establishing register and depth as orthogonal controls. These three dimensions correlated with human-rated NRC lexicons in the expected directions, supporting their validity. Within-domain robustness checks reproduce both the baseline profiles and the titration trajectories across multiple question framings. Because a measurable communicative property can be monitored and audited, these results position IEAP as a practical foundation for studying, comparing, and ultimately governing how AI systems communicate.\n                  <\/jats:p>","DOI":"10.3389\/frai.2026.1815425","type":"journal-article","created":{"date-parts":[[2026,9,3]],"date-time":"2026-09-03T06:44:09Z","timestamp":1788417849000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Words matter: measuring and titrating the communicative character of AI"],"prefix":"10.3389","volume":"9","author":[{"given":"William C.","family":"Kouns","sequence":"first","affiliation":[{"name":"Independent Researcher","place":["Knoxville, TN, United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,9,3]]},"reference":[{"key":"ref1","doi-asserted-by":"publisher","first-page":"139","DOI":"10.1111\/j.2517-6161.1982.tb01195.x","article-title":"The statistical analysis of compositional data","volume":"44","author":"Aitchison","year":"1982","journal-title":"J. R. Stat. Soc. B. Methodol."},{"key":"ref2","doi-asserted-by":"publisher","first-page":"187","DOI":"10.1007\/s11229-025-05022-6","article-title":"Do language models lack communicative intentions?","volume":"205","author":"Attah","year":"2025","journal-title":"Synthese"},{"key":"ref3","author":"Bender","year":"2020"},{"key":"ref4","doi-asserted-by":"publisher","first-page":"108663","DOI":"10.1016\/j.pec.2025.108663","article-title":"Exploring ChatGPT's communication behaviour in healthcare interactions: a psycholinguistic perspective","volume":"134","author":"Biassoni","year":"2025","journal-title":"Patient Educ. Couns."},{"key":"ref5","doi-asserted-by":"crossref","DOI":"10.1017\/CBO9780511621024","volume-title":"Variation Across Speech and Writing","author":"Biber","year":"1988"},{"key":"ref6","doi-asserted-by":"publisher","first-page":"1191","DOI":"10.1037\/0022-3514.47.6.1191","article-title":"Empirical validation of affect, behavior, and cognition as distinct components of attitude","volume":"47","author":"Breckler","year":"1984","journal-title":"J. Pers. Soc. Psychol."},{"key":"ref7","doi-asserted-by":"publisher","first-page":"1482","DOI":"10.1037\/0278-7393.22.6.1482","article-title":"Conceptual pacts and lexical choice in conversation","volume":"22","author":"Brennan","year":"1996","journal-title":"J. Exp. Psychol. Learn. Mem. Cogn."},{"key":"ref8","author":"Carroll","year":"2023"},{"key":"ref9","doi-asserted-by":"crossref","DOI":"10.1017\/CBO9780511620539","volume-title":"Using Language","author":"Clark","year":"1996"},{"key":"ref10","doi-asserted-by":"publisher","first-page":"2063","DOI":"10.3389\/fpsyg.2019.02063","article-title":"Language, gesture, and emotional communication: an embodied view of social interaction","volume":"10","author":"De Stefani","year":"2019","journal-title":"Front. Psychol."},{"key":"ref11","volume-title":"The Psychology of Attitudes","author":"Eagly","year":"1993"},{"key":"ref12","doi-asserted-by":"publisher","first-page":"279","DOI":"10.1023\/A:1023818214614","article-title":"Isometric logratio transformations for compositional data analysis","volume":"35","author":"Egozcue","year":"2003","journal-title":"Math. Geol."},{"key":"ref13","doi-asserted-by":"publisher","first-page":"335","DOI":"10.1017\/S0272263109990544","article-title":"Epilogue: a framework for investigating oral and written corrective feedback","volume":"32","author":"Ellis","year":"2010","journal-title":"Stud. Second. Lang. Acquis."},{"key":"ref14","doi-asserted-by":"publisher","first-page":"59","DOI":"10.3102\/00346543074001059","article-title":"School engagement: potential of the concept, state of the evidence","volume":"74","author":"Fredricks","year":"2004","journal-title":"Rev. Educ. Res."},{"key":"ref15","doi-asserted-by":"publisher","first-page":"16807","DOI":"10.1007\/s12144-025-08376-9","article-title":"Uncovering ChatGPT's narrative identity through a psycholinguistic perspective","volume":"44","author":"Gnerre","year":"2025","journal-title":"Curr. Psychol."},{"key":"ref16","doi-asserted-by":"publisher","first-page":"107","DOI":"10.1002\/1520-6696(198004)16:2<107::AID-JHBS2300160202>3.0.CO;2-Y","article-title":"The trilogy of mind: cognition, affection, and conation","volume":"16","author":"Hilgard","year":"1980","journal-title":"J. Hist. Behav. Sci."},{"key":"ref17","doi-asserted-by":"publisher","first-page":"91","DOI":"10.1016\/0010-0277(96)81418-1","article-title":"When do speakers take into account common ground?","volume":"59","author":"Horton","year":"1996","journal-title":"Cognition"},{"key":"ref18","author":"Hutto","year":"2014"},{"key":"ref19","doi-asserted-by":"publisher","first-page":"1262","DOI":"10.3390\/electronics15061262","article-title":"Evaluation of linguistic consistency of LLM-generated text personalization using natural language processing","volume":"15","author":"Huynh","year":"2026","journal-title":"Electronics"},{"key":"ref20","first-page":"350","volume-title":"Style in Language","author":"Jakobson","year":"1960"},{"key":"ref21","doi-asserted-by":"publisher","first-page":"3605","DOI":"10.48550\/arXiv.2305.02547","article-title":"PersonaLLM: investigating the ability of large language models to express personality traits","volume":"1","author":"Jiang","year":"2024","journal-title":"Find. Assoc. Comput. Linguist. NAACL 2024"},{"key":"ref22","doi-asserted-by":"publisher","first-page":"22199","DOI":"10.52202\/068431-1613","article-title":"Large language models are zero-shot reasoners","volume":"35","author":"Kojima","year":"2022","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref23","doi-asserted-by":"publisher","first-page":"20210483","DOI":"10.1098\/rstb.2021.0483","article-title":"Beyond speaking: neurocognitive perspectives on language production in social interaction","volume":"378","author":"Kuhlen","year":"2023","journal-title":"Philos. Trans. R. Soc. B"},{"key":"ref24","author":"Lu","year":"2022"},{"key":"ref25","doi-asserted-by":"publisher","first-page":"10387","DOI":"10.48550\/arXiv.2601.10387","article-title":"The assistant axis: situating and stabilizing the default persona of language models","volume":"2026","author":"Lu","year":"2026","journal-title":"arXiv"},{"key":"ref26","doi-asserted-by":"publisher","first-page":"961","DOI":"10.3389\/fpsyg.2015.00961","article-title":"The integration of emotional and symbolic components in multimodal communication","volume":"6","author":"Mehu","year":"2015","journal-title":"Front. Psychol."},{"key":"ref27","author":"Mohammad","year":""},{"key":"ref28","author":"Mohammad","year":""},{"key":"ref29","author":"Mohammad","year":""},{"key":"ref30","doi-asserted-by":"publisher","first-page":"23547","DOI":"10.48550\/arXiv.2503.23547","article-title":"NRC VAD Lexicon v2: Norms for Valence, Arousal, and Dominance for over 55k English Terms","volume":"2025","author":"Mohammad","year":"","journal-title":"arXiv"},{"key":"ref31","doi-asserted-by":"publisher","first-page":"436","DOI":"10.1111\/j.1467-8640.2012.00460.x","article-title":"Crowdsourcing a word-emotion association lexicon","volume":"29","author":"Mohammad","year":"2013","journal-title":"Comput. Intell."},{"key":"ref32","doi-asserted-by":"publisher","first-page":"477","DOI":"10.1038\/s41586-019-1138-y","article-title":"Machine behaviour","volume":"568","author":"Rahwan","year":"2019","journal-title":"Nature"},{"key":"ref33","author":"Reynolds","year":"2021"},{"key":"ref34","doi-asserted-by":"publisher","first-page":"1645","DOI":"10.1038\/s41562-025-02194-6","article-title":"On the conversational persuasiveness of GPT-4","volume":"9","author":"Salvi","year":"2025","journal-title":"Nat. Hum. Behav."},{"key":"ref35","doi-asserted-by":"publisher","first-page":"1954","DOI":"10.1038\/s42256-025-01115-6","article-title":"A psychometric framework for evaluating and shaping personality traits in large language models","volume":"7","author":"Serapio-Garc\u00eda","year":"2025","journal-title":"Nat. Mach. Intell."},{"key":"ref36","doi-asserted-by":"publisher","first-page":"68","DOI":"10.1145\/3624724","article-title":"Talking about large language models","volume":"67","author":"Shanahan","year":"2024","journal-title":"Commun. ACM"},{"key":"ref37","doi-asserted-by":"publisher","first-page":"493","DOI":"10.1038\/s41586-023-06647-8","article-title":"Role play with large language models","volume":"623","author":"Shanahan","year":"2023","journal-title":"Nature"},{"key":"ref38","doi-asserted-by":"publisher","first-page":"46","DOI":"10.1038\/s42256-022-00593-2","article-title":"Human-AI collaboration enables more empathic conversations in text-based peer-to-peer mental health support","volume":"5","author":"Sharma","year":"2023","journal-title":"Nat. Mach. Intell."},{"key":"ref39","author":"Sofroniew","year":"2026"},{"key":"ref40","doi-asserted-by":"publisher","first-page":"24","DOI":"10.1177\/0261927X09351676","article-title":"The psychological meaning of words: LIWC and computerized text analysis methods","volume":"29","author":"Tausczik","year":"2010","journal-title":"J. Lang. Soc. Psychol."},{"key":"ref41","doi-asserted-by":"publisher","first-page":"1771992","DOI":"10.3389\/frobt.2026.1771992","article-title":"A platform for investigating prompt framing as interface parameters in foundation models for robotics","volume":"13","author":"Tuladhar","year":"2026","journal-title":"Front. Robot. AI"},{"key":"ref42","doi-asserted-by":"publisher","first-page":"24824","DOI":"10.52202\/068431-1800","article-title":"Chain-of-thought prompting elicits reasoning in large language models","volume":"35","author":"Wei","year":"2022","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref43","doi-asserted-by":"publisher","first-page":"00679","DOI":"10.48550\/arXiv.2505.00679","article-title":"Steering large Language Models with Register Analysis for Arbitrary style Transfer","volume":"2025","author":"Yang","year":"2025","journal-title":"arXiv"},{"key":"ref44","author":"Zhao","year":"2021"},{"key":"ref45","doi-asserted-by":"publisher","first-page":"92","DOI":"10.1177\/09636625241252561","article-title":"Cognitive, affective, and behavioral engagement with science news predicted by the use of accessibility strategies in science-minded and general audiences","volume":"34","author":"Zimmerman","year":"2025","journal-title":"Public Underst. Sci."}],"container-title":["Frontiers in Artificial Intelligence"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frai.2026.1815425\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,3]],"date-time":"2026-09-03T06:44:11Z","timestamp":1788417851000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frai.2026.1815425\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,9,3]]},"references-count":45,"alternative-id":["10.3389\/frai.2026.1815425"],"URL":"https:\/\/doi.org\/10.3389\/frai.2026.1815425","relation":{},"ISSN":["2624-8212"],"issn-type":[{"value":"2624-8212","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,9,3]]},"article-number":"1815425"}}