Effective prompt engineering when seeking psychological support from artificial intelligence
The light and shadows of digital reality. Living intelligence and its avatars
DOI:
https://doi.org/10.17072/2078-7898/2026-3-419-430Keywords:
Artificial Intelligence (AI), Large Language Models (LLMs), psychological support, chat-bot, prompt, prompt engineering, prompt for seeking psychological supportAbstract
Against the backdrop of rising psycho-emotional tension in society, artificial intelligence (AI) technologies are being actively integrated into the field of psychological support. The quality of interaction directly depends on the user’s textual query (prompt). Well-constructed prompts enable users to leverage AI to generate therapeutic content and receive psychological assistance. However, a contradiction exists: users who are most in need of psychological support are often unable to formulate a clear query, which increases the risk of receiving irrelevant responses. The aim of this study was to identify the key linguistic characteristics of prompts that influence the subjective evaluation of AI response usefulness in the context of psychological support. We analyzed 52 cases of user–AI interactions. The analysis showed that, overall, users tend to rate AI responses as helpful (mean score: 3.77 out of 5). Direct and specific queries, these formulated in an anxious or neutral tone, demonstrate the highest effectiveness. Conversely, implicit formulations and a tone of hopelessness are associated with significantly lower perceived usefulness. No significant effect on response evaluation was found for explicitly assigning the AI a “psychologist” role. In practical terms, our research findings can be used to develop guidelines aimed at teaching users effective prompt engineering when seeking psychological assistance. This would improve the quality, relevance, and safety of the psychological support received, particularly for users who experience difficulties in formulating their own needs.
References
Алексеева Л.Г., Алексеев П.С. Язык промптов, или особенности формулирования запросов к генеративным нейросетям для создания изображений // Verba. Северо-Западный лингвистический журнал. 2024. № 3(13). URL: https://cyberleninka.ru/article/n/yazyk-promptov-ili-osobennosti-formulirovaniya-zaprosov-k-generativnym-neyrosetyam-dlya-sozdaniya-izobrazheniy (дата обращения: 16.09.2025). DOI: https://doi.org/10.34680/VERBA-2024-3(13)-50-61 EDN: ZVOBBA
Комаров С.В., Внутских А.Ю. Свет и тени цифровой реальности. Образ инженера XXI века (от редакции) // Вестник Пермского университета. Философия. Психология. Социология. 2025. № 3. URL: https://cyberleninka.ru/article/n/svet-i-teni-tsifrovoy-realnosti-obraz-inzhenera-xxi-veka-ot-redaktsii (дата обращения: 31.05.2026). DOI: https://doi.org/10.17072/2078-7898/2025-3-309-316 EDN: ADMDFX
Ушаков Д.В. Искусственный интеллект как инструмент психологического исследования // Сибирский психологический журнал. 2023. № 90. URL: https://cyberleninka.ru/article/n/iskusstvennyy-intellekt-kak-instrument-psihologicheskogo-issledovaniya (дата обращения: 31.05.2026). DOI: https://doi.org/10.17223/17267080/90/11 EDN: FBMUGM
Фрейманис И.Ф., Шляпов И.В. Риски применения технологии искусственного интеллекта для психологической помощи: взгляд психолога и разработчика // Russian Journal of Education and Psychology. 2025. № 5. URL: https://cyberleninka.ru/article/n/riski-primeneniya-tehnologii-iskusstvennogo-intellekta-dlya-psihologicheskoy-pomoschi-vzglyad-psihologa-i-razrabotchika (дата обращения 30.05.2025). DOI: https://doi.org/10.12731/2658-4034-2025-16-5-829 EDN: WZCUOA
Aschbacher K. How to Build Empathetic AI // Medium. 2023. URL: https://medium.com/@kirstinashe/how-to-build-empathetic-ai-1bb444a8b3fe (дата обращения: 05.07.2026).
Ayers J., Poliak A., Dredze M. et al. Comparing Physician and Artificial Intelligence Chatbot Responses to Patient Questions Posted to a Public Social Media Forum // JAMA Internal Medicine. 2023. Vol. 183, № 6. P. 589–596. URL: https://each.international/wp-content/uploads/2023/05/jamainternal_ayers_2023_oi_230030_1681999216.70842.pdf (дата обращения 30.05.2025). DOI: https://doi.org/10.1001/jamainternmed.2023.1838 EDN: RFVQOH
Banerjee S., Dunn P., Conard S., Ali A. Mental Health Applications of Generative AI and Large Language Modeling in the United States // International Journal of Environmental Research and Public Health. 2024. Vol. 21, № 7. Art. 910. DOI: https://doi.org/10.3390/ijerph21070910 EDN: MXXOZC
Brown T., Mann B., Ryder N. et al. Language models are few-shot learners // Advances in neural information processing systems. 2020. Vol. 33. P. 1877–1901. DOI: https://doi.org/10.48550/arXiv.2005.14165
Filienko D., Wang Y., Jazmi C. et al. Toward Large Language Models as a Therapeutic Tool: Comparing Prompting Techniques to Improve GPT-Delivered Problem-Solving Therapy // arXiv. 2024. Art. 2409.00112. URL: https://arxiv.org/abs/2409.00112 (дата обращения: 17.09.2025).
Lawrence H., Schneider R., Rubin S. et al. The Opportunities and Risks of Large Language Models in Mental Health // JMIR Mental Health. 2024. Vol. 11. Art. e59479. DOI: https://doi.org/10.2196/59479 EDN: CNGJYM
Mesco B. Prompt Engineering as an Important Emerging Skill for Medical Professionals: Tutorial // Journal of Medical Internet Research. 2023. Vol. 25. Art. e50638. DOI: https://doi.org/10.2196/50638 EDN: KUUMXF
Olla P., Elliott L., Abumeeiz M., Mihelich K., Olson J. Promptology: Enhancing Human–AI Interaction in Large Language Models // Information. 2024. Vol. 15, № 10. Art. 634. DOI: https://doi.org/10.3390/info15100634 EDN: LYDEUK
Priyadarshana Y.H.P.P., Senanayake A., Liang Z., Piumarta I. Prompt engineering for digital mental health: a short review // Front. Digit. Health. 2024. Vol. 6. Art. 1410947. DOI: https://doi.org/10.3389/fdgth.2024.1410947
Ray A., Bhardwaj A., Malik Y., Singh S., Gupta R. Artificial intelligence and Psychiatry: An overview // Asian Journal of Psychiatry. 2022. Vol. 70. Art. 103021. DOI: https://doi.org/10.1016/j.ajp.2022.103021 EDN: GBXYZG
Sorin V., Brin D., Barash Y. et al. Large Language Models and Empathy: Systematic Review // J. Med. Internet. Res. 2024. Vol. 26. Art. e52597. DOI: https://doi.org/10.2196/52597 EDN: WVAUHI
Wang S., Fang R., He Z., Song S., Li Y. Emotional Support with LLM based Empathetic Dialogue Generation // arXiv. 2025. Art. arXiv:2507.12820. URL: https://arxiv.org/abs/2507.12820 (дата обращения: 16.09.2025).
References
Alekseeva, L.G. and Alekseev, P.S. (2024). [The language of prompts, or specifics of formulating queries to generative neural networks for image creation]. Verba. Severo-Zapadnyy lingvisticheskiy zhurnal [Verba. North-Western Linguistic Journal]. No. 3(13). Available at: https://cyberleninka.ru/article/n/yazyk-promptov-ili-osobennosti-formulirovaniya-zaprosov-k-generativnym-neyrosetyam-dlya-sozdaniya-izobrazheniy (accessed 16.09.2025). DOI: https://doi.org/10.34680/VERBA-2024-3(13)-50-61
Aschbacher, K. (2023). How to build empathetic AI. Medium. Available at: https://medium.com/@kirstinashe/how-to-build-empathetic-ai-1bb444a8b3fe (accessed 05.07.2026).
Ayers, J., Poliak, A., Dredze, M. et al. (2023). Comparing physician and Artificial Intelligence chatbot responses to patient questions posted to a public social media forum. JAMA Internal Medicine. Vol. 183, no. 6, pp. 589–596. Available at: https://each.international/wp-content/uploads/2023/05/jamainternal_ayers_2023_oi_230030_1681999216.70842.pdf (accessed 30.05.2025). DOI: https://doi.org/10.1001/jamainternmed.2023.1838
Banerjee, S., Dunn, P., Conard, S. and Ali, A. (2024). Mental health applications of generative AI and large language modeling in the United States. International Journal of Environmental Research and Public Health. Vol. 21, no. 7, art. 910. DOI: https://doi.org/10.3390/ijerph21070910
Brown, T., Mann, B., Ryder, N. et al. (2020). Language models are few-shot learners. Advances in Neural Information Processing Systems. Vol. 33, pp. 1877–1901. DOI: https://doi.org/10.48550/arXiv.2005.14165
Filienko, D., Wang, Y., Jazmi, C. et al. (2024). Toward large language models as a therapeutic tool: comparing prompting techniques to improve GPT-delivered problem-solving therapy. ArXiv. Art. 2409.00112. Available at: https://arxiv.org/abs/2409.00112 (accessed 17.09.2025).
Freimanis, I.F. and Shlyapov, I.V. (2025). [Risks of using Artificial Intelligence technology for psychological assistance: a psychologist's and developer's perspective]. Russian Journal of Education and Psychology. No. 5. Available at: https://cyberleninka.ru/article/n/riski-primeneniya-tehnologii-iskusstvennogo-intellekta-dlya-psihologicheskoy-pomoschi-vzglyad-psihologa-i-razrabotchika (accessed 30.05.2025). DOI: https://doi.org/10.12731/2658-4034-2025-16-5-829
Komarov, S.V. and Vnutskikh, A.Y. (2025). [Light and shadows of digital reality. The image of the 21st century engineer (an editorial)]. Vestnik Permskogo universiteta. Filosofiya. Psikhologiya. Sotsiologiya [Perm University Herald. Philosophy. Psychology. Sociology]. Iss. 3. Available at: https://cyberleninka.ru/article/n/svet-i-teni-tsifrovoy-realnosti-obraz-inzhenera-xxi-veka-ot-redaktsii (accessed 31.05.2026). DOI: https://doi.org/10.17072/2078-7898/2025-3-309-316
Lawrence, H., Schneider, R., Rubin, S. et al. (2024). the opportunities and risks of large language models in mental health. JMIR Mental Health. Vol. 11, art. e59479. DOI: https://doi.org/10.2196/59479
Mesco, B. (2023). Prompt engineering as an important emerging skill for medical professionals: tutorial. Journal of Medical Internet Research. Vol. 25, art. e50638. DOI: https://doi.org/10.2196/50638
Olla, P., Elliott, L., Abumeeiz, M., Mihelich, K. and Olson, J. (2024). Promptology: enhancing human–AI interaction in large language models. Information. Vol. 15, no. 10, art. 634. DOI: https://doi.org/10.3390/info15100634
Priyadarshana, Y.H.P.P., Senanayake, A., Liang, Z. and Piumarta, I. (2024). Prompt engineering for digital mental health: a short review. Frontiers in Digital Health. Vol. 6, art. 1410947. DOI: https://doi.org/10.3389/fdgth.2024.1410947
Ray, A., Bhardwaj, A., Malik, Y., Singh, S. and Gupta, R. (2022). Artificial intelligence and psychiatry: an overview. Asian Journal of Psychiatry. Vol. 70, art. 103021. DOI: https://doi.org/10.1016/j.ajp.2022.103021
Sorin, V., Brin, D., Barash, Y. et al. (2024). Large language models and empathy: systematic review. J. Med. Internet. Res. Vol. 26, art. e52597. DOI: https://doi.org/10.2196/52597
Ushakov, D.V. (2023). [Artificial Intelligence as a tool for psychological research]. Sibirskiy psikhologicheskiy zhurnal [Siberian Journal of Psychology]. No. 90. Available at: https://cyberleninka.ru/article/n/iskusstvennyy-intellekt-kak-instrument-psihologicheskogo-issledovaniya (accessed 31.05.2026). DOI: https://doi.org/10.17223/17267080/90/11
Wang, S., Fang, R., He, Z., Song, S. and Li, Y.(2025). Emotional support with LLM based empathetic dialogue generation. ArXiv. Art. 2507.12820. Available at: https://arxiv.org/abs/2507.12820 (accessed 16.09.2025).
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Perm University Herald. Philosophy. Psychology. Sociology
This work is licensed under a Creative Commons Attribution 4.0 International License.
