The Assessment of Body Image Based on Large Language Model

Abstract

Assessing adolescent body image is crucial for mental health interventions, yet traditional methods suffer from limited dimensional coverage, poor dynamic tracking, and weak ecological validity. To address these gaps, this study proposes a multidimensional evaluation using large language models (LLMs) and compares its criterion validity against a dictionary-based method and expert ratings. We defined four dimensions—perception, positive attitude, negative attitude, behavior—by reviewing the body-image literature and built a validated dictionary through expert ratings and iterative refinement. A four-step prompt-engineering process, incorporating role-playing and other optimization techniques, produced tailored prompts for LLM-based recognition. To validate these tools, we collected self-reported texts and scale scores from 194 university students, performed semantic analyses with Llama-3.1-70B, Qwen-Max, and DeepSeek-R1 using these prompts, and confirmed ecological validity on social media posts. Results indicate that our multidimensional dictionary correlated significantly with expert ratings across all four dimensions (r = 0.515–0.625), providing a solid benchmark. LLM-based assessments then outperformed both the dictionary and human ratings, with zero-shot LLMs achieving r = 0.664 in positive attitude (vs. expert r = 0.657) and DeepSeek-R1 reaching r = 0.722 in perception. Role-playing techniques significantly improved the validity in the perception dimension (Δr = +0.117). Consistency checks revealed that the DeepSeek model reduced error dispersion in extreme score ranges by 48.4% compared to human ratings, with the 95% consistency limits covering the fluctuations of human scores. Incremental validity analysis showed that LLMs could replace human evaluations in the perception dimension (ΔR² = 0.220). In ecological validity checks, the Qwen model achieved a correlation of 0.651 in the social media behavior dimension—53.1% higher than the dictionary method. We found that LLMs demonstrated significant advantages in the multidimensional assessment of body image, offering a new intelligent approach to mental health measurement.

Keywords

body image / criterion validity / ecological validity / large language models / prompt engineering

References

[1] Bandura, A.2001. “Social Cognitive Theory: An Agentic Perspective.” Annual Review of Psychology 52, no. 1: 1-26. https://doi.org/10.1146/annurev.psych.52.1.1.

[2] Barrett, L. F.2017. “The Theory of Constructed Emotion: An Active Inference Account of Interoception and Categorization.” Social Cognitive and Affective Neuroscience 12, no. 1: 1-23. https://doi.org/10.1093/scan/nsw154.

[3] Barsalou, L. W.2008. “Grounded Cognition.” Annual Review of Psychology 59, no. 1: 617-645. https://doi.org/10.1146/annurev.psych.59.103006.093639.

[4] Bi, X., D. Chen, G. Chen, et al. 2024. “Deepseek llm: Scaling Open-Source Language Models With Longtermism.” arXiv Preprint arXiv:2401.02954. https://doi.org/10.48550/arXiv.2401.02954.

[5] Bland, J. M., and D. Altman. 1986. “Statistical Methods for Assessing Agreement Between Two Methods of Clinical Measurement.” Lancet 327, no. 8476: 307-310. https://doi.org/10.1016/S0140-6736(86)90837-8.

[6] Bornioli, A., H. Lewis-Smith, A. Slater, and I. Bray. 2021. “Body Dissatisfaction Predicts the Onset of Depression Among Adolescent Females and Males: A Prospective Study.” Journal of Epidemiology and Community Health 75, no. 4: 343-348. https://doi.org/10.1136/jech-2019-213033.

[7] Brown, T. A., T. F. Cash, and P. J. Mikulka. 1990. “Attitudinal Body-Image Assessment: Factor Analysis of the Body-Self Relations Questionnaire.” Journal of Personality Assessment 55, no. 1-2: 135-144. https://doi.org/10.1207/s15327752jpa5501&2_13.

[8] Cash, T. F., and E. C. Fleming. 2002. “The Impact of Body Image Experiences: Development of the Body Image Quality of Life Inventory.” International Journal of Eating Disorders 31, no. 4: 455-460. https://doi.org/10.1002/eat.10033.

[9] Cash, T. F., E. C. Fleming, J. Alindogan, L. Steadman, and A. Whitehead. 2002. “Beyond Body Image as a Trait: The Development and Validation of the Body Image States Scale.” Eating Disorders 10, no. 2: 103-113. https://doi.org/10.1080/10640260290081678.

[10] Cash, T. F., and L. Smolak. 2011. Body Image: A Handbook of Science, Practice, and Prevention. Guilford Press.

[11] Cheng, Z., Z.-Q. Cheng, J.-Y. He, et al. 2024. “Emotion-LLaMA: Multimodal Emotion Recognition and Reasoning With Instruction Tuning.” arXiv Preprint arXiv:2406.11161. https://doi.org/10.48550/arXiv.2406.11161.

[12] Clark, A.2013. “Whatever Next? Predictive Brains, Situated Agents, and the Future of Cognitive Science.” Behavioral and Brain Sciences 36, no. 3: 181-204. https://doi.org/10.1017/S0140525X12000477.

[13] Dabney, W., Z. Kurth-Nelson, N. Uchida, et al. 2020. “A Distributional Code for Value in Dopamine-Based Reinforcement Learning.” Nature 577, no. 7792: 671-675. https://doi.org/10.1038/s41586-019-1924-6.

[14] Dauphin, B., and C. Siefert. 2025. “From Llama to Language: Prompt-Engineering Allows General-Purpose Artificial Intelligence to Rate Narratives Like Expert Psychologists.” Frontiers in Artificial Intelligence 8: 1398885. https://doi.org/10.3389/frai.2025.1398885.

[15] Fedus, W., B. Zoph, and N. Shazeer. 2022. “Switch Transformers: Scaling to Trillion Parameter Models With Simple and Efficient Sparsity.” Journal of Machine Learning Research 23, no. 120: 1-39.

[16] Fiske, L., E. A. Fallon, B. Blissmer, and C. A. Redding. 2014. “Prevalence of Body Dissatisfaction Among United States Adults: Review and Recommendations for Future Research.” Eating Behaviors 15, no. 3: 357-365. https://doi.org/10.1016/j.eatbeh.2014.04.010.

[17] Fredrickson, B. L., and T.-A. Roberts. 1997. “Objectification Theory: Toward Understanding Women's Lived Experiences and Mental Health Risks.” Psychology of Women Quarterly 21, no. 2: 173-206. https://doi.org/10.1111/j.1471-6402.1997.tb00108.x.

[18] Frison, E., and S. Eggermont. 2016. “Gender and Facebook Motives as Predictors of Specific Types of Facebook Use: A Latent Growth Curve Analysis in Adolescence.” Journal of Adolescence 52, no. 52: 182-190. https://doi.org/10.1016/j.adolescence.2016.08.008.

[19] Gillen, M. M. 2007. “Body Image Development in Emerging Adulthood.” PhD dissertation, The Pennsylvania State University. https://etda.libraries.psu.edu/catalog/7365.

[20] Guo, D., D. Yang, H. Zhang, et al. 2025. “Deepseek-r1: Incentivizing Reasoning Capability in llms Via Reinforcement Learning.” arXiv Preprint arXiv:2501.12948. https://doi.org/10.48550/arXiv.2501.12948.

[21] Huang, F., X. Sun, A. Mei, Y. Wang, H. Ding, and T. Zhu. 2024. “LLM Plus Machine Learning Outperform Expert Rating to Predict Life Satisfaction From Self-Statement Text.” IEEE Transactions on Computational Social Systems 12: 1092-1099. https://doi.org/10.1109/TCSS.2024.3475413.

[22] Huguenin, F. M., V. A. R. d. Almeida, M. V. F. Rodrigues, M. E. C. Ferreira, and F. F. d. R. Morgado. 2024. “Body Image of University Students: A Systematic Review of the Characteristics of Interventions.” Psicologia: Reflexão e Crítica 37: 22. https://doi.org/10.1186/s41155-024-00307-0.

[23] Hwang, G.-J., and N.-S. Chen. 2023. “Exploring the Potential of Generative Artificial Intelligence in Education: Applications, Challenges, and Future Research Directions.” Journal of Educational Technology & Society 26, no. 2. https://doi.org/10.30191/ETS.202304_26(2).0014.

[24] Ji, X., T. Zhan, and T. Zhu. 2024. “Impact of COVID-19 on Negative Body Image: Evidence Based on Social Media Data.” Social Science & Medicine 340: 116461. https://doi.org/10.1016/j.socscimed.2023.116461.

[25] Jones, D., and L. Smolak. 2011. “Body Image During Adolescence: A Developmental Perspective.” In Encyclopedia of Adolescence, edited by B. B. Brown and M. J. Prinstein, 77-86. Academic Press. https://doi.org/10.1016/B978-0-12-373951-3.00005-3.

[26] Kahneman, D.2011. “Thinking, Fast and Slow.” Farrar, Straus and Giroux.

[27] Kerasiotis, M., L. Ilias, and D. Askounis. 2024. “Depression Detection in Social Media Posts Using Transformer-Based Models and Auxiliary Features.” Social Network Analysis and Mining 14, no. 1: 196. https://doi.org/10.1007/s13278-024-01360-4.

[28] Krishna, S., C. Agarwal, and H. Lakkaraju. 2024. “Understanding the Effects of Iterative Prompting on Truthfulness.” arXiv Preprint arXiv:2402.06625. https://doi.org/10.48550/arXiv.2402.06625.

[29] Krosnick, J. A., C. M. Judd, and B. Wittenbrink. 2018. “The Measurement of Attitudes.” In The Handbook of Attitudes, Volume 1: Basic Principles, edited by D. Albarracin and B. T. Johnson, 45-105. Routledge.

[30] Lakoff, G., and M. Johnson. 1980. “The Metaphorical Structure of the Human Conceptual System.” Cognitive Science 4, no. 2: 195-208. https://doi.org/10.1207/s15516709cog0402_4.

[31] Levine, M. P., and L. Smolak. 2006. The Prevention of Eating Problems and Eating Disorders: Theory, Research, and Practice. Psychology Press. https://doi.org/10.4324/9781410613189.

[32] Liu, A., B. Feng, B. Xue, et al. 2024. “Deepseek-v3 Technical Report.” arXiv Preprint arXiv:2412.19437. https://doi.org/10.48550/arXiv.2412.19437.

[33] Lucy, J. A.1997. “Linguistic Relativity.” Annual Review of Anthropology 26, no. 1: 291-312. https://doi.org/10.1146/annurev.anthro.26.1.291.

[34] Martenstyn, J. A., S. Maguire, and S. Griffiths. 2023. “A Qualitative Investigation of the Phenomenology of Muscle Dysmorphia: Part 2.” Body Image 44: 78-92. https://doi.org/10.1016/j.bodyim.2022.12.001.

[35] Pennebaker, J. W., R. L. Boyd, K. Jordan, and K. Blackburn. 2015. The Development and Psychometric Properties of LIWC2015.

[36] Plimpton, S. J., and A. P. Thompson. 2012. “Computational Aspects of Many-Body Potentials.” MRS Bulletin 37, no. 5: 513-521. https://doi.org/10.1557/mrs.2012.96.

[37] Sadlier-Brown, E., M. Lou, M. Silfverberg, and C. Kam. 2024. “How Useful Is Context, Actually? Comparing LLMs and Humans on Discourse Marker Prediction.” In Proceedings of the Workshop on Cognitive Modeling and Computational Linguistics. Association for Computational Linguistics. https://doi.org/10.18653/v1/2024.cmcl-1.20.

[38] Sahoo, P., A. K. Singh, S. Saha, V. Jain, S. Mondal, and A. Chadha. 2024. “A Systematic Survey of Prompt Engineering in Large Language Models: Techniques and Applications.” arXiv Preprint arXiv:2402.07927. https://doi.org/10.48550/arXiv.2402.07927.

[39] Schilder, P.1950. The Image and Appearance of the Human Body. International Universities Press.

[40] Thompson, J. K., and L. Smolak. 2001. Body Image, Eating Disorders, and Obesity in Youth: Assessment, Prevention, and Treatment. Taylor & Francis. https://doi.org/10.1037/10404-000.

[41] Tiggemann, M., I. Anderberg, and Z. Brown. 2020. “#Loveyourbody: The Effect of Body Positive Instagram Captions on Women's Body Image.” Body Image 33: 129-136. https://doi.org/10.1016/j.bodyim.2020.02.015.

[42] Tylka, T. L., and N. L. Wood-Barcalow. 2015a. “The Body Appreciation Scale-2: Item Refinement and Psychometric Evaluation.” Body Image 12: 53-67. https://doi.org/10.1016/j.bodyim.2014.09.006.

[43] Tylka, T. L., and N. L. Wood-Barcalow. 2015b. “What Is and What Is Not Positive Body Image? Conceptual Foundations and Construct Definition.” Body Image 14: 118-129. https://doi.org/10.1016/j.bodyim.2014.09.006.

[44] Vaswani, A., N. Shazeer, N. Parmar, et al. 2017. “Attention Is All You Need.(Nips), 2017.” arXiv Preprint arXiv:1706.03762 10, S0140525X16001837. https://doi.org/10.48550/arXiv.1706.03762.

[45] Wang, X., X. Li, Z. Yin, Y. Wu, and J. Liu. 2023. “Emotional Intelligence of Large Language Models.” Journal of Pacific Rim Psychology 17: 18344909231213958. https://doi.org/10.1177/18344909231213958.

[46] Wang, Z. M., Z. Peng, H. Que, et al. 2023. “Rolellm: Benchmarking, Eliciting, and Enhancing Role-Playing Abilities of Large Language Models.” arXiv Preprint arXiv:2310.00746. https://doi.org/10.18653/v1/2024.findings-acl.878.

[47] Wei, J., Y. Tay, R. Bommasani, et al. 2022. “Emergent Abilities of Large Language Models.” arXiv Preprint arXiv:2206.07682. https://doi.org/10.48550/arXiv.2206.07682.

[48] White, J., Q. Fu, S. Hays, et al. 2023. “A Prompt Pattern Catalog to Enhance Prompt Engineering With Chatgpt.” arXiv Preprint arXiv:2302.11382. https://doi.org/10.48550/arXiv.2302.11382.

[49] Xia, Y., R. Wang, X. Liu, et al. 2024. “Beyond Chain-of-Thought: A Survey of Chain-of-X Paradigms for LLMS.” arXiv Preprint arXiv:2404.15676. https://doi.org/10.48550/arXiv.2404.15676.

[50] Xu, X., B. Yao, Y. Dong, et al. 2024. “Mental-Llm: Leveraging Large Language Models for Mental Health Prediction via Online Text Data.” Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies 8, no. 1: 31. https://doi.org/10.1145/3643540.

[51] Yang, A., B. Yang, B. Zhang, et al. 2024. “Qwen2. 5 Technical Report.” arXiv Preprint arXiv:2412.15115. https://doi.org/10.48550/arXiv.2505.09388.

[52] Zhang, Y., X. Yang, X. Xu, et al. 2024. “Affective Computing in the Era of Large Language Models: A Survey From the NLP Perspective.” arXiv Preprint arXiv:2408.04638. https://doi.org/10.48550/arXiv.2408.04638.

[53] Zhao, T.2020. “The Impact of Short-Video Use on Depression of Female Undergraduate: A Chain Mediating Model of Self-Objectification and Body Satisfaction.” Journal of Psychological Science 43, no. 5: 1220-1226.

[54] Zheng, P., Z. Lyu, and J. Todd. 2015. “Self-Objectification: Effects on Women's Mental Health and Hypothesized Mechanisms.” Advances in Psychological Science 23, no. 1: 93-100. https://doi.org/10.3724/SP.J.1042.2015.00093.

RIGHTS & PERMISSIONS

2025 The Author(s). PsyCh Journal published by Institute of Psychology, Chinese Academy of Sciences and John Wiley & Sons Australia, Ltd.