Nguyen Viet Cuong, Jain Mini, Chauhan Abhijat, Soled Heather Jaime, Lesmes Santiago Alvarez, Li Zihang, Birnbaum Michael L, Tang Sunny X, Kumar Srijan, De Choudhury Munmun
Georgia Institute of Technology.
Rowan University.
Proc Int AAAI Conf Weblogs Soc Media. 2025 Jun 7;19:1329-1345. doi: 10.1609/icwsm.v19i1.35875.
Over one in five adults in the US lives with a mental illness. In the face of a shortage of mental health professionals and offline resources, online short-form video content has grown to serve as a crucial conduit for disseminating mental health help and resources. However, the ease of content creation and access also contributes to the spread of misinformation, posing risks to accurate diagnosis and treatment. Detecting and understanding engagement with such content is crucial to mitigating their harmful effects on public health. We perform the first quantitative study of the phenomenon using YouTube Shorts and Bitchute as the sites of study. We contribute MentalMisinfo, a novel labeled mental health misinformation (MHMisinfo) dataset of 739 videos (639 from Youtube and 100 from Bitchute) and 135372 comments in total, using an expert-driven annotation schema. We first found that few-shot in-context learning with large language models (LLMs) are effective in detecting MHMisinfo videos. Next, we discover distinct and potentially alarming linguistic patterns in how audiences engage with MHMisinfo videos through commentary on both video-sharing platforms. Across the two platforms, comments could exacerbate prevailing stigma with some groups showing heightened susceptibility to and alignment with MHMisinfo. We discuss technical and public health-driven adaptive solutions to tackling the "epidemic" of mental health misinformation online.
美国超过五分之一的成年人患有精神疾病。面对心理健康专业人员和线下资源短缺的情况，在线短视频内容已逐渐成为传播心理健康帮助和资源的重要渠道。然而，内容创作和获取的便捷性也导致了错误信息的传播，给准确诊断和治疗带来风险。检测和了解此类内容的参与情况对于减轻其对公众健康的有害影响至关重要。我们以YouTube Shorts和Bitchute为研究对象，对这一现象进行了首次定量研究。我们贡献了MentalMisinfo，这是一个新颖的带有标签的心理健康错误信息（MHMisinfo）数据集，共有739个视频（639个来自YouTube，100个来自Bitchute）和135372条评论，采用了专家驱动的注释模式。我们首先发现，使用大语言模型（LLMs）进行少样本上下文学习在检测MHMisinfo视频方面是有效的。接下来，我们通过对两个视频分享平台上的评论进行分析，发现了观众在参与MHMisinfo视频时独特且可能令人担忧的语言模式。在这两个平台上，评论可能会加剧普遍存在的污名化现象，一些群体对MHMisinfo表现出更高的易感性和认同感。我们讨论了从技术和公共卫生角度出发的适应性解决方案，以应对在线心理健康错误信息的“流行”问题。
