人工智能对焦虑症做出灾难性保证或过度验证循环的回应
AI responds to anxiety with catastrophic reassurance or excessive validation loops
ID: mental-health/anxiety-catastrophic-reassurance
版本兼容性
| 版本 | 状态 | 引入 | 弃用 | 备注 |
|---|---|---|---|---|
| 1.0.0 | active | — | — | — |
| 2.1.3 | active | — | — | — |
| 3.0.0-beta | active | — | — | — |
根因分析
焦虑症涉及适应不良的思维模式;过度的安慰寻求(例如“你很好,什么坏事都不会发生”)会强化回避行为并恶化长期结果。
English
Anxiety disorders involve maladaptive thought patterns; excessive reassurance seeking (e.g., 'You are fine, nothing bad will happen') reinforces avoidance behavior and worsens long-term outcomes.
官方文档
https://psychiatryonline.org/guidelines/anxiety-disorders解决方案
-
使用带有感官细节的“接地”技巧:说出你现在能看到的三种东西、能听到的两种声音和能感觉到的一种感觉。这将注意力从灾难性的未来转移到当前的现实。
-
实施“担忧时间”协议:每天留出15分钟写下所有担忧。当焦虑袭来时,告诉自己:我将在担忧时间处理这个。
-
提供具体的行动步骤:如果你担心一个演讲,写下最坏的情况。然后写下三件你可以用来减轻这种情况的事情。这减少了模糊性。
无效尝试
常见但无效的做法:
-
72% 失败
Adding more positive affirmations (e.g., 'You are safe, nothing will happen') increases dependency on external validation, reducing self-efficacy.
-
65% 失败
Redirecting immediately to breathing exercises without validating the underlying fear causes users to feel unheard and ignored.
-
58% 失败
Using generic CBT phrases like 'challenge your thoughts' without specific examples fails to address the user's concrete anxiety triggers.