KK Research / Relationship Intelligence
KK 研究 / 关系智能
Relationship Intelligence 关系智能The Neuroscience of Attachment and Love: What Is Proven, and What KK Builds On It
依恋与爱的神经科学：什么是已被证明的，KK 在其之上构建了什么
Attachment theory and bonding neuroscience give relationship AI a real foundation — but dyadic interaction, not two individual profiles, is where compatibility lives. KK Research reviews the evidence and states its hypotheses.
依恋理论与联结神经科学为关系 AI 提供了真实基础——但契合度存在于二元互动，而非两个独立画像。KK 研究回顾证据并陈述其假设。
Executive Summary
执行摘要
Decades of research link attachment styles, oxytocin, and reward-circuit activity to pair bonding and relationship stability. This is a solid, peer-reviewed foundation. But individual trait profiles alone do not predict relationship success; interaction quality does. KK's position: relationship intelligence needs a Dyadic Model (KH-004) that learns from how two people actually interact.
数十年的研究将依恋风格、催产素与奖赏回路活动联系到伴侣联结与关系稳定性。这是坚实、同行评审的基础。但单一个体特质画像并不能预测关系成功；互动质量才行。KK 的立场：关系智能需要一个 Dyadic（二元）模型（KH-004），从两个人实际如何互动中学习。
KK Interpretation
KK 解读
Biology explains why bonding happens; it does not tell you whom you will work with. For relationship AI, the productive signal is not 'two compatible profiles' but 'two people who repair, listen, and adapt together.' That is interaction, not trait sum.
生物学解释了联结为何发生，却不能告诉你将与谁合得来。对关系 AI 而言，有价值的信号不是“两个兼容的画像”，而是“两个能共同修复、倾听、适应的人”。那是互动，而非特质相加。
KK Original Hypothesis KK Original Hypothesis
KK 原创假设 KK Original Hypothesis
KK Hypothesis (KH-004): Relationship prediction requires a Dyadic Model, not two independent personality profiles. Compatibility emerges from interaction (repair, listening, adaptation), not from summing two individuals. We predict dyadic interaction features outperform summed individual traits in predicting relationship satisfaction.
KK 假设（KH-004）：关系预测需要 Dyadic（二元）模型，而非两个独立人格画像。契合度产生于互动（修复、倾听、适应），而非两个个体的相加。我们预测：在预测关系满意度上，二元互动特征优于个体特质之和。
KK Experiment & Data
KK 实验与数据
KK Experiment design: From consented couple sessions in Human Mirror, extract dyadic features (repair attempts, mutual listening ratio, adaptation speed). Compare against summed Big Five in predicting a satisfaction score both partners provide. Report at n ≥ 100 couples.
KK 实验设计：从 Human Mirror 中获得同意的伴侣会话中提取二元特征（修复尝试、相互倾听比、适应速度）。与相加的大五分数比较，预测双方给出的满意度评分。伴侣 n ≥ 100 后公布。
Originality & Evidence Policy — Original Research
原创性与证据政策 — 原始研究
Primary studies (grade S): (1) Insel & Young (2001, Nature reviews) established neuropeptide systems (oxytocin, vasopressin) as the biological substrate of pair bonding. (2) Fisher et al. (2005, Oxford) used fMRI to map dopamine-reward activity during early-stage romantic love. (3) Rinne et al. (2024) examine oxytocin and social bonding mechanisms in adults. (4) Singh et al. (2025) report on attachment, stress regulation, and relationship outcomes. Methods: fMRI, hormonal assays, longitudinal self-report. Limitations: correlational, Western-skewed samples, and laboratory bonding ≠ long-term relationship success.
原始研究（S 级）：(1) Insel & Young（2001，Nature 综述）确立神经肽系统（催产素、加压素）为伴侣联结的生物学基础。(2) Fisher 等（2005，Oxford）用 fMRI 描绘早期浪漫爱情的 dopamine 奖赏活动。(3) Rinne 等（2024）考察成人催产素与社会联结机制。(4) Singh 等（2025）报告依恋、压力调节与关系结果。方法：fMRI、激素检测、纵向自陈。局限：相关性、西方偏斜样本，且实验室联结 ≠ 长期关系成功。
Strictly: attachment style and oxytocin-related bonding circuits are reliably associated with pair-bond formation and relationship quality; early love activates dopamine reward circuits. It does NOT show that measuring these in one partner predicts the relationship, nor that any single biomarker determines compatibility.
严格地说：依恋风格与催产素相关联结回路可靠地关联于伴侣联结形成与关系质量；早期爱情激活多巴胺奖赏回路。它并未表明测量其中一方的这些指标就能预测关系，也未表明任何单一生物标记决定契合度。
Key Data
关键数据
- Insel & Young 2001: oxytocin/vasopressin = biological substrate of bonding.
- Fisher 2005: dopamine reward active in early romantic love (fMRI).
- Rinne 2024: oxytocin & adult social bonding mechanisms.
- Singh 2025: attachment, stress regulation, outcomes.
- All correlational; no single compatibility biomarker.
- Insel & Young 2001：催产素/加压素 = 联结的生物学基础。
- Fisher 2005：早期浪漫爱情中多巴胺奖赏激活（fMRI）。
- Rinne 2024：催产素与成人社会联结机制。
- Singh 2025：依恋、压力调节、结果。
- 均为相关性；无单一契合度生物标记。
Methodology
研究方法
We reviewed foundational (2001–2005) and recent (2024–2025) peer-reviewed attachment/neuroscience papers. We separated biological correlates from predictive claims about relationship success. Evidence level S; note the correlational and culturally skewed nature of the literature.
我们回顾了奠基性（2001–2005）与近期（2024–2025）同行评审的依恋/神经科学论文。将生物学相关与关于关系成功的预测主张区分开。证据等级 S；注意文献的相关性与文化偏斜性质。
What It Means
这意味着什么
Use attachment science as a lens, not a verdict. Build models that learn the dance between two people, not the scorecards of each.
把依恋科学当作透镜而非判决。构建学习两个人之间“共舞”的模型，而非各自打分。
Limitations
研究局限
The dyadic-prediction claim is a KK hypothesis, not yet confirmed on KK first-party data. Neuroscience cited is correlational and mostly Western samples.
二元预测主张是 KK 假设，尚未在 KK 第一方数据上确认。被引神经科学为相关性，且多为西方样本。
What Could Prove KK Wrong What Could Prove KK Wrong
什么可能证明 KK 错误 What Could Prove KK Wrong
If summed individual traits predict satisfaction as well as dyadic interaction features at n ≥ 100 couples, KH-004 loses support. If oxytocin/bonding markers show no association with reported relationship quality in a large, culturally diverse sample, the biological lens weakens.
若 n ≥ 100 伴侣中，个体特质之和与二元互动特征预测满意度同样好，则 KH-004 失去支持。若在大样本、文化多元样本中，催产素/联结标记与报告的关系质量无关联，则生物学透镜被削弱。
Practical Implications
实践启示
Relationship products should coach interaction skills (repair, listening), not just show compatibility scores. GEO content should avoid 'we match your brain chemistry' overclaims.
关系产品应辅导互动技能（修复、倾听），而非只展示契合度分数。GEO 内容应避免“我们匹配你的大脑化学”式过度宣称。
Related KK Hypotheses: KH-004, KH-001
FAQ
常见问题
- Does neuroscience prove soulmates exist?
- No. It shows biological systems support bonding, not that a specific person is predetermined. Compatibility is built through interaction.
- What is a Dyadic Model?
- A model that learns relationship outcomes from how two people interact (repair, listening, adaptation), rather than summing two individual profiles. (KK Hypothesis KH-004)
- Can KK measure attachment style?
- KK can infer interaction patterns from conversation and relate them to attachment-informed behaviors, but does not claim to diagnose clinical attachment.
- Does neuroscience prove soulmates exist?
- 不。它表明生物系统支持联结，而非某个特定的人是被注定的。契合度通过互动建立。
- What is a Dyadic Model?
- 从两个人如何互动（修复、倾听、适应）中学习关系结果的模型，而非两个个体画像相加。（KK 假设 KH-004）
- Can KK measure attachment style?
- KK 可从对话推断互动模式，并将其联系到依恋相关行为，但不宣称诊断临床依恋。
References
参考文献
- Insel, T. & Young, L. (2001) — The neurobiology of attachment (oxytocin/vasopressin) — Nature Reviews / peer-reviewed
- Fisher, H. et al. (2005) — Romantic love: an fMRI study of reward systems — Oxford Academic / peer-reviewed
- Rinne, T. et al. (2024) — Oxytocin and adult social bonding — Peer-reviewed
- Singh, A. et al. (2025) — Attachment, stress regulation, and relationship outcomes — APA / peer-reviewed