满月的光只有0.3勒克斯:月光能抑制褪黑素吗?一场可以被证伪的实验

Written by

in

Language

Evidence level: B

满月夜室外照度仅 0.1-0.3 勒克斯,而人类褪黑素抑制在约 1.5 melanopic 勒克斯才触发——”月相影响睡眠”链条上最薄弱的一环,是一个从未被直接做过、但完全可以做的实验。

正文

“月圆之夜睡不着”这句民间经验,近年来被几项流行病学研究重新提起:2013 年《Current Biology》报告满月前后入睡时间延长 5 分钟、总睡眠减少 20 分钟、深睡期脑电活动下降 30%(PMID 23891110)。但这个关联一直有个物理学上说不通的尴尬:满月照到地面的光,满打满算只有 0.1-0.3 勒克斯——比一根蜡烛还暗几十倍。这么弱的光,凭什么穿透眼皮、驱动视网膜里的感光细胞、再去压掉松果体的褪黑素?本文要回答的,正是这个可以被实验直接裁决的问题。

先看生理端已有的证据。人类对光的”非视觉”响应(褪黑素抑制、昼夜节律相移)主要由视网膜上一类自带感光色素 melanopsin 的 ipRGC 细胞驱动。2019 年《J Pineal Res》对 Brainard 数据集的重建给出了一条明确的剂量-反应曲线:褪黑素抑制的触发点在约 1.5 melanopic lux(melanopsin 加权照度,约 0.2 µW/cm²),饱和点在 305 melanopic lux(PMID 30697806)。更早的经典实验甚至报告过,在严格控制暴露条件下,低于 1 lux 的单色光就能显著抑制夜间褪黑素(PMID 12163843);Brainard 本人 1997 年的综述也确认,当光谱与暴露时间优化后,≤200 lux 即可急性抑制褪黑素或推动节律相移(PMID 9406028)。另一组 1998 年的唾液实验测出 30-120 分钟暴露下的最小有效强度在 285-393 lux 之间(PMID 9756329)——注意这用的是白光,而剂量-反应曲线对光谱成分高度敏感。

现在把物理端和生理端放在同一根轴上换算。满月照度 0.1-0.3 lux 是photopic(人眼明视觉加权)照度,而褪黑素抑制看的是 melanopic(melanopsin 加权)照度——两者不是一回事。月光本质是反射太阳光,光谱偏蓝,melanopic 比例并不低,但即便按最乐观估计,0.3 photopic lux 的月光换算成 melanopic 也只在 0.1-0.3 之间,比 1.5 melanopic lux 的触发阈值低了一个数量级。再加上眼睑的遮挡(闭眼时透光率仅约 3-10%)和室内窗帘衰减,实际到达视网膜的光剂量还要再掉一截。按现有剂量-反应曲线外推,”满月光抑制褪黑素”在生理上似乎不成立——但注意,这只是外推,不是实验。

这条链上还有两个让外推失效的变量。其一是光历史与暗适应:在持续昏暗背景(约 0.5 lux)下待 15 小时后,同一个 200 lux 光脉冲引发的褪黑素抑制显著强于在 200 lux 背景下的反应(PMC2717723 综述系统梳理了该效应)——也就是说,睡眠前的暗环境可能把个体的光敏感度顶上去,让阈值下移。其二是个体差异:同是弱光,有人 ipRGC 功能强(可用 PIPR 瞳孔反应无创测量,PMID 30809186)、有人褪黑素受体敏感度不同(来源文章讨论的 MTNR1B 基因型),个体间敏感度可能差 5-10 倍。

所以缺口很具体:“月光光谱 + 0.1-0.3 lux 梯度 + 暗适应后的真实人类”这个组合,从没有人做过实验。 现有文献要么用白光测阈值(数百 lux 级),要么用单色光测极低照度(<1 lux 蓝光,但光谱是纯 446-484nm,与月光全谱不同),中间恰好在”月光这一格”是空的。这也是它可贵之处:一个完全可证伪、成本可控的实验就能填上。

实验设计并不复杂。暗室可控条件:以 LED 复现月光光谱(全谱+蓝峰),设置 0.1 / 0.2 / 0.3 / 0.5 / 1.0 / 3.0 lux 六个照度梯度(含一个明确超过阈值的阳性对照),每位受试者在标准暗适应(≥2 小时 <0.5 lux)后接受夜间光暴露,每小时唾液褪黑素连续采样,辅以瞳孔 PIPR 分型作为 ipRGC 功能协变量。结局是两个预设问题:① 0.1-0.3 lux 区间内是否存在可测的褪黑素抑制(哪怕 5%);② 若存在,是否与 PIPR 强度、MTNR1B 基因型交互。若前 60 分钟在最高月照强度(0.3 lux)下抑制量无法与安慰剂区分,则”月相-睡眠”流行病学关联的光通路解释被直接削弱,转而支持社会行为/噪音/气压等其他解释——这就是证伪的姿势。

证据与局限

已验证: 人类褪黑素抑制由 ipRGC/melanopsin 驱动,触发阈值约 1.5 melanopic lux(PMID 30697806);<1 lux 单色蓝光在严格条件下可抑制褪黑素(PMID 12163843);≤200 lux 优化光谱可急性抑制(PMID 9406028);白光最低有效强度数百 lux 级(PMID 9756329);暗适应提高光敏感度(PMC2717723);月相-睡眠流行病学关联存在但效应微小(PMID 23891110)。

部分验证: 暗适应/光历史对低照度抑制阈值的调制幅度;个体 ipRGC 功能差异(PIPR)与褪黑素抑制敏感度的定量关系(PMID 30809186、38348477)。

待验证: 月光光谱(非单色、非白光)在 0.1-0.3 lux 照度段能否驱动可测褪黑素抑制——尚无直接实验;闭眼+窗帘后实际视网膜光剂量是否达到触发阈值的换算验证。

相关阅读

证据等级: B(科学类比/间接证据——剂量-反应外推+单色光先例,但”月光光谱×0.1-0.3 lux”直接实验缺失)
参考文献: PMID 30697806, PMID 12163843, PMID 9406028, PMID 9756329, PMID 23891110, PMID 30809186, PMID 38348477


免责说明:本文属传统知识与科学文献的整理、对照与研究设计讨论,仅供理解与参考,不构成医疗建议,不能替代专业诊断与治疗。文中涉及的效应量、频率区间与暴露结论多为研究报告层面的推断,个体情况差异很大;如有身体不适或健康疑虑,请就医并由专业医师判断。

Moonlight at 0.3 lux: Can it suppress melatonin? A falsifiable experiment

Evidence level: B

Full moon outdoor illuminance is only 0.1–0.3 lux, whereas human melatonin suppression triggers at approximately 1.5 melanopic lux. The weakest link in the “lunar phase affects sleep” chain is an experiment that has never been conducted directly but could easily be performed.

Main Text

The folk saying “cannot sleep on a full moon” has recently been revisited by several epidemiological studies: a 2013 report in Current Biology noted that sleep onset time increased by 5 minutes, total sleep decreased by 20 minutes, and deep sleep EEG activity dropped by 30% around the full moon (PMID 23891110). However, this association has always had a physically inexplicable awkwardness: the light from a full moon hitting the ground totals only 0.1–0.3 lux—decades dimmer than a candle. How can such weak light penetrate the eyelids, drive photoreceptive cells in the retina, and subsequently suppress melatonin in the pineal gland? This article aims to answer this question, which can be directly adjudicated by experiment.

First, let’s examine the existing physiological evidence. Human “non-visual” light responses (melatonin suppression, circadian phase shifts) are primarily driven by ipRGC cells in the retina that possess their own light-sensitive pigment, melanopsin. A 2019 reconstruction of the Brainard dataset in J Pineal Res provided a clear dose-response curve: the trigger point for melatonin suppression is at approximately 1.5 melanopic lux (melanopsin-weighted illuminance, ~0.2 µW/cm²), with saturation at 305 melanopic lux (PMID 30697806). Earlier classic experiments even reported that under strictly controlled exposure conditions, monochromatic light below 1 lux could significantly suppress nocturnal melatonin (PMID 12163843); Brainard himself confirmed in his 1997 review that with optimized spectral composition and exposure duration, ≤200 lux could acutely suppress melatonin or drive circadian phase shifts (PMID 9406028). Another group of saliva experiments from 1998 measured a minimum effective intensity between 285–393 lux under 30–120 minutes of exposure (PMID 9756329)—note this used white light, while the dose-response curve is highly sensitive to spectral composition.

Now, let’s convert both physical and physiological endpoints onto the same axis. Full moon illuminance of 0.1–0.3 lux is photopic (human eye photopic weighting), whereas melatonin suppression relies on melanopic (melanopsin-weighted) illuminance—these are not equivalent. Moonlight is essentially reflected sunlight with a blue-skewed spectrum, so the melanopic proportion is not low; however, even under the most optimistic estimates, 0.3 photopic lux of moonlight converts to only 0.1–0.3 melanopic, which is one order of magnitude lower than the 1.5 melanopic lux trigger threshold. Additionally, eyelid obstruction (transmittance is only ~3–10% when eyes are closed) and indoor curtain attenuation further reduce the actual light dose reaching the retina. Based on extrapolation from existing dose-response curves, “full moon light suppressing melatonin” seems physiologically unfeasible—but note, this is extrapolation, not experimental verification.

There are two variables on this chain that invalidate simple extrapolation. First is light history and dark adaptation: after 15 hours in a continuously dim background (~0.5 lux), the same 200 lux light pulse induces significantly stronger melatonin suppression than in a 200 lux background (PMC2717723 provides a systematic review of this effect)—meaning the dark environment before sleep may boost individual light sensitivity and lower the threshold. Second is individual differences: under weak light, some individuals have stronger ipRGC function (measurable non-invasively via PIPR pupil response, PMID 30809186), while others have different melatonin receptor sensitivities (discussed MTNR1B genotypes in source articles), with inter-individual sensitivity potentially differing by 5–10 times.

Thus, the gap is specific: no one has ever conducted an experiment on the combination of “moonlight spectrum + 0.1–0.3 lux gradient + real humans after dark adaptation.” Existing literature either uses white light to measure thresholds (hundreds of lux) or monochromatic light at extremely low illuminance (<1 lux blue light, but the spectrum is pure 446–484 nm, differing from moonlight’s full spectrum). There is precisely a gap at “the moonlight slot.” This is also its value: a completely falsifiable, cost-controlled experiment could fill it.

The experimental design is not complex. Controlled dark room conditions: replicate moonlight spectrum using LED (full spectrum + blue peak), set six illuminance gradients of 0.1 / 0.2 / 0.3 / 0.5 / 1.0 / 3.0 lux (including one clearly exceeding the threshold as a positive control). Each subject undergoes nighttime light exposure after standard dark adaptation (≥2 hours <0.5 lux), with continuous saliva melatonin sampling every hour, supplemented by PIPR pupil response typing as an ipRGC function covariate. The outcomes address two preset questions: ① Is there measurable melatonin suppression within the 0.1–0.3 lux range (even if only 5%); ② If yes, does it interact with PIPR intensity and MTNR1B genotype. If suppression under the highest moonlight intensity (0.3 lux) cannot be distinguished from placebo in the first 60 minutes, the photopathway explanation for the “lunar phase–sleep” epidemiological association is directly weakened, shifting support to alternative explanations such as social behavior/noise/atmospheric pressure—this is the posture of falsification.

Evidence and Limitations

Verified: Human melatonin suppression is driven by ipRGC/melanopsin, with a trigger threshold of ~1.5 melanopic lux (PMID 30697806); monochromatic blue light <1 lux can suppress melatonin under strict conditions (PMID 12163843); ≤200 lux with optimized spectra can acutely suppress melatonin (PMID 9406028); white light minimum effective intensity is in the hundreds of lux (PMID 9756329); dark adaptation enhances light sensitivity (PMC2717723); lunar phase–sleep epidemiological association exists but effects are minimal (PMID 23891110).

Partially Verified: Modulation amplitude of low-illuminance suppression thresholds by dark adaptation/light history; quantitative relationship between individual ipRGC function differences (PIPR) and melatonin suppression sensitivity (PMID 30809186, 38348477).

To Be Verified: Whether moonlight spectrum (non-monochromatic, non-white) can drive measurable melatonin suppression at 0.1–0.3 lux illuminance—no direct experiment exists; verification of whether actual retinal light dose after closing eyes + curtains reaches the converted trigger threshold.

Related Reading

Evidence Level: B (Scientific Analogy/Indirect Evidence—dose-response extrapolation + monochromatic light precedents, but direct experiment missing for “moonlight spectrum × 0.1–0.3 lux”)
References: PMID 30697806, PMID 12163843, PMID 9406028, PMID 9756329, PMID 23891110, PMID 30809186, PMID 38348477


Disclaimer: This article compiles traditional knowledge and scientific literature for comparison and research-design discussion only. It is not medical advice and cannot replace professional diagnosis or treatment. Effect sizes, frequency ranges and exposure conclusions cited here are largely inferences at the research-report level, and individual circumstances vary widely. If you have any discomfort or health concerns, please consult a physician.

Comments

Leave a Reply

Your email address will not be published. Required fields are marked *