Evidence level: C
退行速度取决于剂量降到哪,而不是简单地”练就保、停就丢”——最小维持频率拐点是判定脉搏签名算不算”指纹”的判决实验。
正文
训练适应的退行从来不是一根均匀往下滑的斜线。Neufer 的经典综述指出(PMID 2692122),完全停练后最大摄氧量在 2 到 4 周内快速下降,主因是血容量回落导致每搏输出量缩水;但只要每次训练的强度保住,频率与时长减掉三分之一到三分之二,有氧适应可以维持数月。Hickson 等(PMID 7219129)更早用随机对照直接给出频率阈值:减到每周 2 次仍可维持已获得的有氧能力。这两条证据合起来就是”剂量-反应”裁决的骨架:退行是剂量函数,不是开关。
维持剂量可以低到什么程度?Tucci 等的腰伸肌实验(PMID 1471008)给出一个惊人的下限:经过 10 到 12 周训练获得的力量,在单次强度与容量不变的前提下,每 2 周练 1 次、甚至每 4 周才练 1 次,都能把力量整整维持 12 周不降;而完全停练组平均掉了约 55%。对力量这类慢成分,”维持剂量”与”获得剂量”之间可以差一个数量级以上。这与本系列前篇拆出的退功快慢双分量结构吻合:快分量(神经募集、血浆容量类)几天内塌掉,慢分量(结构重塑类)靠极低剂量就能托住。
但”减量能维持”不等于”频率无所谓”。Dalleck 等(PMID 20535658)把受试者随机分入每周 1 次与每周 2 次的高强度间歇附加训练,6 周后乳酸阈提升分别为 4.3% 与 8.2%,组间差异显著——频率本身就是一条独立的剂量轴,不是可以随意归零的变量。这提示”最小维持频率拐点”在原则上存在,且很可能因成分而异。
2026 年发表的系统综述与 meta 分析(PMID 42558561)汇总 86 项随机对照试验共 5670 人,结论是停训后的残余效应高度分项分化:最大摄氧量的改善在短、中、长期随访窗内均保持显著,而静息心率、血脂等读数较快回到基线。换句话说,不同生理读数的退行时间常数天然不同,任何”一刀切”的维持方案都注定顾此失彼。
把这些证据接到脉搏签名上,就得到本篇的裁决设计:练出稳定谐波签名的受试者随机分入完全停练组与每周 2、3、4 次减量组(每次训练的强度与容量保住),逐周追踪 C1 到 C10 谐波、反射波增强指数与心率变异性,拟合各组的保留量曲线并做拐点检测,找最小维持频率;再与最大摄氧量”每周 2 次即可维持”的阈值对照。若逐谐波维持曲线不一致——低阶谐波(心率主导)所需的维持频率远高于高阶谐波——则”签名”更像一束时间常数各异的成分叠加;若各阶同步退行,则支持存在单一的底层调节源。两个结果都有信息量,这正是裁决实验的价值。
设计上要锁三件事。第一,减量组减的必须是频率而不是强度——全部既有证据一致表明,单次强度保住是维持成立的共同前提,强度与频率同时减的”减量”无法与完全停练区分开。第二,拐点检测要以”保留量平台”为判据而不是终点比较:每周 2、3、4 次组的曲线若在某个保留高度上先后走平,各频率组的平台高度之差就是剂量-反应的形状,最小维持频率取”平台不低于停练前 95%”的最低频率档。第三,分析要用逐周重复测量的混合效应模型,谐波按阶分层报告,避免把不同时间常数的成分混进一条平均曲线里。样本量上,前篇的退行离散度显示个体间差异不小,每组至少需要十余例才可能把组间保留量差与个体噪声分开。
这个设计还埋着一个对照性假设:若最小维持频率落在每周 2 次附近,与最大摄氧量的维持阈值同档,则提示谐波签名与有氧适应共享底层调节机制,签名更接近”心肺读数”而非”独立指纹”;若拐点更低——像腰伸肌那样每 2 到 4 周一次就能托住——则签名更接近结构性慢成分,”指纹性”反而更强。传统练功者的经验说法里本就有”功要续、不可断,但养功不必如练功”的分层直觉,此实验相当于给这条直觉一个可测的现代版本。
易学传统把脉象当作长期体质的指纹,《内经》论四时之脉即预设脉象有其稳定结构。但现代证据提醒:多数生理读数是”用进废退”的动态量,可训练性本身不等于稳定性——一个读数要配得上”指纹”二字,必须先证明它在极低维持剂量下不塌。本篇仍属方法论推演,证据等级为 C。
证据与局限
已验证: 频率剂量-反应现象与极低频维持现象在肌力(每周 1 次至每 4 周 1 次可维持 12 周)、乳酸阈(每周 1 次与 2 次呈剂量关系)、有氧能力(每周 2 次可维持)三个领域各有随机对照或系统综述直接支撑。
部分验证: 脉搏谐波签名的可训练性与退行双分量结构(本系列前篇,个体化记录与短期追踪层面)。
待验证: 逐谐波维持曲线、最小维持频率拐点的存在性与数值、谐波维持阈值与摄氧量维持阈值的关系——均无直接文献。
相关阅读
证据等级: C
参考文献: PMID 2692122, PMID 7219129, PMID 1471008, PMID 20535658, PMID 42558561
证据等级:A=有PubMed直接支撑,B=科学类比/间接证据,C=传统文化共识/初步探索
L1 网络状态:该主题尚未进入 L1 对应网络。EDGE_TABLE 中「退功/减量维持/脉搏签名」零命中;最近邻边为「谐波 / 脉象」(A级,PMID 21622820,脉象分类盲评主题),不覆盖减量维持剂量-反应。
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。
Dose-Response Decisions for Pulse Signature Reduction Maintenance: Retention Curves of Complete Cessation and Weekly 2/3/4-Session Reduction Groups
Evidence level: C
Regression speed depends on the dose level at which reduction occurs, rather than simply “maintaining upon practice and losing upon cessation”—the minimum maintenance frequency threshold serves as the decisive experiment for determining whether pulse signatures qualify as “fingerprints.”
Body
Regression of adaptation is never a uniform downward slope. Neufer’s classic review (PMID 2692122) indicates that after complete cessation, maximal oxygen uptake (VO₂max) declines rapidly within 2 to 4 weeks, primarily due to reduced blood volume leading to diminished stroke volume; however, if training intensity is preserved while frequency and duration are reduced by one-third to two-thirds, aerobic adaptation can be maintained for several months. Hickson et al. (PMID 7219129) earlier provided a frequency threshold via randomized controlled trials: reducing to two sessions per week still maintained acquired aerobic capacity. Together, these two lines of evidence form the skeleton of the “dose-response” verdict: regression is a dose function, not a switch.
How low can the maintenance dose go? Tucci et al.’s experiment on the erector spinae (PMID 1471008) reveals a startling lower limit: strength gained over 10 to 12 weeks of training, under conditions where intensity and volume remain constant, can be maintained for 12 weeks without decline when training once every 2 weeks or even once every 4 weeks, whereas the complete cessation group lost approximately 55%. For slow components like strength, the “maintenance dose” can differ by more than an order of magnitude from the “acquisition dose.” This aligns with the dual-component structure of regression (fast and slow) identified in earlier parts of this series: the fast component (neural recruitment, plasma volume) collapses within days, while the slow component (structural remodeling) can be supported by extremely low doses.
However, “maintainable at reduced load” does not mean “frequency is irrelevant.” Dalleck et al. (PMID 20535658) randomly assigned subjects to high-intensity interval training added once or twice weekly; after 6 weeks, lactate threshold increased by 4.3% and 8.2%, respectively, with significant intergroup differences—frequency itself is an independent dose axis, not a variable that can be arbitrarily set to zero. This suggests the “minimum maintenance frequency inflection point” exists in principle, and likely varies by component.
The 2026 systematic review and meta-analysis (PMID 42558561), aggregating 86 randomized controlled trials involving 5,670 participants, concludes that residual effects after cessation are highly component-specific: improvements in VO₂max remain significant across short, medium, and long-term follow-up windows, whereas resting heart rate and lipid profiles return to baseline relatively quickly. In other words, different physiological markers have naturally distinct regression time constants; any “one-size-fits-all” maintenance protocol is destined to be partial.
Connecting these evidence streams to the pulse signature yields the design of this verdict: subjects exhibiting stable harmonic signatures are randomly assigned to complete cessation groups or reduced-frequency groups (2, 3, or 4 times per week), with intensity and volume preserved. Weekly tracking of harmonics C1 through C10, reflected wave enhancement index, and heart rate variability will allow fitting of retention curves and inflection point detection to identify the minimum maintenance frequency, which will then be compared against the VO₂max threshold of “twice weekly suffices.” If maintenance curves diverge by harmonic order—lower-order harmonics (heart rate-dominated) requiring higher frequencies than higher-order ones—the “signature” resembles a superposition of components with varying time constants; if all harmonics regress synchronously, it supports the existence of a single underlying regulatory source. Both outcomes yield information; this is precisely the value of the verdict experiment.
Three constraints must be locked in the design. First, the reduced-frequency group must reduce frequency, not intensity—all existing evidence consistently shows that preserving single-session intensity is the common prerequisite for maintenance; simultaneous reduction of both intensity and frequency cannot be distinguished from complete cessation. Second, inflection point detection must use “retained plateau” as the criterion rather than endpoint comparison: if curves for 2, 3, or 4 times/week flatten at different retained heights, the difference in plateau levels defines the dose-response shape, and the minimum maintenance frequency is taken as the lowest tier where the plateau remains ≥95% of pre-cessation levels. Third, analysis must employ mixed-effects models with repeated weekly measurements, reporting harmonics stratified by order, to avoid conflating components with different time constants into a single average curve. Regarding sample size, the high inter-individual variability in regression observed earlier necessitates at least a dozen participants per group to separate intergroup retention differences from individual noise.
This design also embeds a comparative hypothesis: if the minimum maintenance frequency falls near twice weekly, aligning with the VO₂max threshold, it suggests harmonic signatures and aerobic adaptation share underlying regulatory mechanisms, making the signature closer to “cardiopulmonary readings” than “independent fingerprints”; if the inflection point is lower—such as once every 2 to 4 weeks, akin to erector spinae strength maintenance—then the signature more closely reflects structural slow components, enhancing its “fingerprint-like” character. Traditional martial arts experience already contains the layered intuition that “practice must continue, but cultivation need not match training intensity”; this experiment provides a measurable modern version of that intuition.
Yijing Studies tradition treats pulse signatures as fingerprints of long-term constitution, and the Inner Classic’s discussion of seasonal pulses presupposes stable pulse structures. Yet modern evidence reminds us that most physiological readings are dynamic quantities governed by “use it or lose it,” and trainability itself does not equate to stability—a reading qualifies as a “fingerprint” only after demonstrating it does not collapse under extremely low maintenance doses. This piece remains methodological inference; evidence level: C.
Evidence and Limitations
Verified: Frequency dose–response phenomena and the low-frequency maintenance phenomenon are directly supported in three domains by randomized controlled trials or systematic reviews: muscle strength (maintenance for 12 weeks with training once per week to once every four weeks), lactate threshold (a dose–response relationship between once-per-week and twice-per-week protocols), and aerobic capacity (maintenance with twice-per-week training).
Partially Verified: The trainability of the pulse harmonic signature and its regressive dual-component structure (see the preceding article in this series, addressing individualized records and short-term tracking).
Pending Verification: The existence and numerical values of the per-harmonic maintenance curve and the minimum maintenance frequency inflection point, as well as the relationship between harmonic maintenance thresholds and oxygen uptake maintenance thresholds; no direct literature is available for these.
Related Reading
- Pulse Signature Withdrawal Fast/Slow Bifurcation Level Separation Decision
- Withdrawal Time Constant and Intervention Duration “Depth Effect” Decision
- Pulse Signature Individualized Retention
- Research Pool: Bio-frequency Code
Evidence Level: C
References: PMID 2692122, PMID 7219129, PMID 1471008, PMID 20535658, PMID 42558561
Evidence Level: A = direct PubMed support; B = scientific analogy/indirect evidence; C = traditional consensus/preliminary exploration.
L1 Network Status: This topic has not yet entered the L1 corresponding network. The EDGE_TABLE shows zero hits for “withdrawal/reduction maintenance/pulse signature”; the nearest neighbors are “harmonic/pulse character” (Level A, PMID 21622820, pulse character blind evaluation topic), which does not cover reduction maintenance dose-response.
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.
Leave a Reply