Evidence level: C
同一个人练功时报告的「气感强度」和用力呼吸时报告的「呼吸困难强度」,如果把当时的心情(愉悦度-唤醒度)当作协变量放回回归里,两条标定斜率的差异可能不是感觉系统的差异,而是情绪混进去的部分。
正文
气功练习者常说「热、麻、胀、沉」,这些气感报告在科学语言里属于一类主观内感受量:身体内部信号被感知、被评分。与之对照的是临床里常用的 Borg 呼吸困难量表——用力呼吸时「喘不喘、费不费力」的评分。两者都是「强度随刺激量变化」的标定曲线:横轴是生理刺激(通气量、呼吸阻力),纵轴是主观评分,斜率代表这个人对这类感觉有多敏感。
问题在于:主观评分从来不是纯净的。心理学里早有定论——任何感觉报告都叠着两维情绪底色:愉悦度(舒服-不舒服)和唤醒度(平静-激动),常用 SAM(自我评估小人量表)同批采集。实验室证据显示,情绪效价确实会调制呼吸感觉的加工:看愉悦图片时吸气时间变长、通气反应改变(PMID 15294392);抑郁水平越高,呼吸内感受任务的准确率越低(PMID 36368617);正性情绪情境能增强呼吸感觉门控、改变扣带回与眶额叶激活(PMID 36389230)。也就是说,「你当时心情如何」不是一个旁观变量,它直接参与感觉报告的生成。
这带来一个方法学判决问题:假如研究发现「气感斜率比 Borg 斜率更陡」,这个差异有多少是感觉系统本身的差异,有多少是效价混入?气感报告天然带着好奇、期待甚至愉悦(练功的正面预期),而 Borg 量表挂在「用力、费力」上,天然偏负效价。两条斜率各自泡在不同情绪底色里,直接比较就像用两把不同材质的尺子量两个东西。
本文提出的检验很直接:同一批受试者、两条协议同步采集,除了强度评分外加 SAM 二维效价评分;然后对每条斜率各做两个版本的回归——不带效价协变量的原始斜率,与加入愉悦度、唤醒度作协变量后的校正斜率。判据有三档:若校正后两条斜率差异保留且幅度基本不变,说明差异主要来自感觉系统本身,效价只是次要混入;若差异明显缩小,说明原来那部分「敏感性差异」其实是情绪差,报告层被污染;若效价协变量本身显著但只影响其中一条斜率,说明两条报告通道的情绪耦合不对称——这本身就是一个新发现,指向气感报告与情绪系统的耦合强于临床呼吸困难报告。
Borg 斜率本身的稳定性也值得一提:慢阻肺患者 Borg/通气量斜率的个体内变异系数可达 20%(PMID 7781352),即斜率这个量本身就带噪声,任何「斜率差异」的裁决都必须把这种个体内漂移计入统计功效预算。
具体设计上,检验要在个体内做,而不是跨组比较:每个受试者各自完成气感标定协议与 Borg 标定协议,各自拟出两条斜率,随后把同批采到的愉悦度、唤醒度评分作为协变量分别加入两条回归。协变量进入前后的斜率差值(Δ斜率)就是「效价混入量」的直接读数——混入量越大,说明原来的斜率里情绪成分越多。两条协议在同一次实验里同步采集效价,是为了避免「补测效价」时情绪状态已经漂移;SAM 评分本身只需几秒,不会干扰主任务。若样本量允许,再按修炼年限分层:常年练习者对气感的期待与愉悦耦合可能更深,效价混入也许随年限上升——如果属实,「练得越久气感越准」就要重新解释:准的部分有多少来自知觉系统,有多少来自正性预期放大了报告。
L1 网络状态:该主题尚未进入 L1 对应网络(CORRESPONDENCE_NETWORK.md / EDGE_TABLE_science-yixue.md 中无「气感斜率 vs Borg 斜率 + 效价协变量」对应边;L1 现有 E-061 心肺边及「主观量与客观量分列报告」原则可作远端参照,但不能直接支撑本文命题)。
证据与局限
已验证: 情绪效价与唤醒度确实调制呼吸反应与呼吸感觉加工(PMID 15294392, 36368617, 36389230);Borg 斜率个体内变异显著(PMID 7781352)。
部分验证: 效价作为协变量改变主观感觉回归斜率,在痛觉等域有证据,但未在「气感 vs Borg」双斜率框架下检验。
待验证: 同批受试者双协议下,效价协变量加入后两条斜率差异是否保留;效价耦合是否在气感通道不对称地更强。此为方法学提案,属初步探索。
相关阅读
证据等级: C
参考文献: PMID 15294392, PMID 36368617, PMID 36389230, PMID 7781352
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
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。
Affective Covariate Modulation Test of the Valence-Covariate Calibration Slope for Air Sensation vs. Dyspnea: Do the Two Slope Differences Still Hold After Incorporating Valence…
Evidence level: C
When an individual reports “Qi sensation intensity” during practice and “dyspnea intensity” during forced breathing, treating the concurrent mood (pleasure-arousal) as a covariate in regression may reveal that the difference between the two calibration slopes reflects emotional influence rather than a difference in the sensory system.
Evidence Level A: Structural Analogy of Valence Modulation
| Variable | Traditional Term | Glossary Definition (First Appearance) |
|---|---|---|
| $S_{Qi}$ | 气感强度 (Qi Sensation Intensity) | A subjective report of internal energy flow or presence, distinct from external stimuli. |
| $S_{Dyspnea}$ | 呼吸困难强度 (Dyspnea Intensity) | A subjective report of the sensation of difficulty breathing or air hunger. |
| $V$ | 愉悦度-唤醒度 (Pleasure-Arousal) | A two-dimensional affective space where pleasure ranges from unpleasant to pleasant, and arousal ranges from calm to excited. |
| $\beta_{Qi}$ | 气感标定斜率 (Qi Calibration Slope) | The rate of change in reported Qi sensation intensity per unit of physiological stimulus. |
| $\beta_{Dyspnea}$ | 呼吸困难标定斜率 (Dyspnea Calibration Slope) | The rate of change in reported dyspnea intensity per unit of physiological stimulus. |
| $\Delta\beta$ | 斜率差异 (Slope Difference) | The difference between the Qi calibration slope and the dyspnea calibration slope. |
Hypothesis
If the difference $\Delta\beta$ is significant, it may indicate a structural analogy between the two reporting systems. However, if $V$ is included as a covariate and $\Delta\beta$ becomes non-significant, the difference likely stems from emotional modulation rather than a fundamental sensory distinction.
Evidence Level B: Regression Framework with Valence Covariate
The proposed model incorporates pleasure-arousal ratings to adjust the calibration slopes:
$$ \text{Intensity} = \alpha + \beta \cdot \text{Stimulus} + \gamma \cdot V + \epsilon $$
Where:
– $\alpha$ is the intercept.
– $\beta$ is the slope for the specific sensation.
– $\gamma$ is the coefficient for the valence covariate $V$.
– $\epsilon$ represents residual error.
Methodology
- Data Collection: Simultaneous measurement of $S_{Qi}$, $S_{Dyspnea}$, and $V$.
- Regression Analysis: Fit separate models for Qi sensation and dyspnea, including $V$ as a covariate.
- Comparison: Evaluate the difference in slopes ($\Delta\beta$) after adjustment.
Evidence Level C: Interpretation of Results
| Outcome | Interpretation |
|---|---|
| $\Delta\beta$ remains significant | Suggests a genuine structural analogy between the two sensory systems. |
| $\Delta\beta$ becomes non-significant | Indicates that the difference was likely due to emotional modulation rather than sensory system differences. |
Evidence Level D: Limitations and Future Directions
- Limitation: The current approach assumes that pleasure-arousal ratings fully capture the emotional component influencing sensation reports.
- Future Direction: Further research could explore other affective dimensions or physiological markers to refine the calibration model.
Conclusion
This study proposes a rigorous method for disentangling emotional influences from sensory reporting in traditional practices. By incorporating pleasure-arousal ratings as covariates, we can test whether the observed differences in calibration slopes reflect true structural analogies or merely emotional modulation. This approach enhances the evidence level of our understanding of the interplay between traditional concepts and modern psychophysiological frameworks.
Body
Practitioners of qigong often report sensations described as “heat, numbness, swelling, and heaviness.” These reports of qi-sensations belong to a class of subjective interoceptive measures in scientific terminology: internal bodily signals that are perceived and rated. In contrast, clinical settings commonly use the Borg Dyspnea Scale, which rates how “short of breath” and “effortful” breathing feels during exertion. Both scales represent intensity-calibration curves: the x-axis reflects physiological stimuli (ventilation volume, breathing resistance), while the y-axis represents subjective ratings; the slope indicates an individual’s sensitivity to these sensations.
The problem is that subjective ratings are never pure. Psychology has long established that any sensory report carries two-dimensional emotional coloring: valence (pleasant–unpleasant) and arousal (calm–excited), typically collected simultaneously using the SAM (Self-Assessment Manikin). Laboratory evidence confirms that emotional valence indeed modulates the processing of respiratory sensation: viewing pleasant images prolongs inhalation time and alters ventilatory responses (PMID 15294392); higher depression levels reduce accuracy on respiratory interoception tasks (PMID 36368617); and positive emotional contexts enhance respiratory sensation gating, modifying activation in the cingulate and orbitofrontal regions (PMID 36389230). In short, “how you felt at the time” is not a bystander variable; it directly participates in generating the sensory report.
This raises a methodological judgment question: if research finds that “the qi-sensation slope is steeper than the Borg slope,” how much of this difference stems from inherent differences in the sensory system, and how much arises from valence contamination? Reports of qi-sensations naturally carry curiosity, expectation, and even pleasure (positive expectations from practice), whereas the Borg scale is anchored to “effort, strain,” inherently carrying negative valence. Each slope floats on a different emotional backdrop; comparing them directly is like measuring two objects with rulers made of different materials.
The test proposed here is straightforward: the same cohort of participants completes both protocols synchronously, collecting not only intensity ratings but also the two-dimensional valence ratings via SAM. For each slope, we fit two versions of regression: the original slope without valence covariates, and a corrected slope incorporating arousal and pleasantness as covariates. The criterion has three tiers: if the difference between the two slopes persists after correction and remains largely unchanged, the discrepancy originates primarily from the sensory system itself, with valence playing only a secondary contaminating role; if the difference shrinks significantly, the original “sensitivity gap” was actually an emotional gap, meaning the reporting layer was polluted; if the valence covariates are significant but affect only one of the two slopes, this indicates asymmetric emotional coupling between the two reporting channels—a novel finding suggesting that qi-sensation reports are more tightly coupled to the emotional system than clinical dyspnea reports.
The stability of the Borg slope itself is also worth noting: in patients with COPD, the within-subject coefficient of variation for Borg/ventilation slopes can reach 20% (PMID 7781352), meaning the slope itself carries noise. Any verdict on “slope differences” must therefore account for this within-subject drift in statistical power calculations.
In design, the test must be conducted within subjects, not across groups: each participant completes both the qi-sensation calibration protocol and the Borg calibration protocol independently, generating two slopes for themselves. Subsequently, the pleasantness and arousal ratings collected simultaneously are added as covariates to each regression. The difference in slopes before and after introducing covariates (Δslope) provides a direct readout of “valence contamination amount”—the larger the contamination, the more emotional components were embedded in the original slope. Collecting valence ratings synchronously within the same experiment avoids drift in emotional state during “catch-up” valence assessment; SAM ratings themselves take only a few seconds and do not interfere with the primary task. If sample size permits, stratification by years of practice can be added: long-term practitioners may exhibit deeper coupling between expectation and pleasantness regarding qi-sensations, so valence contamination might increase with years of practice—if true, the claim “the longer one practices, the more accurate the qi-sensation” requires reinterpretation: how much of the accuracy stems from perceptual system improvements, and how much arises from positive expectations amplifying reports.
L1 Network Status: This topic has not yet entered the L1 correspondence network (no “qi-sensation slope vs Borg slope + valence covariates” edge exists in CORRESPONDENCE_NETWORK.md / EDGE_TABLE_science-yixue.md; existing L1 E-061 cardiopulmonary edge and the principle of “separate reporting of subjective and objective measures” can serve as remote references but cannot directly support the proposition of this paper).
Evidence and Limitations
Verified: Emotional valence and arousal do modulate respiratory responses and the processing of respiratory sensations (PMID 15294392, 36368617, 36389230); individual within-subject variability in Borg ratings is significant (PMID 7781352).
Partially Verified: Valence acting as a covariate alters the slope of subjective sensation regression; evidence exists for this in domains such as pain, but it has not been tested within the dual-slope framework of “Qi sensation versus Borg.”
Pending Verification: Under a dual-protocol design with the same cohort, it remains to be determined whether the difference between the two slopes persists after adding valence as a covariate; furthermore, it is unclear whether valence coupling exerts a stronger influence asymmetrically on the Qi sensation channel. This constitutes a methodological proposal and represents preliminary exploration.
Related Reading
- Structured self-assessment of Qi sensation: location, intensity, timing
- Borg slope as a perceptual sensitivity readout
- Resonant consciousness field research pool
Evidence level: C
References: PMID 15294392, PMID 36368617, PMID 36389230, PMID 7781352
Evidence level: A = direct PubMed support, B = scientific analogy/indirect evidence, C = traditional consensus/preliminary exploration
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.