5D Library › Neurobiology of Ritual, Spatial Boundary Transitions and Computational Biomedicine of Self-Transcendence · 中文

Chapter 5 Expectation and Learning: Open and Hidden Administration, Placebo and Conditioned Immunity

The expectation inside a ritual

Every ritual comes with a story: do this, and things will be better. The framework uses two models to understand what the story does: a Bayesian model of open versus hidden administration, showing how expectation changes experience; and Rescorla–Wagner conditioning, showing how the body links a neutral cue to the effect of a drug, even in the immune system. This chapter checks the evidence first and then writes the models.

Evidence

Open and hidden administration (human studies). Colloca, Lopiano, Lanotte and Benedetti (2004) compared the same treatment when patients knew about it (open, given by a doctor in person) and when they did not (hidden, delivered by an infusion pump without their knowledge). In post-operative pain, hidden analgesics were less effective; in anxiety, hidden diazepam was markedly less effective; and in patients with Parkinson’s disease receiving subthalamic stimulation, manipulating the stimulation openly or covertly affected motor performance differently (the exact manipulation could not be checked against the original in this round) (Colloca et al. 2004). The difference between open and hidden can be taken as the contribution of expectation (and the clinical context) to the effect of treatment. The framework attributes the paradigm to “Benedetti” alone; it was the joint work of four authors, and is corrected accordingly.

Büchel, Geuter, Sprenger and Eippert (2014) explained placebo analgesia from the standpoint of predictive coding: perceived pain is a precision-weighted combination of nociceptive input and expectation, and the higher the precision of expectation and the lower that of the input, the more expectation shapes experience (Büchel et al. 2014). This is the same formula recalled in Chapter 1.

The author’s model: a Bayesian decomposition of open and hidden

Untreated pain is 6 points (0–10), and the drug lowers the nociceptive signal by 2 points. Told “you are getting a strong painkiller”, the patient expects pain of 3; given the drug hidden, she still expects 6. Reported pain is the posterior mean wx+(1−w)μew\,x + (1-w)\mu_e with w=πs/(πs+πe)w = \pi_s/(\pi_s + \pi_e). The program computes five cases at four ratios of expectation precision to signal precision, and in part B simulates a phenomenon that has nothing to do with expectation: regression to the mean.

Calculation

# A Gaussian model of open and hidden administration, and regression to the mean
# (Chapter 5). The author's illustrative model in the spirit of Buchel et al. (2014).
# Reported pain = posterior mean = w x + (1 - w) mu_e, where x is the nociceptive signal
# (precision pi_s) and mu_e the expected pain (precision pi_e), w = pi_s/(pi_s + pi_e).
# Untreated pain 6 (0-10 scale); the drug lowers the signal by 2. Told "you are getting a
# strong painkiller", the patient expects pain 3; given it hidden, she expects 6.
import math, random

def report(x, mu_e, ratio):                       # ratio = pi_e / pi_s
    w = 1 / (1 + ratio)
    return w * x + (1 - w) * mu_e

print("A  reported pain           ratio pi_e/pi_s:   0.25    0.5    1.0    2.0")
arms = (("no treatment", 6, 6), ("hidden drug", 4, 6), ("open drug", 4, 3),
        ("open placebo (deceptive)", 6, 3), ("open-label placebo", 6, 5.4))
for name, x, mu in arms:
    cells = "".join(f"{report(x, mu, r):7.2f}" for r in (0.25, 0.5, 1, 2))
    print(f"   {name:26s}" + cells)
for r in (0.25, 0.5, 1, 2):
    w = 1 / (1 + r)
    print(f"   ratio {r:4.2f} (w {w:.2f}): open - hidden ="
          f" {report(4, 3, r) - report(4, 6, r):+.2f} = -(1 - w) x 3;"
          f" hidden drug vs none {report(4, 6, r) - 6:+.2f}")
print("   (open-label placebo: expectation lowered by 10 % only, an assumption)")
# B: regression to the mean. True pain varies between people (mean 5, sd 1.2); each
# measurement adds day-to-day noise (sd 1.2). People enrol when a screening score >= 7.
rng = random.Random(1975)
enrol = []
while len(enrol) < 20000:
    true = rng.gauss(5, 1.2); scr = true + rng.gauss(0, 1.2)
    if scr >= 7:
        enrol.append((true, scr))
ms = math.fsum(s for _, s in enrol) / len(enrol)
mf = math.fsum(t + rng.gauss(0, 1.2) for t, _ in enrol) / len(enrol)
mt = math.fsum(t for t, _ in enrol) / len(enrol)
print(f"B  20,000 people enrolled with a screening score >= 7: screening mean {ms:.2f},")
print(f"   follow-up mean {mf:.2f} with NO treatment (their true mean is {mt:.2f});")
print(f"   the apparent 'improvement' of {ms - mf:.2f} points is selection plus noise")
# A  reported pain           ratio pi_e/pi_s:   0.25    0.5    1.0    2.0
#    no treatment                 6.00   6.00   6.00   6.00
#    hidden drug                  4.40   4.67   5.00   5.33
#    open drug                    3.80   3.67   3.50   3.33
#    open placebo (deceptive)     5.40   5.00   4.50   4.00
#    open-label placebo           5.88   5.80   5.70   5.60
#    ratio 0.25 (w 0.80): open - hidden = -0.60 = -(1 - w) x 3; hidden drug vs none -1.60
#    ratio 0.50 (w 0.67): open - hidden = -1.00 = -(1 - w) x 3; hidden drug vs none -1.33
#    ratio 1.00 (w 0.50): open - hidden = -1.50 = -(1 - w) x 3; hidden drug vs none -1.00
#    ratio 2.00 (w 0.33): open - hidden = -2.00 = -(1 - w) x 3; hidden drug vs none -0.67
#    (open-label placebo: expectation lowered by 10 % only, an assumption)
# B  20,000 people enrolled with a screening score >= 7: screening mean 7.83,
#    follow-up mean 6.41 with NO treatment (their true mean is 6.41);
#    the apparent 'improvement' of 1.42 points is selection plus noise

Part A: the open-minus-hidden difference equals −(1−w)×3-(1-w)\times 3, and grows with the relative precision of expectation (−0.60 at ratio 0.25, −2.00 at ratio 2). A less intuitive result is that, in this model, the hidden drug also looks “discounted”, because the unchanged expectation pulls experience back towards the original 6; at ratio 1, a drug that lowers the signal by 2 points buys only a 1.00-point drop in reported pain. A deceptive open placebo lowers pain from 6.00 to 4.50 at ratio 1; an open-label placebo (the patient knows it is a placebo) lowers it only to 5.70 under this book’s assumption that expectation falls by just 10 %.

Evidence

Open-label placebo (randomised trials and meta-analysis in humans). Kaptchuk and colleagues (2010) ran a randomised trial in patients with irritable bowel syndrome: patients told that they were receiving placebo pills with no active ingredient reported more improvement after three weeks than patients given no treatment (Kaptchuk et al. 2010). The meta-analysis of von Wernsdorff and colleagues (2021) pooled clinical trials of open-label placebos and found a significant overall effect, but the trials were few and small, most outcomes were self-reported, and the risk of bias was high (Wernsdorff et al. 2021). A safer statement is therefore that open-label placebos have moderate effects on self-reported outcomes in some conditions, with little evidence on objective physiological outcomes.

Part B shows a phenomenon often mistaken for a placebo effect. True pain in a population has mean 5 and standard deviation 1.2, each measurement adds day-to-day variation of 1.2, and only people with a screening score of at least 7 are enrolled. The 20,000 enrolled have a screening mean of 7.83 and, with no treatment at all, a follow-up mean of 6.41: an “improvement” of 1.42 points that comes entirely from selection and measurement noise. This is regression to the mean. No study with only before-and-after measurements and no concurrent control group can attribute such a fall to a practice or a ritual. The trial in Chapter 8 therefore has concurrent control groups.

Conditioned immunity: findings in rats and evidence in humans

Evidence

Conditioned immunosuppression in rats (animal experiment). Ader and Cohen (1975) were originally studying taste aversion: rats drank saccharin water while being injected with cyclophosphamide, which causes nausea and is also immunosuppressive. They noticed higher mortality in rats later given saccharin water alone, and went on to show that re-presenting saccharin water by itself suppressed the rats’ antibody response to an antigen (Ader and Cohen 1975). The study opened the field of psychoneuroimmunology: a formerly neutral taste cue, paired with a drug, acquired the power to affect immune function.

Evidence

Conditioned immunosuppression in humans (small human experiments). In healthy men, Goebel and colleagues (2002) paired the immunosuppressant cyclosporin A several times with a drink of unusual smell and colour; when the drink was later given with placebo capsules, lymphocyte proliferation and the expression of interleukin-2 and interferon-γ were suppressed (Goebel et al. 2002). This shows that conditioned immunosuppression can occur in humans; but such studies usually have a few dozen participants, the effect varies greatly between individuals, and the specific figures from the original paper could not be obtained in checking for this book. They demonstrate a learning mechanism, not a therapy that can replace drugs.

Rescorla and Wagner (1972) wrote learning as an update driven by prediction error (Rescorla and Wagner 1972): each time the conditioned stimulus (CS) appears, its associative strength VV is updated by ΔV=αβ(λ−Vtotal)\Delta V = \alpha\beta(\lambda – V_{\text{total}}), where λ\lambda is 1 when the unconditioned stimulus (US) occurs and 0 when it does not, VtotalV_{\text{total}} is the summed strength of all cues present, and α\alpha and β\beta are the salience of the cue and of the US. The next program uses it to illustrate several properties of conditioned immunosuppression. The drug’s own effect is assumed to lower a lymphocyte marker by 60 %, an illustrative value, not a figure from any study.

Calculation

# Rescorla-Wagner learning applied to conditioned immunosuppression (Chapter 5).
# Illustrative: dV = alpha * beta * (lambda - V_total) on every trial with the CS.
# CS = a novel-tasting drink; US = an immunosuppressive drug (lambda = 1 when given).
# The conditioned response is V times the drug's own effect; the drug is assumed to lower
# a lymphocyte marker by 60 % (an assumption for illustration, not a study value).
import math

def train(schedule, alpha, beta=0.5, v=0.0, context=0.0, alpha_ctx=0.0):
    out = []
    for us in schedule:
        err = (1.0 if us else 0.0) - (v + context)
        v += alpha * beta * err; context += alpha_ctx * beta * err
        out.append(v)
    return out

DRUG = 0.60
print("A  acquisition: V after k paired trials (novel taste alpha 0.3; familiar 0.08)")
for alpha, lab in ((0.3, "novel"), (0.08, "familiar")):
    v = train([1] * 8, alpha)
    print(f"   {lab:8s}" + "".join(f"  k={k}: {v[k - 1]:.2f}" for k in (1, 2, 4, 8)))
v4 = train([1] * 4, 0.3)[-1]
print(f"   after 4 novel pairings the conditioned effect is {v4:.2f} x {DRUG:.0%}"
      f" = {v4 * DRUG:.1%} suppression")
print("B  extinction: CS alone at evocation, starting from V after 4 pairings")
ext = train([0] * 6, 0.3, v=v4)
print("   after CS-alone trial " + "  ".join(f"{k}: {ext[k - 1] * DRUG:.1%}"
                                         for k in range(1, 7)))
print("C  partial reinforcement (drug on alternate trials, 'dose reduction'), 40 trials")
for alpha in (0.3, 0.08):
    v = train([1, 0] * 20, alpha)
    lo, hi = min(v[-10:]), max(v[-10:])
    print(f"   alpha {alpha:.2f}: V between {lo:.2f} and {hi:.2f}"
          f" (mean of last 10 trials {math.fsum(v[-10:]) / 10:.2f})")
print("D  blocking: the room (context) already predicts the drug before the drink"
      " is added")
ctx = train([1] * 10, 0.5)[-1]
blk = train([1] * 4, 0.3, context=ctx, alpha_ctx=0.5)
print(f"   context V {ctx:.3f}; drink V after 4 compound trials {blk[-1]:.3f}"
      f" (vs {v4:.3f} without")
print("   pre-training)")
# A  acquisition: V after k paired trials (novel taste alpha 0.3; familiar 0.08)
#    novel     k=1: 0.15  k=2: 0.28  k=4: 0.48  k=8: 0.73
#    familiar  k=1: 0.04  k=2: 0.08  k=4: 0.15  k=8: 0.28
#    after 4 novel pairings the conditioned effect is 0.48 x 60% = 28.7% suppression
# B  extinction: CS alone at evocation, starting from V after 4 pairings
#    after CS-alone trial 1: 24.4%  2: 20.7%  3: 17.6%  4: 15.0%  5: 12.7%  6: 10.8%
# C  partial reinforcement (drug on alternate trials, 'dose reduction'), 40 trials
#    alpha 0.30: V between 0.46 and 0.54 (mean of last 10 trials 0.50)
#    alpha 0.08: V between 0.36 and 0.41 (mean of last 10 trials 0.38)
# D  blocking: the room (context) already predicts the drug before the drink is added
#    context V 0.944; drink V after 4 compound trials 0.018 (vs 0.478 without
#    pre-training)

Part A: a novel taste (α=0.3\alpha = 0.3) reaches V=0.48V = 0.48 after 4 pairings, corresponding to 28.7 % conditioned suppression; a familiar taste (α=0.08\alpha = 0.08) reaches only 0.15 after 4 pairings. This is why experiments use unusual drinks: the more novel the cue, the faster the learning (slow learning about familiar cues, “latent inhibition”, is approximated here only by a smaller α\alpha). Part B: presenting the drink without the drug extinguishes the association trial by trial, and after 6 trials conditioned suppression falls from 28.7 % to 10.8 %. Part C: if the drug is given on alternate occasions to reduce the dose (partial reinforcement), VV swings between 0.46 and 0.54, around 0.50 on average; in this model the conditioned effect can sustain only part of the drug’s effect and cannot replace the drug. Part D is “blocking”: if the room already fully predicts the drug (V=0.944V = 0.944), a drink added later learns almost nothing (0.018, against 0.478 without pre-training).

Part D has a direct implication for ritual: if a person always receives a treatment in the same place and the same routine, the role of any single cue may be spread over the whole context. The Rescorla–Wagner model cannot describe spontaneous recovery, context updating or relapse after extinction, and later learning theories modified it in many ways; only its most basic properties are used here.

Ritual, expectation and regression to the mean

Putting this chapter’s three programs together gives a list of possible explanations for “I felt better after the ritual”: expectation moves experience towards what is expected (A); earlier pairings let the cue itself evoke part of a response (conditioning); people who enrol at a peak fall back naturally (regression to the mean); and the practice has its own physiological effects (Chapter 2). In an observation with only before-and-after measurements the four are completely mixed. The trial design of Chapter 8 aims to separate some of them.

Hypothesis · not established · Conditioning to the ritual context

The author’s hypothesis (not established): after several weeks of slow-breathing practice repeated in a fixed space with a fixed action sequence, merely entering that space and starting the action sequence (without slow breathing) produces a measurable fall in heart rate, larger than on entering a space where the person has never practised. What would show it wrong: under a pre-registered protocol, a confidence interval for the difference in heart-rate change between entering the practice space and entering a new space that lies within the smallest effect of interest; or a difference that appears when participants are told “this is where you practise” and not when they are not told (which would show that expectation, not conditioning, is doing the work).

Chapter summary

The classic open–hidden study is the 2004 joint work of Colloca, Lopiano, Lanotte and Benedetti; the author’s Bayesian model shows that, in this model, the open-minus-hidden difference is (1−w)(1-w) times the change in expectation, and that an unchanged expectation also “discounts” the drug’s effect. Effects of open-label placebos are seen mainly in self-reported outcomes, and the evidence is of limited quality. Regression to the mean can produce an “improvement” of 1.42 points with no treatment at all. Ader and Cohen discovered conditioned immunosuppression in rats, and Goebel and colleagues observed a similar phenomenon in a small sample of healthy people; the Rescorla–Wagner model illustrates the roles of cue novelty, extinction, partial reinforcement and blocking, but conditioning sustains only part of a drug’s effect and is no substitute therapy.

References

Ader, Robert, and Nicholas Cohen. 1975. “Behaviorally Conditioned Immunosuppression.” Psychosomatic Medicine 37 (4): 333–40. https://doi.org/10.1097/00006842-197507000-00007.
Büchel, Christian, Stephan Geuter, Christian Sprenger, and Falk Eippert. 2014. “Placebo Analgesia: A Predictive Coding Perspective.” Neuron 81 (6): 1223–39. https://doi.org/10.1016/j.neuron.2014.02.042.
Colloca, Luana, Leonardo Lopiano, Michele Lanotte, and Fabrizio Benedetti. 2004. “Overt versus Covert Treatment for Pain, Anxiety, and Parkinson’s Disease.” The Lancet Neurology 3 (11): 679–84. https://doi.org/10.1016/S1474-4422(04)00908-1.
Goebel, Marion U., Annette E. Trebst, Jan Steiner, et al. 2002. “Behavioral Conditioning of Immunosuppression Is Possible in Humans.” The FASEB Journal 16 (14): 1869–73.
Kaptchuk, Ted J., Elizabeth Friedlander, John M. Kelley, et al. 2010. “Placebos without Deception: A Randomized Controlled Trial in Irritable Bowel Syndrome.” PLoS ONE 5 (12): e15591. https://doi.org/10.1371/journal.pone.0015591.
Rescorla, Robert A., and Allan R. Wagner. 1972. “A Theory of Pavlovian Conditioning: Variations in the Effectiveness of Reinforcement and Nonreinforcement.” In Classical Conditioning II: Current Research and Theory, edited by Abraham H. Black and William F. Prokasy. Appleton-Century-Crofts.
Wernsdorff, Melina von, Martin Loef, Brunna Tuschen-Caffier, and Stefan Schmidt. 2021. “Effects of Open-Label Placebos in Clinical Trials: A Systematic Review and Meta-Analysis.” Scientific Reports 11: 3855. https://doi.org/10.1038/s41598-021-83148-6.

Fan, Y. (2026). Neurobiology of Ritual, Spatial Boundary Transitions and Computational Biomedicine of Self-Transcendence: From Predictive-Coding Free-Energy Minimisation, Default-Mode Network Decoupling and Vagal–Endocrine Restoration to Classical Chinese Medicine on Shifting the Spirit (Yijing Bianqi) [仪式神经学、空间边界切换与自我转移的计算生物医学] (Version 1.0.0). 5D Library. https://doi.org/10.5281/zenodo.23839910 · Licensed under Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0).