5D Library

Chapter 8 Practice: A Sitting Log and a Pre-Registered Experiment

5D Library › The Default Mode Network, the Entropic Brain and Neural Oscillations · 中文

Chapter 8 Practice: A Sitting Log and a Pre-Registered Experiment

Part One: four weeks of sitting and catching

This part is a practice that anyone can do, legally and at no cost: twenty minutes a day of sitting with attention on the breath, and a record of how often the mind wandered and was brought back. It uses no substance, device or product.

Note

If you have a history of psychosis, trauma, severe depression or seizures, or if practice makes you feel worse, speak to a doctor or an experienced teacher before continuing, and stop if you feel distressed (see the note at the end of Chapter 4).

The practice

  1. Sit upright on a chair or cushion, in a quiet place, at about the same time each day. Set a timer for 20 minutes.
  2. Rest attention on the sensation of breathing, at the nostrils or the abdomen. Do not control the breath.
  3. When you notice that the mind has wandered, note it silently (“wandering”), make one tally (a click counter or a mark on paper), and return to the breath. Each tally is a catch.
  4. After the timer, write down three things: the number of catches, your estimate of how many minutes in total you were “lost” before noticing, and one word for your mood.

What to record

Column Meaning
date, time when you sat
catches number of times you noticed and returned
minutes lost your estimate of total time spent wandering before noticing
sleep hours slept the night before
mood word one word, before and after

Reading four weeks of it

Calculation

An invented four-week log, summarised week by week, with the trend in catches and in estimated minutes lost.

# Four weeks of an invented sitting log: one 20-minute sit a day,
# "catches" = times the sitter noticed the mind had wandered and came back.
catches = [3, 4, 2, 5, 4, 3, 5,   5, 6, 4, 6, 5, 7, 6,
           6, 7, 5, 6, 7, 6, 8,   7, 6, 7, 8, 6, 7, 8]
felt_lost = [9, 8, 9, 7, 8, 9, 7,  7, 6, 8, 6, 7, 5, 6,
             6, 5, 6, 5, 5, 6, 4,  5, 5, 4, 4, 5, 4, 4]   # minutes 'lost' (self-estimate)

def slope(y):
    n = len(y); xs = range(n)
    mx, my = (n - 1) / 2, sum(y) / n
    return sum((x - mx) * (v - my) for x, v in zip(xs, y)) / sum((x - mx) ** 2 for x in xs)

for w in range(4):
    c, l = catches[7 * w:7 * w + 7], felt_lost[7 * w:7 * w + 7]
    print(f"week {w + 1}: {sum(c) / 7:.1f} catches per sit, {sum(l) / 7:.1f} min felt lost")
print(f"trend: {7 * slope(catches):+.2f} catches per week, "
      f"{7 * slope(felt_lost):+.2f} min lost per week")
# week 1: 3.7 catches per sit, 8.1 min felt lost
# week 2: 5.6 catches per sit, 6.4 min felt lost
# week 3: 6.4 catches per sit, 5.3 min felt lost
# week 4: 7.0 catches per sit, 4.4 min felt lost
# trend: +1.09 catches per week, -1.23 min lost per week

In this invented log, catches rise over the four weeks while estimated time lost falls. Chapter 4 explains why this is the expected pattern: noticing sooner ends each episode sooner, so more episodes fit into a sitting, and each costs less. A rising count of catches is progress, not failure. Self-estimates of time lost are rough, and should be compared only with your own earlier estimates.

Part Two: a pre-registered experiment

The question

Does twenty minutes of breath-focused listening reduce lapses of sustained attention, and repetitive self-focused thought, more than twenty minutes of listening to binaural beats? Chapter 6 found that binaural beats have a modest average effect in the literature, but weak evidence that they work by entrainment. Comparing them with an equally quiet, equally engaging practice is the test that separates entrainment from simply listening.

Design

Element Specification
Participants Healthy adults aged 18–60 with normal hearing, not currently meditating regularly; exclusions: history of psychosis or seizures, current treatment for depression, hearing impairment
Design Randomised two-period crossover; each person does both sessions, on separate days at least 48 hours apart, at the same time of day; order (AB or BA) allocated by computer, balanced
Session A: breath 20 minutes of guided breath-focused listening (recorded instructions at the start, then silence with soft room tone)
Session B: binaural 20 minutes of binaural beats (carrier near 200 Hz, 10 Hz difference) under a neutral sound bed, with the instruction simply to listen
Common Same headphones, room, volume and length; no substance of any kind; participants told that the study compares “two listening sessions”
Primary outcome Commission-error rate on the Sustained Attention to Response Task (SART), measured right after each session (Robertson et al. 1997)
Secondary outcomes Brief State Rumination Inventory (Marchetti et al. 2018); SART reaction-time variability; self-rated calm
Blinding Outcome assessment and analysis done by people who do not know the order; the analysis code is frozen before unblinding

Number of participants

Calculation

import math
from statistics import NormalDist

z = NormalDist().inv_cdf

def completers(sd_diff, delta, alpha=0.05, power=0.90):
    """Two-period crossover analysed as paired differences."""
    za, zb = z(1 - alpha / 2), z(power)
    return math.ceil(((za + zb) * sd_diff / delta) ** 2 + za ** 2 / 2)

# Outcome: commission-error rate on the Sustained Attention to Response Task
# (share of no-go trials answered by mistake), measured after each session.
for sd in (0.10, 0.12, 0.15):
    n = completers(sd, 0.05)
    print(f"sd of within-person difference {sd:.2f}: {n} completers "
          f"(recruit {math.ceil(n / 0.8)} if 20 % drop out)")
# sd of within-person difference 0.10: 44 completers (recruit 55 if 20 % drop out)
# sd of within-person difference 0.12: 63 completers (recruit 79 if 20 % drop out)
# sd of within-person difference 0.15: 97 completers (recruit 122 if 20 % drop out)

The protocol plans for the middle case: 63 people completing both sessions, rounded to 64 for balance, recruiting 80 to allow for drop-out. A pilot of 8 people will be used only to check the procedures.

Analysis and decision rule

The primary analysis estimates the within-person difference in commission-error rate (breath minus binaural) with a 95 % confidence interval, in a model that includes period and order to check for carry-over. The result will be read by a rule written before any data are seen, in the spirit of equivalence testing (Lakens 2017).

Calculation

SESOI = 0.05      # 5 percentage points of commission errors: smallest difference worth acting on

def verdict(diff, lo, hi):
    """diff = mindful listening to the breath minus binaural beats, in
    commission-error rate (negative = fewer errors after breath practice),
    with its 95 % confidence interval. Fixed before any data are seen."""
    if hi <= -SESOI:
        return "breath practice clearly better (by at least 5 points)"
    if hi < 0:
        return "breath practice better; size not shown to reach 5 points"
    if lo > 0:
        return "binaural beats better"
    if lo >= -SESOI and hi <= SESOI:
        return "no difference worth acting on (within ±5 points)"
    return "unclear: the interval is too wide"

for d, lo, hi in ((-0.08, -0.12, -0.05), (-0.03, -0.05, -0.01),
                  (0.00, -0.03, 0.03), (-0.02, -0.08, 0.04)):
    print(f"{d:+.2f} ({lo:+.2f} to {hi:+.2f}): {verdict(d, lo, hi)}")
# -0.08 (-0.12 to -0.05): breath practice clearly better (by at least 5 points)
# -0.03 (-0.05 to -0.01): breath practice better; size not shown to reach 5 points
# +0.00 (-0.03 to +0.03): no difference worth acting on (within ±5 points)
# -0.02 (-0.08 to +0.04): unclear: the interval is too wide

The examples are invented, to show in advance how each kind of result will be reported. An interval within ±5 points will be reported as “no difference worth acting on”, not “no effect”; an interval that crosses the boundaries will be reported as unclear, not as “a trend”.

Ethics and conflicts of interest

Both sessions are harmless listening activities. Participants give informed consent, can withdraw at any time, and are screened as above. The protocol will be reviewed by an independent research ethics committee and registered publicly before recruitment, following the principles of pre-registration (Nosek et al. 2018).

Disclosure

The author teaches singing and a listening method on vocal.guru and singingasia.com, and his group runs websites on frequency and resonance, so he has a commercial interest in how listening practices are perceived. To limit that interest: the protocol, analysis code and decision rule will be published before recruitment; data will be collected and analysed by researchers independent of the author’s businesses; and the results, anonymised data and code will be published whatever they show.

What this experiment cannot show

It compares two single sessions in healthy adults. It says nothing about long-term practice, about clinical populations, or about any substance. A difference in attention errors after one session is a difference in attention errors, not a change in the default network, which this design does not measure.

References

Lakens, Daniël. 2017. “Equivalence Tests: A Practical Primer for t Tests, Correlations, and Meta-Analyses.” Social Psychological and Personality Science 8 (4): 355–62.
Marchetti, Igor, Nilly Mor, Carlo Chiorri, and Ernst H. W. Koster. 2018. “The Brief State Rumination Inventory (BSRI): Validation and Psychometric Evaluation.” Cognitive Therapy and Research 42 (4): 447–60.
Nosek, Brian A., Charles R. Ebersole, Alexander C. DeHaven, and David T. Mellor. 2018. “The Preregistration Revolution.” Proceedings of the National Academy of Sciences 115 (11): 2600–2606.
Robertson, Ian H., Tom Manly, Jackie Andrade, Bart T. Baddeley, and Jenny Yiend. 1997. “‘Oops!’: Performance Correlates of Everyday Attentional Failures in Traumatic Brain Injured and Normal Subjects.” Neuropsychologia 35 (6): 747–58.

Fan, Y. (2026). The Default Mode Network, the Entropic Brain and Neural Oscillations: 5-HT2A Receptor Kinetics, Pineal Calcite Microcrystals and the Computational Neuroscience of Meditative Attention [默认模式网络、熵脑与神经振荡] (Version 1.0.0). 5D Library. https://doi.org/10.5281/zenodo.23697320 · Licensed under Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0).