Anda di halaman 1dari 11

Homeopathy (2003) 92, 19–29

& 2003 The Faculty of Homeopathy

PII: S1475-4916(02)00161-3


A kinetic approach to caffeine–Coffea

cruda interaction
G Ruiz-Vega1*, L Pérez-Ordaz1, L Cortés-Galván1, and FM Juárez-G1
Instituto de Fı´sica y Matemáticas, Universidad Michoacana, Morelia, Michoacán, México

The biological effect of Coffea cruda 30c was investigated in rats pre and post treated
with caffeine. The experimental subjects were male Wistar rats. Caffeine was
administered intraperitoneally at the beginning of a sleep period. Coffea cruda 30c
(0.1 ml) was administered orally, a contemporaneous control group was tested. The
Electroencephalogram (EEG) was recorded in the parietal region during the following
sleep cycle. The effect was evaluated by three EEG parameters: the spectral power in
delta (0.5–2.5 Hz) and slow 0.32–0.48 Hz bands and the slow/delta power ratio. These
markers were analyzed vs time for control and homeopathic groups, blind. In the pre-
treated set, a similar pattern was identified for control and verum groups up to the 4th
hour. From the 5th hour on, power in the delta band was statistically higher in the
verum. Spectral power in the slow band and power ratio for the verum group was
smaller than the control group from the 6th hour on. In the post-treated set, two verum
sub-groups were identified: Post v-A: did not exhibit significant differences from
control; Post v-B: displayed an opposite tendency than pre-treatment verum. We
conclude that Coffea cruda 30c modifies sleep pattern increasing sleep intensity with
preteatment. In a subset of the post-treated animals Coffea 30c appeared to reinforce
the effects of caffeine. Homeopathy (2003) 92, 19–29.

Keywords: Coffea cruda; slow rhythm; delta rhythm; sleep; EEG; neocortex;
stochastic resonance

Introduction In general, two conditions have been considered:

In spite of 200 years of clinical practice and various

objective analyses including meta-analyses,1,2 (i) Pretreatment with ultra-low dose followed by
homeopathy still lacks a scientific basis to explain stimulus in conventional pharmacological
its mechanism of action. Many attempts have concentration.
been made to explain the mechanism involved.2–7 (ii) Pretreatment with conventional pharmacological
Among several experimental approaches, those concentrations, followed by ultra-low-dose
that involve the evaluation of homeopathic stimulus stimulus.
interacting with a similar in pharmacological
concentrations,8–10 provide a promising research Interesting results have arisen from such experi-
field. ments, such as an induced host response enhance-
ment,8,9 which are in line with the general homeopathic
*Correspondence: G Ruiz-Vega, Instituto de Fı́sica y concept of stimulating health mechanisms. Indeed,
Matemáticas, Universidad Michoacana 58040, Morelia, specific terminology has been proposed, such as
Michoacán, México. ‘homologous priming’, to describe (i) above in which
Received 8 July 2002; revised 18 September 2002; a response enhancement is observed.8 Reported null
accepted 11 November 2002 effects of mixed treatment10 perhaps reflect the
Caffeine-Coffea interaction
G Ruiz-Vega et al
neutralization of both low- and high-dose activity, sets:
through an unknown mechanism.
This paper deals with both: (1) ‘Pre’, consisted of 36 subjects distributed in two
sub-groups: Pre-control, and Pre verum. Caffeine
(1) Pre-treatment with pharmacological dose treated (15.0 mg/kg i.p.) was administered at time zero, T0;
subjects. half an hour later two drops (0.1 ml) of
(2) Post-treatment with pharmacological dose treated homeopathic solvent (87% ethanol) or Coffea
subjects. cruda 30c (87% ethanol) was given to control
and verum sub-groups respectively.
(2) ‘Post’ also consisted of 36 subjects, it was divided
We hypothesized that the homeopathic stimulus in control and verum. 0.1 ml of homeopathic
would strengthen host response by means of ‘homo- solvent or Coffea cruda 30c was given
logous priming’ in (2) and enhance subject response to respectively, to control or verum groups 1 h
caffeine effect in (1). In addition, the evolution of before caffeine (15.0 mg/kg i.p.) administration at
parameters in time was evaluated and it fitted to a T0 (Table 1).
mathematical pattern. We hypothesize that by analyz-
ing the nature of the relationships that arise, it will be
EEG was recorded by means of three 1.5 mm
possible to get some clues to the mechanism involved; a
diameter stainless-steel electrodes (resistance o0.1 O)
model of homeopathic effect could be developed from
implanted in the animals’ cranial vault. Two electrodes
this collected information.
were implanted bilaterally in the parietal region, and
Stochastic resonance (SR) is a nonlinear phenom-
one in the frontal area for reference (ground).
enon whereby a dynamic system response can actually
Trepanation points were spotted stereotaxically and
be enhanced through noise. The parallelism between
correspond to the following stereotaxic coordinates:
SR and a possible homeopathic mechanism will be
3.8 mm posterior to Bregma and 4.5 mm lateral to the
side of central line for the bilateral electrodes.
Our general method relies on the fact that living
Stereotaxic coordinates were determined according to
beings emit electric signals; these can be detected by
the Paxinos and Watson map. Further details are given
various methods: (EEG), (ECG), plethysmography,
in previous articles.12,13 After implantation there was a
etc. Mathematical analysis (linear and nonlinear)
6-day recovery period followed by a further 1 day light
reveals slight changes, which must be systematic,
conditioning period in which the electrodes were
repetitive and objectively related to the expected
connected to a signal amplifier by a cable that allowed
physiologic outcome. We have focused on electro-
freedom of movement to the animal. The next day
encephalogram (EEG)11-13 and electrocardiogram
measurements started.
(ECG)14, as their alterations are related to relevant
health disturbances. Recording these signals during
sleep avoids bias due to possible interference of subject Animal welfare
awareness. Caffeine at pharmacological dose was Mexican Federal animal welfare guidelines were care-
selected as the high dose stimulus due to its well- fully observed during our handling of rats in this and
known effects on sleep pattern15. Caffeine is also related experiments to minimize their physiological
identified as the main alkaloid16 in green coffee beans, shock.
from which Coffea cruda is obtained.17 It is expected It is worthwhile to point out that
that Coffea cruda exhibits effects like those of
caffeine.  The electrodes do not penetrate the meninges, so
there is no invasion of brain tissue.
 In about 2 weeks, rats expel the electrodes due to
cranial regeneration; they fully recovered from the
Materials and Methods intervention and were released unharmed to lead a
The subjects were young adult male Wistar rats normal life after the experiment.
weighing between 250 and 350 g. Results discussed  Measurements took place during the animals’ sleep
arose from two caffeine administered experimental period, which implies lack of suffering.

Table 1 Administered doses per kilogram of body weight

Set Sub-set Caffeine (T0) (mg/kg) Treatment (T, admininstration time, hours)
Pre (18) Control 15 0.1 ml homeopathic solvent (T=T0+0.5)
Pre (18) Verum 15 0.1 ml Coffea cruda 30c
Pre (18) Control 15 0.1 ml homeopathic solvent (T=T0+0.5)
Pre (18) Verum 15 0.1 ml Coffea cruda 30c(T=T0+0.5)
Column 1 displays the trial set. No. of subjects in parentheses.

Caffeine-Coffea interaction
G Ruiz-Vega et al
 Throughout the experiment the animals were fed were the same in both sets (Table 1). Time zero, T0
normal rat diet and drinking water ad libitum. for records was the administration time for caffeine
and the sessions length was similar to Pre. Due to
technical reasons, first session was carried out with
Materials only five subjects; as in Pre, it was impossible to
Caffeine was supplied by SIGMA, it was diluted in complete 6 records because of caffeine-induced hyper-
double-distilled water. Coffea cruda 30c was prepared sensitivity.
by the centesimal Hahnemannian method according to
the Mexican Homeopathic Pharmacopoeia and sup-
plied by Laboratories Medicor S.A. de C.V., in 87% Mathematical Analysis
pharmaceutical ethanol. Control was 87% pharma-
The record was divided in to 1-h periods, from T0 to
ceutical ethanol.
T8. Before any mathematical operation, every file was
visually examined and noisy segments due to discon-
nection or external artifacts (body movement) were
In both Pre and Post groups, rats under experimenta-
tion were housed individually in wooden cages, built to
In this research non-REM activity, as a marker of
avoid external disturbances but allowing observation.
sleep intensity, was estimated through a parameter,
Temperature was held constant at 2970.51C, in order
considered in previous related work.12,13 It was defined
to avoid sleep pattern alterations.18
as spectral power density in 0.5–2.5 Hz band, evaluated
Each subject was connected to a 15A54 Grass
in noise-free file. The spectral power density in 0.32–
amplifier module, its controls set as follows: sensitivity
0.48 Hz band was also evaluated; power ratio between
20.0 mV/div, display gain 1, band-pass filter 0.3–35 Hz,
0.32–0.48 Hz over 0.5–2.5 Hz (delta) band and ‘slow/
line filter on. The output was sampled at 8 Hz and
delta power ratio’ was calculated. To eliminate
online digitized with a National Instruments AT-MIO-
frequencies outside the selected bands, a low-pass filter
64E-3 card and PolyView v 2.0 software. The resulting
with 2.5 Hz cut frequency was applied. The series of 1-
data was stored in a computer for mathematical
h-filtered data were subjected to a fast Fourier
analysis. Sampling frequency was chosen taking into
transformation algorithm and the chosen band power
account, the features of the delta band19 and the
was computed according to the periodogram meth-
Nyquist theorem.
od.20 Due to different lengths of noisy periods, the
records related to the subjects tested at the time were of
Experimental procedure
different sizes. In order to have a common reference in
this between-subject approach, spectral power in delta
and slow bands were analyzed as percentage of the
Six rats were tested at a time, one to a cage, three for
spectral density in the whole filtered file frequency
each sub-set. Measurements started the day after
range (0 – 2.5 Hz); proportional changes in the delta
conclusion of the preparation detailed above and
and slow bands were possible to detect by this means.
connecting them to the amplifier. The rats were fed
Hence, three markers were considered and evaluated:
their regular diet and allowed purified water without
spectral power density in delta and slow bands and
any medicament. For technical reasons, three sessions
slow/delta power ratio.
were run with only five subjects. At 10.00 the animals
received caffeine (15.0 mg/kg i.p.). All records were
gathered from time zero (caffeine administration T0)
through the 8-h sleep period of the animals (between
10.00 and 18.00). Half an hour after T0, three animals Pre
were given 0.1 ml of Coffea cruda 30c or homeopathic Results arose from between-subject analysis, compar-
solvent, blind. ing the mean value of a single evaluation in the two
All 33 animals were tested in a similar manner. groups. Two and one outliers were identified in verum
Analysis of resulting data was carried out by an and control groups, respectively. They exhibited a
investigator ‘blind’ to the specific source of the file. (delta spectral density value)>(mean72s), and were
Caffeine-induced hyperactivity caused some problems discarded. Comparison between sub-sets was made by
for six rats, we were unable to complete the recording means of ‘t’ test when applicable, (i.e. Shapiro-Wilk’s
because of frequent disconnections and intolerance of normality criteria fulfilled) if not, the Mann–Whitney
being touched for re-connection. test was used.20 Results are displayed in Fig 1a–c.
No significant differences were detected between
Post the control and verum groups from T=1 to 5 and
A scheme similar to Pre was followed; the only T=6 for delta and slow bands, respectively;
differences were the administration times. In this in the latter, P-value differences are close to the
group, Coffea cruda 30c or homeopathic solvent was statistical significance. For slow/delta power ratio,
administered 1 h before caffeine. However, doses significant differences arose also from T=6 on.

Caffeine-Coffea interaction
G Ruiz-Vega et al

* * * *

log (spectral density)




1 2 3 4 5 6 7 8 9
(a) log (time)

-1.9 VERUM
-2.0 * *
log (spectral density)






1 2 3 4 5 6 7 8 9
(b) log (time)


2.2 *
log (slow/delta ratio)

* *




1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0

(c) log (time)
Figure 1(a^c) Variation for delta band, slow rhythm and slow/delta power ratio, respectively. Verum and control follow similar trend
up to T=4. Opposite tendencies observed from T=5. Differences statistically significant appear at T=5 for delta band and T=6 for the
other two parameters.

Parameters value vs time of the three markers was from the Post control sub-set. Two normally
fitted to an analytical function. A log–log function was distributed verum sub-sets were identified in Post
chosen after testing different mathematical relations, for the three monitored parameters: Post v-A and
because of its high correlation. Variations for verum Post v-B sub-groups. The first consisted of eight
were compiled in a single expression throughout the subjects, which exhibited the same tendency as
testing period (Figure 2–c), whereas control pattern Post control for the three parameters, with no
required two equations: first from T=1 to 4, and statistically significant variation from the controls
second from T=5 to 8. Control ran according to at any time. The second sub-set consisted of
bibliography reports. In order to compare the para- seven animals, displayed an opposite tendency than
meters involved, the same log-log fitting was applied in Post v-A sub-group, showing statistically signi-
all cases. A graphic representation is displayed in ficant variation from control during the first
Figure 3–c. hours (Figure 5). Differences between Post v-A
and Post v-B sub-groups were highly statistically
significant during the entire session for the three
Post markers (Figure 4–c). The main issue in the Post
The same criterion was followed as in Pre for the group is the appearance of two sub-groups
identification of outliers and comparison between (Table 2 and Figure 5). This splitting appears
control and verum values. One outlier was discarded to be due to differences in sensitivity in response

Caffeine-Coffea interaction
G Ruiz-Vega et al
log(sd) = 4.3614 + 0.035396*log(t), r = 94%


log(spectral density,sd)




-0.2 0.2 0.6 1.0 1.4 1.8 2.2 2.6
(a) log(time,t)

log(sd)= 2.1436 - 0.1715*log(t), r = 98.5%


log (spectral density,sd)





-0.2 0.2 0.6 1.0 1.4 1.8 2.2 2.6
(b) log (time,t)

log(rat) =- 2.1631 - 0.2308*log(t), r = 98 %


log (slow/delta ratio,rat)








-0.2 0.2 0.6 1.0 1.4 1.8 2.2 2.6

(c) log (time,t)

Figure 2(a^c) Log–log fitting for mean spectral density (sd) vs time in delta band, slow rhythm and slow/power ratio, respectively,
for verum in caffeine pre-administered subjects. Linear fitting was selected because it highlights the high correlation of the
experimental data with the mathematical function. It can be re-arranged in the general form sd=C(t)a where C=Exp(4.3614) and
a=0.035396 in 2a, similar substitutions can be done in 2b and 2c. According to these results, the homeopathic stimulus effect evolves
as predicted by a power law.

to the homeopathic stimulus. Post control behavior 4 th when the greatest non-REM length is reached; a
was in line with reported results, as in Pre; in fact, decrease is observed through the following 4 h, as
overlapping between two control graphics for two of expected.
three evaluated markers, and no statistical significant Regarding, slow band and power ratio, a linear
difference for the third one, supported the experiment decline was observed from the maximum at T=1 to
replication (Figure 6–c). the 4th hour; an increase was detected from the 5th
hour on; as these oscillations are associated with
resting sleep,21 this is the expected outcome. Such a
pattern can be understood if we take into consideration
Discussion that slow rhythm has been related to processes devoted
Results in the control groups in the delta band agree to organizing memory traces acquired during wakeful-
with previous findings15, ie, it goes from a minimum, at ness,21,22 this mechanism could be reflected in the
T=1, due to several awakenings produced immedi- subject behavior modification leading to condition-
ately after caffeine administration to a maximum in the ing.23 Power ratio value increased due to: delta

Caffeine-Coffea interaction
G Ruiz-Vega et al
log(sd) = 4.2843 + 0.054219*log(t) r = 98%


log(spectral density,sd)

4.32 log(sd)= 4.3829 - 0.032*log(t) r = 83.5%

log(spectral density, sd)

4.30 4.328
4.28 4.316
1.55 1.65 1.75 1.85 1.95 2.05 2.15
-0.2 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6
(a) log(time,t)

log(sd) = 2.2817 - 0.16475*log(t) r = 99.9%


log(spectral density, sd)


2.16 log(sd) = 2.064395 + 0.025396*log(t) r = 12.6%


2.04 2.09
1.55 1.65 1.75 1.85 1.95 2.05 2.15
-0.2 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6
(b) log(time,t)

log(rat) = -1.922 - 0.25018*log(t), r = 99%

log(rat) =-2.34 + 0.111213*log(t) r = 49.5%
log(slow/delta ratio,rat)

log(slow/delta ratio,rat)

1.55 1.65 1.75 1.85 1.95 2.05 2.15
-2.1 log (time,t)



-0.2 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6
(c) log(time,t)
Figures 3(a^c) Log–log fitting for mean spectral density (sd) vs time in delta band, slow rhythm and slow/power ratio, respectively,
for Pre control. Because opposite tendencies were detected, correlation for first 4 h are shown in main figure, last four in insert. Log–
log fitting constants for first 4 h closely resemble those of verum.

band diminishing and slow rhythm enhancement. A upward tendency is detected for the delta band in
similar pattern with no significant differences at any verum, this parameter subsided in the control sub-
time was found for Pre and Post control groups, group as time passed, as expected.15 In fact, up to the
suggesting that the method is reliable and reproducible 5th hour in both groups, delta and control followed the
(Figure 6–c). same upward pattern (Figures 2a and 3a), after this
For the sake of clarity, further discussion will be time the verum group continued to exhibit an
around the marker that most characterizes caffeine increasing tendency, that is, as the homeopathic
effect, i.e. delta band spectral density. The following medicament sustained the self-initiated recovery. As-
aspects can be identified: suming caffeine–Coffea cruda interaction according to
similia principle, the latter stimulus should mirror the
Pre first and a healing effect is the expected outcome. Self-
Transient pharmacological sleep deprivation was recovery through an homeostatic response is a
induced due to caffeine administration. A persisting spontaneous process, the homeopathic effect consists

Caffeine-Coffea interaction
G Ruiz-Vega et al


log (spectral density)


Post control
Post verum-A
4.26 Post verum-B

* * *
* *
1 2 3 4 5 6 7 8 9
(a) time

2.4 * *
log (spectral density)



Post control
Post v-B
Post v-A
1 2 3 4 5 6 7 8 9
(b) time


-1.8 * * *
log (spectral density)




Post control
-2.8 Post v-B
Post v-A
1 2 3 4 5 6 7 8 9
(c) time
Figures 4(a^c) Variation for delta band, slow rhythm and slow/delta power ratio respectively in caffeine post-administered subjects.
Mean spectral density values, log (percentages) vs time (hours, log scaling). Differences statistically significant between control and
Post v-B, marked. No differences were detected between control and Post v-A at any time, ie evolution is similar to control.
Differences between Post v-A and Post v-B sub-sets were statistically significant throughout the recording period, highlighting
individual response to homeopathic stimulus.

in a synergic effect on host response strengthening the homeopathic stimulus: Post v-A, did not exhibit any
self-initiated recovery. change in the chosen parameters whereas Post v-B
displayed lower values than control in the first 4 h for
delta band spectral density. This initial decrease in
Post sleep intensity represents an opposite effect than in Pre
In Post, the splitting of the population in two sub- verum. It corresponds to reported effects of caffeine
groups presumably reflects different sensitivity to the plus additional sleep deprivation15, as if the homeo-

Caffeine-Coffea interaction
G Ruiz-Vega et al
Table 2 Post v-A and Post v-B means and variances comparison
HORA P-Mean F ratio P variance
1 0.00500 17.82 0.001290
2 0.00006 38.90 0.000190
3 0.00110 33.28 0.000168
4 0.00100 171.01 0.000001
5 0.01000 34.65 0.000147
6 0.01100 33.52 0.000164
7 0.02800 8.41 0.012822
8 0.01000 15.53 0.001995
P-values from ‘t’ or Mann-Whitney tests (3, 7 and 8 h) are listed in second column. Column 3 shows the variance ratio, Post v-B over Post v-
A. Last column displays variance P-values. Two different populations can be identified.

Postv-B, y = 7 * 5 * normal (x, 64.64077, 9.0219745)

Post v-A, y = 8 * 5 * normal (x, 84.84002, 1.446581)
11 Post v-B
10 Post v-A
No of observations

<= 50 (50,55] (55,60] (60,65] (65,70] (70,75] (75,80] > 80
Sprectral density, percentage
Figure 5 Normal fitting general pattern for Post v-A and Post v-B frequency distribution. Two different populations can be identified,
as the only two required parameters to completely describe the normal distribution, the mean and standard deviation are statistically

pathic medicament had enhanced the effect of caffeine. spring, the Newton’s Law of Gravity, the healthy heart
There are similar reported findings in this field24 rate dynamics, membrane channel openings, human
regarding ‘superimposed’ action of homeopathic med- writings, biological evolution, economics, music, earth-
icines, the so-called ‘priming effect’8 Differences quake structure, etc. which suggest that SOC might be
between Post v-A and Post v-B persisted throughout an universal phenomenon.25 From this point of view,
all the recording period, reflecting differences in the systems would evolve from one critical state to
subjects sensitivity to homeopathic stimulus (Table 2 another one according to this general pattern.
and Figure 5). The main implication of this relationship consists in
There is a high coefficient of correlation in the log– its self-similarity, which means: if ‘x’ is rescaled
log fitting for the three selected markers (i.e. delta and (multiplied by a constant), the f(x) is still proportional
slow spectral density and slow/delta ratio) vs time in to (x)a, although with a different constant of propor-
Pre verum (Figure 2a–c). tionality, C.26
The pattern was identified as y=C(t)a, (where y is According to this model, it seems that the homeo-
the spectral density or power ratio, t, time and C and a pathic medicine drives the system, in accordance to
are constants. This kind of relationship reflects a widespread natural laws, between critical states.
tendency to self-organized critically (SOC).25,26 The Moreover, as self-similarity implies scale-free, (i.e.
tendency to self-organization appears to be a universal the pattern is the same irrespective of the scale) this
tendency of systems that evolve from an initial critical pattern can explain the cyclic appearance of symptoms
state to another. SOC is associated to the tendency of through days, weeks, months, seasons, or annually,
large dissipative systems to drive themselves to a when two suitable pairs of variables (x, y) are
critical state. Under this condition, systems, as whole involved. Indeed in the homeopathic Materia Medica
entities, exhibit power laws, ie relationships between many medicines have aggravations daily at the same
two correlated properties (x, y) that could be fitted to time.
the form y=C(x)a, C and a being constants. Power Regarding the possible mechanism which a system
laws are ubiquitous in nature; they have been identified might enhance its own response, we propose a
in diverse fields, eg the restoring force in a linear parallelism with SR. SR is a cooperative nonlinear

Caffeine-Coffea interaction
G Ruiz-Vega et al


log (spectral density)




1 2 3 4 5 6 7 8 9
(a) time


log (spectral density)





2.04 Post
1 2 3 4 5 6 7 8 9
(b) time


log (spectral density)




1 2 3 4 5 6 7 8 9
(c) time
Figures 6(a^c) Variation for delta band, slow rhythm and slow/delta power ratio respectively for Pre and Post controls. Scaling as in
preceding figures. Similar evolution according to reported findings with no significant differences at any time between two sets. This
supports the reproducibility of the method.

phenomenon wherein the signal-to-noise ratio (SNR) complete mathematical model can be found in
of a noisy nonlinear system, suitable to be in two stable reference27.
states, driven by a weak deterministic modulating Parallel terms can be identified in the homeopathic
signal, can actually be enhanced by increasing effect:
the noise. Switching between the two states would
not be possible in the absence of noise, correlations  The noisy nonlinear system would be the living
in the noise also affect switching probability. A being, which could be in ‘healthy’ or ‘sick’ states.

Caffeine-Coffea interaction
G Ruiz-Vega et al
 Living beings exhibit diverse biological cycles at
different levels: membrane potential, metabolic
oscillations, enzymatic reactions, protein synthesis,
cellular communication by pulses, hormonal sig-
naling, among others.28 Hence, the weak deter-
ministic modulation is self-emitted by the system,
and we propose it could be enhanced via noise.
 Noise is ubiquitous in the nervous systems of living
beings. We suggest that certain kinds of noise (in
moderate amounts)are related to sick or healthy
states. The so-called ‘pink noise’26 would be
associated to healthy state, as a resemblance of
‘yall the parts of the living being in a harmony’;29
disorder would be related to white noise. Internal
noise features would be modified in the sickness
condition: the symptoms associated to each illness
would imprint specific characteristics that could be
related to the ‘terrain’ of the French literature,3
noise also could be associated with the influence of Figure 7 Representative diagram of SNR vs noise intensity.
N* represents noise level at maximum. According to our
the environment.27 Homeopathic stimulus would hypothesis, a similar diagram could be constructed from
modify noise through ‘constructive interference’,30 treatment-to-control ratio measurements recorded over a wide
which means to add point by point, the signal that noise density range, ie different homeopathic medicine
potency. A specific potency, N*, corresponds to the SNR
would be brought by homeopathic medicine to maximum for particular patient/disease characteristics. Be-
internal noise. In order to get an enhancement, two cause of their own noise density, two subjects condition can
signals ought to match perfectly. The homeopathic be detected: For subjects to the left of N*, an enhancement in
SNR can be expected through overlapping the correct
term potency could be related to noise intensity. medicine signal, increasing noise density. For subjects
located to the right, because of the high noise density,
The complete hypothesis would be as follows: enhancement of the auto-initiated response will not be
possible and no homeopathic effect would be expected.
Living beings oscillate between two meta-stable
states: health or sickness. When a healthy subject is
perturbed and moved from the healthy to sick state,
symptoms appear at different levels, eg mental, general, produce a different response to the same stimulus.
particular, modifying internal noise features, eg High sensitivity would relate to the maximum, N*in
intensity, correlation. Simultaneously, cyclical signals, Figure 7. Moreover, in specific neuronal noisy mod-
attempting to return the system to healthy state, are els,33 the output (subject response) at maximum
emitted by the autonomous system. Homeopathic sensitivity is extremely sensitive to only small mod-
medicine promotes the enhancement, via noise, of ifications of driver parameters (71% of the original
auto-initiated signals by superimposing a transient value);33 hence, there is no enhancement possible when
illness, which matches that already present. This is the the system location is slightly moved to the left or right
reason why the symptoms of the chosen medicine of N*(Figure 7). Hence, subtle differences (not
should correspond exactly to those of illness.31-32 The necessarily macroscopic) could be the origin of
term vital force could be related to internal noise differences in the response intensity, which in turn,
features, intensity and the distance from pink noise (ie, could explain population splitting seen in our Post
from harmonic behavior), and would reveal if the verum group (Figure 5). Behavior of SNR as a
system will be capable of enhancing its self-emitted function of noise could be taken as the signature of SR.
signals, ie, to promote its self-recovery. In this At last, but not least important, is to highlight a
approach, it could happen that in particular situations, parallelism regarding the capability of a system,
with very large amounts of noise, the cooperative through a SR mechanism, to adjust itself to operate
behavior that identifies SR is lost (to the right of close to the maximum of the SNR curves that is,
Figure 7), when homeopathic medicine is not able to maximizing the system response, as a self-optimization
promote self-healing. process. This principle matches Hahnemann’s ideal ‘y
Homeopathic pathogenetic trials (provings) could restore the health in the shortest, surest, least harmful
also be explained: Because homeopathic medicines are wayy’29.
applied to a healthy subject, the noise imprint prevails
transiently over the steady-state parameters such as
noise and signal may play interchangeable roles.27
Different susceptibility of subjects could be related to
density noise fluctuations; then it is expected that even This research was possible due to a grant from the
subtle physiological changes in the subject would Scientific Research Council from the University of

Caffeine-Coffea interaction
G Ruiz-Vega et al
Michoacan. Guadalupe Ruiz-Vega thanks Rafael 16 Kirk RE y Othmer DF. Enciclopedia de tecnologı´a quı´mica,
Villalobos-Molina and Bernard Poitevin for their Tomo 3. México: UTHEA, 1961.
comments. She is also grateful to Daniel Garcı́a- 17 Farmacopea Hpmeopática de los Estados Unidos Mexicanos.
Secretarı́a de Salud, México: 1998.
Garrido and Miguel Garcı́a-Silva for their help in 18 Borbély AA, Tobler I, Hanagasioglu M. Effect of sleep
editing. deprivation on sleep and EEG spectra in the rat. Behav Brain
Res 1984; 14: 171–182.
19 Rechtschaffen A, Kales A (eds). A Manual of Standardized
Terminology, Techniques and Score System for Sleep Stages of
References Human Subjects. Bethesda, MD, USA: Department of Health,
Education and Welfare, 1968.
1 Linde K, et al. Are the clinical effects of homeopathy placebo 20 Statistica. User Manual. Tulsa, OK, USA: StatSoft, 1995.
effects? A meta analysis of placebo-controlled trials. Lancet 21 Steriade M, Amzica F. Coalescence of sleep rhythms and their
1997; 350: 9081: 834–843. chronology in corticothalamic networks. Sleep Res Online
2 Ernst E, Hahn EG, (eds). Homoeopathy a Critical Appraisal. 1998, 1: 1–10.
Oxford: Butterworths-Heinemann, 1998, pp101, 145. 22 Steriade M. Impact of network activities on neuronal
3 Poitevin B. Introducción a la Homeopatı´a: Bases experimentales properties in corticothalamic systems J Neurophysiol 2001,
y cientı´ficas. México: Nueva Editorial Homeopática Mexicana 86: 1–39.
SA, 1992. 23 Williams DA, Hurlburt JL. Mechanisms of second-order
4 Belon P (ed). Investigación en Homeopatı´a: Publicaciones y conditioning with a backward conditioned stimulus. J Exp
comentarios. Sainte-Foy-Les Lyon: Editions Boiron, 1999. Psychol Anim Behav Process 2000; 26: 340–351.
5 Anagnostatos GS, Vithoulkas G, Garzonis P, Tavouxoglou C. 24 Bertani S, Lussignoli S, Andrioli G, Bellavite P, Conforti A.
A working hypothesis for homeopathic microdiluted remedies. Dual effects of a homeopathic mineral comples on
Berlin J Res Hom 1991; 1: 141–147. carrageenan-induced oedema in rats. Br Hom J 1999; 88:
6 Towsey MW, Hasan MY. Homoeopathy-a biophysical point 101–105.
of view. Br Hom J 1995; 84: 218–228. 25 Bunde A, Havlin S (Eds). Fractals in Science. Berlin: Springer,
7 Torres JL, Ruiz MG. Stochastic resonance and the 1994.
homoeopathic effect. Br Hom J 1996; 85: 134–140. 26 Schroeder M. Fractal, Chaos, Power Laws. New York: W.H.
8 Bellavite P, Lussignoli S, Semizzi ML, Ortolani R, Signorini A. Freeman and Company, 1991. pp 122, 103.
The similia principle from cellular models to regulation of 27 Millonas M (Ed). Fluctuation and Order. New York: Springer,
homeostasis. Br Hom J 1997; 86: 73–85. 1996.
9 Bonamin LV, Martinho KS, Nina AL, Caviglia F, Do Rio 28 Montero F, Morán F. Biofı́sica: Procesos de autoorganización
RGW. Very high dilutions of dexamethasone inhibit its en biologı´a.
pharmacological effects in vivo. Br Hom J 2001; 90: 198–203. 29 Hahnemann S. Organon of Medicine. Blaine, Washington:
10 Belougne-Malfatti E, Aguejouf O. Combination of two doses Cooper Publishing 1996 (English traslation of the 6th edn of
of acetyl salicylic acid: Experimental study of arterial the original work).
thrombosis, Thromb Res 1998; 90: 215–221. 30 Madrid: EUDEMA, SA, 1992 pp. 222.
11 Ruiz G, Torres J-L. Homeopathic effect on the sleep pattern of 31 Hecht E. Optics. Reading, MA: Adison-Wesley, 1974.
rats. Br Hom J 1997; 86: 201–206. 32 Vijnovksky B. Tratado de Materia Médica Homeopática.
12 Ruiz-Vega G, Pérez-Ordaz L, Proa-Flores P, Aguilar-Dı́az Y Buenos Aires: Macagno, Landa y Cı́a. SRL., 1978.
An evaluation of Coffea cruda on rats. Br Hom J 2000; 89: 33 Vannier L. La Práctica de la Homeopatı´a. México: Editorial
122–126. Porrúa, 1991
13 Ruiz-Vega G, Pérez-Ordaz L, León-Huéramo O, Cruz- 34 Huber MT, Krieg JC, Dewald M, Voigt K, Braun HA,
Vázquez E, Sánchez-Dı́az N, Comparative effect of Coffea Stimulus sensitivity and neuromodulatory properties of noisy
cruda potencies on rats. Homeopathy 2002; 91: 80–84. intrinsic neuronal oscillators. Biosystems 1998; 48: 95–104.
14 Ruiz G, Torres J-L, Michel O, Navarro R. Homeopathic effect 35 Barr M, Kiernan J. The Human Nervous System, 6th edn.
on heart rate variability. Br Hom J 1999; 88: 106–111. Philadelphia: J.B. Lippincott Company, 1993.
15 Schwierin B, Borbély AA, Tobler I. Effects of N6 - 36 Bethea RM, Duran BS, Boullion TL, Statistical Methods for
cyclopentyladenosine and caffeine on sleep regulation in the Engineers and Scientists, 2nd edn. New York: Marcel Dekker
rat. Eur J Pharmacol 1996; 300: 163–171. 1985.