Consciousness

something is qualitatively altered in hypnosis (see Chapter 13),
this change in consciousness helps. Many more experiments
would be needed to establish which of a potentially unlimited
number of untested variables are relevant. The connection
with consciousness seems to come entirely from theoretical
suppositions about how ESP might work. There is no direct
evidence that consciousness is involved in any way (Blackmore,
1998). We have considered evidence in earlier chapters that the
influence of conscious will on our own actions, attention, and
perception, let alone anyone else’s, may be illusory. So maybe
we should expect no such involvement, whether in terms of
the power of conscious will or the power of consciousness per-
ception beyond the material.

CONJURING OTHER WORLDS

All of us conjured other worlds when playing as children: invent-
ing food and drinks for dollies’ tea-time, imagining illnesses to
be cured by ‘doctors and nurses’, and creating invisible cargoes
to be carried by toy trucks on imaginary roads. Many children,
especially only children, have imaginary playmates. Some chil-
dren play and talk with the same friend for many years, though
not often past the age of ten. In the early years, the playmates
are described as solid and real, but older children rarely see
them that way. Most imaginary companions are people, usually
of the same sex as the child, but they can be animals, invisible
toys, storybook characters, and even things like clouds or door-
knobs (Siegel, 1992). These friends take part in conversations,
games, and all sorts of creative activities.

Pretend play is crucial to how children develop their causal
understanding of the physical and mental world. In one series
of experiments, two-year-olds watched Naughty Teddy vic-
timising other toy animals with make-believe substances: for
example, squirting pretend toothpaste on to a rabbit’s ear (Har-
ris, 2000, pp. 17–19). When asked to describe what they saw,
the children referred to the pretend substances and actions
(‘toothpaste’ and ‘squeezed’) and the consequences of those
actions (making the rabbit’s ear ‘dirty’ or ‘wet’) despite having
the objectively dry and clean rabbit in front of them. Then a
brick on a paper plate is pushed towards the rabbit and chil-
dren are told the rabbit likes eating banana. This time Naughty
Teddy squeezes toothpaste onto the brick instead of the rab-
bit, and the children didn’t just infer a non-existent substance
(toothpaste) but also swapped the real object’s name (brick) for the make-believe
name (banana); if they failed to do this, they almost always just pointed to the
brick or remained silent. This suggests that they knew the causal outcome was
directed at the object the prop stands for, not the prop itself. These imaginary
worlds are robust, but they also obey the same causal laws as the real world

at the correct times and the receivers write down which

suit they think the sender is looking at.

When the test is complete, ask the sender to return.

Call out the target sequence and ask each person to

check their neighbour’s scores. If you have a large

enough group (say twenty or more), you can show the

results by building up a histogram for all to see. Ask

each person in turn to say how many hits they got,

and add each result to the growing picture. At first the

results may seem impressive, or strange, but they will

tend ever closer to a normal distribution with a mean

at 10. If the results deviate from 10 and you wish

to test them statistically, use a normal approximation

to binomial, or a one-sample t-test using 10 as the

expected value (but see below).

This method avoids most of the obvious problems, but

some remain, including the ‘stacking effect’, which

means that t-tests are not wholly appropriate when

many people guess at the same target list. Sensory

leakage or fraud might still have taken place, and you

may discuss whether they could ever be completely

ruled out. Other easy psi experiments, with more

detailed instructions, can be found in Blackmore and

Hart-Davis (1995).

8 7 6 5 4 3 2 1 0

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

Number of correct guesses

Number of people

FIGURE 14.9 • Sample histogram of results from a simple

ESP experiment. Unless psi is operating in

your experiment you are likely to get a normal

distribution with a mean of 10. As you gather more

data, the histogram approaches more closely a

normal distribution.