efulness with which I am accustomed to carry on the experiments in
the Harvard Psychological Laboratory.

To give some illustrations of sources of error, I may mention that the
earlier observers were convinced that Beulah could not see slight
movements of the persons in the room when she was looking fixedly at
the ceiling, or that she could not notice the movements of the sister
or the mother when she was staring straight into the eyes of the
experimenter. Any psychologist, on the contrary, would say that that
would be a most favourable condition for watching small signs. He
knows that while we fixate a point with the centre of our eye we are
most sensitive to slight movement impressions on the side parts of our
eye, and that this sensitiveness is often abnormally heightened. Just
when the child is looking steadily into our face or to the ceiling,
the outside parts of her sensitive retina may bring to her the visible
unintentional signs from her sister or mother. The untrained observer
is also usually unaware how easily he helps by suggestive movements or
utterances to the other observers. When Beulah gave a six instead of a
nine, one of our friends whispered that she may have seen it upside
down in her mind, or when she gave a zero instead of a six that it
looked similar. In short, they keep helping without knowing it. Very
characteristic is the habit of unintentionally using phrases which
begin with the letters of which they are thinking. The letter in their
minds forces them to speak words which begin with it. If they start at
a C, we hear "Come, Beulah," if at a T, "Try, Beulah," if at an S,
"See, Beulah." It is very hard to protect ourselves against such
unintended and unnoticed helps. It is still more difficult to keep the
failures in mind. The eager expectancy of hearing the right letter or
number from the lips of the child gives such a strong emphasis to the
right results that the wrong ones slip from the mind of the hearer.
The right figure may be only the third or the fourth guess of the
child, but if then the whole admiring chorus around say emphatically
at this fourth trial that this is quite right, those three wrong
efforts which preceded fade away from the memory. I may acknowledge
for myself that I was mostly inclined to believe that the number of
the correct answers had been greater than they actually were according
to my exact records. For all these reasons I had the very best right
to disregard the rep