1

Experiment A. Problem: If a series of syllables is presented within the range of clear vision alongside of a second series which is to be learned, and engages the observer's attention, to find whether such a side series to which no especial attention has been given will later be more easily learned because of its exposure to possible "unconscious" perception. If Sidis's and Scripture's experiments are valid in their results, it was judged that we should find evidence of "subconscious' learning in the more ready memorizing of series which have been so exposed.

Three of the above mentioned twelve-syllable series were written in three vertical columns, so that the first syllable of each series appeared to the observer through the slit at the same time, then, as the drum turned, the second syllable of the three series together, and so on to the end. These series we shall call a, b, c, reading from left to right, b being the middle series, and the one learned. On a second strip were written three series, d, a, e; d and e being entirely new series and a, the one which had stood to the left of b, the first series learned. On a third strip was written an entirely new set of three series, f, g, h; on a fourth strip, i, h, j, of which h had stood to the right of g, the series previously learned. All ten of these series were "verschärft" series, and the ten together constituted a regular set.

The number of series which could be used for each subject was, from the nature of the case, limited (11 sets for S and 12 for E). The method of obtaining syllables enough for all the experiments and avoiding familiarity with the syllables as much as possible was this. Syllable series which had been side syllables for E, but not learned (there were 6 such in every set), were transposed so that they occupied the places of b, a, g and h which had been learned, and were then given to S, and vice versa. Thus, if series retained their former lettering they would read j, i, h; g, j, f; e, d, c; b, c, a. After each subject had learned these transformed series, E. was given the transformed series for S, and S for E. Then each was given the series the other had learned first, but the syllables of the individual series were shifted. This made a possibility of 552 new syllables for each subject. When these had been run through, the second and third permutation (which the subject had learned before) was repeated.

An extra series (indicated in what follows by x) was also used each day and was obtained by taking a strip from an unused set and learning the middle series. One set thus divided furnished extra series for four days. This extra series was given sometimes first, sometimes third and sometimes fifth in the order of the day, each subject being kept in ignorance of its position, so that he was never aware of which series he was learning. The order which was maintained for the learning was, therefore, either: x, b, a, g, h, or b, a, x, g, h; or b, a, g, h, x.

The odd numbered syllables of the middle series of each strip were written in the centre of a slip fitting the "Treffer" drum; here, therefore, there appeared through the slit only one syllable at a time instead of three. For the first three days, the "Treffer" order was given according to the permutations of Müller and Pilzecker' but afterwards the "Treffer" syllables were written in a direct order from one to eleven, as it was decided that a varied scheme would give less uniform results with a moderate number of experiments, and would not in any case influence the point at issue in the experiment.

The average time taken for ten revolutions of the drum during the practice work and for two days of regular work was 87.1 seconds for ten revolutions or 0.62 + seconds per syllable. This speed was accelerated on March 10 (the third day), and from then on was 80.2 sec. per 10 revolutions or 0.57 + second per single syllable.

1 The middle series is always the one learned in these experiments.

Müller u. Pilzecker: Experimentelle Beiträge zur Lehre vom Gedächtniss, Zeit. f. Psych. Erg. Bd. I, 1900.

Each observer served as experimenter when the other was observer. Both knew the object of the experiment, but neither knew the results until the experiment was finished, as no accounts were cast until the end. With the exception of the first two days of regular experimentation, when E. worked at 2 P. M., E. served as subject from 8 until 9 A. M. and S. from 12 to 1 P. M. (The time for E. was changed because syllables which S. had recited at 12 persisted and acted as a disturbing factor.)

From January 24, 1909, to March 7 practice work was carried on every day except Sunday and four days of the Easter Vacation. Regular work began March 8 and was carried on every day except Sunday and one day when S. was out of town and another day for E. when the machine was out of order, making 25 days of regular work for S. and 24 for E.

At the words "In your place' the observer took his seat before the machine. This was started and the drum allowed to make one revolution to get its speed; as the first blank space appeared the experimenter called "ready," and lowered a shutter previously hiding the drum. The observer began to read as soon as the first syllable appeared, reading through the whole series of twelve syllables. After the two blank spaces had passed the series was read through again, and so on, until it had been repeated twelve times. The syllables were read pair-wise, at first in trochaic rhythm; later both observers fell into the iambic. When the last syllable was read on the twelfth revolution, the experimenter raised the shutter and started a stop watch. The observer began immediately to turn the "Treffer'' drum, which was adjusted so that a single blank space preceded the first "Treffer'' syllable, and read each "Treffer" syllable as it came up, giving, if possible, its associated syllable (i. e., the syllable which had formed the other half of the pair); when he could not recall a syllable he said "don't know'' and passed on to the next. When the last associated syllable had been given (or given up) the stop watch was stopped. This made a somewhat rough method of timing, but was effective enough for this experiment in which time was only a minor consideration. The observer was neither hindered nor helped by the fact that he was being timed. A list of each series to be learned had previously been written in a blank book and opposite each syllable was placed a check mark, if the correct syllable were given, a dash if none came, or the syllable which was given if a wrong one was given. After testing his associations with the "Treffer" syllables the observer's introspections were taken on such items as the difficulty of the series, the conditions of his attention, and influences which might have favored or been disadvantageous to the learning or reproducing of the series. The whole process-repeating the series, giving the associated syllables and the introspections-took, on an average, two and a half minutes. Ten minutes after the first series was begun, the experimenter again called, "In your place," and began the second series by starting the drum and, at a ready signal, letting down the shutter. Between the learnings a free time of about seven and a half minutes elapsed. In this interval the observer was allowed to relax as he pleased, either in looking over books, walking around the room, or gazing out of the window. But any taxing occupation which absorbed the attention to a considerable degree was avoided as it was found to have an unfavorable effect on succeeding series.

In computing the results a unit was allowed for each perfect syllable, 6 therefore being the score for a perfect record, i. e., the recall of each of the six even numbered syllables. An average was taken for each of the series for the whole period of the experiment (24-25 days). A modified average was also made in order to include partial successes as follows: One-third was given for each vowel, diphthong, consonant or double consonant correctly given. Thus a syllable having only the vowel correct would score and one having its two consonants or a consonant and the vowels, .

Fig. III. In this figure the screen of the upper drum is arranged to show but a single syllable, an arrangement used in Experiments B and C below. In Experiment A it was open full width horizontally and showed three syllables at a time. The treffer drum is seen below behind the black cardboard screen.