Problem: The ‘induced reproduction’ here meant is reproduction caused by remote association, which is, to the writer's belief, due to the same mechanism as that of psycho-physical induction on the level of perception.
In his pioneer work entitled ‘Uber das Geddchtnis,’ Ebbinghaus has shown that, in learning a series of syllables, associations are formed not only between immediately adjacent items, but also between those which are not immediately adjacent, the strength of this remote association declining with the increasing degree of remoteness.
The purpose of the present investigation is to attack the same problem from a wholly different angle, studying it by the recall method instead of by the saving method devised by Ebbinghaus. The associative prosesses examined in the following experiments are of various degrees of complexity. They range from a simple process such as an associative response to a single digit, to more complex activities such as the constructing of nonsense syllables, the free-association-test following the Kent-Rosanoff's technique, etc. As we shall see in the following report, all these phenomena can be deduced from the principle of induced reproduction discovered in simple association experiments.
Experiments and Results: Experiment I: Experimental procedure is as follows; subjects were requested to write a digit between 1 and 9 when they heard a stimulus digit (digits used as stimuli are those from. 1 to 9). Instruction was given to avoid mere repetition of a presented digit as well as an immediate sequence in number order suck as 5 (stimulus)6 (response) or 5.(stimulus)4(response)
A large number of responses was collected by successively , presenting a series of digits. Each response was scored for its serial position relative to a stimulus digit; that is, the remoteness of each response digit from a stimulus one was computed. The data were then massed in tabular form and the graph was constructed from this table. Inspection of Fig. 1 in the Japanese text p. 336 shows that the strength of the reproductive influence declines with the increasing distance of the numbers from one another. This finding is in harmony with the most direct expectation from Ebbinghaus experiment.
Experiment II: The same thing holds true of the alphabet (a, b, c) and the musical scale (
do, re, mi, fa) materials. See Table 2 (Fig. 2) and Table 3 in the Japanese text. We thus see how the recall of an image of various sorts may be conditioned by the principle of induced reproduction.
Now, it is clear that, if a term ‘numerical distance’ be substituted for a ‘spatial distance,’ the present results reveal a close resemblance between the induced reproduction and the psycho-physical induction on the level of perception.
Thus the reproductive and perceptual processes must be deducible from a single set of assumotions linking both fields.
Experiment III: In the experiments described thus far, the remote association has been tested with a single-letter-stimulus. In the following experiments an attempt was made to test the concept with combined stimuli such as pt, nb, etc. or 27, 83, etc. These were presented in the same manner as in the previous experiments. Subjects were asked to write down two other letters when they heard a stimulus syllable (two-letters-syllable).
An associative response in this experiment becomes, as might be supposed. somewhat complicated, since new associative connections in addition to the former one come into play. There is good reason to believe that every letter in a response syllable is, in some manner, connected with every other letter of a stimulus syllable. The most important of these, however, is that which leads from an initial letter to a final one in a response syllable. We may briefly show the situation as follows: stimulus response
The experimental result was summarised in Fig.
抄録全体を表示