Let us consider what happens during the experiment in the mind of the person who has read the sentences previously. When one of the words is read out, e.g. "carpet," an association takes place, perhaps "room," in a normal time; and so on for several words. Then a critical word included in the sentence is read out, e.g. "die." The first thought in the mind of the victim will very likely be "Paris," but to say Paris would betray that she had read the paper. So there is hesitation and confusion, and the reaction time is lengthened considerably through the necessity of finding another word, the mind hovering fixedly sometimes over the response word, which must not be given. In most instances, no doubt, it is the mere necessity of rejecting the first association that causes delay; but frequently the delay is obviously accentuated by confusion and mild excitement, which is, indeed, sometimes sufficiently great to cause the patient to blurt out after all the tell-tale word.

The emotional disturbance is still more probably the cause of delayed response in the following illustration of the use of the word-association test, in which it was used by Dr. Jung for the detection of crime.

In a certain hospital a theft was committed. A nurse complained of having been robbed of 70 francs. Circumstances made it almost certain that the theft was committed by one of three nurses. The pocket-book stolen was of dark-reddish leather, and contained a small silver chain, a stencil, and a receipt from a shop in Zurich. Apart from the plaintiff and the guilty person, only the head nurse knew the exact particulars of the theft. Dr. Jung, who performed the experiment, tested each of the three nurses with a long list of words, in the midst of which were inserted such words as "leather," "dark-reddish," "chain," "stencil." These words he calls the "critical" words. He says: "I first undertook the experiment with a friend of the head nurse, and judging by the circumstances, she appeared to be only slightly moved (we will call her Nurse A). The head nurse was next examined; she showed marked excitement, her pulse being 120 per minute immediately after the experiment. The last to be examined was the nurse who attended to the cleaning of the room in which the theft occurred. She was the most tranquil of the three; she displayed but little embarrassment, and only in the course of the experiment did it occur to her that she was suspected of stealing, a fact which manifestly disturbed her towards the end of the experiment.

"The general impression from the examination spoke strongly against the head nurse. It seemed to me that she evinced a very 'suspicious,' or, I might almost say, 'impudent' countenance. With the definite idea of finding her the guilty one, I set about adding up the results."

The "critical" words were selected, and the average length of the reaction time of those words found in the case of each nurse. Then the average reaction time was found for the ordinary words, and it was found that the reaction time for the critical words was much longer than that for the ordinary words in the case of Nurse A, the friend of the head nurse, though the general excitement of the head nurse resulted in the reactions to the ordinary words being longer than were those of Nurse A. Dr. Jung made various other calculations which we need not discuss, except to say that they all pointed to Nurse A being the guilty one. The same evening Nurse A made a complete confession of the theft.

A number of other experiments of a similar type have been made. In such cases the "critical" word tends to rouse emotion in the guilty persons, and this emotional disturbance results in confusion, and delay. Even if there is no delay, the guilty person is likely to give himself away by giving in response a word which shows that he has special knowledge of the details of the crime.

If the slight emotional excitement, accruing in such a simple test as my own, described on page 58, is enough to increase reaction-times, we can imagine what would be the case when an actual criminal is being tested. Of course, an innocent man might be emotionally excited owing to being suspected, but provided he was ignorant of the special conditions of the crime his excitement would not affect the critical words more than the other words. If the criminal understands the nature of the test, he may try to avoid detection by preparing his response word before each word is given to him. He may thus avoid prolonged reactions to the critical words, but the irrelevance of most of his associations is likely to reveal itself. It is not suggested that this method will prove an adequate means of detecting crime in the absence of more definite evidence, but it would seem possible that it may in some cases afford valuable preliminary evidence, certainly more reliable psychologically than some of the absurd tests in which a witness has to pick out of a group of a dozen the man she thinks she saw near the scene of the crime, when she has already seen a photograph of the suspected criminal, as was required in a recent trial for murder.

Now it is easy to understand the delayed reactions in these and similar experiments in which the causes of the delayed reactions, namely, the inhibiting feelings and ideas, are conscious. But what happens in the case of patients who have forgotten the original disturbing experiences, and are unconscious of any reasons for inhibiting the recall of the associated ideas? We may suppose that in some way the critical words rouse hidden emotions, through their connection with the forgotten ideas which have been repressed, and with the feelings attached to them. Sometimes, when patients are subjected to word-association tests, it is evident that there is a rousing of some emotional experience, vague and incomprehensible perhaps, but enough to cause that delay in the association process which suggests the buried "complex" - the word used to denote a group of repressed ideas with their associated feelings.