Jungian Automated Word Association Test

Alexis Champion, PhD and

Bernard Auriol, MD

(Institut Métapsychique International)

 

Champion, A. and Auriol, B.(2002), Le Test d’Association de Mots Automatisé de Jung (TAMAJ), Research in parapsychology, Abstracts and Papers from the forty-fifth Annual Convention of the Parapsychological Association, 385-389.

 

 

Abstract

 

 

Spontaneous  cases of telepathy are generally linked to emotionnal targets even though Delanoy (1988-1989) shows that experimental findings are somewhat equivocal. A way to enhancing this parameter is, leaving behind general ideas, to focus our efforts on individually specific targets. To do so, we propose to use a Automated Jung's Word Association Test.

In the original Jungian test, the experimenter uses a  list of  « inducer words » chosen at random. The experimenter invites the subject to react to each inductive word as quickly as possible by pronouncing only the first word which is elicited in his mind.

In the Automated Test, the subject is told to look at the display of a computer where he/she will see the words which are displayed in turn.; then he/she has to answer as quickly as possible and the vocal answer is recorded.

After the test, the software sorts the answers by duration of reaction time.

After Jung, when a word interests only the surface of the conscience, the reaction is quick; but when, the word goes through the intimacy of ego, it determines a disturbance that prevents the subject complying the instructions given : A longer reaction time occurs.

 

Parapsychology is, from the beginning, confronted to the paradoxical Psi-missing as a counterpart of the Psi-Hitting. This odd thing is probably the clue of a mechanism which is necessary to mark borders between individuals and make the language possible. Otherwise, it appears the ESP is not under the mastering of our conscience, it comes as it wants ! Intention is not required and, its presence is not of interest !

 

We assume the hypothesis that the « long delay words » will give better results in term of transmission (but not necessarily as successes). They would provide more significant results (psi-hitting AND psi-missing) while « short delay words » would be at risk to give « psi-nothing ».

 

The results of our experimental preliminary ESP study seem to be encouraging : The proportion of series at the both tails of distribution is a little bit smaller for short delay words than for long ones. If this tendency was confirmed, it would be interesting to use long delay words as potential targets for future experiments !

 

Under the hypothesis of a better significance for words with long delay, we hope to have discovered a very simple and easy to use instrument to select potential targets for ESP experiments. The main interest is to provide a personalized tool, more precise than general idea (affective targets better than neutral ones) parapsychology has already got.

 

 

TEXT

In this study we hope to show how interesting is the use of a very old test for some parapsychological experiments. I mean the Word Association Test created by Jung, which is a completely different approach as the famous MBTI.

 

Jung's Word Association Test

 

 

In this test, the experimenter uses a  list of  « inducer words » chosen at random. If two words have a common semantic field, one of them is suppressed, as a necessary condition to get mere associations. The experimenter invites the subject to react to each inductive word as quickly as possible by pronouncing only the first word which is elicited in his mind. For example, to the word water, launched by the experimenter, the subject will answer « wet, or green, or H2O, or washing », etc.

 

The experimenter measures the reaction time with a stop watch which indicates the fifth of second (a larger precision would be useless). The stop watch is started each time the last syllable of the inductive word is pronounced and it is stopped as soon as the subject says the first syllable of the induced word. The time thus measured is the reaction time. It is enough to use fifty to hundred inductions to avoid tiredness.

 

One of the possible lists of inducer words (created by C.G. Jung, himself)

« Before the experiment begins the test person receives the following instruction: "Answer as quickly as possible the first word that occurs to your mind." This instruction is so simple that it can easily be followed by anybody. The work itself, moreover, appears extremely easy, so that is might be expected that any one could accomplish it with the greatest facility and promptitude. But contrary to expectation the behavior is quite different. »

 

The reaction times are extremely unequal, sometimes short, sometimes long; it is also noticed that certain answers undergo disturbances: the subject forgets the initial recommendation to answer by only one word and answers by a whole sentence; or it neglects the meaning of the inductive word and reacts by a tonal association; or it repeats the inductive word; or it starts to laughing, or answers « yes » or « no », or it answers several successive inductive words the same induced word.

« Another sign of affective reaction is the repetition of the stimulus words. The test persons repeat the stimulus word as if they had not heard or understood it distinctly. (…)In principle it is the same phenomenon as the subsequent completion of the reaction.

In many experiments we observe that (…) some words seem to possess a special reproduction tendency : they always contain something very important for the individual psychology of the test person ».

Defective Reproductions :

 

After having recorded a certain number of associations one takes again the list of the inductive words and asks the subject to repeat the response made to each one of them. It is recorded if the subject remembers that or not, or gives a different answer.

 

One calls all these disturbances as well as the very long response times, or the absences of reaction: indices of complex. They touch in some way the intimate taboo sphere.

 

When an inductive word interests only the surface of the conscience, the reaction is normal and it does not occur anything unusual; but when, on the contrary, it crosses the protective dams of the interior life and penetrates in the intimacy of ego, it determines a disturbance of the external reaction, while starting an unconscious automatism, to which the subject is not prepared, which catches his  attention, subjugates him to some extent, and prevents him complying with the instructions given.

 

Thus « The first thing that strikes us is the fact that many test persons show a marked prolongation of the reaction time. (…) The association experiments cannot deal with a separated psychic function : Any psychic occurrence is never a thing in itself, but is always the resultant of the entire psychological past ».

 

 

Parapsychological Hypothesis

 

 

We all bear the original sin of the missed conjunction of nascent psychoanalysis with the yet immature parapsychology. That is a way to explain the difficulties for the latter to make its way towards its object. Telepathy comes from one’s unconscious and  sometimes reaches others’ unconscious. We could not actually study it unless we consider the whole dimension of that, in its whole theoretical and practical extent ! Particularly repression should not be forgotten, when facing psi-missing !

 

A lot of facts, gathered by parapsychology, force us to admit, once and for all, that we are confronted by the paradoxical psi-missing as a counterpart of the psi-hitting. This odd thing is probably necessary to mark borders between individuals and make language possible. Otherwise, it appears the ESP is not under the mastering of our conscience, it comes as it wants ! Intention is not required and, its presence is not of interest ! That is why, we do have to consider the following schema which tries to outline the four barrier the message has to go through [1]  !

 

 

Input

à

Sender Consciousness

à

[Censorship]

ß à

Sender Unconscious

ß

? R-Gate Keeper ¿

ß

? Psi Tunnel Psi Channel ¿

ß

¿ T-Gate Keeper ?

ß

Receiver  Unconscious

à

Censorship

à

Receiver Consciousness

à

Output

 

 

Thus, we have to cheat with all these obstacles. Be it psi-hitting or psi-missing, we want to get transmission and eliminate « psi-nothing » ! For doing it, we propose to use the Jung's Word Association Test in order to choose convenient potential targets, personalized for each subject.

 

Generally speaking, indeed, the set of targets is drawn from a list chosen by the experimenter. He more or less proceeds to preliminary choices founded on " emotional nature " of any potential target.

 

On the basis of some experiments, Dr. Barry put forth the assumption that the potential targets chosen by the receiver (for example first names of its close relations) would be more relevant than those chosen by the sender or the experimenter. This idea to personalize the potential targets according to the sender  or the receiver seems to me very interesting. However, the idea of " voluntary choice " contravenes other assumptions which starts to be accepted in our discipline : ESP is probably more effective if it is in touch with a nonconscious emotional dynamics for the agent as well as for the percipient.

 

Thus it would be convenient to test the assumption according to which words (or images very related to these words) " provoking a reaction " at the nonconscious level would be better candidates for an ESP process than " neutral " words, either from the point of view of the agent, or from the point of view of percipient.

 

We are able to test this assumption by proposing an automated Jung's word association test for subjects participating in ESP experiments, using as potential targets :

 

·        In a quarter of cases a set of long delay words for the receiver and for the sender

·        In a quarter of cases a set of short delay words for the receiver and for the sender

·        In a quarter of cases a set of short delay words for the sender  which are long delay words for the receiver.

·        In a quarter of cases a set of long delay words for the sender  which are short delay words for the receiver.

 

Another track of research is trying to detect if the « long delay » words give a longer latency to the answer of the receiver. Is there any increase of the time the receiver needs to answer the words for which he needs a longer time to answer an associated word ? If such a word gives a reaction time longer in the Jung’s word association test, why does it not get a reaction time longer when the subject receives it ? The unconscious is certainly as much at work in this case than in the former one !

 

 

Software prototype

 

One of the objectives of our work was to try to realize a software program which will present inductive words to a subject, obtain an oral answer, and calculate the delay between the presentation of the inductive word and the associated answer (the reaction time). The software distributes the answering times into short, means and long delays (for example by distributing them by quartiles). These delays are the only jungian indicator which we plan to automate at the moment.

 

·        This first part presents the functional characteristics and the technical characteristics of the software.

·        Then, the use and the functioning of the software are explained.

·        Finally, some possible improvements are proposed.

 

Functional characteristics

 

The prototype consists of two different modules and autonomous software. The first module is dedicated to the experiment itself. The second module is dedicated to the analysis of the data collected by the first module during the experiment.

 

1.     The module dedicated to the experiment requires, as input, the list of the inductive words which is supplied by the experimenter as a text file (in the ASCII standard format).

As output, the experiment module supplies,

·        on one hand, sound files (in the wav standard format) - one file for each inductive word - which contains the subject’s answers.

·        and, on the other hand, a text file which contains the same answers (entered by the subject via the keyboard).

 

2.      The module dedicated to the analysis requires, as input, the sound files created during the experiment. The experimenter has also to indicate the parameters for the number of quartiles and the noise level, both of which should be considered during the analysis. The noise level allows detecting the actual beginning of every answer from the pervading noise. As output, the module supplies a text file (legible by any spreadsheet type software) containing the results of the analysis. These results are, on one hand, the list of the inductive words and, on the other hand, a list of reaction times (in tenth of second) corresponding to each inductive word.

 

 

Technical characteristics

 

The software can be used on any type of personal computer working with Microsoft’s Windows operating system and provided with a sound card which will drive the associated mixer. As this is a prototype version of the software, not all of the configurations have been tested. Nevertheless, it is possible to assert that a computer running with a relatively recent operating system (Windows 9x or higher) and equipped with a relatively cheap sound card (for example Creative Labs’ SoundBlaster PCI 128) allows a normal use of the software.

 

Use and functioning

 

The software has to be used by following the four stages described below.

 

1. The first step is the generation, by the experimenter, of the text file containing the inductive words.
2. The second step is the calibration of the recording. The experimenter can do this by himself (herself) or assisted by the subject. The calibration consists in preparing the recording by looking for the optimal quality, relatively to the material which can be used (microphone and sound card) and, also, to the general sound atmosphere of the room where the experiment is going to take place.
3. The third and most important step is the experiment itself. The module used for the realization of the experiment,
1. first displays a presentation about the experimental procedure to the subject.
2. Then, as soon as the subject is ready, the experiment begins when he (she) hits the “Start” button.
3. Then, the first inductive word appears on the screen (figure 1) and the voice recording -via the microphone- begins.
4. As soon as the subject ends answering, he (she) clicks with a mouse button or hits the keyboard.
5. A dialog box asks then to write the word which the subject has just orally expressed. Once done, the subject clicks with the mouse or hits a key. The following inductive word appears and the related recording begins. The process repeats as often as there are still inductive words.

 

 

Figure 1 : Screenshot of the experiment software module

(for have an example, the word « mercy » is an inductive one).

 

 

Figure 2 : Screenshot of the analysis software module.

 

4.      The fourth and last step is the analysis of the experiment results.

a.       The experimenter chooses in the analysis module which sound files he (she) wants to analyze, that is to say files corresponding to the subject’s answers.

b.      Then, the experimenter sets the noise level which he (she) wishes for the analysis.

c.       Finally, he (she) hits the “Start” button and the analysis begins (figure 2).

 

5.      The analysis consists of three phases.

 

a.       The first phase calculates the average sound level of the recordings to optimize the analysis.

b.      The second phase examines each of the recordings, one by one, to determine when, at the beginning of every recording, the first syllable is audible. It is about the time that the subject took to answer every inductive word. This is done by considering the first sound showing a peak level higher than the average noise level (calculated from the parameter set by the experimenter at the beginning of the analysis and from the average sound level obtained during the first phase of the analysis).

c.       The third and last phase of the analysis sorts the reaction times - calculated during the previous phase - and save everything in a text file. The experimenter then knows the inductive words for which the subject had a relatively short or relatively long reaction time.

 

The analysis lasts a few minutes; its duration depends on the number of inductive words used in the experiment and on duration of the reaction times.

 

 

25 Mots		5 Quantiles				1 % de Niveau de Bruit
Temps de réponse courts
			4 4 8 10 10	nid pasteur biscuit enlever godet
Temps de réponse moyens
10 10 12 14 14	idiot ingurgiter radoucir rixe adjudant		14 16 18 20 20	escarpé explication bourse empester intrigant	20 22 22 22 24		opium choc jonction abandonné narquois
Temps de réponse longs
			42 40 34 30 26	chasser compatir considérable pente cuisine
(Reaction Time in 1/10° seconds)

 

Figure 2 : Table of results for one subject

 

 

Possible improvements

 

As the software is a prototype, numerous improvements are possible.

 

A first idea would be to ask the subject to clarify if he (she) finds the inductive word nice or not. For that purpose, it would be useful to ask him (her) to give a mark to every word (1 = not nice, 2 = neutral, 3 = nice) or to click on a bar for getting quantitative evaluation. This would allow one to know how the subject perceives an inductive word like “aggressive”.

 

Another idea would be to display the inductive word on the screen and in the same moment to say it via speech synthesis software. The progress has been considerable these last years in that field and this could make the experiment more naturalistic.

 

Finally, and this is the most important improvement, we plan to use the answers as future inductors in a future session for the same subject or another subject. It would be very interesting in particular to do some group work.

 

Other minor improvements should also be made soon. They will be especially useful in making the software more user-friendly.

 

Pilot work

 

 

The present research, besides the creation of an Automated Jung's Word Association Test, aimed to appreciate the feasibility of a parapsychological comparison between the words with strong emotional resonance (Long Associative Delay or LAD-words) and the words with weak emotional resonance (Brief Associative Delay or BAD-words).

 

Fifteen subjects were tested using a shortened list of 25 words chosen at random from a small French dictionary.

References

 

·        Jung C. G. (1910), The Association Method, American Journal of Psychology, 31, 219-269 (available at http://www.gesher.org/Neurocare/Medical%20Education/Jung.html )

·        Jung C.G. (1928), L’Homme à la découverte de son âme, French translation by R. Cahen – 1987, Chap.4, Albin Michel éd.

·        Barry J. (1971), Journal d’un Parapsychologue, EP, 165.

·        Delanoy, D. (1988). Characteristics of successful free-response
targets: Experimental findings and observations. Proceedings of the
31st Annual Convention of the Parapsychological Association, Montreal,
Quebec, 230-246.

·        Delanoy, D. L. (1989). Characteristics of successful free-response
targets: Experimental findings and observations. In L. A. Henkel, & R.
E. Berger (Eds.), Research in Parapsychology 1988 (pp. 92-95).
Metuchen, NJ: Scarecrow Press.

·        Winer, R. I. (2002)  How to use Jung's Word Association Test : (http://www.gesher.org/Neurocare/Medical%20Education/Jung.html )

 

Acknowledgements

 

The authors wish to thank Monique Widmer for her implication in the collect of data and the participants who accepted to be tested….