Defense Mechanisms and Respiratory Parameters

doi:10.4236/psych.2011.24052

Paper Menu >>

Journal Menu >>

Psychology

2011. Vol.2, No.4, 331-334

Defense Mechanisms and Respiratory Parameters

Uwe Hentschel1, Thomas van Praag2, Manfred Kießling3

1Department of Psychology, University of Leiden, Leiden, The Netherlands;

2Cope, Leiden, The Netherlands;

3GPS, Mainz, Germany.

Email: [email protected]

Received March 16th, 2011; revised May 14th, 2011; accepted June 11th, 2011.

The purpose of the study was to look the relation of stress, respiration and gender to defense mechanisms. A

questionnaire was used to measure Turning against Object, Projection and Reversal and showed a negative rela-

tion to stress, measured by a high-speed task at the PC, representing a cognitive task. Defense mechanisms

showed also a negative relationship to the respiratory variables, Amplitude per minute and Time in. Steepness

and Amplitude per minute showed a greater value for women. Pause at the end of inhalation, Time in and Fre-

quency gave a greater value for men, who also showed more stress. Thus defense mechanisms can obviously

help to prevent cognitive stress, with respiration they showed mixed results. Future research should especially

pay more attention to the existing gender differences in respiration as well as in defense mechanisms.

Keywords: Defense Mechanisms, Respiratory Variables, Gender Differences, Optimal Scaling

Introduction

Defense mechanisms have been introduced as unconscious

variables that can be chosen by the respondents according to

their prevailing needs. In fact, they are trait variables that are

not so easily changed. Vaillant (1977) especially has shown that

they are pretty stable without any psychotherapeutic intervene-

tion and remain constant over more than 30 years. Kragh (1985;

cf. also Smith & Hentschel, 2004) has also found fairly stable

patterns for defense mechanisms.

This report used questionnaire items to measure defense

mechanisms by means of the ‘Fragebogen zu Konfliktbewälti-

gungsstrategien’ (FKBS) [Conflict-Solving Strategies Inven-

tory] (Hentschel, Kießling, & Wiemers, 1998). The FKBS

shows, among other results, relationships to pain (Egle et al.,

1989), psychotherapeutic interventions (cf. Geiser, Imbierovicz,

Conrad, Wegener, & Liedtke, 2005; Liedtke, Künsebeck, &

Lempa, 1990), sleep disturbances (Hermann-Maurer et al.,

1992), and dreams (Hentschel, Van der Voort, & Davids, 2007).

The FKBS has some projective components, in contrast to

many other questionnaire items by its mode of presentation,

like the Defense Mechanisms Inventory (DMI) (Gleser &

Ihilevich, 1969).

Breathing is a variable under unconscious as well as con-

scious control. People can easily shift from one to the other

mode, but mostly, under normal circumstances, we do not pay

attention to our respiration (Harver & Lorig, 2000). The normal

breathing cycle has 3 phases, which in the ideal case, have an

equal length: inspiration, expiration and pause. Respiration

consists mostly of situational aspects (Wientjes, 1993). The

respiratory parameters are thus, to a greater extent, state vari-

ables.

Both measures—respiratory parameters as well as defenses—

usually show gender differences. For defenses, this difference

has been avoided often by using only male respondents. Also

we started our breathing research with a pure male sample (cf.

Van Praag, 1995) and then detected the gender differences.

The first author has together with his co-authors (Hentschel,

Smith, & Draguns, 2004) summarized many results on defense

mechanisms as revealing differences in information processing,

a link that also has been evident in respiration (Wientjes,

Grossman, Gaillard, & Defares, 1986).

Breathing variables can be constructed from the main respi-

ratory cycle, with almost no limit in number, if one admits

higher correlations of them. The main difference is that de-

fenses are trait variables and breathing has to be regarded

mainly as a state.

Method

Sample

The whole sample comprised hundred and three persons, 50

males and 53 females, in the age range of 18 to 55 years. The

age range was equal for the 2 groups. They were students or

members of a Dutch air company. All were volunteers and

received the test results and were invited to a cup of coffee or

tea.

Instruments

To measure defense mechanisms we used the FKBS [Con-

flict-Solving Strategies Inventory], which is mainly docu-

mented in the German manual (cf. Hentschel et al, 1998). A

native speaker made the translation into Dutch. Although there

is a large database, there are no official norms for the Dutch

version, thus the reliability values in Table 1 come from the

German manual. The lack of norms has however no influence

on our results, in which norm values were not used. All FKBS

variables (15 in total) were used in the beginning and exclu-

sions were made later on the basis of the ANOVA results (see

Method of Analysis section). The FKBS has good reliabilities

(all Cronbach’s Alphas equal or are above .78 in the German

U. HENTSCHEL ET AL.

332

standardization sample) and knows 5 defenses (cf. Hentschel et

al., 1998). The FKBS has been developed basically in analogy

to the DMI (Gleser & Ihilevich, 1969), but has fewer defenses,

only 2 modes of answering (feeling and doing) and mainly

other stories. Three mechanisms of defense were included in

the present analysis (TAO, Turning against Object; PRO, Pro-

jection; REV, Reversal). The defense mechanisms in the FKBS

can be grouped according to the ways of responding to the

threat (feeling, doing, and total scores). The defenses included

in this report and the method of response, are given in Table 1.

We have measured respiration by two belts, one around the

chest and one around the stomach. As there were too many

respiratory variables in the beginning, a selection had to be

made there as well. We have used 2 methods for doing this (cf.

Method of Analysis section). The respiration parameters used,

are summarized in Table 2.

Stress was imposed by a speeded multiple decision task,

presented to the respondents on a personal computer. The task

consisted of reacting as soon as possible to 4 different colors

and reacting at the same time to high and low tones presented

via earphones. The stress task is described in more detail in

Table 4.

Procedure

The participants were administered the FKBS. They were

fitted with 2 belts to measure the breathing variables. Then the

timed decision task was administered, measuring the breathing

variables continuously. Their sex was registered; their age was

limited by admission to the tasks. All ordinal groups were

formed by the grouping program of SPSS (1990) for their use

in the further analysis.

Method of Anal ysi s

We shall present our results mainly by means of a nonlinear

canonical correlation analysis; i.e. OVERALS (SPSS, 1990). In

this program, the scaling levels of the variables included can be

chosen (defenses, respiration and the stress variables were

taken as ordinal, gender as single nominal). The program gives

a Fit value, i.e. how much the results correspond with the data

at hand; it has also the advantage that the results can be pre-

sented in form of a graphical model. For OVERALS we have

used a grouping of the cases with different numbers of groups

(cf. Tables 1 and 2 and the Instruments section) and limited the

output to two dimensions.

Fit and Loss are the only estimates one gets, i.e. OVERALS

does not give a real test result for the included variables. Both

values provide only a comparison of the weighted sum of the

included variables with the object scores, which in our case are

the values of the respondents. The indications are in general,

comparable to other statistical procedures, lower with a greater

number of the included variables. For further details of the

method see Van der Burg, de Leeuw and Verdegaal (1988) and

Bijleveld and Van der Burg (1998).

Given the problematic gender differences, among other

things, we have applied many ANOVA’s. Those analyses shall

however not be reported in detail as they only were used to

select the most important variables by the post hoc results of

the ANOVA’s. For the ANOVA analyses we have divided the

whole sample in almost equal parts into an analysis and a vali-

dation group. Only variables that have shown significant results

in both analyses, in the analysis as well as in the validation

group, were included in OVERALS. We thus can lean with our

results also on traditional p-values.

Exclusions of the respiratory variables were based on the

ANOVA results and too high inter-correlations among them.

As the remaining respiratory parameters still were not com-

pletely independent, a Pearson correlation analysis of the used

ones (Minute ventilation, Steepness, Frequency, Pause at the

end of inhalation, Time in) is also presented in Table 3 (see the

Results section).

We ended up finally with 3 defense mechanisms and 5 respi-

ratory variables.

Results

Table 1 gives a description and an overview of the abbrevia-

tions used for the defenses, and the number of groups of the

respondents. Table 2 presents a description for the respiratory

variables and an overview of the abbreviations used, together

with the number of groups of the respondents. Table 3 gives the

inter-correlations of the respiratory variables. In Table 4 the

stress induction is presented. Figure 1 presents the graphical

summary of OVERALS with 2 dimensions. As far as it con-

cerns variables with an ordinal level, projected centroids are

given. For variables with a nominal scale level, centroids are

used. The arrows always represent the highest value of the re-

spective variable. The lengths of the respective lines represent

the importance of a relationship, i.e. a vector. The Fit of the

OVERALS solution (1.02) has to be evaluated as pretty good,

and the Loss is accordingly 0.98. If the solution would be in

perfect agreement with the data, the maximum value would

equal 2.0 (the number of dimensions in the OVERALS solu-

tion).

It is remarkable that a high level of defense mechanisms

(TAO, PRO, REV) has a negative relationship to cognitive

stress (high speed of a multiple decision task on the PC). For

Table 1.

The defense mechanisms used in the analysis.

Defense mechanisms Abbreviation in Figure 1 Form of response usedMeaning or example Cronbach’s α Number of groups

(high 2, low 1) (ordinal)

Reversal REV Feeling

Thinking that the frustrator is a

nice person .80 2

Turning against object TAO Doing Smashing a door .90 2

Projection PRO Feeling

Thinking that the frustrator has

done it with vexatious motives.78 2

U. HENTSCHEL ET AL. 333

Table 2.

The respiratory variables i n the OVERALS solution.

Respiratory parameters Abbreviation in Figure 1 Number of groups

(ordinal)

Minute ventilation Ampl./min. 4

Steepness Steepness 6

Frequency Freq. 6

Pause at the end of

inhalation Pause 4

Time of inhalation Time in 2

Figure 1.

The OVERALS solution for defenses, respiratory parameters, stress and

the gender of the respondent s.

the respiration parameters the results are however mixed. Min-

ute volume (Ampl./min.) and Time in are in contrary direction

to the defenses, whereas Pause is in-between, Steepness and

Frequency go together with the defense mechanisms. Thus

these respiration variables show a higher non-linear relationship

to the defense mechanisms. Male respondents seem to have

experienced more stress than the females. They react more with

respiration also (higher values of Time in, Pause and Fre-

quency). The females show higher values of Minute ventilation

(Ampl./min.) and Steepness.

Discussion

A combination of state and trait variables seems especially

promising, if the trait variables also show some change, like

defense mechanisms do with psychotherapy for example. With

regard to defense mechanisms, questionnaires and other meth-

ods of registering them usually do not correlate to a higher

degree, which is basically a theoretical problem. This holds also

true for the FKBS with at least some projective aspects. The

relationship to different outside criteria may be however never-

theless satisfactory. This has to be cross-validated for respira-

tion, of course. It is relatively new that respiration is regarded

as an information-processing variable. This should open up the

relationship of respiration to more personality variables that

have some connection with information processing.

Usually only the chest changes of air in the lungs are regis-

tered for respiration, but for the relationship with emotions and

defense mechanisms it is better to use two belts: A chest belt

and a stomach belt, or any other measures, are thus recom-

mended.

In the in the beginning of our research, we had high hopes

for differentiation between chest and stomach respiration.

Adding females to the sample ended most of these hopes. More

research seems recommended here.

It is very unfortunate that speech and respiration cannot be

measured together, i.e. that all tasks that are to be studied in

Table 3.

Pearson correlations o f the respiratory parameters.

Minute ventil. Steepness Freq. Pause in

Minute ventilation (Amplitude/min.: chest and stomach) -- -- -- --

Steepness (at the beginning of exhaling; low values = high steepness) −.66** -- -- --

Frequency (Freq.: 1/(cycle length) × 60 .13 .22* -- --

Pause in (pause at the end of inhalation) −.55** .60** .29** --

Time in (% of inhalation of the whole respiration cycle) −.29** .11 −.28** .46**

Note: **Correlation is significant at the 0.01 level (two tailed); *Correlation is significant at the 0.05 level (two tailed).

Table 4.

Stress induction.

Speeded reactions to colours on the PC Red Green Blue Yellow

Arrow keys for: high, low, left, right Colors in squares

Speeded reactions to tones coming via earphones

Low tone z to be pressed

High tone x to be pressed

U. HENTSCHEL ET AL.

334

relation to respiration should be non-verbal. Even if decisions

are required, like in our case, they should happen in silence.

Pressing computer keys, as we used them as answers, would be

a solution here. Given the relationship of stress and Minute

volume a cross-validation is to be found in Wientjes (1993). In

future research also the induction of stress could be evaluated,

as probably not all respondents act according to the high-speed

instruction.

We regard our results in spite of all rigorous controls still as

preliminary, but a publication of them seems, according to our

judgment, justified in order to allow repetitions or refutations.

Perhaps one would come to different relationships with a dif-

ferent form of measuring defense mechanisms. This should be

checked however empirically as well. The result that defense

mechanisms have a moderating effect on stress is according to

the hypothesis, that defenses, if they work, have an anxiety

reducing influence. The inclusion of respiratory parameters

adds also something to the validation of defense mechanisms,

even if the results are mixed.

References

Bijleveld, C. C. J. H., & Van der Burg, E. (1998). Analysis of longitu-

dinal categorial data using optimal scaling techniques. In C. C. J. H.

Bijleveld, L. J. H. Th. Van der Kamp et al. (Eds.), Longitudinal data

analysis. Designs, models and m e th ods (pp. 46-154). London: Sage.

Egle, U. T., Rudolf, M.-L., Hoffmann, S.O., König, K., Schöfer, M.,

Schwab, R., & Wilmowsky, H. von (1989). Personality charachteris-

tics, defensive behavior and illness experiences in patients with with

primary fibromyalgy. Zeitschrift für Rheumatologie, 48, 73-78.

Geiser, F., Imbierovicz, K., Conrad, R., Wegener, I., & Liedtke, R.

(2005). Turning against self and its relation to symptom distress, in-

terpersonal problems, and therapy outcome: A replicated and en-

hanced study. Psychotherapy Research, 15, 357-365.

doi:10.1080/10503300500226241

Gleser, G., & Ihilevich, D. (1969). An objective instrument for meas-

uring defense mechanisms. Journal of Consulting and Clinical Psy-

chology, 33, 51-60. doi:10.1037/h0027381

Harver, A., & Lorig, T. S. (2000). Respiration. In J. T. Cacioppo, L. G.

Tassinari and G. G. Berntson (Eds.), Handbook of psychophysiology

(2nd ed.) (pp. 265-293). Cambridge: Cambridge University Press.

Hentschel, U., Kießling, M., & Wiemers, M. (1998). Conflict-solving

strategies inventory-FKBS. Weinheim: Beltz.

Hentschel, U., Smith, G., & Draguns, J. G. (2004). Defense mecha-

nisms and their psychophysiological correlates. In U. Hentschel, G.

Smith, J. G. Draguns and W. Ehlers (Eds.), Defense mechanisms.

Theoretical, research and clinical perspectives (pp. 611-633).

Amsterdam: Elsevier. doi:10.1016/S0166-4115(04)80052-9

Hentschel, U., Van der Voort, M., & Davids, A. (2007). On the evalua-

tion of creativity in dream reports. In P. Herdina, A. Oberprantacher,

& J. Zelger (Eds.) Learning and development in organizations (pp.

260-279). Wien: Lit Verlag.

Hermann-Maurer, E. K., Drews, U., Imhof-Eichenberger, E., Knab, H.,

Schneider-Helmert, D., Hentschel, U., & Schoenenberger, G. A.

(1992). Sleep disturbances: Conflict solving strategies of patients

suffering from insomnia. Zeitschrift für Klinische Psychologie,

Psychopathologie und Psychoth erapie, 40, 34-46.

Kragh, U. (1985). Defense mechanism Test-DMT manual. Stockholm:

Persona.

Liedtke, R., Künsebeck, H.-W., & Lempa, W. (1990). Changes in cop-

ing with conflicts during inpatient psychotherapy. Zeitschrift für

Psychosomatische Medizin und P s y ch o an a l ys e , 36, 79-88.

Smith, G., & Hentschel, U. (2004). Percept-genetic identification of

defense. In U. Hentschel, G. Smith, J. G. Draguns, & W. Ehlers

(Eds.), Defense mechanisms. Theoretical, research and clinical per-

spectives (pp. 129-152). Amsterdam: Elsevier.

SPSS (1990). SPSS Categories. User’s manual. Chicago, IL: SPSS Inc.

Van der Burg, E., De Leeuw, J., & Verdegaal, R. (1988). Homogeneity

analysis with k-sets of variables: An alternating least squares method

with optimal scaling features. Psychometrika, 53, 177-197.

doi:10.1007/BF02294131

Van Praag, T. (1995). Respiratory movements and personality. Master;s

Thesis, Leiden: Leiden University.

Vaillant, G. E. (1977). Adaptation to life. Boston, MA: Little, Brown &

Co.

Wientjes, C. J. E. (1993). Psychological influences upon breathing: Situ-

ational and dispositional aspects. Dissertation, Tilburg: University of

Tilburg.

Wientjes, C. J. E., Grossman, P., Gaillard, A. W. K., & Defares, P. B.

(1986). Individual differences in respiration and stress. In R. Hockey,

A. W. K. Gaillard and M. Coles (Eds.), Energetics and human in-

formation processing (pp. 317-327). Dordrecht: Nijhoff.