Seediscussions,stats,andauthorprofilesforthispublicationat:https://www.researchgate.net/publication/45390277

Defensemechanismsdevelopmentintypicalchildren

ArticleinPsychotherapyResearch·September2010

DOI:10.1080/10503307.2010.493536·Source:PubMed

CITATIONS

0

READS

58

2authors:

mariaannaTallandini

UniversitàdegliStudidiTrieste

17PUBLICATIONS160CITATIONS

SEEPROFILE

CorradoCaudek

UniversityofFlorence

90PUBLICATIONS655CITATIONS

SEEPROFILE

Allin-textreferencesunderlinedinbluearelinkedtopublicationsonResearchGate,

lettingyouaccessandreadthemimmediately.

Availablefrom:mariaannaTallandini

Retrievedon:27October2016

Full Terms & Conditions of access and use can be found athttp://www.tandfonline.com/action/journalInformation?journalCode=tpsr20

Download by: [University of London] Date: 03 July 2016, At: 07:07

Psychotherapy Research

ISSN: 1050-3307 (Print) 1468-4381 (Online) Journal homepage: http://www.tandfonline.com/loi/tpsr20

Defense mechanisms development in typicalchildren

Maria Anna Tallandini & Corrado Caudek

To cite this article: Maria Anna Tallandini & Corrado Caudek (2010) Defense mechanismsdevelopment in typical children, Psychotherapy Research, 20:5, 535-545, DOI:10.1080/10503307.2010.493536

To link to this article: http://dx.doi.org/10.1080/10503307.2010.493536

Published online: 27 Jul 2010.

Submit your article to this journal

Article views: 418

View related articles

Defense mechanisms development in typical children

MARIA ANNA TALLANDINI1 & CORRADO CAUDEK2

1Department of Psychology, University of Trieste, Trieste, and Department of Psychology, University College London, London

& 2Department of Psychology, University of Florence, Florence, Italy

(Received 6 April 2009; revised 29 March 2010; accepted 7 May 2010)

AbstractThe defense mechanisms (DMs) of 103 nonreferred children ages 47 to 102 months were assessed through dollhouse play.The authors measured the children’s temperament (Temperament Assessment Battery for Children�Teacher Form[TABC]) and verbal capacities (Wechsler Preschool and Primary Scale of Intelligence or Wechsler Intelligence Scale forChildren). Four main findings were derived: (1) DM use decreased with age with different developmental trajectories; (2)regression, displacement, and reaction formation were more frequent in girls and denial more frequent in boys; (3) thenumber of DMs used was negatively associated with the TABC Adaptability score and positively with the TABC Approach/Withdrawal score; and (4) children who used rationalization and did not use identification and suppression scored better onverbal capacities.

Keywords: childhood; defense mechanisms; temperament; verbal intelligence

Defense mechanisms (DMs) have been considered

as strategies*adaptive or maladaptive*that chil-

dren and adults use to deal with the emotional

difficulties of everyday life (Cooper, 1998; A. Freud,

1936/1965; S. Freud, 1926/1959; Hentschel,

Smith, Draguns, & Ehlers, 2004; Laor, Wolmer, &

Cicchetti, 2001). Much of the empirical research

primarily focuses on DMs among adult and adoles-

cent populations (Feldman, Araujo, & Steiner, 1996;

Conte, Plutchick, & Draguns, 2004), even though

DMs in early childhood have also been studied (e.g.,

Araujo, Medic, Yasnovsky, & Steiner, 2006; Cramer,

1997; Cramer & Block, 1998; Wolmer, Laor, &

Cicchetti, 2001). The aim of the present investiga-

tion was to study the development of DMs in

children ages 47 to 102 months. Information gained

from an assessment of DMs, in fact, may prove to be

useful in developing a more effective therapeutic

relationship (Despland, de Roten, Despars, Stigler,

& Perry, 2001).

Initially, defenses were seen as countercathexes

serving to diminish anxiety produced by conflictual

contents of the mind and were linked to pathological

aspects of human functioning (S. Freud, 1894/

1962). Later, Sigmund Freud argued that DMs

are useful mental mechanisms that allow individuals

to cope with internal conflicts: They maintain the

unconscious status of forbidden impulses and,

therefore, mitigate or avoid anxiety (S. Freud,

1923/1961, 1926/1959). Anna Freud (1936/1965)

believed that DMs emerge at different moments of

normal development during childhood with different

intensity, depending on the psychological stage of

development and the psychological context. She

considered denial and projection as ‘‘normal’’ in

early childhood but as indicators of possible pathol-

ogy later on. Repression and sublimation were

expected to emerge relatively late in the develop-

mental process. Other DMs such as ‘‘regression,

reversal or turning round the self are probably

independent of the stage that the psychic structure

has reached’’ (A. Freud, 1936/1965, p. 56). More-

over, Anna Freud posited that DMs are influenced

by libidinal stages and intellectual capacity. Denial

and regression, for instance, require very little

intellectual contribution, whereas rationalization

requires more complex intellectual skills. Overall,

Anna Freud’s position was that DMs emerge at

different ages with different intensity, depending on

the psychological stage of development and the

psychological context.

Studying the developmental aspects of DMs,

Cramer (1996) highlighted the confusion that exists

in the description and classification of defenses.

Correspondence concerning this article should be addressed to Maria Anna Tallandini, Department of Psychology, University of Trieste,

via S. Anastasio 12, Trieste, 34134 Italy. E-mail: tallandi@units.it

Psychotherapy Research, September 2010; 20(5): 535�545

ISSN 1050-3307 print/ISSN 1468-4381 online # 2010 Society for Psychotherapy Research

DOI: 10.1080/10503307.2010.493536

Dow

nloa

ded

by [

Uni

vers

ity o

f L

ondo

n] a

t 07:

07 0

3 Ju

ly 2

016

Different authors use different classification princi-

ples, and no general, common taxonomy of DMs

had yet been proposed (Hauser & Safyer, 1995;

Vaillant, 1992). The classifications and definitions of

the Diagnostic and Statistical Manual of Mental

Disorders (fourth edition, text revision [DSM-

IV-TR]; American Psychiatric Association, 2000),

therefore, represent a useful tool for research on this

topic.1 In the present research, the term defense

mechanism refers to an ‘‘automatic psychological

process that protects the individual against anxiety

and from awareness of internal or external stressors

or dangers. Defense mechanisms mediate the in-

dividual’s reaction to emotional conflicts and to

external stressors’’ (American Psychiatric Associa-

tion, 2000, p. 821). DMs are thought to operate

unconsciously but have observable manifestations

within an individual’s behavior (A. Freud, 1936/

1965; S. Freud, 1926/1959). The individual can be

aware of the presence of DMs in his or her behavior

but unaware of the purposes, drives, and affects

underlying them. DMs have often been studied in

the adult population, but little research has been

conducted on children (e.g., Bond, Gardner, Christian,

& Sigal, 1983; Haan, 1963; Perry & Cooper, 1989;

Sammallahti & Aalberg, 1995). Among these inves-

tigations is the work of Smith and Danielsson

(1982). Using the meta-contrast technique to pro-

duce anxiety reactions and defense strategies, they

found that anxious children show denial and passive

surrender. Laor et al. (2001) developed the compre-

hensive assessment of defense style method to

evaluate child and adolescent defensive behavior.

Such a method, however, relies exclusively on

parental reports and is not framed within a devel-

opmental perspective. Cramer (1991) conducted

several studies in normal children and adolescents

on three DMs: denial, projection, and identification.

Similarly to A. Freud (1936/1965), Cramer found

that these DMs have different developmental

courses: Denial is the earliest to appear; the pre-

cursors of projection develop into proper projection

as a result of cognitive development; and the

precursor of identification is present in early infancy

and reintensifies during adolescence.

In this investigation, we examined the development

of 12 DMs in normal children ranging in age from

47 to 102 months. Within this time frame, from a

psychoanalytic perspective, children develop from

the preoedipal phase, through the oedipal complex,

to the beginning of the latency period (e.g., A. Freud,

1936/1965). Moreover, from the cognitive perspec-

tive, the intellectual functions undergo significant

reorganization as a result of the acquisition of a

wider range of cognitive tools (Fischer, Shaver, &

Carnochan, 1990; Keil, 1989; Sloutsky, Lo, & Fisher,

2001).

It is easy to recognize that the research methods

appropriate for adults are of little use when studying

DMs in children. To address this problem, we

studied the presence of DMs in an almost natural

setting by evaluating children’s overt behavior in

dollhouse play (DHP; see also Emde, Wolf, &

Oppenheim, 2003). Given that DMs are activated

by the presence of emotionally salient individuals

(parents and siblings among others), we hypothe-

sized that the expression of DMs would be induced

by the request to reproduce scenes involving the

children’s own family. In fact, at the earliest age

we consider (47 months), children’s cognitive func-

tioning at a symbolic level is well established

(Lichtenberg, 1983).

Regarding the development of DMs, we hypothe-

sized that the likelihood of children’s use of denial

decreases with age (Cramer & Gaul, 1988; A. Freud,

1936/1965; Glasberg & Aboud, 1982; Hill &

Sarason, 1966; Smith & Danielsson, 1982) and

that the likelihood of their use of rationalization

and idealization increases with age (Cramer, 1991;

Schimek, 1968). Regarding gender, we hypothesized

that boys tend to use a larger number of DMs than

girls (Brody, Rosek, & Muten, 1985), that boys tend

to use denial more often than girls, and that

girls tend to use regression more often than boys

(Feldman et al., 1996). No specific prediction was

made for other DMs (see Laor et al., 2001).

Consistent with A. Freud’s claim that excessive

use of DMs can be interpreted as difficulty in dealing

with reality, we hypothesized that children using a

large number of DMs also show temperamental

difficulties (Cramer, 1991; A. Freud, 1936/1965;

Vaillant, 1992; Wallerstein, 1985; Martin, 1998).

Finally, we expected an association between verbal

intellectual capacity and rationalization, as found in

adult samples by Schimek (1968), Haan (1974), and

Hentschel, Kiessling, Teubner-Berg, and Dreier

(2004).

Method

Participants

The sample was composed of 103 children enrolled in

different kindergartens or schools. They ranged in age

from 47 to 102 months (mean, 73.9891.841 SE). To

analyze the development of DMs and its possible

nonlinear trend, children were separated into three

groups based on age. The youngest group (47�55

months) comprised 27 children who were at the stage

of intuitive thinking and dealing with the precursors

of the oedipal stage (mean age, 49.5 months 9 0.924

536 M. A. Tallandini and C. Caudek

Dow

nloa

ded

by [

Uni

vers

ity o

f L

ondo

n] a

t 07:

07 0

3 Ju

ly 2

016

SE). The intermediate group (60�72 months) com-

prised 37 children in the oedipal phase who had not

obtained the notion of conservation (mean age,

65 months90.74 SE). The older group (90�102

months) comprised 39 children who acquired opera-

tional thinking and whose emotional and social

development reached an equilibrium in the latency

phase (mean age, 96 months 9 0.624 SE).

Because DHP must be presented in a homoge-

neous way, with identical instructions to all participants,

only first-born children from two-parent families

were included in the study. This selection choice was

made so as to reduce the number of variables

affecting the expression of DMs (see also Dunn &

Kendrick, 1982). All children came from families

with an average to high socioeconomic status, with

at least one parent having obtained an upper level

secondary degree. Data of six children from the

initial sample were removed because of technical

problems.

The data were collected by four assistants who

were blind to the hypotheses of the study and

received no psychoanalytic training. All interviews

were conducted in accordance with the ethical

requirements proposed by the Society for Research

in Children. All queried families provided written

consent for their child to take part in the study.

Procedure

Children were examined individually in two separate

sessions 3 to 6 days apart. In the first session, half of

the children engaged in the DHP task and the

remaining children were administered the Verbal

subtest of one of two Wechsler scales: the Wechsler

Preschool and Primary Scale of Intelligence, Revised

(WPPSI; Wechsler, 1989) for the youngest group of

children or the Wechsler Intelligence Scale for

Children, Revised (WISC-R; Wechsler, 2003) for

the two older groups. Each participant was examined

in a specifically organized room. The Temperament

Assessment Battery for Children�Teacher’s Form

(TABC) was individually administered to teachers.

The DHP task was used in a semistructured

manner to investigate the presence and development

of DMs. A pilot study was conducted with nine

participants (three from each age group). Careful

analyses of the recorded play of this preliminary

investigation were used to precisely define the

procedure used in the main study. We implemented

the DHP by using a two-storey furnished dollhouse,

placed on a table to allow children easy access. To

evoke emotional content, children were asked to

pretend that the house and the dolls represented

their own house and their own family (McHale,

Neugebauer, Radin, & Schwartz, 1999; Minnis

et al., 2006). The experimental situation was sub-

divided into a fixed sequence of four typical family

life episodes: (1) mealtime, (2) bedtime, (3) a sad

(bad) time, and (4) a happy time (see McHale et al.,

1999). The mealtime episode could elicit conflicts

concerning food and table manners, the bedtime

episode could favor regression, a sad time could

induce a wish to cancel out or deny a bad event, and

a happy time could provide a reason for idealization.

The happy time episode was always placed at the end

of the sequence in order to restore the child’s

emotional state. Similar to the MacArthur Story

Stem Battery (Emde et al., 2003), this method

allows the free expression of narratives as the

children play with the dollhouse. We were interested

in how children described these events, irrespective

of their historical accuracy, presupposing that the

narrative concerning the four episodes would be

informative about the children’s psychological

reactions.

The four assistants were unaware of the children’s

family situations (with the exception of the number

of children in the family). To stimulate and contain

play without interfering with the children’s sponta-

neous production, we implemented the mirroring

technique (i.e., immediate reformulation of partici-

pants’ expressions; Lumbelli, 1994). By being asked

two alternative questions such as ‘‘Do you go to

sleep by yourself or is there somebody who tucks you

in?’’, children were encouraged to expand on any

theme that they had spontaneously produced. The

assistants learned the general protocol during the

pilot study. Training was conducted by Maria Anna

Tallandini, a trained, experienced psychoanalyst and

expert developmental psychologist.

Each child was presented with a box containing six

dolls: two representing the parents, two representing

the grandparents, one representing the child by

gender, and eventually, an infant doll representing

a younger sibling. The assistant initiated the DHP by

first saying ‘‘Let’s pretend that this is your house’’

and then ‘‘Let’s pretend that these dolls are your

family.’’ Subsequently, the experimenter provided

clear instructions in order to differentiate the four

episodes. After the fourth episode, the experimenter

indicated the end of play by saying, ‘‘We have run

out of time for playing together with the dollhouse.

Thank you.’’

The DHP was recorded by a video camera

positioned in front of the dollhouse at an average

distance of 3.5 meters. In the majority of cases,

children expressed very limited curiosity about the

camera and appeared completely absorbed in the

play. We allotted approximately 20 min for each

DHP session, which comprised all four episodes.

Development of childhood defense mechanisms 537

Dow

nloa

ded

by [

Uni

vers

ity o

f L

ondo

n] a

t 07:

07 0

3 Ju

ly 2

016

Coding system. To code the recordings of children’s

pretend play, we created an initial list of DMs based

on the taxonomy of the DSM-IV-TR (American

Psychiatric Association, 2000), further integrated by

the children’s DM exemplifications from A. Freud

(1936/1965), Sandler and Freud (1985), Cramer

(1996), and Wolmer et al. (2001). Projection

was not considered because the DHP called for

children ‘‘to pretend’’ about dolls, thus masking

other projection instances.

Twenty randomly selected tapes (20% of the

entire sample) were carefully transcribed: six for

the first group and seven each for the second and

third groups (see Oppenheim, Emde, & Warren,

1997). Transcriptions included all verbal utterances

and a complete record of nonverbal behavior, such

as motor behavior, sounds, and gestures made while

children were speaking and commenting on a given

person (e.g., shoulder shrugging or making negative

mouth noises). To transcribe the tapes, the scorers

completed a form comprising five entries: time of the

event, child’s verbal utterances, experimenter’s ver-

bal utterances, child’s motor behavior, and child’s

expressive behavior.

Through a careful study of the video recordings and

the protocol transcriptions, two raters (Maria Anna

Tallandini and a researcher with no psychoanalytic

training) identified a total of 12 DMs: acting out,

denial, displacement, idealization, identification, pas-

sive aggression, rationalization, reaction formation,

regression, repression, suppression, and undoing.

The scoring of the complete set of video recordings

was based on this DM list (see Appendix A).

Scoring procedure. Two raters scored the entire set of

36 video recordings. One of the raters had not

participated in the previous phase, but received

thorough training prior to scoring. In a first phase, a

random selection of the video recordings (20%) was

scored separately by the two raters. In this phase, the

raters reviewed their scoring whenever interrater

reliability did not reach criterion (Cohen’s k�0.80)

and until consensus was reached. A third rater was

consulted when disagreement could not be resolved,

which occurred six times overall. After this ‘‘calibra-

tion’’ phase, each rater evaluated half of the remaining

video recordings. The global kappa was .85 and

ranged from .76 to .96.

Results

DMs Presence as a Function of Age and Gender

Figure I shows the proportion of children manifest-

ing each DM in at least one episode of pretend play

as function of age and gender. Generalized linear

mixed-effects (lme) models with crossed random

effects for participants and episodes were used to

analyze binary presence/absence of each DM (see

Baayen, Davidson, & Bates, 2008; Bates & Sarkar,

2007). The dependent variable was scored as 1 if the

DM appeared in at least one episode of pretend play

and 0 otherwise.

The lme models were fit using the program lmer in

the R package lme4 (Bates & Sarkar, 2007) and the

default penalized quasi-likelihood method for esti-

mating the parameters. Significance tests on the

interactions and the main effects were carried out

by computing the deviance statistic (�2�log-

likelihood). Change in deviance is distributed as

chi-square, with degrees of freedom equal to the

number of parameters deleted from the model.

A first lme model showed that children did not use

the 12 DMs with the same frequency, x2(11,

N�103)�374.920, pB.001. The DMs used more

often were rationalization, passive aggression, displa-

cement, idealization, and regression (see Figure I).

In a second lme model, we analyzed the presence/

absence of each DM as a function of DM type and

child age and gender. Given that the three-way

interaction (DM Type�Age�Gender) was signifi-

cant, x2(35, N�103)�64.757, pB.002, we then

analyzed each DM separately, with age and gender

as independent variables. Table I shows the deviance

statistics and the significance levels.

We found significant age and/or gender effects

for the following DMs: denial, displacement,

Figure I. Proportion of children presenting each defense mec-

hanism (DM) in at least one episode of pretend play.

538 M. A. Tallandini and C. Caudek

Dow

nloa

ded

by [

Uni

vers

ity o

f L

ondo

n] a

t 07:

07 0

3 Ju

ly 2

016

idealization, rationalization, reaction formation, and

regression. No age and/or gender effects were found

for acting out, identification, passive aggression,

suppression, repression, and undoing (see Table I).

For denial, we found a significant Gender�Age

interaction (see Figure II). For boys, no difference

was found in the likelihood of using this defense

between the first and the second age groups,

Figure II. Proportion of defense mechanisms (DMs) present as a function of age and gender for the 12 mechanisms studied: acting out,

denial, displacement, idealization, identification, passive aggression, rationalization, reaction formation, regression, repression, suppression,

and undoing.

Table I. Defense Mechanisms Presence Related to Age and Gender: Deviance Statistics and Significance Levels

Gender�Agea Gender effectb Age effectc

Defense mechanism x2a x2 x2

Acting out 0.215 1.697 0.487

Denial 7.823* 1.340 1.973

Displacement 2.253 9.255** 8.908*

Idealization 1.341 0.944 6.975*

Identification 0.023 0.286 0.454

Passive aggression 0.199 0.616 0.507

Rationalization 2.376 0.017 11.058**

Reaction formation 2.462 7.659** 0.252

Regression 1.001 19.017** 9.283*

Repression 1.154 0.878 1.079

Suppression 1.546 0.079 0.593

Undoing 2.162 0.019 0.531

adf�2. bdf�1. cFor denial, df�1; for all other defense mechanisms, df�2.

*pB.05; **pB.01.

Development of childhood defense mechanisms 539

Dow

nloa

ded

by [

Uni

vers

ity o

f L

ondo

n] a

t 07:

07 0

3 Ju

ly 2

016

t(152)�0.618, ns, and between the first and the

third age groups, t(152)��0.817, ns. For girls, the

odds of not observing denial decreased by 16% from

the first to the second age group, t(131)��2.1709,

pB.05; no difference was found between the first

and the third age groups, t(131)��0.984, ns.

We found age and gender effects for both

displacement and regression. The odds for displace-

ment were 4.3 times larger for girls than for boys

(95% confidence interval [CI]: 1.743�16.168) and

5.4 times lower in the oldest age group than in the

intermediate age group (95%CI: 1.728�23.517).

The odds for regression were almost eight times

larger for girls than for boys, (95%CI: 3.149�20.207) and were 4.08 larger for the intermediate

age group than for the oldest age group (95%CI:

1.767�14.583).

For idealization and rationalization, we found age

effects in opposite directions. The odds for idealiza-

tion decreased by a factor of 5.6 from the first to the

last age group (95%CI: 1.276�24.630), and the odds

for rationalization were 4.7 times larger for the oldest

group than for the youngest group (95% CI: 1.929�11.378). For reaction formation we found a gender

effect: The odds were 3.44 times larger for girls than

for boys (95%CI: 1.339�14.740).

How Many Different DMs?

The number of different DMs (from 0�12) used by

each child across the four episodes of pretend play

was submitted to an analysis of variance, with gender

and age as independent variables. We found a

significant age effect, F(2, 95)�3.356, pB.05,

which indicated that the number of different DMs

decreased by about 20% from the first to the last age

group (see Figure III). Neither the effect of gender

nor the Age�Gender interaction were significant.

Is Temperament Associated with DM Use?

The TABC is used to assess individuals’ capacity to

positively react to new situations, to express positive

and negative feelings, and to persist in a task. The

questionnaire consists of six subscales: Activity,

Adaptability, Approach/Withdrawal, Emotional In-

tensity, Distractibility, and Persistence. We analyzed

the scores of each subscale with an analysis of

covariance, with DM number (ranging from 0�12)

and child age and gender as predictors. Only

Adaptability and Approach/Withdrawal scores re-

sulted associated with DM number.

For the Adaptability scores, we found a significant

interaction between DM number and gender,

F(1, 91)�7.273, pB.01 (Figure IV). The Adapt-

ability scores decreased with the increase of DM

number for boys but not for girls, if the effect of

age was statistically controlled. We also found an

Age�Gender interaction, F(2, 91)�4.790, pB

.002. The Adaptability scores increased with age

for girls but not for boys, if the effect of DM number

Figure III. Number of different defense mechanisms (DMs) used

by children in the four episodes of pretend play as a function of

age. Vertical bars represent91 SEM.

Figure IV. (A and B) Effects of Adaptability on age, gender, and

number of defense mechanisms (DMs). Dashed lines are point-

wise 95% confidence bands around fitted effects.

540 M. A. Tallandini and C. Caudek

Dow

nloa

ded

by [

Uni

vers

ity o

f L

ondo

n] a

t 07:

07 0

3 Ju

ly 2

016

was statistically controlled. Neither the three-way

interaction nor the DM Number�Age interaction

reached significance (R2 � .30).

We found that Approach/Withdrawal scores were

positively associated with DM number, t(95)�2.161, pB.05. We also found a significant age

effect: Older children showed lower Approach/

Withdrawal scores, F(2, 95)�5.033, pB.01, if

DM number was statistically controlled. The

mean Approach/Withdrawal scores for the three

age groups (47�55, 60�72, and 90�102 months)

were 38.125 (0.926 SE), 36.135 (0.881 SE), and

30.658 (0.816 SE), respectively (R2�.16). (See

Figure V.) No other significant main effects or

interactions were found.

Are Verbal Abilities Associated with Specific

DMs?

We computed the average WPPSI/WISC-R Verbal

scores based on the presence/absence of each DM. A

significant difference was found for rationalization,

suppression, and identification (Figure VI). Chil-

dren using rationalization scored significantly higher

(M�12.81, SE�0.247) on the WPPSI/WISC-R

Verbal subtest than those who did not (M�10.950, SE�0.270), t(97)�3.010, pB.005. Chil-

dren using suppression scored significantly lower on

the WPPSI/WISC-R Verbal subtest (M�11.176,

SE�2.404) than those who did not (M�12.671,

SE�2.750), t(97)�2.086, pB.05. Children using

Identification (M�11.210, SE�2.262) scored sig-

nificantly lower on the WPPSI/WISC-R Verbal

subtest than those who did not (M�12.700, SE�0.278), t(97)�2.176, pB.05. No difference was

found when grouping the children according to the

presence/absence of the other DMs.

Discussion

We used DHP to study the development of DMs in

children ages 47 to 102 months. In the four episodes

of pretend play that we had examined, children’s use

of DMs varied with age and gender and was

associated with temperament and verbal abilities.

Age Effects

We found that the number of DMs decreased with

age, as shown in Figure III (see also Brody et al.,

1985). This result is consistent with the hypothesis

that at about 84 to 102 months of age children enter

a latency period, when they overcome their internal

conflicts and direct attention and energy toward the

external world (Freud, 1923/1961). Consistently

with A. Freud’s (1936/1965) hypothesis that de-

fenses have different developmental patterns, we

found that the likelihood of observing each DM

varied with age. For example, displacement, ideali-

zation, or regression was observed less frequently for

children ages 90 to 102 months than for children

ages 60 to 72 months. Denial and regression,

generally viewed as immature modes of functioning,

were used more often by children ages 60 to 72

months, which would explain why their presence

diminished with cognitive development and with

increased complexity in emotional functioning (A.

Freud, 1936/1965; Laor et al., 2001).

The developmental pattern of rationalization was

the clearest of all, showing that the odds of observing

this DM clearly increased with age. For the oldest

age group, moreover, rationalization was the DM

used more frequently. For children ages 90 to 102

months, the odds of observing rationalization were

3.02 to 92.07 times larger than those of any other

DM. This result is consistent with the establishment

of concrete operational thinking and with the

Figure V. (A and B) Effects of Approach/Withdrawal on age, gender, and number of defense mechanisms (DMs). Dashed lines are

pointwise 95% confidence bands around fitted effects.

Development of childhood defense mechanisms 541

Dow

nloa

ded

by [

Uni

vers

ity o

f L

ondo

n] a

t 07:

07 0

3 Ju

ly 2

016

maturation of processes that underlie categorical

classification capacities (Keil, 1989; Sloutsky et al.,

2001).

Gender Effects

Speculations about gender differences in children’s

defenses have been formulated by hypothesizing that

girls tend to use internalizing defenses and that boys

tend to use externalizing defenses (Brody et al.,

1985; Laor et al. 2001). Consistent with this

hypothesis, we found that regression (an internaliz-

ing defense) was used more by girls than by boys;

conversely, denial, which is related to the external

reality, was more present in boys (see also Feldman

et al., 1996).

Gender-related differences in DM use may be

influenced by the different attitudes that parents and

teachers have toward boys and girls (Fagot & Hagan,

1991) but also by a gender-specific reorganization

involving oedipal contents that takes place at about

60 to 78 months of age (Watson & Getz, 1990;

Yorke, Wiseberg, & Freeman, 1989).

Temperament

Although relatively little research has been con-

ducted with children, the temperament�DM con-

nection is well established in adults (Wolmer et al.,

2001). For example, an association between more

‘‘mature’’ defenses and successful life adaptation has

been found in female adolescents (Araujo, Ryst, &

Steiner, 1999). In the present research, when

comparing TABC scores with the number of differ-

ent DMs used by each participant, we found that

children adjusting less easily to changes in rules or to

a new context (Adaptability) used a larger number of

DMs. Moreover, children who were less outgoing

toward peers and adults (Approach/Withdrawal)

tended also to use a larger number of DMs. The

association between TABC scores and DM fre-

quency in the present research is thus consistent

with the hypothesis that, when used too frequently,

DMs can be seen as a sign of maladjustment

(Sandstrom & Cramer, 2003a,b).

Several issues concerning the DMs remain con-

troversial, one of them being whether defenses are

pathological or whether there is such a thing as a

‘‘normal’’ defense. Our findings suggest that DMs

are ‘‘normal,’’ because they were expressed by all our

‘‘normal’’ children, but their disproportionate use is

an indicator of psychological difficulties (see also

Lingiardi & Madeddu, 2002).

Verbal Intelligence

A positive association between IQ and rationaliza-

tion has been found in adults and adolescents (Haan,

1963; Schimek, 1968). Consistent with these results,

we found that children who used rationalization

performed significantly better on the WPPSI/

WISC-R Verbal subtest. Moreover, we also found

that the use of identification and suppression (two

DMs often recognized as primitive; e.g., Vaillant,

1992) was associated with lower WPPSI/WISC-R

Verbal subtest scores.

Conclusion

The DHP task allows normal children to reveal their

DMs in an ecologically satisfactory manner. The

analysis of children’s pretend play provides conver-

ging evidences that DMs relate to central aspects of

the emotional lives of children and manifest them-

selves differently depending on age, gender, tem-

perament, and verbal skills. DM use is consistent

with major shifts in many domains, including emo-

tional, ego, and self-concept development. More-

over, DM presence stems from the ways in which the

individual relates with oneself and the human

environment.

The DHP may prove to be an asset for the

psychotherapist because information gained from

an assessment of DMs is beneficial for treatment

Figure VI. Wechsler Preschool and Primary Scale of Intelligence, Revised (WPPSI) and Wechsler Intelligence Scale for Children, Revised

(WISC-R) Verbal scores as a function of the presence or absence of rationalization, suppression, and identification. Vertical bars represent

91 SEM.

542 M. A. Tallandini and C. Caudek

Dow

nloa

ded

by [

Uni

vers

ity o

f L

ondo

n] a

t 07:

07 0

3 Ju

ly 2

016

planning and intervention (Siefert, Hilsenroth,

Weinberger, Blagys, & Ackerman, 2006). A deeper

knowledge of the nature of the DMs and how they

operate in child development may be useful to bridge

the gap among theoretical knowledge, empirical

data, and clinical intervention.

Acknowledgements

This research was funded by The Tavistock Medical

School (London) and Ministero Istruzione, Univer-

sita Ricerca (Rome). We wish to thank Paola Dario,

Elisa Pavan, and Cristina Suplina for their contribu-

tions in data collection and coding. We wish to thank

the two anonymous reviewers for their valuable

comments and suggestions on a previous version of

this manuscript.

Note1 Note, however, that the DSM-IV-TR (American Psychiatric

Association, 2000) uses the terms defense mechanisms and coping

mechanisms interchangeably to refer to both defense mechanisms

and coping styles, although the overlapping of these two

concepts is disputed (Erdelyi, 2001).

References

American Psychiatric Association. (2000). Diagnostic and statistical

manual of mental disorders (4th ed., text rev.). Washington, DC:

American Psychiatric Association.

Araujo, K., Ryst, E., & Steiner, H. (1999). Adolescent defense

style and life stressors. Child Psychiatry & Human Development,

30, 19�28.

Araujo, K. B., Medic, S., Yasnovsky, J., & Steiner, H. (2006).

Assessing defense structure in school-age children. Using the

Response Evaluation Measure-71-YouthVersion (REM-Y-71).

Child Psychiatry & Human Development, 36, 427�436.

Baayen, R. H., Davidson, D. J., & Bates, D. M. (2008). Mixed-

effects modeling with crossed random effects for subjects and

items. Journal of Memory and Language, 59, 390�412.

Bates, D. M., & Sarkar, D. (2007). lme4: Linear mixed-effects models

using S4 classes, R package version 0.999375�32. Unpublished

manuscript.

Bond, M., Gardner, S. T., Christian, J., & Sigal, J. J. (1983).

Empirical study of self-rated defense styles. Archives of General

Psychiatry, 40, 333�338.

Brody, L. R., Rozek, M. K., & Muten, E. O. (1985). Age, sex, and

individual differences in children’s defensive styles. Journal of

Clinical Child Psychology, 14, 132�138.

Conte, H. R., Plutchick, R., & Draguns, J. G. (2004). The

measurement of ego defenses in clinical research. In U.

Hentschel, G. Smith, J. G. Draguns & W. Ehlers (Eds.),

Defense mechanisms: Theoretical, research and clinical perspectives

(pp. 393�414). Oxford, UK: Elsevier Science.

Cooper, S. H. (1998). Changing notions of defense within

psychoanalytic theory. Journal of Personality, 66, 947�964.

Cramer, P. (1991). The development of defense mechanisms. New

York: Springer Verlag.

Cramer, P. (1996). Storytelling, narrative, and the Thematic

Apperception Test. New York: Guilford Press.

Cramer, P. (1997). Evidence for change in children’s use of

defense mechanisms. Journal of Personality, 65, 233�247.

Cramer, P., & Block, J. (1998). Preschool antecedents of defense

mechanism use in young adults: A longitudinal study. Journal of

Personality and Social Psychology, 74, 159�169.

Cramer, P., & Gaul, R. (1988). The effects of success and failure

on children’s use of defense mechanisms. Journal of Personality,

56, 729�742.

Despland, J.-N., de Roten, Y., Despars, J., Stigler, M., & Perry,

J. C. (2001). Contribution of patient defense mechanisms and

therapist interventions to the development of early therapeutic

alliance in a brief psychodynamic investigation. Journal of

Psychotherapy Practice and Research, 10, 155�164.

Dunn, J. F., & Kendrick, C. (1982). Siblings: Love, envy and

understanding. Cambridge, MA: Harvard University Press.

Emde, R., Wolf, D., & Oppenheim, D. (2003). Revealing the inner

worlds of young children: The MacArthur Story Stem Battery and

Parent-Child Narratives. New York: Oxford University Press.

Erdelyi, M. H. (2001). Defense processes can be conscious or

unconscious. American Psychologist, 56, 761�762.

Fagot, B. I., & Hagan, R. (1991). Observations of parents’

reactions to sex-stereotyped behaviors: Age and sex effects.

Child Development, 62, 617�628.

Feldman, S. S., Araujo, K. B., & Steiner, H. (1996). Defense

mechanisms in adolescents as a function of age, sex, and mental

health status. Journal of American Academy of Child and

Adolescent Psychiatry, 35, 1344�1354.

Fischer, K., Shaver, P. R., & Carnochan, P. (1990). How emotions

develop and how they organise development. Cognition and

Emotion, 4, 81�127.

Freud, A. (1965). The ego and the mechanisms of defense. New York:

International Universities Press. (Original work published

1936).

Freud, S. (1959). Inhibitions, symptoms and anxiety. In J.

Strachey (Ed. and Trans.), The standard edition of the complete

psychological works of Sigmund Freud: Volume 20 (pp. 77�174).

London: Hogarth Press. (Original work published 1926)

Freud, S. (1961). The ego and the id. In J. Strachey (Ed. and

Trans.), The standard edition of the complete psychological works of

Sigmund Freud: Volume 19 (pp. 12�66). London: Hogarth Press.

(Original work published 1923)

Freud, S. (1962). The neuro-psychoses of defence. In J. Strachey

(Ed. and Trans.), The standard edition of the complete psycholo-

gical works of Sigmund Freud: Volume III (1893�1899). Early

psycho-analytic publications (pp. 41�61). London: Hogarth

Press. (Original work published 1894)

Glasberg, R., & Aboud, F. E. (1982). Keeping one’s distance from

sadness: Children’s self-reports of emotional experience. Devel-

opmental Psychology, 18, 287�293.

Haan, N. (1963). Proposed model of ego functioning: Coping and

defense mechanisms in relationship to IQ change. Psychological

Monographs, 77, 1�23.

Haan, N. (1974). The adolescent antecedents of an ego model of

coping and defense and comparisons with Q-sorted ideal

personalities. Genetic Psychology Monographs, 89, 273�306.

Hauser, S., & Safyer, A. (1995). The contributions of ego

psychology to developmental psychopathology. In D. J. Cohen

& D. Cicchetti (Eds.), Developmental psychopathology: 1. Theory

and methods (pp. 555�580). Oxford, UK: Wiley.

Hentschel, U., Kiessling, M., Teubner-Berg, H., & Dreier H.

(2004). Intellectual performance and defence. The measure-

ment of ego defenses in clinical research. In U. Hentschel,

G. Smith, J. G. Draguns, & W. Ehlers (Ed.), Defense mechan-

isms: Theoretical, research and clinical perspectives (pp. 431-451).

Oxford, UK: Elsevier Science.

Hentschel, U., Smith, G., Draguns, J. G., & Ehlers, W. (2004).

Defense mechanisms: Current approaches to research and

measurement. In U. Hentschel, G. Smith, J. G. Draguns, &

Development of childhood defense mechanisms 543

Dow

nloa

ded

by [

Uni

vers

ity o

f L

ondo

n] a

t 07:

07 0

3 Ju

ly 2

016

W. Ehlers (Ed.), Defense mechanisms: Theoretical, research and

clinical perspectives (pp. 3-41). Oxford, UK: Elsevier Science.

Hill, K. T., & Sarason, S. B. (1966). The relation of text anxiety

and defensiveness to test and school performance over the

elementary school years: A further longitudinal study. Mono-

graphs of the Society for Research in Child Development, B(2, Serial

No. 104).

Keil, F. (1989). Concepts, kinds and conceptual development.

Cambridge, MA: MIT Press.

Kragh, U. (1969). The Defense Mechanism Test. Stockholm:

Testfoerlaget.

Laor, N., Wolmer, L., & Cicchetti, D. V. (2001). The compre-

hensive assessment of defense style: Measuring defense

mechanisms in children and adolescents. Journal of Nervous

and Mental Diseases, 189, 360�368.

Lichtenberg, J. D. (1983). Psychoanalysis and infant research.

Hillsdale, NJ: Erlbaum.

Lingiardi, V., & Madeddu, F. (2002). I meccanismi di difesa [The

defense mechanisms]. Milan: R. Cortina Editore.

Lumbelli, L. (1994). Pedagogia della comunicazione verbale [Verbal

communication educational psychology]. Milan: F. Angeli.

Martin, R. P. (1988). The assessment of personality and behavior

problems: Infancy through adolescence. New York: Guilford Press.

McHale, J. P., Neugebauer, A., Radin, A. A., & Schwartz, A.

(1999). Preschooler’s characterization of multiple family rela-

tionships during family doll’s play. Journal of Clinical Child

Psychology, 28, 256�268.

Minnis, H., Millward, R., Sinclair, C., Kennedy, E., Greig, A.,

Towlson, K., et al. (2006). The computerized MacArthur Story

Stem Battery. A pilot study of a novel medium for assessing

children’s representations of relationships. International Journal

of Methods in Psychiatric Research, 15, 207�214.

Oppenheim, D., Emde, R. N., & Warren, S. (1997). Children’s

narrative representation of mothers: Their development

and associations with child and mother adaptation. Child

Development, 68, 127�137.

Perry, J. C., & Cooper, S. H. (1989). An empirical study

of defense mechanisms. Archives of General Psychiatry, 46,

444�452.

Sammallahti, P., & Aalberg, V. (1995). Defense style in personality

disorders. An empirical study. Journal of Nervous and Mental

Disease, 183, 516�521.

Sandler, J., & Freud, A. (1985). The analysis of defense: The ego and

the mechanisms of defense revisited. New York: International

Universities Press.

Sandstrom, M. J., & Cramer, P. (2003a). Defense mechanisms

and psychological adjustment in childhood. Journal of Nervous

and Mental Disease, 19, 487�495.

Sandstrom, M. J., & Cramer, P. (2003b). Girls’ use of defense

mechanisms following peer rejection. Journal of Personality, 71,

605�627.

Schimek, J. G. (1968). Cognitive style and defenses: A long-

itudinal study of intellectualization and field independence.

Journal of Abnormal Psychology, 73, 575�580.

Shaw, J., & Ryst, E. (1996). Temperament as a correlate of

adolescent defense mechanisms. Child Psychiatry & Human

Development, 96, 105�114.

Siefert, C. J., Hilsenroth, M. J., Weinberger, J., Blagys, M. D., &

Ackerman, S. J. (2006). The relationship of patient defensive

functioning and alliance with therapist technique during short-

term psychodynamic psychotherapy. Clinical Psychology and

Psychotherapy, 13, 20�33.

Sloutsky, V. M., Lo, Y., & Fisher, A. V. (2001). How much does

a shared name make things similar? Linguistic labels, similarity,

and the development of inductive inference. Child Development,

72, 1695�1709.

Smith, G. J. W., & Danielsson, A. (1982). Anxiety and defense

strategies in childhood and adolescence. New York: International

Universities Press.

Vaillant, G. E. (1992). Ego mechanisms of defense: A guide for

clinicians and researchers. Washington, DC: American Psychia-

tric Association.

Wallerstein, R. S. (1985). Defenses, defense mechanisms, and the

structure of the mind. In H. P. Blum (Ed.), Defense and

resistance (pp. 201�225). New York: International Universities

Press.

Watson, M. W., & Getz, K. (1990). Developmental shift in oedipal

behaviors related to family role understanding. New Directions

for Child Development, 48, 29�45.

Wechsler, D. (1989). Wechsler Preschool and Primary Scale of

Intelligence, Revised (WPPSI-R). San Antonio, TX: Psycholo-

gical Corporation.

Wechsler, D. (2003). Wechsler Intelligence Scale for Children, Revised

(WISC-IV-R). San Antonio, TX: Psychological Corporation.

Wolmer, L., Laor, N., & Cicchetti, D. V. (2001). Validation of the

Comprehensive Assessment of Defense Style (CADS):

Mothers’ and children’s responses to the stresses of missile

attacks. Journal of Nervous and Mental Disease, 189, 369�376.

Yorke, C., Wiseberg, S., & Freeman, T. (1989). Development and

psychopathology: Studies in psychoanalytic. New Haven, CT: Yale

University Press.

544 M. A. Tallandini and C. Caudek

Dow

nloa

ded

by [

Uni

vers

ity o

f L

ondo

n] a

t 07:

07 0

3 Ju

ly 2

016

Appendix A

Defense Mechanisms: Definitions and Examples

Defense mechanism Definition Examples

Acting out Child deals with emotional conflicts, or with internal/external stressors,

by acting rather than by expressing feelings or by thinking over what happened.

At mealtime, the mother doll reproaches the child doll for being untidy; the child doll throws

his plate against the wall.

Denial Child avoids problems, wishes, feelings, conflicts by putting them on hold. The experimenter asks, ‘‘Who made this noise?’’ The child does not answer.

During sad (bad) time episode, the child says, ‘‘At my home, we never have a bad time.’’

Displacement Child displaces emotional conflicts to less threatening objects. The feelings

stay the same, but their object changes.

During mealtime, the child doll does not talk about her family when it would be appropriate;

rather, she speaks at length about her dog, clearly making the pet act as one of the family

components.

Idealization Child attributes exaggerated positive qualities to an object or person. In the sad (bad) time episode, the child tells a story like the following: During an earthquake,

the dollhouse collapses, but the father doll is able to save the mother doll from the ruins.

Afterward, the father doll rebuilds the house and makes it as nice as it was before the

earthquake.

Identification To avoid internal conflicts, child models self on another person’s character

and behavior.

After mealtime, the child doll greets the mother doll and pretends that he must start his truck

driver’s shift, while leaving the mother doll at home alone.

Passive aggression Child expresses indirect or passive aggression toward others. During the mealtime episode, the father doll reproaches the mother doll with a nasty

comment. The child doll says, ‘‘Dad broke a plate and while going to fetch a broom, he

stumbled and badly hurt himself.’’

Rationalization Through faulty and false reasoning, child modifies emotional conflicts by

elaborating seemingly logical and acceptable explanations.

The father doll and the mother doll quarrel. The child doll says, ‘‘They are not fighting, just

discussing.’’

The child doll wants to sleep in his parents’ bed because ‘‘those two beds have no gap in

between.’’

Reaction formation Child converts behaviors, thoughts, or feelings that are perceived to be

unacceptable into their opposites.

With force, the child doll wraps a scarf around the neck of his sister doll and then turns his

aggressive action into a friendly emotion, saying, ‘‘You are so beautiful, I really love you.’’

(Kragh, 1969)

Regression Rather than handling emotional conflicts and internal/external stressors in

a more mature fashion, child regresses to behaviors from an earlier stage of

development.

The child doll is placed in a high chair as if she was a baby; the child doll sits on the father

doll’s shoulders, sucking her fingers or a pacifier.

Repression Child deals with emotional conflicts or internal/external stressors by expelling

disturbing wishes, thoughts, or experiences from conscious awareness.

The emotion remains conscious but is detached from the idea behind.

The father doll scolds the child doll for his bad behavior. The child doll expresses his anger in a

generalized fashion, never directing it toward the father doll.

Suppression Child deals with emotional conflicts or internal/external stressors by

intentionally avoiding thinking about disturbing problems, wishes, feelings, or

experiences.

The mother doll and father doll have an argument. When the experimenter asks, ‘‘What

happened afterward?’’ the child switches to a different topic.

Undoing Child deals with emotional conflicts by using words or behaviors that are

designed to negate or to symbolically amend unacceptable thoughts, feelings, or

actions.

During the bedtime episode, the child doll is very anxious and reluctant to go to bed.

Suddenly, the child doll starts to behave as if it were already waking-up time and runs down

the stairs and goes to eat breakfast.

Dev

elopm

ent

ofch

ildhood

defen

sem

echanism

s545

Dow

nloa

ded

by [

Uni

vers

ity o

f L

ondo

n] a

t 07:

07 0

3 Ju

ly 2

016