Content uploaded by Daniel Gassmann
Author content
All content in this area was uploaded by Daniel Gassmann on Nov 21, 2017
Content may be subject to copyright.
Clinical Psychology and Psychotherapy
Clin. Psychol. Psychother. 13, 1–11 (2006)
Published online in Wiley InterScience (www.interscience.wiley.com) DOI: 10.1002/cpp.442
Copyright © 2006 John Wiley & Sons, Ltd.
General Change Mechanisms:
The Relation Between Problem
Activation and Resource Activation
in Successful and Unsuccessful
Therapeutic Interactions
Daniel Gassmann* and Klaus Grawe†
Department of Clinical Psychology and Psychotherapy, University of Bern,
Switzerland
Using the Consistency Theory Micro-Process Analysis (CMP), the rela-
tion between two general change mechanisms, problem activation
and resource activation, was studied with the focus on (1) patient
behaviour and (2) therapist intervention. The unit of analysis was one
minute. The results show that problem and resource activation play
different roles in the process of change: problem activation alone did
not reliably lead to therapeutic progress; only when combined with
thorough resource activation could it unfold its therapeutic potential.
Successful therapists chose different degrees of and different timing
in applying the two change mechanisms than unsuccessful ones. The
results indicate that clearer conceptualizations and specific therapist
training are necessary to make better use of resource activation.
Copyright © 2006 John Wiley & Sons, Ltd.
†Prof. Dr. Klaus Grawe is now deceased.
*Correspondence to: Dr. Daniel Gassmann, Universität Bern,
Psychotherapeutische Praxisstelle, Mittelstrasse 42, CH-3012
Bern, Switzerland.
E-mail: daniel.gassmann@ptp.unibe.ch
To improve the benefits of psychotherapy we need
to learn more about the processes underlying ther-
apeutic change. In the past few years psychother-
apy research has shown that therapeutic change is
induced less by specific techniques than by more
pantheoretical factors (Hubble, Duncan, & Miller,
1999). Therefore, the healing power of psychother-
apy arises from the realization of general principles
represented by different therapy schools, although
they might be named and explained differently. In
the field of psychotherapy process research several
significant variables inducing or moderating ther-
apeutic change have been identified (Orlinsky,
Grawe, & Parks, 1994). On the basis of broad
empirical data, Grawe, Donati, and Bernauer (1994;
Grawe, 1997) extracted the following five change
mechanisms: (1) the therapeutic bond, (2) problem
activation, (3) resource activation, (4) mastery and
(5) motivational clarification. They assumed that
the realization of these general change mechanisms
promotes therapeutic progress. However, the
relationship between these different change
mechanisms is as yet unclear. Are some mecha-
nisms more important than others? Can or should
these mechanisms be combined in specific thera-
peutic interventions? In the present study we focus
on the inter-relations between problem activation
and resource activation. Our aim was to investigate
how these two principles should be used in the
clinical practice to support therapeutic change.
The change mechanism problem activation refers
to the finding that a patient must come into direct
contact with painful emotions to overcome his or
her problems. This principle of therapeutic change
2D. Gassmann and K. Grawe
Copyright © 2006 John Wiley & Sons, Ltd. Clin. Psychol. Psychother. 13, 1–11 (2006)
DOI: 10.1002/cpp
is stressed in all major therapy schools (Orlinsky
et al., 1994; Beutler et al., 2004). In behaviour
therapy, problem activation becomes salient while
exposing the patient to previously avoided stimuli
(Barlow, 1993). Exposure therapy is only effective
when the fear-provoking situation actually leads
to psychophysiological arousal (Craske, 1999;
Emmelkamp, 2004). Experiential approaches stress
the importance of arousing a patient’s emotions.
The therapist’s task is seen as focussing on the
emotional core of the theme the patient is working
on in therapy (Gendlin, 1981; Beutler, Booker, &
Peerson, 1998; Greenberg, 2002). In psychody-
namic therapies, problem activation is evident for
example in the process of transference when the
patient projects negative emotions on the therapist
(Blum, 2003; Smith, 2003). In problem activation
we prefer to distinguish between approaching emo-
tional behaviour, in which the patient focusses on
painful emotions (e.g. experiencing fear in a con-
frontation experiment), and avoiding emotional
behaviour, in which the patient experiences painful
emotions but tries to distract his attention from the
source that troubles him (e.g. a patient laughing
while talking about a failure in an important
exam).
Resource activation is an empirically strongly sup-
ported change mechanism. In essence, this princi-
ple is realized in interventions that focus not on
the patient’s problems but rather on the sound
and healthy parts of the patient’s personality
(Gassmann & Grawe, 2004; de Shazer, 1988). This
view of the patient as being capable differs
strongly from the traditional deficit-oriented way
of describing the patient’s problems. One area in
which psychotherapists have interacted with
healthy aspects of the patient’s personality ever
since is the therapeutic relationship. If a bond
develops between the patient and therapist, which
is maintained by the therapist through warm,
empathic, congruent behaviour and which is char-
acterized on the patient’s side by openness and
trust, the chances for a good therapy outcome are
very high (Lambert & Bergin, 1994; Bachelor &
Horvath, 1999; Lambert & Ogles, 2004). However,
resource activation as we refer to it here goes dis-
tinctly beyond the realization of the Rogerian vari-
ables: resources can be found in a wide range of
specific intra- and interpersonal strengths and abil-
ities, for example a person’s looks, intellectual
capability, specific skills, motivational goals, sup-
porting family members and so forth (Bohart &
Tallman, 1999; Grawe & Grawe-Gerber, 1999;
Grawe, 2004). Resource-activating interventions
are specific therapeutic behaviours supporting a
productive therapeutic relationship (Caspar, 1995)
as well as distinctly reinforcing specific strengths
and abilities of the patient (Gassmann & Grawe,
2004).
Problem and resource activation both play a
central role in explaining how patients change
during therapy. On the one hand, the patient must
come into contact with previously avoided emo-
tions to overcome them. On the other, the patient
needs to experience her- or himself as being more
than the sum of the problems. Only then will the
patient be motivated to deal with his or her prob-
lematic sides. To employ these two principles of
change well in the clinical practice, we must know
when and in what manner they should be applied.
To our knowledge no previous study exists on the
relationship between problem and resource acti-
vating processes in the therapeutic interaction. Our
study aims to fill this gap.
From a methodological point of view one impor-
tant step in providing access to these processes is
to leave behind the research paradigm of pre—post
comparisons of treated patient groups. This kind of
research produces average effects of a given thera-
peutic intervention but sheds no light on the
mechanisms generating these effects. To catch the
moments of therapeutic change, research must
focus on units smaller than therapy outcome, such
as on session outcome, or even more appropriate
on sequences within sessions. In order to learn
more about the relationship between problem and
resource activation an exploratory research design
is necessary to study naturally occuring therapeu-
tic interactions (Hill & Lambert, 2004).
METHOD
Design and Procedure
Thirty therapies from a large sample of outpatients
treated at the Outpatient Clinic of the University of
Bern, Switzerland, were selected according to their
therapy outcome (see below). They were divided
into three groups (nine patients with poor, eleven
patients with moderate and ten patients with very
good therapy outcome). From each therapy, four
sessions were selected according to session
outcome (see below) and distributed into two
groups (low and high session outcome), which
added up to a sample of 120 sessions (see Figure 1).
Therapy Outcome
Therapy outcome was determined by a compos-
ite measure consisting of five tests measuring goal
Problem and Resource Activation in Therapeutic Interactions 3
Copyright © 2006 John Wiley & Sons, Ltd. Clin. Psychol. Psychother. 13, 1–11 (2006)
DOI: 10.1002/cpp
attainment (Kiresuk & Sherman, 1968; Kiresuk &
Lund, 1979), emotional and behavioral changes
(Zielke, 1978), changes in important relationships
(Grawe, 1982, 2002) and patient and therapist post-
therapy evaluation (Regli & Grawe, in prepara-
tion). The mean effect size (Cohen’s d) of these five
measures was below 1.0 for the group with low
therapy outcome, between 1.0 and 2.0 for the
group with moderate outcome and above 2.0 for
the group with high outcome.
To ensure that the differences in outcome were
attributed to the therapeutic interventions rather
than to patient variables already existing before
treatment, the scores of a broad pre–post battery
measuring psychopathology (Derogatis, 1994;
Franke, 1995), well-being (Grob, 1995) and interper-
sonal problems (Horowitz, Rosenberg, Bauer,
Ureno, & Villasenor, 1988; Horowitz, Strauss, &
Kordy, 1994) were compared. The three outcome
groups (low, moderate and high) did not differ in any
of these measures at pretest at the level of p<0.05.
Session Outcome
After each session patients completed the Patient
Session Questionnaire (Regli & Grawe, in prepara-
tion). The quality of a session was determined by
the summary score of the scale ‘corrective experi-
ences’1consisting of six items. This scale indicates
the degree of mastery experiences, clarification
experiences or both that the patient made (from his
or her point of view) in the session. From each
therapy the two sessions with the lowest and the
two sessions with the highest scores were selected.
Patients
17 patients were female, 13 male. The average age
was 33.8 years (range 18–56; SD =8.8). All patients
chosen fullfilled the criterion of at least one diag-
nosis on Axis I according to DSM-IV. 40% had an
affective and 33% an anxiety disorder as the main
diagnosis (see Table 1). 47% of the patients met the
criteria of a second and 13% of a third diagnosis on
Axis I. 20% of the patients clearly and 27% less
clearly fullfilled comorbidly the criteria for a per-
sonality disorder on Axis II. Average global assess-
ment of functioning (gaf) was 62.4 (range 37–85,
SD =9.3).
Treatment
Therapists were nine female and 13 male graduate
psychologists. Four therapists contributed two and
one therapist three therapies. 20 therapies were
conducted by experienced therapists, ten therapies
by therapists still in training. All therapies were
under constant supervision by experienced super-
visors. Patients were treated according to the
guidelines of consistency theoretical psychother-
apy (Grawe, 2004), an integrative therapy
approach based on empirically supported general
change mechanisms. The average duration of the
therapies was 23 sessions (range 7–54, SD =12).
Process Analysis
The 120 sessions were videotaped and analysed
minute by minute with the Consistency Theory
Micro-Process Analyses (CMP; Gassmann, 2002).
The CMP is an observation-based rating procedure
therapy outcome
low
(es < 1.0)
moderate
(es < 2.0)
high
(es > 2.0)
low
18
22
20
60
session
outcome
high
18
22
20
60
36
44
40
120
Figure 1. Two factorial design: Factor 1, therapy outcome; Factor 2, session outcome. es =effect size (Cohen’s d)
1The internal consistency (Cronbach’s alpha) of the scale was
r=0.83 in a sample of N=779 sessions.
4D. Gassmann and K. Grawe
Copyright © 2006 John Wiley & Sons, Ltd. Clin. Psychol. Psychother. 13, 1–11 (2006)
DOI: 10.1002/cpp
used by trained raters. Four different perspectives
were judged independently: (1) the content of the
therapeutic interaction, (2) the degree of problem
activation, (3) the degree of resource activation and
(4) the immediate outcome of the therapeutic inter-
action. In the present study we present data from
perspectives (2) and (3). Each perspective focussed
in one rating passage on the patient and in a
second passage on the therapist. In the present
study four graduate students were trained, two in
assessing the degree of problem activation and the
other two in assessing the degree of resource acti-
vation. The raters were trained for two weeks fol-
lowed by several consolidating meetings to avoid
rater drift. Table 2 depicts the inter- and intrarater
reliabilities obtained in the present study.
RESEARCH QUESTIONS
In the present investigation we focussed on three
questions related to different levels of the therapy
process.
1. Therapy level. Are there differences in the real-
ization of problem activation and resource acti-
vation in therapies with low, moderate and high
outcome? This question was addressed by com-
paring the means between the three outcome
groups and checked by t-tests.
Table 1. Main diagnosis on axis I, DSM-IV
Diagnosis Code Quantity %
Major Depressive Disorder, single episode, mild 296.21 1
Major Depressive Disorder, single episode, moderate 296.22 1
Major Depressive Disorder, recurrent, mild 296.31 2
Major Depressive Disorder, recurrent, moderate 296.32 5
Bipolar II Disorder 296.89 1
Depressive Disorder Not Otherwise Specified 311 2
Total mood disorders 12 40.0
Panic Disorder without Agoraphobia 300.01 2
Panic Disorder with Agoraphobia 300.21 2
Social Phobia 300.23 2
Specific Phobia 300.29 2
Obsessive–Compulsive Disorder 300.3 1
Post-traumatic Stress Disorder 309.81 1
Total anxiety disorders 10 33.3
Somatization Disorder 300.81 2
Male Erectile Disorder 302.72 1
Cannabis withdrawal 304.30 1
Pain Disorder Associated with Psychological Factors 307.80 1
Adjustment Disorder with Depressed Mood 309.0 1
Personality Change Due to a General Medical Condition 310.1 1
Academic Problem V 62.3 1
Total other disorders 8 26.7
Table 2. Average inter- and intrarater reliabilities in the
perspectives of processual problem activation and
processual resource activation of the CMP. p.a. =percent
agreement; k=Cohens’s kappa
Perspective p.a. k
Processual problem activation
inter-rater reliability
focus on patient 70.2 0.50
focus on therapist 87.9 0.45
intrarater reliability
focus on patient 87.9 0.80
focus on therapist 95.6 0.82
Processual resource activation
inter-rater reliability
focus on patient 93.2 0.76
focus on therapist 96.8 0.71
intrarater reliability
focus on patient 95.2 0.84
focus on therapist 97.7 0.85
2. Session level. Are there different patterns of
problem activation and resource activation
within the course of successful and unsuccess-
ful sessions? This was answered by a descrip-
tive illustration of the session course. Since the
sessions were not of exactly the same length
(50–60 minutes), each session was transformed
Problem and Resource Activation in Therapeutic Interactions 5
Copyright © 2006 John Wiley & Sons, Ltd. Clin. Psychol. Psychother. 13, 1–11 (2006)
DOI: 10.1002/cpp
to a length of 100%. Thus it was possible to
summarize the course of several sessions. The
sessions with low session and low therapy
outcome were defined as unsuccessful (a total of
18 sessions). Successful sessions were those with
high session and high therapy outcome (total 20
sessions). The curves in Figure 4 represent a pro-
totypical, average session course.
3. Minute level. Are there differences in the
patient–therapist interactions regarding pro-
blem activation and resource activation on the
minute-by-minute level in successful and un-
successful therapeutic interactions? To answer
this, cross-correlations between patient and
therapist behavior were computed. Again, the
sessions with low session and low therapy
outcome were considered unsuccessful (total
18 sessions) and those with high session and
high therapy outcome were successful (total 20
sessions).
RESULTS
Therapy Level
Problem Activation
In relation to the whole sample, painful emotions
were activated on average in patients with low
outcome (see ‘intensity’ on the left-hand side of
Figure 2). Patients in therapies with moderate
outcome experienced more, patients in therapies
with high outcome less, negative emotion. Quali-
tatively unsuccssful patients showed more avoid-
ing and less approaching emotional behaviour,
whereas the successful patients showed the oppo-
site picture: less emotional avoiding and more
emotional approaching behaviour. All of these dif-
ferences are to be understood as trends; none of
them was statistically significant.
There were qualitative differences concerning
therapist intervention style in the three outcome
groups (see Figure 2, right-hand side): Therapists
of low outcome therapies reduced patient emo-
tional involvement clearly beyond average. The
comparison with the moderate outcome therapies
was significant (N=30, t=-2.3, df =18, p<0.05).
Successful therapists also seldom reduced the
patients’ emotional involvement but rather kept
patients on track with the emotional level they
already had reached. Emotion-intensifying inter-
ventions were used only rarely among the suc-
cessful therapists.
Resource Activation
The three outcome groups differed greatly
regarding resource activation: patients in therapies
with low outcome had far less activated resources
than patients with moderate (N=30, t=-2.9, df =
18, p<0.05) and high outcome (N=30, t=-3.5,
df =17, p<0.05) (see Figure 3, left side).
patient’s behaviour therapist’s intervention style
Figure 2. Z-tranformed means of patient processual problem activation and therapist processual problem-
activating interventions in therapies with low, medium and high outcome. The 0-line represents an average
activation. Only the difference between therapist interventions reducing patients emotional involvement in therapies
with low compared to medium outcome was significant at p<0.05
6D. Gassmann and K. Grawe
Copyright © 2006 John Wiley & Sons, Ltd. Clin. Psychol. Psychother. 13, 1–11 (2006)
DOI: 10.1002/cpp
Most interestingly, not only were there consider-
able differences in the patient groups but also in
the resource-activating interventions of the thera-
pists (see Figure 3, right side). Highly significant
differences between the three outcome groups
were observed: Therapists in low outcome thera-
pies activated patients resources far less often than
therapists in moderate outcome (N=30, t=-2.8,
df =18, p<0.05) and high outcome therapies (N=
30, t=-2.9, df =17, p<0.05).
Session Level
Figure 4 shows the patients’ activated resources
and therapists’ resource-activating interventions in
the course of unsuccessful (left-hand side) and suc-
cessful (right-hand side) sessions. In unsuccessful
sessions, patients entered the session with few acti-
vated resources (represented by the grey line in
Figure 4). During the course of the session, there
was no increase but rather a decrease in activated
patient’s behaviour therapist’s intervention style
Figure 3. Z-tranformed means of patient processual resource activation and therapist processual resource-
activating interventions in therapies with low, medium and high outcome. The 0-line represents an average
activation. The comparion of means between therapies with low and medium, as well as the comparison of means
between therapies with low and high, outcome was significant at p<0.05 for patient behaviour (left-hand side) and
therapist intervention style (right-hand side)
unsuccessful sessions successful sessions
Figure 4. Patient activated resources and therapist resource-activating interventions in the course of unsuccessful
(left-hand side) and successful sessions (right-hand side). The 0-line represents an average activation
Problem and Resource Activation in Therapeutic Interactions 7
Copyright © 2006 John Wiley & Sons, Ltd. Clin. Psychol. Psychother. 13, 1–11 (2006)
DOI: 10.1002/cpp
patient resources. The patient left the session with
even fewer activated resources! Therapists in
unsuccessful sessions rarely activated the patient’s
resources at the beginning of a session (represented
by the black line in Figure 4, left-hand side).
This did not change until the last quarter of the
session. Then they suddenly started trying to
activate patient resources, with no success, as can
be seen in the grey line below the black one in
Figure 4.
Figure 4 (right-hand side) demonstrates that the
successful patients (represented by the grey line)
entered the session with equally low activated
resources as the unsuccessful patients. However, in
contrast to the unsuccessful ones, successful
patients experienced a strong activation of their
strengths and abilities from the beginning of the
session on. During the course of the session their
resources remained activated to an above-average
degree. Yet their resource activation did not remain
stable. There was a decrease in activated resources
observed towards the middle and the end of the
session. All in all, the patients’ curve of activated
resources can be described by an inverted U-
distribution. Therapists in successful sessions
(represented by the black line in Figure 4) behaved
similarly to the patients: they distinctly activated
patient resources at the beginning of the session.
Towards the middle of the session, they reduced
that behaviour, but remained above average and
increased resource activation again towards the
end of the session.
When we analyse patients’ activated resources in
relation to patients’ activated negative emotions, it
becomes clear that unsuccessful patients experi-
enced more negative than positive emotions (see
Figure 5, left-hand side). During the course of the
session unsuccessful patients experienced negative
emotions more strongly towards the middle of the
session. This high activation level remained more
or less stable, only to sink shortly before the end of
the session.
When we compare the amount of activated
resources and activated negative emotions in the
course of a session for successful patients, we find
a different pattern (see Figure 5, right-hand side):
the very beginning of the session (about the first
five minutes) was dedicated exclusively to
resource activation. After that, successful patients
also experienced an activation of negative emo-
tions, which decreased over the course of the
session. With almost no exception, successful
patients experienced their resource as stronger
than their problems during the entire course of the
session.
Minute Level
To focus on the direct interaction between the
patient’s activated resources and the thera-
pist’s resource-activating interventions, cross-
correlations between these two variables were
computed. In unsuccessful sessions we found sig-
nificant negative cross-correlations (see Figure 6,
left-hand side). Positive lags in Figure 6 represent
patients’ activated resources as the leading vari-
able and therapists’ resource-activating interven-
unsuccessful sessions successful sessions
Figure 5. Patient activated problems and patient activated resources in the course of unsuccessful (left-hand side)
and successful sessions (right-hand side). The 0-line represents an average activation
8D. Gassmann and K. Grawe
Copyright © 2006 John Wiley & Sons, Ltd. Clin. Psychol. Psychother. 13, 1–11 (2006)
DOI: 10.1002/cpp
tions as the following variable. With negative lags
this was vice versa. The highly significant negative
cross-correlations are to be read as follows. When
resources were activated in unsuccessful patients,
they were not reinforced by resource-activating
interventions from the therapist, and/or when the
patient experienced his resources only to a limited
extent, therapists attempted to activate them.
Negative lags represent the therapist’s behaviour
as the leading variable and the patient’s behaviour
as the following variable. The highly negative cor-
relations indicate that when therapists tried to acti-
vate the patients’ resources there was no effect on
the patient; they did not have a stronger experience
their strengths and abilities. Negative correlations
also resulted when therapists did not intervene
with resource activating in situations in which
resources were activated by the patients.
In successful sessions of therapies with high
outcome we found very different patterns. Here
the cross-correlations were all highly significantly
positive (see Figure 6, right-hand side). Positive
lags again demonstrate the patient’s behaviour to
be the leading variable. In successful sessions we
observed that in situations when the patients expe-
rienced their resources they were reinforced at the
same time by resource-activating interventions by
their therapists. Negative lags represent the thera-
pists’ resource-activating interventions as the
leading variable. Here again we found highly sig-
nificant effects. When successful therapists tried to
activate patient resources, there was a strong pos-
itive response from the patients.
The last result we present relates to the use of
problem- and resource-activating interventions by
the therapists. Again, cross-correlations of sessions
with a poor outcome were compared with sessions
with high outcome (see Figure 7). In unsuccessful
sessions (see Figure 7, left-hand side) we found
strong negative correlations. In positive lags the
problem-activating interventions were the leading
variable and the resource-activating interventions
the following variable; in negative lags this was
vice versa. The strong negative correlations indi-
cate that therapists in unsuccessful sessions inter-
vened by either problem activating or resource
activating, but never by both within the same
minute. In successful sessions (see Figure 7, right-
hand side) the cross-correlation in lag 0—in one
and the same minute—is almost zero. This indi-
cates that successful therapists intervened during
one minute by either problem or resource activat-
ing or both at the same time.
DISCUSSION
In the present study we investigated two general
change mechanisms, problem activation and
resource activation. In evaluating the importance
of emotive interventions (problem activation),
Beutler et al. (2004) report inconsistent findings. In
unsuccessful sessions successful sessions
Figure 6. Cross-correlations between patient activated resources and therapist resource-activating interventions in
unsuccessful (left-hand side) and successful sessions (right-hand side). Positive lags show patient activated resources
as the leading variable; negative lags show the therapist resource-activating interventions as the leading variable
Problem and Resource Activation in Therapeutic Interactions 9
Copyright © 2006 John Wiley & Sons, Ltd. Clin. Psychol. Psychother. 13, 1–11 (2006)
DOI: 10.1002/cpp
some studies a positive correlation between the
patient’s emotional in-session arousal and therapy
outcome was found (e.g. Bond & Bunce, 2000;
Vaughan et al., 1994); in others it was not (e.g.
Borkovec & Costello, 1993; Stiles, Agnew-Davies,
Hardy, Barkham & Shapiro, 1998). As an explana-
tion for this inconsistency, Beutler et al. (2004) con-
sidered moderating factors. One moderating factor
extracted in serveral studies was the quality of the
therapeutic bond (which was studied here under
the perspective of resource activation).
The aggregation on the therapy level in our study
demonstrated that resource activation accounts for
much more of the outcome variance than problem
activation. Considering the fact that all major
therapy schools focus on patient problems, this
result is highly relevant. It indicates that problem
activation alone does not lead to a good therapy
outcome. The degree of resource activation,
however, distinguished unsuccessful therapies sig-
nificantly from therapies with moderate and high
outcome.
Nevertheless, we observed a trend that success-
ful patients showed more emotional approach
behaviour, whereas unsuccessful patients made
greater effort to avoid painful emotions. This cor-
relates well with psychotherapeutic concepts that
stress the role of confronting the patient with pre-
viously avoided emotions. However, activating a
patient’s problems seems to be two edged: our
results suggest that, especially when the patient is
avoiding threatening emotions, a further insisting
on problem-intensifying interventions seems to
be counterproductive. Here levelling or even
emotion-reducing interventions would probably
be more appropriate.
We not only found substantial differences in acti-
vating patient resources on the aggregated level of
therapy outcome, but also discovered different pat-
terns in the course of a session. Unsuccessful
therapists focussed more on the patient’s problems
and tended to overlook the patient’s resources. As
could be seen from the patient’s perspective, their
self-confidence and positive rapport sank the
longer the session lasted. The unsucessful thera-
pists seemed to recognize this; they changed their
intervention style towards the end of the session
dramatically by focussing strongly on the patient’s
better functioning aspects. This clearly indicates
that some patients have resources the therapist can
relate to, and some have not. Furthermore, the
unsuccessful therapists proved at the end of
the session that they had the potential to activate
the patient’s resources. Yet they differed distinctly
from the successful therapists who focussed right
at the beginning of the session markedly on what
worked well for the patient. The invert U-shaped
curve of therapist resource-activating interven-
tions in successful sessions also demonstrates that
these therapists were doing more than just sup-
unsuccessful sessions successful sessions
Figure 7. Cross-correlations between therapist’s problem-activating interventions and therapist’s resource-
activation interventions in unsuccessful (left-hand side) and successful sessions (right-hand side). Positive lags show
therapist’s problem-activating interventions as leading variable, negative lags show therapist’s resource-activating
interventions as leading variable
10 D. Gassmann and K. Grawe
Copyright © 2006 John Wiley & Sons, Ltd. Clin. Psychol. Psychother. 13, 1–11 (2006)
DOI: 10.1002/cpp
porting the patient’s ego. They created an environ-
ment in which the patient felt he was perceived as
a well functioning person. As soon as this was
established, productive work on the patient’s prob-
lems was more likely. Successful therapists also
turned the focus of intervention away from the
patient’s problems in time. They did not let
patients leave the session with aroused emotions
but rather with even higher activated resources
than patients experienced when they entered the
session.
We also found differences in the immediate
interaction between therapist and patient. In
unsuccessful interactions, therapists often did not
respond with reinforcement to the patient’s acti-
vated resources. They let these situations pass and
continued their work on the patient’s problems.
When therapists then addressed the patient’s
resources, there was often a lack of response on the
part of the patient. This must have been quite dis-
couraging for the therapists, which may explain
their rare use of resource-activating interventions.
On the other hand, if these therapists had been
more dedicated to activating the patient’s
resources and let the problems aside for a moment,
a more productive interaction could have resulted.
It seems that the prerequisite for sucessful work on
a patient’s problems is an atmosphere of activated
resources. The therapist can influence this not only
by establishing a good therapeutic bond, but also
by focussing more explicitly on the healthy parts
of the patient’s personality.
REFERENCES
Bachelor, A., & Horvath, A. (1999). The therapeutic rela-
tionship. In M.A. Hubble, B.L. Duncan, & S.D. Miller
(Eds), The heart and soul of change: What works in therapy
(pp. 133–178). Washington, DC: American Psychologi-
cal Association.
Barlow, D.H. (1993). Clinical handbook of psychological dis-
orders: A step-by-step treatment manual (2nd ed.). New
York: Guilford.
Beutler, L.E., Booker, K., & Peerson, S. (1998). Experien-
tal Treatments: Humanistic, Client-Centred and
Gestalt Approaches. In P. Salkovskis (Ed.), Comprehen-
sive Clinical Psychology (Vol. 6, pp. 163–182). Oxford:
Pergamon.
Beutler, L.E., Malik, M., Alimohamed, S., Harwood, T.M.,
Talebi, H., Noble, S., & Wong, E. (2004). Therapist
Variables. In M.J. Lambert (Ed.), Bergin and Garfield’s
Handbook of Psychotherapy and Behavior Change (5th ed.,
pp 227–306). New York: Wiley.
Blum, H.P. (2003). Psychoanalytic Controversies: Repres-
sion, Transference and Reconstruction. International
Journal of Psychoanalysis, 84(3), 497–513.
Bohart, A., & Tallman, K. (1999). How clients make therapy
work: The process of active self-healing. Washington, DC:
American Psychological Association.
Bond, F.W., & Bunce, D. (2000). Mediators of change in
emotion-focused and problem-focused worksite stress
managment interventions. Journal of Occupational
Health Psychology, 5(1), 156–163.
Borkovec, T.D., & Costello, E. (1993). Efficacy of applied
relaxation and cognitive–behavioral therapy in the
treatment of generalized anxiety disorder. Journal of
Consulting and Clinical Psychology, 61(4), 611–619.
Caspar, F. (1995). Plan Analysis. Toward optimizing psy-
chotherapy. Seattle, WA: Hogrefe-Huber.
Craske, M.G. (1999). Anxiety Disorders. Psychological
Approaches to Theory and Treatment. Boulder, CO:
Westview.
Derogatis, L.R. (1994). The Symptom Checklist-90-R:
Administration, Scoring and Procedures Manual. Balti-
more: John Hopkins School of Medicine. Clinical
Psychometrics Unit.
de Shazer, S. (1988). Clues: Investigating solutions in brief
therapy. New York: Norton.
Emmelkamp, P.M. (2004). Behavior therapy with adults.
In M.J. Lambert (Ed.), Bergin and Garfield’s Handbook
of Psychotherapy and Behavior Change (5th ed., pp
393–446). New York: Wiley.
Franke, G.H. (1995). SCL-90-R. Die Symptom-Checkliste
von Derogatis. Deutsche Version. [SCL-90-R. The
Symptom Checklist of Derogatis. German Version]. Wein-
heim: Beltz.
Gassmann, D. (2002). Korrektive Erfahrungen im Psy-
chotherapieprozess. Entwicklung und Anwendung der
Konsistenztheoretischen Mikroprozessanalyse KMP
[Corrective Experiences in the Psychotherapeutic Process.
Development and Application of the Consistency Theory
Micro Process Analysis CMP], Unveröffentlichte
Dissertation, Universität Bern.
Gassmann, D., & Grawe, K. (2004). Die Messung intrap-
ersonaler Ressourcen im Therapieprozess: Ergebnisse
der Mikroprozessanalyse KMP [Measuring Intra-
personal Resources in the Therapeutic Process: The
Results of the Micro Process Analysis CMP]. Verhal-
tenstherapie und psychosoziale Praxis, 36(1), 63–72.
Gendlin, E.T. (1981). Focusing. New York: Bantam.
Grawe, K. (1982). Der Veränderungsprozessbogen.
Ein Instrument zur deskriptiven Einzelfallanalyse
psychotherapeutisch induzierter Veränderung-
sprozesse [Psychotherapeutic Changes Schedule. An
Instrument for Descriptive Single case Analysis of
Psychotherapeutically Induced Change Processes]. In
M. Zielke (Ed.), Diagnostik in der Psychotherapie (pp.
231–252).
Grawe, K. (1997). Research-informed psychotherapy.
Psychotherapy Research, 7, 1–19.
Grawe, K. (2002). VLB Veränderungsfragebogen für
Lebensbereiche [Changes in Life Domains Question-
naire]. In E. Brähler, J. Schumacher, & B. Strauss (Eds),
Diagnostische Verfahren in der Psychotherapie (pp.
378–381). Göttingen: Hogrefe.
Grawe, K. (2004). Psychological Therapy. Bern: Hogrefe.
Grawe, K., Donati, R., & Bernauer, F. (1994). Psychothera-
pie im Wandel. Von der Konfession zur Profession
Problem and Resource Activation in Therapeutic Interactions 11
Copyright © 2006 John Wiley & Sons, Ltd. Clin. Psychol. Psychother. 13, 1–11 (2006)
DOI: 10.1002/cpp
[The change of psychotherapy. From confession to profes-
sion]. Göttingen: Hogrefe.
Grawe, K., & Grawe-Gerber, M. (1999). Ressourcenak-
tivierung. Ein primäres Wirkprinzip der Psychothera-
pie [Resource activation: A primary change principle
in psychotherapy]. Psychotherapeut, 44(2), 63–73.
Greenberg, L. (2002). Emotion-focused therapy: Coaching
clients to work through their feelings. Washington, DC:
American Psychological Association.
Grob, A. (1995). Subjective well-being and significant life-
events across life-span. Swiss Journal of Psychology,
54(1), 3–18.
Hill, C.E., & Lambert, M.J. (2004). Methodological Issues
in Studying Psychotherapy Processes and Outcomes.
In M.J. Lambert (Ed.), Bergin and Garfield’s Handbook of
Psychotherapy and Behavior Change (5th ed., pp 84–135).
New York: Wiley.
Horowitz, L.M., Rosenberg, S.E., Bauer, B.A., Ureno, G.,
& Villasenor, V.S. (1988). Inventory of Interpersonal
Problems: Psychometric Properties and Clinical Appli-
cations. Journal of Clinical and Consulting Psychology, 56,
885–892.
Horowitz, L.M., Strauss, B., & Kordy, H. (1994). Inventar
Interpersonaler Probleme: Manual [Inventory of Interper-
sonal Problems: Manual]. Weinheim: Beltz.
Hubble, M.A., Duncan, B.L., & Miller, S.D. (1999). Intro-
duction. In M.A. Hubble, B.L. Duncan, & S.D. Miller
(Eds), The heart and soul of change: What works in therapy.
pp. 1–19. Washington, DC: American Psychological
Association.
Kiresuk, T., & Lund, S. (1979). Goal attainment scaling:
Research, evaluation and utilization. In H. Schulen-
berg, & F. Baker (Eds), Program evaluation in health fields
(Vol. 2). New York: Human Science.
Kiresuk, T., & Sherman, R. (1968). Goal attainment
scaling: A general method for evaluating comprehen-
sive community mental health programs. Community
Mental Health Journal, 4, 443–453.
Lambert, M.J., & Bergin, A.E. (1994). The effectiveness of
psychotherapy. In A.E. Bergin, & M. Lambert (Eds),
Handbook of Psychotherapy and Behaviour Change (4th
ed., pp. 143–189). New York: Wiley.
Lambert, M.J., & Ogles, B.M. (2004). The efficacy and
effectiveness of psychotherapy. In M.J. Lambert (Ed.),
Bergin and Garfield’s Handbook of Psychotherapy and
Behavior Change (5th ed., pp 139–193). New York: Wiley.
Orlinsky, D.E., Grawe, K., & Parks, B. (1994). Process and
Outcome in Psychotherapy—noch einmal. In A.E.
Bergin, & S.L. Garfield (Eds.), Handbook of Psychother-
apy and Behaviour Change (4th ed.). New York: Wiley.
Smith, J.D. (2003). Transference, Triangles and Trajecto-
ries. Psychodynamic Practice: Individuals, Groups and
Organisations,9(4), 439–462.
Stiles, W.B., Agnew-Davies, R., Hardy, G.E., Barkham,
M., & Shapiro, D.A. (1998). Relations of the alliance
with psychotherapy outcome: Findings in the second
Sheffield Psychotherapy Project. Journal of Consulting
and Clinical Psychology, 66(5), 791–802.
Vaughan, K., Armstrong, M.S., Gold, R., O’Connor, N.,
Jenneke, W., & Tarrier, N. (1994). A trial of eye-
movement desensitization compared to image habi-
tuation training and applied muscle relaxation in
posttraumatic stress disorder. Journal of Behaviour
Therapy and Experimental Psychiatry, 25(4), 283–291.
Zielke, M. (1978). Veränderungsfragebogen des Erlebens und
Verhaltens (VEV) [Emotional and Behavioral Changes in
Psychotherapy Questionnaire]. Weinheim: Beltz Test.
