Tackling depressive disorder and anxiety: A working memory intervention as addition to treatment as usual

First we will tell you about depression, then we will focus on anxiety and
finally both disorders will be linked in the part about working memory training:

Depression:
The life of individuals with a depressive disorder changes significantly. A
depression changes the way people feel and how they perceive themselves as well
as the world around them. According to the World Health Organisation (WHO),
around 12% of the population is suffering from clinical depression, making it
among the most prevalent psychiatric disorders. The WHO estimates that this
number is increasing, and that by the year 2020, depression will be the most
prevalent disorder causing disability for all ages in men and women. In
addition to the distress depression causes to individuals and their families,
this emotional disorder also incurs extensively direct and indirect economic
costs, which for instance in the Netherlands exceeds one billion Euros annually
and in the United States of America 65 billion dollars. Clearly, more research
is needed to increase the understanding of the causes and maintenance of this
disorder, and to enhance prevention and treatment. Therefore the current
proposal is aimed at examining crucial underpinnings of depression and also
aspires to yield a novel approach of clinical treatment by targeting these
deficits. Apart from important neurobiological research examining the onset and
maintenance of depression, a dominant focus in the past 30 years has been on
cognitive models of depression. These posit that selective information
processing plays a crucial role in the development and maintenance of this
disorder (for a review, see Williams, Watts, MacLeod, & Mathews, 1988, 1997).
That is, how people think, make inferences, approach certain situations, attend
to certain events, and how they recall these events determine their emotional
responses and, as a consequence, whether or not they are likely to incur a
depression. Clearly, cognitive processes play a crucial role in how much people
are affected by negative experiences and determine whether these events will be
followed by quick recovery or by recurring depressive episodes. These models,
therefore, make the important assumption that investigating the content of
cognition and the nature of cognitive processes in depression is essential for
our understanding of the onset and maintenance of this disorder. The extensive
research programs generated by these cognitive models have shown that depressed
individuals are characterised by preferential processing of negative material,
difficulties in disengaging attention from negative information, interpreting
ambiguous information in a negative way and recalling events in a more negative
and more general fashion than they originally were (Mathews & MacLeod, 2005).
Recently, new procedures (i.e., cognitive bias modification; CBM) have been
developed and studied to manipulate these biases and the first steps have now
been made to experimentally employ these CBM procedures for improving cognitive
deficits in depression. For instance, Watkins, Baeyens, and Read (2009)
administered a concreteness training that successfully overcame the
depression-related cognitive bias to process self-relevant information in an
overgeneralised manner. In a similar vein, Holmes and colleagues demonstrated
that modifying maladaptive interpretations reduces depressive intrusions (e.g.,
Holmes, Lang, & Shah, 2009). An important concept in understanding these
dysfunctional cognitive processes is working memory. Working memory is commonly
described as a system for the active maintenance and manipulation of
information in memory and for the control of attention (Baddeley & Hitch,
1974). The capacity of this system is limited; therefore it is important that
its contents are updated efficiently, which is controlled by executive
processes (e.g., Friedman & Miyake, 2004). Executive processes direct the
access to working memory, by removing information that is no longer relevant,
as well as protecting it from intrusions. If these processes perform poorly,
cognitive and emotional functioning are likely to be affected. For example,
poor interference resolution may lead to more intrusive thoughts. In fact,
increased interference from irr



























elevant intrusions has been suggested as a source of low working memory
capacity (Geraerts, Merckelbach, Jelicic, & Habets, 2007). Irrelevant negative
intrusions are an important characteristic of depression. Indeed, such
deficient executive functioning has been linked to depression (Joormann, 2010).
Emerging evidence now shows that depression is characterized by difficulties in
the inhibition of mood-congruent material, resulting in prolonged processing of
negative, goal-irrelevant aspects of presented information. This in turn
hinders recovery from negative mood and leads to sustained negative affect,
which is typical for depressive episodes. Accordingly, theorists have suggested
that deficits in executive functioning lie at the heart of biases in attention,
interpretation, and memory in depression. They are said to lead to ruminative
responses to negative events and, consequently, negative mood states. Indeed, a
study by Joormann and Gotlib (2008) has shown that interference control was
decreased in depressive patients. This means that they experienced difficulty
from removing irrelevant material from working memory. Noticeably, this
increased interference was linked to rumination, one of the hallmark symptoms
of depression. This association was still evident after a 6-month period
(Zetsche & Joormann, in press). Similarly, Goeleven, de Raedt, Baert, and
Koster (2006) found that depressed patients showed strongly impaired inhibition
of negative affect. These findings of executive deficits in depression have
been backed up by neuroscientific work, which indicated abnormalities in neural
function underlying difficulties in inhibition of negative thoughts in
depressed individuals (Koster, De Lissnyder, Derakshan, & De Raedt, in press).
One wonders whether such executive deficits can be trained in the same manner
as those deficits targeted in cognitive bias modification procedures. Is it
possible to improve executive processes, which then in turn influence
higher-order cognitive abilities and even overt behaviour?

Anxiety:
An anxiety disorder consists of an ongoing and severe kind of anxiety without
the presence of a realistic threat. There are
different kinds of anxiety disorders, for example panic disorders, agoraphobia,
social phobia, generalized anxiety disorders
(GAD) and obsessive compulsive disorders (OCD). Together with mood disorders
and substance use disorders, anxiety
disorders are among the most prevalent mental disorders (Brysbaert, 2006). The
World Health Organisation (WHO; 2010)
estimates that around 12% of the population suffers from clinical anxiety every
year. The life of people with an anxiety disorder
changes significantly. Symptoms of anxiety are frequently associated with a
variety of physical symptoms like sweating, heart
palpitations and trembling. Anxiety causes a great deal of distress to the
patients and to their families. In addition, anxiety
disorders cause significant economic costs. For example, these disorders were
costing the Netherlands 285,6 million euros in
2005 (Van Wieren, Schoemaker, & Van Balkom, 2010). Treatment strategies for
anxiety disorders include cognitive therapy,
cognitive-behavioral therapy, psychopharmacology, exposure therapy, relaxation
training, biofeedback, meditation, supportive
psychotherapy, psychodynamic psychotherapy, and other forms of psychotherapy
(Miller, Fletcher, & Kabat-Zinn, 1995). The
most common treatment strategies these days are cognitive-behavioral therapy
and psychopharmacology. Past research
indicates that both genetics and important events in a person*s life are
playing a significant role in the development of anxiety
disorders. However, because there is still a lot unknown about the aetiology
and maintenance of these disorders, more research
is needed to address these issues and to enhance prevention and treatment.
Therefore the current proposal aims to examine
important cognitive processes involved in anxiety and aspires to set up a new
approach to a clinical treatment method targeting
these processes.
Over the past three decades cognitive models of anxiety disorders have
demonstrated that selective information processing
plays an important role in the development and maintenance of anxiety
(Williams, Watts, MacLeod, & Mattews, 1988). More
specifically, recent studies provide considerable evidence to state that
anxiety is strongly associated with an attentional bias
towards threatening stimuli and biases in interpretation and memory (Mathews &
MacLeod, 1994; Mathews & MacLeod, 2005).
Patients with an anxiety disorder tend to interpret ambiguous information in a
negative way. Several researches demonstrate that
individuals reporting high levels of anxiety display a disproportionate ability
to identify or detect emotionally negative words (e.g.
Foa & McNally 1986; as described in Mathews & MacLeod, 1994). For example,
during Stroop tasks, anxious individuals display
problems ignoring the emotionally negative content of threat-related stimulus
words (e.g. Mathews & MacLeod 1985; as
described in Mathews & MacLeod, 1994). However, the nature of the relation
between anxiety and cognition is far from clear.
Therefore, it is necessary to get more insight in the cognitive processes
behind anxiety.
A lot of past research shows that high levels of anxiety are associated with a
reduced ability to perform complex cognitive tasks
(Mueller 1992, Watts & Cooper 1989; as described in Mathews & MacLeod, 1994). A
lot of researchers state that these
reductions are being caused by a depletion of capacity-limited cognitive
resources, especially working memory (Eysenck & Calvo
1992, Ellis & Ashbrook 1988; as described in Mathews & MacLeod, 1994).
Therefore, working memory is an important concept
in understanding the cognitive biases associated with anxiety disorders.
Working memory can be described as a limited capacity
system for the temporary, active maintenance and storage of information
(Baddeley, 2003). This system is critical for human
thought processes. The ability to retain and manipulate information in working
memory is linked with the prefrontal cortex (Fuster,
1989; Goldman-Rakic, 1987; as described in Klingberg, Forssberg, & Westerberg,
2002) and underlies different executive
functions, such as problem solving and reasoning (Engle, Kane, & Tuholski,
1999; Hulme & Roodenrys, 1995; Klingberg, 2000;
as described in Klingberg et al., 2002). The theoretical concept of working
memory argues that working memory is important for
human thought processes because it provides an interface between perception,
long-term memory and action (Andrade, 2001;
Miyake & Shah, 1999; Conway, Jarrold, Kane, Miyake, & Towse, 2007; as described
in Klingberg et al., 2002). Reduced working
memory capacity is associated with several neurological and psychiatric
disorders like schizophrenia and ADHD (Goldman-
Rakic, 1994; Castellanos & Tannock, 2002; as described in McNab et al., 2009).

Working memory training:
Seminal work by Klingberg and colleagues has demonstrated that training the
working memory is possible. These researchers showed that training of working
memory in both children and adults improved their executive functioning and
higher-order abilities such as reasoning (Klingberg, Forssberg, & Westerberg,
2002). This improvement was related with changes in cortical activity (McNab et
al., 2009). Interestingly, in a sample of children with attention
deficit/hyperactivity disorder (ADHD) a working memory training improved
executive functioning but also led to a significant reduction in the severity
of ADHD symptoms (Klingberg et al., 2005). Likewise, Jaeggi, Buschkuehl,
Jonides, and Perriq (2008) showed that a working memory training improved
participants* reasoning and problem solving skills. Recent studies showed that
even in schizophrenia (Subramaniam et al., 2012) and in people with alcohol
problems (Houben, Nederkoorn, Wiers & Jansen, 2011) the complaints decreased
after training the working memory. Noticeably, these different lines of
research all point towards one conclusion: individual differences in the
ability to control the contents of working memory may be related to the onset
and maintenance of depressive disorder. Improving working memory abilities
could therefore tackle what may be at the root of depression.

Results from previous studies from our research team (not yet published) showed
that people who already did cognitive behavioral therapy showed the best
improvement when executing a working memory training. Cognitive behavioral
therapy has good results but this can be increased when combined with another
therapy. We also aim to increase the effectivity of the other treaments of PsyQ
for anxiety and depression complaints. As all therapies share the cognitive
component with cognitive behavioural therapy we think the change is high that
the addition of a working memory training will indeed increase the effecitivty.

Study whether a working memory training as addition to treatment as usual
decreases the depression and anxiety complaints.

128 patients in double blind RCT in two groups: working memory training or a
bogus working memory training as addition to their treatment as usual.

Pre-test:
• SCID-I: depression and anxiety modules
• BDI-II: depression
• RRS: Rumination (standard, frequently used test for measuring rumination)
• STAI: anxiety state and trait
• AMT: Autobiographical Memory Test (standard, frequently used to test the
specificity of autobiographical memory)

Working memory training (and half of the patients bogus training):
4 weeks, six times a week 25 minutes on the computer
N-back task and symmetry span task: both reliable, valid and often used working
memory tasks. The tasks in the working memory training condition will be
adapted to the level of the patient to train the working memory. Because of the
adaption it is never too hard or easy for the patient. The same tasks will be
executed in the bogus training, but the level will not adapt and will stay at a
really easy level. The training can be executed via a website whenever and
where the patient like to do it.


Post-test (after 4 weeks):
• SCID-I: depression and anxiety modules
• BDI-II: depression
• RRS: Rumination (standard, frequently used test for measuring rumination)
• STAI: anxiety state and trait
• AMT: Autobiographical Memory Test (standard, frequently used to test the
specificity of autobiographical memory)

Trainingtasks which have to be executed by the patient six times a week for 4
weeks:

- Symmetry Span Task: The symmetry span task consists of images, presented on a
computer screen, that have to be judged by the participant on symmetry. After
the judgment a stimulus will be presented ( a red square in a grid of 4x4),
from which the participant has to remember the location. The participant get
the possibility to practice the judgment of symmetry in 15 pictures. After
that, they practice the square part (2 times 2 squares and 2 times 3 squares).
As last practice phase they can practice the combination of these two parts (3
times 2 symmetry pictures and squares). The experiment consists of 17 trials in
which the participant has to remember 2 to 17 squares, they start with 2
squares. If they remember the squares correctly and judge the symmetry pictures
correctly as well, the level will increase (3 squares). When they make a
mistake on one or both parts they will stay on the same level. The computer
programme randomly pick the pictures and squares. The symmetry span is a task
in the same category as the reading- or digit span and is a good method to
train the working memory.

The symmetry span is the same for the control group, however they stay on the
same level with only 2 symmetry pictures and squares. Because the task is easy
and does not adapt, the working memory will not be trained.

- N-back task:

In the N-back task the participant sees a black computer screen with a white
cross. On the screen a blue square can light up in 8 different positions. The
participant sees the square for 1 second and on the same time a letter will be
heard (letters: F, H, C, L, K, Q, S, T). The location of the square and the
letter has to be remembered by the participant. On every presentation the
square and letter has to be compared with the square and letter a certain steps
before. The subject presses the a-key if the square is the same, the l-key if
the letter is the same, and both keys if both stimuli are the same. The subject
get a visual instruction before starting the real experiment. They begin with
the 1-back task where they have to compare the letter and square with the
stimuli 1 step back. The pre- and post task consists of 10 trials. If the
subject makes less than 3 errors on the trial (in both parts) the level will
increase with one; the distance between the to remember stimuli and showed
stimuli will get bigger (1-back, 2 back, maximum of 5-back). If they make more
than 5 errors they will go to a lower level. The training consists of 13
trials. The computer randomly generate the squares and letters. The N-back task
is a frequently used test for training the working memory and studying it
(Jeaggi, Studer-Luethi, Buschkeuhl, Su, Jonides & Perrig; 2010). Tje control
group does the same task but they execute the 0-back; they have to compare the
present stimuli with the first stimuli. Because the task is easy and does not
adapt to the level of the subject, the working memory is not trained.

The task will cost the participant one and a half hour at the pre- and
post-test and weeks 6 times a week 25 minutes. They can execute the training
whenever and where they want. There are no risks linked with this study.