Task 3. How to improve exposure therapy: what about a cognitive approach
and safety behaviors?
Learning goals
What is information processing?
- What is the interpretation/evaluation bias and how can it be measured?
- What is attention bias and how can it be measured?
- What is memory bias and how can it be measured?
- Is there evidence for cognitive biases in anxiety disorders?
- How do these biases maintain the anxiety disorders?
De Jong, P. J. (2014) Information Processing, in The Wiley Handbook of Anxiety Disorders
Introduction
Cognitive models imply that emotional disorders critically depend on the existence of
maladaptive cognitive structures in memory. These schemas are assumed to automatically
influence all stages of individuals’ information processing. The basic assumption of the
information-processing models of anxiety disorders (ADs) is that these processing biases are
not merely symptoms but play a vital role in the maintenance and causation of ADs. It is
assumed that there is a reciprocal relationship between biased information processing and
anxiety symptoms.
Threat-related automatic associations are assumed to reflexively activate a threat-
processing template, which would guide automatically initiated cognitive and emotional
responses on the basis of the initial threat impression. Current dual-process models
emphasize the importance of differentiating between this type of reflexive, associative
process and more deliberate, reflective processes. The reflexive processes may be followed
by a more slow and effortful stage allowing propositional processes to reconsider the
“validity” of the original reflexive responses. This rule-based system may thus help interrupt
the automatically initiated fear responses and may counteract the reflexively initiated
escape and avoidance tendencies in the presence of sufficient cognitive resources.
Transdiagnostic information-processing biases
Attentional bias
Characteristics
If threatening and neutral stimuli occur together, the attention of individuals suffering from
AD will likely be biased toward threat. A feature of threat-related attentional bias (AB) is its
temporal unfolding. Early stages of threat processing are characterized by increased
attention to threat, whereas later stages are characterized by attentional avoidance of
threat.
Underlying mechanisms
Different processes may be involved in the various components of AB, than just the
mechanisms underlying the pattern of an apparently reflexive AB toward threat that is
quickly followed by a bias away from threat. it has been argued that attention is not a
,unitary system, but rather a set of different components that are functionally and
structurally independent, although they may work cooperatively.
Three major components can be differentiated, each of which might be differentially
involved over the course of information processing of threatening stimuli:
- Alerting is involved in maintaining an appropriate sensitivity level to perceive and
process relevant stimuli. It is supported that anxiety patients show an enhanced
sensitivity to perceive and process threatening stimuli. The attentional system would
preferentially select threat over safe/control signals. It has been shown that specifically
in high fearful individuals, phobia-relevant stimuli are frequently detected even when
presented in the attentional blink interval, indicating that phobic stimuli receive
prioritized access to limited cognitive resources. The increased sensitivity for threat
stimuli comes with a cost, as the attentional capture of threat stimuli interfered with
individuals’ ability to identify the targets they were looking for. It may obstruct detecting
safe/helpful/reassuring stimuli that occur in close temporal proximity to the threat
stimuli.
- The orienting network is involved in the selection of relevant information and is
therefore especially relevant for the facilitated engagement toward concern-relevant
stimuli in AD patients. Anxious individuals are typically faster in responding to probes
that replace briefly presented threat, which is evidence for a facilitated spatial
orientation toward threat. the failure to find evidence for facilitated attention might be
explained by the presentation duration of the threat stimuli. Enhanced attention might
be especially relevant at an early stage of stimulus presentation. Thus, anxiety seems
associated with both increased engagement and a difficulty to disengage, yet the
relevance of these processes depends on the processing stage. There is evidence for the
orienting system regarding stimuli with (high) threat value, suggesting that facilitated
attention is reflected in relatively fast responses on valid-threat trials compared to valid-
neutral trials. The difficulty to disengage is reflected in relatively slow responses on
invalid-threat trials compared to invalid-neutral trials. So, facilitated attention would be
most relevant during the early stages of processing, whereas disengagement difficulties
become increasingly important during the later stages of stimulus processing.
- The executive control network is involved in conflict resolution and voluntary action
control. This network is especially relevant with respect to the conflict that might arise
between the motivational relevance of concern-relevant threat stimuli and the urge to
avoid the same stimuli. This conflict might give rise to an initial difficulty in disengaging
attention. During later stages of information processing, this conflict might then be
resolved when additional strategic resources can be used to modulate the more bottom-
up processing of threat-relevant stimuli. Attentional capture and hold at least partly
reflect an inability to inhibit automatic threat alerts from being further processed. The
capacity to voluntarily regulate attention to threat stimuli may vary as a function of the
stimulus’ threat value and the temporarily available cognitive resources. Moreover, the
efficiency of the executive network may vary across individuals. Deficiencies in executive
control will logically enhance individuals’ sensitivity to threat stimuli.
Prognostic value and causal status
If AB plays a critical role in the persistence of symptoms, it should be reduced following
successful treatment. However, evidence is limited. Only one study tested the influence of
cognitive-behavior therapy (CBT) on AB in adult anxiety patients. The study demonstrated
, that the pretreatment tendency to direct attention toward threatening words and angry
faces changed into a bias away from these threat stimuli. This has to be further examined. In
a study, the induced difficulty to disengage from threats was paralleled with increased
anxiety responding.
Bias-induction studies primarily focused on the later stages of AB. It remains therefore to be
demonstrated also whether early AB causally mediates emotional reactivity.
The finding that a bias induction in nonanxious individuals can lower the threshold for
evoking fear responses suggests that a modified visual probe task may also be used to
reduce an already existent AB in anxiety patients. However, there are some failures to
replicate these findings.
Thus far, CBM primarily targeted the later components of AB related to engagement
difficulties. It would be important to explore further whether modifying early AB might
perhaps be even more effective as it might prevent cascading effects. Furthermore, it might
be relevant to test the synergistic effect of combining cognitive bias modification (CBM) with
CBT. Recent research has provided evidence to suggest that the strength as well as the
pattern of pretreatment attentional bias (avoidant vs. attention) moderates the efficacy of
CBT.
Interpretation bias
AD patients tend to show threatening interpretations of ambiguous stimuli relevant to their
concerns. They typically tend to select more often the threatening than the safe/benign
alternative. This interpretation bias (IB) may occur as a consequence of post-event
processing but could also occur during the actual encounter with particular ambiguous
information. A study shows that the absence of positive rather than the presence of a
negative IB seems most critical in SAD. This may depend on the potential threat value of the
ambiguous context. Given the benefits of an early identification of severe threat, it seems
reasonable to assume that even in low anxious individuals information with potentially high
threat value will give rise to negative rather than positive interpretations. In the context of
ambiguous information that may signify severe threats, the presence of relatively strong
negative interpretations instead of the absence of a positive IB may be more strongly
involved in the maintenance of fearful concerns.
Another important issue regarding IB is its time course. High trait anxious individuals get
stuck to a threatening interpretation during the later stages of processing.
Experimentally reducing IB is paralleled with a reduction in anxiety vulnerability. Reducing
negative IB and/or enhancing positive IB reduces anxiety symptoms. This points to the
relevance of using CBM as a clinical tool.
Reasoning bias
Individuals tend to search for belief-confirming information and their habitual reasoning
pattern is biased in a way that confirms rather than falsifies prior beliefs. The reflective
system seems, just as the reflexive system, inclined to sustain rather than to correct faulty
anxiogenic convictions.
Hypothesis testing
The frightening convictions of anxiety patients can be typically condensed into conditional
propositions of the type “If P, then Q”. Testing the logical validity of such conditional beliefs
requires the investigation whether P is indeed followed by Q and a careful check whether
and safety behaviors?
Learning goals
What is information processing?
- What is the interpretation/evaluation bias and how can it be measured?
- What is attention bias and how can it be measured?
- What is memory bias and how can it be measured?
- Is there evidence for cognitive biases in anxiety disorders?
- How do these biases maintain the anxiety disorders?
De Jong, P. J. (2014) Information Processing, in The Wiley Handbook of Anxiety Disorders
Introduction
Cognitive models imply that emotional disorders critically depend on the existence of
maladaptive cognitive structures in memory. These schemas are assumed to automatically
influence all stages of individuals’ information processing. The basic assumption of the
information-processing models of anxiety disorders (ADs) is that these processing biases are
not merely symptoms but play a vital role in the maintenance and causation of ADs. It is
assumed that there is a reciprocal relationship between biased information processing and
anxiety symptoms.
Threat-related automatic associations are assumed to reflexively activate a threat-
processing template, which would guide automatically initiated cognitive and emotional
responses on the basis of the initial threat impression. Current dual-process models
emphasize the importance of differentiating between this type of reflexive, associative
process and more deliberate, reflective processes. The reflexive processes may be followed
by a more slow and effortful stage allowing propositional processes to reconsider the
“validity” of the original reflexive responses. This rule-based system may thus help interrupt
the automatically initiated fear responses and may counteract the reflexively initiated
escape and avoidance tendencies in the presence of sufficient cognitive resources.
Transdiagnostic information-processing biases
Attentional bias
Characteristics
If threatening and neutral stimuli occur together, the attention of individuals suffering from
AD will likely be biased toward threat. A feature of threat-related attentional bias (AB) is its
temporal unfolding. Early stages of threat processing are characterized by increased
attention to threat, whereas later stages are characterized by attentional avoidance of
threat.
Underlying mechanisms
Different processes may be involved in the various components of AB, than just the
mechanisms underlying the pattern of an apparently reflexive AB toward threat that is
quickly followed by a bias away from threat. it has been argued that attention is not a
,unitary system, but rather a set of different components that are functionally and
structurally independent, although they may work cooperatively.
Three major components can be differentiated, each of which might be differentially
involved over the course of information processing of threatening stimuli:
- Alerting is involved in maintaining an appropriate sensitivity level to perceive and
process relevant stimuli. It is supported that anxiety patients show an enhanced
sensitivity to perceive and process threatening stimuli. The attentional system would
preferentially select threat over safe/control signals. It has been shown that specifically
in high fearful individuals, phobia-relevant stimuli are frequently detected even when
presented in the attentional blink interval, indicating that phobic stimuli receive
prioritized access to limited cognitive resources. The increased sensitivity for threat
stimuli comes with a cost, as the attentional capture of threat stimuli interfered with
individuals’ ability to identify the targets they were looking for. It may obstruct detecting
safe/helpful/reassuring stimuli that occur in close temporal proximity to the threat
stimuli.
- The orienting network is involved in the selection of relevant information and is
therefore especially relevant for the facilitated engagement toward concern-relevant
stimuli in AD patients. Anxious individuals are typically faster in responding to probes
that replace briefly presented threat, which is evidence for a facilitated spatial
orientation toward threat. the failure to find evidence for facilitated attention might be
explained by the presentation duration of the threat stimuli. Enhanced attention might
be especially relevant at an early stage of stimulus presentation. Thus, anxiety seems
associated with both increased engagement and a difficulty to disengage, yet the
relevance of these processes depends on the processing stage. There is evidence for the
orienting system regarding stimuli with (high) threat value, suggesting that facilitated
attention is reflected in relatively fast responses on valid-threat trials compared to valid-
neutral trials. The difficulty to disengage is reflected in relatively slow responses on
invalid-threat trials compared to invalid-neutral trials. So, facilitated attention would be
most relevant during the early stages of processing, whereas disengagement difficulties
become increasingly important during the later stages of stimulus processing.
- The executive control network is involved in conflict resolution and voluntary action
control. This network is especially relevant with respect to the conflict that might arise
between the motivational relevance of concern-relevant threat stimuli and the urge to
avoid the same stimuli. This conflict might give rise to an initial difficulty in disengaging
attention. During later stages of information processing, this conflict might then be
resolved when additional strategic resources can be used to modulate the more bottom-
up processing of threat-relevant stimuli. Attentional capture and hold at least partly
reflect an inability to inhibit automatic threat alerts from being further processed. The
capacity to voluntarily regulate attention to threat stimuli may vary as a function of the
stimulus’ threat value and the temporarily available cognitive resources. Moreover, the
efficiency of the executive network may vary across individuals. Deficiencies in executive
control will logically enhance individuals’ sensitivity to threat stimuli.
Prognostic value and causal status
If AB plays a critical role in the persistence of symptoms, it should be reduced following
successful treatment. However, evidence is limited. Only one study tested the influence of
cognitive-behavior therapy (CBT) on AB in adult anxiety patients. The study demonstrated
, that the pretreatment tendency to direct attention toward threatening words and angry
faces changed into a bias away from these threat stimuli. This has to be further examined. In
a study, the induced difficulty to disengage from threats was paralleled with increased
anxiety responding.
Bias-induction studies primarily focused on the later stages of AB. It remains therefore to be
demonstrated also whether early AB causally mediates emotional reactivity.
The finding that a bias induction in nonanxious individuals can lower the threshold for
evoking fear responses suggests that a modified visual probe task may also be used to
reduce an already existent AB in anxiety patients. However, there are some failures to
replicate these findings.
Thus far, CBM primarily targeted the later components of AB related to engagement
difficulties. It would be important to explore further whether modifying early AB might
perhaps be even more effective as it might prevent cascading effects. Furthermore, it might
be relevant to test the synergistic effect of combining cognitive bias modification (CBM) with
CBT. Recent research has provided evidence to suggest that the strength as well as the
pattern of pretreatment attentional bias (avoidant vs. attention) moderates the efficacy of
CBT.
Interpretation bias
AD patients tend to show threatening interpretations of ambiguous stimuli relevant to their
concerns. They typically tend to select more often the threatening than the safe/benign
alternative. This interpretation bias (IB) may occur as a consequence of post-event
processing but could also occur during the actual encounter with particular ambiguous
information. A study shows that the absence of positive rather than the presence of a
negative IB seems most critical in SAD. This may depend on the potential threat value of the
ambiguous context. Given the benefits of an early identification of severe threat, it seems
reasonable to assume that even in low anxious individuals information with potentially high
threat value will give rise to negative rather than positive interpretations. In the context of
ambiguous information that may signify severe threats, the presence of relatively strong
negative interpretations instead of the absence of a positive IB may be more strongly
involved in the maintenance of fearful concerns.
Another important issue regarding IB is its time course. High trait anxious individuals get
stuck to a threatening interpretation during the later stages of processing.
Experimentally reducing IB is paralleled with a reduction in anxiety vulnerability. Reducing
negative IB and/or enhancing positive IB reduces anxiety symptoms. This points to the
relevance of using CBM as a clinical tool.
Reasoning bias
Individuals tend to search for belief-confirming information and their habitual reasoning
pattern is biased in a way that confirms rather than falsifies prior beliefs. The reflective
system seems, just as the reflexive system, inclined to sustain rather than to correct faulty
anxiogenic convictions.
Hypothesis testing
The frightening convictions of anxiety patients can be typically condensed into conditional
propositions of the type “If P, then Q”. Testing the logical validity of such conditional beliefs
requires the investigation whether P is indeed followed by Q and a careful check whether
