Cognitive therapy neural networks are increasingly well known

Cognitive therapy neural networks are changed bottom-up in antidepressant therapy and top-down in cognitive therapy

Hypothetical time course of the changes to amygdala and prefrontal function that are associated with antidepressant medication and cognitive therapy, illustrating major cognitive therapy neural networks

Hypothetical time course of the changes to amygdala and prefrontal function that are associated with antidepressant medication and cognitive therapy.

a | During acute depression, amygdala activity is increased (red) and prefrontal activity is decreased (blue) relative to activity in these regions in healthy individuals.

b | Cognitive therapy (CT) effectively exercises the prefrontal cortex (PFC), yielding increased inhibitory function of this region.

c | Antidepressant medication (ADM) targets amygdala function more directly, decreasing its activity.

d | After ADM or CT, amygdala function is decreased and prefrontal function is increased. The double-headed arrow between the amygdala and the PFC represents the bidirectional homeostatic influences that are believed to operate healthy individuals.[1]

Cognitive therapy neural networks –
– work together (along the black lines) to produce depressed symptoms
– feed back the results (along the gray line) to generate depressed symptoms in the future

Information processing in the cognitive model of depression illustrates cognitive therapy neural networks, showing feedback loops

Information processing in the cognitive model of depression.

  • Activation of depressive self-referential schemas by environmental triggers in a vulnerable individual is both the initial and penultimate element of the cognitive model.
  • The initial activation of a schema triggers biased attention, biased processing and biased memory for emotional internal or external stimuli.
  • As a result, incoming information is filtered so that schema-consistent elements in the environment are over-represented.
  • The resulting presence of depressive symptoms then reinforces the self-referential schema (shown by a grey arrow), which further strengthens the individual’s belief in its depressive elements.
  • This sequence triggers the onset and then maintenance of depressive symptoms.[2]

Untreated cognitive therapy neural networks take negative schema information and fan it out, and add in overgeneral negative information

Cognitive functioning in a healthy individual vs. in a depressed individual illustrates functionality in major cognitive therapy neural networks

Cognitive functioning in a healthy (a) or depressed (b) individual.

  • In a depressed individual, a negative self-schema and an over-general mode of processing concur to automatically prime and activate information that is congruent with the negative self-schema, via a cognitive interlock (resulting in rumination), biased memory and attention.
  • In a healthy individual, a concrete mode of processing counteracts these automatic activations.

Cognitive therapy neural networks information flow (in the diagrams above) maps directly to neural regions (in the pictures below)

Brain networks involved in various cognitive functions of cognitive therapy neural networks

Brain networks involved in
(a) self-referential processes and rumination,
(b) cognitive interlock and mood congruent processing,
(c) episodic buffer,
(d) attention bias,
(e) memory bias,
(f) overgeneral processing.

dmPC: dorsomedial prefrontal cortex,
vmPFC: ventromedial prefrontal cortex,
mPFC: medial prefrontal cortex,
iPFC: inferior prefrontal cortex,
mOFC: medial orbitofrontal cortex,
aOFC: anterior orbitofrontal cortex,
dlPFC: dorsolateral prefrontal cortex,
aITC: anterior inferotemporal cortex,
STG: superior temporal gyrus,
AnG: angular gyrus,
Ins: insula,
ACC: anterior cingulate cortex,
PCC: posterior cingulate cortex,
PCun: precuneus,
Rsp: retrosplenial cortex,
dmTh: dorsomedial thalamus,
HPC: hippocampus,
Amy: amygdala,
Hab: habenula,
Acc: nucleus accumbens,
Cd: caudate,
Pu: putamen,
Re: nucleus reuniens,
DG dentate gyrus of the hippocampus.[3]


  1. DeRubeis, Robert J., Greg J. Siegle, and Steven D. Hollon. “Cognitive therapy versus medication for depression: treatment outcomes and neural mechanisms.” Nature Reviews Neuroscience 9.10 (2008): 788-796.
  2. Disner, Seth G., et al. “Neural mechanisms of the cognitive model of depression.” Nature Reviews Neuroscience 12.8 (2011): 467-477.
  3. Belzung, Catherine, Paul Willner, and Pierre Philippot. “Depression: from psychopathology to pathophysiology.” Current opinion in neurobiology 30 (2015): 24-30.

Cognitive therapy gains can be sudden and large

Cognitive therapy gains can be sudden and can be early in therapy.

Cognitive therapy gains can be sudden, and can spiral upward

Psychotherapy does not always follow a linear path.

Most… research is based on the assumption that treatment progress in psychotherapy is linear, or log-linear, and follows some form of regular dose-response relationship… However, this assumption regarding the macro, or average level of change, does not necessarily hold for a finer grained analysis of individual progress…

The prevalence of sudden gains found in several studies from different research groups varies between 17% and 50%…

…sudden gains are a phenomenon of cognitive behavioral therapies (CBT) and result from CBT-specific techniques. …substantial cognitive changes could be observed in the therapy session preceding the sudden gains. Sudden gains were followed by a better therapeutic alliance and more cognitive changes in the session after the sudden gain. Thus, the authors postulated an upward spiral, i.e., cognitive changes during the pregain sessions foster the therapeutic alliance and eventually result in additional cognitive changes…

In this paper the frequency of sudden gains as well as sudden losses will be investigated in a large naturalistic outpatient sample (n=1500) with repeated measurements of session progress.

Sudden losses are less frequent, and are more-or-less random

…23.4%… of the patients in the sample experienced at least one sudden gain…

…4.5%… of those experienced a sudden gain and a sudden loss…

…18.9%… had only sudden gains…

…5.47%… experienced only sudden losses.

In contrast to sudden gains, sudden losses occur over the course of treatment without a typical peak of occurrence.

Cognitive therapy gains are greater for people who are hurting more

…those patients who experienced no sudden shifts at all and followed a more linear trajectory had the shortest therapies. At intake patients with no sudden shifts tended to be significantly less disturbed on average… than the three groups with sudden shifts… … post-hoc tests revealed that patients with no sudden shifts were less impaired… compared to sudden gain patients…

…patients with both sudden gains and sudden losses stayed in treatment longer than those in the other groups.

Patients with sudden gains… did not rate the therapeutic relationship significantly higher on average than patients with sudden losses… Results further confirm previous findings that a large amount (about 42%) of sudden gains tend to take place early in therapy and they are part of the phenomenon of early change… But the phenomena also occur later in treatment (about 58%). …we found shorter treatments for patients experiencing sudden gains early in treatment. Patients with early sudden gains showed the highest effect sizes… at the end of treatment.[1]


  1. Lutz, Wolfgang, et al. “The ups and downs of psychotherapy: Sudden gains and sudden losses identified with session reports.” Psychotherapy Research 23.1 (2013): 14-24.

Thoughts determine perceptions, feelings, and actions

Thoughts bias memories and attention, and are crucial to psychological disorders.

A 19-year-old college junior, Louis, learned that his girlfriend had committed suicide…

One year after… Louis… feels “on-edge” and “down” most of the time… stopped playing sports… no longer spends much time with his friends who knew his deceased girlfriend… feels tired all the time… sleeps 10 to 12 hours a day… has trouble focusing in school… “constantly snaps” when he’s around people… frequently has upsetting dreams… drinks six to eight alcoholic drinks each night to fall asleep.

… when Louis was 7… his younger sister was killed… he was supposed to be in the same car… it took him a long time to deal with her death.

Louis… feels the worst in the evening… alone… “I’m never going to be happy again,” “There is no point getting close to people,” and “I should have known she was going to kill herself. ”

The clinician noted… biased expectancies…. (… “I’m never going to be happy again”), overgeneralization (… “There is no point getting close to people”), and emotional reasoning (… “I should have known she was going to kill herself”)…

Louis and the clinician collaboratively identified three main goals:
(a) identify, examine, and modify inaccurate or unhelpful thoughts Louis has about his future and others;
(b) examine his thoughts about his perceived role in his girlfriend’s suicide; and
(c) increase his level of social interaction.

With practice, Louis was able to independently modify his thoughts so that they were more accurate and helpful.

The clinician dedicated a significant amount of time to helping Louis examine the evidence for his thoughts… that he should have known his girlfriend was suicidal… and adopt a… thought that his girlfriend’s desire to commit suicide was something he could not have predicted.

Louis agreed to slowly increase his social activity… evaluate his belief that it would be too unbearable to see his friends who had known his girlfriend… Louis reported feeling grateful that he had reconnected with some of his old friends.

Louis… independently designed a behavioral experiment to test his thought that it will take him “forever to fall asleep without drinking.”[1]


  1. Beck, Aaron T., and Emily AP Haigh. “Advances in cognitive theory and therapy: The generic cognitive model.Annual review of clinical psychology 10 (2014): 1-24.