Biological Underpinnings of the cognitive Model of …...Biological Underpinnings of depression 517...

Psychoanalytic Review, 99(4), August 2012 © 2012 N.P.A.P.

Biological Underpinnings of the cognitive Model of depression:

a prototype for psychoanalytic research

aaron t. Beck emily a. p. haigh

Kari f. Baber

The phenomena we deal with do not belong only to psychology; they have also an organic and biological aspect, and accordingly in the course of our efforts at building up psychoanalysis we have also made important biological discoveries and have not been able to avoid framing new biological hypotheses.

—Sigmund Freud, An Outline of Psychoanalysis

Decades before writing An Outline of Psychoanalysis, Freud’s Project for a Scientific Psychology attempted to define the unconscious in neurological terms. Due to the technological restraints of the time, Freud was forced to abandon the Project and turn his focus to investigating the unconscious though dream analysis (Roffman & Gerber, 2008). In modern times, biological discoveries contin-ue to inform and support hypotheses about the development and amelioration of psychological distress. The body of research iden-tifying biological support for the cognitive theory of depression may serve as an exemplar for furthering the study and practice of psychoanalysis.

siMilarities Between cognitive theory and psychoanalysis

The cognitive theory and psychoanalytic approaches are con-cerned with similar phenomena. Both approaches involve explo-ration of content that is private, nonmaterial, and nonspatial (Beck, 2004). The individual’s introspective report of physically

516 BecK et al.

intangible phenomena provides the basis for intervention in both cognitive theory and psychoanalytic approaches. Likewise, chang-es in an individual’s symptoms, level of distress, and/or function-ing are theorized to result from changes in this private content.

In addition, both cognitive theory and psychoanalytic ap-proaches are concerned with the importance assigned to the meaning of events. Both acknowledge that meanings enhance the richness of our experiences as well as contribute to our personal and interpersonal problems. Both approaches suggest that mean-ing exists at several levels and is arranged hierarchically. Cogni-tive theory and classical psychoanalytic theory differ in the degree to which the deeper—and more crucial—levels are accessible. In cognitive theory, the deeper cognitive levels are generally not ex-plicit (Alford & Beck, 1997) but may be made conscious through skillful questioning. In psychoanalysis, the aversive meanings are presumed to be ego-alien and thus are sealed off by repression or other defense mechanisms. In cognitive theory, conscious mean-ing may be derived from basic beliefs embedded in cognitive structures (known as schemas). Thus, a person may become very upset by a presumed slight (“She is putting me down”) because the interaction activates a lower-level belief (e.g., “I am undesir-able”). At the lowest level the beliefs are overly general, abstract, and dysfunctional. In severe psychopathology (e.g., clinical de-pression), the absolute negative beliefs are preemptive, persis-tent, and conscious (Beck, 1967). In the psychoanalytic approach, the explanation for the individual’s distress might be that it re-sults from an intolerable repressed meaning or an ego-alien emo-tion/drive, such as unconscious rage.

In addition to similarities in their content, both cognitive theory and psychoanalysis postulate dual systems of cognitive pro-cessing. The basic cognitive processes in cognitive theory and psychoanalysis have been labeled primal process and primary process, respectively (Beck, 1999). These automatic and resource-sparing processes are presumed to account for emotions and urg-es. The secondary processing in both theories is reflective and re-source demanding, and is designed to correct or inhibit primary processes. Improved functioning, as described by psychoanalysis,

Biological Underpinnings of depression 517

results from the resolution of conflict between primary and sec-ondary processes. Improved functioning from a cognitive theory perspective is due to addressing dysfunctional patterns of primary and/or secondary processing. The underlying similarities be-tween cognitive theory and psychoanalysis suggest that the meth-ods by which biological underpinnings of the cognitive model of depression have been identified may be similarly valuable for identifying biological correlates of psychoanalytic theory.

cognitive Model of depression

The prototype cognitive model of depression was based on obser-vations of depressed patients in psychotherapy. The model was based on a developmental perspective that presupposed a consti-tutional vulnerability that sensitized the patient to negative expe-riences. It was proposed that later aversive experiences analogous to those in childhood would precipitate symptoms of depression. Early research by our group indicated that severely depressed pa-tients who had lost a parent in childhood and subsequently expe-rienced a significant loss (breakup of relationship, divorce, or be-reavement) would experience depression (Beck, Sethi, & Tuthill, 1963). We theorized that the earlier experience of severe loss bi-ased the child’s view of subsequent negative experiences, leading to the formation of a negative memory schema. When this schema was activated by a significant loss later in life, it formed a system-atic negative bias toward the patient’s view of experiences and ex-pectancies referred to as the negative cognitive triad (Beck, 1967). Thus the patient was very critical of the self, and considered the past filled with failures and the future as hopeless. In addition to distorting the patient’s interpretations, the systematic negative bias was found to impair other cognitive processes such as atten-tion and memory, whereby the depressed patient would selective-ly attend to and recall negative information. Ultimately, the sys-tematic negative bias produces a feedback loop that initiates and maintains an episode of depression (see Figure 1).

The original cognitive model of depression could account for many of the symptoms of depression, such as loss of interest,

518 BecK et al.

loss of motivation, withdrawal, and in many cases suicidal tenden-cies; however, two important pieces of the model were missing. The model could not account for the fact that only a subset of in-dividuals who had traumatic childhood experiences followed by later trauma actually developed depression. Also the model did not account for “sickness behaviors” such as fatigue, loss of appe-tite, sleep disturbances, and loss of libido. Recent biologically based depression research has helped fill the gaps in the cognitive model, and efforts to synthesize research across disciplines have yielded a more comprehensive understanding of depression and its treatment (Disner, Beevers, Haigh, & Beck, 2011).

The following section will outline the biological underpin-nings of the cognitive model of depression (see Figure 2 for a graphical representation). In addition, we will discuss how gaps in the cognitive model have been filled by recent research, specifi-cally genetic studies and investigations of the inflammatory re-sponse in depression.

figure 1. cognitive model of depression. note that the environmental trigger activates a latent depressogenic schema. the activated schema biases cognitive processing that leads to the development of depressive symptoms. a positive feedback loop is initiated, which contributes to the development and maintenance of a depressive episode.


Biological correlates of the cognitive Model

Dual Processing

Several lines of research suggest that cognitive biases in de-pression are related to maladaptive biological “bottom-up pro-cesses” that are analogous to primal process in cognitive theory. Bottom-up processes are patterns of activation starting in subcor-tical brain regions that are lower along the cognitive hierarchy and are associated with quick, low-level, affective analyses of stim-

figure 2. visual representation of the cognitive model of depression with the associated biological correlates. 5-httlpr, serotonin-transporter-linked polymorphic region; acc, anterior cingulate cortex; Mpfc, me-dial prefrontal cortex; vlpfc, ventral lateral prefrontal cortex; dlpfc, dorsal lateral prefrontal cortex; pfc, prefrontal cortex; na, nucleus ac-cumbens.

520 BecK et al.

uli. Brain regions associated with bottom-up processing include the amygdala, thalamus, nucleus accumbens (NA), hippocampus, caudate, putamen, and the anterior cingulate cortex (ACC). In contrast, “top-down” processes, which are analogous to secondary processing in cognitive theory, refer to higher-level cognitive pro-cesses, such as cognitive appraisal, that draw upon stored knowl-edge. Brain regions associated with top-down processing include the prefrontal cortex (PFC) medial PFC (MPFC), ventral lateral PFC (VLPFC), dorsal lateral PFC (DLPFC), rostral ACC, and the superior parietal cortex.

Bottom-up processes proceed sequentially to connected cor-tical areas higher up and are generally perpetuated by attenuated cognitive control. In this sense, top-down processes associated with regions higher on the cognitive hierarchy fail to effectively regulate activity in those lower regions; for a detailed review see Disner, Beevers, Haigh, and Beck (2011). Broadly stated, growing research suggests that the cognitive model of depression can be understood biologically as the failure of top-down cortical pro-cesses to dampen overactive bottom-up limbic processes (Disner et al., 2011; Mayberg, 1997, 2003; Phillips, Drevets, Rauch, & Lane, 2003a, 2003b).

Negative Self-referential Schemas

The cognitive model presupposes that negative self-referen-tial schematic activation sets in motion the negative cognitive bias characteristic of depression, and there is preliminary evidence for the biological underpinnings of depressogenic schemas. Research has shown that when depressed individuals complete negative self-referential tasks (e.g., judging whether a trait adjective is rep-resentative of oneself), performance is associated with increased activation in the amygdala, ACC and MPFC, and with depression severity (Craik et al.,1999; Fossati et al., 2003; Gusnard, Akbudak, Shulman, & Raichle, 2001; Kelley et al.,, 2002). The amygdala is implicated in emotionality and emotional processing, while the MPFC is a region thought to be related to internal representation of the self. The ACC serves two roles: The ventral ACC influences whether incoming stimuli is labeled with emotional valence, and


the rostral ACC influences the extent to which the stimuli is la-beled as self-referential. Hyperactivation of the amygdala, ACC, and MPFC likely facilitates and maintains a person’s negative self-referential thoughts (Disner et al., 2011). Another line of research has established a link between serotonin, pessimism (a specific self-referential schema), and depression by demonstrating that depressed participants who endorsed marked pessimism evi-denced decreased serotonin expression (Meyer, 2007; Meyer et al., 2003). In summary, self-referential schemas are related to ac-tivity in the amygdala, which detects emotion (bottom-up process-ing); in the intermediate ACC regions involved in coding emo-tional stimuli; and in the MPFC, a top-down region involved in self-referential thought. Finally, alteration in serotonin expres-sion is associated with the presence of dysfunctional attitudes or schemas.

Negatively Biased Attention for Emotional Stimuli

The cognitive model predicts that activated depressogenic schemas lead to information processing biases in attention. As such, depressed individuals selectively attend to (Gotlib, Krasno-perova, Yue, & Joormann, 2004) and have difficulty disengaging from negative stimuli (Hasler, Drevets, Manji, & Charney, 2004). The inability to disengage from negative stimuli is thought to ex-acerbate symptoms of dysphoria and perpetuate the positive feed-back loop of depressive symptoms. Research suggests that im-paired attentional disengagement from negative stimuli is associated with decreased activity in the right VLPFC, DLPFC, and right superior parietal cortex for depressed individuals com-pared to healthy controls (Beevers, Clasen, Stice, & Schnyer, 2010). One factor that may contribute to inability to disengage is deficient inhibition or the impaired ability to inhibit attention for negative stimuli. Inhibitory processing in healthy individuals is as-sociated with the rostral ACC. Depressed individuals show greater activation in the rostral ACC when successfully inhibiting atten-tion to negative stimuli as compared to controls (Elliott, Rubinsz-tein, Sahakian, & Dolan, 2002; Eugene, Joorman, Cooney, Atlas, & Gotlib, 2010; Mitterschiffthaler et al. 2008). These lines of re-

522 BecK et al.

search suggest that negatively biased attention and inhibitory def-icits are associated with cortical regions associated with top-down processing (VLPFC, DLPFC, and right superior parietal cortex, rostral ACC).

Negatively Biased Processing of Emotional Stimuli

A central element of the cognitive model is the putative role of negatively biased processing of internal and external emotion-al stimuli. Several lines of research have identified neural regions associated with negatively biased processing of emotional stimuli. Research has found that when depressed individuals process neg-ative stimuli, they demonstrate more intense (by up to seventy percent) and longer lasting (up to three times as long) amygdala reactivity than healthy controls, even when an emotional task is immediately followed by a nonemotional task (Drevets, 2001; Siegle, Steinhauer, Thase, Stenger, & Carter, 2002). Genetic fac-tors have been implicated in the biased processing of negative stimuli. For example, recent research has shown that hyperactivity of the amygdala in the short 5-HTTLPR variant carriers is associ-ated with increased sensitivity to negative stimuli (Munafò, Brown, & Hariri, 2008).

Research on biased processing of positive stimuli has re-vealed a bias against attending to positive stimuli as reflected by difficulty maintaining activation of the reward circuitry (Heller et al., 2009). Research has found decreased activity in the NA and PFC among depressed subjects asked to consciously maintain or enhance positive emotion, which suggests an impaired capacity to maintain positive affect through top-down control.

In addition to providing evidence for positive and negative emotional processing biases, research has revealed that these bi-ases can be identified at various levels of awareness. In particular, a recent study investigated amygdala response to sad, happy, and neutral faces presented below the level of conscious awareness in medicated participants with major depressive disorder (MDD) and healthy controls (Suslow et al., 2010). Depressed participants exhibited potentiated right amygdala reactivity to masked nega-tive stimuli along with a reduced responsiveness to masked posi-


tive stimuli compared to healthy controls (Suslow et al., 2010). The work by Suslow and colleagues (2010) was replicated and findings were extended to unmedicated depressed and remitted depressed participants (Victor, Furey, Fromm, Ohman, & Dre-vets, 2010). In addition, Victor and colleagues (2010) included a longitudinal component and reexamined cognitive processing bi-ases following eight weeks of antidepressant treatment. Results showed that both the currently and remitted depressed partici-pants had greater amygdala activity than healthy controls when processing masked sad versus happy faces. The healthy control group demonstrated the reverse pattern for happy and neutral faces. Following antidepressant treatment for the MDD group, re-sults showed that the negative bias resolved and a positive bias developed (Victor et. al., 2010). These data suggest that cognitive processing biases are evident at an automatic or early processing level.

A related line of research examining the neuropharmacolog-ical action of antidepressants has provided biological support for the primary role of cognitive processing biases in the cognitive theory of depression. For example, a recent study found differen-tial responses to a cognitive processing task following the single administration of an antidepressant or placebo among unmedi-cated depressed participants and healthy controls (Harmer et al., 2009). The medicated depressed participants demonstrated in-creased recognition of positive facial expressions, increased pro-cessing speed for positive self-relevant personality traits and in-creased memory for positive information compared to healthy volunteers receiving placebo (Harmer et al., 2009). In another study, participants being treated with either a selective serotonin reuptake inhibitor (SSRI) antidepressant or a noradrenaline reuptake inhibitor (NaRI) antidepressant were examined in the context of a facial recognition task (Tranter et al.,2009). Results showed that increased recognition of happy faces after two weeks of antidepressant use preceded changes in emotions and other symptoms. These results suggest that antidepressant drug use may influence cognitive processing before changes in mood and symp-toms are reported, which is compatible with cognitive theories of depression (Harmer et al., 2009).

524 BecK et al.

In sum, research examining the biological underpinnings of biased emotional processing has implicated both bottom-up (in-creased amygdala activity, hypoactive NA) and top-down (DLPFC and PFC hypoactivity) processes. Research also has revealed that processing biases exist below the level of awareness and can be resolved following antidepressant treatment. Finally, research demonstrating that antidepressant administration leads to chang-es in positive cognitive processing prior to changes in emotion and other depressive symptoms has provided support for primacy of cognitive processing in depression.

Biased Memory for Negative Stimuli

An important component of the cognitive model of depres-sion is biased memory or the extent to which negative stimuli are encoded and recalled. Research has identified brain regions im-plicated in biased memory in depressed individuals. In particular, the amygdala facilitates encoding and retrieval of emotional stim-uli in healthy individuals by influencing activity in the hippocam-pus and in the caudate and putamen regions (Adolphs, Cahill, Schul, & Babinsky, 1997). In depressed participants, hyperactivity in the amygdala has been associated with enhanced encoding and recall for negative as compared to positive stimuli (Hamilton & Gotlib, 2008). The amygdala also has been implicated in biased memory through bottom-up influence (Hamilton, & Gotlib, 2008). Another line of research examining recall of happy and sad events has found a differential association with the MPFC, a region associated with abstract representations of reward (Elliott, Friston, & Dolan, 2000; Knutson, Fong, Adams, Varner, & Hom-mer, 2001). Finally, compared to healthy controls, depressed par-ticipants exhibit hyperactivity in the MPFC during recall of self-relevant happy events and hypoactivity during recall of self-relevant negative events (Keedwell, Andrew, Williams, Brammer, & Phil-lips, 2005). In sum, both bottom up (increased amygdala) and top-down processing (alterations in MPFC activity) have been as-sociated with biases in encoding and recall of events.

While the cognitive model has been successful in accounting for the mood component of depression, the model has struggled


to account for two important aspects of depression. First, only a subset of individuals who initially experience a stressor and pre-sumably develop depressogenic schemas actually go on to develop depression after experiencing a stressor that activates latent sche-ma. Second, the model does not account for “sickness behaviors” such as fatigue, loss of appetite, sleep disturbances, and loss of li-bido. These gaps in the model have been filled by recent research, specifically genetic research and investigations of the inflamma-tory response in depression.

Cognitive Vulnerability to Depression

Whether someone who experiences a negative life event goes on to develop depression as the result of activated depressogenic schemas may be due in part to specific biological factors. A num-ber of genetic variants have been associated with the likelihood of developing depression. The most widely researched genetic vari-ance has been the 5-HTTLPR short allele (or serotonin transport-er gene variation), which was found to interact with life experi-ences to produce depression (Caspi et al., 2003; Karg, Burmeister, Shedden, & Sen, 2010).

Sickness Behaviors in Depression

A body of immunological research developed over the past thirty years has provided a key to understanding the origin of fa-tigue, loss of appetite, sleep disturbance, and loss of libido in de-pression. It is now well-established that elevated levels of inflam-matory cytokines have been associated with these physiological and behavioral symptoms, termed “sickness behaviors” (Kent, Bluthé, Kelley, & Dantzer, 1992). Moreover, the current body of research on inflammatory cytokines supports the cognitive mod-el’s proposition that the onset of depressive symptoms, including sickness behaviors, is catalyzed by an aversive event experienced by a vulnerable individual. Specifically, psychosocial stress such as one might experience following an aversive event has been shown to activate the inflammatory response (Miller, Maletic, & Raison,

526 BecK et al.

2009). Individuals’ responses to social rejection appear to be par-ticularly illustrative.

Life events associated with rejection confer significantly greater risk for depression than other life events (Kendler, Hettema, Butera, Gardner, & Prescott, 2003), and rejection also precipitates an immune response and subsequent depressive symptoms (Slavich, O’Donovan, Epel, & Kemeny, 2010). In the first study of its kind, Slavich and his colleagues demonstrated not only that healthy individuals who completed a socially threaten-ing task in the laboratory demonstrated increased levels of inflam-matory activity, which is consistent with previous research (see Dickerson, Kemeny, Aziz, Kim, & Fahey, 2004), but also that greater increases in inflammatory activity were associated with greater activation of neural regions implicated in the processing of rejection-related negative affect (Slavich, Way, Eisenberger, & Taylor, 2010). Subsequently, Slavich and his colleagues proposed that differences in individuals’ neural processing of social rejec-tion may influence the extent to which they demonstrate an in-flammatory response and “sickness behaviors” following rejection (Slavich, O’Donovan, et al., 2010).

Also consistent with the cognitive model, certain psychologi-cal, physiological, and social factors appear to render some indi-viduals particularly vulnerable to depressive symptoms associated with cytokine activity. For example, Capuron and her colleagues (Capuron et al., 2002; Capuron, Ravaud, Miller, & Dantzer, 2004) reported that baseline levels of sadness predicted the severity of depressive symptoms following one month of cytokine treatment for cancer. Physiological factors that appear to impact susceptibil-ity to cytokine-induced depressed mood include sleep disturbance and increased responsiveness of the HPA system (Capuron et al., 2004). Social factors including socioeconomic status (SES) and social support also appear to increase vulnerability to sickness be-haviors. Yirmiya et al. (2000) reported that low SES conferred vul-nerability to more severe depressed mood, poor concentration, and social problems in response to acute infection and the ac-companying elevated levels of pro-inflammatory cytokines, and Capuron et al. (2004) found that low social support predicted greater severity of depressed mood following cytokine treatment.


These findings appear consistent with the cognitive model of de-pression, in that specific psychological, physiological, and social factors may confer vulnerability to depression via immune system pathways.

Although much of the research on inflammatory cytokines and depression supports a unidirectional relationship (i.e., that cytokines induce sickness behaviors in depression), the relation between cytokines and depression may turn out to be more com-plex than a unidirectional hypothesis suggests. In a recent study, Thornton and her colleagues (Thornton, Andersen, Schuler, & Carson, 2009) found that cancer patients who demonstrated im-provement in depressive symptoms following psychological inter-vention exhibited reductions in inflammatory markers and that these reductions appeared to be mediated by change in depres-sive symptoms. Further research examining the mechanisms by which psychological interventions impact the relation of inflam-matory activation and depressive symptoms is warranted and may provide additional support for the cognitive model of depression.

In sum, significant advances have been made in terms of in-dentifying the biological correlates of the cognitive model of de-pression (see Table 1 for an overview). Furthermore, biologically based discoveries have helped to inform and refine our under-standing of the cognitive model of depression. It follows that re-search has also focused on identifying the biological correlates of cognitive therapy for depression. The following section will re-view the biological correlates of cognitive therapy and associated mechanisms.

Biological correlates of cognitive therapy for depression

Kandel (1998) proposed, “Insofar as psychotherapy or counseling is effective and produces long term changes in behavior, it pre-sumably does so through learning, by producing changes in gene expression that alter the strength of synaptic connections and structural changes that alter the anatomical pattern of intercon-nections between nerve cells of the brain” (p. 460). This principle is evident in the substantial body of research that identifies bio-

528 BecK et al.

TABLE 1Overview of Biological Correlates of Cognitive Model of Depression

Cognitive Model

Associated Neural Findings in Depressed Individuals

Schema activation Performance on negative self-referential tasks associated with increased activation in the amygdala, ACC, MPFCa

Marked pessimism associated with decreased serotonin expressionb

Biased attention Impaired attentional disengagement from negative stimuli is associated with decreased activity in the right VLPFC, DLPFC, and right superior parietal cortexc

Rostral ACC hyper-reactivity associated with successful inhibition of attention to negative stimuli as compared to controlsd

Biased processing Amygdala hyper-reactivity associated with processing negative stimulie

Amygdala hyper-reactivity in short 5-HTTLPR variant carriers associated with increased sensitivity to negative stimulif

Difficulty maintaining or enhancing positive emotion (reward region) as evidenced by decreased activity in NA and PFCg

Amygdala hyper-reactivity to masked negative stimuli and reduced responsiveness to masked positive stimuli compared to healthy controlsh,i; negative bias resolved and a positive bias developed following ADMi

Single dose of ADM associated with increased recognition of positive facial expressions, increased processing speed for positive self-relevant personality traits, and increased memory for positive information compared to placebo controlj

Increased recognition of happy faces after two weeks of ADM preceded changes in emotions and other symptomsk

Biased memory Amygdala hyper-reactivity associated with enhanced encod-ing and recall for negative as compared to positive stimulil

Recall of self-relevant happy and negative events associated with MPFC hyper-reactivity/hyporeactivity, respectivelym

Note. ACC, anterior cingulate cortex; MPFC, medial prefrontal cortex; VLPFC, ventral lat-eral prefrontal cortex; DLPFC, dorsal lateral prefrontal cortex; 5-HTTLPR, serotonin-transporter-linked polymorphic region; NA, nucleus accumbens; PFC, prefrontal cortex; ADM, antidepressant medication.aCraik et al., 1999; Fossati et al., 2003; Gusnard et al., 2001; Kelley et al., 2002. bMeyer, 2007; Meyer et al., 2003. cBeevers et al., 2010. dElliott et al., 2002; Eugene et al., 2010; Mitter-schiffthaler et al., 2008. eDrevets, 2001; Siegle et al., 2002. fMunafò et al., 2008. gHeller et al., 2009. hSuslow et al., 2010. iVictor et. al., 2010. jHarmer et al., 2009. kTranter et al., 2009. lHamilton & Gotlib, 2008. mKeedwell et al., 2005.


logical correlates of change in cognitive therapy, consistent with Kandel’s call for an intellectual framework that synthesizes mod-ern psychiatry with modern biology.

Cognitive therapy is an active, time-limited, structured ap-proach used to treat a variety of mental disorders including de-pression (Beck, Rush, Shaw, & Emery, 1979). Cognitive therapy for depression involves a variety of cognitive and behavioral strat-egies to help the client critically examine the accuracy and help-fulness of negatively biased beliefs about the self, world, and fu-ture. Successful cognitive therapy is associated with a reduction in schema-based information-processing biases.

Relatively recent advances in neuroimaging techniques have allowed researchers to examine the neural mediators of change in cognitive therapy (for a review see Etkin, Pittenger, Polan, & Kandel, 2005; Frewen, Dozois, & Lanius, 2008; Linden, 2006; Por-to et al., 2009). In response to this growing line of research, Clark and Beck (2010) expanded Beck’s cognitive formulation of de-pression and anxiety to include neural mediational processes of cognitive therapy. Briefly stated, this expanded model suggests that depression and anxiety are related to impaired reflective cog-nitive controls (i.e., top-down processes) that fail to inhibit reflex-ive, automatic cognitive processes (i.e., bottom-up processes; Clark & Beck, 2010). The expanded model suggests that top-down and bottom-up processes can be represented on a neural level and are sensitive to cognitive therapy. As previously described, bottom-up processes appear to be associated with limbic process-es while top-down processes are associated with activity in cortical regions (Clark & Beck, 2010).

In a seminal study investigating the neural responses to cog-nitive therapy, Goldapple and colleagues showed that the neural pattern of change associated with recovery from depression with cognitive therapy is consistent with top-down mechanisms theo-rized to mediate cognitive theory response (Goldapple et al., 2004). Further, in a follow-up study, Kennedy and colleagues (2007) found that response to cognitive therapy was related to cortical-limbic connectivity.

Several studies have examined neural activity following cog-nitive therapy by examining group differences in emotional pro-

530 BecK et al.

cessing tasks (Costafreda, Khanna, Mourao-Miranda, & Fu, 2009; Fu et al., 2008; Ritchey, Dolcos, Eddington, Strauman, & Cabeza, 2011; Siegle, Carter, & Thase, 2006). Insofar as the main goal of cognitive therapy is to alter the way individuals process emotion-relevant stimuli, imaging studies that record neural activity in the context of an affective or cognitive task are likely to uncover im-portant neural indicators of cognitive therapy–related change. In one such study, Siegle and colleagues (2006) examined whether pretreatment neural response to emotional stimuli predicted re-sponse to cognitive therapy. The results suggested that cognitive-behavioral therapy CBT is most useful to those who demonstrate increased limbic activity and who cannot engage regulatory struc-tures associated with cortical regions (Siegle et al., 2006). In a re-lated study, Fu and colleagues (2008) examined CBT influences on neural activity during an implicit facial affect processing task. Following CBT, decreased bias toward sad faces in depressed pa-tients was associated with normalization of bottom-up processing. Changes in cortical regions commonly impaired in depression in-creased following CBT and were predictive of treatment response with CBT (Fu et al., 2008). The finding that neural activity in cor-tical regions was predictive of successful response to CBT was rep-licated in a recent prospective study (Costafreda et al., 2009). In another study, Ritchey and colleagues (2011) examined neural activity associated with an emotional processing task in depressed patients before and after CBT. The results suggested that hypoac-tivation of cortical regions and hyper-responsivity of the limbic structures change in the direction of normalization after CBT. These findings are consistent with the view that cognitive therapy treatment response is mediated by its impact on neural regions associated with higher-order cognitive processes.

The studies reviewed here suggest that the neural correlates of depression are responsive to cognitive therapy and may predict treatment outcome. While the pattern of interactions is not fully understood, there is convincing preliminary evidence that cogni-tive therapy may be effective in increasing inhibitory executive control, which in turn interrupts or dampens automatic limbic reactions characteristic of depression.

In addition to evidence suggesting that cognitive therapy,


as a complete treatment modality, is associated with important changes in neural activity, recent research also supports the mechanisms of change proposed by the cognitive model. For ex-ample, a preliminary fMRI study examined the neural correlates of cognitive reappraisal, the emotion-regulation strategy of pur-posefully changing thinking about situations to produce a change in affect (Cristea et al., 2011). Cognitive reappraisal is similar to the cognitive therapy approach to cognitive restruc-turing, in which the individual is encouraged to consider a dis-tressing cognition from alternative perspectives in order to de-velop a more accurate, adaptive view. Results showed that cognitive reappraisal was associated with areas involved in high-er cognitive functions (DLPFC), theory of mind (temporal pole, TP) and self-mentalizing (posterior cingulate cortex, PCC). In addition, amygdala activity was reduced during practice of cog-nitive reappraisal, which the authors speculated might represent a possible neurobiological correlate of down-regulation of nega-tive emotions.

Another mechanism of change similar to cognitive restruc-turing is self-distancing, a type of reflection in which the perspec-tive of a distanced observer is adopted (Ayduk & Kross, 2010; Kross, Davidson, Weber, & Ochsner, 2009). Ayduk & Kross (2010) examined the neural and physiological correlates of self-distanc-ing and reported a relation between higher levels of spontaneous self-distancing and decreased peripheral autonomic nervous sys-tem activity during recall of a distressing event, suggesting that a self-distancing perspective is associated with a more adaptive pro-file of physiological response to stress. This is particularly reveal-ing in light of findings that rumination, which is quite distinct from the cognitive restructuring approach used in cognitive ther-apy, is associated with delayed physiological recovery from stress (Gerin, Davidson, Christenfeld, Goyal, & Schwartz, 2006). Kross and colleagues (2009) also reported that individuals who used a self-distanced perspective when reflecting on a distressing event demonstrated less activity in the subgenual anterior cingulate cor-tex (sgACC), MPFC, and right VLPFC than individuals who ad-opted a self-immersed perspective, in which they were instructed to reflect on the feelings experienced in response to the event.

532 BecK et al.

The self-distancing strategy was also associated with a significantly less self-reported negative affect than the self-immersed stance.

iMplications for research on the psychoanalytic perspective

The preceding sections described how the cognitive model of de-pression has been supported and refined by biological research. The cognitive model has not only served to guide interpretation of biological data, but new biological advances also have helped to expand the cognitive model. It is now possible to consider how depression-related phenomena can be represented both cogni-tively and in biological terms. The integration of biological and psychological research to enhance the cognitive model of depres-sion may provide a useful model for psychoanalytically oriented researchers seeking to define the psychoanalytic model in biologi-cal terms.

Biological findings similar to those presented here also have been cited as support for psychodynamic theory and therapy (e.g., Roffman & Gerber, 2008; Shedler, 2010). This observation sug-gests that psychoanalytically oriented researchers are adopting an approach similar to that initially taken by cognitively oriented re-searchers, namely, using their theoretical orientation to guide in-terpretation of the existing biological data. The extent to which future biological research will provide distinct support for, and enhancement of, psychoanalytic constructs will depend on the ex-tent to which the psychoanalytic constructs measured in relation to biological correlates are clearly and uniquely operationalized.

The strongest support for the psychoanalytic perspective would result from such operationalization at two levels: at the lev-el of construct definition and at the level of measurement. At the first level, it will be important that researchers focus on psycho-logical constructs that are uniquely linked to psychoanalytic theo-ry (such as the Unconscious) rather than constructs (such as un-conscious processing) that are part and parcel of cognitive psychology. Similarly, constructs such as therapeutic alliance, which have been described in the literature as distinct to psycho-dynamic theory/therapy (Blagys & Hilsenroth, 2000), are now


considered inherent to most psychotherapeutic modalities. As such, it will be critical for researchers to carefully isolate con-structs that distinguish psychoanalytic theory from other explana-tions for psychological distress or impairment. At the second lev-el, the psychoanalytic constructs being measured in relation to biological correlates must, themselves, be defined in measurable terms. For example, as described in greater detail by Roffman and Gerber (2008), a behavioral method used by Susan Andersen and her colleagues (e.g., Andersen, Glassman, Chen, & Cole, 1995; Berk & Andersen, 2000) to measure transference has yielded promising findings and is now being employed in neuroimaging studies. The success of Andersen and her colleagues in employing a behavioral method to measure the transference construct pro-vides a model for researchers interested in similarly operational-izing other key constructs of psychoanalytic theory. Clear opera-tionalization of constructs unique to psychoanalytic theory/therapy will be a necessary step toward identifying the type of bio-logical support for psychoanalytic theory that exists for cognitive therapy.

references

Adolphs, R., Cahill, L., Schul, R., & Babinsky, R. (1997). Impaired declara-tive memory for emotional material following bilateral amygdala damage in humans. Learn. Memory, 4(3):291–300.

Alford, B. A., & Beck, A. T. (1997). The integrative power of cognitive therapy. New York: Guilford Press.

Andersen, S. M., Glassman, N. S., Chen, S., & Cole, S. W. (1995). Transfer-ence in social perception: The role of chronic accessibility in significant-other representations. J. Personality Soc. Psychology, 69(1):41–57.

Ayduk, O., & Kross, E. (2010). From a distance: Implications of spontaneous self-distancing for adaptive self-reflection. J. Personality Soc. Psychology, 98:809–829.

Beck, A. T. (1967). Depression: Causes and treatment. Philadelphia: University of Pennsylvania Press.

______ (1999). Prisoners of hate: The cognitive basis of anger, hostility, and violence. New York: Harper.

______ (2004). Cognitive therapy, behavior therapy, psychoanalysis, and phar-macotherapy: A cognitive continuum. In A. Freeman, M. J. Mahoney, P. Devito, & D. Martin, eds., Cognition and psychotherapy (2nd ed., pp. 197–220). New York: Springer.

______ Sethi, B. B., & Tuthill, R. W. (1963). Childhood bereavement and adult depression. Arch. Gen. Psychiatry, 9:295–302.

534 BecK et al.

Beevers, C. G., Clasen, P., Stice, E., & Schnyer, D. (2010). Depression symp-toms and cognitive control of emotion cues: A functional magnetic reso-nance imaging study. Neurosci., 167:97–103.

Berk, M. S., & Andersen, S. M. (2000). The impact of past relationships on in-terpersonal behavior: Behavioral confirmation in the social-cognitive pro-cess of transference. J. Personality Soc. Psychology, 79:546–562.

Capuron, L., Gumnick, J. F., Musselman, D. L., Lawson, D. H., Reemsnyder, A., Nemeroff, C. B., & Miller, A. H. (2002). Neurobehavioral effects of interferon-alpha in cancer patients: Phenomenology and paroxetine re-sponsiveness of symptom dimensions. Neuropsychopharmacology, 26:643–652.

______ Ravaud, A., Miller, A. H., & Dantzer, R. (2004). Baseline mood and psychosocial characteristics of patients developing depressive symptoms during interleukin-2 and/or interferon-alpha cancer therapy. Brain, Be-hav., Immunity, 18:205–213.

Caspi, A., Sugden, K., Moffitt, T. E., Taylor, A., Craig, I. W., Harrington, H., McClay, J., Mill, J., Martin, J., Braithwaite, A., & Poulton, R. (2003). Influence of life stress on depression: Moderation by a polymor-phism in the 5-HTT gene. Science, 301:386–389.

Clark, D. A., & Beck, A. T. (2010). Cognitive theory and therapy of anxiety and depression: Convergence with neurobiological findings. Trends Cognitive Sci., 14:418–424.

Costafreda, S. G., Khanna, A., Mourao-Miranda, J., & Fu, C. H. Y. (2009). Neural correlates of sad faces predict clinical remission to cognitive behav-ioural therapy in depression. NeuroReport, 20:637–641.

Craik, F. I. M., Moroz, T. M., Moscovitch, M., Stuss, D. T., Winocur, G., Tulving, E., & Kapur, S. (1999). In search of the self: A positron emission tomography study. Psycholog. Sci., 10(1):26–34.

Cristea, A., Gentili, C., Ricciardi, E., Rota, G., Bonino, D., David, D., Pie-trini, P., & Guazzelli, M. (2011, June). Brain correlates of rational and irra-tional beliefs: An fMRI study of cognitive reappraisal. Poster session presented at annual meeting of the Organization for Human Brain Mapping, Qué-bec, Canada.

Dickerson, S. S., Kemeny, M. E., Aziz, N., Kim, K. H., & Fahey, J. L. (2004). Im-munological effects of induced shame and guilt. Psychosom. Med., 66:124–131.

Disner, S. G., Beevers, C. G., Haigh, E. A. P., & Beck, A. T. (2011). Neural mechanisms of the cognitive model of depression. Nature Reviews Neurosci., 12(8):467–477.

Drevets, W. C. (2001). Neuroimaging and neuropathological studies of depres-sion: Implications for the cognitive-emotional features of mood disorders. Current Opin. Neurobiology, 11:240–249.

Elliott, R., Friston, K. J., & Dolan, R. J. (2000). Dissociable neural responses in human reward systems. J. Neurosci., 20(16):6159–6165.

______ Rubinsztein, J. S., Sahakian, B. J., & Dolan, R. J. (2002). The neural basis of mood-congruent processing biases in depression. Arch. Gen. Psy-chiatry, 59:597–604.

Etkin, A., Pittenger, C., Polan, H. J., & Kandel, E. R. (2005). Toward a neu-robiology of psychotherapy: Basic science and clinical applications. J. Neu-ropsychiatry Clin. Neurosci., 17(2):145–158.

Eugene, F., Joorman, J., Cooney, R. E., Atlas, L. Y., & Gotlib, I. H. (2010).


Neural correlates of inhibitory deficits in depression. Psychiatry Res.: Neuro-imaging, 181:30–35.

Fossati, P., Hevenor, S. J., Graham, S. J., Grady, C., Keightley, M. L., Craik, F., & Mayberg, H. (2003). In search of the emotional self: An fMRI study using positive and negative emotional words. Amer. J. Psychiatry, 160:1938–1945.

Freud, S. (1940). An outline of psychoanalysis (J. Strachey, trans.). New York: Nor-ton, 1949.

Frewen, P. A., Dozois, D. J. A., & Lanius, R. A. (2008). Neuroimaging studies of psychological interventions for mood and anxiety disorders: Empirical and methodological review. Clinical Psychol. Review, 28(2):228–246.

Fu, C. H. Y., Williams, S. C. R., Cleare, A. J., Scott, J., Mitterschiffthaler, M. T., Walsh, N. D., Donaldson, C., Suckling, J., Andrew, C., Steiner, H., & Murray, R. M. (2008). Neural responses to sad facial expressions in major depression following cognitive behavioral therapy. Biolog. Psychiatry, 64(6):505–512.

Gerin, W., Davidson, K. W., Christenfeld, N. J. S., Goyal, T., & Schwartz, J. E. (2006). The role of angry rumination and distraction in blood pres-sure recovery from emotional arousal. Psychosom. Med., 68:64–72.

Goldapple, K., Segal, Z., Garson, C., Lau, M., Bieling, P., Kennedy, S., & Mayberg, H. (2004). Modulation of cortical-limbic pathways in major de-pression: Treatment-specific effects of cognitive behavior therapy. Arch. Gen. Psychiatry, 61:34–41.

Gotlib, I. H., Krasnoperova, E., Yue, D. N., & Joormann, J. (2004). Atten-tional biases for negative interpersonal stimuli in clinical depression. J. Ab-normal Psychology, 113:127-135.

Gusnard, D. A., Akbudak, E., Shulman, G. L., & Raichle, M. E. (2001). Medial prefrontal cortex and self-referential mental activity: Relation to a default mode of brain function. Proceed. Nat. Acad. Sci. USA, 98(7):4259–4264.

Hamilton, J. P., & Gotlib, I. H. (2008). Neural substrates of increased memory sensitivity for negative stimuli in major depression. Biolog. Psychiatry, 63: 1155–1162.

Harmer, C. J., O’Sullivan, U., Favaron, E., Massey-Chase, R., Ayres, R., Reinecke, A., Goodwin, G. M., & Cowen, P. J. (2009). Effect of acute anti-depressant administration on negative affective bias in depressed patients. Amer. J. Psychiatry, 166:1178–1184.

Hasler, G., Drevets, W. C., Manji, H. K., & Charney, D. S. (2004). Discover-ing endophenotypes for major depression. Neuropsychopharmacology, 29(10): 1765–1781.

Heller, A. S., Johnstone, T., Shackman, A. J., Light, S. N., Peterson, M. J., Kolden, G. G., et al. (2009). Reduced capacity to sustain positive emotion in major depression reflects diminished maintenance of fronto-striatal brain activation. Proceedings National Acad. Sci., 106(52):22445–22450.

Kandel, E. R. (1998). A new intellectual framework for psychiatry. Amer. J. Psy-chiatry, 155:457–469.

Karg, K., Burmeister, M., Shedden, K., & Sen, S. The serotonin transporter promoter variant (5-HTTLPR), stress, and depression meta-analysis revisit-ed: Evidence of genetic moderation. Archives of General Psychiatry, 68(5):444.

Keedwell, P. A., Andrew, C., Williams, S. C. R., Brammer, M. J., & Phillips, M. L. (2005). A double dissociation of ventromedial prefrontal cortical re-

536 BecK et al.

sponses to sad and happy stimuli in depressed and healthy individuals. Bio-log. Psychiatry, 58(6):495–503.

Kelley, W. M., Macrae, C. N., Wyland, C. L., Caglar, S., Inati, S., & Heath-erton, T. F. (2002). Finding the self? An event-related fMRI study. J. Cogni-tive Neurosci., 14(5):785–794.

Kendler, K. S., Hettema, J. M., Butera, F., Gardner, C. O., & Prescott, C. A. (2003). Life event dimensions of loss, humiliation, entrapment and dan-ger in the prediction of onsets of major depression and generalized anxi-ety. Arch. Gen. Psychiatry, 60:789–796.

Kent, S., Bluthé, R. M., Kelley, K. W., & Dantzer, R. (1992). Sickness behav-ior as a new target for drug development. Trends Pharmacolog. Sci., 13:24–28.

Knutson, B., Fong, G. W., Adams, C. M., Varner, J. L., & Hommer, D. (2001). Dissociation of reward anticipation and outcome with event-related fMRI. NeuroReport, 12(17):3683–3687.

Kross, E., Davidson, M., Weber, J., & Ochsner, K. (2009). Coping with emo-tions past: The neural bases of regulating affect associated with negative autobiographical memories. Biolog. Psychiatry, 65:361–366.

Linden, D. (2006). How psychotherapy changes the brain—The contribution of functional neuroimaging. Molecular Psychiatry, 11(6):528–538.

Mayberg, H. (1997). Limbic-cortical dysregulation: A proposed model of de-pression. J. Neuropsychiatry Clin. Neurosci., 9(3):471–481.

______ (2003). Modulating dysfunctional limbic-cortical circuits in depression: Towards development of brain-based algorithms for diagnosis and opti-mised treatment. Brit. Med. Bull., 65(1):193–207.

Meyer, J. H. (2007). Imaging the serotonin transporter during major depressive disorder and antidepressant treatment. J. Psychiatry Neurosci., 32(2):86–102.

______ McMain, S., Kennedy, S. H., Korman, L., Brown, G. M., DaSilva, J. N., Wilson, A. A., Blak, T., Eynan-Harvey, R., Goulding, V. S., Houle, S., & Links, P. (2003). Dysfunctional attitudes and 5-HT2 receptors during de-pression and self-harm. Amer. J. Psychiatry, 160:90–99.

Miller, A. H., Maletic, V., & Raison, C. L. (2009). Inflammation and its dis-contents: The role of cytokines in the pathophysiology of major depres-sion. Biolog. Psychiatry, 65:732–741.

Mitterschiffthaler, M., Williams, S., Walsh, N., Cleare, A., Donaldson, C., Scott, J., & Fu, C.H.Y. (2008). Neural basis of the emotional Stroop interference effect in major depression. Psycholog. Med., 38(2):247–256.

Munafò, M. R., Brown, S. M., & Hariri, A. R. (2008). Serotonin transporter (5-HTTLPR) genotype and amygdala activation: A meta-analysis. Biolog. Psychiatry, 63:852–857.

Phillips, M. L., Drevets, W. C., Rauch, S. L., & Lane, R. (2003a). Neurobiolo-gy of emotion perception I: The neural basis of normal emotion percep-tion. Biolog. Psychiatry, 54:504–514.

______ (2003b). Neurobiology of emotion perception II: Implications for major psychiatric disorders. Biolog. Psychiatry, 54:515–528.

Porto, P. R., Oliveira, L., Mari, J., Volchan, E., Figueira, I., & Ventura, P. (2009). Does cognitive behavioral therapy change the brain? A systematic review of neuroimaging in anxiety disorders. J. Neuropsychiatry Clin. Neuro-sci., 21(2):114–125.

Ritchey, M., Dolcos, F., Eddington, K. M., Strauman, T. J., & Cabeza, R. (2011). Neural correlates of emotional processing in depression: Changes


with cognitive behavioral therapy and predictors of treatment response. J. Psychiatr. Res., 45:577–587.

Roffman, J. L., & Gerber, A. J. (2008). Neural models of psychodynamic con-cepts and treatments: Implications for psychodynamic psychotherapy. In R. Levy & J. S. Ablon, eds., Handbook of evidence-based psychodynamic psycho-therapy: Bridging the gap between science and practice (pp. 305–338). New York: Humana.

Shedler, J. (2010). The efficacy of psychodynamic psychotherapy. Amer. Psychol-ogist, 65:98–109.

Siegle, G. J., Carter, C. S., & Thase, M. E. (2006). Use of fMRI to predict recov-ery from unipolar depression with cognitive behavior therapy. Amer. J. Psy-chiatry, 163:735–738.

______ Steinhauer, S. R., Thase, M. E., Stenger, V. A., & Carter, C. S. (2002). Can’t shake that feeling: Event-related fMRI assessment of sustained amyg-dala activity in response to emotional information in depressed individu-als. Biolog. Psychiatry, 51:693–707.

Slavich, G. M., O’Donovan, A., Epel, E. S., & Kemeny, M. E. (2010). Black sheep get the blues: A psychobiological model of social rejection and de-pression. Neurosci. Biobehavioral Rev., 35:39–45.

______ Way, B. M., Eisenberger, N. I., & Taylor, S. E. (2010). Neural sensitiv-ity to social rejection is associated with inflammatory responses to social stress. Proceed. Nat. Acad. Sci. USA, 107(33):14817–14822.

Suslow, T., Konrad, C., Kugel, H., Rumstadt, D., Zwitserlood, P., Schon-ing, S., Ohrmann, P., Bauer, J., Pyka, M., Kersting, A., Arolt, V., Hein-del, W., & Dannlowski, U. (2010). Automatic mood-congruent amygdala responses to masked facial expressions in major depression. Biolog. Psychia-try, 67(2):155–160.

Thornton, L. M., Andersen, B. L., Schuler, T. A., & Carson, W. E., III. (2009). A psychological intervention reduces inflammatory markers by al-leviating depressive symptoms: Secondary analysis of a randomized con-trolled trial. Psychosom. Med., 71:715–724.

Tranter, R., Bell, D., Gutting, P., Harmer, C., Healy, D., & Anderson, I. M. (2009). The effect of serotonergic and noradrenergic antidepressants on face emotion processing in depressed patients. J. Affective Disorders, 118(1–3):87–93.

Victor, T. A., Furey, M. L., Fromm, S. J., Ohman, A., & Drevets, W. C. (2010). Relationship between amygdala responses to masked faces and mood state and treatment in major depressive disorder. Arch. Gen. Psychiatry, 67(11): 1128–1138.

Yirmiya, R., Pollak, Y., Morag, M., Reichenberg, A., Barak, O., Avitsur, R., Shavit, Y., Ovadia, H., Weidenfeld, J., Morag, A., Newman, A., & Pollmächer, T. (2000). Illness, cytokines and depression. Ann. NY Acad Sci, 917:478–487.

Aaron T. Beck The Psychoanalytic Review Psychopathology Research Center vol. 99, no. 4, august 2012Department of Psychiatry3535 Market St., Room 2031Philadelphia, PA 19104E-mail: [email protected]

Biological Underpinnings of the cognitive Model of …...Biological Underpinnings of depression 517...

Documents

Transcript of Biological Underpinnings of the cognitive Model of …...Biological Underpinnings of depression 517...