competing visions of the implications of...

Journal of Constructivist Psychology, 2008, in press

COMPETING VISIONS OF THE IMPLICATIONSOF NEUROSCIENCE FOR PSYCHOTHERAPY

BRIAN TOOMEY1 AND BRUCE ECKER2

1Clinical Psychology Department, University of Memphis, Memphis, Tennessee2Private practice in psychotherapy, Oakland, California

In this third and final article of a series on the confluence of neurobiology andpsychotherapy, we consider three current, influential interpretations of the implicationsof neuroscience for psychotherapy: pharmacological treatment, reparative attachmenttherapy, and the cognitive regulation of emotion and behavior. We critically examinethese clinical strategies, reviewing efficacy data, neuroscientific research, and the modelof symptom production by coherent implicit memory as articulated in coherencepsychology. We argue that according to current knowledge, (a) each of the three clinicalinterpretations of neuroscience implements only part of the brain’s known capabilitiesfor change, (b) those capabilities are more fully utilized and can yield greater clinicaleffectiveness for the majority of psychotherapy clients through a therapeutic strategy ofselective depotentiation of implicit memory, as epitomized by coherence therapy, and(c) counteracting an implicit memory, whether cognitively or psychopharmacologically,is only moderately effective, is inherently susceptible to relapse, and entails a range ofundesirable collateral effects.

The field of psychotherapy has wrestled sinceits inception with the challenging task ofintegrating an understanding of subjective mentalexperience with an understanding of how the brainand neural networks function. Freud drafted anattempt to formulate a neurologically basedpsychology (Freud, 1895/1950) only three yearsafter the discovery of the synapse was published(Ramon y Cajal, 1892). Since then, the sought-forsynthesis of the physiological and thephenomenological has proven elusive and hasmore often than not degenerated into a polarizeddebate over mind versus brain, holism versusreductionism, top-down versus bottom-up.

The present authors are not reductionists, yetwe fully value and welcome what the neuralviewpoint contributes to an accurate understandingof the functioning of the brain-mind-body system.________________________________________Correspondence: Bruce Ecker, 445 Bellevue Ave.,Suite 101, Oakland, CA, USA 94610.E-mail: [email protected]

The latter is so complex and rich as to allow for awide range of useful approaches for facilitatingchange and growth. The almost dizzying diversityof psychotherapies created in the second half ofthe twentieth century has been a positivedevelopment in many ways. Nevertheless,“complex” does not mean “arbitrary.” The brain-mind-body system consists of certain structures,processes, forms, laws and rules, and not others.Having an accurate knowledge of these couldallow us to discern psychotheraputic methods thatcan be reliably effective from those that cannot,and to this end we believe neuroscience has muchto contribute. The psychotherapy field alreadyowes much to authors such as Schore (2001a,2001b) and Siegel (1999) for calling attention soeffectively to the value of understanding theneurodynamic substrates of symptoms thattherapists work with every day.

In this article we attempt to sort out theimplications for psychotherapy of current know-ledge in neuroscience. In doing so, we know we

Competing Visions2

are adding only a few notes to a more thanhundred-year-long, ongoing conversation.

There are three major current trends inneurodynamically informed mental health treat-ment: pharmacological treatment, reparativeattachment therapy, and the cognitive regulation ofemotion and behavior. As we review and criticallyexamine these approaches, it will emerge that eachcaptures only a part of the implications ofneuroscience research for psychotherapy andleaves out important areas that have significantclinical relevance. We will conclude that thenative, neurodynamic capabilities for therapeuticchange are more fully utilized by a therapy ofimplicit memory depotentiation, as described inthe second of this set of three articles (Ecker andToomey, 2007). As a useful context for ourdiscussion, we begin with a short review of theimplicit memory perspective afforded bycoherence therapy.

THE VIEW THAT FINDINGS INNEUROSCIENCE IMPLY A THERAPY OFIMPLICIT MEMORY DEPOTENTIATION

In the first two articles of this series (Eckerand Toomey, 2007; Toomey and Ecker, 2007) weoffered a detailed, neural and phenomenologicalmodel of a psychotherapy—coherence therapy—that putatively ends symptom production byselectively depotentiating symptom-requiringimplicit memory. We now raise the question: Howgeneral is this form of symptom causation? Whatclinical symptoms and problems in living arecaused by subcortical implicit memory? Howgenerally applicable is a therapy of implicitmemory depotentiation?

We argue below that the great majority ofclinical symptoms are caused by implicit memory.If this proves to be true, then the selectivedepotentiation of implicit memory would logicallymaximize therapeutic efficacy and consistency.

Implicit memory differs qualitatively from thevernacular connotation of the word “memory,”which implies the subjective recall of either pastpersonal experiences (episodic, autobiographicalmemory) or of facts (semantic memory), both ofwhich are stored cortically. In contrast, anactivated implicit emotional memory in the limbicsystem is experienced as an immersion in aparticular emotional state, such as strong anxiety

or anger, with or without thematic or narrativemeaning (such as a nonverbal expectation ofimminent abandonment or a construal of beingused), as well as a compulsion to carry out (or notto carry out) a particular behavior. There is noconscious thought as to why the feeling and thebehavior are occurring, no conscious recall of theexperiences and learnings in the past that createdthis response, and indeed no sense that one isexperiencing a memory at all. Yet, as the caseexamples in the two prior articles have illustrated,full conscious retrieval of the coherent materialgenerating this response usually elicits well-defined, specific knowledge structures as well asexplicit, episodic memory of the original, concretescenes and experiences in which these implicitknowings were formed.

The result of successful implicit memorydepotentiation is the uncreation of the implicitknowings driving symptom production. Neurally,then the synaptic circuits encoding those knowingsis no longer operational. Phenomenologically, theconstituent knowings and constructs then nolonger have an emotionally compelling, subjectiverealness (though they may remain in factual andautobiographical memory, having been translatedinto explicit memory in the course of therapy). Forexample, in the main case example usedthroughout this series, it no longer seemed real toCarol that for her to enjoy marital sex would resultin repeating her mother’s egregious sexualbehavior, and to Tina in our second article (Eckerand Toomey, 2007), it no longer seemed real thatshe is powerless to keep her mother from takingpossession of her interests and pursuits. When thesymptom-requiring constructs are depotentiated,symptom production ceases permanently becauseits very basis and cause is eliminated permanently.

A therapy of implicit memory depotentiationconsists, by definition, of locating, accessing anddepotentiating the specific, unconscious personalconstructs that require production of the presentingsymptom. Coherence therapy (Ecker and Hulley,1996, 2000a, 2004) is a particular methodologydesigned for efficiently carrying out theseprocesses.

At the time of writing—thirteen years sincecoherence therapy was first presented topsychotherapists—its practitioners have, collect-ively, several hundred years of experience inobserving the effects of its methodology. Anec-

Competing Visions 3

dotal evidence is a notoriously unreliable basis forassessing the effectiveness of a psychotherapy andthe veracity of its model of change, yet a numberof positive indications regarding coherencetherapy, taken together, appear promising andwarrant systematic, rigorous study:

a. Many experienced practitioners of otherwidely used therapies—such as cognitive-behavioral therapy, psychodynamic psychotherapyand existential-humanistic therapies—have de-scribed a decisive enhancement in both theirnumber of therapeutic successes and the degreeand quality of those successes, across a wide rangeof symptoms of mood, thought and behavior.1

b. In agency settings, supervisors’ recognitionof heightened effectiveness appears to be a regularoccurrence. For example, a coherence therapypractitioner who works in a large outpatient clinichas been promoted to core staff because theaverage number of sessions of his clients isdramatically lower than that of all other clinicians.Another practitioner who is an intern in an urbancommunity mental health center was askedrepeatedly by his supervisor to teach her hismethods and subsequently, while still an intern,was offered a post-internship supervisory position,the first such offer to an intern in the long historyof this agency.

c. A partial verification of coherence therapy’smodel of change is built into the methodology inthe form of strong evidence that an implicitmemory has actually been depotentiated: thepreviously vivid, emotionally laden content of thememory (the pro-symptom position) fails to bereactivated by potent reminders and triggers, asdescribed in our previous article (Ecker andToomey, 2007).

d. The methodology embodies a logic thatpotentially also lends support for the model(Ecker, 2006): In the course of coherence therapyproperly executed, nothing whatsoever is done tocounteract, avoid or prevent the symptom or the

1 Symptoms that have been dispelled by coherencetherapy include depression, anxiety, panic, agora-phobia, low self-worth, attachment problems, sequelaeof childhood abuse, sexual problems, food/eating/weight problems, rage, attention deficit, complicatedbereavement, fidgeting, codependency, underachieve-ment, procrastination, and a wide range of inter-personal, couple and family problems.

implicit, pro-symptom constructs maintaining it. Ifa symptom then ceases to occur solely as a resultof depotentiating a specific implicit memory, thecausation of the symptom by that implicit memoryis indicated. The clearest case occurs the casewhen the juxtaposition experience thatdepotentiates a pro-symptom position is delib-erately, overtly orchestrated by the therapist, as inthe case of Tina in our prior article. Up until thatpoint, the work has amply (and verifiably)incorporated all of the nonspecific common factorswidely regarded as the basis of clinicaleffectiveness (see for example Hubble, Duncan,Miller and Hubble, 1999), but the symptompersists, and it is only after a well-defined event,the juxtaposition experience, that symptomcessation immediately occurs. (In order toconclude reliably that this verifies symptomcausation by implicit memory [pro-symptompositions], qualitative analysis is needed toestablish that indeed nothing counteractive hasoccurred and that the usual common factors wereindeed present but did not dispel the symptomapart from the juxtaposition process.)

The foregoing four points constitute thecurrent basis for the plausibility of coherencetherapy. Of particular note, coherence therapy isregularly effective for achieving permanentcessation of longstanding mood problems widelyregarded as requiring neuromodulatory drugs forsuccessful relief. (For detailed case studies ofcoherence therapy for depression, see Ecker andHulley (1996 [pp. 63-90], 2002a), and for casestudies of anxiety and panic see Ecker (2003) andEcker and Hulley (2000b, 2002b).) We take thoseresults to indicate that the cause of the moodproblem was in those cases psychological, notbiochemical. This means that the neural andbiochemical substrates were driven by the person’sphenomenological and emotional material, not theother way around.

Lastly, we suggested in the previous articlethat coherence therapy’s methodology for implicitmemory depotentiation fulfills plausible criteriafor the optimal use of neuroplasticity inpsychotherapy: (a) recruitment of the types ofsynaptic change known to be most potent (b) inthe brain regions causing symptom production, (c)as rapidly as these neuroplasticity mechanisms canwork.

Competing Visions4

From this perspective, next we offer ourcritiques of other, widespread views of neuro-dynamically informed clinical treatments.

THE VIEW THAT FINDINGSIN NEUROSCIENCE IMPLY

PHARMACOLOGICAL TREATMENT

As knowledge of the cellular and molecularcorrelates of mental life grows, inevitably manynew strategies will arise for chemical, bottom-uptreatment of mental, emotional and behavioralsymptoms and sufferings. In this section weexamine the notion that pharmaceutical treatmentis implied by neuroscience. We first consider whatthe drug effectiveness research shows, and thenexamine the role and relevance of psychiatricmedications in treating symptoms that are causedby implicit memory. Our discussion primarilyaddresses the most widely used medications, theSSRIs (selective serotonin reuptake inhibitors).

Letting the drug research speak for itself

Meta-analyses of clinical trials funded by thedrug companies themselves have consistentlyshown that SSRIs are not more effective thanplacebos to a clinically significant degree. In 27 of47 studies used to approve them, Prozac, Paxil,Zoloft, Effexor, Serzone, and Celexa failed tosignificantly outperform placebos (Kirsch andSapirstein, 1998). Reviewing these data Kirsch,Moore, Scoboria and Nicholls (2002) summarize:

Approximately 80% of the response tomedication was duplicated in placebo controlgroups, and the mean difference between drugand placebo was approximately 2 points on the17-item (50-point) and 21-item (62-point)Hamilton Depression Scale. Improvement atthe highest doses of medication was notdifferent from improvement at the lowestdoses. …If drug and placebo effects areadditive, the pharmacological effects ofantidepressants are clinically negligible.

Fischer and Greenberg (1998) conclude fromtheir review of the SSRI research that in studiesshowing a slight superiority of SSRIs overplacebos, the difference can more plausibly be

ascribed to questionable experimental method-ology than to true drug effect.

Likewise, in a more recent meta-analysis,Moncrieff and Kirsch (2005, p. 157) conclude,“recent meta-analyses show selective serotoninreuptake inhibitors have no clinically meaningfuladvantage over placebo” and “methodologicalartifacts may account for the small degree ofsuperiority shown over placebo.”

In our experience, our psychotherapycolleagues meet these clear facts regarding thedrug companies’ own data with great skepticism,describing clients whose depression diminishedsignificantly after using these drugs. In this regard,an account by journalist Roger Greenberg (2003)of a participant in a clinical trial at UCLA forEffexor (venlafaxine) is instructive.

Janet Schonfeld had suffered seriousdepression for over two decades when she readabout a trial for antidepressant medication. She felthopeful and excited about the possibility of a cure,and within a few weeks of enrolling in the studyshe was largely relieved of feelings ofworthlessness and suicidal ideation, which sheviewed as a dramatic improvement. She alsoexperienced nausea, one of the drug’s known sideeffects, leading her and her nurse to assume thatshe was receiving the active drug, not the placebo.However, at the completion of her six-monthparticipation she was alarmed to learn that she hadbeen taking an inert placebo and that herimprovement could not be attributed topharmacological action.

For what were no doubt both personal andsocial reasons, she had been keen to accept thepharmacological narrative stressing biochemicaldeficiency, corrected imbalances and exogenouscures. Amazingly, when her doctors told her thatdespite her great improvement, she would likelybe better off on Effexor, she agreed and took theactual drug for an additional two and a half years.Apparently she was unable to accept that she hadbeen able to shed her symptoms entirelypsychologically, on the basis of perceptions andexpectations.

In drug tests, drug companies often identifyand remove from their data pools subjects whodemonstrate a strong and early response toplacebos and other non-medical treatments. In aprocedure known as a placebo washout strategy,all subjects are placed on a placebo for the first

Competing Visions 5

week or two of a drug trial, and those who show alarge recovery response are removed from the datapool before randomization, which significantlybiases results in favor of drug efficacy. As manyas 20% of participitants can be “washed out” inthis way (Antonuccio, Danton, DeNelsky,Greenberg and Gordon 1999; Kirsch et al., 2002).

Additionally, many doctors and patientscorrectly infer from the absence or presence ofside effects whether they are in the control ortreatment group, thus penetrating the double blindand compromising experimental design. Flawedplacebo design also is indicated by the observationthat when drugs believed not to havean t idepressan t p roper t i e s ( such asmethylphenidate, benzodiazep ines , andantipsychotics) are used as “active” placebos inorder to mimic side effects, the already clinicallyinsignificant superiority of the SSRI shrinksfurther or disappears (Moncrieff and Kirsch,2005).

Despite the widespread clinical notion thatSSRIs are more effective in more acute cases,Moncrieff and Kirsch (2005, p. 157) conclude intheir recent meta-analysis, “claims thatantidepressants are more effective in more severeconditions have little evidence to support them.”

The inability of researchers to clinicallydifferentiate SSRIs from placebos (except withrespect to negative side effects) seriouslyundermines the theory that these drugs reducesymptoms of depression through a targetedreadjustment of neurotransmitter levels.

Drug makers maintain that SSRIs are effectiveat reducing depressive symptomology becausethey correct an imbalance of neurotransmitters.This occurs, it is maintained, by blocking thereuptake of serotonin back into the cell after aneuron has fired, causing serotonin to remainlonger in the synaptic junction between neurons.This results in an increase in overall serotoninlevels, which, it is claimed, in turn alleviatesdepression. There is some evidence that depressedsubjects do show differences in serotoninmetabolism and utilization (Meltzer, 1989) and thepartial blocking of serotonin reuptake by the drugsappears scientifically confirmed (Mann, 1999).

Nonetheless, while these two facts can beinterpreted to support manufacturers’ arguments,the placebo equivalence suggests that the decreasein depression experienced by those taking SSRIs is

largely due to brain changes that occur naturally inresponse to expectancy of healing—the placeboeffect—rather than a true drug effect. (For reviewsof placebo research see Benedetti, Helen,Mayberg, Wager, Stohler and Zubieta, 2005;Vallance, 2006). Indeed, recent evidence suggeststhe placebo responses to treatment for depressionare mediated through changes in serotonerigicfunctioning (de la Fuente-Fernandez and Stoessl,2002) The fact that dosage and blood plasmalevels of SSRIs show minimal or no causalrelation to treatment outcome (Amsterdam et al.,1997; Preskorn, 1997) raise further questions onthe drug effects of SSRIs. Truly effective drugsare expected to show a distinct dose-responserelationship.

Hyman and Nestler (1996) stress that drugs ingeneral and psychiatric drugs in particular do not“correct imbalances,” but rather create them. Theydescribe the process as follows:

Chronic administration of psychotropic drugscreates perturbations in neurotransmitterfunction that likely exceed the strength or timecourse of almost any natural stimulus... Theresult of these types of repeated perturbationsor initiating events is to usurp normalhomeostatic mechanisms within neurons,thereby producing adaptations that lead tosubstantial and long-lasting alterations inneural function. (Hyman and Nestler, 1996, p.154)

They note that in the case of drugs of abuse,the cascade of compensatory adjustments initiatethe well-studied negative biochemical feedbackloops and dependencies that define physicaladdiction. They observe the same initial pattern inthe major classes of prescription psychiatric drugs,which likewise initiate an imbalance or “chronicperturbation,” and they comment that “it is lessclear how compensatory adaptations to a drugwould result in a therapeutic response.”

Adding weight to the addiction hypothesis, anSSRI “discontinuation syndrome” has beenidentified and is under study. Black, Shea, Dursunand Kutcher (2000) report fifty-three symptomsthat can accompany or follow tapereddiscontinuation and “are not due to a…recurrenceof a mental disorder” (p. 255). These symptomscease within 72 hours of restarting the SSRI. In a

Competing Visions6

review, Tamam and Ozpoyraz (2002, p. 17)emphasize (a) many “somatic and psychologicalsymptoms” of discontinuation were “not notedduring short-term efficacy studies,” and (b) thesesymptoms “cannot be explained as aremanifestation of the original disorder.” Theimplication is that SSRIs produce a biochemicaldependency with withdrawal symptoms, inaddition to failing to demonstrate clinicallysignificant antidepressant drug properties.

Further evidence suggests that serotoninreuptake inhibition does not remedy a depression-producing, serotonergic imbalance: (a) A serotoninreuptake accelerator, tianeptine, with precisely theopposite effect on serotonin reuptake than theSSRIs have, was found to be as effective an anti-depressant as Prozac (Lôo, et al., 2001). (b) Ablocker of the neurotransmitter glutamate,ketamine, produced far stronger and far fasterantidepressant effects than do any of the SSRIs,without affecting serotonin levels (Zarate et al.,2006).

In our view, the totality of the data that wehave reviewed above indicates that the therapeuticeffects of SSRIs are due almost entirely to theendogenous placebo effect, possibly enhanced to asmall, clinically insignificant degree by a complexcascade of neural responses constituting anabnormal brain state (as distinct from a correctedimbalance) that has many effects, coincidentallyincluding a reduction of mood symptoms alongwith a large number of other, unwanted (side)effects. (The array of effects of SSRIs has evenbeen shown recently to include enhancedproduction of new neurons in the hippocampus(Encinas, Vaahtokari and Enikolopov, 2006;Santarelli et al., 2003), bringing the hypothesisthat this, rather than the increase of synapticserotonin levels, could be the real antidepressantmechanism.)

Psychiatrist Pedro Delgado summarizes threedecades of research on the neurotransmitterimbalance/deficiency model of depression with thefollowing frank admission: “Intensiveinvestigation has failed to find convincingevidence of a primary dysfunction of a specificmonoamine2 system in patients with majordepressive disorders” (Delgado, 2000, p. 7). He

2 Monoamines are a class of neurotransmitters thatinclude serotonin, dopamine, and norepinephrine.

further summarizes that while experimentallyinduced serotonin depletion causes a drop in moodfor clients currently taking SSRIs, it has no effecton the mood of unmedicated depressives and non-depressed controlled subjects. This indicates fromyet another angle that the serotonin depletionmodel of depression is fallacious.

Thus, there are two major gaps that all of theSSRI research has done nothing to close: first,depressives show no “primary dysfunction” intheir neurotransmitter function, and second, thebiochemical effects of SSRIs do not produceclinically significant anti-depressant drug effects.

The research, including the drug companies’own data, forces us to conclude that SSRI effectsare primarily, and perhaps entirely, due tononpharmacological effects. When inert placebos,other drugs not claimed to have antidepressanteffects, and both the inhibition and theacceleration of serotonin reuptake have all provenroughly equivalent across dozens of wellcontrolled trials, there is a serious problem withthe argument that the neuroscientific data impliesthat depression is caused by a neurotransmitterimbalance that should be corrected exogenouslyby psychiatric drugs.

In contrast to these impressively unimpressivedata we can consider a drug with a well-documented drug effect. Lipitor (atorvastatin), forexample, regulates blood cholesterol through awell-understood biochemical process—theinhibition of a specific enzyme in the liver thatmetabolizes cholesterol—and was found effectivefor achieving target cholesterol levels for 85percent of subjects. This rate is over five times the16 percent efficacy of placebos (P≤0.001) (Harris,Wheeler and Chong, 2002).3

A meta-analysis by Walsh, Seidman, Syskoand Gould (2002) showed that the rate of responseto both placebos and medications grew steadilybetween 1981 and 2000 at the rate of about 7percent per decade. As it is absurd to argue that

3 This efficacy alone does not indicate Lipitor as atreatment of choice. For instance, using a dietaryintervention based on whole plant foods, Jenkins et al.(2003) found no difference between the effects of theirregimen and a drug of the same class as Lipitor. Thedietary intervention was also accompanied by a rangeof positive effects, in contrast to the array of negativeside effects and medical risks attendant to the drug.

Competing Visions 7

sugar pills had steadily improved in quality, webelieve that a constructivist, meaning-basedaccount of the increase is indicated. Lacasse andLeo (2005) present a side-by-side comparisonshowing a striking disparity between thepharmaceutical companies’ advertising rhetoric,on the one hand, and the weakness of the scientificevidence for their claim that depression is causedby a serotonin deficit, on the other.

Psychiatric medications for symptomscaused by implicit memory?

Throughout this three-article series we haveargued that subcortical implicit memory is thecause of most of the symptoms and problemspresented by most people receiving psycho-therapy. This includes depression and anxiety,4 thesymptoms that draw the heaviest use ofmedications, including SSRIs. If indeed the causeof these symptoms lies in implicit memory, howdoes the argument that neuroscience impliespharmacological treatment fare?

If a mood symptom is generated by an implicitmemory, the cause is a functional, not structural,impairment. As noted earlier, an implicit memoryoperates not as an autobiographical, episodicmemory-of-the-past, but as a model of a particularaspect of the world and a knowing of what to feeland how to act in that context (though only thefeeling and the action are conscious, not theknowingness). An implicit memory is amodifiable, functional schema that is created, heldand applied by the normal operation of the mentalsystem.

In a neural network with an encoded implicitmemory that maintains a mood problem, thesynapses and neurons are not damaged ormalfunctioning. For instance, Tina, whose therapyfor depression we studied in our previous article(Ecker and Toomey, 2007), had, in addition to thepro-symptom position (implicit knowing) wedescribed, another that consisted of , expecting tobe sharply attacked verbally and emotionally by a

4 An example of depression caused by implicit memorywas given in the example of Tina in our prior article(Ecker and Toomey, 2007) and is described more fullyin Ecker and Hulley (2002a). For numerous examplesof anxiety and panic caused by implicit memory, seeEcker (2003) and Ecker and Hulley (2002b).

family member if she expresses any knowing orfeeling of her own, so keeping her energy andfeelings deadened (depressed) is urgent in order tobe safe. In the subcortical neural circuits encodingthat coherent schema, all of the neurons andsynapses were working properly. It is the memoryas a whole that maintains a high or low synapticrelease of certain neurotransmitters and theassociated mood or behavior problem, not anydefect of neurons or synapses.

This understanding of the normal functioningof implicit memory can be applied to the findingthat the brain region known as the subgenualcingulate, or Brodmann area 25, is chronicallyhighly active in depressed persons (Mayberg et al.,1999), and that applying ongoing electricalstimulation (“deep brain stimulation”) to thisregion largely eliminates depressive symptoms intwo-thirds of tested subjects (Mayberg et al.,2005). Area 25 is in the paralimbic cortex, anolder part of the cortex that is in close contact withthe limbic system and is an important pathway ofcortico-limbic communication and coordination.Based on our central tenet that depression, likemany other symptoms, is generated by coherentpersonal constructs in implicit memory, weconjecture that what keeps area 25 highlyactivated is not a neurological malfunction, asneuroscientists hypothesize, but rather a chronicactivation of implicit personal constructs. Thisimplies that brain scans of depressed clients incoherence therapy should show a quieting of area25 as soon as their depressogenic implicitconstructs have been revised or depotentiated. Incontrast, neither deep brain stimulation norantidepressant drugs can remove the implicitconstructs that may be driving the hyperactivationof area 25, though they can reduce the depressiveeffects of those constructs.

Perhaps ultimately most important inunderstanding the brain’s response to SSRIs is asystemic view of the hierarchical relationshipbetween the operations of biochemistry, implicitmemory and psychology. Consider the situation ofmost people who take an SSRI to counteractdepression. If it is the case that a person’sdepression is generated by an implicit memory, thedepressed state is emotionally necessary accordingto the content of that subcortical implicit memory(pro-symptom position). Can we assume that theblocking of particular molecular pathways by a

Competing Visions8

drug prevents the implicit memory systems of theamygdala, striatum or brain stem from creating thenecessary state of depression? As a rule, eachneural, synaptic and molecular process in the brainhas several independent, alternate pathways,making the brain a supremely flexible andadaptive system (Denes and Pizzamiglio, 1999;Panksepp, 2004). Subcortical systems that “know”they must create a state of depression in order toavoid a feared attack have many options forgetting the brain to do so, such as reducing thetotal number or sensitivity of serotonin receptors;creating changes in any of the molecular pathwaysof many other types of neurotransmitters; alteringthe strengths and numbers of synapses; oradjusting any of the mood-affecting endocrinehormones in any number of ways.

The same argument can be applied to otherclasses of antidepressants, which includemonoamine oxidase inhibitors, dopamine reuptakeinhibitors, norepinephrine reuptake inhibitors,serotonin-norepinephrine reuptake inhibitors,tricyclics, and tetracyclics. All of these drugstarget a small number of specific elements within adizzyingly complex system involving hundreds ofgenes, proteins, synapses, receptors, transmitters,enzymes and functionally encoded neuralnetworks.

Looking specifically at the complexity of theserotonin system, Carrasco and Sander (2005, p.1428) describe the remarkable diversity ofbiological mechanisms recruited across thefunctioning of different SSRIs, noting that inaddition to modulating serotonin reuptake,"transsynaptic effects such as modulation ofsignaling cascades, gene expression processes andneuroplasticity are also important in themechanism of action of antidepressants." Synapseshave well over a dozen distinct serotonin receptorsubtypes, which respond differently to structurallydistinct SSRIs such as paroxetine (Paxil) andfluoxetine (Prozac) (Olivier, VanOorschot andWaldinger, 1998). Likewise, on synapse surfacesthere are several different enzymes (proteins) thatcarry out serotonin reuptake, and these are alsosubject to different levels of inhibition by differentSSRIs (Preskorn, 1997). Again, given thiscomplexity, it appears that in any attempt tobiochemically blockade a symptom there will beholes through which a pro-symptom position couldfind a way to express itself. To assume that a

pharmaceutical intervention can trump theoperation of coherent implicit memory may bewishful, naïve thinking that greatly underestimatesthe systemic hierarchy of the brain.

Proponents of SSRIs would argue,presumably, that the drugs subdue the brainsystems that implicit memory activates. Studieshave detected a suppression of sympatheticnervous system activity in response to SSRI use(Shores, Pascually, Lewis, Flatness and Veith,2001). However, such suppression is non-selectiveand likely causes the troubling and oftenunacceptable emotional flattening described bymany SSRI users. Conceivably, for some users thedrugs may induce an actual drug effect in dullingemotional responsiveness broadly. However,suppression of the central nervous system by nomeans necessarily translates into an effectivetreatment for depression. Many drugs, includingalcohol and barbiturates, also suppress centralnervous system activity.

The possible role of nutritional deficiency maybe a noteworthy exception to our argumentsagainst the view that psychological symptoms areproduced by biological defects or imbalances. Forexample, there is evidence that depressed personshave decreased blood plasma levels of omega-3fatty acids and that the reduced levels “areassociated with the severity of depression”(Edwards, Peet, Shay and Horrobin, 1998, p. 149.)Omega-3 fatty acids, like certain other micro- andmacronutrients, are an essential building blockneeded by the brain, and it is biochemicallyplausible that a depletion could generate astructural neural deficiency with deleteriouspsychological effects, such as depressed mood.Treatment in the form of orthomolecularsupplementation and dietary adjustments wouldideally occur in tandem with coherence-focuseddepotentation of any implicit memory that isfurther amplifying the mood symptom.

Genes, drugs, and serotonin

In contrast to the functional causation ofsymptoms by implicit memory, neuroscientistshave recently begun to identify true structuraldiseases of the synapse, such as a genetically-caused, synaptic molecular deficiency thatmanifests as fragile X, the most common form ofhereditary mental retardation (Huber, Gallagher,

Competing Visions 9

Stephen, Warren and Bear, 2002). Medicationscreated for such conditions could be designed totruly remedy their causes, rather than merelysuppress symptoms.

The distinction between functional andstructural causation of symptoms can also shedlight on recent research into the powerful role ofneural genes in our emotional life. Genes havebeen identified that synthesize proteins that play acritical role in the creation of specific emotionalstates, such as fear or depression, by controllingsynapse strength in particular subcortical circuits.For example, mice genetically altered to lack thegene stathmin show no fear in situations thatreliably engender innate fear responses in normalmice, and also exhibit decreased memory forconditioned fears (Shumyatsky et al., 2005). Micegenetically altered to lack the gene GRP showed,in contrast, a marked intensification of learnedfear, but no change of instinctive, innate fear(Shumyatsky et al., 2002). Deletion of the geneTREK-1 yields mice that shows clear signs ofbeing resistant to depression and stress (Heurteauxet al., 2006).

What are the implications for psychotherapyof these and similar discoveries of how geneexpression influences emotional states? EricKandel, co-author of the stathmin and GRP studiesjust cited, comments that the results indicatepotential drug targets for instinctive fear andlearned anxiety (HHMI, 2005). He states also,"Since GRP acts to dampen fear, it might bepossible in principle to develop drugs that activatethe [protein that GRP produces], representing acompletely new approach to treating anxiety... [I]nstudies of fear learning we could well have anexcellent beginning for animal models of a severemental illness” (HHMI, 2002). Similar commentshave been offered by authors of the TREK-1study.

Only if the cause of a person's anxiety ordepression happened to be a structural impairmentof stathmin, GRP, TREK-1 or some other geneinvolved in the neural circuitry controlling theparticular mood-state would the effect of a gene-targeting drug be a true corrective to the actualcause of the mood symptom. We contend that thiswould prove to be the case for at most a very smallpercentage of therapy clients.

If, on the other hand, the cause of theindividual's mood disturbance is an implicit

memory, then a gene-targeting drug treatmentcould only serve to suppress the symptom and itsactual cause. Such a switching-off of fear circuitrywould be global, possibly eliminating appropriate,adaptive fear responses that maintain personalsafety. Likewise, switching off the depressioncircuitry might eliminate the capacity to feelappropriate sadness, grief and sorrow, which areintegral to the human experience of love andmeaning.

Moreover, there may be other importantfunctions served by these genes and the proteinsthey synthesize. Even if the drug counteracts theexperience of anxiety or depression, the implicitmemory responsible for the troubled mood wouldcontinue to exist, leaving the person fullysusceptible to relapsing upon discontinued use ofthe drug, or to symptom expression via alternateneurophysiological pathways.

By analogy, if the temperature in one's homeis chronically too high or too low, the appropriateresponse depends on whether the cause is adefective central furnace (corresponding to agenetic impairment) or the setting on thethermostat (a functional problem corresponding toan implicit memory producing, for example,anxiety or depression). One should not tamperwith the furnace if the thermostat is causing theproblem.

Effective, parsimonious treatment ofneurogenetic impairments must also take intoaccount the fact that genetic expression is, as arule, highly regulated by both the externale n v i r o n m e n t a n d t h e i n t e r n a l ,physiological/psychological environment. Genesalone do not determine manifested traits. Genescan be silenced, regulated and flexibly switched onand off through interactions with the environmentand the organism itself (Cirelli, 2005; Lewinton,1980, 2000; Meaney, 2004). As the followingexample shows, research is already finding thatcertain genetic vulnerabilities are expressed aspsychological symptoms only if the neural systemis driven by an implicit memory involvingexpectations that entail anxiety or depression. Insuch cases, the psychotherapeutic clearing of thoseimplicit memories should be the first strategy oftreatment.

Consider the much-discussed short alleleversion of 5-HTT, a gene implicated in impairedutilization of serotonin. In human studies, Hariri et

Competing Visions10

al. (2002) found that subjects with this genedemonstrated increased amygdalar activation inresponse to fearful stimuli, implying that the geneenhances threat appraisal and, we would add, actsas an amplifier on any fear-based implicitmemories. Caspi et al. (2003) then found that thisgene in humans correlates with an increasedsusceptibility to depression in subjects with majorlife stresses or trauma. However, subjects with theshort-allele, depression-prone gene who had notsuffered major stressors had the same probabilityof developing depression as subjects with thenormal, long allele version of 5-HTT. Theimplication of the study is that only in tandemwith negative implicit memories does the shortallele form of 5-HTT cause expression ofpsychological symptoms.

Likewise, Bennett, Lesch, Heils, Long, Lorenzand Shoaf (2002) found that the same short alleleform of 5-HTT did not have deleterious effects inmonkeys who had normal amounts of nurturingfrom their mothers, but monkeys with deficientrearing showed a wide range of symptoms,including deficiencies in early neurobehaviorialdevelopment, central nervous system integrity,serotonin metabolism, increased antisocialaggression and propensity for alcoholconsumption. These researchers conclude that thegene's regulatory region interacts with earlyexperience to affect central 5-HTT functioning.These findings indicate an environment-dependenteffect of the 5-HTTL genotype (Bennett et al.,2002, p. 119).

These observations suggest a model in whichconstructions, meanings, and protective strategiesthat are formed in response to stressors interactwith the 5-HTT gene to increase its depression-inducing activity. In such emotional-stress-relatedcases, the resultant depression is neitherirreversible nor entirely genetically determined.Transformation of the implicit schema(s) affectingthis gene could be expected to prevent depressiondespite the presence of the gene.

Similarly, symptom-producing imbalances ofhormones (such as cortisol or adrenaline) could bemaintained by a chronic reactivation of implicitmemory. For instance, an implicit memory thatencodes some aspect of a client's usualenvironment as threatening could maintainchronically high levels of adrenaline.

The recognition of bottom-up causation ofmood states—the knowledge that ingestion of asubstance can strongly influence subjectiveexperience—is prehistoric. However, the fact thata symptom such as a mood state has measurablechemical and/or neural correlates does notnecessarily imply bottom-up causation of thatsymptom. We contend that top-down processeslaunched by subcortical implicit memory play aprimary role in the creation of most psychologicalsymptoms. In support of that contention is ourobservation that the methodology of coherencetherapy, which very specifically targets causationby implicit memory, shows consistency of clinicaleffectiveness across a wide range of clients andsymptoms. Controlled studies are of course neededto verify this inference.

Coherence therapy implements in thesubjective domain the same meta-strategy thatneuroscientists seek to implement on the cellularand molecular level: a precise targeting andneutralizing of a well-defined cause of thesymptom. If a person's symptom is caused by acoherent, implicit memory, the best treatmentwould be a psychotherapy designed specificallyfor targeted depotentiation of implicit memory.Lasting removal of the actual, primary cause of asymptom would always be superior to symptomsuppression via counteractive methods that operateat either the neural substrate (SSRIs, for example)or at the psychological/behavioral level (as inpsychotherapies of symptom suppression).

THE VIEW THAT FINDINGS INNEUROSCIENCE IMPLY A THERAPY OF

REPARATIVE ATTACHMENT

Neuroscience shows that new experience cansignificantly shape the brain’s neural wiringthroughout the lifespan. During infancy andchildhood, neuropsychological development of thebrain is powerfully affected by interpersonalinteractions and attachment patterns with primarycaregivers. It is in response to these interactionsthat the developing infant organizes and regulatesits own subjective states and, to a significantdegree, the wiring of its brain. Non-attuned, non-nurturing and traumatic interaction patterns canhave deleterious, symptom-generating effects onbrain development, including the non-integrationand discoordination of various brain systems; the

Competing Visions 11

formation of behavior- and mood-destabilizingimplicit memories with no corresponding explicitmemories or coherent autobiographical narrative;and reduced hippocampal volume (Bremner et al.,1997; Schore, 2001a, 2001b; Siegel, 1999).

Therefore, as indicated in the followingquotations, a central goal of psychotherapy in suchcases should be to bring about the neuralintegration and the concomitant, coherent narrativethat were disrupted by disturbed attachmentpatterns and trauma early in childhood.

Coherent narratives can thus be proposed to bea product of the integration of left and righthemisphere processes: the drive to explaincause and effect relationships (left) and tounderstand the minds of others and of the selfwithin autonoetic consciousness (right). In thismanner, we can propose that coherentnarratives reflect the mind's ability to integrateits processes across time and across therepresentat ional processes of bothhemispheres. (Siegel, 2001, p. 87)

The co-construction of a coherent narrative ofthe trauma may emerge in a relational[therapeutic] context which promotes acallosal transfer of affective information fromthe right to left orbitofrontal regions. Thisstructural advance allows for left hemisphericretrieval and explicit semantic processing ofright hemispheric emotional states encoded inimplicit memory. (Schore, 2001b, p. 245)

A sizable fraction of working therapists haveinterpreted these conclusions from neuroscience tomean that because relational experiences were thecause of a person’s neuro-developmentaldisruptions, they need to be the cure: effectivepsychotherapy should centrally consist of the useof the client-therapist relationship to reverse theeffects of the client’s original, negativeexperiences of relationship by reliably producingpositive experiences of reparative attachment inthe form of the therapist’s caring attention andattuned, empathic understanding (see for exampleLewis, Amini and Lannon, 2000). It is theseexperiences, in this approach, that will bring abouta reorganization of neural linkages, buildingneural integration and manifesting as emotionaland behavioral stabiliz-ation.

According to this widespread view,neuroscience implies a therapy of reparativeattachment, and the essence of therapy should be ahighly skillful reparenting, in the sense that theaim is for the client to receive and experience thereliable, accurate empathy and understanding thatwere not provided earlier in development. Theclient’s felt experience of receiving the therapist’sempathy and understanding is seen as the crucialtransformative ingredient in this therapy. Thework therefore entails a central, explicit focus onthe client’s experience of interactions with thetherapist.

A therapy of reparative attachment is optimal,undoubtedly, for a certain population of therapyclients. But how generally applicable is thisapproach? What determines whether reparativeattachment is a suitable form of therapy fordispelling a given symptom presented by a givenclient? We suggest that the situation is defined bya decision tree consisting of two questions:

Question 1: Is the presenting symptom based in atroubled attachment pattern?

Question 2: If an attachment pattern or schema isthe basis of the presenting symptom, is thatattachment pattern amenable to being changedthrough new experiences of interaction withthe therapist?

We suggest that only when the answer to eachof these questions is yes can a therapy ofreparative attachment be appropriate. If the answerto either question is no, reparative attachmentwork is inappropriate because it cannot dispel thesymptom. The fact that the answer to eitherquestion can be no and often is no is illustrated bycase examples below.

Answering these two questions theoretically isa complex and subtle matter that would go wellbeyond our scope here. However, they can beanswered pragmatically with each therapy clientthrough coherence therapy’s phenomenologicalmethods, without relying on interpretations ortheorizing. We will consider each of the twoquestions in turn.

Competing Visions12

Question 1

Is the presenting symptom based in a troubledattachment pattern?

A therapy client’s attachment pattern orschema is made up of specific, implicit, proceduralconstructs, or knowings, of the forms of availableconnection with others and the specific rules,behaviors and conditions that constituteconnection and relatedness with others. Coherencepsychology refers to this material as the terms ofattachment that caregivers imposed and/or that theindividual has construed in response tointerpersonal experiences.

As described in our previous article (Eckerand Hulley, 2007) and recapitulated earlier here,the great majority of clinical symptoms are causedby specific personal constructs in implicitmemory. Question 1, then, is equivalent to asking:Does the client’s presenting symptom, such asinsomnia or reactive anger, arise from implicitconstructs that define terms of attachment?

The therapist cannot answer that question untilthe discovery work has revealed the pro-symptomconstructs. One person’s reactive anger will proveto be arising from an attachment-defining pro-symptom position, whereas the pro-symptomposition responsible for the anger of another clientwill have nothing to do with terms of attachment.

For example, a man began therapy for his hair-trigger, excessive anger whenever he perceivedanyone as departing even slightly from what hethought had been agreed. He became aware intherapy that he was “stuck” in anger over beingbetrayed by a business partner who robbed him ofa small fortune five years ago. He then furtherdiscovered that he remains “stuck” in this angerfor several well-defined purposes: in order not tofeel powerless; in order not to feel intense grief; inorder to sustain a fantasy of demanding andgetting accountability from the betrayer or fromthe universe; and in order to fiercely protest andprevent an intolerable degree of injustice thatthreatens to disconfirm the moral nature of theworld. The emotional truth of his anger, in otherwords, was that he needs it to protect himself fromcollapsing into feeling powerless, grief-stricken,and in despair and disillusionment over a worldthat has no moral order.

Guided experientially to see if any of thesethemes and purposes had a resonant feeling of

similarity or connection to what he experienced inhis family of origin—his terms of attachment—hefound that they did not. These were existentialissues that he was solving with anger, notattachment issues. For him, the answer to question1 was no—the symptom was not based in atroubled attachment pattern. Because hissymptom-requiring constructs did not define orpertain to terms of attachment (how bondedpersonal relationship works), the therapist’sattuned empathy could not, in itself, create anexperience that would disconfirm and transformhis symptom-requiring constructs. Therefore, atherapy of reparative attachment was notappropriate for dispelling his anger.

Question 2

If an attachment pattern is the basis of thepresenting symptom, is it amenable to beingchanged through new experiences of interactionwith the therapist?

Consider, for example, Carol in our earliercase example. Her presenting symptom was heraversion to sexuality in her marriage. Herdiscovered pro-symptom position was directlybased in a major pattern of noxious, sexualizedrelationship with her mother and in herunquestioned, implicit expectation that she wouldenact that same relational pattern with her owndaughter (which in turn necessitated her symptomof suppressed sexuality). This is a case, then, inwhich an attachment pattern (the client’sconstruction of terms of attachment) was the basisof the presenting symptom.

However, this was not an attachment patternamenable to being changed through new relationalexperiences with the therapist. Her relationshipwith her therapist had no subjective relevance towhat Carol expected herself to enact in herrelationship with her daughter. Carol’s positiveexperience of her therapist, no matter howempathic, safe and trustworthy, could neverspecifically disconfirm her expectation ofreplicating her mother’s sexualized relational role.Therefore reparative attachment methods were notsuitable. The therapist’s accurate empathy couldnot, in itself, transform the implicit personalconstructs maintaining the symptom. Theknowings that would successfully disconfirm


Carol’s expectations were not to be found inexperiences of the client-therapist relationship.

As the foregoing two examples suggest,reparative attachment therapy is not suitable for alarge fraction of clients and therefore cannot be thegeneral form of psychotherapy implied byneuroscience.

Of course, with some clients the client-therapist relationship is subjectively relevant to anattachment schema generating a presentingsymptom. In that case, that attachment schema isamenable to being changed by the client’sexperience of the therapist.5 Then the answer isyes to both questions and reparative attachmentmethods are a valid option—among variousalternatives, as described below.

Reparative attachment workwithin coherence therapy

From the perspective of coherencepsychology, the implicit, subcortical personalconstructs that make up an attachment schema arenot fundamentally different from the implicitconstructs that define and govern other areas ofpersonal experience and behavior. The same meta-methodology of change applies: the creation ofexperiences that discover, integrate and thentransform the symptom-requiring constructions, orpro-symptom positions.

Within coherence therapy, a client’s attach-ment schemas can be transformed by various typesof experiences. The use of experiences of thetherapist’s empathic understanding (reparative

5 It is when the client-therapist relationship i ssubjectively relevant to an attachment schema that atransference projection tends to arise. Use oftransference remains central to therapies of thepsychoanalytic school, but not to the more recentlydeveloped, experiential, neuroscience-orientedtherapies of reparative attachment, as represented byFosha (2001) and Lewis, Amini and Lannon (2000).The latter emphasize the healing effect of the client’sexperience of the therapist’s non-interpretive, accurateempathy, regardless of the presence or absence of atransference projection. Coherence therapy likewise canutilize, but does not rely upon, the arising of clienttransference. As described in the next subsection,coherence therapy also differs significantly fromreparative attachment therapies regarding the curativerole of the therapist’s empathy.

attachment work) is one option among many.When the therapist opts to use reparativeattachment (a valid choice when both questionsdefined above are answered in the affirmative),coherence therapy’s methodology tends toenhance its effectiveness.

According to coherence therapy’s principles ofhow constructs are transformed (see the previousof this series), positive experiences of thetherapist’s empathy and sensitive understandingdo not automatically revise a client’s troubled,original terms of attachment held in implicitmemory. The transformative effect occurs only asa result of a deep disconfirmation produced by aconscious juxtaposition of the new relationalexperience and the longstanding attachmentschema—that is, the two emotional realities areexperienced simultaneously, and both cannot betrue. By actively structuring this disconfirmingjuxtaposition experience, a coherence therapistensures that new relational experiences have atransformative (as distinct from counteractive)effect.

In addition, the coherence-guided discoveryprocess of coherence therapy (also described in theprevious article) typically yields a swift elicitationof the client’s specific attachment schemas,bringing them from the implicit background to theexplicit foreground of awareness, usually in one ortwo sessions. The high level of explicitverbalization sought in coherence therapy in turnmakes the transformation process more accurateand reliable, because the therapist knows exactlywhat constructs need to be juxtaposed anddisconfirmed.

For example, a 42-year-old man was intherapy because none of the marriages or othercouple relationships in his life had ever developedan emotional closeness, and he had finally realizedthat “it must be my fault.” He was the middle offive siblings, with authoritarian parents whoserange of behaviors did not include supplyingchildren with personal understanding or emotionalattunement. In a guided visualization he was backin that family at six years of age. When he wasabsorbed in the scene and describing details, thetherapist, a woman, asked him, “What’s it like foryou, there in your family, when you’re scared orhurting over something?” Persisting with thisfocus for five minutes resulted in his firstawareness of a lifelong emotional truth, verbalized

Competing Visions14

as, “This is not a world where anyone will payattention to what I feel or give me understandingfor how I’m hurting. I don’t matter and I’m all onmy own.”

This brought quiet tears. The therapist said,“There you are, a little boy of six, and sometimesyou’re really hurting or really scared, and youneed a grown-up to understand that and take careof you, but it never, ever happens, so you feel veryalone, like you’re all on your own in this world. Isthat right?” His tears increased as he noddedassent. The therapist allowed a short silence andthen created an experience of disconfirmingjuxtaposition by softly saying, “And how is it foryou right now to be so in touch with that—yourcertainty that you will never get caringunderstanding for what hurts—and at the sametime, to recognize that you are actually getting thatkind of emotional understanding from me, rightnow?”

At this he burst into deep crying, fully feelingboth the intensity of his long-suppressed anguishand a kind of amazed relief that the impossiblewas suddenly happening. When his cryingsubsided, the therapist repeated the same question,again guiding him into the juxtapositionexperience. The man thought for a moment andsaid, “If it’s possible for that to happen, then whydid I get parents who couldn’t do it?” Here heagain cried quietly, the beginning of a grievingprocess that typically accompanies the recognitionand revision of troubled attachment schemas.

In subsequent sessions during the next twomonths, the therapist continued to findopportunities to create disconfirmationexperiences. For example, when the client gave adry, matter-of-fact account of being spoken todemeaningly by his boss at work, the therapistsaid, “I notice that you mention nothing at allabout how you feel from being talked to in thatway by him. And I don’t know if you’re assumingI’ll be like your parents and won’t care about whatyou’re feeling, or if you’re assuming that I’mdifferent from them and will pay attention andunderstand what you’re going through. Could youtell me about both sides of that?” In that way thetherapist was deliberately guiding him to againhold both emotional realities, side by side,juxtaposed. Because attachment schemas are suchcentral, life-organizing constructs with far-reaching ramifications, a series of disconfirmation

experiences across a range of contexts tends to benecessary in order to transform them.

Contrary to the view often implicit in therapiesoriented to reparative attachment, in coherencetherapy the therapist’s empathy is not regarded asbeing curative in itself, but only as serving todisconfirm troubled attachment schemas viajuxtaposition with them. What is curative is theclient’s experiential act of retrieving unconsciousconstructs into awareness and subjecting them toreassessment and disconfirmation. If the client’sattachment schemas do not come into experientialjuxtaposition with the therapist’s empathy andunderstanding, those attachment schemas will notbe altered no matter how much empathy isdelivered by the therapist and enjoyed by theclient.

Similarly, original, troubled terms ofattachment that have become conscious andintegrated, successfully creating a coherentautobiographical narrative, do not automaticallytransform as a result, because this much can occurwithout the original attachment schema beingsubjected to a transforming disconfirmation. Theattachment schema continues to feel real, eventhough fully explicit and conscious, and eventhough the client might now recognize itcognitively as being true in “the past,” not in “thepresent.” Reparative attachment work that reliescentrally on creating a past-present distinction isbound to falter because, while the client’sneocortex finds the past-present distinctionfactually meaningful, terms of attachment are heldas unconscious knowings in implicit memory,which is timeless and immune to that distinction.The past-present distinction is therefore only acounteractive, non-transformational measure thatpits the neocortex against subcortical systems ofimplicit memory. An attachment schema that hasnot been experientially disconfirmed continues tofeel potently real even when fully conscious andeven if the neocortex regards it as being from thepast. The neocortex’s notion of past versus presentcan be truly effective only as an adjunct to adisconfirming juxtaposition.

Transformation of attachment schemasin coherence therapy

Even when use of the client-therapistrelationship for reparative attachment work would


be a suitable option in the sense just described, itis not the therapist’s only way to createexperiences that transform a client’s troubledattachment schema, as noted above. That is,reparative attachment work is not essential foreffective transformation of attachment patterns.Coherence therapy provides for the creation of awide range of other kinds of experience that candiscover, integrate and transform the personalconstructs, or pro-symptoms positions,maintaining negative attachment patterns.

Imaginal techniques provide myriad ways ofcreating potent experiences of old and newattachment schemas in relation to any significantattachment figure, past and present. Suchtechniques become indispensable when reparativeattachment work via the client-therapistrelationship is not suitable. The therapist tailorsimaginal experiences in which the client interactsexperientially with a selected attachment figure.This is effective because, as described in ourprevious article (Ecker and Toomey, 2007),personal constructs in implicit memory activate inresponse to both real and imagined percepts.

For example, an integration experience couldbe created for the discovered attachment schemaof the man described above by guiding him toimagine being six, looking at his parents andprivately thinking explicitly, “If they don’t want topay any attention to me and how I’m feeling, I’msure this means nobody every will, so I’d betternot try to tell anyone what I’m feeling, becausethey won’t care.” This could be followed byvisualizing all of his former wives and girlfriendsand saying to all of them an overt statement of hisemotional truth: “My parents didn’t care about mefeeling hurt or troubled, so I’m sure you won’teither, and I’d only get hurt worse if I tried to tellyou what I’m feeling. I wish we were closer, but Iknow better than to trust you or anybody to careabout what I feel.”

In an imaginal version of the symptomdeprivation technique (Ecker and Hulley, 2000a,2004), this man would be guided into a previewexperience of “how it will be when you’re nolonger viewing all others as not caring about you,and instead you know that some people will care,pay attention and understand.” The purpose of thisexercise is for the client to encounter unwelcomeconsequences that make it more important tomaintain the symptom than live without it, even

with the suffering it brings. Every ecologicalconsequence of transformation that is daunting is atheme of unconscious necessity for maintainingthe symptom. In coherence therapy, as thesedaunting consequences are made conscious,addressed as legitimate challenges, and renderedworkable, the resistance fades and transformationproceeds.6

These are just a few examples of a very widerange of possibilities for experiential work onattachment schemas, other than through relying onthe client-therapist relationship to providereparative attachment experiences.

Coherence, conscious and unconscious,and attachment

Main, Kaplan and Cassidy (1985) determinedacross multiple studies that the security ofattachment of a child is predicted to better than 80percent accuracy by how coherent (well-knit) theparents’ conscious narratives (sense-making) oftheir own childhoods are (for a review see Hesse,1999). This finding is widely understood asindicating the great importance of forming acoherent, conscious narrative for the sufferingsand ordeals one passed through in childhood.Parents whose narratives lack coherence hadtroubled attachment in their own childhoods.

Coherence therapy brings benefits that relatedirectly to these findings. As understood withinthe framework of coherence psychology, when theconscious narrative of an adult lacks coherence,the person’s implicit, unconscious knowledge ofhis or her childhood ordeals nevertheless isalready coherent and, if retrieved into consciousexperience, is the therapeutically optimal guide formaking sense of what was suffered and forforming the needed conscious narrative. Not anycoherent narrative, not any way of making senseof what happened in childhood is emotionally andneurologically healing and symptom-dispelling.The most therapeutic narrative is formed as thehonest facing, verbalizing and declaring of what

6 A video example of imaginal work on an attachment-defining pro-symptom position can be viewed, alongwith transcript and commentaries, in an online shortcourse available at:www.dobt.com/catalog_online_training.htm. Scroll toCourse 700: Obsessive Attachment to Former Lover.

Competing Visions16

the subcortical brain already knows in implicitmemory. Furthermore, this verbalization is animportant part of integrating that subcorticalmemory and setting it up for transformation, as wehave described in our previous article.

Coherence therapy appears to offer amethodology and conceptual framework that iswell-suited to the therapeutic agenda described inSiegel’s call for an "understanding [of] the factorsand mechanisms the mind can use to achieve acoherent integration…in the face of a suboptimalattachment history" (Siegel, 1999, p. 90). Thenarrative and neurological integration that Schore,Siegel, and Main emphasize is a built-in, centralresult of coherence therapy methodology.

How the therapist’s empathy fosters change

As noted above, in coherence therapy the useof the therapist’s empathy to create adisconfirmation of attachment schemas is a specialcase, not a defining feature of the methodology.Yet the therapist’s empathy does reliably haveother, catalytic effects that always support the coremethodology and with some clients are critical toits success. Ecker and Hulley (2004, p. 59) providethe following short account of these functions ofcoherence empathy, a phrase that denotes empathyfocused on the content of a pro-symptom position(as distinct from empathy toward .

1. Calming accompaniment. For some clients,even to approach their pro-symptom material ontheir own would bring feelings of aloneness andfear that are unbearable. Feeling closelyaccompanied by the therapist greatly reduces thesedaunting feelings, making it workable to go intothe material.

2. Fostering alignment of neocortical andsubcortical knowledges. Going into unconsciousemotional material alone (without the therapistbeing present) would, for some clients, result inmerely merging into the troubled material andhelplessly re-suffering it in its original intensity, aplunge into earlier, developmental states withoutretaining the neocortical, adult, observing self. Incontrast, being in empathic connection with thetherapist while going into the troubled, pro-symptom material helps retain the participation ofthe client’s adult, observing self, which has twocrucial effects that bring about the sought-afterneural integration: (a) As experienced by the

client’s full adult self, the unresolved emotionalmaterial is now tolerable (whereas it is intolerableif merely merged into) and the client can open to itand allow integration of it. (b) Retaining theperspective of the adult, neocortical state to somedegree while immersing in the material allows theclient to apprehend, verbalize and integrate thesymptom-generating meanings (pro-symptomconstructs) she or he originally formed, which setsthe stage for transforming them.

3. Creating the state of neuro-affective well-being. In fully integrating a pro-symptom position,whatever its specific content, so that the deepsense and emotional truth is known in symptomsthat had seemed irrational and defective, clients incoherence therapy experience:

• A deep connection-with-self• The actual coherence of self• A profound authenticity of self• Being seen, heard, intimately understood,

honored and affirmed at this same, vulnerablelevel of depth by another human being (thetherapist)

In this state of both connection-with-self andconnection-with-other, the client’s limbic systemand right brain have recovered their native state ofintegrated well-being. This is the same statesought by therapists oriented around reparativeattachment and the use of the client-therapistrelationship. Achieving this state is built intocoherence therapy by virtue of centering the worknot on attachment or the dyadic relationship, buton the coherence of the underlying emotionaltruths generating the client’s symptoms. This statecomes about even with clients for whom noattachment issues arise.

THE VIEW THAT FINDINGS INNEUROSCIENCE IMPLY A THERAPY OF

COGNITIVE REGULATION

Interconnections between brain systems allowfor a wide range of mutual influences betweentheir various functions. For example, recruiting thecognitive talents of the cortex to exert acounteractive, regulating influence on symptomsdriven by emotional brain centers is the strategy ofcognitive-behavioral therapy (CBT), various formsof which are in widespread use today. Cognitive


regulation is the essence of many counteractivemethods designed to stop, fix, prevent, get awayfrom, override, or correct a symptom and replace itwith a desired state. Examples include positivethinking, self-soothing, narrative reframing,rational refuting of irrational beliefs, and thestandard protocols of anger management. Thestrategy of cognitive regulation is regarded insome quarters as the primary implication ofneuroscience for psychotherapy.

The interest of neuroscientists in cognitiveregulation is based on the long hegemony of theview that long-term, emotionally intense memoryin the amygdala is indelible, and that conditionedresponses of mood and behavior that are rooted inamygdalar memory correspondingly cannot beeliminated. With the assumption of indelibilityregarding the brain’s primary emotional center, theonly strategy for remediation of amygdala-generated symptoms has seemed to be theenlisting of other brain regions, particularly thecortex, to compete against, moderate and managethe unwanted responses of amygdalar and otherimplicit memory systems.

However, as described in our previous articleof this series (Ecker and Toomey, 2007),neurodynamic studies since 2000 have shown thatamygdalar implicit memory is not indelible afterall and can be depotentiated. This occurs throughthe reconsolidation of the memory, which is adirect synaptic readjustment in the amygdala,rather than through pitting a cognitive brainsystem against this emotional center.

This major discovery is so recent that theneuroscience field is by no means reoriented to itas yet. Furthermore it is not yet apparent toneuroscientists how reconsolidation can be utilizedin humans. Therefore the counteractive, corticalregulation of symptoms arising subcorticallycontinues to be of strong interest. For example:

A characteristic difficulty in manypsychological disorders, such as depression oranxiety, is the maladaptive cognitiveinterpretation of situations or events. Acomponent of treatment for these disorders isto teach active coping skills, consciouslyapplied strategies that help assure adaptiveinterpretations or reactions to emotionalstimuli… The habits and skills we develop toguide these internally generated emotional

events are critical. Recent research on theneural systems of emotion regulation exploresthe mechanisms underlying the ability to usecognitive and active coping strategies to alteremotional reactions. (LeDoux, 2005, p.183)

The strategy of cognitive regulation exists inrelation to the presumed disregulation of theemotional brain systems. However, coherencepsychology emphasizes and coherence therapydemonstrates that subcortical, emotional brainsystems are fully coherent in their internalknowings and functioning. Symptoms seemirrational and out of control only from the point ofview of cortical consciousness, which is not privyto the subcortical knowings that are coherentlygenerating them. Symptoms, as a rule, are beyondconscious control not because subcortical brainsystems are malfunctioning, but only because alack of neural and psychological integration keepsthe cortex disconnected from the knowings and theagency actually governing the person’s responses,as illustrated in our case examples in the previoustwo articles (Toomey and Ecker, 2007; Ecker andToomey, 2007).

There is a great deal of neurodynamic,empirical evidence concerning cognitiveregulation. Whereas the long-term memory of alearned fear is stored in the amygdala itself(Maren, 1999a and 1999b), a subsequent,counteractive learning is stored elsewhere, in aparticular region of the cortex (Milad and Quirk,2002; Phelps, Delgado, Nearing and LeDoux,2004; Quirk, Likhtik, Pelletier and Paré, 2003).This leaves the amygdalar fear memory unchangedand pits one brain system against another with thehope that the cortex reliably prevails. Thisstructural opposition is the case in classicalextinction learning as well as for diverse forms ofcognitive regulation of emotion, ranging fromorthodox behaviorist therapy to the current “thirdwave” of cognitive-behavioral therapies, such asacceptance and commitment therapy (ACT)(Hayes, Strosahl and Wilson, 2003).

In one line of cognitive regulation research,subjects first viewed a photo or a word that elicitsan emotional arousal, and then carried out acognitive reappraisal of that stimulus, designed todiminish the arousal. Successful reappraisal isfound to result in decreased subjective ratings ofemotional arousal and a measurable decrease in

Competing Visions18

both physiological responses and activation of theamygdala (Gross, 2002; Ochsner, Bunge, Grossand Gabrielli, 2002; Schaefer, Jackson, Davidson,Aguirre, Kimberg and Thompson-Schill, 2002).Brain imaging in such studies has identifiedspecific cortical regions recruited in the cognitivecontrol of emotion (Ochsner et al., 2004; Oschnerand Gross, 2005). The images show an increase incortical (specifically prefrontal) activation coupledwith a decrease in amygdalar activation, both ofwhich are correlated with subjects’ reports ofreduced emotion. These observations support thecognitive-behavioral tenet that the cognitionsattached to emotionally laden stimuli andbehaviors can amplify or weaken emotionalresponses.

Oschner and Gross (2005, p. 243) define aspectrum of cognitive regulatory strategies, inorder of increasing benefit and effectiveness:

• Attentional control via deliberate suppression• Attentional control via distracting secondary

task• Attentional control via attending to non-

emotional features of stimulus• Cognitive change via classical extinction

learning• Cognitive change via the placebo response• Cognitive change via cognitive reappraisal

Contrasting the two ends of that spectrum inlight of their brain imaging studies, these authorsstate that deliberate suppression “of negativeemotions might limit expressive action but doesnot dampen unpleasant experience, worsensmemory, and increases sympathetic nervoussystem activation” (Oschner and Gross, 2005, p.243). At the more effective end of the spectrum,“using reappraisal to regulate emotions isassociated with healthier patterns of affect, socialfunctioning, and well-being than is usingsuppression” (John and Ochsner, 2004, p. 1301).The meta-strategy of some therapies of cognitiveregulation is to help clients develop a moreadaptive repertoire of strategies across thatspectrum.

Ochsner and Gross (2005) report that, withminor variance, the same cortical regions areobserved to activate across the spectrum ofcognitive control strategies listed above. Phelpsand LeDoux (2005) make the same point:

[The cortical] region that has previously beenlinked to extinction learning in humans(Phelps et al., 2004) also showed a similarpattern of response when conditioned fear wasdiminished with a cognitive strategy, suggest-ing that overlapping neural mechanisms foramygdala inhibition/ regulation may supportboth cognitive emotion-regulation strategiesand extinction learning. (P. 183)

For all practical purposes then, and in light ofcurrent brain science, therapies of counteractivecognitive regulation can be regarded as variationson extinction learning.

In summary, it is well established neurally thatcognitive counteracting of emotional reactions hassome degree of effectiveness and verifiabilityunder the short-term conditions of the brainstudies.

A somewhat different picture is encounteredin the context of psychotherapy, however.Outcome studies have shown that cognitive-behavioral therapies, which follow cognitiveregulation strategies, have efficacy, but are only aseffective as placebo treatments that are structurallyequivalent (Baskin, Tierney, Minami andWampold, 2003). This can be interpreted to meanthat the real-life effects of cognitive counteractingare moderate. Indeed, there are several known,built-in weaknesses to this strategy:

Prone to relapse. In an authoritative review ofextinction learning, Bouton (2002, p. 982)comments that because "the original learning is notdestroyed by a retroactive interference treatment,and if relapse is therefore always possible, we needto know how to optimize the new learning so as toprevent the phenomena of relapse." Relapse(recurrence of symptoms after counteracting/extinguishing seems effective) is fairly easilyproduced by (a) an established trigger (conditionedstimulus) occurring in a new context (termedreinstatement); (b) an established trigger occurringin an old context covered by counteractive training(termed spontaneous recovery); (c) relatively highlevels of stress; or (d) passage of time since thecounteractive/extinction learning.

Inherent, unending struggle. The main sourceof unwanted emotional reactions, the amygdala,carries out its own appraisal and reactsindependently of, and far faster than, any


conscious, cortical reappraisal or other type ofregulation can possibly be implemented (Whalen,Rauch, Etcoff, Mcinerney, Lee and Jenike, 1998).This means that a person using a cognitiveregulation strategy is forever on the defensive andthat full suppression of emotional reactions isusually impossible, even when the amygdala’sresponse is conscious. When it is unconscious, asall amygdalar implicit memory is, cognitiveregulatory leverage is particularly weak. Bydesign, cognitive counteracting prepares a personto perpetually oppose symptoms that will continueto tend to arise from an amygdala that still harborsthe same reactive, implicit memory.

Inherent ineffectuality. Amygdalar fearreactions studied in brain imaging studies ofcognitive regulation were relatively mild responsesto photos or words in a safe, experimental setting.Among therapy clients presenting raw, intenseanxiety or panic symptoms, cognitive antidotes aregenerally ineffectual, for these reasons: (a)cognitive antidotes lack compelling realness andfail to alter the emotional root of the reaction; and(b) the stronger the emotional reaction, the weakerand less available is the person’s cognitiveattention to any cognitive antidote. (Behaviortherapists have therefore sought a pharmaceutical“cognitive enhancer” to increase the strength ofnew cognitive learning during exposure therapyfor phobias. Tests have been conducted using D-cycloserine, a compound previously found toincrease the strength of new synapses formedduring extinction learning in rats. In an initial trial(Ressler et al., 2004), administering D-cycloserineto therapy clients just before each session ofexposure therapy for fear of heights resulted in twosessions being about as effective as eight sessionswithout the drug (Travis, 2004). Long-term effectsand relapse rates have not yet been reported.)

In the case example of Carol in the first ofthese three articles (Toomey and Ecker, 2007), acognitive regulation approach to her distaste andcoldness toward marital sex could have takenseveral different forms. For example, the therapistcould have worked along any of these lines:

• Cultivate and make vivid the ways in whichCarol “truly wanted to enjoy the sexualdimension of her marriage.”

• Teach Carol to notice and interrupt negativeself-talk around sex, and replace it withpositive thinking and a focus on anticipatingthe positive effects of physical intimacy withher husband.

• Guide Carol to identify the situationalvariables that constitute her key conditions forrelaxing comfortably into sexuality, and coachher in setting up those conditions.

• Guide Carol and her husband to engage insensate focus exercises that would allow Carolto expect a well-defined, positive experience.

Any of these or other methods ofcounteracting might have partially amelioratedCarol’s sexual aversion, at times. If we consider,however, the power and urgency of the implicit,subcort ical knowings underlying heraversion—the revulsion toward her mother’svoyeuristic, incestuous interest in her own sexualbehavior as a young girl, and her expectation thatsurrendering to sexual feelings would make herperpetrate the same mortifications on her owndaughter—it is obvious that the cognitive antidoteslisted above could not possibly have enoughrealness and strength to override that intenseemotional schema.

However, if that symptom-requiring emotionalschema is first dissolved, ending symptomproduction, then the same methods listed aboveare no longer counteractive and now caneffectively help cultivate new knowings andbehaviors for a satisfying sex life. This is thestrategy of coherence therapy.

In the very recent discovery that theamygdala’s emotional memory is notindelible—that the brain can revise and evendepotentiate a specific implicit memory orsubcortical emotional appraisal—neuroscience hasrecognized a strategy for dispelling emotionalreactions that is a fundamental alternative tocognitive regulation. The option of implicitmemory depotentiation removes the main reasonfor resorting to cognitive regulation and becomesthe psychotherapeutic strategy of choice for thesereasons:

Competing Visions20

• Relapse is impossible and effectiveness isdecisive if the implicit memory driving symp-tom production no longer exists.

• Removing the cause of symptom productionmeans that the symptom ceases to occur withno need for symptom-counteracting measures,cognitive or behavioral.

• An ongoing conflict of one part of thepersonality against another is not created andperpetuated.

Our previous article (Ecker and Toomey,2007) details the theoretical and clinical evidencethat coherence therapy’s methodology is effectivefor implicit memory depotentiation.

SUMMARY AND CONCLUSION

The staggering complexity of the brain andmind will forever allow for a rich diversity ofvalid approaches to alleviating suffering andfostering change, growth and well-being.However, despite the proliferation of psycho-therapeutic models and methods, so far the fieldhas been constrained by a glass ceiling onefficacy. We have argued in this and the previoustwo articles in this series that the marriage ofneuroscience and psychotherapy could helpidentify more reliable, effective methodologiesand techniques. The brain sciences can guidepsychotherapists to aim their creativity and skillsin directions that work most effectively andparsimoniously.

Here we have attempted to help map this newterritory. Our findings, based on extensive clinicalexperience and a review of the research literature,suggest that to counteract a symptom, whetherpharmaceutically or cognitively, is to settle for anefficacy no better than is achieved by placebotreatments that are properly designed to havestructural equivalence. The reason, we havesuggested, is that these methods—drugs andcognitive regulation—strive to compete againstand override the powerful, coherent, adaptive, pro-symptom constructs in implicit memory. This is aninternally oppressive strategy that is inherentlylimited in effectiveness and reliability because itdoes not actually eliminate the roots of symptomproduction. In contrast, we have described

coherence therapy as an internally unifyingmethodology for fundamentally depotentiating theimplicit, symptom-requiring constructs, elimin-ating the need for struggling against them. Wehave argued that this transformation represents anoptimal therapeutic use of neuroplasticity and anoptimal outcome for most clients. Why aim tocompete against the cause of symptom productionwhen the cause can usually be eliminated?Shouldn’t psychotherapy reduce rather thanincrease the amount of internal conflict?

We also have argued that a neuroscientificallyinformed understanding of symptom causation byimplicit memory implies that the therapeuticstrategy of reparative attachment, with its primaryfocus on use of the client-therapist relationship,needs to be selectively applied and is contra-indicated in a sizable fraction of cases. Coherencetherapy provides a meta-psychology and meta-methodology within which (a) the suitability orunsuitability of reparative attachment work can bedetermined, (b) reparative attachment work can becarried out, when suitable, in service ofdepotentiating an attachment schema in implicitmemory (pro-symptom position), and (c) either theclient-therapist relationship or a range of methodsother than use of the client-therapist relationship isavailable for transforming attachment schemas.Furthermore, the achieving of both narrative andneural integration is intrinsic to the methodology.

The arguments and analyses that we haveoffered regarding coherence therapy requireverification through controlled studies. Theputative advantages of coherence therapy relativeto pharmaceuticals and to psychotherapies basedon cognitive regulation, such as CBT, could betested by two types of investigation: controlledcomparative study of outcome efficacy withclients randomly assigned to the three modes oftreatment, and functional imaging studies ofregional changes in brain activity before, duringand after treatment.

Brain imaging studies of the reduction ofdepression by SSRIs and by CBT show that thosetwo types of treatment have quite different effectson over a dozen brain regions (Goldapple et al.,2004), yet the two treatments are known to haveessentially equal efficacy in producing symptomrelief (DeRubeis, Gelfand, Tang and Simons,1999; Hollon et al., 1992). The post-CBT brainscans show regional effects that are consistent


with the psychological model of counteractivechange posited by CBT, lending support for thatmodel. A corresponding study carried out forcoherence therapy could indicate whether co-herence therapy’s models of symptom productionand symptom cessation are likewise consistentwith changes of brain activation resulting fromcoherence therapy. Distinct differences should beapparent between post-treatment brain scans ofCBT and coherence therapy responders. Forexample, as noted earlier, a hyperactive subgenualcingulate (Brodmann area 25) is a key char-acteristic of depressed persons (Mayberg et al.,1999). An important component of cortico-limbicpathways, this region has been shown to be a mainneural correlate of the feeling of sadness (Liotti,Mayberg, Brannan, McGinnis, Jerabek and Fox,2000). Curiously, CBT, even when effective inreducing depressive symptoms, does not diminishthe activation of this region (Goldapple et al.,2004). We have conjectured an explanation basedon coherence psychology: hyperactivation of area25 may be caused by chronic activation ofspecific, unconscious, sadness-inducing personalconstructs held in subcortical implicit memory.CBT by design does not access or depotentiatesuch deeply unconscious material, but we believecoherence therapy will prove to do precisely that,and would therefore yield post-therapy brain scansof responders that show diminished activity in area25.

In conclusion, we hope to have providedperspectives that are useful to clinicians andresearchers seeking to bring psychotherapy to newlevels of effectiveness.

ACKNOWLEDGEMENTS

The authors are grateful to Laurel Hulley andTimothy Desmond for many helpful discussionsand clarifications of drafts of this article, and toBret Lyon for influential critiques of ourarguments regarding reparative attachmenttherapy. Much benefit came also from critiquesprovided by this article’s anonymous reviewers,whom we thank sincerely.

REFERENCES

Amsterdam, J. D., Fawcett, J., Quitkin, F. M.,Reimherr, F.W., Rosenbaum, J.F., Michelson, D.,

Hornig-Rohan, M., & Beasley, C.M. (1997).Fluoxetine and norfluoxetine plasmaconcentrations in major depression: A multicenterstudy. American Journal of Psychiatry, 154, 963-969.

Antonuccio, D., Danton, W., DeNelsky, G., Greenberg,R., & Gordon, J. (1999). Raising questions aboutantidepressants. Psychotherapy andPsychosomatics, 68, 3-14.

Baskin, T.W., Tierney, S.C., Minami, T., & Wampold,B.E. (2003). Establishing specificity inpsychotherapy: A meta-analysis of structuralequivalence of placebo controls. Journal ofConsulting and Clinical Psycholology. 71, 973-979.

Benedetti, F., Helen, S., Mayberg, H.S., Wager, T.D.,Stohler, C.S., & Zubieta, J.-K. (2005).Neurobiological mechanisms of the placebo effect.Journal of Neuroscience, 25, 10390-402.

Bennett, A. J., Lesch, K. P., Heils, A., Long, J. C.,Lorenz, J. G., & Shoaf, S. E. (2002). Earlyexperience and serotonin transporter gene variationinteract to influence primate CNS function.Molecular Psychiatry, 7, 118-22.

Black K., Shea C., Dursun S., & Kutcher S. (2000).Selective serotonin reuptake inhibitordiscontinuation syndrome: proposed diagnosticcriteria. Journal of Psychiatry and Neuroscience,25, 255-261.

Bouton, M. E. (2002). Context, ambiguity, andunlearning: Sources of relapse after behavioralextinction. Biological Psychiatry, 52, 976-986.

Bouton, M. E. (2004). Context and behavioralprocesses in extinction. Learning and Memory, 11,485–494.

Bowlby, J. (1998). Attachment and Loss, Vol. 2:Separation. New York: Vintage/Ebury.

Bremner, J. D., Randall, P., Vermetten, E., Staib, L.,Bronen, R. A., Mazure, C., Capelli, S., McCarthy,G., Innis, R.B., & Charney, D.S. (1997). Magneticresonance imaging based measurement ofhippocampal volume in posttraumatic stressdisorder related to childhood physical and sexualabuse—a preliminary report. BiologicalPsychiatry, 41, 23-32.

Carrasco, J. L., & Sandner, C. (2005). Clinical effectsof pharmacological variations in selective serotoninreuptake inhibitors: An overview. Internationaljournal of clinical practice, 59, 1428-1434.

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

Cirelli, C. (2005). A molecular window on sleep:

Competing Visions22

changes in gene expression between sleep andwakefulness. The Neuroscientist, 11, 63-74.

Cozolino, L. (2002). The neuroscience ofpsychotherapy: Building and rebuilding the humanbrain. New York & London: W.W. Norton.

de la Fuente-Fernández, R., & Stoessl, A.J. (2002). Thebiochemical bases for reward: Implications for theplacebo effect. Evaluation & the HealthProfessions, 25, 387-398.

Delgado, P.L. (2000). Depression: the case for amonoamine deficiency. Journal of ClinicalPsychiatry. 61, 7-11.

DeRubeis, R.J., Gelfand, L.A., Tang, T.Z., & Simons,A.D. (1999). Medications versus cognitivebehavior therapy for severely depressedoutpatients: mega-analysis of four randomizedcomparisons. American Journal of Psychiatry, 156,1007-1013.

Dobbs, D. (2006, August/September). Turning offdepression. Scientific American Mind, 17(4), 26-31.

Ecker, B. (2003). The hidden logic of anxiety: Look forthe emotional truth behind the symptom.Psychotherapy Networker, 27 (6), 38-43, 58.Retrieved March 2, 2006, fromwww.findarticles.com/p/articles/mi_qa4016/is_200311/ai_n9313906

Ecker, B. (Speaker) (2005). Utilizing the brain’s hiddenrules for change. (Audio CD Item No. 715-506).Psychotherapy Networker Symposium,Washington, DC. Available from:www.playbacknow.com

Ecker, B. (Speaker) (2006). The effectiveness ofpsychotherapy: Constructivism to the rescue! 12th

Biennial Conference of the ConstructivistPsychology Network (Keynote Address), SanMarcos, California.

Ecker, B. & Hulley, L. (1996). Depth oriented brieftherapy: How to be brief when you were trained tobe deep, and vice versa. San Francisco: Jossey-Bass.

Ecker, B. & Hulley, L. (Producers). (1997). Compulsiveunderachieving. [Video and viewer’s manual].Oakland, CA: Pacific Seminars.

Ecker, B. & Hulley, L. (2000a). Depth-oriented brieftherapy: Accelerated accessing of the coherentunconscious. In J. Carlson & L. Sperry (Eds.),Brief therapy with individuals and couples (pp.161-190). Phoenix, AZ: Zeig, Tucker & Theisen.

Ecker, B. & Hulley, L. (2000b). The order in clinical“disorder”: Symptom coherence in depth orientedbrief therapy. In R. A. Neimeyer & J. Raskin(Eds.), Constructions of disorder (pp. 63-89).Washington, DC: American PsychologicalAssociation Press.

Ecker, B. & Hulley, L. (2000c). A new zone of

effectiveness for psychotherapy. New Therapist, 6,31-33. Retrieved March 10, 2006, fromhttp://www.newtherapist.com/ecker6.html

Ecker, B. & Hulley, L. (2002a). Deep from the start:Profound change in brief therapy. PsychotherapyNetworker, 26(1), 46-51, 64. Retrieved March 2,2006, fromwww.findarticles.com/p/articles/mi_qa4016/is_200201/ai_n9033893

Ecker, B. & Hulley, L. (2002b). DOBT toolkit for in-depth effectiveness: Methods and concepts ofdepth-oriented brief therapy. New Therapist, 20,24-29.

Ecker, B. & Hulley, L. (2004). Depth-oriented brieftherapy practice manual and training guide.Oakland, CA: Pacific Seminars. Available fromwww.dobt.com/manual.htm

Ecker, B., & Toomey, B. (2007). Depotentiation ofsymptom-producing implicit memory in coherencetherapy. Journal of Constructivist Psychology, ,- .

Edwards, R., Peet, M., Shay, J., & Horrobin, D. (1998).Omega-3 polyunsaturated fatty acid levels in thediet and in red blood cell membranes of depressedpatients. Journal of Affective Disorders, 48, 149-155.

Encinas, J.M., Vaahtokari, A., & Enikolopov, G.(2006). Fluoxetine targets early progenitor cells inthe adult brain. Proceedings of the NationalAcademy of Sciences, 103, 8233-38.

Fisher, S., & Greenberg, R. (Eds). (1997). Fromplacebo to panacea: Putting psychiatric drugs tothe test. New York: John Wiley & Sons.

Fosha, D. (2001). The dyadic regulation of affect.Journal of Clinical Psychology. 57, 227-42.

Freud, S. (1895/1966). Project for a scientificpsychology. In Complete psychological works,Standard Ed., Vol. 1 (1966) (pp. 295–397).London: Hogarth Press.

Goldapple, K., Segal, Z., Garson, C., Lau, M., Bieling,P., Kennedy, S., & Mayberg, H. (2004).Modulation of cortical-limbic pathways in majordepression: Treatment-specific effects of cognitivebehavior therapy. Archives of General Psychiatry,61, 34-41.

Greenberg, G. (2003, November/December). Is itProzac? Or placebo? Mother Jones. RetrievedMarch, 4, 2006, fromwww.motherjones.com/news/feature/2003/11/ma_565_01.html

Greenberg, G. (2005, August). Misery’s fogs. Harper’sMagazine, 89-94.

Greengard, P. (1976). Possible role for cyclicnucleotides and phosphorylated membrane proteinsin postsynaptic actions of neurotransmitters.Nature, 260, 101–108.


Gross, J.J. (2002). Emotion regulation: Affective,cognitive, and social consequences.Psychophysiology, 39, 281–291.

Harris, K.P., Wheeler, D.C., & Chong, C.C. (2002). Aplacebo-controlled trial examining atorvastatin indyslipidemic patients undergoing CAPD. KidneyInternational, 61, 1469–1474.

Hariri, A.R., Mattay, V.S., Tessitore, A., Kolachana, B.,Fera, F., Goldman, D., Egan, M.F., & Weinberger,D.R. (2002). Serotonin transporter geneticvariation and the response of the human amygdala.Science, 19, 400-403.

Hayes, S.C., Strosahl, K.D., & Wilson, K.C. (2003).Acceptance and commitment therapy: Anexperiential approach to behavior change. NewYork: Guilford.

Hebb, D. 0. (1949) The organization of behavior: Aneuropsychological theory. New York: Wiley.

Hesse, E. (1999). The adult attachment interview:Historical and current perspectives. In J. Cassidy & P.Shaver (Eds.), Handbook of attachment (pp. 395–433).New York: Guilford Press.

Heurteaux, C., Lucas, G., Guy, N., El Yacoubi, M.,Thümmler, S., Peng, X.-D., Noble, F., Blondeau, N.,Widmann, C., Borsotto, M., Gobbi, G., Vaugeois, J.-M., Debonnel G., & Lazdunski, M. (2006). Deletionof the background potassium channel TREK-1 resultsin a depression-resistant phenotype. NatureNeuroscience, 9, 1134-1141.

HHMI News (2002, Decemeber 13). Researchersdiscover gene that controls learned fear. HowardHughes Medical Institute News. Retrieved March2, 2006, from www.hhmi.org/news/kandel3.html

HHMI News (2005, November 18). Loss of fear factormakes timid mouse bold. Howard Hughes MedicalInstitute News. Retrieved March 3, 2006, fromwww.hhmi.org//news/kandel20051118.html

Hollon, S.D., DeRubeis, R.J., Evans, M.D., Wiemer,M.J., Garvey, M.J., Grove, W.M., & Tuason, V.B.(1992). Cognitive therapy and pharmacotherapy fordepression. Singly and in combination. Archives ofGeneral Psychiatry, 49, 774-781.

Hubble, M.A., Duncan, B.L., Miller, S.D., Hubble, B.L(1999) The heart & soul of change: What works intherapy. Washington, DC: American PsychologicalAssociation.

Huber, K.M., Gallagher, S.M., Stephen, T., Warren,S.T., & Bear, M.F. (2002). Altered synapticplasticity in a mouse model of fragile X mentalretardation. Proceedings of the National Academyof Sciences, 99, 7746–7750.

Hyman, S.E., & & Nestler, E.J. (1996). Initiation andadaptation: a paradigm for understandingpsychotropic drug action. American Journal ofPsychiatry, 153, 151-162.

Jenkins, D.J., Kendall, C.W., Marchie, A., Faulkner, D.,

Wong, J., deSouza, R., Emam, A., Parker, T.L.,Vidgen, E., Lapsley, K.G., Trautwein, E.A., Josse,R.G., Leiter, L.A., & Connelly, P.W. (2003). Adietary portfolio of cholesterol-lowering foodsversus lovastatin on serum lipids and C-reactiveprotein. Journal of the American MedicalAssociation, 290, 502-510.

John, O.P., & Gross, J.J. (2004). Healthy and unhealthyemotion regulation: Personality processes,individual differences, and lifespan development.Journal of Personality, 72, 1301-1334.

Kalisch, R., Wiech, K., Critchley, H.D., Seymour, B.,O'Doherty, J.P., Oakley, D.A., Allen, P., & Dolan,R.J. (2005). Anxiety reduction throughdetachment: Subjective, physiological, and neuraleffects. Journal of Cognitive Neuroscience, 17,874-883.

Kandel, E.R. (1998). A new intellectual framework forpsychiatry. American Journal of Psychiatry, 155,457–469.

Kandel, E.R. (2001). The molecular biology of memorystorage: a dialog between genes and synapses.Science, 294, 1030-1038. retrieved March 2, 2006,fromhttp://caspar.bgsu.edu/~courses/Reading/Papers/2000Kan.pdf

Karen, R. (1997). Becoming attached: Psychology'seffort to understand the power of first relationshipsand how they impact our capacity to love. Oxford,UK: Oxford University Press.

Kelly, G.A. (1955). The psychology of personalconstructs (Vols. 1 & 2). New York: W.W. Norton.

Kirsch, I., Moore, T., Scoboria, A., & Nicholls, S.(2002). The emperor's new drugs: An analysis ofantidepressant medication data submitted to theU.S. food and drug administration. Prevention &Treatment, 5. Retrieved March 2, 2006, fromwww.journals.apa.org/prevention/volume5/pre0050023a.html.

Kirsch, I., & Sapirstein, G. (1998). Listening to Prozacbut hearing placebo: a meta analysis ofantidepressant medication. Prevention &Treatment, 1, Article 0002a. Retrieved Feb. 11,2007 from:www.journals.apa.org/pt/prevention/volume1/pre0010002a.html

Lacasse, J.R., & Leo, J. (2005) Serotonin anddepression: A disconnect between theadvertisements and the scientific literature. PLoSMedicine, 2. Retrieved March 1, 2006 fromhttp://medicine.plosjournals.org/perlserv/?request=get-document&doi=10.1371/journal.pmed.0020392

LeDoux, J.E. (2000). Emotion circuits in the brain.Annual Review of Neuroscience, 23, 155-184.

LeDoux, J.E., Xagoraris, A., & Romanski, L. (1989).

Competing Visions24

Indelibility of subcortical emotional memories.Journal of Cognitive Neuroscience, 1, 238-243.

Lewis, T., Amini, F., & Lannon, R. (2000). A generaltheory of love. New York: Random House.

Lewontin, R. (1980). Sociobiology: Another biologicaldeterminism. International Journal of HealthServices, 10, 347–363.

Lewontin, R. (2001). Triple helix: Gene, organism,environment. Cambridge, MA: Harvard.

Liotti, M., Mayberg, H.S., Brannan, S.K., McGinnis, S.,Jerabek, P., & Fox, P.T. (2000). Differentiallimbic-cortical correlates of sadness and anxiety inhealthy subjects: implications for affectivedisorders. Biological Psychiatry, 48, 30-42.

Lôo, H., Saiz-Ruiz, J., Costa e Silva, J., Ansseau, M.,Herrington, R., Vaz-Serra, A., Dilling, H., & deRisio, S. (1999). Efficacy and safety of tianeptinein the treatment of depressive disorders incomparison with fluoxetine. Journal of AffectiveDisorders, 56, 109-18.

Main, M., Kaplan, N., & Cassidy, J. (1985). Security ininfancy, childhood, and adulthood: A move to thelevel of representation. In I. Bretherton & E.Waters (Eds.), Growing points of attachmenttheory and research. Monographs of the Societyfor Research in Child Development, 50(1,2), 67-104.

Mann, J.J. (1999). Role of the serotonergic system inthe pathogenesis of major depression and suicidalbehavior. Neuropsychopharmacology, 21, 99S-105S.

Maren, S. (1999a). Long-term potentiation in theamygdala: A mechanism for emotional learningand memory. Trends in Neuroscience, 22,561–567.

Maren, S. (1999b). Neurotoxic basolateral lesionsimpair learning and memory but not performanceof conditioned fear in rats. Journal ofNeuroscience, 19, 8696–8703.

Mayberg, H. S., Liotti, M., Brannan, S. K., McGinnis,S., Mahurin, R. K., Jerabek, P. A., Silva, J. A.,Tekell, J. L., Martin, C. C., Lancaster, J. L. & Fox,P. T. (1999). Reciprocal limbic-cortical functionand negative mood: converging pet findings indepression and normal sadness. American Journalof Psychiatry, 156, 675-682.

Mayberg, H.S., Lozano, A.M., Voon, V., McNeely,H.E., Seminowicz, D., Hamani, C., Schwalb, J.M.,& Kennedy, S.H. (2005). Deep brain stimulationfor treatment-resistant depression. Neuron, 45,651-660.

Meaney, M. (2004). The nature of nurture: Maternaleffects and chromatic remodeling. In J.T. Cacioppo& G.G. Berntson (Eds.), Essays in socialneuroscience. Cambridge, MA: MIT Press.

Meltzer, H. (1989). Serotonergic dysfunction indepression. The British Journal of PsychiatrySuppliment, 8, 25–31.

Milad, M.R., & Quirk, G.J. (2002). Neurons in medialprefrontal cortex signal memory for fear extinction.Nature, 420, 70–74.

Moncrieff, J., & Kirsch, I. (2005). Efficacy ofantidepressants in adults. British Medical Journal,331, 155-157.

Myers, K.M. & Davis, M. (2002). Behavioral andneural analysis of extinction. Neuron, 36, 567–584.

Nader, K. (2003). Memory traces unbound. Trends inNeuroscience, 26, 65-72.

Nader, K., Schafe, G.E., & LeDoux, J.E. (2000). Fearmemories require protein synthesis in the amygdalafor reconsolidation after retrieval. Nature, 406,722-726.

Neimeyer, R. A., & Bridges, S. K. (2003). Postmodernapproaches to psychotherapy. In A. S. Gurman & S. B.Messer (Eds.), Essential psychotherapies (2nd ed.) (pp.272-316). New York: Guilford.

Neimeyer, R. A., & Raskin, J. D. (2001). Varieties ofconstructivism in psychotherapy. In K. S. Dobson(Ed.), Handbook of cognitive-behavioral therapies(pp. 407-411). New York: Guilford.

Ochsner, K.N., Bunge, S.A., Gross, J.J., & Gabrieli,J.D. (2002). Rethinking feelings: An fMRI study ofthe cognitive regulation of emotion. Journal ofCognitive Neuroscience, 14, 1215–1229.

Ochsner, K.N., & Gross, J.J. (2005). The cognitivecontrol of emotion. Trends in Cognitive Science, 9,408-409.

Ochsner, K.N., Ray, R.D., Cooper, J.C., Robertson,E.R., Chopra, S., Gabrieli, J.D., & Gross, J.J.(2004). For better or for worse: Neural systemssupporting the cognitive down- and up-regulationof negative emotion. NeuroImage, 23, 483–499.

Olivier, B., VanOorschot, R., & Waldinger, M.D. (1998).Serotonin, serotonergic receptors, selective serotoninreuptake inhibitors and sexual behaviour.International Journal of ClinicalPsychopharmacology, 13 (Suppl. 6), S6-S14.

Olson, S. (2004). The synapse revealed. HowardHughes Medical Institute Bulletin, 17(3), 14-23.Retrieved March 3, 2006, fromwww.hhmi.org/bulletin/fall2004/synapses/synapses2.html

Panksepp, J. (2004). Affective neuroscience: Thefoundations of human and animal emotions.Oxford, UK: Oxford University Press.

Pedreira, M.E., Perez-Cuest, L.M. & Maldonado, H.(2004). Mismatch between what is expected andwhat actually occurs triggers memoryreconsolidation or extinction. Learning & Memory,11, 579-585.

Phelps, E.A., Delgado, M.R., Nearing, K.I., & LeDoux,


J.E. (2004). Extinction learning in humans: Role ofthe amygdala and vmPFC. Neuron, 43, 897–905.

Phelps, E.A. & LeDoux, J.E. (2005). Contributions ofthe amygdala to emotion processing: From animalmodels to human behavior. Neuron, 48, 175–187.

Preskorn, S.H. (1997). Clinically relevantpharmacology of selective serotonin reuptakeinhibitors. An overview with emphasis onpharmacokinetics and effects on oxidative drugmetabolism. Clinical Pharmacokinetics, 32 (Suppl.1), 1-21.

Quirk, G.J., Likhtik, E., Pelletier, J.G. & Paré, D.(2003). Stimulation of medial prefrontal cortexdecreases the responsiveness of central amygdalaoutput neurons. Journal of Neuroscience, 23, 8800-8807.

Ramon y Cajal, S. R. (1892). A new concept of theconcept of the histology of the central nervoussystem. In Rottenberg, D.A. and Hochberg, F.H.(Eds.) (1977), Neurological classics in moderntranslation (pp. 7–29). New York: Hafner.

Rescorla, R.A. (2001). Experimental extinction. In R.R.Mowrer & S. Klein (Eds), Handbook ofcontemporary learning theories (pp. 119-154).Mahwah, NJ: Erlbaum.

Ressler, K.J, Rothbaum, B.O., Tannenbaum, L.,Anderson, P., Graap, K., Zimand, E., Hodges, L.,& Davis, M. (2004). Cognitive enhancers asadjuncts to psychotherapy: use of D-cycloserine inphobic individuals to facilitate extinction of fear.Archives of General Psychiatry, 61, 1136-44.

Santarelli, L., Saxe, M., Gross, C., Surget, A., Battaglia,F., Dulawa, S., Weisstaub, N., Lee, J., Duman, R.,Arancio, O., Belzung, C., & Hen, R. (2003).Requirement of hippocampal neurogenesis for thebehavioral effects of antidepressants. Science, 301,805-809.

Sara, S.J. (2000). Retrieval and reconsolidation:Toward a neurobiology of remembering. Learning& Memory, 7, 73-84.

Schaefer, S. M., Jackson, D. C., Davidson, R. J.,Aguirre, G. K., Kimberg, D. Y., & Thompson-Schill, S. L. (2002). Modulation of amygdalaractivity by the conscious regulation of negativeemotion. Journal of Cognitive Neuroscience, 14,913-921

Schore, A.N. (2001a). The effects of a secureattachment relationship on right braindevelopment, affect regulation, and infant mentalhealth. Infant Mental Health Journal, 22, 7–66.

Schore, A.N. (2001b). The effects of relational traumaon right brain development, affect regulation, andinfant mental health. Infant Mental Health Journal,22, 201–269.

Schore, A.N. (2003). Affect dysregulation and disordersof the self. New York: W.W. Norton.

Shores, M.M., Pascually, M., Lewis, N.L., Flatness, D.,& Veith, R.C. (2001) Short-term sertralinetreatment suppresses sympathetic nervous systemactivity in healthy human subjects.Psychoneuroendocrinology, 26, 433-439.

Shumyatsky G.P., Tsvetkov E., Malleret G., Hatton M.,Vronskaya S., Hampton L., Battey, J.F., Dulac, C.,Kandel, E.R., & Bolshakov, V.Y. (2002).Identification of a signaling network in the lateralnucleus of Amygdala important for inhibitingmemory specifically related to learned fear. Cell,111, 905-918.

Shumyatsky, G.P., Malleret, G., Shin, R.-M., Takizawa,S., Tully, K., Tsvetkov, E., Zakharenko, S.S.,Joseph, J., Vronskaya, S., Yin, D., Schubart, U.K.,Kandel, E.R., & Bolshakov, V.Y. (2005). Stathmin,a gene enriched in the amygdala, controls bothlearned and innate fear. Cell, 123, 697-709.Retrieved March 3, 2006, fromwww.cell.com/cgi/content/abstract/123/4/697

Siegel, D. (1999). The developing mind: Toward aneurobiology of interpersonal experience. NewYork: Guilford Press.

Seigel, D. (2001). Toward an interpersonalneurobiology of the developing mind: Attachment,relationships, mindsight, and neural integration.Infant Mental Health Journal (Special Edition),22(1,2), 67–94.

Stricker, G., & Trierweiler, S. J. (1995). The localclinical scientist: A bridge between science andpractice. American Psychologist, 50, 995–1002.

Tamam, L., & Ozpoyraz, N. (2002). Selective serotoninreuptake inhibitor discontinuation syndrome: areview. Advances in Therapy, 19, 17-26.

Toomey, B., & Ecker, B. (2007). Of neurons andknowings: Constructivism, coherence psychologyand their neurodynamic substrates. Journal ofConstructivist Psychology, , -.

Travis, J. (2004). Fear not. Science News, 165, 42-44.Vallance, A.K. (2006). Something out of nothing: the

placebo effect. Advances in Psychiatric Treatment,12, 287-296.

Walsh, B.T., Seidman, S.N., Sysko, R., & Gould, M.(2002). Placebo response in studies of majordepression: Variable, substantial, and growing.Journal of the American Medical Association, 287,1840-1847.

Whalen, P.J., Rauch, S.L., Etcoff, N.L., Mcinerney,S.C., Lee, M.B., & Jenike, M.A. (1998). Maskedpresentations of emotional facial expressionsmodulate amygdala activity without explicitknowledge. Journal of Neuroscience, 18, 411-418.Zarate, C.A., Singh, J.B., Carlson, P.J.,Brutsche, N.E., Ameli, R., Luckenbaugh, D.A.,Charney, D.S., & Manji, H.K. (2006). Arandomized trial of an N-methyl-D-aspartate

Competing Visions26

antagonist in treatment-resistant major depression.Archives of General Psychiatry, 63, 856-864.

Zarate, C.A., Singh, J.B., Carlson, P.J., Brutsche, N.E.,Ameli, R., Luckenbaugh, D.A., Charney, D.S., &Manji, H.K. (2006). A randomized trial of an n-

methyl-d-aspartate antagonist in treatment-resistantmajor depression. Archives of General Psychiatry,63, 856-864.

competing visions of the implications of...

Documents