|Home | About | Journals | Submit | Contact Us | Français|
Ever since the idea was accepted that memory is associated with alterations in synaptic strength, studies on the cellular and molecular mechanisms responsible for the plastic changes in neurons have attracted wide interest in the scientific community. This article explores the process of memory consolidation leading to persistent modifications in synaptic plasticity as a mechanism by which psychotherapy facilitates changes in the permanent storage of information acquired throughout the individual's life. The psychobiological interrelationships of affect, attachment, and memory offer a perspective regarding the etiology and treatment of clinical disturbances of affect. Analogies between brain physiology and modes of psychotherapy provide the foundation for a review of psychiatric disorders involving the inability to control fear, obsessions, compulsions, and delusions, all of which respond to psychotherapeutic interventions.(The Journal of Psychotherapy Practice and Research 1999; 8:103–114)
The role of learning in the conduct of psychotherapy dates back to mid-century. The neuroanatomist Ram¢n y Cajal1 had earlier discovered that information could be stored by modifying the connections between communicating nerve cells in order to form associations. In 1949 this idea was formalized by Hebb,2 who suggested that such modifications should take place between the connected cells if and only if both neurons were simultaneously active. From this idea emerged the well-known Hebb's rule: When an axon of cell A is near enough to excite a cell B and repeatedly and persistently takes part in firing it, some growth process or metabolic change takes place in one or both cells such that A's efficacy, as one of the cells firing B, is increased. Information is therefore encoded by strengthening the connections between neurons that are simultaneously activated. The discovery of Hebb-like synaptic plasticity in a putative memory structure was one of the most important neurophysiological finds of the 1950s. Hebb's main interest was considering how complex neural networks could account for phenomena such as perception and memory. His postulate of synaptic changes offered a theory of how such changes could support the formation of neural networks. Studies of various forms of synaptic plasticity in the central nervous system provided insights into the cellular and molecular mechanisms for certain types of learning and memory.3–5 Synaptic plasticity became the target of much neurobiological research as its role in memory formation was elucidated. Evidence indicated that activity-dependent short-term and long-term changes in strength of synaptic transmission are important for memory processes.
Cortical maps are dynamic constructs that are remodeled in detail by behaviorally important experiences through life. Over the past decade, the number of experimental papers reporting physiological plasticity in primary neocortical regions, following certain types of controlled sensory experience, has increased greatly.6–9 These reports have been characterized by specific changes in receptive fields of individual neurons and/or the distributions of receptive fields across cortical maps. In parallel with developments in the field of cortical map plasticity, studies of synaptic plasticity characterized specific elementary forms of plasticity, including associative long-term potentiation (LTP) and long-term depression (LTD) of excitatory postsynaptic potentials.10,11 A great deal of the experimental work linking modifiable synapses to macroscopic brain behavior has been carried out in the hippocampal formation, a part of the limbic system that has long been implicated in memory formation.12–15 There is now a multitude of experimental evidence linking hippocampal long-term potentiation to memory formation.16–21
In this article, we examine a memory model of psychotherapy based on brain plasticity as a prerequisite for any long-lasting change in behavior, cognition, and emotions, and therefore for all psychotherapeutic effects. After considering learning-induced anatomical changes demonstrated by psychotherapy trials, we present analogies between brain physiology and psychotherapy schools. Finally, we discuss brain plasticity from the perspective of psychiatric disorders involving the inability to control fear, obsessions, compulsions, and delusions, all of which respond to psychotherapeutic interventions.
What happens to information once it has been transduced by the sensory receptors and passed into the brain? Long-term potentiation is currently regarded as the best existing memory model. There are three known forms of synaptic plasticity at CNS synapses: 1) LTP mediated by N-methyl-d-aspartate (NMDA) receptor activation; 2) LTP mediated by voltage-dependent calcium channel activation; and 3) LTD mediated by the NMDA receptor. A detailed review of the status of research into the mechanisms underlying LTP has been presented by Bliss and Collingridge.22 Suffice it to say that the neural mechanism underlying cortical representational remodeling is that it is a result of synaptic plasticity, primarily LTP of excitatory synapses following a Hebbian learning rule. Post and Weiss23 posit that mechanisms involved in neuronal learning and memory, such as LTP and LTD, are used and reused in the molding of personality and behavior based on experience. They postulate that for higher order processes such as emotional memory, such neuroplasticity is occurring at increasingly larger numbers of synapses and cell assemblies with increasing mechanistic complexity and self-organization. Experiments conducted by Bear24 reveal that many synapses in the hippocampus and neocortex are bidirectionally modifiable, that the modifications persist long enough to contribute to long-term memory storage, and that key variables governing the sign of synaptic plasticity are the amount of NMDA receptor activation and the recent history of cortical activity (Figure 1). The principle common to representation formation in nearly all neural networks is that of associability, which is the idea that streams of information are combined by forming, strengthening, or pruning connections between them to form new representations that can later be retrieved.25,26
Most accounts of perceptual learning are concerned with changes in neuronal sensitivity or changes in the way a stimulus is represented across two distinct memory systems, depicted in Figure 2 as explicit and implicit memory. These two types of memory are fundamentally separate brain functions that rely on different sets of neural structures and physiologic properties, and they result in distinct patterns of neural activity as detected by positron emission tomography (PET)27 and electroencephalography.28 The explicit memory system records experience for later recall by utilizing temporal lobe structures, especially the hippocampus.29,30 In contrast, implicit memory is an enduring neural structure that depends on the basal ganglia and whose existence is inferred from observable influence on emotional behaviors related to early attachment experiences. Information from this system is not available for conscious recall.31–36 Comprehensive reviews of implicit memory have been provided by Reber,37 Schacter,38 Schacter et al.,39 and Squire et al.30
The distinction between the two memory systems is significant both in the course of development over the life cycle and in the context of memory modulation through psychotherapy. Amini et al.40 emphasize that evidence exists to support the hypothesis that human infants are equipped with a functional memory system at birth and that memory is more capable of implicit learning than explicit learning at this neurodevelopmental stage. Information regarding affect undergoes processing in the implicit system. The implicit system is capable of extracting and storing prototypes and rules from exposure to large amounts of complex information. Once learned implicitly, rules may exert a self-perpetuating bias for interpreting later experience in a manner consistent with past experience, regardless of the appropriateness of such an interpretation. Therefore, the information learned in this way is not available for conscious processing and reflection, but rather guides behavior without impinging on consciousness.
In psychotherapy, these patterns of implicit rules are revealed and reflected upon, and change occurs through the learning of new patterns explicitly repeated until the new habit-based manner is engrained in the implicit memory system. Within this proposition, psychotherapeutic change may be attributed to a process of insight or to the provision of abstract explanations regarding underlying relationship patterns.
Animal research examining habituation, sensitization, and classical conditioning has offered some intriguing insights into how learning affects the brain. The simplest neural chain investigated was the monosynaptic reflex arc, which is used to describe the mechanism of primary learning in the gill withdrawal reflex of the marine snail Aplysia californica.41–43 In habituation, an animal learns to suppress its response to a stimulus that is neither rewarding nor harmful. In contrast, sensitization occurs when the animal encounters a harmful stimulus and learns to respond more vigorously to a variety of other stimuli. Classical conditioning describes learning by associating one type of stimulus with another. The animal is exposed to an initially weak or ineffective stimulus that becomes highly effective in producing a response after it has been associated with a strong unconditioned stimulus.44 Because gill withdrawal in the Aplysia is a change that occurs within the reflex arc, it is sometimes referred to as an intrinsic change.45 Bernibeu et al.46 reported that there is a direct link between learning and learned behavior and neurotransmitters. At the synaptic level in the Aplysia, serotonin may be one of the mediators in behaviorally learned changes in stimulus response.47 Many simple neural circuits receive input from superordinate circuits, and learning-induced plasticity may occur at the superordinate level. In eyelid conditioning in mammals, it appears that the site of plasticity necessary for the conditioning is in a higher order circuit in the cerebellum.48,49 Considered in relation to the reflex arc, this is sometimes called an extrinsic change.
Extrapolating from this animal research, it is believed that implicit memory provides an empirical basis for postulating the mechanisms by which people might come to have memories of which they are unaware, that are based on early emotional experiences and that act in an unconscious way to guide behavior, expectations, and impressions relating to relationships.40,50 In order for these interactions to exert long-lasting influences on behavior, they must be encoded in the type of memory that is believed to be implicit in nature.51 The existence of implicit memory represents an important part of the theory of maladaptive affective learning characteristic of psychiatric disorders.
The assertion that affective attunement can result in changes in neural structure and is critical to the long-term stability of the emotional life of the individual clearly implies that early affective interactions are permanently encoded in memory.40 Central to the memory model of psychotherapy is attachment research demonstrating that early patterns of responsiveness exhibited by attachment figures can have far-reaching consequences during neural development.
The capacity for affective self-regulation is postulated to be minimal at birth and is thought to be enhanced by exposure to experiences of appropriate attachment relationships. Notable deficits in the adequacy of the attachment relationship result in disorganized neurobehavioral repertoires and impaired capacity for internal self-regulation.40 The inability to self-regulate is clinically manifested by an inability to self-soothe or to modulate anger.
Rosenblum et al.52 demonstrated this model of attachment with respect to the rhesus monkey. Exposure of monkeys in early life to an inadequate attachment figure engendered permanent vulnerability to anxious and depressed states and to poor social functioning. Additionally, there are intriguing similarities between the behaviors of animals that exhibit the isolation syndrome and the behaviors of character-disordered patients.53 In these personality disorders, a particular consequence of early attachment failure is an exaggerated and prolonged reliance on external sources of regulation.
Because early attachment memories may be implicitly unconscious—rather than repressed, as stated in classic analytic literature—the memory model of psychotherapy offers an interesting psychobiologic perspective. Amini et al.40 posit that psychotherapy is not merely a conversation or an intellectual exchange of words and ideas. Instead, it is an attachment relationship, which is a physiologic process capable of regulating neurophysiology and altering underlying neural structure.
The psychobiologic conceptualization of psychotherapy offered by Amini and colleagues is that psychotherapy may function as an attachment relationship whose purpose is to regulate affective homeostasis and restructure attachment-related implicit memory. Therefore, when patients participate in psychotherapy, they first of all activate the implicit memory system and then engage the mechanism whereby implicitly stored material can be modified.
Whether learning-induced anatomical changes in the nervous system are necessary for storage of long-term memory has been discussed by several authors, including Morris,5 Greenough et al.,54 and Martinez and Derrick.55 The search for a model of the neural mechanisms of memory is based on discoveries that training or differential experience leads to significant changes in brain neurochemistry, anatomy, and electrophysiology. Consequently, it is generally accepted that psychotherapy is a powerful intervention that directly affects and changes the brain.
Landmark investigations by Baxter et al.56 and Schwartz et al.57 demonstrated that psychotherapy and fluoxetine produced decreased cerebral metabolic rates in the right caudate nucleus. The two interventions appeared to have similar physiological effects. Additionally, van der Kolk58 conducted clinical trials of patients diagnosed with posttraumatic stress syndrome. After treatment with eye movement desensitization and reprocessing (EMDR), they reported increased prefrontal metabolism and decreased limbic system activation in single-photon emission computed tomography (SPECT) scans of the patients. The imaging data provide additional support for the findings of controlled clinical studies that have shown that EMDR may be effective. In EMDR, patients focus on their memories of a traumatic event while moving their eyes rapidly back and forth. When used several times, this exercise somehow appears to change the patient's cognition of the traumatic incident.
Viinamaki et al.59 used SPECT imaging before and after 1 year of dynamic psychotherapy in a patient with bipolar personality disorder and depression. Her imaging studies were compared at the same timepoints to those of a control bipolar patient who received no therapy and to those of 10 healthy control subjects. Both patients initially had decreased serotonin uptake in prefrontal cortex and thalamus compared with the healthy control subjects. After 1 year, the psychotherapy patient had normal serotonin uptake. These findings suggest that dynamic psychotherapy can affect serotonin metabolism.
Viinamaki's clinical work was preceded by evidence from animal studies establishing a relationship between serotonin and learning. Injecting serotonin into the anterior limbic cortex of dogs with low serotonin blood levels has been shown to reduce the effects of classical conditioning.60 A high level of conditioned and unconditioned reflexes in dogs was accompanied by low serotonin blood levels.61 Although Viinamaki's study has a number of methodological problems (such as a relatively small N ) and has not yet been replicated, the clinical significance of these findings is that most exposure treatments for anxiety disorder are based on the principles of classical conditioning.62 The underlying concept is that successful exposure therapy reduces patient anxiety through extinction of a classically conditioned reflex.
Joffe et al.63 reported that in treating patients with depression, those who responded to cognitive-behavioral therapy had significant decreases in measures of thyroxine (T4), whereas those who did not respond to the therapy had increases in T4. Therefore, cognitive-behavioral therapy had an effect similar to that of antidepressant medication on the thyroid axis.
Brain plasticity refers to the brain's ability to change structure and function. Brain plasticity is regarded as prerequisite for any long-lasting change in behavior, cognition, and emotion, and thus for all measurable psychotherapeutic effects.64 Of particular interest to the current review is Hebbian synaptic plasticity thought to underlie the experience-dependent changes in discrete brain areas consequent to psychotherapy. Although there is no evidence that psychotherapy directly stimulates brain plasticity, there are many reasons to assume this mechanism of action. It is postulated that in psychotherapy, learning through exploration gives increased synaptic field potentials in the perforant path synapses. This increase has been demonstrated in animal models in which a group of spatially trained adult rats showed faster spatial learning and higher basal dendritic spine density compared with two control groups. On the basis of the unchanged dendritic length and branching pattern, the results suggest the formation of new synapses.65 Greenough et al.54 note several observations that relate number of synapses and degree of dendritic branching to the amount and sites of learning or experience in the rat. Environmental stimulation has been found to increase brain weight (especially forebrain), cortical thickness, the number of glial cells, the glia-to-neuron ratio, and neuronal cell body and nucleus size, and to alter synaptic profiles by increasing dendritic branching, dendritic spine density, and the number of discontinuous synapses.66 Note that the changes occur in brain regions involved in the learned tasks; if learning is confined to one side of the brain, the synaptic and dendritic changes are also confined to that side. These findings were replicated by Kolb and Whishaw,67 who found that experience produces multiple, dissociable changes in the brain, including increases in dendritic length, increases (or decreases) in spine density, synapse formation, increased glial activity, and altered metabolic activity. For more comprehensive reviews of synaptic and activity-dependent plasticity in cortex as well as in other areas and systems, see Byrne,68 Brown et al.,69 Tyler et al.,70 Madison et al.,71 Tsumoto,72 Bear and Kirkwood,73 Bliss and Collingridge,22 Linden and Connor,74 Bear and Abraham,75 and Katz and Shatz.76
In psychotherapy, some kinds of affective dysfunction can be successfully managed by the re-establishment of homeostatic regulation through an improved and functioning attachment relationship. The major forms of psychotherapy (behavioral, cognitive, and psychodynamic) can be conceptualized as reflecting interventions at different levels of psychological organization.77 Depending on where in the brain one measures and on the kind of therapy or differential experience the individual has undergone, one may expect to find an increase in number of synapses and an increase in their size. Whether the brain shows plastic changes in response to a particular kind of experience depends on the brain region, the kind of experience, and special circumstances or treatments that enhance or impair plasticity.78 For example, behavioral psychotherapy focuses on dysfunction in simple forms of learning and memory (operant and associative conditioning) and related motor behavior.79 This paradigm involves brain structures such as the amygdala, basal ganglia, and hippocampus.80–82
Cognitive psychotherapy focuses on specific patterns of information processing. These symptoms are used as clues by which to define specific verbal thoughts and assumptions or schemata that account for both the symptomatic state and the psychological vulnerability to that state.83 Cognitive theory predicts the kinds of thinking patterns encountered with each disorder. For example, anxious patients perceive danger in situations that are not dangerous; depressed patients see evidence of personal defect in situations that offer no objective reasons for self-deprecation. Paranoid patients may misconstrue situations in terms of being deceived or attacked.84,85 According to cognitive theory, negative cognitions play a pivotal role in the development and maintenance of the psychopathological state. Thus, the initial objectives of treatment are to teach the patient to recognize these cognitions. Thereafter, the patient learns to evaluate and modify such thinking patterns. Therapy then focuses on the identification and modification of dysfunctional attitudes that are inferred from the patient's stereotyped thinking and behavioral patterns. These schemata are values derived from early life experiences and support moment-to-moment thinking patterns.86 Putative brain areas include the neocortex, specifically the frontal cortex.
Psychodynamic psychotherapy has as its central focus interpersonal representation: a set of expectations about self, others, and their relationship that organizes related affect, thought, and behavior.87–92 The neuropsychological underpinnings of interpersonal representations probably involve complex neurocircuitry incorporating lateralized cerebral hemispheres and subcortical areas.93
The concept of lateralization was documented by Tucker's94 observations that the left cerebral hemisphere specializes in verbal, linear, and analytic functions (in right-handed individuals) and that the right cerebral hemisphere specializes in intuitive, spatial, and holistic activities. Tucker also reported that negative emotions are more often lateralized to the left hemisphere. Gottschalk et al.95 endorsed the significance of correlation between the left temporal lobe and social alienation/personal disorganization scores derived from analysis of cerebral glucose metabolic rates in wakeful subjects. Although it may be premature to conclude that the left temporal lobe is a cerebral focus for certain types of psychopathological processes that can be ameliorated by psychodynamic psychotherapy, these findings encourage further exploration of the mechanism of psychotherapeutic action.
Many psychiatric disorders involve the inability to control fear: anxiety disorders, phobias, and posttraumatic stress disorder. Neurons in a number of brain regions undergo physiological changes during aversive classical conditioning.96 Studies of the physiology of learning have suggested that many brain regions show physiological changes during learning. Thus, it is not surprising that plasticity has been found throughout the fear conditioning circuitry: in the auditory thalamic areas that project to the amygdala;97–100 in the auditory cortex;101,102 in the lateral, basolateral, and central nuclei of the amygdala;103,104 and in the lateral hypothalamus.105
The amygdala is involved in both the acquisition and the expression of fear conditioning.106–112 Numerous studies in both rats and humans indicate the importance of the amygdala in the acquisition and expression of learned fear. The identification of the amygdala as an essential neural substrate for fear conditioning has permitted neurophysiological examinations of synaptic processes in the amygdala that may mediate this conditioning. Recent studies of synaptic transmission and plasticity in the amygdala shed light on the relationships of these processes to aversive learning and memory.113 For example, Adamec114 reports that lasting changes in anxiety-like behavior may be produced in several ways. These include partial limbic kindling, injection of a beta-carboline, and brief, noninjurious exposure of rodents to cats (predator stress).
Emerging data support the idea that behavioral changes following induced anxiety-like behavior may model anxiety associated with posttraumatic stress disorder. These mechanisms likely involve initiation of long-term potentiation by NMDA receptors and prolongation of LTP by creatinine kinase (CCK) receptors. To the extent that response to the stressors mimics the symptoms of PTSD, the data implicate NMDA-mediated processes in the creation of what van der Kolk115 has called permanent emotional memories in PTSD. Because the blocking of CCK receptors before and after the stressor prevents lasting increases in anxiety-like behavior, interventions to block CCK receptors shortly after a traumatic stressor might be efficacious in mitigating the permanence of these emotional memories. Transmitter systems involved in neural plasticity underlie increased anxiety and defense. This high level of physiological arousal and inability to regulate autonomic responses to both internal and external stimuli markedly diminish the individual's ability to make use of feelings as important signals.58 To compensate for the hyperarousal, the victim attempts to avoid feelings and situations reminiscent of the trauma by shutting down psychologically and becoming numb to stimuli.
Correlates of gross neuronal activity can be assessed by observing which territories are relatively more or less active. This is accomplished by imaging studies that reflect cerebral metabolism or blood flow, since both are tightly coupled to neuronal activity. Functional brain imaging techniques help characterize how multiple mental operations and the spatially distributed processes that subserve them work in concert to produce normal human emotions and how their dysfunction produces disorders of fear.97 The explicit memory system is implicated in the encoding of facts related to the trauma. Activity of the amygdala associated with long-term storage of memories of emotionally arousing versus neutral events viewed on film has been measured by PET imaging.116 In that study, emotional films enhanced glucose metabolism in the right amygdala.
The amygdala is probably responsible for extinction of sensory and cognitive associations to original trauma and activation of traumatic memories. This would explain increased activity of the amygdala in PTSD. Extinction is the process through which the strength of a conditioned response is weakened by repeated exposure to the conditioned stimulus in the absence of the unconditioned stimulus. Considerable evidence suggests that extinction of conditioned fear does not occur passively; that is, the memory persists in the absence of explicit extinction training, and when extinction occurs it is not passive forgetting but instead is an active process involving new learning.117 Data from studies of emotional conditioning in rats suggest that neocortex, although not necessary for the acquisition of conditioned fear, is necessary for extinction once fear conditioning has been acquired.118 In this way, psychotherapy deals with the deliberate act of recollecting explicit memories.43 Theoretically, working with emotional memories in psychotherapy by asking patients to recall and associate activates the amygdala. The patient participates in an active process of new learning, resulting in extinction of sensory and cognitive associations to the original trauma. Thus, undoing traumatic memories is not passive forgetting. It is an active process involving new learning, probably as the orbitofrontal cortex becomes more activated.
Obsessive-compulsive disorder (OCD) is a neurobiological disorder caused by an imbalance in the brain neurotransmitter serotonin. Jenike et al.119 presented results from a magnetic resonance imaging (MRI)-based brain segmentation study demonstrating that patients with OCD had significantly less total white matter, greater total neocortex, and greater opercular volumes. Functional neuroimaging studies of OCD have primarily implicated orbitofrontal and anterior cingulate cortex, as well as striatum, showing hyperactivity during neutral states that is accentuated during symptom provocation.120 PET scans have shown that patients with OCD have significantly elevated glucose metabolic rates in both cerebral hemispheres, the heads of the caudate nuclei, the orbital gyri, and the orbital gyri relative to the ipsilateral hemisphere.121 Glucose metabolic rate is closely tied to neuronal functioning. Therefore, it is clear that at some level, obsessive-compulsive disorder is mediated through serotonin-based neurochemical processes of the brain.
Behavioral therapy for OCD consists of exposure and response prevention interventions. Exposure consists of asking the patient to interact with stimuli that result in the obsession or ritualistic behavior. Response prevention consists of delaying, diminishing, or discontinuing anxiety-reducing rituals. The response prevention principle is based on the idea that anxiety slowly decreases when the performance of the ritual is blocked. When anxiety reduction no longer depends on the rituals, the ritual compulsion will be extinguished. Current research in animals and humans has demonstrated that neurochemical changes due to behavioral interventions are similar to the effects of pharmacologic therapies.122
Especially in the treatment of OCD, there is overwhelming evidence that behavior therapy is capable of producing changes in brain chemistry that may be similar to the effects of serotonergic medications.123 Investigators found that 3 weeks of intensive behavior therapy had an effect on serotonin activity as measured by imipramine binding and changes in platelet serotonin level.124 owever, the major neurobiological evidence comes from two recent studies using PET conducted by Baxter et al.56 and replicated by Schwartz et al.57 The findings of Baxter et al.56 provided evidence that glucose metabolic rates in the right head of the caudate nucleus change when OCD is treated successfully with either fluoxetine or behavior therapy. Schwartz et al.57 found that after successful behavior modification, there is significant bilateral decrease in glucose metabolism in the caudate nucleus. These studies are significant because they confirm that behavior therapy can cause neurochemical changes in the brain.
Chronic schizophrenia is characterized by change in the normal cortical structure of the brain, asymmetric reduction, and often ventricular enlargement.125–129 The debate continues as to whether these anomalies occur early in development or represent an active, progressive process continuing after the onset of psychosis. DeLisi130 proposes that the underlying basis for the neuropathology of schizophrenia resides in the periodic activation of a defective gene or genes that determine the rate of cerebral growth. This process causes subtle cortical maldevelopment prenatally and through early childhood, is activated again during adolescent pruning of neurons, and again during the gradual aging process in the brain throughout adulthood.
Structural change in the hippocampal formation has become popular as a proposed neurobiological substrate for schizophrenic disorders. Hippocampal function is particularly sensitive to neurochemical modulation, and the expression of monoamine receptors in the temporal lobe is altered in schizophrenic patients.131 An attractive proposal is that behavioral plasticity, in the form of long-term potentiation of hippocampal synaptic transmission, may mediate transient psychosis.132 Moreover, the disturbed hippocampal neuroarchitecture found in schizophrenic patients may be susceptible to potentiation and may be dysfunctional to the degree that delusions and hallucinations develop.132
It has been hypothesized that delusions and related phenomena arise from some disturbance of the cognitive machinery of belief, memory, and their associated neural representations.18,133 Delusions are characterized by an aberrant belief plus a gradual spreading of the belief system to incorporate new information as the disease progresses. It is as if the associations between ideas that form the framework of a normal belief system had become so immutable that the usual processes of assimilation and incorporation of new information can no longer operate. This observation leads to the hypothesis that mechanisms of synaptic plasticity may be abnormal in these patients.
The increasing empirical support for a limbic substrate of schizophrenic symptomatology is based on three factors:
Defect or negative symptoms, which apparently do not respond well to neuroleptic treatment134 but do respond to atypical antipsychotics, have been proposed to reflect frontal lobe dysfunction.135 These symptoms may be attributable to changes in prefrontal cortex metabolism that are secondary to limbic pathology.136 Conceivably, then, hippocampal dysfunction could be directly responsible for positive symptoms and indirectly involved, via frontal projections, in the negative symptoms.132
The implications of this theoretical framework support psychotherapeutic interventions in the treatment of schizophrenia. There is a growing body of literature promoting cognitive-behavioral therapies aimed at the modification of delusions. Psychological approaches to treatment-resistant symptoms of psychosis all emphasize the importance of normalization of even the most bizarre personal experiences. Although antipsychotic drugs are generally successful at attenuating behavioral dysfunction in schizophrenic patients, prevention of further accrual of hippocampal damage would also be beneficial. Given the prominent role of stress and related hormones in the degeneration of hippocampal pyramidal neurons, it may be advantageous for at-risk persons to minimize the level of environmental stress and maximize their capacity to cope adaptively with stressors. Hogarty et al.137 devised a disorder-relevant and disorder-specific therapy for schizophrenia. Four key fundamentals of this personal therapy are 1) theoretical grounding in the stress-vulnerability model; 2) consideration of stress-related dysregulation as proximate to symptom exacerbation; 3) sequential use of interventions of graded complexity based on the patient's stage of recovery; and 4) flexible use of a range of therapeutic techniques to accommodate the individual needs, deficits, and preferences of patients with this heterogeneous disorder. Goals include developing self-recognition of the links between stresses, maladaptive responses, the reactions of others, and symptoms.
Supporting Hebb's 1949 hypothesis of use-induced plasticity of the nervous system, learning has been found to alter the structure and function of nerve cells and their connections. Further studies revealed changes in cortical thickness, size of synaptic contacts, number of dendritic spines, and dendritic branching, constituting a mechanism by which psychotherapy facilitates changes in the permanent storage of memory. Psychotherapy alters the neurochemistry and physiology of the brain by providing a stimulus that leaves a memory trace. Neural network research and memory research have proceeded in parallel to elucidate the theoretical properties of ideal neural networks and the actual properties of information storage in the brain. Long-term potentiation serves as a model of activity-dependent synaptic plasticity proposed to underlie memory.
The study of learning and memory not only brings us closer to understanding how psychotherapy produces emotional and behavioral changes in patients, but also gives us new perspectives on psychiatric disorders involving the inability to control fear, obsessions, compulsions, and delusions. We anticipate that more direct evidence of links between neurobiology and the psychotherapeutic processes will be available as neurobiologists and psychotherapists collaborate on future studies.