Review of the Brain Substrates of FEAR and Anxiety

The experiences of fear and anxiety reflect the actions of complex, poorly understood emotional systems of the brain, for which no common neural denominator—no generally accepted mechanistic explanation—yet exists. In order to make sense of how these substrates are functionally organized, we must currently simplify to a substantial extent. In any event, the capacity of organisms to respond effectively to threats to survival was such an important evolutionary issue that it was not simply left to individual learning. As already noted, the study of the evolved neurochemistries of these mechanisms provides an optimal strategy for yielding new, clinically useful information.

The trajectory of one major fear system (e.g., Fig. 16.1) courses between basolat-eral and central regions of the amygdala [and other higher brain zones such as the bed nucleus of the stria terminalis (BNST) and perhaps lateral septal area] and projects









Figure 16.1. This schematic of the FEAR system depicted on a sagittal section of the rat brain (the background of which highlights high-density acetylcholine esterase staining in black). This transhypothalamic executive systemforFEARorchestrates many cognitive,affective,behavioral, hormonal, and physiological changes that characterize various fearful states. The executive circuit is a two-way avenue of communication between the BNST and the central regions of the amygdala, which transmits information caudally primarily by the ventral amygdalofugal pathway and the mesencephalic periaqueductal or central gray (CG). This circuit courses through the anterior and medial hypothalamic areas of the diencephalon, where it is especially easy to elicit fearful behaviors (both freezing and flight) using ESB. There are multiple entry and exit points in this circuit (as depicted by the branching bubbles) that synchronize the many brain and bodily processes that must be concurrently influenced when an animal is threatened. Anatomical designations are as follows: AHA, anterior hypothalamic area; CA, caudate nucleus; Ce, cerebellum; CG, central gray; BN, bed nucleus of the stria terminalis; FC, frontal cortex, Hc, hippocampus; LC, locus coeruleus; MH, medial hypothalamus; NA, nucleus accumbens; Th, thalamus; V, motor nucleus of the trigeminal nerve; and VII, nucleus of the facial nerve. [This figure is reprinted from Panksepp (1998a), Affective Neuroscience, with the kind permission of Oxford University Press.]

downward through the anterior and medial hypothalamus to the periaqueductal gray (PAG) of the midbrain and adjacent tegmental fields. Henceforth, this neural trajectory will be called the FEAR system, to distinguish it from other, less well-understood, negative affective systems, including those that precipitate panic attacks, social disgust, and separation distress, to name a few. Electrical and chemical stimulation along the circuits generates fearful states along with many fear-related behaviors and autonomic changes in both experimental animals and humans (Depaulis and Bandler, 1991; Panksepp, 1985, 1998a, 2000).

It has long been known that one can arouse coherent freezing, flight, and other defensive responses, as well as associated autonomic changes, with electrical and chemical stimulation along this extended circuit. Animals readily learn to turn off this type of brain stimulation, even though under some testing conditions they do not exhibit efficient learned avoidance of such apparently aversive central states. This problem—the failure to obtain certain types of avoidance behavior—permitted investigators to devalue affective issues, claiming that the striking emotional behaviors were sham emotions with no experiential contents. However, the failure of avoidance behavior to become manifest appears to have arisen from straightforward methodological problems such as the failure to use sensitive measures (e.g., place avoidance paradigms), and perhaps from neural circuit "quirks" such as how exteroceptively driven learning processes interface with primitive emotional systems (Panksepp et al., 1991).

It is now clear that an enormous amount of learning can influence the FEAR system through higher limbic areas (most prominently various amygdaloid-hippocampal-temporal and frontal cortical regions). This conditioning can emerge, as described by LeDoux (1996, 2000), through short-loop sensory inputs such as those arising from thalamus (the so-called low road to fear conditioning) as well as higher sensory-perceptual processing (the so-called high road to fear). In addition there is an evolutionarily created royal road to understanding fear—a FEAR circuit that descends from amygdala, BNST and other telencephalic areas that converges on the PAG (Fig. 16.1 and 16.2) and coordinates the many evolved behavioral, physiological, and primitive affective aspects of fear (Panksepp, 1982, 1990). The importance of such primitive FEAR circuitry in conceptualizing the nature of human anxiety has been affirmed by recent brain imaging studies (Chapter 2 and Damasio et al., 2000) and is gradually gaining acceptance in behavioral neuroscience (Rosen and Schulkin, 1998). Only the higher amygdalar reaches are currently well recognized in psychiatry (Charney et al., 1999; Johnson and Lydiard, 1995) and human experimental psychology (Ohman and Mineka, 2001). The full extent of the circuitry provides the optimal approach for detailing the underlying causes of anxiety and is a clarion call for psychiatry and other mind sciences to reinvest in animal brain research.

This parsimonious view—that affect is largely a subcortical brain function shared homologously with other mammals—which entails no need for cortical re-representation or readout of affect, may require a neural conceptualization of a primordial "core self" (Damasio, 1999; Panksepp, 1998 a,b). Many higher cortical regions of the brain are essential for regulating (e.g., sustaining, dampening, as well as restructuring) emotions, but, to the best of our knowledge, those higher brain regions do not have the intrinsic capacity to create the primal valenced quality of affective experience. Indeed, many of the higher regions, in their important regulatory roles, may actually dampen the affective features (e.g., consider that young children with immature cortical controls generally feel affect more intensely than adults, even though they do not yet have the

Figure 16.2. General overview of the anterograde and retrograde connectivities of the mesencephalic defense systems (FEAR and RAGE) as they converge on the periaqueductal gray (PAG) of the mesencephalon. The two emotions are so closely integrated that current anatomical techniques cannot easily discriminate between the two. Putative functional integrations are indicated. [This figure is reprinted from Panksepp (1998a), Affective Neuroscience, with the kind permission of Oxford University Press.]

Figure 16.2. General overview of the anterograde and retrograde connectivities of the mesencephalic defense systems (FEAR and RAGE) as they converge on the periaqueductal gray (PAG) of the mesencephalon. The two emotions are so closely integrated that current anatomical techniques cannot easily discriminate between the two. Putative functional integrations are indicated. [This figure is reprinted from Panksepp (1998a), Affective Neuroscience, with the kind permission of Oxford University Press.]

cognitive capacities to sustain affective states through ruminations). In this context, it is noteworthy that there are rich anatomical connections from frontal cortical areas to the PAG (Holstege et al., 1996), and higher cognitive processes can regulate the arousal of these lower brain regions (Tracey et al., 2002). Such "mental" dampening and restructuring of emotional arousal is surely more refined in humans than any other species.

It may well be that most anxiety disorders arise from constitutional shifts in the sensitivity of these core systems as opposed to merely the sustained incoming impact from learned stimuli. In other words the FEAR system, like all emotional systems, may sensitize in response to chronic overarousal (Adamec and Young, 2000; Maren, 1999).

When the arousal of such systems becomes free floating, disconnected from external perceptions, various pathological states emerge, ranging from generalized anxiety to posttraumatic stress disorders (PTSD). During maturation, the genetically provided and learned aspects of fear become inextricably blended. The most useful new information for biological therapeutics will probably be derived from a better understanding of the unconditional neurochemical substrates of fear and a study of how such systems become sensitized, and overresponsive, rather than through a study of how unconscious fear reflexes are linked to neutral stimuli through classical conditioning. Such conditioning studies may be more germane to understanding the emergence of specific phobias, providing ideas on how therapists may be better able to de-condition acquired fears (Ohman and Mineka, 2001).

The FEAR system, as all other major emotional systems of the brain, is hierarchically arranged. Higher brain regions harvest perceptual/cognitive information (the periamygdaloid cortex of the temporal lobe); the middle hypothalamic zones control autonomic/hormonal responses that bias fear in reference to activities of homeostatic detectors that monitor bodily needs (e.g., animals will be less afraid of approaching resources in potentially dangerous situations if they are hungry); the critical lower zones in the PAG and surrounding midbrain orchestrate the integrated behavioral/bodily responses, with most of the individual response elements being situated in yet lower regions of the brainstem (Fig. 16.1). The more caudally such electrical stimulation of the brain (ESB) is imposed, the more rapid and intense is the evoked fear response and to all appearances (including human subjective reports) the resulting affective experiences. Responses evoked from higher brain areas (e.g., amygdala) are critically dependent on the integrity of the lower brain regions (e.g., PAG) but not vice versa (Panksepp, 1998a).

Of course, the arousal of this system has widespread consequences on the brain, partly through direct interactions with higher brain areas such as the frontal and temporal cortices. There are also indirect consequences through interactions with various general-purpose cholinergic and biogenic-amine (e.g., norepinephrine and serotonin) arousal/attention systems arising from the brain stem. These effects surely modulate the quality of the resulting subjective experiences. Anxious ruminations require those higher brain areas, but it is important to reemphasize that there is no evidence that neocortical tissue has the intrinsic capacity to generate affective states. The cortico-cognitive realms parse and re-represent primal feelings through their capacity to make finer and finer discriminations and distinctions. Thus, the position that affect is largely generated subcortically does not deny that primitive emotional dynamics can be used as tokens of information in the deliberative systems of the neocortex that regulate and fine-tune emotional arousal.

There will be many ways to regulate fears, but a reasonable working hypothesis is that the most powerful and clinically useful effects will be those that act directly on the specific neurochemistries of the FEAR system. Pharmacological dampening of this system facilitates calmness. So far this has been achieved with rather general modulators of brain gamma-aminobutyric acid (GABA), norepinephrine, and serotonin activities. It will soon be achieved through our increasing knowledge of more specific chemistries such as the neuropeptides (Chapter 21), as well as neurosteroids that can modulate GABA receptors (Heilig, 1995; Paul and Purdy, 1992) and new biogenic amine GABA facilitators (Skolnick et al., 2001).

Before proceeding to therapeutic issues, let us briefly consider the abundance of existing animal models for studying various types of anxiety. Parenthetically, the large variety of preclinical measures of anxiety may reflect the diversity of psychobiological states of trepidation that may exist within the brain. We might recall that the complex hierarchy of anxiety Freud advocated consisted of (i) fear of loss of object, (ii) fear of loss of love of the object, (iii) castration anxiety, and (iv) superego anxiety [as detailed in Inhibitions, Symptoms and Anxiety (volume 20) of the Standard Edition as well as in the New Introductory Lectures (volume 22)]. Such issues cannot be studied in animal research, and future taxonomies of anxiety should be based as much on neurobiological data as on more theoretical psychological perspectives.

0 0

Post a comment