The smell of danger: a behavioral and neural analysis of predator odor-induced fear

The odors of predators used in animal models provide, in addition to electric footshock, an important means to investigate the neurobiology of fear. Studies indicate that cat odor and trimethylthiazoline (TMT), a synthetic compound isolated from fox feces, are often presented to rodents to induce fear-related responses including freezing, avoidance, stress hormone and, in some tests, risk assessment behavior. Furthermore, we report that different amounts of cat odor impregnated on small-, medium-, or large-sized cloths impact the display of fear-related behavior when presented to rats. That is, rats exposed to a large cat odor containing cloth exhibit an increase in fear behavior, particularly freezing, which remains at high levels in habituation tests administered over a period of 7 days. The large cloth also induces a long-lasting increase in avoidance behavior during repeated habituation and extinction tests. A review of the brain regions involved in predator odor-induced fear behavior indicates a modulatory role of the medial amygdala, bed nucleus of the stria terminalis, and dorsal premammillary nucleus. In addition, the basolateral amygdala is involved in fear behavior induced by cat odor but not TMT, and the central amygdala does not appear to play a major behavioral role in predator odor-induced fear. Future research involving the use of predator odor is likely to rapidly expand knowledge on the neurobiology of fear, which has implications for understanding fear-related psychopathology.     

Inactivation of the lateral septum blocks fox odor-induced fear behavior

A number of lesion studies have shown that the lateral septum plays an important role in the modulation of innate fear. Furthermore, an increased c-fos expression in the lateral septum was demonstrated after exposure to natural predator odors and 2,3,5,-trimethyl-3-thiazoline (TMT), a component of fox odor. This study investigates, on a behavioral level, whether the lateral septum plays a role in TMT-induced fear. Temporary inactivation of the lateral septum by local muscimol injections clearly blocked TMT-induced fear behavior but had no effect on behavior in a controlled condition. This indicates that the lateral septum is important for the processing of TMT-induced fear and suggests that the lateral septum is also involved in fear behavior induced by natural predator odors.     

Chronic corticosterone administration does not potentiate unconditioned freezing to the predator odor, trimethylthiazoline

Chronic high levels of corticosterone (CORT) are known to facilitate learning and memory of aversive events. Whether this effect of chronic CORT also generalizes to unconditioned or unlearned fear behavior is not known. The present study investigated whether high levels of chronic CORT enhance unconditioned fear to a predator odor, trimethylthiazoline (TMT), an innate fear stimulus to rodents. TMT induces a dose-related freezing response, a prototypical behavior to fearful stimuli, in rats. The first experiment demonstrated that dose-related freezing to repeated exposures of TMT does not habituate, sensitize or produce contextually conditioned fear, and therefore can be used to measure the effects of chronic CORT on unconditioned fear to repeated exposures of TMT. In Experiment 2, 21-day release corticosterone pellets (200mg) were implanted subcutaneously in male, Sprague-Dawley rats. Control rats received sham implantation. On days when TMT was not present, chronic CORT rats froze significantly more than sham rats. However, while TMT-induced freezing in both chronic CORT and sham rats, freezing during exposure to TMT was not further enhanced in chronic CORT rats. Thus, chronic CORT appears to increase fear as measured by freezing, possibly by enhancing vigilance, but does not facilitate fear behavior induced by the innate fear stimulus, TMT.     

2,3,5-Trimethyl-3-thiazoline (TMT), a component of fox odor - just repugnant or really fear-inducing?

During recent years, an increasing number of studies have used 2,3,5-trimethyl-3-thiazoline (TMT), a component of fox feces, as a stimulus to induce fear in predator naive rodents. The use of TMT is controversially discussed: There are some clear advantages of TMT against natural predator odors (e.g. stimulus intensity can be better controlled) but also still some open questions and objections regarding TMT. The aim of the present article is to discuss four often mentioned objections against TMT: (1) In some cases, TMT failed to produce fear behavior, (2) TMT is rather a noxious than a fear-inducing stimulus, (3) TMT does not support fear conditioning, and (4) there are different neural pathways processing natural predator odors and TMT. We summarize data showing different sensitivity to TMT in different rat strains. Then, new data are presented showing that TMT concentrations which are not avoided by rats induce fear behavior, and that concentrations of TMT and of the control odor butyric acid, which are similarly avoided, are totally different in their ability to induce fear behavior. Furthermore, we summarize and discuss data showing that fear conditioning to a TMT-paired context is possible and that there is an overlap between the neural basis for TMT- and cat odor-induced fear behavior. In conclusion, the recent data do not support the idea that TMT is simply a noxious stimulus which non-specifically induces fear behavior. Therefore, TMT is still a viable alternative stimulus to natural predator odors to investigate effects of predator odors on behavior.     

Analysis of behavioral constraints and the neuroanatomy of fear to the predator odor trimethylthiazoline: a model for animal phobias

Specific phobias, including animal phobias, are the most common anxiety disorders, and have a strong innate and genetic component. Research on the neurobiology and environmental constraints of innate fear of predators in rodents may be useful in elucidating mechanisms of animal phobias in humans. The present article reviews research on innate fear in rats to trimethylthiazoline (TMT), an odor originally isolated from fox feces. TMT induces unconditioned freezing and other defensive responses that are regulated by the dose of TMT and the shape of the testing environment. Contextual conditioning induced by TMT occurs, but is constrained by the environment. Lesion studies indicate the amygdala circuitry subserving fear conditioning is not necessary for unconditioned fear to TMT. Additionally, a medial hypothalamic defensive circuit also appears not necessary for unconditioned freezing to TMT, whereas circuits that include the medial nucleus of the amygdala and the bed nucleus of the stria terminalis are essential. The importance of these findings of innate predator odor fear in rodents to animal phobias in humans is discussed.     

Inactivation of the prelimbic cortex enhances freezing induced by trimethylthiazoline, a component of fox feces

Previous research has demonstrated that the rodent medial prefrontal cortex (mPFC) is critical for the expression of unconditioned defense behaviors. The prelimbic (PL) and infralimbic (IL) cortices comprise the majority of the mPFC, but the role of these regions in mediating unconditioned defense behaviors is not well understood. In order to address this, we temporarily inactivated the PL or IL and documented the effects of these manipulations on freezing induced by trimethylthiazoline (TMT), a component of fox feces, and center region avoidance in the open field (OF). PL inactivation enhanced TMT-induced freezing, but had no effect on OF behavior. IL inactivation had no effect on any behavioral measure. The results of this study are the first to demonstrate that the PL can have an inhibitory role with regard to unconditioned defense behavior. Further research is needed to define the parameters under which the PL inhibits unconditioned defense behavior.     

Corticosterone controls the developmental emergence of fear and amygdala function to predator odors in infant rat pups

In many altricial species, fear responses such as freezing do not emerge until sometime later in development. In infant rats, fear to natural predator odors emerges around postnatal day (PN) 10 when infant rats begin walking. The behavioral emergence of fear is correlated with two physiological events: functional emergence of the amygdala and increasing corticosterone (CORT) levels. Here, we hypothesize that increasing corticosterone levels influence amygdala activity to permit the emergence of fear expression. We assessed the relationship between fear expression (immobility similar to freezing), amygdala function (c-fos) and the level of corticosterone in pups in response to presentation of novel male odor (predator), littermate odor and no odor. CORT levels were increased in PN8 pups (no fear, normally low CORT) by exogenous CORT (3 mg/kg) and decreased in PN12 pups (express fear, CORT levels higher) through adrenalectomy and CORT replacement. Results showed that PN8 expression of fear to a predator odor and basolateral/lateral amygdala activity could be prematurely evoked with exogenous CORT, while adrenalectomy in PN12 pups prevented both fear expression and amygdala activation. These results suggest that low neonatal CORT level serves to protect pups from responding to fear inducing stimuli and attenuate amygdala activation. This suggests that alteration of the neonatal CORT system by environmental insults such as alcohol, stress and illegal drugs, may also alter the neonatal fear system and its underlying neural control.     

The role of the dorsomedial part of the prefrontal cortex serotonergic innervation in rat responses to the aversively conditioned context: behavioral, biochemical and immunocytochemical studies

In this study we have explored differences in animal reactivity to conditioned aversive stimuli using the conditioned fear test (a contextual fear-freezing response), in rats subjected to the selective lesion of the prefrontal cortex serotonergic innervation, and differing in their response to the acute painful stimulation, a footshock (HS--high sensitivity rats, and LS--low sensitivity rats, selected arbitrarily according to their behavior in the 'flinch-jump' pre-test). Local administration of serotonergic neurotoxin (5,7-dihydroxytryptamine) to the dorsomedial part of the prefrontal cortex caused a very strong, structure and neurotransmitter selective depletion of serotonin concentration. In HS rats, the serotonergic lesion significantly disinhibited rat behavior controlled by fear, enhanced c-Fos expression in the dorsomedial prefrontal area, and increased the concentration of GABA in the basolateral amygdala, measured in vivo after the testing session of the conditioned fear test. The LS animals revealed an opposite pattern of behavioral and biochemical changes after serotonergic lesion: an increase in the duration of a freezing response, and expression of c-Fos in the basolateral and central nuclei of amygdala, and a lower GABA concentration in the basolateral amygdala. In control conditions, c-Fos expression did not differ in LS and HS, na?ve, not conditioned and not exposed to the test cage animals. The present study adds more arguments for the controlling role of serotonergic innervation of the dorsomedial part of the prefrontal cortex in processing emotional input by other brain centers. Moreover, it provides experimental data, which may help to better explain the anatomical and biochemical basis of differences in individual reactivity to stressful stimulation, and, possibly, to anxiolytic drugs with serotonergic or GABAergic profiles of action.     

The dorsal periaqueductal and basolateral amygdala are necessary for the expression of conditioned place avoidance induced by semicarbazide stimulation of the dorsal periaqueductal region

The amygdala is critically involved in the regulation of unconditioned and conditioned reactions to threatening stimuli. It has been suggested that a neural circuit responsible for the production of defensive behavior elicited by the dorsal periaqueductal gray (dPAG) stimulation may project through ascending fibers to forebrain structures such as the basolateral complex of the amygdala (BLA). The present study evaluates the involvement of the dPAG and BLA in the mediation of unconditioned and conditioned responses organized in the dPAG using the open field and the conditioned place aversion (CPA) tests. In both tests, the intra-dPAG injections of semicarbazide (SEM), an inhibitor of the GABA synthesizing enzyme, was used as unconditioned stimulus (US). Using the open field test, we examine the effects of BLA inactivation with the GABA-(A) receptor agonist muscimol (MUS) on the unconditioned fear. We also investigated, through the CPA test, the effects of BLA and/or dPAG inactivation with MUS on the acquisition and the expression of the fear conditioned response. Our results showed that intra-BLA injections of MUS did not change the unconditioned fear elicited by dPAG injections of SEM. As for the CPA test, intra-BLA and intra-dPAG injections of MUS impaired the expression of CPA behavior induced by SEM injections into the dPAG. However, this inactivation of BLA did not impair the acquisition of the CPA behavior induced by injections of SEM into the dPAG. Altogether, these findings suggest that BLA does not participate in the mediation of unconditioned fear induced by dPAG chemical stimulation or in the acquisition of CPA in which aversive stimulation of the dPAG was used as US. In contrast, our results indicate that the activation of the dPAG and BLA is essential to the expression of the conditioned aversive response.     

Targeting ASIC1a reduces innate fear and alters neuronal activity in the fear circuit.

BACKGROUND: The molecular mechanisms underlying innate fear are poorly understood. Previous studies indicated that the acid sensing ion channel ASIC1a influences fear behavior in conditioning paradigms. However, these differences may have resulted from an ASIC1a effect on learning, memory, or the expression of fear. METHODS: To test the hypothesis that ASIC1a influences the expression of fear or anxiety independent of classical conditioning, we examined the effects of disrupting the mouse ASIC1a gene on unconditioned fear in the open field test, unconditioned acoustic startle, and fear evoked by the predator odor trimethylthiazoline (TMT). In addition, we tested the effects of acutely inhibiting ASIC1a with PcTx, an ASIC1a antagonist in tarantula venom. Our immunohistochemistry suggested ASIC1a is expressed in the bed nucleus of the stria terminalis, medial amygdala, and periaqueductal gray, which are thought to play important roles in the generation and expression of innate fear. Therefore, we also tested whether ASIC1a disruption altered c-fos expression in these structures following TMT exposure. RESULTS: We found that the loss of ASIC1a reduced fear in the open field test, reduced acoustic startle, and inhibited the fear response to TMT. Similarly, intracerebroventricular administration of PcTx reduced TMT-evoked freezing in ASIC1a(+/+) mice but not ASIC1a(-/-) mice. In addition, loss of ASIC1a altered TMT-evoked c-fos expression in the medial amydala and dorsal periaqueductal gray. CONCLUSIONS: These findings suggest that ASIC1a modulates activity in the circuits underlying innate fear. Furthermore, the data indicate that targeting the ASIC1a gene or acutely inhibiting ASIC1a suppresses fear and anxiety independent of conditioning.     

Dissociable roles of the central and basolateral amygdala in appetitive emotional learning

The amygdala is considered to be a core component of the brain's fear system. Data from neuroimaging studies of normal volunteers and brain-damaged patients perceiving emotional facial expressions, and studies of conditioned freezing in rats, all suggest a specific role for the amygdala in aversive motivation. However, the amygdala may also be critical for emotional processing in positive or appetitive settings. Using an appetitive Pavlovian approach procedure we show a theoretically important dissociation in the effects of excitotoxic lesions of the central nucleus and basolateral area of the amygdala, in the rat. Whilst central nucleus lesions impair appetitive Pavlovian conditioning, basolateral lesions do not. Together with other data, these results not only support the hypothesis that the amygdala is critical for appetitive as well as aversive learning, but are also consistent with amygdala subsystems subserving distinct aspects of emotional learning. Lesions of the dorsal or ventral subiculum were without effect on autoshaping, indicating the lack of involvement of hippocampal processing in this form of emotional behaviour and emphasizing further the neural specificity of the effects seen following central amygdala lesions.     

Distinct Fos expression in the brain following freezing behavior elicited by stimulation with NMDA of the ventral or dorsal inferior colliculus

The inferior colliculus (IC) is an important relay station for ascending auditory information to the medial geniculate nucleus (MGN) and temporal cortex. It has been reported that the ventral (ICv) and dorsal (ICd) regions of the IC are involved with the defensive reaction and audiogenic seizures, respectively. As freezing is the first response induced by stimulation of these IC nuclei with increasing doses of N-methyl-d-aspartate (NMDA), a question that arises is whether or not fear and audiogenic seizures generated at the IC level are interrelated processes. To address this issue, the Fos distribution in selected limbic structures following injections of NMDA into the ICv or ICd at freezing (7 nmol)- and escape (20 nmol)-producing doses was examined. Freezing behavior induced by intra-ICd NMDA caused an increase of Fos expression in the MGN, superior colliculus, dorsal columns of the periaqueductal gray and locus coeruleus while freezing induced by intra-ICv NMDA caused a significant Fos immunoreactivity in the prelimbic (PrL) and cingulate (Cg) cortices, basolateral and medial nuclei of the amygdala, ventrolateral periaqueductal gray, cuneiform nucleus and locus coeruleus. Escape behavior induced by NMDA injections into both nuclei caused a widespread Fos labeling in all limbic structures examined in this study. These results suggest that distinct circuits underlie the freezing behavior generated at the level of ICd and ICv. This is the first study to map Fos distribution associated with the stimulation of the ICv and ICd, regions supposed to be involved with fear and audiogenic seizures, respectively.     

Extinction of auditory fear conditioning requires MAPK/ERK activation in the basolateral amygdala

Whereas the neuronal substrates underlying the acquisition of auditory fear conditioning have been widely studied, the substrates and mechanisms mediating the acquisition of fear extinction remain largely elusive. Previous reports indicate that consolidation of fear extinction depends on the mitogen-activated protein kinase/extracellular-signal regulated kinase (MAPK/ERK) signalling pathway and on protein synthesis in the medial prefrontal cortex (mPFC). Based on experiments using the fear-potentiated startle paradigm suggesting a role for neuronal plasticity in the basolateral amygdala (BLA) during fear extinction, we directly addressed whether MAPK/ERK signalling in the basolateral amygdala is necessary for the acquisition of fear extinction using conditioned freezing as a read-out. First, we investigated the regional and temporal pattern of MAPK/ERK activation in the BLA following extinction learning in C57Bl/6J mice. Our results indicate that acquisition of extinction is associated with an increase of phosphorylated MAPK/ERK in the BLA. Moreover, we found that inhibition of the MAPK/ERK signalling pathway by intrabasolateral amygdala infusion of the MEK inhibitor, U0126, completely blocks acquisition of extinction. Thus, our results indicate that the MAPK/ERK signalling pathway is required for extinction of auditory fear conditioning in the BLA, and support a role for neuronal plasticity in the BLA during the acquisition of fear extinction.     

The fear circuit revisited: contributions of the basal amygdala nuclei to conditioned fear

The lateral nucleus (LA) is the input station of the amygdala for information about conditioned stimuli (CSs), whereas the medial sector of the central nucleus (CeM) is the output region that contributes most amygdala projections to brainstem fear effectors. However, there are no direct links between LA and CeM. As the main target of LA and with its strong projection to CeM, the basomedial amygdala (BM) constitutes a good candidate to bridge this gap. Consistent with this notion, it was reported that combined posttraining lesions of the basal nuclei [BM plus basolateral nucleus (BL)] abolish conditioned fear responses, whereas selective BL inactivation does not. Thus, we examined the relative contribution of BM and BL to conditioned fear using unit recordings and inactivation with muscimol microinfusions in rats. Approximately 30% of BM and BL neurons acquired robust responses to auditory CSs predicting footshocks. While most BL cells stopped firing at CS offset, BM responses typically outlasted the CS by ≥ 40 s, paralleling the persistence of conditioned fear after the CS. This observation suggests that BM neurons are not passive relays of rapidly adapting LA inputs about the CS. Surprisingly, independent inactivation of either BM or BL with muscimol did not cause a reduction of conditioned freezing even though an extinction recall deficit was seen the next day. In contrast, combined BL-BM inactivation did. Overall, there results support the notion that the basal nuclei are involved in conditioned fear expression and extinction but that there is functional redundancy between them.     

Fear-like biochemical and behavioral responses in rats to the predator odor,TMT, are dependent on the exposure environment

Several laboratories have reported that exposure to predator odor can result in stress-like effects in rodents. While some laboratories have reported fear-like alterations in behavior, other laboratories, including our own, have failed to consistently observe fearful behaviors in rats exposed to the predator odor TMT. One potential contributing factor to this discrepancy is the handling of the rat and its test environment. In the current report, we examine biochemical, endocrinological, and behavioral effects of TMT in two distinct open fields: one small, familiar, and dimly lit, while the other was large, novel, and brightly lit. Only exposure to TMT in the large, novel open field resulted in fearful behavior; however, no increase in dopamine turnover was noted compared to no odor and control odor rats. As expected, the different open fields resulted in some biochemical and behavioral differences, including more horizontal locomotion and less grooming, higher serum corticosterone, and increased dopamine turnover in the ventral prefrontal cortex in the large open field. Finally, compared to the same open field controls, TMT exposure elevated rat serum corticosterone levels in both open fields and dopamine turnover in the dorsal and ventral medial prefrontal cortex and amygdala of rats only in the small, familiar open field. These results indicate that the TMT-induced biochemical activation of may occur without detectable fearful behaviors and may indicate a mechanism that prepares the animal for the expression of a fearful response if additional provocative stimuli are present.     

Effects of cyclic adenosine monophosphate response element binding protein overexpression in the basolateral amygdala on behavioral models of depression and anxiety.

BACKGROUND: Chronic antidepressant administration increases the cyclic adenosine monophosphate response element binding protein (CREB) in the amygdala, a critical neural substrate involved in the physiologic responses to stress, fear, and anxiety. METHODS: To determine the role of CREB in the amygdala in animal models of depression and anxiety, a viral gene transfer approach was used to selectively express CREB in this region of the rat brain. RESULTS: In the learned helplessness model of depression, induction of CREB in the basolateral amygdala after training decreased the number of escape failures, an antidepressant response. However, expression of CREB before training increased escape failures, and increased immobility in the forced swim test, depressive effects. Expression of CREB in the basolateral amygdala also increased behavioral measures of anxiety in both the open field test and the elevated plus maze, and enhanced cued fear conditioning. CONCLUSIONS: Taken together, these data demonstrate that CREB expression in the basolateral amygdala influences behavior in models of depression, anxiety, and fear. Moreover, in the basolateral amygdala, the temporal expression of CREB in relation to learned helplessness training, determines the qualitative outcome in this animal model of depression.     

Impaired extinction of learned fear in rats selectively bred for high anxiety – evidence of altered neuronal processing in prefrontal‐amygdala pathways

The impaired extinction of acquired fear is a core symptom of anxiety disorders, such as post‐traumatic stress disorder, phobias or panic disorder, and is known to be particularly resistant to existing pharmacotherapy. We provide here evidence that a similar relationship between trait anxiety and resistance to extinction of fear memory can be mimicked in a psychopathologic animal model. Wistar rat lines selectively bred for high (HAB) or low (LAB) anxiety‐related behaviour were tested in a classical cued fear conditioning task utilizing freezing responses as a measure of fear. Fear acquisition was similar in both lines. In the extinction trial, however, HAB rats showed a marked deficit in the attenuation of freezing responses to repeated auditory conditioned stimulus presentations as compared with LAB rats, which exhibited rapid extinction. To gain information concerning the putatively altered neuronal processing associated with the differential behavioural response between HAB and LAB rats, c‐Fos expression was investigated in the main prefrontal‐amygdala pathways important for cued fear extinction. HAB compared to LAB rats showed an attenuated c‐Fos response to repeated conditioned stimulus presentations in infralimbic and cingulate cortices, as well as in the lateral amygdala, but facilitated the c‐Fos response in the medial part of the central amygdala. In conclusion, the present results support the notion that impaired extinction in high anxiety rats is accompanied by an aberrant activation profile in extinction‐relevant prefrontal‐amygdala circuits. Thus, HAB rats may represent a clinically relevant model to study the mechanisms and potential targets to accelerate delayed extinction processes in subjects with enhanced trait anxiety.     

Enriched environment impacts trimethylthiazoline‐induced anxiety‐related behavior and immediate early gene expression: critical role of Crhr1

It has been shown previously (Sotnikov et al., 2011 ) that mice selectively inbred for high anxiety‐related behavior (HAB) vs. low anxiety‐related behavior in the elevated plus maze differentially respond to trimethylthiazoline (TMT), a synthetic fox fecal odor. However, less is known about whether environmental factors can rescue these extreme phenotypes. Here, we found that an enriched environment (EE) provided during early adolescence induced anxiolytic effects in HAB (HAB‐EE) mice, rescuing their strong avoidance behavior induced by TMT. In a series of experiments, the contribution of maternal, juvenile and adolescent behavior to the anxiolytic effects elicited by EE was investigated. At the molecular level, using c‐fos expression mapping, we found that the activity of the medial and basolateral amygdala was significantly reduced in HAB‐EE mice after TMT exposure. We further analysed the expression of Crhr1, as its amount in the amygdala has been reported to be important for the regulation of anxiety‐related behavior after EE. Indeed, in situ hybridisation indicated significantly decreased Crhr1 expression in the basolateral and central amygdala of HAB‐EE mice. To further test the involvement of Crhr1 in TMT‐induced avoidance, we exposed conditional glutamatergic‐specific Crhr1‐knockout mice to the odor. The behavioral response of Crhr1‐knockout mice mimicked that of HAB‐EE mice, and c‐fos expression in the amygdala after TMT exposure was significantly lower compared with controls, thereby further supporting a critical involvement of Crhr1 in environmentally‐induced anxiolysis. Altogether, our results indicate that EE can rescue strong avoidance of TMT by HAB mice with Crhr1 expression in the amygdala being critically involved.     

The cerebellum: synaptic changes and fear conditioning.

In addition to coordinating movement, the cerebellum participates in motor learning, emotional behavior, and fear memory. Fear learning is reflected in a change of autonomic and somatic responses, such as heart rate and freezing, elicited by a neutral stimulus that has been previously paired with a painful one. Manipulation of the vermis affects these responses, and its reversible inactivation during the consolidation period impairs fear memory. The neural correlate of cerebellar involvement in fear consolidation is provided by a behaviorally induced long-term increase of synaptic efficacy between parallel fibers and a Purkinje cell. Similar synaptic changes after fear conditioning are well documented in the amygdala and hippocampus, providing a link between emotional experiences and changes in neural activity. In addition, in hotfoot mice, with a primary deficiency of parallel fiber to Purkinje cell synapse, short- and long-term fear memories are affected. All these data support the idea that the cerebellum participates in fear learning. The functional interconnection of the vermis with hypothalamus, amygdala, and hippocampus suggests a more complex role of the cerebellum as part of an integrated network regulating emotional behavior.