Psychopharmacology (1995) 121 : 38-56 © Springer-Verlag 1995

K. A. Miczek • E. M. Weerts • J. A. Vivian H. M. Barros , anxiety and vocalizations in animals: GABAAand 5-HT

Received: 4 February 1995 / Final version: 19 April 1995

Abstract A continuing challenge for preclinical promise other physiological and behavioral processes. research on is to capture the affective Antagonists at the benzodiazepine reverse the dimension that characterizes anxiety and aggression, effects of full on vocalizations, particularly either in their adaptive forms or when they become of when these occur in threatening, startling and dis- clinical concern. Experimental protocols for the pre- tressing contexts. With the development of antagonists clinical study of anxiolytic drugs typically involve the at 5-HT receptor subtypes, it can be anticipated that suppression of conditioned or unconditioned social and similar receptor-specificity can be established for the exploratory behavior (e.g., punished drinking or social effects of 5-HT anxiolytics. interactions) and demonstrate the reversal of this behavioral suppression by drugs acting on the benzo- Key words Anxiolytics - Benzodiazepines • 5-HT diazepine-GABAA complex. Less frequently, aversive agonists • Benzodiazepine receptors • 5-HIAA • events engender increases in conditioned or uncondi- Aggression " Punishment' Startle • Vocalizations - tioned behavior that are reversed by anxiolytic drugs Exploratory behavior • Social behavior • Dominance • (e.g., fear-potentiated startle). More recently, putative Maternal behavior. Pain. Defense " Stress • anxiolytics which target 5-HT receptor subtypes pro- Opiates. Ultrasounds duced effects in these traditional protocols that often are not systematic and robust. We propose ethological studies of vocal expressions in rodents and primates Introduction during social confrontations, separation from social companions, or exposure to aversive environmental The neurobiological and psychopharmacological study events as promising sources of information on the of aggression and anxiety is faced with the initial chal- affective features of behavior. This approach focusses lenge of how to relate clinical and social concerns with on vocal and other display behavior with clear func- pathological excesses to our understanding of these tional validity and homology. Drugs with anxiolytic behaviors' ontogenetic and phylogenetic adaptive ori- effects that act on the benzodiazepine-GABAa recep- gin. Clinicians and public health officials focus on an tor complex and on 5-HT~A receptors systematically excess or, alternatively, a lack of anxious and aggres- and potently alter specific vocalizations in rodents and sive behavior, while ethological analyses delineate the primates in a pharmacologically reversible manner; the adaptive nature of these behavior patterns. The most specificity of these effects on vocalizations is evident global definitions of aggression and anxiety attempt to due to the effectiveness of low doses that do not corn- encompass adaptive as well as pathological forms. Alternatively, psychiatric and ethological definitions K. A. Miczek ([]) • E. M. Weerts 1 • J. A. Vivian z . H. M. Barros are offered that emphasize either the pathological Research Building, Tufts University, 490 Boston Ave, Medford, extremes of aggression and anxiety or, alternatively, MA 02155, USA refer to their evolutionary roots and importance in the Present address: survival of the individual and the species. One impor- IDivision of Behavioral Biology, Johns Hopkins University tant dimension of aggressive and anxious behavior pat- School of Medicine, Hopkins Bayview Research Campus, terns that has eluded adequate quantitative assessment Baltimore, MD 21224-6823, USA in clinical, as well as preclinical research, is the affective 2Department of , 1301 Medical Science Research Building III, University of Michigan Medical School, Ann Arbor, or emotional nature of these activities. Here we pro- MI 48109-0632, USA pose that certain types of vocalizations may represent 39 aftbctive expressions that are accessible to quantitative Munizz et al. 1993). The historic Phineas Gage episode behavioral and neuropharmacological preclinical inves- points to traumatic and toxic insults to the brain as tigations. sources for aggressive episodes (Siegal and Mirsky 1994). The approach in the social sciences highlights Clinical and preclinical traditions environmental determinants of aggressive behavior and refers to aggression as antisocial behavior (e.g., The nosological situation is considerably more satis- Berkowitz 1993). Here, the causative factors are factory for "anxiety" than for "aggression" in clinical assigned to the distal living conditions, such as isolated, populations (Eichelman and Hartwig 1993). The impoverished or crowded housing, or to proximal Diagnostic and Statistical Manual of Mental Disorder triggers, such as noxious and aversive stimuli. For more IV (DSM IV) of the American Psychiatric Association than 50 years, the omission of scheduled reinforcement (1994) clearly distinguishes several types of anxiety dis- or "frustrative non-reward" has been a prime example orders, such as panic disordm, phobic disorder (e.g., for aggression as antisocial activity (Dollard et al. agoraphobia, social phobia, and simple phobia), obses- 1939). sive-compulsive disorder, post-traumatic stress disor- On the other hand, the ethological objectives in the der, generalized anxiety disorder (GAD), and organic study of aggressive behavior seek to understand the anxiety syndrome. By contrast, the preclinical proto- evolutionary origin and functional significance of these cols have not achieved this degree of differentiation. In behavior patterns (Lorenz 1966; Marler 1976). The addition, anxiety can appear concurrently with other adaptive significance of aggressive behavior in repro- pathological states such as alcoholism, depression dis- ductive contexts is apparent in dominance, territorial orders, schizophrenia and organic brain disorders. Here and maternal aggression when potential rivals or we focus on benzodiazepine and 5-HT~A anxiolytics intruders are repulsed and the brood is defended that are most effective in GAD (e.g., Lader and (Huntingford and Turner 1987; Archer 1988). Petursson 1981; Greenblatt et al, t983; De Vry 1995). Preclinical research strategies for psychopharmaco- By contrast, aggression remains unrecognized as a logical and neurobiological investigations of aggressive diagnostic category in the DSM IV (American and anxious behavior patterns are primarily guided by Psychiatric Association 1994), but aggressive behavior pharmacological validation, with homologies across appears as a symptom in several disorders that include: species remaining elusive. While aggressive and anx- mental retardation, attention deficit disorder, organic ious behaviors are characterized by affective or emo- mental syndromes, schizophrenia, delusional disorder, tional features, this dimension is rarely part of brief reactive psychosis, mood disorders, anxiety preclinical psychopharmacological research. In fact, disorder, disorders, impulse control disor- most research during the past three decades suggests der, adjustment disorder and personality disorders that anxiolytics with sites of action at the GABAa-ben- (Eichelman 1987). Mysteriously, while ultimately every zodiazepine receptor complex and, more recently, the behavior is a function of brain activity, the recent pub- 5-HTIA receptor profoundly and systematically modu- lic debate in the US reveals considerable reluctance in late animal behaviors that represent adaptations to the social science community" to relate aggressive behav- aversive environmental events. ior to any neural function. On the other hand, pre- clinical research of aggressive behavior has been guided mainly by the objectives and methods of biomedical, Major preclinical research methods for predicting social and ethological scientific traditions. anxiolytic effects From a biomedical perspective, aggression under inappropriate conditions is considered a disease caused The many preclinical tests or "models" for the study by disordered neural activity that needs to be treated of anxiolytic drugs can be grouped according to the (Mark and Ervin 1970). Aggressive outbursts are part type of experimental manipulations, the nature and of the symptomatology of a range of psychiatric and direction of the behavioral change, and the resulting neurological diseases (vide supra). In spite of many clin- effects as predicting anxiolytic effects. Based on the ical case reports, it still remains poorly understood how, effects of prototypic benzodiazepine drugs, the pre- for example, seizure disorders are related to aggressive clinical tests predict anxiolytic activity either by (i) outbursts, and whether or not the neurobiological restoring behavior that is suppressed by aversive con- mechanisms for ictal and interictal events are related tingencies, (2) attenuating behavior that is increased by to those for explosive aggressive behavior (Eichelman aversive manipulations, or (3) reversing a pharmaco- 1983). In fact, the limbic dyscontrol syndrome has been logically induced "anxiogenic" state (e.g., File 1985; suggested to encompass both aggressive and seizure Handley 1991). Table 1 compares the effects of chlor- events (e.g., Monroe 1978). There is some success with diazepoxide and diazepam as prototypic benzodi- the phenobarbital or carbamazepine azepines with the preclinical work on the potentially in the reduction of aggressive behavior (Tardiff 1983; anxiolytic effects of buspirone as the first clinically 40

Table 1 Prectinical methods predicting anxiolytic effects - acute prototypic effects

Method Benzodiazepine anxiolytics 5-HT agonists Dose Effect Reference Drug Dose Effect Reference (mg/kg) (mg/kg)

Restoration of reversal of behaviors suppressed by aversive contingency C Punished CDP 3-10 1" Barrett (1976) BUS 0.1-40 0/q" Barrett and Witkin (1991) Operant (P) (R,P) Response 3-30 $ Barrett et al. (t986) BUS 0.001-0.3 I" Barrett et al. (1986) (PG) (PG) U Plus-maze DZP 1-25 $ Klint (1991) BUS l $ Klint (1991) Open-arm CDP 1-7.5 $ BUS 0.8 $ File and Andrews (1991) entry BUS 0.14.4 0 File and Andrews (1991) Suppression of behaviors increased by aversive stimulus C Fear- DZP 0.3-2.5 -], Davis (1979) BUS 0.(>10 $ Davis (1986) potentiated startle U Defensive DZP 0.5-2 $ Treit et al. (1981) BUS 0.05-1 $ Treit and Fundytus (1988) burying Drug-induced anxiogenic states C PTZ- DZP 5-10 $ Lal and Shearman BUS 0.56 1" Ator et al. (1989) discrimination (1982) (IM)(P) BUS 0.56 1 0 Ator (1991) U Anxiety-like DZP 1-2 $ Ninan et al. (1982) responses

CDP chlordiazepoxide, DZP diazepam, BUS Buspirone. All doses refer to intraperitoneal administration, unless stated: (IM) intramus- cular route; (IV) intravenous route. C conditioned behavior, U unconditioned behavior, R rats, M mice, P primates, PG pigeons

introduced agent. Major reviews have attenuated by antagonists such as flumazenil (Liljequist summarized the effects of drugs that act at least par- and Enget 1984). tially as 5-HTIA receptor agonists in procedures that So far, the most persuasive preclinical evidence on were developed for the benzodiazepine anxiolytics, i.e. systematic, reliable, dose-dependent attenuation of that either (1) suppress behavior as a result of aversive punishment-suppressed behavior by drugs with affinity contingencies or (2) engender behavior due to aversive for 5-HT1A receptors emerged from studies with stimulation (Chopin and Briley 1987; Barrett and pigeons, whereas similar studies in rats and primates Witkin 1991; Barrett and Vanover 1993; Handley and remain only partially successful (Barett 1992; Sanger McBlane 1993; Handley et al. 1993). 1992). Buspirone, an anxiolytic with some affinity for 5-HTIA receptors, increases specifically punishment- suppressed responding without significantly affecting Restoration or reversal of behavior that is suppressed non-punished food-reinforced key-pecking in pigeons by aversive consequences (Barrett and Witkin 1991). An agent with a more selec- tive affinity for the 5-HT~A receptor subtype, flesinoxan, The classic preclinical findings were obtained with pro- proved to be also highly effective in increasing food- cedures that highlighted how benzodiazepines reversed reinforced responding that was suppressed by punish- the behaviorally suppressive effects of punishment such ment (Barrett et al. 1989). Notably, when given as those introduced by Geller and Seifter (1960), and chronically, buspirone effectively attenuates punish- by Vogel et al. (1971). For example, Sepinwall and Cook ment-suppressed behavior even in rats (Schefke et al. demonstrated systematic, dose-dependent benzodi- 1989). A critical next step will be the reversal of these azepine effects on punishment-suppressed responding effects by newly developed so-called silent in squirrel monkeys that clearly differ from those on antagonists in order to implicate the 5-HT1A receptor concurrently assessed non-punished responding more adequately in these behavioral effects. (Sepinwall and Cook 1984). Of course, these proce- In a research method that does not require condi- dures require considerable conditioning of the animal's tioning, the focus is on exploratory behavior. In behavior in order to obtain stable baselines of punished Montgomery's plus maze, rats readily enter into walled and unpunished responding. Yet, with a few well- arms, but rarely into open arms of the apparatus trained animals systematic, dose-dependent, repeatable (Montgomery 1955). This procedure does not require punishment-attenuating effects are seen with all ben- any training, but necessitates relatively inexperienced zodiazepine anxiolytics. The punishment-attenuating subjects for each test. Usually, behavioral specificity is effects of full or partial benzodiazepine agonists are assessed separately by measuring benzodiazepines' 41

effects on motor activity. Peltow et al. (1985) demon- exposure to aversive environmental events; these effects strated how benzodiazepines increase the time that rats occur at non- doses and without apparent spend in the open arms of a plus-maze which is inter- development of tolerance (Sepinwall and Cook 1984; preted to reflect anxiolytic effects of these drugs. although see, for example, Vellucci and File 1979). Flumazenil antagonizes the benzodiazepine effect on open arm entries (Wada and Fukuda 1991; Rex et al. 1993). Attenuation of drug-induced "anxiogenic" state Similarly, buspirone and gepirone increase the time spent in the usually avoided open arm of a plus-maze, A so-called anxiogenic state can be induced by phar- (S6derpatm et al. 1989). Similar effects have been macological agents of several classes. Some procedures reported for ondansetron, a drug acting at 5-HT3 recep- require elaborate conditioning such as the drug dis- tors (Costall et al. 1988, 1989) but this effect appears crimination procedure, others simply engender a behav- less robust (File and Johnston 1989). ioral, physiological and neurochemical profile of "anxiety" responses as a result of drug administration. Lal and associates trained rats to discriminate Attenuation of behavior that is increased by aversive pentylenetetrazol from saline in a two-lever choice task manipulations (Lal et al. 11988). The injection of either PTZ or saline served as the discriminative stimulus or "cue" defining When exposed to certain aversive conditions animals which response on one of two levers was reinforced, display characteristic responses, some of them require i.e. "drug-appropriate". After the rats responded more conditioning such as the "fear"-potentiated startle, oth- than 90% of the time on the drug-appropriate lever, ers simply measure how much effort is exerted to they were pretreated with benzodiazepine anxiolytics remove the aversive, noxious stimulus such as in a rat's and demonstrated a systematic, dose-dependent atten- "defensive burying." uation of the PTZ discriminative stimulus. Moreover, In the "fear"-potentiated startle procedure, increases animals that are withdrawn from morphine, cocaine, in the startle reflex are measured when the presenta- alcohol or benzodiazepines generalize to the previously tion of a light that has previously been paired repeat- established PTZ discriminative stimulus, presumably on edly with electric shock, immediately precedes a loud the basis of their withdrawal distress (Emmett-Oglesby acoustic stimulus. Benzodiazepines are among the et al. 1983, 1984; Lal et al. 1988; Wood et al. 1989). drugs that attenuate this increase in a dose-dependent Recently, this pharmacological method to induce an fashion without altering the non-potentiated startle anxiety-like state has attained considerable face-valid- reflex (Davis 1979). The antagonist flumazenil blocks ity. For example, rats that are exposed to a predator or the diazepam effect on potentiated startle suggesting to an attacking or threatening conspecific generalize the benzodiazepine receptor as the relevant site of this state to the previously conditioned PTZ action for this effect (Berg and Davis 1984). discriminative stimulus (Vellucci et al. 1988; Gauvin Partial and full 5-HT1A receptor agonists have also and Holloway 1991; Vivian et al. 1994b). Most effectively attenuated the "fear"-potentiated startle. For saline-injected rats, when exposed to an attacking or example, Buspirone and gepirone decreased in a dose- threatening opponent, respond on the pentylenetetra- dependent, systematic manner the potentiated startle zol-appropriate operandum. Midazolam pretreatment reflex (Kehne et al. 1988). So far, however, attempts to prevents the substitution of the social stress for the PTZ block buspirone effects with a range of serotonergic stimulus (Vellucci et al. 1988; Vivian et al. 1994b). antagonists have been unsuccessful. Moreover, the pro- In chaired macaque monkeys, the IV injection of ~- totypic 5-HTIA agonist 8-OH-DPAT, over a range of CCE induced behavioral changes that included agitated several doses, did not alter fear-potentiated startle, sug- responses and distress vocalizations (Ninan et al. 1982). gesting non-serotonergic mediation (Davis et al. 1988). These responses correlated with endocrine measures of Benzodiazepines suppress burying behavior of an stress and were attenuated by benzodiazepines and electrified probe in rats that have been shocked previ- other anxiolytics (Crawley et al. 1985). While some ously by the probe (Treit et al. 1981). This effect appears behavioral features of these fl-CCE-treated monkeys to be mediated via the benzodiazepine receptor, since are reminiscent of macaques that are threatened ftumazenil blocked the suppressive effects of chlor- (Redmond and Huang 1979), the prerequisite of diazepoxide (Treit 1987). Similar to the benzodi- adaptation to chair-restraint for a stress-free baseline azepines, buspirone decreased defensive burying of the renders this experimental protocol in terms of face- shock probe (Treit and Fundytus 1988), although these validity. effects depend on specific testing parameters (Craft To summarize, benzodiazepines consistently and et al. 1988). effectively attenuate the suppressive effects of punish- Anxiolytics with different mechanisms of action ment on conditioned and unconditioned behavior in appear to selectively attenuate conditioned and uncon- various animal species including humans. Similarly, ditioned behavior that is engendered or amplified by these drugs reduce so-called fear- or stress-induced 42 conditioned and unconditioned responses. Drug- few preclinical studies have given these drugs repeat- induced "anxiogenic" states such as the pentylenete- edly or chronically which is necessary for buspirone trazol "cue" or the fl-CCE behavioral and physiological and ipsapirone to become effective in ameliorating responses are attenuated by benzodiazepines. These GAD in humans (e.g., Pecknold t994). anxiolytics achieve their behavioral effects via benzo- Most current experimental preparations have been diazepine receptors which is demonstrated by the rever- effective predictors for treatment of GAD, but have sal of the behavioral and physiological responses by been less successful for identifying drugs for the other administration. These observations types of anxiety. GAD, obsessive-compulsive disorders in animals point to the early recognition of the anxi- and panic disorders often benefit from different phar- olytic effects of benzodiazepines through the use of an macological treatments and appear to have different impressive range of procedures. While generally vali- neurobiological mechanisms requiring separate pre- dated with benzodiazepines through the use of an clinical behavioral procedures. impressive range of procedures. While generally vali- dated with benzodiazepine agonists and antagonists, the available experimental protocols differ considerably Major experimental methods for studying the effects of in terms of face validity, economy of training time and anxiolyUcs on aggression animal use. The implicit assumption for attenuating and reversing behavioral changes by benzodiazepine The research strategies fbr delineating the role of 5-HT anxiolytics is that these behavioral effects are due to in aggressive behavior in animals and in clinical pop- changes in affective or emotional processes. ulations differ substantially from those for character- Partial and full 5HTIA agonists are characterized by izing the benzodiazepines' effects on these behavior systematic and potent effectiveness only in some of the patterns. While many studies on 5-HT have relied upon "classic" punishment and fear-conditioning procedures correlating an index of serotonergic activity in brain (e.g., Przegalinski et al. 1992), but as will be discussed tissue, or CSF with an "aggressive" trait, ben- in detail below, more convincingly in those experi- zodiazepine research has developed a profile of drug mental procedures that engender so-called fear or dis- effects on dominance, territorial and maternal aggres- tress responses such as in rat pups separated from their sion as well as defensive reactions to attacks by rivals mother and nest, in adult rats that are threatened by and predators. Illustrative evidence from these latter conspecifics, or that are in withdrawal "distress"; only procedures is summarized in Table 2.

Table 2 Experimental methods for studying the effects of anxiotyt- 8-OH-DPAT 8-hydroxy-dipropylaminotetralin. All doses refer to ics on aggression. CDP chtordiazepoxide, DZP diazepam, ALP intraperitoneal administration, unless stated: (PO) oral route; (SC) alprazolam, BUS Busirone, GEP Gepirone, ELT , subcutaneous route. IM intramuscular route

Method Benzodiazepine anxiolytics 5-HT agonists Drug Dose Effect Reference Drug Dose Effect Reference (mgfkg) (mg/kg)

Dominance (rats) CDP 2.5-5 $ Miczek, (1974) BUS 2-8 $ Mos et at. (1993) 20(IM) $ Miczek, (1974) 8-OH-DPAT 0.1-0.2 $ (sc) ELT 1.25-5- $ Territorial (mice) a) Male-female CDP 5 $ Miczek and O'Donnell, BUS 1-10 0 Gao (1993) housing 20 $ (1980) GEP 3-30 $ Miczek, unpublished b) Single-housed CDP 5-20 $ Rodgers and Waters, BUS 1-10 $ Olivier et al. (1985) (1989) 8-OH-DPAT 1-10 Maternal aggression (rats) ALP 1.2 q" Mos and Olivier (1989) BUS 0,3-3 0 Mos et al. (1992) 5 + CDZ 5-20 1" 8-OH-DPAT 0,10.2 $ (PO) DZP 0.3-1 $ Olivier et al. (1991) 3-17 $ ELT 2-8 $ Defense Intruder rats DZP 3 lO +/0 Vivian and Miczek GEP 3---6 -J./O Vivian and Miczek, (1993a) (1993a) Tornatzky and Miczek Tornatzky and Miczek (I995) (1995) 43

Benzodiazepines or by a lactating female defending her litter, i.e. lower doses increased aggressive behavior and higher doses Benzodiazepine effects have been studied on all major decreased these behaviors. Defensive responses to types of aggressive behavior in animals (Table 2; e.g., conspecifics and predators are routinely decreased by Miczek and Krsiak 1979; Rodgers and Waters 1985; benzodiazepines, although only at intermediate and Miczek 1987; Mos et al. 1987; Sulcova and Krsiak higher doses that are sedative. 1989; Krsiak and Sulcova 1990). Diazepam and other Benzodiazepines' and alcohol's enhancement of sedative-~pnotics such as alcohol share the feature of aggressive behavior have been attributed to a disin- increasing aggressive behavior by resident rats con- hibitory action similar to the release of inhibition of fronting an intruder, when given at lower doses (Miczek punished behavior. However, the similarity appears to 1974; Miczek and Barry 1977; Olivier et al. 1984; be only superficial. First, in rats, the doses that are Olivier et al. 1991; Miczek et al. 1992); however, at sufficient to increase aggressive behavior are almost ten- higher doses, these drugs decrease aggressive behavior, fold lower than those necessary to increase punished primarily due to their sedative and muscle-relaxant behavior. Also, increases in aggressive behavior after effects. The aggression-heightening effects of alcohol benzodiazepine treatment are not universal. For exam- can be antagonized by benzodiazepine receptor antag- ple, recently we observed large increases in low-rate onists in resident rats and in dominant squirrel operant behavior after alprazolam administration, but monkeys, but the aggression-decreasing effects remain concurrent decreases in aggressive behavior in the same unaltered by these antagonists (Weerts et al. 1993b). mice (Miczek, unpublished observations). In order to This is illustrated in alcohol-treated dominant squirrel maintain a disinhibition hypothesis as a unifying prin- monkeys that display aggressively and emit aggressive ciple for benzodiazepine effects on suppressed behav- vocalizations toward lower ranking group members or ior and on aggressive behavior, it is necessary to rivals (Weerts and Miczek, unpublished). postulate multiple inhibitory mechanisms, each depen- In singly or pair-housed mice that confront intrud- dent on the GABAa-benzodiazepine receptor complex. ers into their territory, chlordiazepoxide and midazo- lam increase aggressive behavior at low doses and decrease it at higher doses (Miczek and O'Donnell 5-HT drugs 1980; Rodgers and Waters 1985). Similarly, in certain mice, particularly those with elevated levels of testos- No other neurotransmitter system has been more fre- terone, low alcohol doses increase aggressive behavior, quently linked to a broad range of aggressive behavior whereas higher doses decrease it (DeBold and Miczek than . Several major families of serotonin 1985; Miczek et al., submitted). Antagonism and recep- receptors, each with several subtypes, are now recog- tor binding studies point to the GABAa-benzodi- nized (Hoyer and Boddeke 1993; Hoyer et al. 1994). azepine receptor complex as the most likely site for the While the molecular cloning of these receptors pro- alcohol aggression-heightening effects (e.g., Weerts et ceeds rapidly, understanding the functional significance al. 1992; Grant 1994; Miczek et al. 1994a). of many of the newly identified receptor subtypes in Similarly, the biphasic nature of the benzodiazepine different types of aggression and anxiety remains a chal- effect on aggression is seen in maternal female rats and lenge at present (Wilkinson and Dourish 1991; Mos mice (Olivier et al. 1985; Yoshimura 1987). The increase et al. 1992; Olivier et al. 1992; Barrett and Vanover in aggressive behavior is evident as more frequent pur- 1993). suits, threats and attacks after administration of low The preclinical stud?" of the role of 5-HT receptor doses of benzodiazepines. Curiously, flumazenil failed subtypes in different types of aggressive behavior is at to antagonize the increased aggressive behavior of lac- an early stage. This is surprising since the 5-HT tating female rats after chlordiazepoxide treatment deficiency hypothesis of violent and aggressive behav- (Mos et al. t987). ior has been reiterated during the past 2 decades with In contrast to the aggression-enhancing effects of vigor and enthusiasm. Ever since Brodie and Shore benzodiazepines, these drugs dose-dependently attenu- (1957) proposed brain 5-HT to subserve an inhibitory ate flight responses in selectively bred timid mice role in behavior and autonomic nervous activity, Hess' (Krsiak 1975) and defensive responses in feral rats tropotrophic system appeared to have a neurochemi- (Blanchard et al. 1989). In feral rats, intermediate doses cal basis (Hess 1954). In fact, the tropotrophic role of of benzodiazepines specifically reduce defensive threat serotonin continues to be the guiding principle for and attack without altering the remainder of the ago- much preclinical and clinical aggression research. nistic repertoire. The antagonist flumazenil attenuated During the 1960s and 1970s, preclinical studies of the benzodiazepine inhibition of flight behavior in mice fighting in singly housed mice and in mouse-killing rats (Sulcova and Krsiak 1984). associated low levels of brain 5-HT metabolites with a To summarize, benzodiazepines exert biphasic effects high incidence of fighting and killing (e.g., Garattini on aggressive behavior by a resident male confronting et al. 1967; Giacalone et al. 1968; Valzelli and Garattini a territorial intruder, by rivals in dominance contests 1968; Valzelli and Bernasconi 1979; Valzelli 1982; 44

Pucilowski and Kostowski 1983; Valzelli and Galateo is more than double that of heterozygous mice (Saudou 1984). Evidence from studies with highly aggressive et al. 1994). Interestingly, no unusual changes in repro- humans with various antisocial, alcoholic or other ductive or feeding behavior, pain responses or sleep- diagnoses showed often an association with low 5- waking activity are seen in the 5-HT~B "knock-out" HIAA levels in spinal CSF (Brown et al. 1979; Linnoila mice pointing to a relatively selective change in aggres- et al. 1983; Virkkunen et al. 1989). sive behavior. However, eltoprazine remains effective in These correlational studies between a single value of decreasing aggressive behavior of these 5-HT~ B mutant 5-HT or its metabolite level in whole brain or cortical mice suggesting that the 5-HT1A receptor continues to regions or CSF from spinal taps on the one hand, and modulate this behavior (Hen et al. 1994). a characteristic trait of excessive aggressive behavior So far, the most specific modulations of aggressive on the other, remain inconclusive as to the causal rela- behavior appear to be achieved via manipulations of tionship between these neurochemical and behavioral the 5-HT~ receptor family, whereas 5-HT2 and 5-HT3 variables. Single measures of blood or CSF antagonists suppress aggressive behavior with limited 5-HIAA are ditficult to relate to regulatory events at specificity. Remarkably, no 5-HT agonist or antagonist, pre- and postsynaptic sites of anatomically discrete acting pre-or post-synaptically, has been found to 5-HT neurons (Miczek et al. 1994b). increase any type of aggressive behavior in animals. The development of more selective agonists and From a pharmacological viewpoint, the current antagonists at the 5-HT receptor subtypes that act at characterization of the various 5-HT receptor subtypes somatodendritic, presynaptic autoreceptors or postsy- in anxiety and aggression is tantalizing and frustrat- naptic sites, associated with ion channel proteins or G ing. For example, which are the requisite mechanistic proteins, is prompting research on their respective role changes during the period of chronic treatment with in dominance, territorial and maternal aggressive partial 5-HT~Aagonists that are necessary for clinical behavior as well as defensive responses (Miczek et at. effects to emerge? Pharmacologically, adequate "silent" 1989; Olivier et al. 1989; Mos et al. I992). Table 2 sum- antagonists only now are becoming available in order marizes the key findings of drugs acting on 5-HT~ to establish the 5-HTIA receptor as the important site receptors on various aggressive behavior patterns in of action for buspirone, ipsapirone or gepirone (e.g., comparison with those of benzodiazepines. Hoyer and Boddeke 1993; Hoyer et al. 1994; Fletcher In confrontations with an intruder, resident male and et al. 1993) Similarly, the pharmacological tools for female rats and mice engage in less frequent aggressive study of the 5-HTlwD and 5-HT3 receptors remain too behavior after treatment with full agonists at 5-HT1A limited in order to adequately characterize the role of receptors, such as 8-OH-DPAT or flesinoxan, and par- these receptors in experimental protocols for the study tial agonists, such as buspirone, ipsapirone and of anxiety and aggressive behavior. gepirone (Oliver et al. 1990; Miczek and Haney, unpub- lished observations). By far the most promising evidence on the role of 5- Anxiolytics and affective vocalizations during aggression HT receptors in aggressive behavior has been collected and anxiety with experimental manipulations that target the 5-HT1A. B receptors. Highly effective and behaviorally specific A seemingly insurmountable challenge to prectinical reductions in a range of aggressive behavior patterns research on anxiolytics is the experimental have been achieved with piperazine derivatives that act quantification of the affective or emotional dimension as mixed agonists at 5-HT1B and 5-HT1A receptors. of anxiety and aggression. The premise for this quest Olivier and associates have summarized the evidence may be traced to Darwin's argument that emotions have that eltoprazine and the more 5-HTIB selective agonist evolved in species other than humans. As a matter of TFMPP (m-trifluoromethylphenyl-piperazine) which fact, the neurobiological mechanisms tbr emotional also causes 5-HT release, decrease a resident rat's or behavior have been studied in a range of mammalian mouse's attacks and threats toward an intruder, decrease species, including non-human primates (MacLean a lactating female rat's attack toward an unfamiliar mate 1949; Skolnick and Paul 1982; Ploog 1989). Yet, approaching the nest, increase the electrical current preclinical psychopharmacological research on aggres- threshold for evoking attack leaps and bites by hypo- sion and anxiety in the behavior-analytic tradition has thalamic stimulation in rats toward a stimulus rat, and largely circumvented the direct study of emotional reduce the incidence of biting during the formation of behavior, by inferring emotional processes from dominance hierarchies among young pigs (Olivier et al. disrupted ongoing behavioral performance (e.g., Estes 1990). The potent anti-aggressive effects across a range and Skinner 1941; Brady 1956). of animal species and situations is promising, although As summarized above, animal models are validated these "serenic" drugs may also increase responses to pharmacologically by their sensitivity to produce aversive events (Rodgers et al. 1992). behavioral changes as a result of specific treatments When the 5-HT1B receptor is "knocked out" in rather than attempting to model specific changes in mutant male mice, their frequency of attack behavior affect, i.e. pharmacological versus external validity. It 45

is important to recall that the critical defining features may detect affective aspects of pain (Levine et al. 1984; of many affective disorders, including anxiety, are their Borszcz et al. 1994). Yet, the rat's audible squeal actu- emotional nature and disruption of normal social func- ally represents only a small fraction of its vocal tioning. Modeling specific affective expressions such as responses immediately before, in reaction to a painful "fear" or "contentment" under socially relevant con- stimulus or after the termination of the stimulus. ditions may provide valuable insight in the neurobiol- Additionally, single or bursts of ultrasonic vocaliza- ogy of anxiety that have not been adequately addressed tions, in anticipation or in reaction to painful stimuli, in animal models (Mark and Ervin 1970; Redmond are emitted in the 20- to 30-kHz frequency range with and Huang i979; Winslow and Insel 1991b). A com- little frequency modulation and are the prevailing vocal mon characteristic of intense excitation when in pain expressions (Tonoue et al. 1986; Van der Poel et al. or fear, is the emission of high-frequency calls. Both in 1989; Van der Poel and Miczek 1991; De Vry et al. mammals and in birds, the average frequency increases 1993; Sanchez 1993; Molewijk et al. 1995). and the band of frequencies of individual vocalizations Opioid peptides, such as /~-endorphin, met- and widens during fearful situations (Scherer 1986; Aubin leu-enkephalin, and dynorphin proved effective in and Bremond 1992). reducing pain-induced ultrasonic vocalizations, but Vocalizations in mammals are the product of audible squeals in reaction to the electric shock respiratory, laryngeal and articulatory movements that pulse remained unaffected (Tonoue et al. 1986). The are the result of activity of cranial and spinal motor ultrasonic vocalizations in the "anticipatory" phase, neurons in the brain stem that in turn receive modu- before the delivery of electric pulses to the tail, were latory input from higher cerebral levels (Jfirgens particularly sensitive to morphine's suppressive effects, and Ploog 1981). For example, anatomical studies in which were reversed by naloxone (Van der Poel et al. squirrel monkeys identify the anterior cingulate cortex, 1989). The environmental context that is associated the basal amygdaloid nucleus, dorsomedial and lateral with past electric shock deliveries is sufficient to induce hypothalamus, and midline thalamus in addition to the ultrasonic vocalizations, and these calls may be inter- caudal periaqueductal grey area as the most prominent preted as affective expressions in anticipation of pain structures that yield vocalizations when electrically (Miczek et al. 1991b). Opioid modulation of ultrasonic stimulated. Also, in rats, electrical stimulation of vocalizations by rats that anticipate the delivery of a thalamic, hypothalamic and mesencephalic as well as painful stimulus may in fact point to a significant role medullary structures evokes ultrasonic vocalizations in of opioid peptides as well as mu and delta receptors in the 20-30 kHz range (Yajima et al. 1980, 1981). Among affective responses. Further support for a significant the neural structures that are most effective for role of opioid peptides in the modulation of affective evoking vocalizations are those where GABAA-benzo- responses may be adduced from the effects of specific diazepine receptor complexes and 5-HT receptors are opioid receptor subtype agonists on vocalizations of localized (Young and Kuhar 1980; Zifa and Fillion rats that have been startled by an acoustic or tactile 1992). Whether modulation of vocalizations by anxi- stimulus. Mu and delta selective receptor agonists, olytic drugs reflects an alteration in affective processes such as DAGO and DPDPE, potently inhibit 20- or is a modification of activity in laryngeal or respira- to 32-kHz ultrasounds in a naltrexone-reversible man- tory activity remains a critical question. ner in male intruder rats (Vivian and Miczek, in prepa- Before highlighting the significance of vocalizations ration). These observations implicate mu and delta in socially arousing situations, it may be useful to con- receptors in vocalization that may reflect affective sider recent evidence on attempting to quantify the distress. emotional component of behavior in response to painful Consistent with this hypothesis is the observation and startling stimuli. that rats that are withdrawn from opiate treatment emit ultrasonic vocalizations (Vivian and Miczek 1991). In addition to the well-known "scream-on-touch" and Vocalizations in the context of pain and startle other prominent signs of autonomic distress, morphine- withdrawn rats emit ultrasonic calls in the 20- to 30- Preclinical methods in pain research include those that kHz as well as the 40- to 60-kHz range for a longer evoke vocalizations by rats; these sounds are typically period of time than the somatic and autonomic with- in the hearing range of the human experimenter (e.g., drawal symptoms. These ultrasounds are particularly Hammond 1989). The stimulus-evoked audible vocal- prominent when the morphine-withdrawn rat is chal- izations are sensitive to opioids' effects lenged in a social situation or by a startling stimulus. (Levine et al. 1984), and when withdrawn from chronic Withdrawal from 30- or 60-day drinking of cocaine opiate treatment, rats "scream on touch" (Blfisig et al. solution is also characterized by the emission of high 1973). Additionally, rats emit audible vocalizations rates of ultrasounds. This effect is most prominent dur- after the painful electrical stimulus has terminated; ing the first week after cocaine termination when the these so-called vocalization afterdischarges are sensi- rats are challenged by tactile startle stimuli (Barros and tive to opiates and benzodiazepine anxiolytics, and they Miczek 1994). 46

Evidence from recent studies with anxiolytic drugs more than other calls (Harris and Newman 1988; provides important phaI~acological validation for the Newman 1988a; Kalin and Shetton 1989). Infant and contention that ultrasonic vocalizations may represent juvenile primates separated from their mother or social affective expressions by rats that are exposed to star- group immediately respond with increased vocaliza- tling or painful stimuli. Anxiolytics that target espe- tions, agitation, and motor activity (Reite et al. 198t; cially 5-HTIA and benzodiazepine-GABAa receptors Wiener et al. 1988; Suomi 1991), as well as increases effectively decrease the rate and duration ofultrasounds in heart rate, body temperature, and circulating corti- in those rats that emit these 20- to 30-kHz calls, when sol levels and contact with the mother attenuates these exposed to acoustic startle stimuli (Kaltwasser 1991) changes (Mendoza et al. 1978; Coe et al. 1985). As or to foot shock, but not tail shock (Cuomo et al. 1988; social separation is prolonged, "protest" behaviors Rowan et al. 1990; De Vry et al. 1993; Sanchez 1993; diminish and infants display "despair" (Suomi 1991). Cullen and Rowan 1994; but see Van der Poel et al. During the despair phase, infants typically display 1989). No tolerance to the suppression of shock-elicited fewer vocalizations, a slouched posture, more self ultrasounds was seen with repeated administration of directed behaviors, less motor activity and fewer social ipsapirone, and no rebound increase in calls was evi- interactions. dent when ipsapirone administrations were terminated In infant rhesus monkeys, separation-induced "dis- (De Vry and Schreiber 1993). One day after the last of tress" vocalizations, defensive behaviors (e.g., freezing ten diazepam administrations rats emit increased ultra- and crouching) and elevations in plasma ACTH and sonic vocalizations, when prompted by acoustic startle cortisol levels were reduced by diazepam (Kalin et al. stimuli (Miczek and Vivian 1993). Gepirone and 1987). Flumazenil blocked the decreases in plasma diazepam reversed the increased ultrasound emission ACTH and cortisol produced by diazepam, but not the during diazepam withdrawal (Vivian et al. 1994a). It reductions in vocalizations. Specifically, defensive is, hovcever, noteworthy that buspirone has not been "bark" calls were significantly decreased by diazepam, shown to be effective in the clinical management of and unaffected by morphine or naloxone, whereas the benzodiazepine withdrawal (Schweizer and Rickels affiliative "coo" calls were reduced by morphine and 1986). When exposed to the environment where rats increased by naloxone but are unaltered by diazepam have received electric foot shock previously, they emit- (Kalin et al. 1988; Kalin and Shelton 1989). ted fewer ultrasounds when given drugs with anti-panic SociMly separated infant and adult squirrel monkeys activity such as alprazolam and 5-HT uptake blockers generally emit isolation "peeps" and show increased (Molewijk et al. 1995). adrenal levels (Winter 1968; Symmes Evidence from comparative, chronic, withdrawal and et al. 1979). In young animals these calls may solicit antagonism studies has begun to pharmacologically social contact with mother and group members characterize the vocalizations by rats exposed to situa- (Newman 1988b), and tactile contact with the mother tions where painful and startling stimuli are likely to significantly reduces adrenal and vocal "distress" occur. Potent and efficacious modulation of ultrasonic responses (Mendoza et al. 1978; Wiener et al. 1988; vocalizations and certain audible vocalization after dis- Weerts and Miczek, submitted). Recently, behavioral charges in rats appear to detect the selective effects of "protest" in reaction to social separation in juvenile drugs acting on opiate, 5-HT~A and benzodiazepine squirrel monkeys was confirmed by characteristically receptors that may play a role in affective expressions. high rates of "peep" vocalizations and explosive motor activity (Fig. 1; Weerts and Miczek, submitted). These vocalizations were hypothesized to be affective expres- Vocalizations during social separation sions of anxiety-like states produced by separation from the mother and familiar group members. Both vocal- Separation from a major attachment figure is consid- izations and explosive motor behaviors were highly sen- ered as a primary cause of childhood separation anx- sitive to full and partial benzodiazepine receptor iety, anticipatory anxiety and over-anxiousness as well agonists and to antagonists. The full agonist, chlor- as for generalized anxiety and panic disorders in adults diazepoxide, as well as the antagonist, flumazenil, dose- (DSM IV, American Psychiatric Association 1994). The dependently reduced calls and explosive motor etiology and pharmacological modulation of separa- behaviors (e.g. rapid locomotion, jumping, leaping at tion distress in primates and other animals may be anal- and shaking the cage; Fig. 1). In contrast, the partial ogous to that observed in humans (Panksepp et al. agonist bretazenil, which may be less sedative at anxi- 1978; Reite et al. 1981). olytic doses (Haefely et al. 1990), reduced calls only at very high doses. In addition, alcohol did not attenuate vocalizations, except at an ataxic dose. The potent Primates anxiolytic-like effects of flumazenil on squirrel monkey "peeps" were closely similar to effects reported for iso- Vocalizations during social separation (i.e. "isolation lated rat pup vocalizations (Insel et al. 1986; vide infra), calls") have been characterized pharmacologically although this effect is not always as significant as in 47

A. Squirrel monkeys affective state of the animal, as suggested for separated kHz rat pups (Inset et al. 1989). Although attractive, the evi- dence for an endogenous ligand remains to be sub- stantiated. Flumazenil's effects in preclinical procedures 16 have been interpreted as both "anxiolytic" (e.g., increased social interactions, exploratory behaviors and I2 feeding) and "anxiogenic" (e.g., reduced social behav- iors, partial generalization to PTZ) depending on the dose and testing situation (File and Pellow 1986; Haefely 1988; Weerts et al. 1993a). Also, squirrel mon- keys may be unusually sensitive to the partial agonist 0.5 sec actions of flumazenil. Previously, squirrel monkeys were found to be more sensitive to benzodiazepine B. Benzodiazepines: receptor antagonists such as flumazenil and ZK 93426 Separation induced vocalization and partial inverse agonist such as Ro 15-4513 than rodents. Flumazenil's hyperphagic effects as well as the Monkeys Rat pups proconvulsant effects of Ro 15-4513 were readily detected in squirrel monkeys, but not in rats (Weerts ---t I et al. 1993a). 12- T1 "{00 In selected adult squirrel monkeys, isolation "peep" calls were sensitive to opiate modulation; morphine sup- 9 pressed isolation calls, whereas naloxone increased iso- .~_ :z E 60 E lation calls and blocked the reductions in calls produced 6 by morphine (Newman t988a). In addition, "peeps" o O - 40 were also suppressed by imipramine and clonidine, when tested 1 h after administration (Harris and 3 I1~ 20 Newman 1988). Clonidine-induced suppression of iso- lation "peeps" were reversed by yohimbine, but not by 0 I ...... ~ ...... i ...... 0 prazosin. However, yohimbine increased isolation Veh 0.1 t 10 "peeps" when administered alone (Harris and Newman Dose (mg/kg) 1987). In addition to benzodiazepine anxiolytics, opi- oid and adrenergic systems also appear to modulate iso- C. Ratpups lation-induced vocalizations in adult squirrel monkeys. kHz

Young rodents 64 When separated from the litter and mother and when 48 exposed to cold or rough handling, neonatal rats and mice emit ultrasonic vocalizations as pure tone whis- tles, ranging in frequency from 20- to 60-kHz, and 0.1- 32 to 3.5-s in duration (Fig. 1C). Benzodiazepine anxi- olytics, such as chlordiazepoxide and diazepam, 16 0.1 sec decreased pup uttrasounds without altering locomotor behavior (Gardner 1985), while inverse agonists, such as DMCM, FG 7142 and pentylenetetrazol, increased Fig. IA Sonogram of "peep" vocalizations emitted by a juvenile squirrel monkey when separated from the native social group. B ultrasounds (Insel et al. 1986; Gardner and Budhram Effects of chlordiazepoxide on "peep" vocalizations by juvenile 1987; Nastiti et al. 1991). Flumazenil reversed the effects squirrel monkeys, and of diazepam on ultrasonic vocalization by of agonists and inverse agonists implicating the benzo- rat pups. The data are expressed as a function of dose. C Sonogram diazepine receptor in the mediation of pup ultrasounds. of ultrasonic vocalizations emitted by a rat pup when separated from the littermates and dam. (Data from Weerts and Miczek, sub- Furthermore, socially isolated pups show decreased mitted, and from Vivian et aL, submitted) flumazenil binding in limbic structures (Insel 1989). GABA agonists such as muscimol and baclofen were found to decrease ultrasounds in mouse pups (Nastiti juvenile rhesus monkeys' "distress" vocalizations et al. 1991), and recently the role of the benzodiazepine- (Kalin et al. 1987). GABAA receptor complex in rat pup ultrasonic vocal- Flumazenil may interact with endogenous sub- ization was more fully characterized (Fig. 1; Vivian stances and exert different effects depending on the et al., submitted). , a 48

Table 3 Vocalizations during aggression and anxiety -prototypic Buspirone, GEP Gepirone; IPS ipsapirone, 8-OH-DPAT 8-OH- effects of anxiolytics. A V Audible vocalizations, USV Ultrasonic dipropylaminotetralin. All doses refer to intraperitoneal adminis- vocalizations, High USV 31~0 kHz, Low USV 20-30 kHz, CDP tration, unless stated: (IM) intramuscular route chlordiazepoxide, DZP diazepam, FNP flunitrazepam, BUS Method Benzodiazepine anxiolytics 5-HT agonists Drug Dose Effect Reference Drug Dose Effect Reference (mg/kg) (mg/kg)

Aversive environmental stimuli (adult rats') Footshock DZP 0.5 1 $ Cuomo (1988) BUS 0.3-10 $ De Vry et aI. (1993) Low USV GEP 1 5 $ Cullen and Rowan (1994) 8-OH-DPAT 0.06-0.24 Sanchez (1993) Startle FNP 0.5 $ Kaltwasser (1991) IPS 5 ,[, Kaltwasser (1991) Low USV DZP 1 3 $ Vivian et al. (1994a) GEP 0.6-1 $ Vivian et al. (1994a) Social separation AV Primates DZP 1 $ Kalin et al. (1987) COP 3-10 (IM) $ Weerts and Miczek, submitted USV Rat pups DZP 2.5-10 Mos and Olivier (1989) BUS 1-3 $ Mos and Olivier (1989) 8-OH-DPAT 0.1~.2 $ Aggression AV Primates CDP 0.3(IM) 1" Weerts and Miczek, submitted Response to threat of aggression (adult rats') a) "anticipatory" High USV DZP 6-10 $ Tornatzky and Miczek GEP 0.3-6 0 Tornatzky and Miczek (1995) (1995) Low USV DZP 1 10 0 GEP 36 $ b) "reactive" High USV DZP 10 0 Tornatzky and Miczek GEP 0.3-6 ,[, Tornatzky and Miczek 0995) (1995) Low USV DZP 1 10 0 Vivian and Miczek GEP 3 6 .1, Vivian and Miczek (1993a) (1993a) acting on this receptor complex, proved active in sup- clomipramine proved effective in suppressing pup ultra- pressing pup ultrasounds (Zimmerberg et al. 1994). sounds (Hard and Engel 1988; Mos and Olivier 1989; Using the suppression of pup ultrasounds as a func- Winslow and Insel 1990a, b). Interestingly, agonists at tional endpoint, the order of relative potency of agents the 5-HT/u receptor TFMPP and CGS12066B increase acting on this receptor complex was muscimol > alpra- the rate of pup ultrasounds. The present evidence points zolam > diazepam > pentobarbital = allopregnanolone to 5-HTIA receptors as being significant targets for > ethanol (Vivian, Ban-os and Miczek, submitted). The modulating pup ultrasounds. However, in order to effects of diazepam and alprazolam were reversibly implicate the 5-HTIa receptors in a selective suppres- antagonized by flumazenil, and those of muscimol by sion of pup ultrasounds it will be important to reverse biccuculine confirming the benzodiazepine and agonist effects. GABAA receptors as relevant for the suppression of Separation-induced vocalizations were initially pup ultrasounds. Allopregnanolone potentiated the linked to opioid peptides and their receptors (Panksepp suppressive effects of diazepam and atprazolam et al. 1978). Rat pup ultrasounds are suppressed by markedly, and those of pentobarbital moderately. morphine and by selective mu and delta receptor ago- Conversely, none of the antagonists of this receptor nists such as DAGO and DPDPE, while the kappa complex (i.e. flumazenil, bicuculline, and picrotoxin) receptor selective agonist U50488H increased these altered the effects of allopregnanolone on pup ultra- vocalizations (Carden et al. 1990, 1994; Kehoe and sounds. Shoemaker 1991; Barr et al. 1994; but see Winslow and The benzodiazepine-GABAA receptor complex con- Inse11991a). The opposite effects ofmu and delta recep- tains receptor and recognition sites for chemically tor agonists versus kappa receptor agonist parallels the diverse agents, extending most recently to neuros- putative reinforcing and aversive effects of these drugs teroids, that are significant targets for modulating pup and are consistent with their proposed role in affective ultrasounds. It will be essential to decipher the neces- behavior. sary molecular configuration of novel substances with anxiolytic-like effects on these distress signals. The 5-HT~A agonists 8-OH-DPAT, flesinoxan, and Vocalizations during threatening situations the partial agonists gepirone, ipsapirone and buspirone, the 5-HT2 agonist DOI as well as the 5-HT uptake Rodents and primates emit a range of loud and blockers such as fluvoxamine, zimelidine, and frequent vocalizations in response to the threats and 49 attacks of an aggressive opponent and as anti-preda- (Vivian and Miczek 1993b; Haney and Miczek 1994, tor signals. Functionally, these calls may represent submitted). affective expressions, but they also may provide infor- When shielded from physical harm, intruder rats mation about the sender, are part of defensive and sub- continue to emit low- and high-frequency ultrasonic missive displays, and may warn conspecifics about vocalizations in response to the threats of an aggres- imminent danger. Recent evidence indicates that these sive opponent. The intensity of a confrontation that calls can be modified by mood-altering drugs, partic- relies upon postural displays and vocal responses is ularly benzodiazepine and 5-HT drugs with anxiolytic considerably lower than that during the actual fight as and potentially effects. indicated by the magnitude of endocrine, cardiovascu- lar and thermoregulatory responses (Miczek et al. 1991a). Under these threatening conditions gepirone Vocal responses to threats and attacks decreased low-frequency ultrasounds in experienced and inexperienced intruder rats (Vivian and Miczek As many currently available anxiolytics are chiefly 1993a; Tornatzky and Miczek 1995). Diazepam atten- effective in anticipatory or generalized anxiety disor- uated the emission of high-frequency ultrasounds in ders (Molander 1982), it is not surprising that the most animals that were threatened for the first time. successful preclinical tests involve animals who antici- Interestingly, drugs such as opiates and clonidine that pate fear-provoking events as a result of conditioning have anxiolytic activity in some individuals, also were (e.g., Estes and Skinner 1941). When a rat enters a effective in reducing high-frequency ultrasounds in this locale that signals the presence of a potential attack- phase of the social confrontation (Vivian and Miczek ing opponent, it begins to emit bursts of ultrasonic 1993b; Haney and Miczek 1994; Tornatzky and Miczek vocalizations in the 20-30 as well as 40-60 kHz range 1994). However, the suppression of high-frequency (Tornatzky and Miczek 1994, 1995). ultrasounds may be part of the sedative effects of Diazepam and alcohol effectively attenuated high- diazepam, morphine and clonidine, since these effects frequency ultrasounds by intruder rats when they were are seen only at higher doses. exposed to the environment where they were previously A most intriguing pharmacological profile of effects attacked, while gepirone suppressed low-frequency on ultrasonic vocalizations during social confronta- ultrasounds (Tornatzky and Miczek 1995). In this early, tions is emerging. Drugs with known or putative anx- anticipatory phase of a social confrontation, the emis- iolytic activity attenuate ultrasonic vocalizations in sion of ultrasounds is accompanied by tachycardia and animals that anticipate a confrontation with an aggres- hyperthermia, all of which were reduced by alcohol, sive opponent, but not in those who actually defend diazepam and gepirone. These observations point to and submit in an intense fight. This distinction between the phase preceding a social confrontation as engen- anticipatory and reactive phases of an encounter offers dering a constellation of behavioral acts, including the opportunity to differentiate between drug effects vocal responses, and physiological changes that are that may be relevant to the generalized anxiety disor- particularly sensitive to the effects of drugs with an der vs those that may correspond to panic states. anxiolytic-like profile, albeit each with a different mech- anism of action. In fact, the marked behavioral and physiological Anti-predatory vocalizations activation in reaction to attacks by an aggressive opponent remains unaffected by anxiolytics in both Under field conditions, squirrel monkeys and vervet experienced and inexperienced intruder rats (Vivian monkeys confronting aerial and carnivorous predators and Miczek 1993a; Tornatzky and Miczek 1995). emit vocalizations with distinct structural elements During the reactive phase of a social confrontation, and in specific frequency ranges (Struhsaker 1967; intruder rats' cardiovascular and thermoregulatory Seyfarth et at. 1980; Newman et al. 1983; Marler 1982). functions are maximally activated and upright and Benactyzine, an anticholinergic drug, prompted adult supine postures are accompanied by bursts of loud squirrel monkeys to emit higher rates of alarm calls and frequent ultrasonic vocalizations. Even at sedative in response to threatening objects presented by the and muscle-relaxant doses, neither diazepam, alcohol experimenter, but not to increase isolation calls nor gepirone affected the emission of audible and (Glowa and Newman 1986; Glowa et al. 1988). This low-frequency ultrasonic vocalizations. It appears that effect was blocked by the cholinesterase inhibitor, drugs that are clinically effective in ameliorating physostigmine, indicating pharmacological specificity generalized anxiety are ineffective in modulating to cholinergic mechanisms. By contrast, yohimbine vocalizations under conditions of high-intensity and naloxone increased squirrel monkey isolation confrontations. It is noteworthy that morphine and calls out did not alter alarm calls (Harris and Newman the mu and delta receptor-selective agonists, DAGO 1988). Vocalizations elicited during situations of and DPDPE decreased ultrasounds by male or isolation and alarm can be pharmacologically dis- female intruders while they were being attacked sociated. 50

Whether or not antipredator alarm calls are first minute of aggressive encounters and declined dur- modified by anxiolytic or antidepressant drugs, and ing the course of the confrontation (Weerts and Miczek, whether these effects are selective, has begun to be submitted), as this habituation is characteristic for addressed in confrontations between a cat and rats. aggressive interactions in other species (e.g., Winslow When presented with a potential predator, a cat, male and Miczek 1984; Miczek et at. 1992). However, nei- and particularly female rats emit ultrasonic vocaliza- ther chlordiazepoxide nor alcohol "disinhibited" vocal tions, when in the presence of other colony members behavior that was suppressed in habituated opponents. (Blanchard et al. 1990, 1991, 1992). Morphine, at an Chtordiazepoxide and alcohol substantially increased analgesic dose, suppressed these antipredator vocal- threat peeps during the first minute of aggressive inter- izations, but even sedative doses of alcohol and actions when they were most frequent. Aggressive dis- diazepam left these calls unchanged (Blanchard et al. plays and threat peeps were maintained across a wide 1990; Shepherd et al. 1992; Blanchard, unpublished range of chlordiazepoxide doses, even though the high- observations). It appears that the antipredator vocal- est dose increased inactivity and produced ataxia. izations are important in the survival of the species, By contrast, vocalizations during feeding and affilia- assuming a functional significance beyond expressions tive behavior include peeps, twitter, chuck and cackle of "affect" or "distress", as these calls remain intact calls directed at familiar group members (Weerts, even in profoundly sedated individuals. Macey and Miczek, submitted). Under these condi- tions, peep vocalizations were not altered by benzodi- azepines including chlordiazepoxide. It is evident that Vocalizations during aggressive confrontations similarly structured calls are differentially altered by anxiolytic drugs depending on the social contexts. Vocalizations may represent signals that are critically These observations suggest that situational influences significant to sender as well as receiver, in addition to such as provocation and interaction with other indi- being affective expressions. Successful communication viduals within the social context are critical determi- of information during situations of conflict increases nants of benzodiazepine and alcohol effects on the survival of individuals as well as increasing the evo- vocalizations and social behavior. lutionary fitness of the species (Smith 1985). For exam- The production and structure of vocalizations as well ple, when confronting a rival, and prior to physical as subsequent behavioral responses are indeed context attack, dominant squirrel monkeys emit vocalizations specific. In pigtailed macaques and rhesus monkeys, such as cackle and harsh display calls (Jfirgens 1979). victims of aggression elicit structurally distinct vocal- It is possible to evoke these types of calls (and others) izations when attacked by a dominant vs. a subordi- by activating glutamate receptors or by blocking nate (Gouzoules et al. 1984), "recruiting" aide in a GABAA receptors in the periaqueductal grey region of rank-dependent manner from other group members or squirrel monkeys, even in the absence of social provo- kin. Playback studies indicate that vocalizations in the cation (Jiirgens and Lu 1993; Lu and Jfirgens 1993). absence of the appropriate contextual cues fail to elicit Drugs acting on the benzodiazepine-GABAa recep- predictable behavioral responses (Gouzoules et al. tor complex may interfere with the production and 1984; Biben and Symmes 1991). perception of social signals that communicate subor- dination or appeasement during social confrontations (e.g., Miczek et al. 1994). Initiation of aggressive behav- Conclusions iors can be enhanced in alcohol-treated dominant male monkeys, whereas subordinates treated with alcohol are Preclinical research on drugs that modulate normal and more often the rec!pient of aggressive interactions, pre- pathological affective processes can derive considerable sumably due to inappropriate social signalling (Miczek benefits from the study of vocalizations in animals. The et al. 1984; Winslow and Miczek 1985). validity of vocalizations as a focus for the study of anx- When adult male squirrel monkeys confront an unfa- iolytic and other mood-altering drugs derives primar- miliar male rival under experimental conditions, they ily from their occurrence in intense phases of display various visual and vocal threats (Weerts and reproduction, maternal separation, social conflict and Miczek, submitted), closely similar to those observed confrontations with predators and distressing events. in the field (Newman 1985) as well as to those elicited It has become apparent under which experimental con- by electrical brain stimulation of limbic structures ditions it is possible to engender different types of loud (Jfirgens 1988). Vocalizations that accompany aggres- and frequent vocalizations in laboratory rodents and sive displays are "noisier", and calls with wider band primates that correspond to those under field condi- frequency ranges have been interpreted to reflect greater tions. The methodological requirements for their accu- levels of arousal and aversion (Jfirgens 1982), as well rate recording and quantitative analysis have become as "distress" and "maximum intensity" aggressive dis- adequately established. In well-defined contexts, typi- plays (Newman 1985). Dominant squirrel monkeys' cally including an audience, certain vocalizations may threat vocalizations, especially peeps, peaked during the represent affective expressions which renders them 51 particularly attractive for research with anxiolytic and Barrett JE, Stanley JA, Brady LS, Mansbach RS, Witkin JM (1986) other mood-altering drugs. Pharmacological validity BehavioraI studies with anxiolytic drugs. II. Interactions of zopiclone with ethyl-beta-carboline-3-carboxylate and Ro has become apparent when, for example, benzodi- 15-1788 in squirrel monkeys. J Pharmacol Exp Ther azepines selectively attenuate a rat's ultrasonic vocal- 236:313-319 ization in an antagonist-reversible manner. Drugs Barrett JE, Gleeson S, Nader MA, Hoffmann SM (1989) acting on 5-HT receptors, particularly on the 1A and Anticonflict effects of the 5-HT~a compound flesinoxan. J 1B/D subtypes effectively suppress vocalizations by Psychopharmacol 3: 64-69 Barros HMT, Miczek K (1994) Withdrawal from oral cocaine: maternally separated pups or by adults rats that ultrasonic vocalizations and tactile startle. Neurosci Abstr 20: anticipate a social confrontation or aversive environ- 594 mental event. It appears quite feasible that experi- Berg WK, Davis M (1984) Diazepam blocks fear-enhanced startle mental conditions can be developed prompting vocal elicited electrically from the brainstem. Physiol Behav 32: expressions that are sensitive to drugs effective in the 333 336 Berkowitz L (1993) Aggression. Its causes, consequences and con- treatment of affective disturbances ranging from trol. Mc Graw Hill, New York panic to depression. In addition to communicating Biben M, Symmes D (1991) Playback studies of affiliative vocaliz- affective expressions, vocalizations may also represent ing in captive squirrel monkeys: familiarity as a cue to response. important messages that inform the receiver about crit- Behaviour 117:1-17 ical features of the sender. Many psychotropic Blanchard DC, Hori K, Rodgers R J, Hendrie CA, Blanchard RJ (1989) Attenuation of defensive threat and attack in wild rats drugs such as alcohol distort this communication and (Rattus rattus) by benzodiazepines. Psychopharmacology 97: thereby disrupt the patterns of social and aggressive 392-401 behavior. Blanchard RJ, Blanchard DC, Rodgers J, Weiss SM (1990) The characterization and modelling of antipredator defensive behav- ior. Neurosci Biobehav Rev 14:463-472 Acknowledgements We dedicate this review to our colleague Blanchard RJ, Blanchard DC, Agullana R, Weiss SM (1991) and friend Berend Olivier, Professor of Psychopharmacology, Twenty~two kHz alarm cries to presentation of a predator, by Rijksuniversiteit Utrecht, The Netherlands. This review is based on laboratory rats living in visible burrow systems. PhysioI Behav the Solvay Duphar Award lecture by K.A.M. at the 101st Annual 50:967 972 Convention of the American Psychology Association in Toronto, Blanchard RJ, Agullana R, McGee L, Weiss S, Blanchard DC Ontario, Canada. The research in our laboratory was supported by (1992) Sex differences in the incidence and sonographic charac- U.S.P.H.S. research grants AA05122 and DA02632. We are grate- teristics of antipredator ultrasonic cries in the laboratory rat ful for the comments by Drs. C.-E. Johanson, J.O. Valentine, (Rattus norvegicus). J Comp Psychol 106:270 277 A.M.M. van Erp, M. Haney, J.W, Tidey, W. Tornatzky, J.C. Cole, Blasig J, Herz A, Reinhold K, Zieglgansberger S (1973) A.M. van der Poel. Development of physical dependence on morphine in respect to time and dosage and quantification of the precipitated withdrawal syndrome in rats. Psychopharmacologia 33: 19-38 References Borszcz GS, Johnson CP, Fahey KA (1994) Comparison of motor reflex and vocalization thresholds following systemically admin- American Psychiatric Association (1994) Diagnostic and statistical istered morphine, fentanyl, and diazepam in the rat: assessment manual of mental disorders, Fourth Edition. Washington, of sensory and performance variables. PharmacoI Biochem DC Behav 49:827-834 Archer J (1988) The behavioural biology of aggression. Cambridge Brady JV (1956) Assessment of drug effects on emotional behav- University Press, Cambridge ior. Science 123:1033-1034 Ator NA (t991) Discriminative stimulus effects of the novel anxi- Brodie BB, Shore PA (1957) A concept for a role of serotonin and olytic buspirone. Behav Pharmacol 2:3-14 norepinephrine as chemical mediators in the brain. Ann NY Ator NA, Cook JM, Griffiths RR (1989) Drug discrimination in Acad Sci 66: 631-642 pentylenetetrazol-trained baboons: generalization to buspirone Brown GL, Goodwin FK, Ballenger JC, Goyer PF, Major LF and beta-carboline-3-carboxylic acid ethyl ester but not (1979) Aggression in humans correlates with cerebrospinal fluid lorazepam or pentobarbital. Drug Dev Res 16:257-267 amine metabolites, Psychiatry Res 1:131-139 Aubin T, Bremond JC (1992) Perception of distress call harmonic Carden S, Barr G, Hofer M (1990) Mu and delta opioid agonists structure by the starling (Sturnus vutgaris). Behaviour quiet isolation-induced vocalizations in rat pups. Soc Neurosci 120:151-163 Abstr 16:1192 Barr G, Wang S, Carden S (1994) Aversive properties of the k opi- Carden SE, Davachi L, Hofer MA (1994) U50, 488 increases ultra- old agonist U50, 488 in the week-old rat pup. Psycho- sonic vocalizations in 3-, 10-, and 18-day, old rat pups in isola- pharmacoIogy 113 : 422-428 tion and the home cage. Dev Psychobiol 27:65-83 Barrett JE (1976) Effects of alcohol, chlordiazepoxide, cocaine and Chopin P, Briley M (1987) Animal models of anxiety: the effect of pentobarbital on responding maintained under fixed-interval compounds that modify 5-HT neurotransmission. Trends schedules of food or shock presentation. J Pharmacol Exp Ther Pharmacol Sci 8 : 383-388 196:605-615 Coe CL, Smith ER, Levine S (1985) The endocrine system of Barrett JE (1992) Studies on the effects of 5-HT(la) drugs in the squirrel monkey. In : Rosenbtum LA, Coe CL (eds) Handbook pigeon. Drug Dev Res 26:299-317 of squirrel monkey research. Plenum Press, New York, pp Barrett JE, Vanover KE (1993) 5-HT receptors as targets for the 191-218 development of novel anxiolytic drugs: models, mechanisms and Costall B, Domeney A, Gerrard P, Kelly M, Naylor R (1988) future directions. Psychopharmacology 112:1-12 Zacopride: anxiolytic profile in rodent and primate models of Barrett JE, Witkin JM (1991) Buspirone in animal models of anx- anxiety. J Pharm PharmacoI 40:302-305 iety. In: Buspirone: mechanisms and clinical aspects. Academic Costall B, Jones BJ, Kelly ME, Naylor R J, Oaktey NR, Onaivi ES, Press, New York, pp 3%79 Tyers MB (1989) The effects of ondansetron (GR38032F) in rats 52

and mice treated subchronically with diazepam. Pharmacol Fletcher A, Cliffe IA, Dourish CT (1993) Silent 5-HTIA receptor Biochem Behav 34:769 778 antagonists: utility as research tools and therapeutic agents. Craft RM, Howard JL, Pollard GT (1988) Conditioned defensive Trends in Pharmaceutical Sciences 14:444-448 burying as a model for identifying anxiolytics. Pharmacol Gao B, Cutler MG (1993) Buspirone increases social investigation Biochem Behav 30:775-780 in pair-housed male mice - comparison with the effects of chlor- Crawley JN, Ninan PT, Pickar D, Chrousos GP, Linnoila M, diazepoxide. Neuropharmacology 32: 429-437 Skolnick P, Paul SM (1985) Neuropharmacological antagonism Garattini S, Giacalone E, Vatzelli L (t967) Isolation, aggressive- of the beta-carboline induced "anxiety" response in rhesus mon- ness and brain 5-hydroxytryptamine turnover. J Pharm keys. J Neurosci 5:47~485 Pharmacol 19:338-339 Cullen WK, Rowan MJ (1994) Gepirone and 1-(2-pyrimidinyl)- Gardner CR (1985) Distress vocalization in rat pups. A simple piperazine-induced reduction of aversively evoked ultrasonic screening method for anxiolytic drugs. J Pharmacol Methods vocalisation in the rat. Pharmacol Biochem Behav 48: 14:181-187 301-306 Gardner CR, Budhram P (1987) Effects of agents which interact Cuomo V, Cagiano R, De Salvia MA, Maselli MA, Renna G, with central benzodiazepine binding sites on stress-induced ultra- Racagni G(1988) Ultrasonic vocalization in response to sounds in rat pups. Era: J Pharmacol 134:275-283 unavoidable aversive stimuli in rats : effects of benzodiazepines. Gauvin DV, Holloway FA (1991) Cross-generalization between an Life Sci 43:485-491 ecologically relevant stimulus and a pentylenetetrazole-discrim- Davis M (1979) Diazepam and flurazepam: effects on conditioned inative cue. Pharmacol Biochem Behav 39:521-523 fear as measured with the potentiated startle paradigm. Geller I, Seifter J (1960) The effects of meprobamate, barbiturates, Psychopharmacology 62:1-7 d-amphetamine and promazine on experimentally induced Davis M (1986) Pharmacological and anatomical analysis of fear conflict in the rat. Psychopharmacologia 1:482-492 conditioning using the fear-potentiated startle paradigm. Behav Giacalone E, Tansella M, Valzelli L, Garattini S (1968) Brain sero- Neuro 100: 814-824 tonin in isolated aggressive mice. Biochem Davis M, Cassella JV, Kehne JH (1988) Serotonin does not medi- Pharmacol 17:1315-1327 ate anxiolytic effects of buspirone in the fear-potentiated startle Glowa JR, Bergman J, Insel T, Newman JD (1988) Drug effects on paradigm: comparisons with 8-OH-DPAT and ipsapirone. primate alarm vocalizations. In: Newman JD (ed) The physio- Psychopharmacology 94:14-20 logical control of mammalian volcalization. Plenum Press, New De Bold JF, Miczek KA (1985) modulates the effects York, pp 343 366 of ethanol on male mouse aggression. Psychopharmacology 86 : Glowa JR, Newman JD (1986) Benactyzine increases alarm call 286 290 rates in the squirrel monkey. Psychopharmacology 90:457-460 De Vry J (1995) 5-HTIA receptor agonists: recent developments and Gouzoules S, Gouzoules H, Marler P (1984) Rhesus money (Macaca controversial issues. Psychopharmacolog3~ (in press) muIatta) screams: representational signaling in the recruitment De Vry J, Schreiber R (1993) Comparison of acute and repeated of agonistic aid. Anita Behav 32:182-193 treatment with the 5-HT(1A)receptor ligands 8-OH-DPAT and Grant KA (1994) Emerging neurochemical concepts in the actions ipsapirone in animal models of anxiety and depression. Drug of ethanol at ligand-gated ion channels. Behav Pharmacol 5: Dev Res 30:91 103 383-404 De Vry J, Benz U, Schreiber R, Traber J (1993) Shock-induced Greenblatt DJ, Shader RI, Abernethy DR (1983) Current status of ultrasonic vocalization in young adult rats: a model for benzodiazepines. N Engl J Med 309:401-416 testing putative anti-anxiety drugs. Eur J Pharmacol Haefely WE (1988) Benzodiazepines. Int Anesthes Ctin 26:262--272 249 : 331-339 Haefely W, Martin JR, Schoch P (1990) Novel anxiolytics that act Dollard J, Doob L, Miller N, Mowrer O, Sears R (1939) Frustration as partial agonists at benzodiazepine receptors. Trends and aggression. Yale University Press, New Haven Pharmacol Sci 11:452-453 Eichelman B (1983) The limbic system and aggression in humans. Hammond DL (1989) Inference of pain and its modulation from Neurosci Biobehav Rev 7: 391-394 simple behaviors. In: Chapman CR, Loeser JD (eds) Advances Eichelman B (t987) Neurochemical and psychopharmacologic in pain research and therapy: issues in pain measurement (vol. aspects of aggressive behavior. In: Meltzer HY (ed) Psycho- 12). Raven Press, New York, pp 69-91 pharmacology: the third generation of progress. Raven Press, Handley SL (1991) Serotonin in animal models of anxiety: the New York, pp 697 704 importance of stimulus and response. In: Idzikowski C, Cowen Eichelman B, Hartwig A (1993) The clinical psychopharmacology PJ (eds), Serotonin, sleep and mental disorder. Wrightson of vioience: toward a nosology of human ag~essive behavior. Biomedical Publishing Ltd, Petersfield, pp 8%115 Psychopharmacol Bull 29: 57-63 Hanley SL, McBlane JW (1993) 5HT drugs in animal models of Emmett-Oglesby M, Spencer D Jr, Lewis M, Elmesallamy F, Lal anxiety. Psychopharmacology 112:13-20 H (1983) Anxiogenic aspects of diazepam withdrawal can be Handley SL, McBlane JW, Critchley MAE, Njunge K (1993) detected in animals. Eur J Pharmacol 92:127d30 Multiple serotonin mechanisms in animal models of anxiety: Emmett-Oglesby MW, Harris CM, Lane JD, Lal H (1984) environmental, emotional and cognitive factors. Behav Brain Withdrawal from morphine generalizes to a pentylenetetrazol Res 58:203-210 stimulus. Neuropeptides 5:37-40 Haney M, Miczek KA (1994) Uttrasounds emitted by female rats Estes WK, Skinner BF (1941) Some quantitative properties of anx- during agonsitic interactions: effects of morphine and nattrex- iety. J Exp Psychol 29:390-~400 one. Psychopharmacology 114:441-448 File SE (1985) Animal models for predicting clinical efficacy of Hard E, Engel J (1988) Effects of 8-OH-DPAT on ultrasonic vocal- anxiolytic drugs: social behaviour. Neuropsychobiology ization and audiogenic immobility reaction in pre-weanling rats. 13 : 55-62 Neuropharmacology 27 : 981-986 File SE, Andrews N (1991) Low but not high doses of buspirone Harris JC, Newman JD (1987) Mediation of separation distress by reduce the anxiogenic effects of diazepam withdrawal. Psycho- alpha2-adrenergic mechanisms in a non-human primate. Brain pharmacology 105:578 582 Res 410:353-356 File SE, Johnston AL (1989) Lack of effects of 5HT3 receptor antag- Harris JC, Newman JD (1988) Combined opiate/adrenergic :recep- onists in the social interaction and elevated plus-maze tests of tor blockade enhances squirrel monkey vocalization. Pharalacol anxiety in the rat. Psychopharmacology 99: 248-251 Biochem Behav 31:223-226 File SE, Pellow S (1986) Intrinsic actions of the benzodiazepine Hen R, Saudou F, Amara DA, Belzung C, Dierich A, LeMeur M, receptor antagonist Ro t5-1788. Psychopharmacology 88:1-11 Ramboz S, Segu L, Misstin R, Buhot MC (1994) Mice lacking 53

5-hydroxytryptamine 1B receptors display aggressive behavior. interoceptive stimulus produced by ethanol withdrawal. J Neurosci Abstr 20:1266 Pharmacol Exp Ther 247:508 518 Hess WR (t954) Das Zwischenhirn. Benno Schwabe & Co., Basel Levine JD, Feldmesser M, Tecott L, Gordon N C, Izdebski K (1984) Hoyer D, Boddeke HWGM (1993) Partial agonists, full agonists, Pain-induced vocalization in the rat and its modification by phar- antagonists: dilemmas of definition. Trends Pharmacol Sci 14: macological agents. Brain Res 296:121-127 270-275 Liljequist S, Engel JA (1984) The effects of GABA and benzodi- Hoyer D, Clarke DE, Fozard JR, Hartig PR, Martin GR, azepine receptor antagonists on the anti-conflict actions of Mylecharane EJ, Saxena PR, Humphrey PP (I994) VII. diazepam or ethanol. Pharmacol Biochem Behav 21:521 525 International Union of Pharmacology classification of receptors Linnoila M, Virkkunen M, Scheinin M, Nuutila A, Rimon R, for 5-hydroxytryptamine (serotonin). Pharmacot Rev 46: Goodwin FK (1983) Low cerebrospinal fluid 5-hydroxyin- 157-203 doleacetic acid concentration differentiates impulsive from non- Huntingford FA, Turner AK (1987) Animal conflict. Chapman and impulsive violent behavior. Life Sci 33:2609-2614 Hall, London Lorenz K (1966) On aggression. Methuen, London Insel TR (1989) Decreased in vivo binding to brain benzodiazepine Lu CL, Jiirgens U (1993) Effects of chemical stimulation in the peri- receptors during social isolation. Psychopharmacology 97: aqueductal gray on vocalization in the squirrel monkey. Brain 142-144 Res Bull 32:143-151 Insel TR, Hill JL, Mayor RB (1986) Rat pup ultrasonic isolation MacLean PD (1949) Psychosomatic disease and the "visceral brain". calls: possible mediation by the benzodiazepine receptor com- Recent developments bearing on the Papez theory of emotion. plex. Pharmacol Biochem Behav 24:1263-1267 Psychosom Med 11:338 Insel TR, Gelhard RE, Miller LP (1989) Rat pup isolation distress Mark VH, Ervin FR (1970) Violence and the brain. Harper and and the brain benzodiazepine receptor. Dev Psychobiol 5: Row, New York 509-525 Marler P (1976) On animal aggression: the roles of strangeness and Jtirgens U (1979) Vocalization as an emotional indicator neuro- familiarity. Am Psychol 31:239~46 ethological study in the squirrel monkey. Behaviour 69:88-117 Mendoza SP, Smotherman WP, Miner MT, Kaplan J, Levine S Jtirgens U (1982) Amygdalar vocalization pathways in the squirrel (t978) Pituitary-adrenal response to separation in mother and monkey. Brain Res 241 : t89-t96 infant squirrel monkeys. Dev Psychobiol 1 t : 16%175 Jfirgens U (1988) Central control of monkey calls. In: Todt D, Miczek KA (1974) Intraspecies aggression in rats: effects of d- Goedeking P, Symmes D (eds) Primate vocal communication. amphetamine and chtordiazepoxide. Psychopharmacologia 39: Springer, Berlin, pp 163-167 275-301 Jiirgens U, Lu CL (1993) The effects of periaqueductally injected Miczek KA (1987) The psychopharmacology of aggression. In: transmitter antagonists on forebrain-elicited vocalization in the Iversen LL, Iversen SD, Snyder SH (eds) Handbook of psy- squirrel monkey. Eur J Neurosci 5:735-741 chopharmacology, vol 19: new directions in behavioral phar- Jtirgens U, Ploog D (1981) On the neural control of mammalian macology. Plenum, New York, pp 183-328 vocalization. Trends Neurosci: 135-137 Miczek KA, Barry HIII (1977) Effects of alcohol on attack and Kalin NH, Shelton SE (1989) Defensive behavior in infant Rhesus defensive-submissive reactions in rats. Psychopharmacology 52: monkeys: environmental cues and neurochemical regulation. 231-237 Science 243 : 1718--1721 Miczek KA, Krsiak M (1979) Drug effects on agonistic behavior. Kalin NH, Shelton SE, Barksdale CM (1987) Separation distress In: Thompson T, Dews PB (eds) Advances in behavioral phar- in infant rhesus monekys: effects of diazepam and RO 15 1788. macology (vol. 2). Academic Press, New York, pp 87-162 Brain Res 408:192-198 Miczek KA, O'Donnell JM (1980) Alcohol and chlordiazepoxide Kalin NH, Shelton SE, Barksdale CM (1988) Opiate modulation increase suppressed agression in mice. Psychopharmacology 69 : of separation-induced distress in non-human primates. Brain Res 39-44 440: 285-292 Miczek KA, Vivian JA (1993) Automatic quantification of with- Kaltwasser MT (199I) Acoustic startle induced ultrasonic vocal- drawaI from 5-day diazepam in rats: ultrasonic distress vocal- ization in the rat - a novel animal model of anxiety. Behav Brain izations and hyperreflexia to acoustic startle stimuli. Res 43: t33-137 Psychopharmacology 110:37%382 Kehne JH, Cassella JV, Davis M (1988) Anxiolytic effects of bus- Miczek KA, Winstow JT, DeBold JF (t984) Heightened aggressive pirone and gepirone in the fear-potentiated startle paradigm. behavior by animals interacting with alcohol-treated con- Psychopharmacology 94: 8-13 specifics: studies with mice, rats and squirrel monkeys. Kehoe P, Shoemaker W (1991) Opioid-dependent behaviors in Pharmacol Biochem Behav 20:349-353 infant rats: effects of prenatal exposure to ethanol. Pharmacol Miczek KA, Mos J, Olivier B (1989) Brain 5-HT and inhibition of Biochem Behav 39:389-394 aggressive behavior in animals : 5-HIAA and receptor subtypes. Klint T (1991) Eft~cts of 8-OH-DPAT and buspirone in a passive Psychopharmacol Bull 25:399 403 avoidance test and in the elevated plus-maze test in rats. Behav Miczek KA, Thompson ML, Tornatzky W (1991a) Subordinate Pharmacol 2:481-489 animals: behavioral and physiological adaptations and opioid Krsiak M (1975) Timid singly-housed mice: their value in tolerance. In: Brown MR, Koob GF, Rivier C (eds) Stress : neu- prediction of psychtropic activity of drugs. Br J Pharmacol robiology and neuroendocrinology. Marcel Dekker, New York, 55:141-150 pp 323 357 Krsiak M, Sulcova A (1990) Differential effects of six structurally Miczek KA, Tornatzky W, Vivian JA (1991b) Ethology and neu- related benzodiazepines on some ethological measures of timid- ropharmacology: rodent ultrasounds. In: Olivier B, Mos J, ity, aggression and locomotion in mice. Psychopharmacology Stangen JL (eds) Animal models in psychopharmacology. 101:396-402 Birkhauser, Basel, pp 409-427 Lader MH, Petursson H (1981) Benzodiazepine derivatives: side Miczek KA, Weerts EM, Tornatzky W, DeBold JF, Vatne TM effects and dangers. Biol Psychiatry 16:1195-1201 (1992) Alcohol and "bursts" of aggressive behavior: ethologi- Lal H, Shearman GT (1982) Attenuation of chemically induced cal analysis of individual differences in rats. Psychopharmaco- anxiogenic stimuli as a novel method for evaluating anxiolytic logy 107:551-563 drugs: a comparison of clobazam with other benzodiazepines. Miczek KA, DeBold JF, Van Erp AMM (1994a) Neuro- Drug Dev Res Suppl 1 : 12%134 pharmacological characteristics of individual differences in Lal H, Harris CM, Benjamin D, Springfield AC, Bhadra S, Emmett- alcohol effects on aggression in rodents and primates. Behav Oglesby MW (1988) Characterization of a pentylenetetrazol-like PharmacoI 5 : 407-421 54

Miczek KA, Haney M, Tidey J, Vivian J, Weerts E (1994b) Olivier B, Mos J, Tulp MTM, Van der Poel AM (1992) Animal Neurochemistry and pharmacotherapeutic management of vio- models of anxiety and aggression in the study of serotonergic lence and aggression. In: Reiss A J, Miczek KA, Roth JA (eds) agents. In: Langer SZ, Brunello N, Racagni G, Mendlewicz J Understanding and preventing violence : biobehavioral (eds), Serotonin receptor subtypes : pharmacological significance influences on violence (vol. 2). National Academy Press, and clinical implications (vol. 1). Karger, Basel, pp 6%79 Washington, DC, pp 244-514 Panksepp J, Herman B, Conner R, Bishop P, Scott JP (1978) The Molander L (1982) Effect of melperone, chlorpromazine, haloperi- biology of social attachments: Opiates alleviate separation dis- dol, and diazepam on experimental anxiety in normal subjects. tress. Biol Psychiatry 13: 607-618 Psychopharmacology 77: 10% 113 Pecknold JC (1994) Serotonin 5-HT~a agonists: A comparative Molewijk HE, Van der PoeI AM, Mos J, Van der Heyden JAM, review. CNS drugs 2:234-251 OIivier B (1995) Conditioned ultrasonic distress vocalizations in Pellow S, Chopin P, File SE, Britey M (1985) Validation of adult male rats as a behavioural paradigm for screening anti- open:closed arm entries in an elevated plus-maze as a measure panic drugs. Psychopharmacology 117:32-40 of anxiety in the rat. J Neurosci Methods 14:14%167 Monroe RR (1978) Brain dysfunction in aggressive criminals. DC Ploog D (1989) Human neuroethology of emotion. Prog Heath and Company, Lexington Neuropsychopharmacol Biol Psychiatry 13:15-22 Montgomery KC (1955) The relation between fear induced by novel Przegalinski E, Chojnacka-Wojcik E, Filip M (1992) Stimulation stimulation and exploratory behaviour. J Comp Physiol Psychol of postsynaptic 5-HTIA receptors is responsible for the 48 : 254-260 anticonflict effect of ipsapirone in rats. J Pharm Pharmacol Mos J, Olivier B (1989) Ultrasonic vocalizations by rat pups as an 44: 780-782 animal model for anxiolytic activity: effects of serotonergic Pucilowski O, Kostowski W (1983) Aggressive behaviour and the drugs. In: Bevan P, Cools AR, Archer T (eds) Behavioural phar- central serotonergic systems. Behav Brain Res 9:33-48 macology of 5-HT. Lawrence Erlbaum, Hillsdale, pp 361-366 Redmond DE, Huang YH (1979) New evidence for the locus Mos J, Olivier B, Van der Poel AM (1987) Modulatory actions of coeruleus-norepinephrine connection with anxiety. Life Sci 25: benzodiazepine receptor ligands on agonistic behaviour. Physiol 2149-2162 Behav 41:265-278 Reite M, Short R, Seiler C, Pauley JD (1981) Attachment, loss and Mos J, Olivier B, Tulp MTHM (1992) Ethopharmacological stud- depression. J Child Psychol Psychiatry 22:141-169 ies differentiate the effects of various serotonergic compounds Rex A, Marsden CA, Fink H (1993) Effect of diazepam on corti- on aggression in rats. Drug Dev Res 26:343-360 cal 5-HT release and behaviour in the guinea pig on exposure Munizz C, Furlan PM, d'Elia A, D'Onofrio MR, Leggero P, Punzo to the elevated plus maze. Psychopharmacology 110:490-496 F, Vidini N, Villari V (1993) Emergency psychiatry: a review Rodgers R J, Cole JC, Cobain MR, Daly P, Doran PJ, Eells JR, of the literature. Acta Psychiatr Scand 88 [Supp. 374]: t-5t Wallis P (1992) Anxiogenic-like effects of and elto- Nastiti K, Benton D, Brain PF (1991) The effects of compounds prazine in the mouse elevated plus-maze: profile comparisons acting at the benzodiazepine receptor complex on the ultrasonic with 8-OH-DPAT, CGS 12066B, TFMPP and mCPP. Behav calling of mouse pups. Behav Pharmacol 2:121-128 Pharmacol 3: 621-634 Newman JD (1985) Squirrel monkey communication. In: Rodgers RJ, Waters AJ (1985) Benzodiazepines and their antago- Rosenblum LA, Coe CL (eds) Handbook of squirrel monkey nists: a pharmacoethological analysis with particular reference research. Plenum, New York, pp 9%125 to effects on "aggression". Neurosci Biobehav Rev 9:21-35 Newman JD (1988a) Ethopharmacology of vocal behavior in pri- Rowan M J, Cullen WK, Moulton B (t990) Buspirone impairment mates. In: Todt D, Goedeking P, Symmes D (eds) Primate vocal of performance of passive avoidance and spatial learning tasks communication. Springer, Berlin, pp 146-153 in the rat. Psychopharmacology 100:393-398 Newman JD (1988b) Investigating the physiological control of Sanchez C (1993) Effect of serotonergic drugs on footshock-induced mammalian vocalizations. In: Newman JD (ed) The physio- ultrasonic vocalization in adult male rats. Behav Pharmacol 4: logical control of mammalian vocalizations. Plenum Press, New 269-277 York, pp 1-5 Sanger DJ (1992) Increased rates of punished responding produced Newman JD, Smith HJ, Talmage-Riggs G (1983) Structural vari- by buspirone-like compounds in rats. J Pharmacol Exp Ther ability in primate vocalizations and its functional significance: 261:513-517 an analysis of squirrel monkey chuck calls. Folia Primatol 40 : Saudou F, Amara DA, Dierich A, LeMeur M, Ramboz S, Segu L, 114-124 Buhot MC, Hen R (1994) Enhanced aggressive behavior in mice Ninan PT, Insel TM, Cohen RM, Cook JM, Skolnick P, Paul SM lacking 5-HT1B receptor. Science 265:1875 1878 (1982) Benzodiazepine receptor-mediated "anxiety" in primates. Schefke DM, Fontana DJ, Commissaris RL (1989) Anti-conflict Science 218:1332-1334 efficacy of buspirone tbllowing acute versus chronic treatment. Olivier B, Van Aken H, Jaarsma I, van Oorshot R, Zethof T, Psychopharmacology 99: 427-429 Bradford D (1984) Behavioural effects of psychoactive drugs on Scherer KR (1986) Vocal affect expression: a review and a model agonistic behaviour of male territorial rats (resident-intruder for future research. PsychoI Bull 99:143-165 model). In: Miczek KA, Kruk MR, Olivier B (eds) Schweizer E, Rickels K (1986) Failure of buspirone to manage ben- Ethopharmacological aggression research. Alan R. Liss, New zodiazepine withdrawal. Am J Psychiatry t43: 1590-1592 York, pp 137-156 Sepinwall J, Cook L (1984) Behavioral pharmacology of antianxi- Olivier B, Mos J, Van Oorschot R (1985) Maternal aggression in ety drugs. In: Iverson LL, Iversen SD, Snyder SH (eds) rats: effects of chlordi~epoxide and fluprazine. Psychopharma- Handbook ofpsychopharmacology (vol. 13). Plenum Press, New cotogy 86: 68-76 York, pp 345--393 Olivier B, Mos J, Tulp M, Schipper J (1989) Modulatory action of Seyfarth RM, Cheney DL, Marter P (1980) Monkey responses to serotonin in aggressive behaviour. In: Archer T, Bevan P, Cools three different alarm calls: Evidence of predator classification A (eds) Behavioral pharmacology of 5- HT. Lawrence Erlbaum, and semantic communication. Science 2t0:801 803 Hittsdale, N.J., pp 8%116 Shepherd JK, Blanchard DC, Weiss SM, Rodgers R J, Blanchard Olivier B, Mos J, Rasmussen D (1990) Behavioural pharmacology RJ (1992) Morphine attenuates antipredator ultrasonic vocal- of the serenic, eltoprazine. Rev Drug Metab Drug Interact 8: izations in mixed-sex rat colonies. Pharmacol Biochem Behav 31-83 41:551-558 Olivier B, Mos J, Miczek KA (1991) Ethopharmacological studies Siegel A, Mirsky AF (1994) The neurobiology of violence and of anxiolytics and aggression. Eur Neuropsychopharmacol 1: aggression. In: Reiss A J, Miczek KA, Roth JA (eds) Under- 97-100 standing and preventing violence, biobehavioral influences on 55

vidence (vol 2). National Academy Press, Washington DC, pp violent offenders and impulsive fire setters. Arch Gen Psychiatry 59-172 46: 600 603 Skolnick P, Paul S (1982) Benzodiazepine receptors in the central Vivian JA, Miczek KA (1991) Ultrasounds during morphine with- . Int Rev Neurobiol 23:103-140 drawal in rats. Psychopharmacology 104:187-193 Smith WJ (1985) Consistency and change in communication. In: Vivian JA, Miczek KA (1993a) Diazepam and gepirone selectively Zivin G (ed) The development of expressive behavior. Biology- attenuate either 20-32 or 32 64 kHz ultrasonic vocalizations environment interactions. Academic Press, Orlando, pp 51-76 during aggressive encounters. Psychopharmacology 112: 66-73 S6derpalm B, Hj6rth S, Engel JA (1989) Effects of 5-HT~Areceptor Vivian JA, Miczek KA (1993b) Morphine attenuates ultrasonic agonists and L-5-HTP in Montgomery's conflict test. PharmacoI vocalizations during agonistic encounters in adult rats. Biochem Behav 32:259 265 Psychopharmacotogy 111 : 367-375 Struhsaker TT (1967) Auditory communication among vervet mon- Vivian JA, Farrell WJ, Sapperstein SB, Miczek KA (1994a) keys (Cercopithecus aethiops). In: Altmann SA (ed) Social com- Diazepam withdrawal: effects of diazepam and gepirone on munication among primates. University of Chicago Press, acoustic startle-induced 22 kHz ultrasonic vocalizations. Chicago, pp 281-324 Psychopharmacology 114:101-108 Sulcova A, Krsiak M (1984) The benzodiazepine-receptor antago- Vivian JA, Weerts EM, Miczek KA (1994b) Defeat engenders nist Ro 15-1788 antagonizes effects of diazepam on aggressive pentylenetetrazole-appropriate responding in rats. Antagonism and timid behaviour in mice. Activ Nerv Super 26:255-256 by midazolam. Psychopharmacology 117: 491-498 Sulcova A, Krsiak M (1989) Differences among nine 1,4-benzodi- Vogel JR, Beer B, Clody DE (1971) A simple and reliable conflict azepines: an ethopharmacological evaluation in mice. Psycho- procedure for testing anti-anxiety agents. Psychopharmacologia pharmacology 97: t 57-159 21:1-7 Suomi SJ (1991) Primate separation models of affective disorders. Wada T, Fukuda N (1991) Effects of DN-2327, a new anxiolytic, In: Madden J (ed) Neurobiology of learning, emotion and affect. diazepam and buspirone on exploratory activity of the rat in an Raven Press, New York, pp 195-.2t4 elevated plus-maze. Psychopharmacology 104:444-450 -- Symmes D, Newman JD, Talmage-Riggs G, Lieblich AK (1979) Weerts EM, Miller LG, Hood KE, Miczek KA (1992) Increased Individuality and stability of isolation peeps in squirrel mon- GABA A-dependent chloride uptake in mice selectively bred for keys. Anita Behav 27:1142-1152 low aggressive behavior. Psychopharmacology 108 : 196--204 Tardiff K (1983) A survey of drugs used in the management of Weerts EM, Tornatzky W, Miczek KA (I993a) "Anxiolytic" and assaultive inpatients. Bull Am Acad Psychiatr Law 1 t:215 222 "anxiogenic" benzodiazepines and beta-carbolines: effects on Tonoue T, Ashida Y, Makino H, Hata H (1986) Inhibition of shock- aggressive and social behavior in rats and squirrel monkeys. elicited ultrasonic vocalization by opioid peptides in the rat: a Psychopharmacology 110: 451-459 psychotropic effect. Psychoneuroendocrinology 11:177-184 Weerts EM, Tornatzky W, Miczek KA (1993b) Prevention of the Tornatzky W, Miczek KA (1994) Behavioral and autonomic proaggressive effects of alcohol by benzodiazepine receptor responses to intermittent social stress: differential effects ofcloni- antagonists in rats and in squirrel monkeys. Psychopharma- dine and metoprolol. Psychopharmacology 116:346-356 cology 111:144-152 Tornatzky W, Miczek KA (1995) Alc0hol, anxiolytics and Social Wiener SG, Coe CL, Levine S (1988) Endocrine and neurochemi- stress in rats. Psychopharmacology 121:135 144 cal sequelae of primate vocalizations. In: Newman JD (ed) The Treit D (1987) Ro 15-1788, CGS 8216, picrotoxin, and pentylenete- physiological control of mammalian vocalizations, Plenum trazol: do they antagonize through an anxiogenic action? Brain Press, New York, pp 367-394 Res BulI 19:401-405 Wilkinson LO, Dourish CI" (1991) Serotonin and animal behavior. Treit D, Fundytus M (1988) A comparison of buspirone and chlor- In: Serotonin receptor subtypes: basic and clinical aspects. diazepoxide in the shock-probe/burying test for anxiolytics. Wiley-Liss, New York, pp 147~10 Pharmacol Biochem Behav 30:1071-I075 Winslow JT, Insel TR (1990a) Serotonergic and catecholaminergic Treit D, Pinel JPJ, Fibiger HC (1981) Conditioned defensive bury- reuptake inhibitors have opposite effects on the ultrasonic iso- ing: a new paradigm for the study of anxiolytic agents. lation calls of rat pups. Neuropsychopharmacology 3:51-59 Pharmacol Biochem Behav 15:6t%626 Winslow JT, Insel TR (1990b) Serotonergic modulation of rat pup Valzelli L (1982) Serotonergic inhibitory control of experimental ultrasonic vocal development - studies with 3, 4-methylene- aggression. Pharmacol Res Commun 14:1-13 dioxymethamphetamine. J Pharmacol Exp Ther 254:212-220 Valzelli L, Bernasconi S (1979) Aggressiveness by isolation and Winslow JT, Insel TR (1991a) Endogenous opioids, do they mod- brain serotonin turnover changes in different strains of mice. ulate the rat pup's response to social isolation? Behav Neurosci Neuropsychobiology 5:129-135 105:31 41 Valzelli L, Galateo E (1984) Serotonergic control of experimental Winslow JT, Insel TR (1991 b) The infant rat separation paradigm: aggression. Pol J Pharmacol Pharm 36:495-503 a novel test for novel anxiolytics. Trends Pharmacol Sci 12: Valzelli L, Garattini S (1968) Behavioral changes and 5-hydroxy- 402-404 tryptamine turnover in animals. Adv Pharmacol 6B:249 260 Winslow JT, Miczek KA (1984) Habituation of aggressive behav- Van der Poel AM, Noach EJK, Miczek KA (1989) Temporal pat- ior in mice: a parametric study. Aggress Behav 10:103-113 terning of ultrasonic distress calls in the adult rat: Effects of mor- Winslow JT, Miczek KA (1985) Social status as determinant of phine and benzodiazepines. Psychopharmacology 97:147-148 alcohol effects on aggressive behavior in squirrel monkeys Van der Poe1 AM, Miczek KA (1991) Long ultrasonic calls in ( Saimiri sciureus). Psychopharmacology 85:167-172 male rats following mating, defeat and aversive stimulation: Winter P (1968) Social communication in the squirrel monkey. In: frequency modulation and bout structure. Behaviour Rosenblnm LA, Cooper RW (eds) The squirrel monkey. 119:127-142 Academic Press, New York, pp 235-253 Vellucci SV, File SE (1979) Chlordiazepoxide loses its anxiolytic Wood DM, Laraby PR, Lal tI (I989) A pentylenetetrazoMike stim- action with long-term treatment. Psychopharmacology 62: 61-65 ulus during cocaine withdrawal: blockade by diazepam but not Vellucci SV, Martin PJ, Everitt BJ (1988) The discriminative stim- haloperidol. Drug Dev Res 16:269-276 ulus produced by pentylenetetrazol: effects of systematic anxi- Yajima Y, ttayashi Y, Yoshii N (1980) The midbrain central olytics and anxiogenics, aggressive defeat and midazolam or gray substance as a highly sensitive neural structure for the pro- muscimol infused into the amygdala. J Psychopharmacol 2: duction of ultrasonic vocalization in the rat. Brain Res 80-93 t 98: 446-452 Virkkunen M, De Jong J, Barko J, Goodwin FK, Linnoila M (1989) Yajima Y, Hayashi Y, Yoshi N (1981) Identification of ultrasonic Relationship of psychobiological variables to recidivism in vocalization substrates determined by electrical stimulation 56

applied to the medula oblongata in the rat. Brain Res 229: Zifa E, Fillion G (1992) 5-Hydroxytryptaminereceptors. Pharmacot 353 362 Rev 44: 401-457 Yoshimura H (1987) Studies contrasting drug effects on reproduc- Zimmerberg B, Brunetli SA, Hofer MA (1994) Reduction of rat tion induced agonistic behavior in male and female mice. In: pup ultrasonic vocalizations by the neuroactive steroid allo- Olivier B, Mos J, Brain PF (eds) Ethopharmacology of agonis- pregnanolone. Pharmacol Biochem Behav 47:735-738 tic behavior in animals and humans. Martinus Nijhoff, Dordrecht, pp 94~ 109 Young WS, Kuhar MJ (1980) Radiohistochemical localization of benzodiazepine receptors in rat brain. J Pharmacol Exp Ther 212:337-346