Midbrain pathways for prepulse inhibition and startle activation in rat.

The midbrain is essential for prepulse inhibition (PPI) of the startle reflex, but the exact neural circuits for PPI are not yet determined. Electrical stimulation of the superior colliculus (SC) or pedunculopontine tegmentum was used to characterize the neurons and pathways that mediate PPI and the activation of startle that also occurs at higher currents in the same sites. Startle was inhibited by prepulses in most, but not all SC sites, with the lowest intensity sites in intermediate layers of SC. PPI latencies in SC sites were 4-6 ms longer than in inferior colliculus, intercollicular nucleus or pedunculopontine sites. Contrary to previous serial models, there must be two parallel midbrain pathways for PPI, a faster auditory pathway from inferior colliculus to pedunculopontine tegmentum, and a slower multimodal SC output for PPI. Double-pulse stimulation of SC sites shows that PPI results from direct stimulation of neurons with moderate refractory periods (0.4-1.0 ms), similar to SC neurons that mediate contraversive turning responses. By contrast, startle activation occurring at higher currents in all SC sites (even sites where PPI could not be elicited) results from stimulation of very short refractory period neurons (0.3-0.5 ms) and very long refractory period neurons (1.0-2.0 ms), with startle inhibition often found from 0.5-1.0 ms. Startle activation appears to result from stimulation of short refractory period neurons in deep SC layers that mediate fear-potentiated startle, plus long refractory period substrates in more dorsal SC sites.

Midbrain muscarinic receptors modulate morphine-induced accumbal and striatal dopamine efflux in the rat.

Midbrain dopamine neurons are critical in mediating the rewarding effects of opiates in dependent rats, as well as modulating some manifestations of opiate withdrawal. Morphine is known to excite dopamine neurons and thereby facilitate forebrain dopamine transmission through inhibition of GABA neurons. Cholinergic neurons in the mesopontine laterodorsal and pedunculopontine tegmental nuclei provide the principal source of excitatory cholinergic input to ventral tegmental area and substantia nigra pars compacta dopamine-containing neurons, via actions on midbrain muscarinic and nicotinic acetylcholine receptors. The present study hypothesized that a reduction in tonic cholinergic input via blockade of midbrain muscarinic receptors would reduce the pharmacological effects of morphine on forebrain dopamine release. Using in vivo chronoamperometry, alterations in morphine-evoked dopamine efflux were monitored at stearate-graphite paste electrodes implanted unilaterally in the nucleus accumbens and striatum of urethane (1.5 g/kg) anesthetized rats, following the pharmacological inhibition of ventral tegmental area/substantia nigra pars compacta muscarinic receptors. The facilitatory effects of morphine (2.0 mg/kg, i.v.) on accumbens and striatal dopamine efflux were markedly reduced by prior infusion of the non-selective muscarinic receptor antagonist scopolamine (200 microg/microl) into the ventral tegmental area or substantia nigra pars compacta, respectively. These findings demonstrate that decreased activation of midbrain muscarinic receptors attenuates the excitatory effects of morphine on mesoaccumbens and nigrostriatal dopaminergic transmission.

Kynurenate in the pontine reticular formation inhibits acoustic and trigeminal nucleus-evoked startle, but not vestibular nucleus-evoked startle.

The startle reflex is elicited by acoustic, trigeminal or vestibular stimulation, or by combinations of these stimuli. Acoustic startle is mediated largely by ibotenate-sensitive neurons in the ventrocaudal pontine reticular formation (PnC). In these studies we tested whether startle elicited by stimulation of different modalities is affected by infusion of the non-selective glutamate antagonist, kynurenate, into the PnC. In awake rats, startle responses evoked by either acoustic or spinal trigeminal nucleus stimulation were inhibited by kynurenate, but not saline, infusions, with the most effective placements nearest PnC. In chloral hydrate-anesthetized rats, kynurenate in the PnC reduced trigeminal nucleus-evoked hindlimb EMG responses, but not vestibular nucleus-evoked startle. Kynurenate in the vestibular nucleus had no effect on trigeminal nucleus-evoked startle. These results indicate that trigeminal nucleus stimulation evokes startle largely through glutamate receptors in the PnC, similarly to acoustic startle, but vestibular nucleus-evoked startle is mediated through other pathways, such as the vestibulospinal tract.

Contributions of the vestibular nucleus and vestibulospinal tract to the startle reflex.

The startle reflex is elicited by strong and sudden acoustic, vestibular or trigeminal stimuli. The caudal pontine reticular nucleus, which mediates acoustic startle via the reticulospinal tract, receives further anatomical connections from vestibular and trigeminal nuclei, and can be activated by vestibular and tactile stimuli, suggesting that this pontine reticular structure could mediate vestibular and trigeminal startle. The vestibular nucleus, however, also projects to the spinal cord directly via the vestibulospinal tracts, and therefore may mediate vestibular startle via additional faster routes without a synaptic relay in the hindbrain. In the present study, the timing properties of the vestibular efferent pathways mediating startle-like responses were examined in rats using electrical stimulation techniques. Transient single- or twin-pulse electrical stimulation of the vestibular nucleus evoked bilateral, startle-like responses with short refractory periods. In chloral hydrate-anesthetized rats, hindlimb electromyogram latencies recorded from the anterior biceps femoris muscle were shorter than those for stimulation of the trigeminal nucleus, and similar to those for stimulation of the caudal pontine reticular nucleus or ventromedial medulla. In awake rats, combining vestibular nucleus stimulation with either acoustic stimulation or trigeminal nucleus stimulation enhanced the whole-body startle-like responses and led to strong cross-modal summation without collision effects. In both chloral hydrate-anesthetized and awake rats, combining vestibular nucleus stimulation with ventromedial medulla stimulation produced a symmetrical collision effect, i.e. a loss of summation at the same positive and negative stimulus intervals, indicating a continuous connection between the vestibular nucleus and ventromedial medulla in mediating vestibular startle. By contrast, combining trigeminal nucleus stimulation with ventromedial medulla stimulation resulted in an asymmetric collision effect when the trigeminal nucleus stimulation preceded ventromedial medulla stimulation by 0.5 ms, suggesting that a monosynaptic connection between the trigeminal nucleus and ventromedial medulla mediates trigeminal startle. We propose that the vestibulospinal tracts participate strongly in mediating startle produced by activation of the vestibular nucleus. The convergence of the vestibulospinal tracts with the reticulospinal tract within the spinal cord therefore provides the neural basis of cross-modal summation of startling stimuli.

Brain stem circuits mediating prepulse inhibition of the startle reflex.

RATIONALE: Prepulse inhibition (PPI) of the startle reflex occurs when brief, non-startling tactile, acoustic or visual stimuli are presented 20-500 ms before the startling stimulus. OBJECTIVE: To review information about PPI-mediating brain stem circuits and transmitters, and their functions. RESULTS: Midbrain systems are most critical for the fast relay of these PPI stimuli. Acoustic prepulses for PPI are relayed through the inferior colliculus (IC). The superior colliculus (SC) is important for acoustic PPI, and may be important for the mediation of tactile and visual prepulses. This collicular activation for PPI is quickly relayed through the pedunculopontine tegmental nucleus (PPTg), with lesser contributions to PPI from the laterodorsal tegmental nucleus (LDTg) and substantia nigra, pars reticulata (SNR). The transient activation of midbrain nuclei by PPI stimuli is converted into long-lasting inhibition of the giant neurons of the caudal pontine reticular nucleus (PnC). We propose that muscarinic and GABA(B) inhibitory receptors (both metabotropic receptors) on PnC giant neurons combine to produce the long-lasting inhibition of startle. Activation of mesopontine cholinergic neurons leads to cortical arousal, turning and exploratory approach responses. CONCLUSION: PPI is mediated by a circuit involving the IC, SC, PPTg, LDTg, SNR and PnC. By reducing startle, PPI allows the execution of approach responses and perceptual processing following salient stimuli.

Conditioned brain-stimulation reward attenuates the acoustic startle reflex in rats.

The acoustic startle reflex (ASR) in rats is attenuated by a light paired with food or, in humans, by "pleasant" pictures. Rats were trained to barpress for lateral hypothalamus (LH) stimulation. ASR amplitudes were then measured at 4 intensities, with or without a light. Control rats that did not receive brain-stimulation reward (BSR) showed initially lower ASR amplitudes than did rats exposed to BSR, but both groups responded similarly with or without light. Next, experimental rats were given BSR in the presence of light but not in its absence. After conditioning, ASR amplitudes were reduced, and ASR thresholds were raised by a mean of 2.6 dB in the light but remained at preconditioning levels without light. No such change was found for control rats or rats with placements outside the LH.

Using intracranial electrical stimulation to study the timing of prepulse inhibition of the startle reflex.

Due to the short latency and briefness of the startle reflex, event-related inhibition of startle has high temporal resolution and is useful for studying the hierarchical organization of sensorimotor gating and motive-motor gating. In this article, we describe methods for measuring the inhibitory effects of electrically stimulating each of the following four brain structures on startle in awake rats: the inferior colliculus (IC), the deeper layers of the superior colliculus (SC), the pedunculopontine tegmental nucleus (PPTg), and the ventral pallidum (VP). These four brain structures have been reported to be important in mediating sensorimotor or motive-motor gating. Startle responses are elicited by either intense noise bursts or electrical stimulation of the principal trigeminal nucleus. The time course of the IC-inhibited startle reflex is used as a standard for estimating timing of the neural transfer of startle-inhibitory information to motor outputs. We also discuss how these methods can be used in combination with neuropharmacology.

Role of the laterodorsal tegmental nucleus in scopolamine- and amphetamine-induced locomotion and stereotypy.

Scopolamine (1.5 mg/kg; i.p.) or amphetamine (3 mg/kg; i.p.) increases locomotion and stereotyped behavior patterns in rats. Previous studies suggest that scopolamine acts via muscarinic receptors near the midbrain-pons border. In this study, unilateral microinjections in N-methyl-scopolamine (2.5-10 microg) into the laterodorsal tegmental nucleus (LDT) increased locomotion. Bilateral ibotenate lesions of the LDT attenuated scopolamine-induced locomotion by 68% 7 days postlesion, and by 35% 28 days postlesion. LDT lesions reduced scopolamine-induced stereotypy less than locomotion. The sensitization to amphetamine observed on repeated tests was attenuated by LDT lesions for stereotypy, but not for locomotion. These findings suggest that scopolamine induces locomotion largely, but not exclusively, by blocking muscarinic receptors in LDT.

Brain-stimulation reward thresholds raised by an antisense oligonucleotide for the M5 muscarinic receptor infused near dopamine cells.

Oligonucleotides targeting M5 muscarinic receptor mRNA were infused for 6 d into the ventral tegmental area of freely behaving rats trained to bar-press for lateral hypothalamic stimulation. The bar-pressing rate was determined at a range of frequencies each day to evaluate the effects of infusions on reward. M5 antisense oligonucleotide (oligo) infusions increased the frequency required for bar pressing by 48% over baseline levels, with the largest increases occurring after 4-6 d of infusion. Two control oligos had only slight effects (means of 5 and 11% for missense and sense oligos, respectively). After the infusion, the required frequency shifted back to baseline levels gradually over 1-5 d. Antisense oligo infusions decreased M5 receptors on the ipsilateral, but not the contralateral, side of the ventral tegmentum, as compared with a missense oligo. Therefore, M5 muscarinic receptors associated with mesolimbic dopamine neurons seem to be important in brain-stimulation reward.

Effects of bilateral electrical stimulation of the ventral pallidum on acoustic startle.

The ventral pallidum (VP) is believed to occupy a critical position between the limbic and the motor systems, for transferring motive information into motor commands. To estimate the time course of signaling from the VP to motor outputs, in the present study we examined the effects of bilateral electrical stimulation of the VP on the acoustic startle reflex in awake rats. When the interstimulus interval (ISI) between VP stimulation and acoustic stimulation was shorter than 5 ms, VP stimulation potentiated acoustic startle. When the ISI was longer than 5 ms, VP stimulation inhibited acoustic startle over a large range of ISIs with the maximum inhibition at ISIs between 15 and 25 ms. In contrast, bilateral electrical stimulation of the amygdala did not have a significant inhibitory effect on acoustic startle, but strongly augmented acoustic startle at shorter ISIs (0-10 ms). Compared to unilateral electrical stimulation of the inferior colliculus (IC), bilateral stimulation of the VP gave rise to a rightward shift of the ISI curve, indicating that the neural pathways conveying the inhibitory influence from the VP to the acoustic startle circuit are longer than those from the IC.

Cochlear and trigeminal systems contributing to the startle reflex in rats.

The startle reflex is evoked by strong acoustic or tactile stimuli, or by electrical stimulation of acoustic or tactile pathways. To dissociate the contributions of acoustic and tactile pathways, stimulating electrodes were placed in adjacent cochlear and trigeminal nuclei. The currents needed to evoke startle-like responses were an order of magnitude lower in ventral trigeminal sites (12-80 microA for a 0.1-ms pulse) than in cochlear nucleus sites (150-800 microA). At low threshold sites in both areas, brief acoustic stimuli were followed 0-4 ms later by a single electrical pulse and the current required to evoke startle was measured at several interstimulus intervals. Summation between the acoustic and electrical stimuli for startle was strong in both cochlear and trigeminal sites. Collision effects were found in the anteroventral cochlear nucleus when the electrical stimulus followed the ipsilateral acoustic stimulus by 2.0 ms, suggesting that acoustic startle is mediated by axons in the anteroventral cochlear nucleus. Collision effects were found at 4.0 ms if the electrical stimulus was presented in the contralateral pontine reticular formation, suggesting that acoustic signals mediating startle mainly cross to the pontine reticular formation. Collision effects were not found in medial or posterior sites in the cochlear nucleus, or trigeminal sites, suggesting that the neurons that mediate startle in these sites do not mediate acoustic startle. Therefore, acoustic startle is mediated through high threshold cochlear nucleus sites, while low threshold sites are non-acoustic, probably as a result of trigeminal or vestibular stimulation.

Summation between acoustic and trigeminal stimuli evoking startle.

Electrical stimulation of the spinal trigeminal pathway evokes a short-latency startle-like response in rats. To explore the relationship between acoustic and tactile systems mediating startle, we studied temporal summation between pairs of startle-evoking stimuli in awake rats by varying the interstimulus interval. The stimuli were: (i) two noise bursts; (ii) two unilateral electrical stimuli near the principal nucleus of the trigeminal nerve; (iii) electrical stimulation of the left and right trigeminal nucleus; or (iv) a noise burst and unilateral stimulation of the trigeminal nucleus. Following two noise bursts, the amplitude of startle increased as the interval increased from 0 to 4 ms, then declined smoothly as the interval increased to 15 ms. Unilateral stimulation of the trigeminal nucleus resulted in a sharper summation effect, with maximal summation at 2 ms, and refractory periods estimated at 0.4-0.8 ms. Bilateral stimulation of the trigeminal nucleus resulted in broader summation without a refractory period, and maximal summation when the stimuli on both sides of the trigeminal nucleus were presented simultaneously. The combination of acoustic and trigeminal stimulation was most effective in enhancing startle amplitudes, and summation peaked when the noise burst preceded the trigeminal stimulation by 5 ms. Similarly, electromyogram latencies measured in the hindlimb were 3-4 ms shorter for trigeminal stimulation than for the noise burst. Startle appears to be optimally activated by simultaneous acoustic and tactile stimuli, as occurs during head blows.

Prepulse inhibition of acoustic or trigeminal startle of rats by unilateral electrical stimulation of the inferior colliculus.

Transient electrical stimulation of the inferior colliculus (IC) of adult male rats had a strong and long-lasting inhibitory effect on startle responses elicited by either intense noise bursts or unilateral electrical stimulation of the principal trigeminal nucleus. Startle elicited by noise bursts was inhibited over a wide range of interstimulus intervals (ISIs) with the maximum inhibition at ISIs between 15 and 30 ms. Startle elicited by trigeminal stimulation was inhibited more sharply than acoustic startle, with the maximum inhibition at ISIs between 20 and 35 ms. These data support the view that the IC is a critical part of the pathway mediating prepulse inhibition (D. S. Leitner & M. E. Cohen, 1985). More important, the data reveal the time course of the inhibitory influence of the IC on startle and indicate that the inhibitory effects of IC stimulation have higher temporal resolution on trigeminal startle than on acoustic startle.

Activation of amygdala cholecystokininB receptors potentiates the acoustic startle response in the rat.

The acoustic startle reflex is a sensitive index of "anxiety" and "fear." Potentiation of startle by conditioned and unconditioned fear stimuli appears to be mediated by the amygdala. CholecystokininB (CCKB) agonists increase "anxiety" in laboratory animals and induce "panic" in humans. Here, we investigate the role CCKB receptor-mediated mechanisms in the amygdala in the potentiation of startle. First, intra-amygdala infusions of the CCKB receptor agonist pentagastrin (0, 0.01, 0.1, 1, and 10 nM) produced a dose-related potentiation of acoustic startle responses. At the highest dose, startle amplitudes were increased up to 90% above preinfusion baseline levels. Second, similar infusions of pentagastrin had no effect on locomotor activity over the same time course, showing that increases in startle responsivity after infusions of pentagastrin are not attributable to nonspecific changes in motor activity. Third, infusions of similar doses of pentagastrin into the striatum or nucleus accumbens did not potentiate startle responses. Fourth, pretreatment with the CCKB receptor antagonist L-365,260 (0.1 mg/kg, i.p.) attenuated the potentiation of startle produced by intra-amygdala infusions of pentagastrin. Finally, intra-amygdala infusion of the CCKB receptor-selective antagonist PD-135158 (10 micro;g) blocked the potentiation of startle produced by i.c.v. infusions of pentagastrin, suggesting that i.c.v. infusions of pentagastrin potentiate startle responses via activation of amygdala CCKB receptors. These results show that amygdala CCKB receptor-mediated mechanisms are involved in the potentiation of acoustic startle responses.

Increased striatal dopamine efflux follows scopolamine administered systemically or to the tegmental pedunculopontine nucleus.

The cholinergic cells of the tegmental pedunculopontine nucleus monosynaptically excite dopaminergic neurons of the substantia nigra. In vivo electrochemical methods were used to monitor dorsal striatal dopamine efflux in awake rats following intraperitoneal scopolamine injections and following the direct application of scopolamine to the tegmental pedunculopontine nucleus. Systemic injections of scopolamine (1.0, 3.0 or 10.0 mg/kg) resulted in dose-related increases in peak striatal dopamine oxidation currents of between 1.1 and 2.0 nA. Increases began within 10-20 min after injection and peaked after 40-90 min. Unilateral microinjections of scopolamine into the tegmental pedunculopontine nucleus (10, 50 or 100 micrograms/0.5 microliter) resulted in dose-related increases in dopamine oxidation currents that peaked 60-90 min postinjection (2.9-5.0 nA). Carbachol (4.0 micrograms/0.5 microliter) injected unilaterally into the tegmental pedunculopontine nucleus 20 min before 100 micrograms tegmental pedunculopontine nucleus scopolamine, or injected bilaterally 20 min before 3.0 mg/kg systemic scopolamine, attenuated the increases produced by scopolamine alone. The carbachol preinjection tests suggest that the effects of both systemic and tegmental pedunculopontine nucleus scopolamine treatments are mediated largely by muscarinic receptors near the tegmental pedunculopontine nucleus. These findings are consistent with the proposal that enhanced activation of substantia nigra dopamine cells results from scopolamine-induced disinhibition of the tegemental pedunculopontine nucleus cholinergic cell group via blockade of their inhibitory autoreceptors.

Locomotion and stereotypy induced by scopolamine: contributions of muscarinic receptors near the pedunculopontine tegmental nucleus.

In this study, we test whether blockade of muscarinic receptors near mesopontine cholinergic cell groups may contribute to locomotor activation induced by scopolamine. Unilateral or bilateral injections of scopolamine (10-150 micrograms) into the pedunculopontine tegmental nucleus (PPT) increased horizontal locomotion by 2-15 times in a dose-related way. Unilateral or bilateral injections of scopolamine into the PPT increased stereotypic behaviors (such as sniffing in one location or over large areas), self-biting and grooming. Carbachol (4 micrograms) injected into PPT reduced locomotion for 20 min, followed by 70 min of increased locomotion. When carbachol (4 micrograms) was injected into the PPT before scopolamine (3 mg/kg, i.p.), the activating effect of scopolamine was attenuated, but not when carbachol was injected after scopolamine. Therefore, carbachol appears to compete with scopolamine for muscarinic receptors near the PPT that mediate locomotor activating effects of systemic scopolamine. Haloperidol (0.1 mg/kg, i.p.) also attenuated the stereotypy and locomotion induced by scopolamine in the PPT. We hypothesize that scopolamine acts by blocking muscarinic receptors on mesopontine cholinergic neurons, thereby disinhibiting cholinergic neurons that can activate dopamine neurons.

Intracerebroventricular infusion of the CCKB receptor agonist pentagastrin potentiates acoustic startle.

The acoustic startle reflex is increased by stimuli associated with aversive events (such as the delivery of shock) and so has been used as a sensitive index of 'anxiety' or 'fear'. Administration of cholecystokininB (CCKB) receptor agonists produces a constellation of behaviors associated with 'anxiety' in laboratory animals and humans. Here, intracerebroventricular infusions of the CCKB agonist, pentagastrin (0, 1, 10, 100 nM), produced a long-lasting, dose-related potentiation of acoustic startle responses. Similar infusions of pentagastrin had no effect on locomotor activity over the same time course, showing that changes in startle responses following infusions of pentagastrin are not due to nonspecific changes in motor activity.

The acoustic startle reflex: neurons and connections.

The startle reflex protects animals from blows or predatory attacks by quickly stiffening the limbs, body wall and dorsal neck in the brief time period before directed evasive or defensive action can be performed. The acoustic startle reflex in rats and cats is mediated primarily by a small cluster of giant neurons in the ventrocaudal part of the nucleus reticularis pontis caudalis (RPC) of the reticular formation. Activation of these RPC neurons occurs 3-8 ms after the acoustic stimulus reaches the ear. Undetermined neurons of the cochlear nuclei activate RPC via weak monosynaptic and strong disynaptic connections. The strong disynaptic input occurs via neurons of the contralateral ventrolateral pons, including large neurons of the ventrolateral tegmental nucleus that integrate auditory, tactile and vestibular information. RPC giant neurons, in turn, activate hundreds of motoneurons in the brain stem and the length of the spinal cord via large reticulospinal axons near the medial longitudinal fasciculus. To hindlimb motoneurons, monosynaptic connections from the reticulospinal tract are weak, but disynaptic connections via spinal cord interneurons are stronger and show temporal facilitation, like the startle response itself.

Median raphe injections of 8-OH-DPAT lower frequency thresholds for lateral hypothalamic self-stimulation.

The selective 5-HT1A agonist 8-hydroxy-2-(di-n-propylamino)tetralin (8-OH-DPAT) reduces the activity of brain 5-HT neurons via somatodendritic autoreceptors located in the midbrain raphe nuclei. This action of 8-OH-DPAT results in reduced 5-HT synthesis and release of 5-HT in terminal regions. Previous studies have shown that injecting 8-OH-DPAT into the raphe nuclei stimulates feeding, sexual behaviour, and locomotor activity, and serves as an unconditioned stimulus for inducing a conditioned place preference. This behavioural profile suggests that raphe injections of 8-OH-DPAT facilitate reward-related behaviour. The present study tested this hypothesis by investigating the effects of median raphe injections of 8-OH-DPAT on sensitivity to lateral hypothalamic (LH) self-stimulation. Frequencies required to sustain half-maximal rates of responding were determined following injection of saline or various doses of 8-OH-DPAT (0.2-5 micrograms) into the median raphe. In four rats with accurate injection sites 8-OH-DPAT dose-dependently lowered frequency thresholds by up to 40%. In the remaining rats injection sites were located outside the median raphe, and 8-OH-DPAT either slightly raised or failed to lower frequency thresholds. These results show that 8-OH-DPAT injected into the median raphe facilitates brain stimulation reward, and suggest that acute reductions in 5-HT neurotransmission may enhance sensitivity to rewarding stimuli. The possible interactions between 5-HT neurons and efferent systems utilizing dopamine and acetylcholine as neurotransmitters in mediating this effect are discussed.

Fear-potentiated startle and electrically evoked startle mediated by synapses in rostrolateral midbrain.

Startle amplitudes are increased when acoustic startle responses are elicited in the presence of a stimulus that has previously been paired with shock. This "fear-potentiated" startle response appears to be mediated via the caudal ventral amygdalofugal pathway to the brainstem. Electrical stimulation of this pathway evokes unconditioned startlelike responses. Collision tests have shown that a monosynaptic connection from amygdala to midbrain mediates these responses. Collision tests here localize these synapses mediating electrically evoked startlelike responses to the rostrolateral midbrain in awake rats. To test whether rostrolateral midbrain synapses also mediate fear-potentiated startle, we lesioned cells in these sites with ibotenic acid. These lesions completely blocked fear potentiation of acoustic startle. These same lesions did not block potentiation of startle by d-amphetamine (6 mg/kg).