Men are more inhibited than women by weak prepulses

BIOL PSYCHIATRY 1993;34:253.-260

253

Men Are More Inhibited than Women by Weak Prepulses Neal R. Swerdlow, Pamela Auerbach, Sean M. Monroe, Heidi Hartston, Mark A. Geyer, and David L. Braff

The acoustic startle reflex is normally inhibited when the startling stimulus is preceded by a weak prepulse. We studied "prepulse inhibition '~ (PPI) to assess potential gender differences in this operational measure of sensorimotor gating. A review of data from our previously published studies in psychiatric patientsand normal controls indicated that startle in women was less inhibited by weak prepulses than was startle in men, and that this gender difference narrowed when stronger prepulses were used to elicit maximal levels of PP1. Based on these observations, new subjects were selected for study using our established criteria to exclude individuals with psychiatric disorders, substance abuse, or serious neurologic or medical illness. Replicating our preliminary observation, women exhibited significantly less PP! than men, with the gender difference being most notable under conditions with weak prepulses, Potentially confounding variables, including electrode impedance, startle amplitude, habituation and latency did not differ between men and women. PP! was then measured in male and female rats, where no gender differences were noted. These findings identify significant gender differences in PP1 in humans, and suggest that inhibitory "gating" processes are more robust in men than in women, when assessed using a sensitive range of inhibitory stimuli.

Key Words: Gating, gender, sensorimotor, startle

Introduction Prepulse inhibition (PPI) is the normal reduction of a startle reflex that occurs when the startling stimulus is preceded 30-500 msec earlier by a weak prepulse (Ison et al 1973; Graham 1975). This partially automatic (DelPezzo and Hoffman 1980) involuntary inhibitory process provides an operational measure of sensorimotor gating (Braff et al From the Department of Psychiatry, University of California, San Diego, LetJo!la, CA. Address reprint requests to Neal R. Swerdlow, MD, PhD, UCSD Department of Psychiatry, La Jolla, CA 92093-0804. Received January 4, 1993; revised May 6, 1993. © 1993 Society of Biological P~ychiatry

1978). Importantly, PPI is impaired in patients with three specific neuropsychiatric disorders that are characterized clinically by impaired ability to gate or inhibit extraneous or nonsalient cognitive, motor or sensory information, including schizophrenia (Braff et al 1978, 1992), obsessive compulsive disorder (Swerdlow et al 1993) and Huntington's disease (Swerdlow et al 1991). Thus, the central inhibitory processes responsible for PPl may be quite relevant to defective cognitive and sensorimotor gating in certain forms of mental illness. In a recent study, we reported our preliminary observation that PPl is impaired in patients with obsessive compulsive disorder (OCD) (Swerdlow et al 1993). Control 0006.3223/93/$06.00

254

BIOLPSYCHIATRY

N.R. Swerdlow et al

1993:34:253-260

and OCD groups were balanced for several variables, including gender. Still, an analysis of data from this preliminary study subsequent to its publication revealed a significant effect of gender, and no significant gender × diagnosis interaction. Inspection of the data revealed that, independent of diagnosis, female subjects exhibited significantly less PPl overall than did male subjects (Figure 1A). Similar gender differences have been reported in 8year-old children, but not in 5-year olds or adults, in a paradigm that employed prepulses that were approximately 45 dB above background (Ornitz et al 1991). Using tactile prepulses and acoustic startle stimuli, Blumenthal and Gescheider (1987) reported significantly lower response probability in males than in females across all prepulse trials, which might suggest greater prepulse inhibition in males. Following our unexpected observation, we examined data from a larger set of control subjects who had been rigoroL:slyscreened to rule out psychopathology. These subjects had been tested using only very salient prestimuli [15 dB(A) above a 65 dB(A) background]. Under these conditions, there was a nonsignificant trend toward reduced PPI in female subjects compared to males (Figure IB). Based on these suggestive post-hoc analyses, we designed a separate study to assess effects of gender on PP! in both humans and laboratory rats. Consistent with our previous observations in humans, female subjects exhibited significantly less PPl than did males; no such gender difference was noted in laboratory rats.

Methods Human Studies Normal men (n = 18) and women (n ffi 19) participated in this study [mean ages: men - 31.2 ± 8.4 (SD); women -- 29.1 -+ 10.5 (SD)]. Subjects were screened and excluded for a history of mental illness, substance abuse, recreational drug use, schizophrenia in a first-degree relative, sustained loss of consciousness, severe neurologic or medical illness, electroconvulsive therapy, pregnancy, or psychotropic medication use. All subjects received a urine toxicology screen, and were excluded on the basis of any positive illicit drugs. The acoustic startle response was measured between 9 AM and 5 PM using electmmyographic (EMG) systems (San Diego Instruments). Hearing was screened using a Saico Audiometer (lnterr,coustics, Assens, Denmark); no subjects met the exclusion criterion of impairment at 40 dB(A) (1000 Hz). The eyeblink component of startle was assessed using EMG measurements of orbicularis oculi, as described elsewhere (Braff et al 1992). Startle stimuli were delivered by Telephonics (TDH-39-P, Maico, Minneapolis, Minn) headphones. Subjects sat upright, and were directed to look straight ahead and to stay awake. A back-

ground 70 dB(A) white noise continued throughout the session, and was followed 5 min after its onset by 72 startle trials, consisting of three conditions: (1) a 118 dB(A) 40 msec noise burst presented alone (P-ALONE); (2) the same 118 dB(A) 40 msec noise burst preceded at a 100 msec interval by prepulses (20 msec noise bursts) that were 2, 4, 8, or 16 dB(A) above background (PP2, PP4, PP8 or PPI6); or (3) no stimulus (NOSTIM) condition. These six trial types were repeated six times in pseudorandom order; this block of 36 trials was repeated twice for a total of 72 trials. A variable intertrial interval averaged 30 sec. PPI was defined as the percentage of reduction in startle amplitude in the presence of the prepulse compared to the amplitude in the absence of the prepulse [ 100 minus (100 × amplitude on prepulse trial/amplitude on P-ALONE trial)]. Thus, a large "percentage score" indicates a high degree of PPI, while a smaller "percentage score" indicates less PPl. Consistent with previous studies, subjects with negligible startle responses (mean startle amplitude < 10) were identified as "nonresponders," and their data was not analyzed further (n = l).One additional subject was excluded from analysis for scores that exceeded the sample mean by > 5.5 SD.

Rat Studies The eight male and eight female Sprague-Dawley rats (250300 g; Harlan, San Diego, CA) were housed in genderdivided rooms in single-sex pairs, handled prior to startle testing and maintained on a 12 hr: 12 hr light/dark schedule with food and water provided ad libitum. Testing occurred at the beginning of the dark cycle. Each startle chamber (SR-LAB, San Diego Instruments) consisted of a Plexiglas cylinder 8,2 cm in diameter resting on a 12.5 × 25.5 cm Plexiglas frame in a ventilated enclosure, Noise bursts were presented via a speaker mounted 24 cm above the rat. A piezoelectric sensor (assembled using a Blatek Audio Transducer Model 6030, Blatek Inc,, State College PA) mounted below the Plexiglas frame detected and transduced motion within the cylinder. Stabilimeter readings are rectified and recorded by a computer and interface assembly, with 100 l-msec readings collected starting at stimulus onset. Startle amplitude is defined as the average of these 100 readings. Tests began 7 days after shipment arrival. Rats were placed in a startle chamber for 5 min with a 70 dB[A] background noise, followed by a startle session that was identical to that used in humans. Statistical analyses for both human and rat studies consisted of mixed-design repeated measure analysis of variance (ANOVAs) with Student's t-tests or ANOVAs as post-hoc analyses. Statistical analyses were completed using BMDP (BMDP Statistical Software, Inc., Los An-

Gender Differences in Prepulse In~bition

~IOL PSYCHIATRY 1993;34:2';3-260

A. i--ZcoNTROLS(n=13)

70 -~ r'IMALE(n=I2)

Z (: 60 I-~" 40 m

50

m

wZ

3O

m

.J =D n_ tu

I0

0

eL Ill -2O

-lO 2

4

8

16

2

PREPULSE I N T E N S I T Y ( l I B ( A ) e b o v e

4

B

background)

16

.

80 z 0 i

am

ea

60

m

~e =e m

t~J U) eL_ uJ ne

40

20

O. it 30

60

120

PREPULSE I N T E R V A L ( m s e c )

Figure I. Prepulse inhibition of acoustic startle in male and female subjects. A. Data from our previously published report of impaired PPl in patients with Obsessive Compulsive Disorder (n = 24; M:F = 12:12) (Swerdlow et al 1993), modified to demonstrate gender differences. The startle session was identical to that described in the present studies. Statistical analysis revealed a significant effect of gender (F -- 7,23, df !,20, p < 0.015) (#) and a significant interaction of diagnosis x prepulse intensity (F = 3.73, df 3,60, < 0.016), but no significant interaction of gender x diagnosis (F < !) or gender x diagnosis x prepulse intensity (F < 1). Post-hoc analysis revealed that PPl was significantly (p < 0.05) reduced in OCD subjects compared to controls for 4rib trials, as previously reported (Swerdlow et al 1993 ( % B. Data from control subjects (n -- 36; M:F -- 19:i7) tested in a paradigm described in a previous report from our group (Braff et ai 1992). Some of these subjects (n -- 25) served as a comparison group in studies of PPi in psychiatric patients (Braff et al 1992), but data had not previously been examined according to gender. Prepulse intensity was 15 dB above background. Statistical analysis revealed a nonsignificant trend toward less PPi in female subjects compared to males (F -- 3.03, dr, 1,34, -- 0.09). Adding gender as a factor in a re-analysis of published data (Braff et al 1992) comparing schizophrenic patients (n = 22) to a subset of these control subjects (n = 25) revealed a significant effect of diagnosis (F -- 6.31, df 1,43, p < 0.02), no significant effect of gender (F = 1.47, df 1,43, NS), and no significant diagnosis x gender interaction (F < 1) (data not shown). Thus, the previously reported significant reduction of PPl in schizophrenic patients cannot be explained I:.y gender-specific patterns of PP1.

255

BIOLPSYCHIATRY

256

N.R. Swerdlow et al

1993:34:253-260

150 "

60

s peak female

s

z

SESSION BLOCK

'~).

i uJ ffl _1

g~ 50 w peak male .a ..d 4o, onset male onset female

["1 MALE ( n - I e )

6o~ •r(MAL((n-~g) 30 30"

16;I0 TRIAL TYPE

2fl

Figure 3, Onset and peak startle latency in normal control subjects, Onset and peak startle latencies were reduced significantly on prepulse trials compared to P-ALONE trials ("latency facilitation"). These effects did not differ significantly between males and females.

el_ LU me 0 n. N "10 "20

pro'.

2

4 O 15 PREPULSE INTENSITY (dO(A) obove bockground)

Figure 2. PP! and startle amplitude (approximately 7.7 pN/digital unit) in norn~al control males (n = 18) and females (n = 19). PPI was significantly reduced in females compared to males. Startle amplitude did not significantly differ between males and females, but there was a significant interaction of gender and block (*).

geles, CA) and SuperANOVA (Abacus Concepts, Inc., Berkeley, CA) statistical packages. P-ALONE amplitude and PPI were analyzed using 2-way ANOVAs with repeated measures on gender (between-subject) and ~al block (within-subject) for both P-ALONE and PPl analyses, and with repeated measures on trial type (within-subject) for PPI analyses. Alpha was set at 0.05. Results

Findings from measures of startle and PPI in humans are seen in Figure 2. Analysis of PPI revealed a significant effect of gender (F = 4.38, df 1,35, p < 0.05), a significant effect of prepulse intensity (F = 50.71, df 3,35, p < 0.0001) and no significant effect of block (F = 1.48, df 1,35, NS). There were no other significant two- or threeway interactions. Since inspection of the data reveals that neither males nor females exhibited prepulse inhibition for 2 dB(A) prepulse conditions, it could be argued that data from 2 dB(A) trials is not relevant to measures of PPI. Separate analysis of 4, 8, and 16 dB(A) trials, howe;,er, yields similar results (significant effect of gender: F 6.01, df 1,35, p < 0.02; significant effect of prepulse

intensity: F = 64.07, df 2,70, p < 0.0001; no significant effect of block: F = 1.75, dr, !,35, NS; no significant two- or three-way interactions). Comparison of P-ALONE amplitude revealed no significant effect of gender (F = 1.45, df i,35, NS), a significant effect of block (F = 50.39, df 1,35, p < 0.0001), and a significant gender x block interaction (F = 4.65, df 1,35, p < 0.05). This significant interaction reflected a non-significant trend towards lower startle amplitude in females in the first block (F = 2.23, df 1,35, NS). Analysis of onset startle latency (time from stimulus onset to reflex onset) revealed normal onset latency facilitation in both male and female subjects, as indicated by a significant reduction in latency on prepulse trials compared to P-ALONE trials (Figure 3). There was no significant effect of gender (F = 0.00), a significant effect of trial type (F = 2.94, df 4,140, p < 0.025), and no significant gender x trial type interaction (F < 1). Analysis of peak startle latency (time from stimulus onset to peak reflex amplitude) also revealed significant latency facilitation in both males and females. ANOVA revealed that there was no significant effect of gender (F < 1), a significant effect of trial type (F = 20.56, df 4,140, p < 0.0001), and no significant gender x trial type interaction (F = 2.19, df 4,140, NS). Thus, consistent with our preliminary and post-hoe observations, PPI was significantly lower in feai,des compared to males; no gender differences were noted in reflex amplitude, latency facilitation, or onset or peak latency. Findings from measures of startle and PPI in rats are

Gender Differences in Prepulse Inhibition

BIOL PSYCHIATRY

1993;34:253-26o

257

6OO

tU

g

500

-~ 400 300

Im-

100

f 100

A

.

.

2 .

.

.

.

.

.

.

.

.

80

E m _z

6o

lu

[] []

,,,,i Itl E ¢I.

4O

male female

20

2

4

8

16

PREPULSE INTENSITY ( d B ( A ) above b a c k g r o u n d )

Figure 4. PPl and startle amplitude in males (n -- 8) and female (n -- 8) rats. PPi and startle amplitude did not differ significantly between males and females. Although startle habituation appeared to be tess robust in females than males, this difference was not statistically significant. seen in Figure 4. Analysis of PPl revealed no significant effect of gender (F < I), a significant effect of prepulse intensity (F = 62.47, df 3,42, p < 0.0001), and no significant gender × intensity interaction (F -- 1.71, df 3,42, NS). Although inspection of the data suggests that female rats exhibited a tendency toward more PPI on 4dB trials and less overall amplitude habituation than did male rats, these trends did not reach statistical significance. Comparison of P-ALONE amplitude revealed no significant effect of gender (F < 1), a significant effect of block (F = 15.02, df 1,14, p < 0.002), and no significant gender × block interaction (F = 1.79, df 1,14, NS). It may be important that the startle session used in humans and rats was not designed to be optimally sensitive

to changes in reflex habituation. Gender differences in startle habituation might be better studied using longer sessions with more P-ALONE trials and longer intertrial intervals. Startle stimuli in the present session did, however, elicit a full range of prepulse inhibition in both humails and rats. Unlike our findings in humans, PPl in rats does not differ significantly between males and females.

Discussion Behavior is significantly influenced by an organism's ability to suppress or gate sensory, cognitive and motor information. Impaired sensorimotor gating has been found

258

BIOLDSYCHIATRY 1993,34:253-260

in patients with three specific neuropsychiatric disorders-schizophrenia (Braff et al 1978, 1992), OCD (Swerdlow et al 1993) and Huntington's disease (Swerdlow et al 1991)--that are characterized phenomenologically by a reduced ability to gate intrusive sensory, cognitive, or motor information. These three specific disorders of impaired cognitive and sensorimotor gating are linked in an anatomical dimension by dysfunction within different--but overlapping---elements of neural circuitry connecting the limbic system and basal ganglia (cf Swerdlow et al 1992). Deficient PPI in children with primary nocturnal enuresis has been attributed to a failure of early sensory processing resulting from dysfunction in the mesopontine reticular formation (Ornitz et al 1992), In nonpathological states, however, an organisms's operational gating' capacity might contribute to, and help define, a structure or style of organizing information for problem solving and other cognitive tasks. Our findings suggest that an absolute amount of PPI is not predictive of psychopathology. Thus, although PPI is clearly impaired in OCD patients compared to controls of the same gender (Swe.rdlnw et a! 1993), female controls and male OCD patients exhibit comparable amounts of PPI. In the present study, female controls clearly demonstrate less PPl than do male control subjects, despite the absence of psychopathology in both groups. These findings underscore the need to carefully match control and patient groups for gender, to ensure that any observed differences in PPl do not reflect gender effects (Josiassen and Johnson 1988). Importantly, deficits in PPl detected in patients with schizophrenia (Braff et al 1992) and OCD (Swerdlow et al 1993) do not simply reflect gender effects (see legend to Figure I). It is possible that the observed gender differences in PPl might reflect differences in sensorimotor gating that contribute to gender differences in patterns of problem solving or other cognitive tasks in normal individuals. Thus, a substantial body of literature documents consistent gender differences in mathematical, spatial, and verbal abilities (of, Maccoby and Jacklin 1974; Petersen and Winig 1979). The relationship of gender differences in sensorimotor gating to differences in these measures of cognitive function might be direct, reflecting overlapping differences in the neural control of cognition and sensorimotor gating. Alternatively, this relationship might be indirect, a result of the effects of different levels or rates of development of s~:nsorimotor gating in childhood on the unfolding of cognitive schema and problem solving strategies in boys and girls. Omitz et al (1991) report that PPl develops during the Piagetian period of concrete operations, and speculate that inhibitory processes reflected by PPI might contribute to "the capacity of the child to separate himself affectively and conceptually from

N.R. Swerdlow et al

external objects and events" (p. 19). Conceivably, reduced sensorimotor gating in females might lead to a more receptive and less focused cognitive style, where attention is allocated to greater amounts of diverse forms of stimuli, promoting more analytical and informationinclusive strategies for problem solving (Cooper 1976, 1982; Kail et al 1979). Such a capacity of women to engage and process information that would be "gated" by males might have important implications for the survival of offspring and the species. It would be more speculative to suggest that gender differences in PPl are relevant to geMer-related patterns noted in the clinical characteristics of disorders of impaired cognitive, sensory, or motor gating. Thus, rates and severity of schizophrenia are reportedly lower among females than males (Nicole et al 1992). Structural brain abnormalities are more evident among male schizophren. ics than females (Andreasen et al 1986); and the patterns of neuropsychologicai abnormalities in schizophrenia differ between males and females (Pedick et al 1992). Comparisons of gender effects in OCD patients reveal that the course of OCD differs between men and women, with earlier onset of symptoms noted in men (Rasmussen and Eisen 1990). Gender effects are also noted in Huntington's disease, with earlier onset and more severe forms of the illness associated with inheritance from males (Bruyn 1968). At present, it is not possible to identify a mechanism that might link intrinsic gender differences in the neural control of sensorimotor gating to differences in the clinical presentation of these disorders of impaired cognitive, sensory, or motor gating. Our previous work suggests that PPI of the startle reflex is modulated by neural circuitry connecting portions of the limbic system and basal ganglia (of, Swerdlow et al 1992). There are reported gender differences in the structure of several iimbic system nuclei, including the amygdala, the bed nucleus of the stria tetminalis (Hines et al 1992) and the medial preoptic area (Allen et al 1989). More generalized gender differences have been reposed in interhemispheric connectivity in the corpus callosum (Allen et al 1991), anterior commissure and massa intermedia (Allen and Gorski 1991). These structural gender differences in limbic nuclei and interhemispheric connections might produce different patterns of activity within limbic-motor circuitry to produce the observed gender-specific patterns of reflex inhibition. Neuroendocrine regulation might also differentially alter neurotransmission in brain systems that are critical substrates of startle gating: for example, estrogen effects on dopamine receptor sensitivity in women might modify the potent dopaminergic modulation of PPl (Hruska 1986). Although the lack of gender difference in PPl in rats suggests that animal models might not be readily used to

Gender Differences in Prepulse Inhibition

study the physiological basis for this gender difference in humans, one alternative strategy in progress is to assess the relationship between sensorimotor gating in women and hormonal levels throughout the menstrual cycle. In summary, PPI of the acoustic startle reflex is reduced in women compared to men. Gender differences were not noted in the amplitude, habituation, or latency of the startle reflex; nor were gender differences in PPI noted in rats. It is possible that intrinsic differences in sensorimotor gating characteristics between normal men and women contribute to differences in cognitive and problem-solving styles. We might also speculate that less

BIOLPSYCHIATRY 1993;34:253-260

259

sensorimotor gating in women compared to men might contribute to gender differences in the clinical presentation of several disorders characterized by impaired sensorimotor gating. This work was supported by NIMH Grants R29 MH48381 (NRS) and MH-42228 (DLB, MAC;, NRS), and RSDA Grant MH-O0188 (MAG). The authors gratefully acknowledge the assistance of Mr. Navid Taaid, Ms. Debra Bortz, Mr. Sam Zinner, and Ms. Patricia Hartman. Some methods in this study are identical to those previously reported by the authors (Swerdlow et al 1993); the description of this methodology may duplicate these previous reports.

References Allen LS, Gorski RA (1991): Sexual dimorphism of the anterior commissure and massa intermedia of the human brain. J Comp Neuroi 312:97-104. Allen LS, Hines M, Shryne JE, Gorski RA ( 1989)', Two sexually dimorphie cell groups in the human brain. J Neurosci 9:497506. Allen LA, Richey MF, Chai YM, Gorski RA (1991): Sex differences in the corpus callosum of the living human being. J Neurosci I ! :933-42. Andreasen NC, Nasrallah HA, Dunn V, et al (1986): Structural abnormalities in the frontal system in schizophrenia: A magnetic resonance imaging study. Arch Gen Psychiatry 43:136144. Blumenthal TD, Gescheider GA (1987): Modification of the acoustic startle reflex by a tactile prepulse: The effects of stimulus onset asynchrony and prepulse intensity. Psycho. physiology 24:320-327. Braff D, Stone C, Callaway E, Geyer M, Glick I, Bali L (1978): Prestimulus effects on human startle reflex in normals and schizophrenics. Psychaphysiology 15:339-343. Braff DL, Grillon C, Geyer MA (1992): Gating and habituation of the startle reflex in schizophrenic patients. Arch Gen Psychiatry 49:206--215. Bmyn GW (1986): Huntington's chorea: Historical, clinical and laboratory synopsis. In Vinken P, Bruyn GW reds), Handbook of Clinical Neurology vol. 6. Amsterdam: North Holland, pp 298-378. Butler RW, Jenkins MA, Braff DL. On the abnormality of normal comparison groups: The identification of psychosis-proneness and substance abuse in putative normal research subjects. Am J Psychiatry (in press). Cooper LA (1976): Individual differences in visual comparison processes. Percept Psychophysics 19:433-444. Cooper L (1982): Strategies for visual comparison and representation: Individual differences. In RJ Steinberg (ed), Advances in the Psychology of Human Intelligence. Hillsdale, NJ: Erlbaum, pp 77-174. DelPezzo EM, Hoffman HS (1980): Attentional factors in the inhibition of a reflex by a visual stimulus. Science 210:673674. Davis M (1984): The mammalian startle response. In Eaton RC (ed), Neural Mechanisms of Startle Behavior. New York: Plenum Press, pp 287-342.

Graham F (1975): The more or less startling effects of weak prestimulation. Psychophysiology 12:238-248. Hines M, Allen LS, Gorski RA (1992): Sex differences in subregions of the medial nucleus of the amygdala and the bed nucleus of the stria terminalis of the rat. Brain Res 579:321326. Hruska RE (1986): Elevation of striatal dopamine receptors by estrogen: Dose and time studies. Y Neurochem 47:1908-1915. lson JR, McAdam DW, Hammond GR (1973): Latency and amplitude changes in the acoustic startle reflex produced by variations in auditory prestimulation. PhysiolBehav 10:10351039. Josiassen RC, Johnson MM (1988): Age, gender and ERPs in schizophrenia: Let's at least handle the easy ones. Schizophr Res i :365-367. Kail R, Carter P, Pellegrino J (1979): The locus in sex differences in spatial ability. Percept Psychophysics 26:182-186. Maccoby EE, Jacklin CN (1974): The Psychology of Sex Differeme.~..~tanford: Stanford Unive~ity .Pr.es~ McGhie A, Chapman J (1961): Disorders of attention and per. ception in early schizophrenia. Br J Med Psyehol 34:102116. Nicole L, Lesage A, Lalonde P (1992): Lower incidence and increased male:female ratio in schizophrenia. Br J Psychiatry 161:556--557. 'Dmitz EM, Guthrie D, Sadeghpour M, Sugiyama T (1991): Maturation of prestimulation-induced startle modulation in girls. Psychophysiology 28: I 1-20. Ornitz EM, Hanna GL, de Traversay J (1992): Prestimulationinduced startle modulation in attention-deficit hyperactivity disorder and nocturnal enuresis. Psyehaphysiology 29:437451. Perlick D, Stastny P, Mattis S, Teresi J (1992): Contribution of family, cognitive and cliqiea! dime,si~ng to long-term outcome in schizophrenia. Schizophr Res 6:257-265. Petersen AC, Wittig MA (1979): Sex-related differences in cognitive functioning: An overview. In Wittig MA, Petersen AC (eds), Sex-RelatedDifferences in Cognitive Functwning. New York: Academic Press. Rasmussen SA, Eisen JL (1990): Epidemiology of obsessive compulsive disorder. J Clin Psychiatry 51 (suppl 2):10-13. Swerdlow NR, Caine SB, Geyer MA, Swenson MR, Downs NS,

260

BIOL PSYCHIATRY

N.R. Swerdlow et al

1993~34:253.--260

Braff DL (1991): Sensorimotor gating deficits in Schizophrenia and Huntington's Disease and in rats after manipulation of hippocampus-accumbens circuitry. (Abstr) lnt Cong Schiz Rex. Swerdlow NR, Caine SB, Braff DL, Geyer MA 0992): Neural substrates of sensorimotor gating of the startle reflex: A review

of recent findings and their implications. J Psychopharmacol 6:176-190. Swerdlow NR, Benbow CH, Zisook S, Geyer MA, Braff DL 0993): A preliminary assessment of sensorimotor gating in patients with Obsessive Compulsive Disorder, Biol Psychiatry 33:298-301.