xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam Ka
controlfield tag 001 001430553
007 cr mnu|||uuuuu
008 031007s2003 flu sbm s000|0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0000072
O'steen, Jennifer Robin.
Prepulse inhibition and the acoustic startle response in nine inbred mouse strains
h [electronic resource] /
by Jennifer Robin O'steen.
[Tampa, Fla.] :
University of South Florida,
Thesis (Au.D.)--University of South Florida, 2003.
Includes bibliographical references.
Text (Electronic thesis) in PDF format.
System requirements: World Wide Web browser and PDF reader.
Mode of access: World Wide Web.
Title from PDF of title page.
Document formatted into pages; contains 18 pages.
ABSTRACT: This study examined the effects of genetic background on the acoustic startle response (ASR) and its modulation by prepulse inhibition (PPI) by comparing nine inbred strains of mice. The ASR, a jerk-like motor reflex, is elicited by bursts of noise or tones with sound pressure levels of 80-90 dB and greater. PPI is a type of modulation of the ASR, requires no training, and results in observable response in both mice and humans. Data were obtained from nine inbred mouse strains, sixteen per strain, which were shipped at approximately 3-5 weeks old from The Jackson Laboratory. In general, ASRs were generally smaller when the startle stimulus was less intense. PPI was relatively weak for the 4 kHz prepulse, and stronger with prepulses of 12 kHz and 20 kHz. However, means varied widely across strains for both ASR and PPI, suggesting a strong influence of genetic background on these behaviors. In addition to genetic influences, peripheral hearing loss and central auditory processing factors must be taken into consideration.
Co-adviser: Hurley, Raymond M.
Co-adviser: Willott, James F.
Mice as laboratory animals.
acoustic startle response.
t USF Electronic Theses and Dissertations.
Prepulse Inhibition and the Acoustic Startle Response in Nine Inbred Mouse Strains By JENNIFER R. OSTEEN A Professional Research Project submitted to the Faculty of the Department of Communication Sciences and Disorders University of South Florid a in partial fulfillment of the requirements for the degree of Doctor of Audiology James F. Willott, Ph.D., Co Chair Raymond M. Hurley, Ph.D., Co Chair Theresa Hnath Chisolm, Ph.D., Committee Member March 25, 2003 Tampa, Florida Key Words: Acoustic startle response Prepulse inhibition Copyright 2003, Jennifer R. OSteen
Jennifer R. OSteen 2 Prepulse Inhibition and the Acoustic Startle Response in Nine Inbred Mouse Strains Jennifer R. OSteen (ABSTRACT) This study examined the effects of genetic background on th e acoustic startle response (ASR) and its modulation by prepulse inhibition (PPI) by comparing nine inbred strains of mice. The ASR, a jerk like motor reflex, is elicited by bursts of noise or tones with sound pressure levels of 80 90 dB and greater. PPI is a type of modulation of the ASR, requires no training, and results in observable response in both mice and humans. Data were obtained from nine inbred mouse strains, sixteen per strain, which were shipped at approximately 3 5 weeks old from The Jackso n Laboratory. In general, ASRs were generally smaller when the startle stimulus was less intense. PPI was relatively weak for the 4 kHz prepulse, and stronger with prepulses of 12 kHz and 20 kHz. However, means varied widely across strains for both ASR and PPI, suggesting a strong influence of genetic background on these behaviors. In addition to genetic influences, peripheral hearing loss and central auditory processing factors must be taken into consideration.
Jennifer R. OSteen 3 Introduction The genetics of h earing, central auditory processing, and hearing loss are poorly understood. One problem has been the difficulty in studying genetics in humans because it is difficult to separate the effects of genes from numerous possible other factors. For example, ir respective of genes that might cause hearing loss, people are typically exposed to multiple other factors that may affect hearing such as noise exposure, disease, drug therapy, and trauma. The present study used highly inbred strains of mice to study aud itory behavior. Because members of inbred strains are as genetically homozygous as monozygotic twins, differences among individuals of an inbred strain are due to non genetic variables. On the other hand, if environmental variance is minimized, reliable d ifferences between strains should largely represent differences in genetic makeup of those strains. Animal models show us what kinds of anatomic and physiologic phenomena can and do occur in mammalian auditory systems that are, in most respects, similar t o our own (Willott, 1996). Mice provide good models with which to study genetic types of hearing loss as they are genetically well defined and there is a large literature on many aspects of mouse hearing (Willott, 2001). Behavioral responses to auditory stimuli can be studied in mice using the acoustic startle response (ASR) and prepulse inhibition (PPI). Both are fundamental human/murine auditory behaviors, occurring widely in mammals and other vertebrates (Davis, 1984; Ison, 2001). While the focus of the present project is to examine ASR and PPI in mouse models, these behaviors have been well studied with human subjects (Dawson, Schell, & Bohmelt, 1999; Reiter & Ison, 1979). The ASR, a jerk like motor
Jennifer R. OSteen 4 reflex, is elicited by bursts of noise or tones wi th sound pressure levels of 80 90 dB SPL and greater in the mouse. At the onset of the noise or tone burst, mice will startle, or jump, in response to the stimulus. The reflex is easily measured using movement sensitive devices, and ASR amplitude can be obtained (Willott, 2001). The mouses movement produces a spike like voltage change in the startle measuring device. Startle amplitude is defined according to the largest peak to peak voltage deflection. Since the ASR occurs only to high intensity sound s, it cannot be used to determine thresholds of auditory sensitivity. PPI is a type of modulation of the ASR. PPI requires no training and results in observable responses in both mice and humans. A nonstartling stimulus (S1) is presented 10 100 ms be fore an intense stimulus (S2) used to elicit the acoustic startle response. S1 may be a tone burst, a gap in the background noise, or any other sensory change. S1 activates the PPI circuit that inhibits the startle pathway for a period of several hundred seconds resulting in reduced, or inhibited ASR amplitude. PPI can be expressed by a ratio of ASR amplitude with presentation of a prepulse (S1 S2) to ASR amplitude without the prepulse (S2 only). The degree to which a tone produces PPI indicates the beha vioral salience of the tone, providing a test of auditory behavior (e.g. Hoffman & Ison, 1980; Ison, 2001; Willott & Turner, 2000). Anatomical pathways mediating the ASR and PPI are well established. These pathways are among the best understood of any ve rtebrate behaviors. The primary neural pathway for the ASR resides in the lower brainstem, with auditory neurons in the cochlear nucleus projecting to startle triggering neurons of the reticular formation (Davis et al., 1982; Huffman & Henson, 1990; Kandl er & Herbert, 1991; Koch, 1999; Leitner &
Jennifer R. OSteen 5 Cohen, 1985); Lingenhohl & Friauf, 1994; Yeomans & Frankland, 1996). In PPI, the S1 is first processed by the central auditory system (minimally to the midbrain level). Neural output from the auditory system then activates other post auditory components of the PPI neural circuitry, including pathways that ultimately descend to the reticular formation and inhibit the neurons that trigger the startle reflex (Carlson & Willott, 1998; Davis, 1984; Koch, 1999; Ison, 2001; Hoffman & Ison, 1980; Li, Priebe, & Yeomans, 1998; Willott, Carlson, & Chen, 1994). The neural circuits provide numerous potential targets for gene actions that can contribute to behavioral differences among individuals. The overall goal of this pro ject was to examine the effects of genetic background on the ASR and its modulation by PPI by comparing nine inbred strains of mice. To do so, ASR amplitude was measured in response to a standard 100 dB SPL broadband noise startle stimulus, intensity fu nctions were obtained using 90, 80, and 70 dB SPL startle stimuli, and PPI was evaluated using a non startling stimulus of 70 dB SPL at frequencies of 4, 12, and 20 kHz. Test reliability was also evaluated by immediately re testing all mouse strains. Meth ods Inbred mice were obtained from The Jackson Laboratory (TJL), Bar Harbor, ME. Mice were tested from the following strains: C57BL/6J, C3H/HeJ, BALB/cJ, A/J, 129/SVlmJ, CBA/J, FVB/N3, CAST/Ei, and DBA/2J. DBA/2J mice show rapid inner ear degeneration, which becomes evident as early as two to three weeks after birth; the other strains have normal hearing at the ages used here. Mice were shipped from TJL to the University of South Florida (USF), where they were housed in a vivarium within the USF medical facility. Sixteen mice per strain were shipped approximately once a month
Jennifer R. OSteen 6 from TJL to USF for testing. Eight male and eight female mice, age ranged from 3 5 weeks old, were shipped each time. Testing was conducted when mice were approximately 4 5 weeks old, after a 48 hour acclimatization period. Mice were housed in cages with four same sex members in each cage and were given free access to food and water. Room temperature and humidity were optimally maintained under a 12 hour light/dark schedule. Thi s study complied with the National Institute of Health (NIH) guidelines for the care and use of animals and was approved by the Institutional Animal Care and Use Committees of the University of South Florida. Apparatus Disposable plastic cups (32 oz) were used to hold the mice during testing. The cup was placed in a Med Associates, Inc. (Georgia, Vermont) sound attenuated test chamber, where it rested on a movement sensitive load cell. This load cell was sensitive to any movements the mouse made, and, in t urn, produced a voltage. Acoustic stimuli were generated by a Radio Shack Supertweeter that fit snugly above the rim of the plastic cup, inside the sound attenuated test chamber. The system was initially calibrated using a 1/8 inch Bruel & Kjaer condens er microphone placed through a hole in a plastic cup, where SPL values were measured in various locations. Mean attenuation values producing the highest sound pressure levels at each frequency were determined and set. The standard startle stimulus (S2) c onsisted of a 100 dB SPL (sound pressure level re: 20 m Pa ) broad band noise (10 msec duration, msec rise/fall time based on dial setting). To generate amplitude intensity functions, S2 stimuli were also presented at 90, 80, and 70 dB SPL. Prepulse s timuli (S1) consisted of 70 dB SPL tones (3 msec rise/fall, 10 msec duration) at frequencies of 4, 12, and 20k Hz. The S1 S2
Jennifer R. OSteen 7 interval was 100 msec. Testing was conducted in a quiet research laboratory, with no continuous background masking noise. Proce dure Testing began after the mouse had been in the plastic cup for at least one minute. The first 40 trials of each test were used to establish baseline startle amplitude in response to the 100 dB SPL S2 and to measure PPI for the three S1 tones. For PPI trials, the S1 tone prepulse preceded the S2 startle stimulus. Stimuli were presented at a variable interval every 3 8 seconds and consisted of sixteen S2 only and eight S1 S2 pairings for each of the three S1 frequencies. Each ten trials contained an equ al number of each stimulus in a variable sequence. After these 40 trials, an additional 15 unmodified startle responses (S2 only) were obtained using stimuli of 90, 80, and 70 dB SPL to determined startle intensity function. Each test procedure required approximately six minutes, and included a total of 55 trials. As an indication of reliability, each mouse was immediately retested. Upon test completion, mice were returned to their original cages in the vivarium. The ASR appears as an abrupt, spike l ike voltage change, with a peak occurring 20 30 msec after the onset of the stimulus. ASR amplitude is defined by the peak voltage deflection. For each session, a mean startle amplitude was computed in response to the S2 alone and in response to S1 S2 pai rings (S1 = 4, 12, and 20 kHz). The degree of PPI for each frequency prepulse is expressed as a ratio of the amplitude of the modified startle response (S1 S2) with respect to the amplitude produced by the S2 alone. A lower percent or proportional value is indicative of greater PPI. Data Analysis
Jennifer R. OSteen 8 The data used for analysis in each session were modified by eliminating the highest and lowest ASR amplitudes in each frequency series (4, 12, 20 kHz) and the highest and lowest two values for startle only. Thi s was done due to the occasional occurrence of very large or small ASRs that might distort the mean. Preliminary analyses were performed for test retest reliability for each strain using a repeated measure ANOVA. For PPI, S1 frequency (4, 12, and 20 kHz) was a repeated measure. For ASR amplitude, S2 intensity (70, 80, 90, and 100 dB SPL) was a repeated measure. Tukey tests were used when significant ANOVAs were found. For test retest reliability of PPI and ASR, two way repeated ANOVAs were used with te st session and S1 frequency (PPI) or S2 intensity (ASR amplitude) as variables. Results Acoustic Startle Response Figure 1 presents a summary of the acoustic startle response for the nine strains, at intensities of 70, 80, 90, and 100 dB SPL. ASRs were generally smaller when the startle stimulus (S2) was less intense. The 70 dB SPL S2s produced virtually no evoked ASRs. These values were considered non ASR movement, and were not computed in the statistical analysis of specific S2 intensities. Means v aried widely across strains. For example, ASRs were extraordinarily large in A/J mice, but were small at all S2 intensities in C3H/HeJ, DBA/2J, and CAST/Ei strains. The statistical analysis showed these observations to be reliable. An ANOVA was computed for ASR at 80, 90, and 100 dB SPL. The main effect of strain (F (8,179)=20.7; p<0.0001) and the main effect of S2 intensity (F(2,358)=243.6; p<0.001) were both statistically significant.
Jennifer R. OSteen 9 Figure 1. Acoustic startle response for nine strains The interaction between strain and S2 intensity (F(16,358)=11.87; p<0.001) was also statistically significant. Due to these findings, an additional one way ANOVA was run at each intensity. Find ings were significant at 80 dB (F(8,179)=10.01, p<0.0001), 90 dB (F(8,179)=17.77, p<0.0001), and 100 dB (F(8,179)=20.64, p<0.0001). As expected, ASRs were generally smaller when the S2 stimulus was less intense. Table 1 presents Tukey test results (*= p< 0.05; **= p<0.01) for the ASR at 80, 90, and 100 dB SPL. As S2 intensity increased, more strains demonstrated significant differences. 129 A/J BALB C3H C57 CAST CBA DBA FVB 129 -** A/J -* * BALB -** ** ** ** C3H -C57 -CAST -CBA -DBA -FVB -ASR: 80 dB
Jennifer R. OSteen 10 129 A/J BALB C3H C57 CAST CBA DBA FVB 129 -** A/J -** ** ** ** BALB -** ** ** ** C3H -* ** C57 -* CAST -* CBA -* DBA -* F VB -ASR: 90 dB 129 A/J BALB C3H C57 CAST CBA DBA FVB 129 -** ** A/J -* ** ** ** ** ** BALB -** ** ** ** C3H -** ** C57 -* CAST -** CBA -DBA -** FVB -ASR: 100 dB Table 1 Tukey test results for ASR at 80, 90, and 100 dB SPL Prepulse Inhibition Figure 2 presents a summary of the PPI effect on the ASR for the nine strains. Scores represent the S1 frequency of 4, 12, and 20 kHz, respectively. In general, the effect of PPI is relatively weak for the 4 kHz S1 and stronger with S1s of 12 kHz and 20 kHz. Like the non PPI ASR data, means varied widely across strains. For example, the mice of the 129/SV1mJ strain had rather strong P PI whereas C3H/HeJ, DBA/2J, and CAST/Ei had weak PPI. These observations are supported by the statistical analysis with the main effect of strain (F (8,179)=7.54; p<0.0001) and the main effect of S1 frequency (F (2,358)=16.7; p<0.001) both statistically si gnificant. The interaction between strain and S1 frequency (F (16,358)=4.34; p<0.001) was also statistically significant. One way
Jennifer R. OSteen 11 ANOVAs were computed for each S1 frequency. The 4 kHz (F(8, 179) = 4.05, p = 0.002), 12 kHz (F(8, 179) = 6.81, p = 0.0001), and 20 kHz (F(8, 179) = 8.63, p<0.0001) were each statistically significant. Figure 2. Mean PPI scores for nine strains Table 2 presents Tukey test results (*= p<0.05; **= p<0.01) for PPI at 4, 12, and 20 kH z. 129 A/J BALB C3H C57 CAST CBA DBA FVB 129 -** ** A/J -BALB -C3H -C57 -CAST -CBA -DBA -FVB -PPI: 4kHZ
Jennifer R. OSteen 12 129 A/J BALB C3H C57 CAST CBA DBA FVB 129 -** ** A/J -BALB -C3H -* ** C57 -* CAST -CBA -* DBA -* FVB -PPI: 12kHZ 129 A/J B ALB C3H C57 CAST CBA DBA FVB 129 -* ** A/J -* BALB -C3H -** C57 -CAST -* CBA ** DBA -* FVB -PPI: 20kHZ Table 2. Tukey test results for PPI at 4, 12, and 20 kHz Reliability Mice were immediately retested after the initial run of 55 trials. No significant differences were found betwee n first and second test scores for either ASR or PPI for the nine strains. Heritability Heritability is defined as the amount of variance that can be attributed to genetic factors. A basic example of this can be seen in eye and hair color, in contrast to language spoken. Genetics contribute fully to natural hair and eye color, whereas environment dictates the native language that a person acquires. Using this example, eye and hair
Jennifer R. OSteen 13 color would have a heritability value of 1.0, while language would have a value of 0.00. The closer the number is to 1.0, the more genetic factors contribute to the behavior. If we assume that the within strain variance is primarily environmental and the between strain variance is primarily genetic, then we can roughly estima te the heritability as between strain variance divided by (within strain variance + between strain variance). For 12 kHz PPI, standard deviation among strain means (N=9) = 0.13. Heritability values for each strain are as follows, with a range of 0.30 in CBA/J to 0.46 in C57. Strain 129 (0.31), A/J (0.37), BALB/c (0.42), C3H (0.39), C57 (0.46), CAST (0.32), CBA/J (0.30), DBA/2J (0.38), and FBV (0.34). For the 100 dB ASR, mean values were 0.55, ranging from 0.41 in A/J to 0.80 in C3H. Strain 129 (0.45), A/J (0.41), BALB/c (0.42), C3H (0.80), C57 (0.61), CAST (0.64), CBA/J (0.46), DBA/2J (0.61), and FVB (0.51). Discussion A goal of the present study was to examine the effects of genetic background on the ASR and its modulation by PPI. This was accomplish ed by comparing nine inbred strains of mice. Data were collected and analyzed using ASR amplitude in response to a standard 100 dB SPL broadband noise startle stimulus at three discrete intensities. In addition, the PPI effect was evaluated using a non startling stimulus of 70 dB at frequencies of 4, 12, and 20 kHz. Data obtained from this study show clearly that inbred strains of mice vary greatly on both the ASR and PPI. Relatively high heritability indicates a strong genetic component to the betwe en strain differences in these behaviors. This conclusion is in agreement with previous studies of ASR in rats (Glowa & Hansen, 1994) and ASR and
Jennifer R. OSteen 14 PPI in inbred strains of mice (Bullock, Slobe, Vazquez, & Collins, 1997; Paylor & Crawley, 1997). The present data are particularly impressive because the earlier studies did not include females in testing, implemented a continuous 70 dB SPL background noise during testing, used longer inter stimulus intervals, and used a broadband noise for the S1s, rather than tones. The latter variables could have contributed to strain differences irrespective of genetic differences. In addition to genetic influences that may affect S1 or S2 responses, peripheral hearing loss of inbred mouse strains must be taken into consider ation. Mice of some strains exhibit genetically determined hearing loss. In hearing impaired mice, the ability of the S1 and/or S2 could be compromised. This would result in weak ASRs or ASR PPI amplitudes. Relative differences among animals in the str ength of PPI appear to be due, in part, to changes in peripheral auditory sensitivity (Willott & Turner, 2000). This likely contributes to the poor performance in DBA/2J mice, which exhibit early hearing loss (Erway et al., 2001). However, the other strai ns hear normally when young, so PPI and ASR differences may be due to central auditory processing. Prepulse inhibition is generally viewed as a measure of central auditory processing because the inferior colliculus and other higher order structures compri se the pathway(s) by which the prepulse modulates the startle response (Willott & Turner, 1999). Thus, the data suggest that central auditory processing is strongly influenced by genetic background. This finding may have clinical implications: even with a normal peripheral auditory system, central auditory processes can contribute to performance on auditory tests.
Jennifer R. OSteen 15 The use of inbred strains of mice holds great promise for understanding the genetic basis of the ASR and PPI. Future research may use some of t he mouse models shown here to elucidate genetic influences on ASR and ASR PPI.
Jennifer R. OSteen 16 References Bullock, A.E., Slobe, B.S., Vazquez, V, and Collins, A.C. (1997). Inbred mouse strains differ in the regulation of startle and prepulse inhibi tion of the startle response. Behavioral Neuroscience, 111, 1353 1360. Carlson, S., & Willott, J. F. (1998). Caudal pontine reticular formation of C57BL/6J: Responses to startle stimuli, inhibition by tones, plasticity. Journal Neurophysiology, 79, 260 3 2614. Davis, M. (1984). The mammalian startle response. In R.C. Eaton (Ed.), Neural Mechanisms of Startle Behavior (287 351). Plenum Publishing: New York. Davis, M., Gendelman, D. S., Tischler, M.D., & Gendelman, P. M. (1982). A primary acoustic ci rcuit: Lesion and stimulation studies. Journal of Neuroscience, 2, (6), 791 805. Dawson, M.E., Schell, A.M., & Bohmet, A.H. (1999). Startle Modification. Cambridge Press: New York. Erway, L.C., Zheng, Q.Y., & Johnson, K.R. (2001). Inbred strains of mic e for genetics of hearing in mammals: Searching for genes for hearing loss. In J.F. Willott (Ed.), Handbook of Mouse Auditory Research: From Behavior to Molecular Biology (pp. 429 440). Boca Raton: CRC Press Glowa, J.R., & Hansen, C. T. (1994). Differen ces in response to an acoustic startle stimulus among forty six rat strains. Behavioral Genetics, 24(1), 79 84. Hoffman, H.S., & Ison, J.R. (1980). Reflex modification in the domain of startle. I. Some empirical findings and their implications for how the nervous system processes sensory input. Psychology Bulletin, 87, 175 189.
Jennifer R. OSteen 17 Huffman, R. F., & Henson, O. W. (1990). The descending auditory pathway and acousticomotor systems: connections with the inferior colliculus. Brain Research Reviews, 15, 295 323. Ison, J.R. (2001). The acoustic startle response: Reflex elicitation and reflex modification by preliminary stimuli. In Willott JF (Ed), Handbook of Mouse Auditory Research: From Behavior to Molecular Biology Boca Raton: CRC Press, pp 59 82. Kan dler, K., & Herbert, H. (1991). Auditory projections from the cochlear nucleus to pontine and mesencephalic reticular nuclei in the rat. Brain Research, 562, 230 242. Koch, M. (1999). The neurobiology of startle. Progress in Neurobiology, 59, 107 128. Leitner, D.S., & Cohen, M.E. (1985). Role of the inferior colliculus in the inhibition of acoustic startle in the rat. Physiology & Behavior, 34, 65 70. Li, L., Priebe, R. P., & Yeomans, J.S. (1998). Prepulse inhibition of acoustic or trigeminal startl e of rats by unilateral electrical stimulation of the inferior colliculus. Behavioral Neuroscience, 112, 1187 1198. Lingenhohl, K., & Friauf, E. (1994). Giant neurons in the rat reticular formation: A sensorimotor interface in the elementary acoustic st artle circuit? Journal of Neuroscience, 14, 1176 1194. Marks, M. J., Stitzel, J. A., & Collins, A.C. (1989). Genetic influences on nicotine respones. Pharmacology, Biochimistry and Behavior, 33 667 678. Paylor, R., & Crawley, J. N. (1997). Inbred st rain differences in prepulse inhibition of the mouse startle response. Psychopharmacology, 132, 169 180.
Jennifer R. OSteen 18 Reiter, L. A., & Ison, J. R. (1979). Reflex modulation and loudness recruitment. Journal of Auditory Research, 19, 201 207. Willott, J.F. (1996). A natomic and physiologic aging: A behavioral neuroscience perspective. Journal of the American Academy of Audiology, 7, 141 151. Willott, J.F. (2001). Handbook of Mouse Auditory Research: From Behavior to Molecular Biology Boca Raton: CRC Press. Willo tt, J.F., Carlson, S., and Chen, H. (1994). Prepulse inhibition of the acoustic startle response in mice: Relationship to hearing loss and auditory system plasticity. Behavioral Neuroscience, 108, 703 713. Willott, J.F., & Turner, J. G. (1999). Prolong ed exposure to an augmented acoustic environment ameliorates age related auditory changes in C57BL/6J and DBA/2J mice. Hearing Research, 135, 77 88. Willott, J. F., & Turner, J. G. (2000). Neural plasticity in the mouse inferior colliculus: Relationship t o hearing loss, augmented acoustic stimulation, and prepulse inhibition. Hearing Research, 147 275 281. Yeomans, J. S. & Frankland, P. W. (1996). The acoustic startle reflex: neurons and connections. Brain Research Reviews, 21, 301 314.