United States       Prevention, Pesticides     EPA712-C-98-242
          Environmental Protection    and Toxic Substances     August 1998
          Agency         (7101)
&EPA    Health Effects Test
          Guidelines
          OPPTS 870.6855
          Neurophysiology:
          Sensory Evoked
          Potentials

-------
                           INTRODUCTION
     This guideline is one  of a  series  of test  guidelines that have been
developed by the Office of Prevention, Pesticides and Toxic Substances,
United States Environmental  Protection Agency for use  in the testing of
pesticides and toxic substances, and the  development of test data that must
be submitted to the Agency  for review under Federal regulations.

     The Office of Prevention, Pesticides and Toxic Substances (OPPTS)
has  developed this guideline through  a process of harmonization that
blended the testing  guidance  and requirements that  existed in the Office
of Pollution Prevention and  Toxics  (OPPT) and appeared in Title  40,
Chapter I,  Subchapter R of the Code of Federal Regulations  (CFR),  the
Office of Pesticide Programs (OPP) which appeared in publications of the
National Technical  Information Service (NTIS) and the guidelines pub-
lished by the Organization  for Economic Cooperation and Development
(OECD).

     The purpose of harmonizing these  guidelines  into a single set of
OPPTS guidelines is to minimize  variations among the testing procedures
that must be performed to meet the data  requirements of the U. S. Environ-
mental Protection Agency  under  the Toxic  Substances  Control Act  (15
U.S.C. 2601) and the Federal Insecticide, Fungicide and Rodenticide Act
(7U.S.C. I36,etseq.).

     Final  Guideline Release: This guideline  is available from the U.S.
Government Printing Office,  Washington, DC 20402 on disks or paper
copies: call (202) 512-0132. This  guideline is also available electronically
in PDF (portable document format) from EPA's  World Wide Web  site
(http://www.epa.gov/epahome/research.htm) under the heading "Research-
ers and Scientists/Test Methods and Guidelines/OPPTS  Harmonized Test
Guidelines."

-------
OPPTS 870.6855   Neurophysiology: sensory evoked potentials.
     (a) Scope—(1) Applicability. This guideline is intended to meet test-
ing  requirements   of both  the  Federal  Insecticide,  Fungicide,   and
Rodenticide Act (FIFRA) (7 U.S.C. 136, et seq.} and the Toxic Substances
Control Act (TSCA) (15 U.S.C. 2601).

     (2) Background. This guideline was developed jointly between the
Office of Pesticide Programs and  the Office of Pollution Prevention and
Toxics in cooperation with the Office  of Research and Development of
the Environmental Protection Agency. The source material used in devel-
oping this harmonized guideline is 40  CFR 798.6855 Neurophysiology:
sensory evoked potentials.

     (b) Purpose.  The techniques  in this guideline  are designed to detect
and characterize changes in the  sensory aspects of nervous system function
that result from exposure to chemical substances. The techniques involve
neurophysiological measurements from  adult animals  and are sensitive to
changes in the  function of auditory, somatosensory (body sensation) and
visual sensory systems. These procedures can be used in two ways:

     (1) To  detect  sensory dysfunction produced by compounds in the ab-
sence of relevant information.

     (2) When there are reasons to expect  that particular sensory functions
are specifically sensitive to the test compound. The procedures employed
during a particular study will be selected on a case-by-case basis depend-
ing on information available at the time of the study design, signs of tox-
icity observed during the  study, and/or the  purpose of the study.  It will
be the responsibility of those submitting to justify the selection of a spe-
cific test  from  the categories of electrophysiological  tests available. The
tests are adaptable so that they may be  used in sum or in part,  and either
alone or in  conjunction with other tests including:  A functional observa-
tional battery, motor activity, neuropathology, and general toxicity studies.
These studies may involve acute, subchronic, or chronic exposures.

     (c) Definitions. The definitions in  section 3  of the Toxic Substances
Control Act (TSCA) and the definitions in 40 CFR Part 792—Good Lab-
oratory Practice Standards apply to this  test guideline. The following defi-
nitions also apply to this test guideline.

     Neurotoxicity is  any  adverse effect on the  structure  or function of
the nervous  system related to exposure to a chemical substance.

     Toxic effect is any adverse  change in structure or function of an exper-
imental animal as a result of exposure to a  chemical substance.

     (d) Principle  of  the test method. The test substance is administered
to several groups of experimental animals,  each group  receiving  a different
dose level. Electrodes for recording brain electrical activity are temporarily

-------
or permanently affixed to the test animals. After electrodes are in place
and, if appropriate surgical recovery, stimuli for the visual, auditory  or
somatosensory sensory systems are presented to the test animals,  and the
resulting brain electrical potentials are recorded. The electrical potentials
recorded from animals treated with the test compound are  compared  to
those recorded from control animals.  The results are interpreted regarding
the extent to which treatment with the  test compound altered the normal
function of the sensory systems tested.

     (e) Test procedure—(1) Animal selection—(i) Species and strain.
Testing should generally be performed  in the laboratory rat, preferably a
pigmented  strain. Albino strains of animals  have abnormalities of the vis-
ual and auditory systems (see paragraphs (g)(4), (g)(5), and (g)(14) of this
guideline),  including:  The  absence of pigment from the retinal pigment
epithelial layer, high incidence of spontaneous retinal pathologies,  prob-
lems of photogenic retinopathy (under paragraph (g)(8) of this guideline),
abnormal pattern visual evoked potentials (under paragraph (g)(l) of this
guideline),  lack of normal pigment in the stria vascularis of the  cochlea
(see paragraph (g)(18) of this guideline)  and differential  susceptibility  to
ototoxic noise and drugs from pigmented strains (under paragraphs (g)(2)
and (g)(3)  of this guideline). However, it is recognized that under some
circumstances, use of another species or an albino strain may be justified.
If another species or an albino strain is used, the user must submit jus-
tification.

     (ii) Age. Animals should be young adults (42-120 days of age, for
rats) at the start  of exposure. Implantation of chronic electrodes in rats
younger than 60 days of age is not advised.

     (iii) Sex. (A) In order to reduce the number of animals  used,  only
one sex is required. Male rats may  be preferred because there is more
existing data on them. If, on the other  hand, females are known or expected
to be more sensitive to the test agent, they may be used. The user should
provide justification for the sex selected.

     (B) If females are used, they should  be nulliparous and  nonpregnant.

     (2) Number of animals. A  final sample size of at least 10  animals
should be used in each dose and control group. The number  of  animals
to be used should be based on appropriate  statistical methods and an allow-
ance for attrition due  to anticipated  problems such  as loss of electrode
preparations, etc.  Note that the rate of attrition should be estimated based
on the  experience  of the  laboratory performing the testing.  If interim
neurophysiological  evaluations are planned in long-term  dosing  studies,
it may be advisable to include an additional number of animals sufficient
for the interim studies. Animals should be randomly assigned to treatment
and control groups. If groups are not randomly assigned, justification must
be provided.

-------
     (3)  Control groups, (i) A concurrent ("sham" exposure) vehicle
control group is required. Subjects should be treated in the same way as
an exposure group except that administration of the test substance is omit-
ted. If the vehicle used has known or potential neuroactive properties, both
untreated and vehicle-control groups are required.

     (ii)  Positive control groups exhibiting functional  changes in the sen-
sory systems to be tested are required in order to demonstrate the capabil-
ity of the laboratory performing the testing to conduct the procedures. Sep-
arate  setups  for each sensory system are acceptable, but  not necessary.
In addition, for each sensory modality in vehicle or untreated control group
data,  it  should be demonstrated that the  mean of an amplitude-sensitive
dependent measure increases monotonically as a function of stimulus in-
tensity as defined in paragraph (e)(7)(v)(D) of this  guideline. Historical
data may be used if the essential  aspects of the  experimental procedure
remain  the   same.  Periodic updating  of positive  control data is rec-
ommended. New positive control data should also be collected when per-
sonnel or some other critical element in the testing laboratory has changed.

     (4)  Dose levels  and dose selection, (i)  At least  three dose  levels
should be used in addition  to the vehicle control  group. Ideally, the data
should be sufficient to produce a dose-effect curve. We encourage the use
of equally spaced  doses on a logarithmic scale,  and a rationale for dose
selection that will enable detection of dose-effect relations to the highest
degree possible. For acute  studies,  dose  selection may be made relative
to the establishment of a benchmark  dose (BD). That  is,  doses may be
specified as successive  fractions, e.g. 0.5, 0.25 of the BD. The  BD itself
may be estimated as the highest nonlethal dose as determined in a prelimi-
nary range-finding  study. A variety of test methodologies may be used
to determine  a BD, and the method chosen may influence subsequent dose
selection. The goal is to use a dose level that is sufficient to be judged
a limit dose, or clearly toxic. Alternatively, the BD  may be specified as
a dose of the test compound producing clear neurotoxic effects in previous
studies.

     (ii)  Acute studies. The high dose need  not  be  greater than 2 g/kg.
Otherwise, the high dose should result in significant neurotoxic  effects
or other  clearly toxic effects, but not result in an incidence of fatalities
that would preclude a meaningful  evaluation of the data. This dose may
be estimated by a BD procedure  as described above, with the middle and
low dose levels chosen as  fractions of the BD.  The  lowest dose should
produce minimal effects or,  alternatively, no effects.

     (iii) Subchronic and  chronic studies. The  high dose need not be
greater than  1 g/kg/day.  Otherwise, the high dose level should result in
significant neurotoxic effects or other clearly toxic effects, but not produce
an incidence of fatalities that would preclude a meaningful evaluation of

-------
the data. The middle and low dose  should be fractions of the high dose.
The low dose should produce minimal effects, or alternatively, no effects.

    (5) Route of exposure. Selection  of route may be based on several
criteria: The most likely route  of human exposure,  the greater likelihood
of observing effects, the practical difficulties, the  likelihood of producing
nonspecific  effects,  and existing data regarding the test compound.  Be-
cause more  than one route of exposure may be important for many mate-
rials, these criteria may conflict with one another. The route that best meets
these criteria should be selected.

    (6) Combined protocol. The tests  described herein may be combined
with any other toxicity study, as long as none of the  requirements of either
is violated by the combination.

    (7) Study  conduct—(i) Preparation of animals for recording. (A)
For electrophysiological  recording it usually will be necessary to implant
animals with chronic in-dwelling electrodes using stereotaxic surgical pro-
cedures. In some circumstances, acute attachment  of temporary electrodes
may be acceptable if criteria for humane treatment of animals, for record-
ing without undue anesthesia,  and  for data acceptability detailed below
can be met.  Chronic implantation of electrodes will  require surgical anes-
thesia  and surgical techniques  appropriate for the species  as outlined in
current laboratory animal care  guidelines under paragraph (g)(17) of this
guideline. Standard  animal  surgical  practices should be followed as out-
lined in a number of standard references (e.g. paragraph (g)(ll)  of this
guideline). Once anesthetized, animals  are usually placed in a stereotaxic
device in  order to position the head firmly. The stereotaxic device should
be designed to  prevent trauma  to the tympanic membranes and,  for audi-
tory studies, tympanic membranes should be examined after removal from
the stereotaxic device.

    (B)  Care should be taken during surgery to prevent drying of the
cornea through  means such as regular application of fluids such as mineral
oil, saline, or artificial tear  solution.  Once the animal is positioned  in the
stereotaxic device and the  implantation  site is prepared,  the electrodes
should be positioned with reference to standard skull markings and/or to
published brain atlas coordinates (see paragraph (g)(12) or (g)(13) of this
guideline).

    (C)  For recording potentials  which are generated in  sensory  cortex,
the recording or active electrodes are to be positioned as close as possible
to  the  brain sites generating the response. For "far-field" potentials such
as  the  brainstem auditory evoked potential, which are  conducted through
cranial tissues from  sites relatively distant to the recording electrodes, the
location of  the recording electrodes should be such as to provide good
resolution of the major waveform components, but need not be as close
as  possible to the generator sites. The  site of the reference electrode for

-------
differential recordings  should  be indifferent with  respect  to  stimulus-
evoked electrical activity to the extent possible. A ground electrode should
also be included.  The electrodes should be made of a material that is not
toxic to neural tissue. The electrodes should be made of a nonpolarizable
material, such as silver-silver chloride, if potentials of a frequency less
than approximately  1 Hz are to be reported. However, extra caution should
be exercised  in this case to avoid toxic effects  of such electrodes.  Elec-
trodes  should be  described  as to composition, size, shape, and position.

    (D) Wound sites should be treated and closed so as to prevent infec-
tions and to protect  the integrity of the electrodes. Following surgery, elec-
trode impedance should be measured in order to verify that electrode con-
nections are  intact  and functional. Following  surgery, animals should be
given sufficient time to  recover from  the  anesthetic and  surgical trauma
prior to testing;  ordinarily a period  of one week is recommended.  Prior
to testing,  the wound site  should be inspected  for  signs  of infection or
inflammation, and any animal showing  such signs should be removed from
the experiment.

    (E) For  acute studies, it is necessary to perform surgery prior to ad-
ministering the test compound. In repeated dosing, it may be necessary
to implant the electrodes during  the  course  of treatment with the test
compound  due to the  limited time which  electrode preparations may be
expected to remain  intact. It may be advisable to prevent exposure to the
test compound on the day of surgery  due to  possible interactions of the
effects of the test compound and the anesthetic agent. The time between
surgical implantation of electrodes and testing should be equal for all ani-
mals or, if unequal, balanced across  the treatment groups. Animals losing
electrode preparations  between time of surgery and testing should be re-
moved from the study and not submitted to  another surgery. At termination
of the  study, postmortem examination of each animal should be used to
determine if the  electrodes were incorrectly positioned,  or if the presence
of epidural electrodes  damaged the  underlying neural tissue. If either of
these two conditions is found, the data from  the involved animal should
be discarded.

    (ii) Testing environment. (A) Electrophysiological testing should be
conducted in a chamber or room which is isolated from extraneous light
and noise and controlled for temperature.  Background noise, light, and
room temperature should be reported.  Where possible, testing should be
conducted without the use of restraint  or anesthetics. The use of restraint
may be necessary when a specific  orientation to, or distance from, the
stimulation equipment is required, or when  movement of the animal would
interfere with recording.

    (B)  For  auditory  testing  of unrestrained,  unanesthetized  animals,
acoustic stimuli should be of equal sound pressure level wherever the ani-
mals'  ears  may be  placed during data acquisition. The animal  enclosure

-------
should be acoustically transparent to sound in the frequency band of the
stimulus spectrum. The  use of anesthetic is  permissible when  consider-
ations  of animal discomfort prohibit testing awake animals,  and when it
can be  reasonably expected that the effects of the test agent would not
be substantially different under anesthesia. If recording is to be performed
on anesthetized animals,  body temperature should be maintained within
a normal range  during testing. In addition, if recording from  anesthetized
animals, the power spectrum of the spontaneous electroencephalogram
(EEG) should be measured in order to control  depth of anesthesia. Position
of the stimulation device or devices relative to  the appropriate sensory
organs should be specified.

     (iii) Electrophysiological  recording. (A)  Electrophysiological re-
cording procedures should  follow generally accepted practice such  as  is
found  in standard reference texts (under paragraph (g)(9) or (g)(16) of
this guideline). Typically, it is appropriate to differentially record between
active and reference electrodes using an amplifier with high common mode
rejection and high input impedance relative to that of the electrodes.

     (B) Electrical shielding and grounding should be used  to  eliminate
activity in the electrophysiological recording at the frequency of the power
lines reflecting inductive noise. Electrophysiological amplifiers  and  filter
bandpass settings  must be appropriate to the signal being measured. For
example, ac-coupling is appropriate for most applications, but de-coupling
should be used  if steady  potentials are to be  measured. Analog  electrical
activity is typically digitized using an analog-to-digital converter  operating
at a  rate at least twice,  preferable higher, the highest frequency passing
the input filters. Analog  or  digital filtering of data is  appropriate to im-
prove signal-to-noise ratios.  If using analog filtering, the decay  functions
of the  high  and low bands of the filters should be specified,  and filtering
parameters should be selected to avoid amplitude reductions or phase shifts
in the frequency bands of the signal to be measured.

     (iv) Signal averaging. (A)  Signal  averaging of the input  data syn-
chronized with the stimulation is appropriate to increase the signal-to-noise
ratio. The number of trials  averaged should be sufficient to yield reliable
data, and a  waveform in control animals for  which the maximum ampli-
tude measure to be reported at the minimum  intensity  stimulus is at least
50 percent greater in amplitude than the  recording noise level. The dura-
tion  of the  sampling epoch should  be  sufficient  to encompass  all major
components  of  transient  evoked potentials.  Automated artifact  rejection
routines may be employed to reject occasional spurious data provided that
no greater than 50 percent of the original trials are rejected for any given
waveform, and all final averaged waveforms for a given stimulus  condition
are based on an equal number of trials.

     (B) The data for cases in which greater than 50 percent of the trials
are rejected due to artifacts should be discarded,  and the recording condi-

-------
tions  improved, if possible, to  allow more  suitable data collection.  The
recording noise level should be determined in order to illustrate the signal-
to-noise ratio. Acceptable methods to do so  include: examining  a portion
of the data preceding the presentation of the stimulus, but following the
last of the evoked activity from the previous stimulus in transient evoked
potentials; averaging  an equal number of trials as  the recording session
but without presentation of the eliciting  stimulus (for visual  stimuli this
should be accomplished by temporarily  blocking  the test subjects  view
of the stimulus with an opaque material); averaging alternate  trials of in-
verted polarity during the standard recording session, or reaveraging the
original raw data from the standard recording session over a similar num-
ber of trials in a manner nonsynchronous with the eliciting stimulus.  The
voltage gain  and temporal response properties of the electrophysiological
recording equipment  should be calibrated for each  experiment,  or more
frequently as needed. The  same recording conditions should  be used on
all animals tested.

    (v) Stimulation. The selection of stimulation parameters will depend
upon  the  specific goals of the  study.  Tests  should include measures of
visual, auditory and somatosensory systems, unless there are reasons to
limit testing to particular aspects of sensory  function. The testing of mul-
tiple sensory systems in the same animal has been demonstrated (see para-
graphs (g)(10) and (g)(15) of this  guideline). The order of presentation
of the different tests to individual subjects  be either random,  balanced
across treatment groups, or fixed.  If the test order  is fixed, justification
for the test order should be provided.

    (A) Visual stimuli.  Visual testing should include  separate  tests in-
volving light flashes and patterned stimuli.  Visual  pattern testing should
employ stimuli with a sinusoidal spatial luminance profile, and should in-
clude a range of pattern sizes which encompass the low,  middle and high
spatial frequency ranges of the contrast sensitivity function of the test spe-
cies. Techniques for recording visual evoked potentials using both flashed
(under paragraph  (g)(6) of this guideline) and patterned stimuli (under
paragraph (g)(l) of this guideline) are available for laboratory rats.

    (B) Auditory stimuli. Auditory testing should include  a  stimulus of
broadband frequency characteristics, such as a click. In addition, pure tone
stimuli  reflecting  the  low, middle and  high frequency  portions of the
audiometric function  of the test species  should be used. Techniques for
recording auditory evoked potentials using both click and  pure tone stimuli
(under paragraph (g)(15) of this  guideline) are available for laboratory rats.

    (C) Somatosensory  stimuli.  Somatosensory testing should include
electrical stimuli delivered to the tail or distal portions of the lower extrem-
ities. Techniques for recording somatosensory evoked potentials are avail-
able for laboratory rats (under paragraph (g)(15) of this guideline).

-------
     (D) Stimulus levels. (7) For studies designed to detect a change in
sensory function produced by a  compound for which little is known, it
is  sufficient to use a single stimulus  level for each visual,  auditory or
somatosensory  evoked potential.  Justification for the  stimulus  level se-
lected should be provided. For  studies designed to characterize a sensory
effect,  at least three different levels  of flash intensity,  pattern contrast,
acoustic click stimulus  sound pressure level  or  somatosensory stimulus
current  are  required for  each  stimulus condition.  The  stimulus  levels
should be chosen  on the basis of prior experience and/or pilot studies.

     (2)  In control animals, the  low level  should be  near the response
threshold, but large enough to  produce a response amplitude at least 50
percent  greater than the  recording noise level. Specification of the high
stimulus level varies with stimulus type. For flashed visual stimuli the high
stimulus level should either produce a  maximal response,  or be the maxi-
mum output  available from a conventional stimulator (e.g. Grass model
PS-22 photic stimulator). For patterned visual stimuli the high stimulus
level should either produce a maximal response,  or be below the highest
level of contrast within the linear range of the input-voltage/stimulus-con-
trast calibration function of the stimulus screen. For somatosensory stimuli
the high stimulus level should  either produce a maximal  response, or be
at  or below a current which produces minimal reflexive muscle movement
or other indications of discomfort. For acoustic stimuli the high stimulus
level should be below approximately  80 decibels sound pressure level in
order to avoid production of temporary  or permanent threshold shifts.
        For all types of stimuli, the middle stimulus level should produce
a response intermediate between high and low stimulus levels. The phys-
ical and temporal parameters of stimuli should be calibrated against known
standards for each experiment using commonly accepted procedures.

     (4) Visual stimulus luminance and contrast, or for flashes integrated
power, should be calibrated with an appropriate radiometer or photometer
against a known standard. Acoustic stimuli should be calibrated for level
using equipment which meets standards of the American National Stand-
ards Institute (ANSI) for sound level meters. Stimuli should be measured
and reported as peak levels, maximum root mean square (max RMS), or
"peak equivalent"  sound pressure levels. In addition,  the polarity of the
electrical stimulus and  the transduction system for acoustic stimuli should
be reported.

     (vi) Measurement.  (A) Dependent measures should be taken from
each evoked potential  which are sensitive to changes  of both amplitude
(voltage) and latency (time after stimulus onset) for transient evoked po-
tentials  or phase for  steady-state evoked potentials.  Enough measures
should be taken to adequately reflect  the shape  of the evoked  potential
in control animals.

                                 8

-------
     (B) A variety of measurement schemes may be acceptable, provided
they are specified a priori, do not ignore major portions of the waveform,
yield ratio-scale values, and meet the criterion for positive  controls speci-
fied in paragraph (e)(3)(ii) of this guideline. If the measurement scheme
involves inspection of each waveform by an operator and scoring  of the
waveform, the criteria for scoring should be objective and stated.

     (C) The experiment also must be conducted  so that the experimental
personnel  are unaware of the treatment of individual animals at the time
of scoring. Measurement schemes that minimize the  scoring by personnel
of each response are preferred. The same data scoring procedure must be
used on all subjects.

     (D) Previously collected data demonstrating selective effects  of the
test compound may be used to restrict the number of test parameters and/
or the number of measured endpoints in order to  examine more restricted
hypotheses.  Colonic temperature should be measured at the time of each
electrophysiological recording. Body weight should be measured on each
test day.

     (vii)  Acute. Testing  should be timed to  include the estimated time
of peak effect.

     (viii) Repeated dosing. Testing  should be conducted  after the com-
pletion of dosing with the test compound when  it can  be expected that
transient effects of the final treatment have dissipated. Additional testing
may be conducted during the course of treatment in order to provide infor-
mation on the emergence of toxic effects.

     (f) Data  reporting and evaluation. The following  should be re-
ported:

     (1) Description of the test methods. This must include:

     (i) Positive control data from the laboratory performing the test which
demonstrate the sensitivity of the procedure being used. Historical  control
data can be critical in the interpretation of study findings. We require sub-
mission of such data  to facilitate the rapid review of the significance of
the observed effects.

     (ii) Procedures for calibrating the stimulation and  recording  equip-
ment and balancing the groups.

     (2) Results. The following must be  arranged  by  test  group (dose
level).

     (i) In tabular form, data must be provided showing for each animal:

     (A) Its identification number.

     (B) Body weight for each day tested.

-------
     (C) Values of each evoked potential dependent measure

     (D) Body temperature of the animal at the time of acquisition of each
evoked potential.

     (iii) Group summary data should also be reported.  Data reporting
should include in tabular form measures of central tendency and variability
for each combination of stimulus conditions  and treatment with  the test
compound, and the  statistical  significance  level for  effects of treatment
with the test compound associated with each set of values. The final sam-
ple size should be reported  along with reasons for excluded or  missing
data. The noise level should be reported. Graphic presentation of the data,
or portions thereof, may also  be included. Also, samples of typical  individ-
ual animal evoked potential waveforms illustrating all stimulus conditions
and the noise level, or preferably  group mean waveforms of the same,
should be included.

     (3) Evaluations of the data—(i) Data analysis.  (A)  Numerical data
analysis should include a measure of central tendency, such as mean, and
a measure of variability, such as standard error of the mean, for each stim-
ulus and dose treatment combination, and for each dependent variable.

     (B) Statistical analysis should test the null hypothesis of no statis-
tically significant overall effect of treatment with the test compound across
stimulus conditions for each type of sensory evoked potential. In addition,
statistical tests  for interactions  between treatment with the test compound
and the manipulation of the stimulus parameters should be performed and
the results  reported.  The choice of statistical analysis  should consider the
experimental design and address the problem of adjustments for multiple
statistical analyses.

     (ii) Interpretation. The  report should include an interpretation of the
neurotoxicological   significance  of   the   findings,   and  relate   the
neurophysiological results to  those of other neurotoxicological results, and
to other data to the extent possible. Guidance  for interpretation of sensory
evoked potential data is described under paragraph (g)(18) of this guide-
line.

     (g) References. The following references should be consulted for ad-
ditional background information on this guideline.

     (1) Boyes, W. K.  and R. S. Dyer. Pattern reversal visual evoked po-
tentials in awake rats. Brain Research Bulletin  10:817-823 (1983).

     (2) Conlee J.W. et al. Differential susceptibility to noise-induced per-
manent threshold shift between albino and pigmented guinea pigs. Hearing
Research 23:81-91 (1986).

                                 10

-------
     (3)  Conlee  J.W.  et  al.  Differential susceptibility  to  gentamicin
ototoxicity between albino  and pigmented guinea pigs. Hearing Research
41:43-52(1989).

     (4) Creel, D. Inappropriate use of albino animals in research. Phar-
macology andBioochemical Behavior 12:969-977 (1980).

     (5) Creel, D. Albinism and evoked potentials: Factors in the selection
of infrahuman models in predicting the human response  to neurotoxic
agents. Neurobehavioral Toxicology and Teratology 6:447-453 (1984).

     (6) Dyer, R. S. and H. S.  Swartzwelder.  Sex and strain differences
in the visual evoked potentials of albino  and hooded rats. Pharmacology
and Biochemical Behavior 9:301-306 (1978).

     (7) Herr, D.W.  and Boys, W.K. Electrophysiological analyses of com-
plex brain systems;  Sensory evoked potentials and their  generators. In:
Neurotoxicology Approaches and Methods, L.W. Chang and W. Slikker,
eds. Academic Press, NY, 205-221 (1955)

     (8) Heywood, R. and C. Gopinath. Morphological assessment of vis-
ual dysfunction. Toxicology and Pathology 18:204-217 (1990).

     (9) International Federation of Societies for Electroencephalography
and Clinical Neurophysiology. Recommendations for the Practice of Clini-
cal Neurophysiology.  Elsevier  Science,  Amsterdam,  The Netherlands
(1983).

     (10) Mattsson,  J. L. and R. R. Albee. Sensory evoked potentials in
neurotoxicology. Neurotoxicology and Teratology 10:435-443 (1988).

     (11) Myers,  R.D.  Methods  in Psychobiology. Academic Press, NY
(1971).

     (12) Paxinos, G. and Watson, C. The Rat Brain in Stereotaxic Coordi-
nates. Academic Press,  Sydney (1982).

     (13) Pelligrino, L. J. et al. A Stereotaxic Atlas of the Rat Brain, 2nd
ed., Plenum, NY (1979).

     (14) Prieur, D.  J. Albino animals: Their use and misuse in biomedical
research. Comparative Pathology Bulletin XIV(3):l-4 (1982).

     (15) Rebert, C. S. Multisensory evoked potentials in experimental and
applied  neurotoxicology. Neurobehavioral  Toxicology and  Teratology
5:659-671  (1983).

     (16) Thompson, R.F. Methods in Physiological Psychology,  Vol 1:
Bioelectric Recording Techniques; Parts A, B and C. Academic Press, NY
(1973 and 1974).

                                 11

-------
    (17) U.S. Department of Health and Human Services,  Public Health
Service, National Institutes of Health,  Publication No. 85-23, Guide for
the care and use of laboratory animals, Revised 1985.

    (18)  U.S.   Environmental   Protection  Agency.   Guidelines  for
Neurotoxicity Risk Assessment. FEDERAL REGISTER 63 FR 26926-26954,
May 14, 1998.

    (19) Witkop, C. J. et al. (eds), The Metabolic Basis of Inherited Dis-
eases, McGraw-Hill, NY, pp. 301-346 (1983).
                                12

-------