Content uploaded by Caio Maximino
Author content
All content in this area was uploaded by Caio Maximino
Content may be subject to copyright.
NATURE PROTOCOLS | VOL.5 NO.2 | 2010 | 209
p
uo
r
G
gn
i
h
s
i
lb
uP eru
t
a
N
010
2
©natureprotocols
/
m
o
c
.
e
r
u
t
a
n
.
w
w
w
/
/
:
pt
t
h
PROTOCOL
INTRODUCTION
The dark/light preference model is already established as an
‘ethoexperimental’ anxiety model in rodents1. It is based on the
natural aversive quality of brightly lit environments for mice,
shaping a conflict situation in which the animal must deal with
its natural tendency to explore in the face of an unpleasant envi-
ronment. The rodent dark/light preference model is an explora-
tion model, in the context that it measures locomotor activity
in both environments as an index of anxiety2–5. The proposed
actinopterygian dark/light preference task is a modification of
an experimental manipulation used by Edward Thorndike6 dur-
ing the beginning of the twentieth century to study motivation
and learning in fish, and afterward by Satake and Morton7 in the
1970s to establish the effects of noradrenergic substances on the
scotophobic (i.e., dark-avoiding) behavior of pinealectomized or
scotophobin-injected goldfish. Recently, the proposed model was
used to establish dark/light preference in zebrafish Danio rerio8,
bluegill Lepomis macrochirus, crucian carp Carassius langsdorfii9,
goldfish Carassius auratus9–12, guppy Poecilia reticulata12, Nile tila-
pia Oreochromis niloticus12, lambari Astyanax altiparanae, cardi-
nal tetra Paracheirodon axelrodi12 and banded knifefish Gymnotus
carapo12, and to screen for the neurobehavioral effects of ethanol13
and dopaminergic drugs14 in zebrafish. The main advantage of
this protocol is the presentation of a clear conflict situation for
the fish; however, most models that investigate innate ‘fear’- and
‘anxiety’-like behavior in fish do not use such conflict. With the
exception of predator inspection tests15–17, most innate anxiety
tests use either the exploration of an open field or the vertical
distribution of the animal in the water column to measure this
variable14,18–32 and aim to describe individual variability across
a ‘shyness–boldness’ continua15,33–37. This continuum can be
mapped to Budaev’s15,34,35 two dimensions of ‘temperament’ in
fish (activity-exploration and fear-avoidance), which in turn
are trait instantiations of approach–avoidance state dimen-
sions38–40. Ex hypothesi, these dimensions are best analyzed using
conflict models41,42. This protocol will focus on the cyprinids
zebrafish and goldfish; however, it has been successfully applied
to non-cyprinids without alterations to it12 (with the exception
of the gymnotid G. carapo, for which different dimensions of
the test tank were needed).
Behavioral responses in the dark/light preference protocol are
readily assessable and quantifiable by an observer, without the need
for extensive training. Briefly, animals are placed in an intersection
compartment, located between one white and one dark compart-
ment; these spectra can be discriminated by both zebrafish43 and
goldfish44, and even though the authors are unaware of behavio-
ral spectral sensitivity experiments in the other species in which
scototaxis has been observed, the very existence of this trait is
indirect evidence for this sensitivity. This intersection compart-
ment is enclosed by two sliding doors, which are removed follow-
ing a habituation interval of 5 min. The animal is then allowed to
explore the apparatus freely, and the observer records its behavior
(number and/or duration of entries in either the black or white
compartments) for 15 min. This procedure is based on the work of
Mattioli’s group8, which observed the natural preference zebrafish
for the black compartment. The duration of the session was
increased to 15 min, which is 5 min more than the original experi-
ment described by Serra et al.8, because the time course of explo-
ration in this apparatus reveals significant changes after the tenth
minute of observation. Behavior in this task (i.e., activity in the
white compartment) reflects a conflict between the preference of
the animal for protected areas (e.g., black substrata, in a process of
crypsis) and an innate motivation to explore novel environments.
Crypsis in fish depends on the dorsal distribution of melanophores
in the fish surface; the presence of dorsal and dorsally oriented
melanophores tends to minimize refraction and reflection of
light incident on the fish, reducing the effectiveness of visual infor-
mation for predators45. Appropriate cryptic coloration requires
the appropriate background and behavior (in the present task,
preference for dark substrata). Antianxiety effects (i.e., increased
time in the white compartment) can be determined simultane-
ously with a measure of increased or decreased swimming activity
(i.e., total number of entries in both compartments). Other meas-
ures that can be observed are thigmotaxis (i.e., the tendency for the
animals to stay nearer to the walls of the apparatus, especially
Scototaxis as anxiety-like behavior in fish
Caio Maximino1, Thiago Marques de Brito2, Claudio Alberto Gellis de Mattos Dias1, Amauri Gouveia Jr1, 3
& Silvio Morato2
1Laboratório de Neurociências e Comportamento, Instituto de Ciências Biológicas, Universidade Federal do Para, Belém, Brazil. 2Laboratório de Comportamento
Exploratório, Departamento de Psicologia e Educação, FFCLRP, Ribeirão Preto, Brazil. 3Núcleo de Estudos e Pesquisa em Comportamento, Departamento de Psicologia,
Universidade Federal do Para, Belém, Brazil. Correspondence should be addressed to C.M. (caio@ufpa.br).
Published online 14 January 2010; doi:10.1038/nprot.2009.225
The scototaxis (dark/light preference) protocol is a behavioral model for fish that is being validated to assess the antianxiety
effects of pharmacological agents and the behavioral effects of toxic substances, and to investigate the (epi)genetic bases of
anxiety-related behavior. Briefly, a fish is placed in a central compartment of a half-black, half-white tank; following habituation,
the fish is allowed to explore the tank for 15 min; the number and duration of entries in each compartment (white or black)
are recorded by the observer for the whole session. Zebrafish, goldfish, guppies and tilapias (all species that are important in
behavioral neurosciences and neuroethology) have been shown to demonstrate a marked preference for the dark compartment. An
increase in white compartment activity (duration and/or entries) should reflect antianxiety behavior, whereas an increase in dark
compartment activity should reflect anxiety-promoting behavior. When individual animals are exposed to the apparatus on only
one occasion, results can be obtained in 20 min per fish.
210 | VOL.5 NO.2 | 2010 | NATURE PROTOCOLS
p
uo
r
G
gn
i
h
s
i
lb
uP eru
t
a
N
010
2
©natureprotocols
/
m
o
c
.
e
r
u
t
a
n
.
w
w
w
/
/
:
pt
t
h
PROTOCOL
while exploring the white compartment)14,22,25,26,28,29,31,46,47; the instant
velocity of entries in each side (i.e., mean duration of each entry,
instead of the total duration of entries) as well as the frequency of
body and caudal fin (BCF) transient propulsions48; frequency of
‘freezing’ events (equivalent to Webb’s ‘non-swimming’48); vertical
distribution (top versus bottom of the water column)29,32; ventila-
tion rate49–52; and changes in the coloration of the animal13,14,47,53–55;
however, this final measure can be difficult to obtain when the
animal is in the black compartment.
Negative controls for this test can be made using all-white and
all-black tanks, with the same dimensions as the tank described
below8,12; if different animals are used for each tank in the
validation procedure, then no statistically significant lateral
preference (i.e., time spent in each compartment) should be
observed, showing that the preference in the dark/light tank
is indeed for the dark side of the apparatus and not for some
other spatial constraint.
Face validity of scototaxis
The dark/light preference protocol has face validity (i.e., the ability
of a given task to appear to measure what it is supposed to measure,
in that the given task ‘looks like’ the endophenotype56 that it is mod-
eling57). Bakshi and Kalin58 proposed that the main endophenotype
of anxiety is an aberrant expression of defensive behavior. As an
example, the anxiety or fear of brightly lit spaces is measured in this
model; the white compartment is avoided, and animals (zebrafish)
spend most time (71.3 ± 26.2%) in the black compartment12. Other
anxiety-related behaviors, such as thigmotaxis, freezing and BCF
transient swimming, also occur more in the white than in the black
compartment (unpublished observations).
Predictive validity of scototaxis
The dark/light preference protocol has predictive validity, defined,
based on a psychometric point of view59–61 as the extent to which
the dependent measure predicts behavior on a related meas-
ure62. Unpublished data from our laboratory (Laboratório de
Neurociências e Comportamento, UFPA) show that increased
dark compartment activity occurs in zebrafish, which also dem-
onstrate increased central area entries in the open field (r2
[gl = 13] =
0.89, P < 0.01). Predictive validity is also defined as the degree to
which a model can predict the effect of antianxiety compounds
with proved clinical efficacy2,63–65. It has been reported66 that anxi-
olytic compounds predictably alter the time that zebrafish spent
in the white compartment of the apparatus, and similar results
were obtained, using chlordiazepoxide (0.0, 0.02 and 0.2 mg kg − 1),
in the Laboratório de Neurociências e Comportamento (153.4 ±
11.4% [0.02 mg kg − 1] and 121.7 ± 4.1% [0.2 mg kg − 1] in relation
to controls) (unpublished data).
Construct validity of scototaxis
The construct validity of most anxiety models rely on linking them
to species-specific defense reactions67. In this model, a behavioral
co-adaptation to heightened dorsal distribution of melanophores
(i.e., preference for dark substrata to enable crypsis) is central to the
construct in question. In animal models, the construct of anxiety
is sometimes defined in terms of opponent motivational processes
being ‘activated’ simultaneously, creating an approach–avoidance
conflict: at the same time when animals tend to explore novel
environments, they also tend to express fear of those novel
locations, which are potentially dangerous38,40,68–72. The result is
that fish will spend more time in the ‘safe’ environment (i.e., the
dark compartment of the tank), making short incursions into the
potentially ‘risky’ environment (i.e., the light compartment of the
tank)12, a pattern of exploratory behavior called ‘risk assessment’
or ‘antipredator apprehension’67.
Effects of rearing in ‘enriched’ environments
Rearing in ‘enriched’ environments has been used in studies assessing
anxiety behavior of rodents73–75. In the Laboratório de Neurociências
e Comportamento, we compared the effects of rearing in an enriched
with impoverished environment on the exploratory behavior of
zebrafish in the dark/light preference task; significant increases were
observed in the time spent in the white compartment (150.1 ± 15.3%,
in relation to the impoverished environment) and decreases in the
frequency of BCF transient swimming (78.7 ± 3.5%) and freezing
(67.3 ± 9.8%) in this compartment. Our ‘enriched’ tank had hiding
places, rocks, vegetation and natural substratum, whereas the ‘impov-
erished’ tank had only water quality, temperature, aeration and light-
ing controls. These observations support the hypothesis that rearing
in ‘enriched’ environments reduces anxiety-like behavior in fish, an
effect that has been observed in rodents73–75—in which it has been
suggested that environmental enrichment reduces emotionality and
induces ‘facilitation’ of motor skills73.
Multiple test sessions in the dark/light preference task
Contrary to rodent models such as, e.g., the elevated plus-maze and
dark/light transitions test, the dark preference (at least in zebrafish)
does not present the so-called ‘one-trial tolerance,’ an effect in
which animals exposed a second time to the same apparatus tend
to present sensitization76 (i.e., increased time spent in the ‘safe’
environment) and there is lack of effect of benzodiazepines77–80.
In a 1-week experiment in which animals were tested daily (at the
same photoperiod), no evidence of sensitization or habituation
was observed in any parameter tested81 (time spent in each com-
partment, number of transitions, latency for exploration) (average
measures intraclass correlation r = 0.983, F(8,36) = 58.095, P < 0.001,
in a one-way random effects model where subject effects are ran-
dom, and H0 for the F-test is that the true value for the intraclass
correlation coefficients is 0).
Developmental factors in the dark/light preference task
Some results indicate that the preference for dark environments
in zebrafish are age-dependent82, which is probably an effect of
the maturation of melanophores83,84. Thus, in larval zebrafish,
which show no pigmentation, a preference for light environments
is observed14,85, whereas a preference for dark environments is
observed in adult zebrafish8,12. These developmental differences
should be taken into account, given that melanophores are impor-
tant for the ontogeny of defensive behavior in fish45.
Comparison with other tests
Currently, there are few behavioral tests of anxiety, temperament
or emotionality in fish. The open-field, a test in which the animal’s
exploration of a novel environment is the dependent measure, has
been used to test dopaminergic and glutamatergic drugs14,23,26,29,
whereas the vertical distribution test has been used to assess sub-
stances that act on the nicotinic receptor30,32. However, none of
these tests have been validated using antianxiety drugs, such as
benzodiazepines and serotonergic agents. Currently, this validation
NATURE PROTOCOLS | VOL.5 NO.2 | 2010 | 211
p
uo
r
G
gn
i
h
s
i
lb
uP eru
t
a
N
010
2
©natureprotocols
/
m
o
c
.
e
r
u
t
a
n
.
w
w
w
/
/
:
pt
t
h
PROTOCOL
is ongoing in our laboratory. Comparison between the results of
these tests is difficult because different drugs and treatments have
been used in each model. In terms of ease of application, as well as
predictive, face and construct validity, the scototaxis test is at least
as good as the other tests. However, the open-field is likely to suffer
from difficulty in isolating locomotor and emotional effects (as is
the case with rodents86), which should not happen in the proposed
dark/light test and in the vertical distribution test.
General background information relating to the procedure
Timing of test. Zebrafish present circadian rhythmicity in
locomotor behavior that is entrained by environmental markers
(‘zeitgebers’) such as feeding regimens and the dark–light cycle87;
this rhythmicity is probably related to diel vertical migrations, which
impose a tradeoff between predation risk and the existence of anti-
predator ‘windows’ for feeding in near-surface waters at dawn and
dusk88–90, akin to the tradeoffs that exist in situations in which risk
is temporally variable72. Consistent with these observations, we have
observed in the Laboratório de Neurociências e Comportamento
that the preference for dark environments is inverted when animals
are tested during the night (from 1800 to 0000 hours)(H[df = 3] =
10.787, P = 0.013) (C.A.G.M. Dias, unpublished data). As such, it
is preferable that animals are tested during the day.
Handling of animals before testing. In procedures such as the
rodent-elevated plus-maze62 and the dark/light preference test1,5,62,
‘it is important to ensure that in [such experiments], handling
of rodents and any experience with prior stressors, particularly
immediately before testing, is consistent across animals and treat-
ment groups’62. The stressful character of handling is prominent in
fish91–95, especially given that prolonged handling can rapidly lead to
hypoxia, a further source of stress in these animals96. The transfer
of animals from housing to test tanks should therefore be as quick
as possible. If any injection procedure is necessary, animals should
be anesthetized. For anesthesia, 100–200 mg liter − 1 tricaine meth-
anesulfonate (MS-222) solution can be used for induction, and
50–100 mg liter − 1 for maintenance; alternatively, menthol97 (100
mg liter − 1) or clove oil98,99 (150 mg liter − 1) can be used. Recovery is
attained by returning the animals to fresh, well-aerated water100.
Choice of material for the tank. The tank should be constructed
of matte acrylic that is as non-reflective as possible. Zebrafish are
a naturally shoaling species101, and the use of reflective surfaces for
the tank can affect stimulus control, as the animals might display
social behavior in relation to their own reflections.
Behavioral sampling procedures and session duration. Although
using what Dunbar102 calls ‘point sampling,’ in which the behavior
state that the animal is performing at a predetermined ‘point’ in
time and its duration are recorded, can be convenient for observers,
this sampling method is not enough to establish preference103. In
our experiments, we use ad libitum sampling for the whole 15-min
session, which not only results in better measures of preference104
but also enables observation of other behavioral acts.
Other potential uses. An identical apparatus has also been used
to assess learning in goldfish105. In this procedure, the animal is
placed in the white compartment of the apparatus, and the time it
takes to swim to the black compartment is the dependent measure;
decreasing latencies over trials represent the acquisition of inhibi-
tory avoidance.
Another use of this protocol, although indirect, is the design of
‘biased’ tanks for place preference conditioning106. The initial aver-
sion to the light compartment is overcome by the rewarding effect of a
drug, and drugs can be compared in terms of their reward potency by
assessing the degree to which this aversion is supplanted across trials.
MATERIALS
REAGENT SETUP
Fish species The choice of species for use in this experiment depends mainly
on a high density of dark melanophores in the dorsum of the animal. In the
Laboratório de Neurociências e Comportamento, we demonstrated the
existence of this pattern in zebrafish, goldfish, fighting-fish Betta splendens
(Bruno Rodrigues dos Santos, unpublished data), Nile tilapia, Cardinal tetra,
guppy, lambari and banded knifefish12. Each of these species is also being
used as a model animal in other types of research, including speciation
events, development, social behavior and sexual selection. Thus, choosing
species is more a matter of validating the preference for them and matching
scototaxis with the more general questions being asked by the experimenters.
Zebrafish was one of the first species to be tested in this paradigm8 because
it has become one of the preferred laboratory model organisms in genetics
and developmental biology107.
The description of the behavior of some of these species was made
elsewhere12; some species differences are of note (Fig. 1). Briefly,
preference for dark environments was found in all these species; however,
the banded knifefish and the Nile tilapia failed to present enhanced locomo-
tion in an all-white tank (a positive control for the validation of scototaxis;
see ‘Validation of test in laboratory’, under ‘Equipment Setup’). As such,
use of species other than the ones presented here demand validation.
Animal housing Animals should be housed in accordance with guidelines
for the use of fish in research108,109, as well as in accordance with the more
general guidelines for the use of animals in research. Briefly, animals should
be group housed in water conditions that are adequate for the species.
The main water quality parameters are temperature, oxygen saturation,
nitrogen compounds, carbon dioxide, pH and salinity; information about
ideal levels for each species can be found in databases such as FishBase.
Table 1 presents optimal values for these parameters for zebrafish, guppy
and goldfish. Other environmental parameters, such as light and noise
levels, also need to be considered.
For example, as schooling fish, zebrafish can be kept at fairly high
densities: ~20 fish can be kept in volumes ranging from 400 ml as young larvae
(1–5 d post-fertilization; dpf) to 3 liters as the fish approach juvenile stage
(3 months, defined as the beginning of reproductive age)100. For general activity
assessment, tanks should be made of transparent glass, Plexiglas (acrylic) or
polycarbonate, enabling easy observation of the animals. A different tank, for
breeding, should be smaller (24 cm × 12 cm × 12 cm), and present either artificial
or natural glass in the bottom, as well as a bottom layer of glass marbles.
Preparation of drugs and toxins Administration routes for drugs and
toxins can vary; most used are intraperitoneal injection and waterborne
administration. The first procedure has the disadvantage of stressing the
animal (even when the animal is anesthetized, as it should be in any
procedure involving injections); thus, identical handling and injection
procedures must be used for control and treated fish. The second procedure
is used in most of the studies so far24,25,27,28,47,110,111 but has the disadvantage
of producing uncertainty in relation to the system bioavailability of the
drug in question; this limitation can be avoided if the researcher has access
to high-speed liquid chromatography and tandem mass spectrometry to
measure the presence and amount of the drug in the nervous system, plasma
or urine112. When using this latter route of administration, we recommend
dissolving the stock solutions in a 250-ml beaker containing 200 ml of
de-chlorinated tank water, at a temperature of ~27 °C. Drug doses should be
calculated based on the weights of the salts. Stock solutions to be dissolved
212 | VOL.5 NO.2 | 2010 | NATURE PROTOCOLS
p
uo
r
G
gn
i
h
s
i
lb
uP eru
t
a
N
010
2
©natureprotocols
/
m
o
c
.
e
r
u
t
a
n
.
w
w
w
/
/
:
pt
t
h
PROTOCOL
in de-chlorinated tank water should contain the
drug of interest, plus associations. For diazepam,
e.g., a 5 mg ml − 1 stock solution should contain
40% propylene glycol, 10% ethyl alcohol, 5%
sodium benzoate and 1.5% benzyl alcohol. This
stock solution can be either injected or
dissolved in 200 m of de-chlorinated tank water.
EQUIPMENT
Buckets or 2 L beakers for temporary housing
Fish nets
Black and white acrylic tank (description in text)
Video camera
Automated (e.g., Ethovision, MouseTracker) or
nonautomated (e.g., Etholog, JWatcher, X-Plo-
Rat) behavioral transcription software
Discard tank, glass
EQUIPMENT SETUP
Transport and temporary housing equipments
Animals should be transferred from temporary
housing tanks (bucket or beakers) by using fish nets.
Dark/light preference tank Here we use the apparatus described else-
where8,12 and presented in Figure 2, although variations exist. An acrylic tank
(15 cm × 10 cm × 45 cm height × width × length) is used that is divided
equally into one-half black and one-half white. Walls and bottom are either
black or white, so as to warrant uniform substrata for each compartment.
The water column is kept at 10 cm, yielding a final volume of
4.5 liters. Water should be de-chlorinated. The colored material chosen
should not be reflective, to avoid the tendency of those animals that present
shoaling tendencies113,114 to behave in relation to their own reflection. The
tank contains central sliding doors, colored with the same color of the
aquarium side, thereby defining an uncolored central compartment
measuring 15 cm × 10 cm × 10 cm.
The species were used in our laboratory are small, ornamental fish, meas-
uring a maximum 5 cm long. For longer animals, the size of the tanks should
be changed accordingly. For example, in our tests with the banded knifefish,
which measures ~10 cm, we used a tank measuring 15 cm × 10 cm × 55 cm,
resulting in a central compartment of 15 × 10 × 20 cm12.
Location and/or illumination Tanks should be illuminated by environmental
light (75 W light bulb, located at 1.80 m above the aquarium top), which keeps
illumination uniform and constant between trials. At this distance, environ-
mental light tends to produce an average of 975 lux right above the tank.
Data collection apparatus The primary method for data collection used in
our laboratory is the manual transcription of video-recorded experiments
•
•
•
•
•
•
using an adapted behavioral transcription software (X-Plo-Rat v1.1.0). Other
software are available for this type of data collection, following a similar
rationale (i.e., recording the duration of ‘behavioral states’ such as entrance
in a given compartment, e.g., Ethovision, EthoLog and JWatcher). If compu-
terized transcription is not available, an observer can make hatchmarks on
a data sheet for each compartment entry that the fish makes and use a timer
to determine the duration spent in either compartment. It is also possible to
use automated video tracking software if the contrast between background
and the animal are altered; this can be done using GPL-licensed video editing
software such as VirtualDub.
Equipment for validation of test in laboratory Interspecific variation in this
test has been observed12, and scototaxis should not be treated as a fixed trait in
a given species. As such, cross-validation should be made in each laboratory, as
to warrant that the population or species in question presents this trait. Nega-
tive controls for this test can be made using all-white and all-black tanks, with
the same dimensions as the tank described above8,12; if different animals are
used for each tank in the validation procedure, then no statistically significant
lateral preference (i.e., time spent in each compartment) should be observed,
showing that the preference in the dark/light tank is indeed for the dark side
of the apparatus, and not for some other spatial constraint. Also, the all-white
tank should produce increased locomotion, as a signal of increased anxiety-
like behavior8,12; however, the opposite result (decreased locomotion in the all-
white tank) can also be expected, as freezing can be an anxiety-like behavior.
Discard tank After animals are tested, they should be removed carefully
from the test tank and transferred to a discard tank, with the same conditions
found in the housing tank (see ‘Animal housing’). Although multiple testing
does not affect the behavior of zebrafish (see ‘Multiple test sessions in the
dark/light preference task’), discarded animals should preferentially not be
used in another scototaxis session, especially if they were subjected to drug
or toxicant treatments.
1,000
800
600
400
Guppy Lambari Zebrafish Tetra Knifefish SwordtailMosquitofish
Total time (s)
200
0
Figure 1 | Different
species of fish vary
in the intensity of
their preference for
the dark environment,
and some species
might not present a
preference for either
dark (black bars) or
light (white bars) environments. This appears to be the case with mosquitofish (Gambusia holbrooki),
which, contrary to other poeciliids such as guppy and swordtail (Xiphophorus helleri), does not present
a preference for dark environments. Other species of fish, however, spend more time in the dark than
in the bright portion of the aquarium. Data for guppies, lambaris, zebrafish, cardinal tetras and banded
knifefish taken from Maximino et al.12; data for mosquitofish and swordtail are unpublished. The
experiments were performed in compliance with the recommendations of SBNeC (Brazilian Society
for Neuroscience and Behavior), which are based on National Institutes of Health’s Guide for Care and
Use of Laboratory Animals, and also complied to the Canadian Council on Animal Care’s Guidelines
on the Care and Use of Fish in Research, Teaching and Testing.
TABLE 1 | Water quality parameters for three species of fish.
Parameter Species
Zebrafish
Danio
rerio
Guppy
Poecilia
reticulata
Goldfish
Carassius
auratus
Temperature (°C) 18–24 18–28 < 41
Dissolved oxygen
(mg liter − 1)
6.98–7.65 4.5 5.5–7.0
Nitrogen compounds
(mg liter − 1)
NH3–N:
< 0.16
NH3–N:
0.11–0.16
NH3–N:
< 0.15
NO2–N:
< 0.009
NO2–N:
0.008–0.009
NO2–N:
< 0.01
pH 6.0–8.0 7.0–8.0 6.0–8.0
Salinity (%) < 6 < 35 < 2
Information for zebrafish and goldfish were included because these species are being increasingly
used in behavioral studies in neurosciences and pharmacology; information for guppy was included
because this species is widely used in behavioral ecology experiments. Information retrieved from
FishBase.
5 cm
22.5 cm
10 cm
15 cm
Figure 2 | Test apparatus for the proposed protocol. The apparatus should
be made of matte acrylic, as described in the text, and the size depicted
is suitable for fish that have a maximum body length of 5 cm. Longer
species require proportionally longer tanks.
NATURE PROTOCOLS | VOL.5 NO.2 | 2010 | 213
p
uo
r
G
gn
i
h
s
i
lb
uP eru
t
a
N
010
2
©natureprotocols
/
m
o
c
.
e
r
u
t
a
n
.
w
w
w
/
/
:
pt
t
h
PROTOCOL
PROCEDURE
Setup ● TIMING 0.5–1 h per cohort of experimental animals
1| Ensure the apparatus is filled with de-chlorinated water before use (water column = 10 cm). Fill out data sheets with
subject number of animal, date, coded condition and experimenter initials before testing.
2| Bring the animal in an individual temporary transport recipient (e.g., bucket or beaker) into the behavioral test room.
3| Transfer the fish to the central compartment of the test tank with a net.
CRITICAL STEP Ensure that all animals are handled in a consistent manner and that each animal is placed in the central
compartment in the same position.
Testing ● TIMING 15 min per animal
4| Following a 5-min habituation period, remove the central sliding doors, allowing the fish to swim freely in the apparatus.
5| Record the locomotor activity for 15 min.
CRITICAL STEP During the habituation period, take precautions to begin data collection as soon as the central sliding
doors are removed. If sessions are video-recorded, this can be done by beginning records during the habituation period.
CRITICAL STEP Entry into either compartment is recorded when approximately two-thirds of the body length of the animal
has crossed the midline.
CRITICAL STEP Observers must avoid unnecessary movements and noise. If possible, sessions should be recorded with
an overhead camera, so that behavioral recording sessions take place in an isolated room. If this is not possible, the use of
physical barriers to block vision to the outside of the aquarium is useful103.
? TROUBLESHOOTING
Cleanup ● TIMING 1–2 min per animal
6| Transfer the animal from the test apparatus to a discard tank with a net.
7| Discard the water used for test, wash the tank thoroughly with tap water and refill it with de-chlorinated water before
testing with another animal.
Data analysis
8| Perform preferred data analysis. In a previous study12, we used non-parametric (Kruskal–Wallis) analyses of variance to
assess the effects of treatment on the following measures: latency for first choice of environment; duration of entries in each
compartment (Fig. 1); and number of midline crossings. However, further unpublished study revealed that other variables,
such as latency to begin exploration after an environment was first chosen, frequency of entries in each compartment and
mean duration of entries, are sensitive to manipulations. In addition, non-parametric analyses of variance are not the only
way to assess the effects of independent variables on dependent variables; however, normality of the data should be
assessed before parametric tests are used. For more information, refer to standard statistics texts.
● TIMING
Steps 1–3, setup: for cohort of experimental animals, 0.5–1 h
Steps 4 and 5, testing: 15 min per animal.
Steps 6 and 7, cleanup: 1–2 min per animal.
? TROUBLESHOOTING
Step 5: Fish jumps out of the aquarium
Infrequently (this happened twice in a block of 60 animals in our laboratory, under the conditions described above), the ani-
mal ‘jumps’ out of the test tank and falls off the apparatus. When this occurs, the experimenter must rapidly pick up the ani-
mal and discard it in the discard tank. Behavioral data from an animal that does this should be excluded from the analyses.
Step 5: Fish freezes on a given compartment after first choice
Occasionally (this happened only once in our laboratory, probably owing to an accident that happened in the testing room),
animals might freeze after they choose a compartment and no longer explore the apparatus for the whole 15-min session.
This is especially true with very stressful manipulations, or if there is noise or unnecessary movements in the experimenta-
tion room. If this happens, data from this animal should not be discarded from the analyses; however, the experimenter
should keep track of the freezing fish, and freezing behavior itself should be analyzed (either by recording its frequency in
individual fish or recording the number of fish which froze).
214 | VOL.5 NO.2 | 2010 | NATURE PROTOCOLS
p
uo
r
G
gn
i
h
s
i
lb
uP eru
t
a
N
010
2
©natureprotocols
/
m
o
c
.
e
r
u
t
a
n
.
w
w
w
/
/
:
pt
t
h
PROTOCOL
Step 5: Fish demonstrates excessive thigmotaxis
If the animal spends too much time (~30% of the total test time) clinging to a particular wall (or to walls in general) of the
apparatus (maximum distance of ~2 cm from the wall), this might be a sign that the fish is behaving in response to its own
reflection. In this case, changing the material of which the apparatus is made should be considered, to avoid confounding
variables. If the material used is absolutely non-reflective, however, this could represent a pattern of exploratory behavior22.
The experimenter should keep track of the thigmotactic fish, and thigmotaxis itself should be analyzed (either by recording
its frequency in individual fish or by recording the number of fish that displayed it).
ANTICIPATED RESULTS
The following results exemplify data that we have obtained using the dark/light preference protocol to investigate the an-
tianxiety effects of rearing in an enriched environment in goldfish and zebrafish115. The results are robust and replicable, and
show generalization across both species. This type of manipulation yields consistent antianxiety effects in rodent models of
anxiety73–75; an experiment conducted in our laboratory (Laboratório de Neurociências e Comportamento), using the scoto-
taxis test, is described below.
In this study, adult zebrafish and goldfish were divided in four groups, depending on species (zebrafish versus goldfish)
and rearing environment (enriched versus impoverished, as described above). Animals were kept in their respective rearing
environment for 2 consecutive months before testing. One hour before testing in the dark/light preference tank, animals
were drawn randomly from the rearing tanks and transported to the behavioral observation room. An observer, blind to the
experimental condition of the experimental animal, recorded the behavioral data using the protocol described above. Data
were analyzed using two-way analyses of variance, followed by Tukey’s HSD as post-hoc tests. Results showed that being
reared in an enriched environment for 2 months increases the time that both species of fish spent in the white compartment
of the test tank, compared with those reared in an impoverished environment. Thus, we can conclude that environmental
enrichment can decrease anxiety-like behavior in this test, perhaps by reducing emotionality, as seems to be the case with
rodents73–75.
Summary
The scototaxis protocol is a novel behavioral assay of anxiety-like behavior in fish. It is easy to use, low-cost and valid
results can be obtained in a short, 15-min testing period. The patterns of results obtained using this protocol are replicable
across fish species, studies and laboratories.
ACKNOWLEDGMENTS Part of this research was supported by grants from CAPES
to C.A.G.d.M.D. and T.M.d.B. The authors thank the Dark/light Preference Team
at the defunct Laboratório de Psicobiologia e Psicopatologia Experimental from
Unesp/Bauru for support with data collection.
AUTHOR CONTRIBUTIONS C.M. and T.M.d.B. conceived and collected data for
most experiments regarding preference for dark environments in the Laboratório
de Neurociências e Comportamento and the Laboratório de Comportamento
Exploratório, and wrote this paper; C.A.G.d.M.D. conceived the experiments
regarding photoperiod and contributed to that section; A.G. and S.M. contributed
to sections regarding validity and to the theoretical background related to
scototaxis, and also wrote this paper.
Published online at http://www.natureprotocols.com.
Reprints and permissions information is available online at http://npg.nature.com/
reprintsandpermissions.
1. Bourin, M. & Hascöett, M. The mouse light/dark box test. Eur. J. Pharmacol.
463, 55–65 (2003).
2. Belzung, C. & Griebel, R. Measuring normal and pathological anxiety-like
behaviour in mice: a review. Behav. Brain Res. 138, 200–209 (2001).
3. Green, S. & Hodges, H. Animal models of anxiety. In Behavioral Models in
Psychopathology: Theoretical, Industrial and Clinical Perspectives (ed. Willner, P.)
21–49 (Cambridge University Press, Cambridge, 1991).
4. Prut, L. & Belzung, C. The open field as a paradigm to measure the effects of
drugs on anxiety-like behaviors: a review. Eur. J. Pharmacol. 463, 3–33 (2003).
5. Hascöett, M., Bourin, M. & Dhonnchadha, B.A.N. The mouse light–dark
paradigm: a review. Progr. NeuroPsychopharmacol. Biol. Psychiatr y 25,
141–166 (2001).
6. Thorndike, E.L. A note on the psychology of fishes. Am. Nat. 33, 923 (1911).
7. Satake, N. & Morton, B.E. Scotophobin A causes dark avoidance in goldfish
by elevating pineal N-acetylserotonin. Pharmacol. Biochem. Behav. 10,
449–456 (1979).
8. Serra, E.L., Medalha, C.C. & Mattioli, R. Natural preference of zebrafish (Danio
rerio) for a dark environment. Braz. J. Med. Biol. Res. 32, 1551–1553 (1999).
9. Yoshida, M., Nagamine, M. & Uematsu, K. Comparison of behavioral
responses to a novel environment between three teleosts, bluegill Lepomis
macrochirus, crucian carp Carassius langsdorfii, and goldfish Carassius auratus.
Fish. Sci. 71, 314–319 (2005).
10. Gazolla, R.A. Preference for dark substr ates in C. auratus: influence of
lighting conditions in housing environment. MSc thesis, 25 pp. (Universidade
Estadual Paulista, Bauru/SP, Brazil, 2008).
11. Gouveia, A. Jr. et al. Preference of goldfish (Carassius auratus) f or dark
places. Rev. Etol. 7, 63–66 (2005).
12. Maximino, C. et al. A comparative analysis of the preference for dark
environments in five teleosts. Int. J. Comp. Psychol. 20, 351–367 (2007).
13. Gerlai, R., Lahav, M., Guo, S. & Rosenthal, A. Drinks like a fish: zebra fish
(Danio rerio) as a behavior genetic model to study alcohol ef fects.
Pharmacol. Biochem. Behav. 67, 773–782 (2000).
14. Bjerke, S. Developing behavioral assays to study dopamine-related disorders
in zebrafish (Daniorerio). PhD thesis, 104 pp. (The University of Oslo, Oslo,
Sweden, 2002).
15. Budaev, S.V. ‘Personality’ in the guppy (Poecilia reticulata): a correlational
study of exploratory behavior and social tendency. J. Comp. Psychol. 111,
399–411 (1997).
16. McCartt, A.L., Lynch, W.E. Jr. & Johnson, D.L. How light, a predator, and
experience influence bluegill use of shade and schooling. Environ. Biol.
Fishes 49, 79–87 (1997).
17. Bleakley, B.H., Mar tell, C.M. & Brodie, E.D. III. Variation in anti-predator
behavior among five strains of inbred guppies, Poecilia reticulata.
Behav. Genet. 36, 783–791 (2006).
18. Kleerekoper, H. et al. An analysis of locomotor behaviour of goldfish
(Carassius auratus). Anim. Behav. 18, 317–330 (1970).
19. Crawshaw, L.I. Twenty-four hour records of body temperature and activity
in bluegill sunfish (Lepomis macrochirus) and brown bullheads (Ictalurus
nebulosus). Comp. Biochem. Physiol. A 51, 11–14 (1975).
NATURE PROTOCOLS | VOL.5 NO.2 | 2010 | 215
p
uo
r
G
gn
i
h
s
i
lb
uP eru
t
a
N
010
2
©natureprotocols
/
m
o
c
.
e
r
u
t
a
n
.
w
w
w
/
/
:
pt
t
h
PROTOCOL
20. Warren, E.W. & Callaghan, S. Individual differences in response to an open
field test by the guppy Poecilia reticulata (Peters). J. Fish Biol. 7, 105–113
(1976).
21. Gervai, J. & Csányi, V. Behavior-genetic analysis of the paradise fish,
Macropodus opercularis. I. Characterization of the behavioral responses of
inbred strains in novel environments: a factor analysis. Behav. Genet. 15,
503–519 (1985).
22. Mikheev, V.N. & Andreev, O.A. Two-phase exploration of a novel environment
in the guppy, Poecilia reticulata. J. Fish Biol. 42, 375–383 (1993).
23. Mok, E.Y. & Munro, A.D. Effects of dopaminergic drugs on locomotor activity
in teleost fish of the genus Oreochromis (Cichlidae): involvement of the
telencephalon. Physiol. Behav. 64, 227–234 (1998).
24. Giacomini, N.J., Rose, B., Kobayashi, K. & Guo, S. Antipsychotics produce
locomotor impairment in larval zebrafish. Neurotoxicol. Teratol. 28, 245–250
(2006).
25. Baraban, S.C., Taylor, M.R., Castro, P.A. & Baier, H. Pentylenetetrazole
induced changes in zebrafish behavior, neural activity and c-fos expression.
Neuroscience 131, 759–768 (2005).
26. Anichtchik, O.V., Kaslin, J., Peitsaro, N., Scheinin, M. & Panula, P.
Neurochemical and behavioural changes in zebrafish Danio rerio after
systemic adminis tration of 6-hydroxydopamine and 1-methyl-4-phenyl-
1,2,3,6-tetrahydropyridine. J. Neurochem. 88, 443–453 (200 4).
27. Aihart, M.J. et al. Movement disorders and neurochemical changes in
zebrafish lar vae after bath exposure to fluoxetine (PROZAC). Neurotoxicol.
Teratol. 29, 652–664 (2007).
28. Swain, H.A., Sigstad, C. & Scalzo, F.M. Effects of dizocilpine (MK-801) on
circling behavior, swimming activity, and place preference in zebrafish
(Danio rerio). Neurotoxicol. Teratol. 26, 725–729 (2004).
29. Echevarria, D.J., Hammack, C.M., Pratt, D.W. & Hosemann, J.D. A novel tes t
battery to assess global dr ug effects using the zebrafish. Int. J. Comp.
Psychol. 21, 19–34 (2008).
30. Bencan, Z. & Levin, E.D. The role of α7 and α4β2 nicotinic receptor s in the
nicotine-induced anxiolytic effect in zebrafish. Physiol. Behav. 95, 408–412
(2008).
31. López-Patiño, M.A., Yu, L., Cabral, H. & Zhdanova, I.V. Anxiogenic effects
of cocaine withdrawal in zebrafish. Physiol. Behav. 93, 160–171 (2008).
32. Levin, E.D., Bencan, Z. & Cerutti, D.T. Anxiolytic effects of nicotine in
zebrafish. Physiol. Behav. 90, 54–58 (2007).
33. Wilson, D.S., Coleman, K., Clark, A.B. & Biederman, L. Shy-bold continuum
in pumpkinseed sunfish (Lepomis gibbosus): an ecological study of a
psychological trait. J. Comp. Psychol. 107, 250–260 (1993).
34. Budaev, S.V. Alternative styles in the European wrasse, Symphodus ocellatus:
boldness-related schooling tendency. Environ. Biol. Fishes 49, 71–78 (1997).
35. Budaev, S.V. How many dimensions are needed to describe temperament in
animals: a factor reanalysis of two data se ts. Int. J. Comp. Psychol. 11, 17–29
(1998).
36. Brown, C., Jones, F. & Braithwaite, V. In situ examination of boldness–
shyness traits in the tropical poeciliid, Brachyraphis episcopi. Anim. Behav.
70, 1003–1009 (2005).
37. Moretz, J.A., Martins, E.P. & Robison, B.D. Behavioral syndromes and the
evolution of correlated behavior in zebrafish. Behav. Ecol. 18, 556–562
(2007).
38. Montgomery, K.C. The relation between fear induced by novel stimulation
and exploratory behavior. J. Comp. Physiol. Psychol. 48, 254–260 (1955).
39. Craig, W. Appetites and aversions as constituents of instincts. Proc. Natl.
Acad. Sci. USA 3, 685–688 (1917).
40. McNaughton, N. & Corr, P.J. A two-dimensional neuropsychology of defense:
fear/anxiety and defensive distance. Neurosci. Biobehav. Rev. 28, 285–305
(2004).
41. Rodgers, R.J., Cao, B-J. & Holmes, A. Animal models of anxiety: an
ethological perspective. Braz. J. Med. Biol. Res. 30, 289–304 (1997).
42. Rodgers, R.J. Animal models of ‘anxiety’: where next? Behav. Pharmacol. 8,
477–496 (1997).
43. Risner, M.L., Lemerise, E., Vukmanic, E.V. & Moore, A. Behavioral spectral
sensitivity of the zebrafish (Danio rerio). Vision Res. 46, 2625–2635 (2006).
44. Yager, D. Behavioural measures of t he spectral sensitivity of dark-adapted
goldfish. Nature 220, 1052–1053 (1968).
45. Fuiman, L.A. & Magurran, A.E. Development of predator defenses in fishes.
Rev. Fish Biol. Fish. 4, 145–183 (1994).
46. Peitsaro, N., Kaslin, J., Anichtchik, O.V. & Panula, P. Modulation of the
histaminergic system and behaviour by a-fluoromethylhistidine in zebrafish.
J. Neurochem. 86, 432–4 41 (2003).
47. Lockwood, B., Bjerke, S., Kobayashi, K. & Guo, S. Acute effects of alcohol on
larval zebrafish: a genetic system for large-scale screening. Pharmacol.
Biochem. Behav. 77, 647–654 (2004).
48. Webb, P.W. Body for m, locomotion and for aging in aquatic ver tebrates.
Am. Zool. 24, 107–120 (1984).
49. Altimiras, J. & Larsen, E. Non-invasive recording of hear t rate and
ventilation rate in rainbow trout during rest and swimming. Fish go
wireless! J. Fish Biol. 57, 197–209 (2000).
50. Barreto, R.E. & Volpato, G. Caution for using ventilator y frequency as an
indicator of stress in fish. Behav. Processes 66, 43–51 (2004).
51. Barreto, R.E., Luchiari, A.C. & Marcondes, A.L. Ventilatory frequency
indicates visual recognition of an allopatric predator in naïve Nile tilapia.
Behav. Processes 60, 235–239 (2003).
52. Sager, D.R., Hocutt, C.H. & St aufer, J.R. Jr. Base and stressed ventilation
rates for Leiostomus xanthurus Lacépède and Morone americana Gmelin
exposed to strobe lights. J. Appl. Ichtyol. 16, 89–97 (2000).
53. Fujii, R. The regulation of motile activity in fish chromatophores. Pigment Cell
Res. 13, 300–319 (2000).
54. Kawauchi, H., Kawazoe, I., Tsubokawa, M., Kishida, M. & Baker, B.I.
Characterization of melanin-concentrating hormone in chum salmon
pituitaries. Nature 305, 321–323 (1983).
55. Hoglund, E., Balm, P.H. & Winberg, S. Skin darkening, a potential social
signal in subordinate arctic charr (Salvelinus alpinus): the regulatory role of
brain monoamines and pro-opiomelanocortin-derived peptides. J. Exp. Biol.
203, 1711–1721 (2000).
56. Gottesman, I.I. & Gould, T.D. The endophenotype concept in psychiatr y:
etymology and strategic intentions. Am. J. Psychiatry 160, 636–645 (2003).
57. Gould, T.D. & Gottesman, I.I. Psychiatr ic endophenotypes and the
development of valid animal models. Genes Brain Behav. 5, 113–119 (2006).
58. Bakshi, V.P. & Kalin, N.H. Animal models and endophenotypes of anxiety and
stress disorders. In Neuropsychopharmacology: The Fif th Generation of
Progress (eds. Davis, K.L., Charney, D., Coyle, J.T. & Nemerof f, C.) 885–900
(American College of Neuropsychopharmacology, Nashville, Tennessee,
2002).
59. Messick, S. Validity of psychological assessment: validation of inferences
from persons’ responses and perfor mances as scientific inquir y into score
meaning. Am. Psychol. 50, 741–749 (1995).
60. Trout, J.D. Measurement. In A Companion to the Philosophy of Science
(ed. Newton-Smith, W. H.) 265–276 (Blackwell Publishing, Oxford, 1999).
61. Li, H. The resolution of some paradoxes related to reliability and validity.
J. Edu. Behav. Stat. 28, 89–95 (2003).
62. Walf, A.A. & Frye, C.A. The use of the elevated plus maze as an assay of
anxiety-related behavior in rodents. Nat. Protoc. 3, 322–328 (2007).
63. Willner, P. Behavioural models in psychopathology. In Behavioural Models
in Psychopathology: Theoretical, Industrial and Clinical Perspectives
(ed. Willner, P.) 3–18 (Cambr idge University Press, Cambr idge, 1991).
64. Segal, D.S. & Geyer, M.A. Animal models of psychopathology. In
Psychobiological Foundations of Clinical Psychiatry (eds. Judd, L.L. & Groves,
P.M.) 1–14 (J.B. Lippincott, Philadelphia, Pennsylvania, 1985).
65. Matthews, K. & Reid, I. Animal models for depression: the anhedonic rat —
theory and practice. In New Models for Depression (eds. Eber t, D. & Ebmeier,
K.P.) 49–71 (Karger, Basel, 1998).
66. Guo, S. Linking genes to brain, behavior and neurological diseases: what
can we learn from zebrafish? Genes Brain Behav. 3, 63–74 (2000).
67. Kavaliers, M. & Choler is, E. Antipredator responses and defensive behavior:
ecological and ethological approaches for the neurosciences. Neurosci.
Biobehav. Rev. 25, 577–586 (2001).
68. Gray, J.A. & McNaughton, N. Neuropsychology of Anxiety: An Enquiry into the
Functions of the Septo-hippocampal System (Oxford University Press, Oxford,
2000).
69. Toth, M. & Zupan, B. Neurobiology of anxiety. In Handbook of Contemporary
Neuropharmacology Vol. 2 (eds. Sibley, D.R., Hanin, I., Kuhar, M. & Skolnick, P.)
3–58 (John Wiley & Sons, Hoboken, New Jersey, 2007).
70. Abrams, P.A. Should prey overestimate the r isk of predation? Am. Nat. 144,
317–328 (1994).
71. Sih, A. Prey uncertainty and the balancing of antipredator and feeding
needs. Am. Nat. 139, 1052–1069 (1990).
72. Lima, S.L. & Bednekoff, P.A. Temporal variation in danger dr ives antipredator
behavior: the predation risk hypothesis. Am. Nat. 153, 649–659 (1999).
73. Prior, H. & Sachser, N. Effects of enriched housing environment on the
behaviour of young male and female mice in four exploratory tasks.
J. Exp. Anim. Sci. 37, 57–68 (1994).
74. Chapillon, P., Mannechpe, C., Belzung, C. & Cas ton, J. Rearing environmental
enrichment in two inbred strains of mice: 1. Effects on emotional reactivity.
Behav. Genet. 29, 41–46 (1999).
75. Roy, V., Belzung, C., Delarue, C. & Chapillon, P. Environmental enrichment in
BALB/c mice: effects in classical tests of anxie ty and exposure to a predatory
odor. Physiol. Behav. 74, 313–320 (2001).
216 | VOL.5 NO.2 | 2010 | NATURE PROTOCOLS
p
uo
r
G
gn
i
h
s
i
lb
uP eru
t
a
N
010
2
©natureprotocols
/
m
o
c
.
e
r
u
t
a
n
.
w
w
w
/
/
:
pt
t
h
PROTOCOL
76. Bertoglio, L.J. & Carobrez, A.P. Previous maze experience required to
increase open arms avoidance in rats submitted to the elevated plus-maze
model of anxiety. Behav. Brain Res. 108, 197–203 (2000).
77. File, S.E. One trial tolerance to the anxiolytic ef fects of chlordiazepoxide in
the plus-maze. Psychopharmacology 100, 281–282 (1990).
78. File, S.E. & Zangrossi, H. Jr. ‘One trial toler ance’ to the anxiolytic actions of
benzodiazepine in the elevated plus-maze, or the development of a phobic
state? Psychopharmacology 110, 240–244 (1993).
79. Rodgers, R.J. & Shepherd, J.K. Influence of prior maze experience on
behaviour and responses to diazepam in the elevated plus-maze in male mice
depends upon treatment regimen and pr ior maze exper ience.
Psychopharmacology 106, 102–110 (1993).
80. Holmes, A. & Rodgers, R.J. Responses of Swiss-Webster mice to repeated
plus-maze exper ience: further evidence for qualitative shift in emotional
state? Pharmacol. Biochem. Behav. 60, 473–488 (1998).
81. Maximino, C. et al. Reliability of dark preference in zebr afish: test-retest
correlations, dif fering black:white proportions of the apparatus, and inter-
laboratory replicability. Behav. Processes (in press).
82. Miklósi, Á. & Andews, R.J. The zebrafish as a model for behavioral studies.
Zebrafish 3, 227–234 (2006).
83. Langsdale, J.R.M. Development al changes in the opacity of lar val herring,
Clupea harengus, and their implications for vulnerability to predation.
J. Mar. Biol. Assays 73, 225–232 (1993).
84. McClure, M. Development and evolution of melanophore patterns in
fishes of the genus Danio (Teleostei: Cyprinidae). J. Morphol. 241, 83–105
(1999).
85. Watkins, J., Miklósi, Á. & Andrew, R.J. Early asymmetries in the behavior of
zebrafish lar vae. Behav. Brain Res. 151, 177–183 (2004).
86. Walsh, R.N. & Cummins, R.A. The open-field test: a critical review.
Psychol. Bull. 83, 482–504 (1976).
87. Hurd, M.W., Debruyne, J., Straume, M. & Cahill, G.M. Circadian rhythms of
locomotor activity in zebrafish. Physiol. Behav. 65, 465–472 (1998).
88. Clark, C.W. & Levy, D.A. Diel vertical migrations by juvenile sockeye salmon
and the antipredation window. Am. Nat. 131, 271–290 (1988).
89. Levy, D.A. Sensor y mechanisms and selective advantage for diel vertical
migration in juvenile sockeye salmon, Oncorhynchus nerka. Can. J. Fish.
Aquat. Sci. 47, 1796–1802 (1990).
90. Neilson, J.D. & Perry, R.J. Diel vertical migrations of marine fishes: an
obligate or facultative process? Adv. Mar. Biol. 26, 115–167 (1990).
91. Wedemeyer, G.A., Barton, B.A. & McLeay, D.J. Stress and acclimation. In
Methods for Fish Biology (eds. Schreck, C.B. & Moyle, P.B.) 451–489
(American Fisheries Society, Bethesda, Maryland, 1990).
92. Barton, B.A. Salmonid fishes dif fer in their cortisol and glucose responses
to handling and transport stress. North Am. J. Aquacult. 62, 12–18
(2000).
93. Barton, B.A. Physiological and condition-related indicators of environmental
stress in fish. In Biological Indicators of Aquatic Ecosystem Health (ed. Adams,
S.M.) 111–148 (American F isheries Society, Bethesda, Maryland, 2002).
94. Frisch, A.J. & Anderson, T.A. The response of coral trout (Plectropomus
leopardus) to capture, handling, transpor t and shallow water stress. Fish
Physiol. Biochem. 23, 23–34 (2000).
95. Kuwada, H. et al. Effect of fish size, handling stresses and training procedure
on the swimming behavior of hatchery-reared striped jack: implications for
stock enhancement. Aquaculture 185, 245–256 (2000).
96. van Raaij, M.T.M., Pit, D.S.S., Balm, P.H., Steffens, A.B. & van der Thillart,
G.E.E. Behavioral strategy and the physiological stress response in rainbow
trout exposed to severe hypoxia. Horm. Behav. 30, 85–92 (1996).
97. Façanha, M.F. & Gomes, L.d.C. Efficacy of menthol as an anesthetic for
tambaqui (Colossoma macropomum, Characiformes: Characidae). Acta
Amazon. 35, 71–75 (2005).
98. Keene, J.L., Noakes, D.L.G., Moccia, R.D. & Soto, C.G. The efficacy of clove
oil as an anaesthetic for rainbow trout, Oncorhynchus mykiss (Walbaum).
Aquacult. Res. 29, 89–101 (1998).
99. Holloway, A.C., Keene, J.L., Noakes, D.L.G. & Moccia, R.D. Effects of clove
oil and MS-222 on blood hormone profiles in rainbow trout Oncorhynchus
mykiss, Walbaum. Aquacult. Res 35, 1025–1030 (2004).
100. Westerfield, M. The Zebrafish Book: A Guide for the Laboratory Use of
Zebrafish Danio (Brachydanio) rerio. (University of Oregon Press, Eugene,
Oregon, 2000).
101. Spence, R., Gerlach, G., Lawrence, C. & Smith, C. The behaviour and ecology
of the zebrafish, Danio rerio. Biol. Rev. 83, 13–34 (2008).
102. Dunbar, R.I.M. Some aspects of research design and their implications in
the obser vational study of behaviour. Behaviour 58, 78–98 (1976).
103. Noakes, D.L.G. & Baylis, J.R. Behavior. In Methods for Fish Biology
(eds. Schreck, C. B. & Moyle, P. B.) 555–574 (American Fisher ies Society,
Bethesda, Mar yland, 1990).
104. Volpato, G. Consider ações metodológicas sobre os testes de preferência na
avaliação do bem-estar em peixes [Methodological considerations about
preference tests for the evaluation of well-being in fish]. Rev. Brasil.
Zootecnia 36, 53–61 (2007).
105. Faganello, F.R. & Mattioli, R. Anxiolytic-like ef fect of chlorpheniramine
in inhibitor y avoidance in goldfish submitted to telencephalic ablation.
Progr. Neuropsychopharmacol. Biol. Psychiatr y 31, 269–274 (2007).
106. Ninkovic, J. & Bally-Cuif, L. The zebrafish as a model system for assessing
the reinforcing properties of dr ugs of abuse. Methods 39, 262–274 (2006).
107. Grunwald, D.J. & Eisen, J.S. Headwaters of the zebrafish—emergence of a
new model vertebrate. Nat. Rev. Genet. 3, 717–724 (2002).
108. Johansen, R., Needham, J.R., Colquhoun, D.J., Poppe, T.T. & Smith, A.J.
Guidelines for health and welfare monitoring of fish used in research.
Lab. Anim. 40, 323–340 (2006).
109. American Society of Ichthyologists and Herpetologists, American Fisheries
Society & American Institute of Fisheries Research Biologists. Guidelines for
use of fishes in field research. Fisheries 13, 16–23 (1988).
110. Zhdanova, I.V., Wang, S.Y., Leclair, O.U. & Danilova, N.P. Melatonin
promotes sleep-like state in zebrafish. Brain Res. 903, 263–268 (2001).
111. Magalhães, D.d.P., Cunha, R.A.d., Santos, J.A.A.d., Buss, D.F. & Baptista,
D.F. Behavioral response of zebrafish Danio rerio Hamilton 1822 to
sublethal stress by sodium hypochlorite: ecotoxicological assay using an
image analysis biomonitoring system. Ecotoxicology 16, 417–422 (2007).
112. Covey, T.R., Lee, E.D. & Henion, J.D. High-speed liquid chromatography/
tandem mass spectrometry for the determination of drugs in biological
samples. Analyt. Chem. 58, 2453–2460 (1986).
113. Bloom, H.D. & Perlmutter, A. A sexual aggregating pheromone system in
the zebrafish, Brachydanio rerio. J. Exp. Zool. 199, 215–226 (1977).
114. Breder, C.M. Jr. & Halpern, F. Innate and acquired behavior affecting the
aggregation of fishes. Physiol. Zool. 19, 154–190 (1946).
115. Maximino, C. et al. Tank enrichment alters explorator y behavior of zebrafish
and goldfish. J. Exp. Anim. Sci. (in press).
