A

GD8

a

ARRAA

KETAAVM

1

wEMie(hRmbE

oogrtAt

h0

Behavioural Processes 107 (2014) 106–111

Contents lists available at ScienceDirect

Behavioural Processes

jo ur nal home p ag e: www.elsev ier .com/ locate /behavproc

version in the elevated plus-maze: Role of visual and tactile cues

uilherme Bracarense Filgueiras, Eduardo F. Carvalho-Netto, Celio Estanislau ∗

epartamento de Psicologia Geral e Análise do Comportamento, Centro de Ciências Biológicas, Universidade Estadual de Londrina, PR 445, Km 380,6051-990 Londrina, PR, Brazil

r t i c l e i n f o

rticle history:eceived 11 November 2013eceived in revised form 12 August 2014ccepted 13 August 2014vailable online 23 August 2014

eywords:levated plus-maze

a b s t r a c t

Thigmotaxis, a tendency to be close to vertical surfaces, leads rats to avoid open arms in the elevated plus-maze. Evidences support a role in thigmotaxis for the vibrissal sense as well as for vision. In this study,sensory inputs for both senses were manipulated in order to identify which of them mainly contributesto thigmotaxis. This was achieved by manipulating the length of rats’ mystacial vibrissae, the presence ofwalls in the “open” arms and their transparency. As expected, rats avoided arms which lacked walls. Onthe other hand, rats did not avoid “open” arms surrounded by transparent walls as one could expect ifthey were using mainly vision while exploring the maze. Furthermore, these “open” arms were explored

higmotaxisnxietyversionisionystacial vibrissae

similarly to arms surrounded by opaque walls. Acute vibrissotomy resulted in minor effects in rats testedin a conventional elevated plus-maze. These findings suggest that vibrissotomized rats seemed to beable to compensate the absence of mystacial vibrissae by means of other sensory pathways (tactile ornon-tactile) and by adjusting some exploratory aspects. Thus, the current results indicate that rats relymore on other sensory cues than on vision in avoiding open arms in the elevated plus-maze.

© 2014 Elsevier B.V. All rights reserved.

. Introduction

The elevated plus-maze (EPM) is currently one of the mostidely used animal models of anxiety. While being tested in an

PM, rodents show marked open arms avoidance (Handley andithani, 1984; Pellow et al., 1985). Anxiolytic or anxiogenic drugs

ncrease or decrease, respectively, open arms exploration (Pellowt al., 1985). Aside from the conventional spatiotemporal measuresi.e., entries into and time spent in the arms), ethological measuresave been indicated as reliable anxiety indexes (Cruz et al., 1994;odgers and Johnson, 1995). For instance, measures of risk assess-ent (primarily, stretched-attend postures – SAP) have proved to

e highly valuable in identifying anti- and proanxiety drugs in thePM (Rodgers and Cole, 1994).

It has been well established that anxiety-like behaviorsbserved in the EPM are triggered mainly by aversion of rodents topen spaces. A crucial motive for open arm avoidance has been sug-ested to be the impossibility of thigmotaxis––a natural defensiveesponse in which rats remain close to vertical surfaces––rather

han, for example, fear of heights or novelty (Treit et al., 1993).ccording to Grossen and Kelly (1972), rats naturally do prefer

o be close to vertical surfaces in order to avoid avian predation.

∗ Corresponding author. Tel.: +55 43 3371 4261; fax: +55 43 3371 4227.E-mail addresses: celio.estanislau@gmail.com, estanislau@uel.br (C. Estanislau).

ttp://dx.doi.org/10.1016/j.beproc.2014.08.005376-6357/© 2014 Elsevier B.V. All rights reserved.

Whereas thigmotaxis is an important determinant of rat behaviorin the EPM, it is still not well established the sensorial modalityby which rodents perceive the openness of an environment. Mostof the senses can possibly contribute to openness perception (forexample, the rat can use odors, air flow or sound reflection cues fornavigation – see, for example Wallace et al., 2002). However, stud-ies about sensorial modulation of open space perception in rodentsare concentrated in two types of inputs, those from the vibrissalsense and from vision.

There are suggestions that the vibrissae play a key role inthigmotaxis and other defensive behaviors. For example, besidesopenness, rats also avoid cliffs and, to evaluate depth, they seemto rely more on the vibrissal sense than on vision (Shiffman et al.,1970). Furthermore, there is evidence that animals submitted tounilateral vibrissotomy exhibit directional thigmotaxis asymmetrytoward the intact vibrissae side while navigating in aquatic or ter-restrial environments (Meyer and Meyer, 1992). In the EPM test,Belzung (1999) reports no open arm avoidance by mice withoutvibrissae.

On the other hand, evidences from studies which manipulatedthe amounts of light during the test session and the presence oftransparent vertical surfaces in the open arms of the EPM sug-

gest that vision is important for rodent perception of open spaces(Cardenas et al., 2001; Martinez et al., 2002). For example, ratsexplore more the open arms under complete darkness or dim light(01 lux) than under higher illuminations (Garcia et al., 2005, 2011).

ural Processes 107 (2014) 106–111 107

Ib(pov

ttesbtotmotstidc

2

2

fhktmto

2

BtoaU

2

(

Table 1Illumination (lux) in each part of each type of maze.

Location Conventional Transparent Opaque

“Open” arms 66.7/66.7 71.0/73.2 42.0/42.0

G.B. Filgueiras et al. / Behavio

n addition, a study shows that the larger the gradient of luminosityetween open and closed arms, the lower is open arm explorationPereira et al., 2005). When the open arms are provided with trans-arent walls, maze exploration is reported to be similar to thebserved in a regular EPM (Martinez et al., 2002), suggesting thatisual cues prevail over tactile ones.

Many previous studies provided evidences about the con-ributions from the vibrissae and from vision as mediators ofhigmotaxis in the EPM. Indeed, no available evidence allows thexclusion of one of these two possible mediators. In this sense, atudy in which both senses are evaluated in the same test can possi-ly be heuristic. Therefore, the current investigation was designedo test these two senses by manipulating vibrissae length (intactr cut), the presence of vertical surfaces (present or absent) andheir transparency (opaque or transparent). Thus, if vision is the

ain sensory modality for thigmotaxis, similar exploration wouldccur in conventional open arms or in “open” arms surrounded byransparent walls – vibrissotomy would, at best, strengthen thatimilarity. On the other hand, if rats rely more on vibrissal sense,hey would explore transparent walled “open” arms as if they weren regular enclosed arms – vibrissotomy would disturb explorationepending on the degree to which the sensory deficit could beompensated by other sensory inputs.

. Materials and methods

.1. Subjects

Subjects were 71 male Wistar rats weighing 250–300 g, obtainedrom Londrina State University/UEL (Paraná, Brazil) which wereoused in groups of 5 per cage (40 cm × 34 cm × 17 cm). They wereept under regular 12:12 h light cycle (lights on at 7:00 a.m.) in aemperature (25 ± 1 ◦C) and humidity (55 ± 5%) controlled environ-

ent. Food and drinking water were freely available, except duringhe brief test periods. All rats were experimentally naïve and usednly once.

.2. Ethics

Experiments were carried out in accordance with the norms ofrazilian Neuroscience and Behavior Society, which are based onhe US National Institutes of Health Guide for Care and Use of Lab-ratory Animals. All experimental procedures had been previouslypproved by the Committee for Animal Research of Londrina Stateniversity (CEEA/UEL 11/09).

.3. Apparatus

Three types of plus-mazes were used in the experiments (Fig. 1):

A) Conventional maze (CM): a standard elevated plus-maze whichconsisted of two open arms (50 cm × 12 cm × 1 cm) at rightangles with two closed arms (50 cm × 12 cm × 40 cm) con-nected to a central platform (12 cm × 12 cm). The apparatus wasconstructed from varnished wood (floor) and transparent Plex-iglas (walls) and was raised to a height of 50 cm above floorlevel. To prevent the rats from falling, a rim of Plexiglas (1 cmhigh) surrounded the perimeter of the open arms. The walls inthe closed arms were covered with brown paper in their outersurface.

The other two plus-mazes were similarly constructed, but

they were different in the open arms construction. Actually, forthese mazes, the word “open” was used only for labeling theposition of the arms (in reference to the closed arms and to theroom).

Closed arms 44.1/44.1 43.0/43.0 42.0/44.1Central square 58.1 58.1 45.2

(B) Transparent walls maze (TW): the “open” arms were surroundedby 40-cm high transparent Plexiglas walls.

(C) Opaque walls maze (OW): similar to the TW, however with theouter Plexiglas walls surface of the “open” arms covered withbrown paper.

2.4. Procedure

All experimental procedures were conducted between 01 p.m.and 04 p.m. under the illumination of a 60-W bulb light fixed 2.4 mabove the apparatus (Table 1 shows illumination levels in eachmaze).

Seventy-one rats were randomly allocated to three groups(n = 23–24); one group to be tested in the CM, another one in theTW maze and the third group in the OW maze (in each testing day,one maze was replaced by another in the same testing room). For allthese groups, 5 min before testing, each animal was subjected to avibrissae removal procedure. Cut group: whole extension of mysta-cial vibrissae was cut with a scissor (no more than 1 mm remained)(n = 11–12) and Intact group: mystacial vibrissae were sham cut(i.e., vibrissae were touched with the side of a scissor and then itwas closed in order to produce the same sound) (n = 12). The vibris-sae (or the sham) removal procedure duration was about 30 s. Fortesting, each rat was carefully placed in the center of the CM, TWor OW maze, facing one of the closed arms. All procedures wereperformed by the same experimenter. Test sessions lasted 5 minand the apparatus was cleaned with 5% ethanol and dried withpaper towels in between trials. All sessions were video-recordedby a camera fixed above the apparatus.

2.5. Behavioral scoring

Behavioral scoring was performed by a trained observer(intra-observer concordance: >90%). The observer, while sco-ring, was unavoidably aware of the type of maze, but wasnot aware about the vibrissae condition (video quality didnot allowed such identification). An ethological scoring soft-ware package developed by Dr. Morato’s group at Faculdade deFilosofia, Ciências e Letras de Ribeirão Preto, USP, Brazil (retrievedfrom http://scotty.ffclrp.usp.br/page.php?6) was used. Behavioralparameters comprised conventional spatiotemporal and etholog-ical measures which are regarded as important in the elevatedplus-maze (see, for example, Cruz et al., 1994; Rodgers and Johnson,1995) and their latencies. Entries into and time spent in the closedand “open” arms were recorded. Total distance traveled in each typeof arm by the animals was estimated from the number of crossings(the maze arms were divided into 10-cm squares on a transparentplastic mask placed over the TV set screen). An entry into an arm (orinto a square within an arm) was scored after all four paws enteredit. The frequencies of the following behaviors were recorded: Rear-ing (animal rising on the hind limbs both touching and not touchinga wall surface), Stretched Attend Posture (SAP) (when the animalstretches its body forward and then retracts to the original positionwith no locomotion) and Head Dipping (animal sticking the head

outside the maze border and toward the floor). Head dippings andSAPs are often assigned (see, for example, Rodgers and Johnson,1995), depending on the part of the maze where they occur, asunprotected (in the open arms) or protected (in the center or the

108 G.B. Filgueiras et al. / Behavioural Processes 107 (2014) 106–111

F ndingw ar to tb

ctbat

2

ap(umbta

u

csawToBtihbhSgw

3

duP(st(

The frequency of SAPs in the open arms is shown in panel Aof Fig. 3. The ANOVA showed a significant main effect of the fac-tor maze (Table 2). Indeed, in sharp contrast with the observed

Table 2Summary of the two-way ANOVAs with the factor type of maze and vibrissae length.

Measure Maze Vibrissae Maze × vibrissae

df(2,65) df(1,65) df(2,65)

OA entries (%)F 6.947 0.227 0.487p <0.01 ns ns

OA time (%)F 13.155 0.044 0.278p <0.001 ns ns

OA distance (%)F 15.836 0.172 0.908p <0.001 ns ns

CA entriesF 1.065 0.540 0.537p ns ns ns

Center time (s)F 2.033 0.035 0.139p ns ns ns

OA SAPsF 7.164 0.188 0.403p <0.01 ns ns

Center SAPsF 16.984 4.519 4.476p <0.001 <0.05 <0.05

CA SAPsF 25.279 0.212 0.117p <0.001 ns ns

OA rearingsF 60.469 0.006 0.145p <0.001 ns nsF 12.751 3.403 2.219

ig. 1. Plus-mazes used in the study. They differed in respect to the open arms surroualls (TW, 40-cm high transparent Plexiglas walls) or (C) opaque walls (OW, simil

rown paper).

losed arms). Taking into account that in this experiment there arehree maze configurations and that in the OW maze no area coulde called “unprotected”, a strictly descriptive approach was appliednd the behaviors were separately recorded as occurring either inhe center, closed or “open” arms.

.6. Data analysis

Entries into and time spent in the “open” arms, as wells the distance traveled in these arms, were converted inercentage, which was calculated by means of the formula:“open”/“open” + closed) × 100. These percentage values were eval-ated regarding their difference from the 50% chance level byeans of t tests. The studied behavioral measures were compared

y means of two-way analyses of variance (ANOVA) with the factorsype of maze (Conventional × Transparent walls × Opaque walls)nd vibrissae (Intact × Cut).

Whenever necessary, the Newman–Keuls post hoc test wassed.

We also evaluated if behavioral effects were present during spe-ific parts of the session. Preliminary minute by minute analyseshowed, within the 5-min session, similar values between the 1stnd 2nd minutes. Similarities between the 4th and 5th minutesere also found. Third minute scores showed intermediate values.

his temporal pattern was similar to the one described in previ-us works (see Rosa et al., 2000 and, for a review, Carobrez andertoglio, 2005). Accordingly, in order to simplify and present syn-hetic results, instead of five 60-s periods, the session was dividedn two 150-s blocks. That split session approach only proved to beeuristic in respect to our objectives when applied to head dippingehavior. Obviously, this behavior is only possible in the CM. Asead dipping data did not meet normality criteria according to thehapiro–Wilk test, Mann–Whitney U-test was used for betweenroups comparisons and Wilcoxon signed rank test was used forithin group (time blocks) comparisons.

In all cases, the significance level was set at p < 0.05.

. Results

A summary of the two-way ANOVAs is shown in Table 2.Fig. 2 shows the percentages of entries into, time spent and

istance run in the “open arms”. For all these behavioral meas-res, significant main effects of the type of maze were found.ost hoc comparisons (Newman–Keuls) showed that both groups

intact and cut vibrissae) tested in the conventional maze (CM)pent less time and traveled shorter distances in the open armshan their corresponding groups tested in the transparent wallsTW) or in the opaque walls (OW). The percentage of entries in the

s: (A) conventional maze (CM, 1-cm high transparent Plexiglas rim), (B) transparenthe transparent walls maze, however with the outer Plexiglas surface covered with

open arms showed similar results; however, differences among theintact vibrissae groups did not reach the significance level. Thesethree percentage indexes were found as significantly below the 50%chance level only for the intact (t[11] < −3.01, p < 0.05) and the cutvibrissae (t[10] < −2.90, p < 0.05) groups tested in the conventionalmaze, but not for those tested in the other mazes.

Table 3 shows the frequency of open arms rearings. A signifi-cant main effect of the type of maze was found (Table 2). Post hoccomparisons showed that the cut and intact vibrissae groups testedin the CM displayed less rearings than their corresponding groupstested in the other mazes. Decreases in rearing in the CM could beexpected due to the absence of vertical surfaces. It worth to note thesimilarity between the levels observed in the TW and OW “open”arms.

Table 3 also shows the frequency of rearing in the centralsquare. ANOVA indicated a significant main effect of the factor maze(Table 2). Post hoc comparisons showed that the intact vibrissaegroup tested in the CM displayed less rearings than its correspond-ing groups tested in the other mazes. The frequency of closed armsrearings is also shown in Table 3. No significant effects were found.

Center rearingsp <0.001 ns ns

CA rearingsF 2.737 1.182 0.257p ns ns ns

OA, open arms; CA, closed arms; ns, non significant.

G.B. Filgueiras et al. / Behavioural Processes 107 (2014) 106–111 109

Table 3Scores of complementary behavioral measures.

Behavioral measure Conventional Transparent walls Opaque walls

Intact Cut Intact Cut Intact Cut

CA entries 8.8 ± 0.6 10.4 ± 1.1 11.1 ± 0.9 10.8 ± 0.7 10.3 ± 0.9 10.7 ± 1.3Center time (s) 51.3 ± 6.5 51.0 ± 8.5 43.7 ± 3.0 45.5 ± 3.8 41.5 ± 7.4 37.3 ± 3.6OA rearings 1.0 ± 0.3 0.5 ± 0.3 11.4 ± 1.1* 12.0 ± 1.8* 10.6 ± 1.1* 10.3 ± 1.1*

Center rearings 0.7 ± 0.2 1.2 ± 0.2 4.5 ± 0.8 2.8 ± 0.6 3.8 ± 0.7 2.5 ± 0.5CA rearings 15.2 ± 1.2 16.9 ± 1.5 14.3 ± 1.2 14.3 ± 1.4 12.3 ± 1.3 13.8 ± 1.2

ONewm

imc

ssceovpt

F“ptt(tl

A, open arms; CA, closed arms.* p < 0.05 as compared to the corresponding group tested in conventional maze (

n the CM, this behavior was nearly absent in the TW and OWazes. However, no significant differences were found in post hoc

omparisons.Panel B of Fig. 3 shows the frequency of center SAPs. ANOVA

howed main effects of the maze and vibrissae factors, and aignificant interaction between these factors (Table 2). Post hocomparisons showed that the cut vibrissae group tested in the CMxhibited more SAPs than its corresponding groups tested in thether mazes. This group also displayed more SAPs than the intact

ibrissae group tested in the same type of maze. Panel C of Fig. 3resents SAPs in the closed arms. A significant main effect of theype of maze was found (Table 2). Post hoc comparisons showed

ig. 2. Percentages of entries into (A), time spent in (B) and distance run in (C) theopen” arms (mean ± SEM). Intact and cut vibrissae rats were tested in one of threelus-mazes which differed in respect to the “open” arms surroundings (conven-ional, transparent walls or opaque walls). Numbers inside the columns correspondo mean raw values of the frequency of entries, time spent in (s) and distance runm) in the open arms. *p < 0.05 as compared to the corresponding group tested inhe conventional maze (Newman–Keuls). #p < 0.05 as compared to the 50% chanceevel (t-test).

an–Keuls).

that rats tested in the CM, regardless of the vibrissae length, dis-played more SAPs than the rats tested in the other mazes, whichvirtually did not show this behavior in the closed arms.

Table 4 shows head dipping results. Between groups compar-isons revealed no significant differences in this behavior. Withingroups comparisons revealed that the intact vibrissae groupshowed decreases in open arm head dipping during the session(Wilcoxon signed rank test). The cut vibrissae group also showeddecreases in head dipping from the center during the session(Wilcoxon signed rank test).

The frequencies of entries into the closed arm and the time spentin the central square are shown in the Table 3. No significant effectsor interactions were found. Latencies for all the studied behavioralmeasures also did not show significant effects (data not shown).

Fig. 3. SAPs performed in the open arms (A), the center square (B) and in the closedarms (C) (mean ± SEM). Intact and cut vibrissae rats were tested in one of three plus-mazes which differed in respect to the “open” arms surroundings (conventional,transparent walls or opaque walls). *p < 0.05 as compared to the correspondinggroup tested in the conventional maze. ap < 0.05 as compared to the correspondingintact vibrissae group (Newman–Keuls).

110 G.B. Filgueiras et al. / Behavioural Pr

Table 4Head dippings in different moments of the conventional elevated plus-maze session.

Moment Intact Cut

OA headdippings

1st half 5.3 ± 1.2 3.5 ± 1.12nd half 0.7 ± 0.3 2.4 ± 1.0

Center headdippings

1st half 2.2 ± 0.5 2.3 ± 0.72nd half 1.3 ± 0.4 0.4 ± 0.2

CA headdippings

1st half 0.3 ± 0.2 0.3 ± 0.12nd half 0.3 ± 0.1 0.3 ± 0.1

O

4

attmiawawwtidwt

pRtortiotB

tir“itottaoatAaawcsbswtr

Likewise, the mentioned differences in exploration shown by vib-

A, open arms; CA, closed arms.

. Discussion

The current study investigated the roles of mystacial vibrissaend vision on exploratory behavior in the EPM. In order to evaluatehe roles played by vision and vibrissal sense, the vibrissal length,he presence of walls in the open arms and their transparency were

anipulated. The main results found were as follows. Rats testedn a conventional maze showed decreased open arm exploration,s compared with those tested in a maze in which the “open” armsere surrounded by opaque walls. On the other hand, rats tested in

maze in which the “open” arms were surrounded by transparentalls explored these arms similarly to those tested in the opaquealls maze. The cut vibrissae groups were found to show roughly

he same amount of “open” arms exploration as their correspondingntact vibrissae groups. However, in exploring a conventional EPM,ifferences in the time course of some behaviors along the sessionere found between these groups. These results are discussed in

he paragraphs below.The rats tested in the conventional maze were found to dis-

lay behaviors that can be understood as representative of anxiety.ates of open arms exploration (percentage of entries, time, dis-ance run, and rearing) indicate a preference for the closed over thepen arms, or, in other words, aversion for the open arms. Theseesults, together with those from SAPs performed from the cen-er and the closed arms, support the notion of increased anxietyn the rats tested in the conventional maze. Both aversion for thepen arms and SAPs are typical anxiety markers in conventionalests using the elevated plus-maze (see, for example, Carobrez andertoglio, 2005; Rodgers et al., 1997).

In order to evaluate the importance of vision in the perception ofhe openness of the open arms, part of the rats were tested in a mazen which these arms were surrounded by transparent walls. Theseats (and those tested in the opaque walls maze) showed moreopen” arms horizontal and vertical exploration than those testedn the conventional maze and barely exhibited SAPs (although, ver-ical exploration, i.e., rearing, could possibly be led by the presencef vertical surfaces). In fact, “open” arms exploration was similar tohat of the closed, i.e., no arm preferences were established. Sincehe rats tested in the transparent walls maze did not show closedrms preference, the current results do not support the notion thatpen space perception through vision is an important trigger of ratnxiety. Instead, current results indicate that anxiety was mainlyriggered when tactile contact with vertical surfaces was prevented.ccordingly, in an experiment aimed at testing whether open armsvoidance is dependent on the absence of vertical surfaces in theserms, it was found that rats explore more an open arm providedith a clear sheet of Plexiglas placed in one of its sides than a

onventional open arm (Treit et al., 1993). Furthermore, anothertudy shows that thigmotaxis (being near to a vertical surface) cane suppressed by anxiolytic drugs (Treit and Fundytus, 1989). Aimilar anxiolytic effect over thigmotaxis is described in a study

ith the ‘elevated minus-maze’ (Pickles and Hendrie, 2013). Taken

ogether, these studies and the current one suggest an importantole for tactile cues in rat anxiety.

ocesses 107 (2014) 106–111

In another study dedicated to evaluate what aspect of themaze leads to open arms aversion (Martinez et al., 2002), groupsof intact vibrissae rats were tested in plus-mazes provided with“open” arms surrounded by transparent walls of different heights.Groups tested with most of the heights (5 cm, 10 cm and 20 cm)showed more open arms exploration than a control group testedin a conventional maze. The group tested in the maze with 40-cm high transparent walls in the open arms, however, differentlyfrom its correspondent group in the current study, explored theopen arms similarly to the control group. By looking at differencesthat could account for the discrepant results between both proce-dures, one can note that the tested rats from the cited study wereslightly lighter (200–230 g) and, consequently, they were possi-bly younger (i.e., perhaps they were not fully adult). Adolescencecan lead to changes in the way rats respond to novelty: pubertymakes male rats apparently more fearful (as suggested by the studyof Primus and Kellogg, 1989). Nevertheless, this information donot account for the discrepancy between the studies because thepresent study’s rats – the apparently older ones – in fact did notshow more, but less avoidance of visual openness. It is worth notic-ing, however, the similarity between the results: in both studiesthe group tested in the maze with 40-cm high transparent wallsin the open arms virtually did not display SAP. This later similar-ity could be considered as suggesting low level of anxiety in bothgroups.

Curiously, by looking at the luminosity levels of each kind ofarm, it could be easily concluded that luminosity was not a predic-tor of the level of exploration in our study. A previous study suggeststhat open arm avoidance is proportional to the open:enclosed gra-dient of luminosity (i.e., when the open arms are much illuminatedas compared to the closed ones, they are greatly avoided) (Pereiraet al., 2005). In the current study, such gradient was higher in thetransparent walls maze than in the conventional maze. However, aspreviously discussed, the rats tested in the transparent walls maze,in contrast with those tested in the conventional one, exploredsimilarly the “open” and the closed arms. Therefore, rat behaviorin the current study seemed not to be sensitive to the gradient ofluminosity.

Rats which had their vibrissae cut before the test presentedroughly the same amount of open arm exploration than the intactvibrissae ones. This suggests that the acute absence of such senso-rial input leads to no deficit. In sharp contrast with this, accordingto Belzung (1999), vibrissotomized mice show no open arm avoid-ance in the elevated plus-maze. It is possible that both species (ratsand mice) use the vibrissae in different ways in assessing openspaces. Unfortunately, as the author mentions that as an unpub-lished observation, it is not possible to verify details about itsprocedure and results. In our study, in opposition to the appar-ent absence of effects of vibrissotomy, a closer look reveals somedifferences in the exploratory behavior. For example, among ratstested in the conventional maze, while intact rats showed decreasesin head dips in the open arms from the 1st to the 2nd half ofthe test, vibrissotomized rats kept a regular moderate level ofexploration from the session beginning. Further, although theirfrequency of head dipping from the center was moderate in theinitial part of the test, a significant decrease was found duringthe second half of the test. This group also showed more SAPsfrom the central square (but not from the closed arms) than itsintact corresponding group. In another study, rats which had theirvibrissae cut also showed subtle differences in exploration ascompared to intact ones, the authors interpreted this as a resultof impairment in information gathering (Cardenas et al., 2001).

rissotomized rats may represent a compensation of a possibleimpairment in information gathering caused by the absence of thistactile input.

ural Pr

csew

mimElbiHttlm

A

G

R

B

C

C

C

G

G.B. Filgueiras et al. / Behavio

On the other hand, the absence of differences between intact andut vibrissae groups in the entirely walled mazes (i.e., TW and OW)trengths the notion that the contribution of the vibrissal sense toxploratory behavior is only important when it comes to dealingith (real, not virtual) open spaces.

In conclusion, the current results do not support that rats relyuch on vision in exploring an elevated plus-maze. Although

mportant in tactile investigation, mystacial vibrissae absenceainly resulted in subtle behavioral alterations in the conventional

PM. These quite subtle alterations suggest that vibrissotomy wasargely compensated in thigmotaxis by other tactile inputs andy adjusting some exploratory aspects. While there is variability

n the plus-maze construction (and procedure, see, for example:ogg, 1996), current results strengthen the notion that an essen-

ial feature which is needed for it to produce reliable results ishe real (not virtual) openness of the open arms. Such openness isargely assessed through somatosensory input. Accordingly, plus-

aze anxiety seems to be mostly triggered through this sense.

cknowledgements

This research was supported by CNPq (proc. 400735/2009-1).BF was recipient of a CAPES fellowship.

eferences

elzung, C., 1999. Measuring rodent exploratory behavior. In: Crusio, W.E., Gerlai,R.T. (Eds.), Handbook of Molecular–Genetic Techniques for Brain and BehaviorResearch. Techniques in the Behavioral and Neural Sciences, vol. 13. Elsevier,Amsterdam, pp. 738–749.

ardenas, F., Lamprea, M.R., Morato, S., 2001. Vibrissal sense is not the main sensorymodality in the rat exploratory behavior in the elevated plus-maze. Behav. Brain.Res. 122, 169–174.

arobrez, A.P., Bertoglio, L.J., 2005. Ethological and temporal analyses of anxiety-likebehavior: the elevated plus-maze model 20 years. Neurosci. Biobehav. Res. 29,

1193–1205.

ruz, A.P., Frei, F., Graeff, F.G., 1994. Ethopharmacological analysis of rat behavioron the elevated plus-maze. Pharmacol. Biochem. Behav. 49, 171–176.

arcia, A.M.B., Cardenas, F.P., Morato, S., 2005. Effect of different illumination levelson rat behavior in the elevated plus-maze. Physiol. Behav. 85, 265–270.

ocesses 107 (2014) 106–111 111

Garcia, A.M.B., Cardenas, F.P., Morato, S., 2011. The effects of pentylenetetrazol,chlordiazepoxide and caffeine in rats tested in the elevated plus-maze dependon the experimental illumination. Behav. Brain Res. 217, 171–177.

Grossen, N.E., Kelly, M.J., 1972. Species-specific behavior and acquisition of avoid-ance behavior in rats. J. Comp. Physiol. Psychol. 81, 307–310.

Handley, S.L., Mithani, S., 1984. Effects of alpha-adrenoceptor agonists and antag-onists in a maze-exploration model of ‘fear”-motivated behaviour. N-S Arch.Pharmacol. 327, 1–5.

Hogg, S., 1996. A review of the validity and variability of the elevated plus-maze asan animal model of anxiety. Pharmacol. Biochem. Behav. 54, 21–30.

Martinez, J.C., Cardenas, F., Lamprea, M., Morato, S., 2002. The role of vision andproprioception in the aversion of rats to the open arms of an elevated plus maze.Behav. Process. 60, 15–26.

Meyer, M.E., Meyer, M.E., 1992. The effects of bilateral and unilateral vibrissotomyon behavior within aquatic and terrestrial environments. Physiol. Behav. 51,877–880.

Pellow, S., Chopin, P., File, S.E., Briley, M., 1985. Validation of open-closed arm entriesin an elevated plus-maze as a measure of anxiety in the rat. J. Neurosci. Methods14, 149–167.

Pereira, L.O., Cunha, I.C., Marino-Neto, J., Faria, M.S., Paschoalini, M.A., 2005. Thegradient of luminosity between open/enclosed arms, but not the absolute levelof lux, predicts the behavior of rats in the plus-maze. Behav. Brain Res. 159 (1),55–61.

Pickles, A.R., Hendrie, C.A., 2013. Anxiolytic-induced attenuation of thigmotaxis inthe elevated minus maze. Behav. Process. 97, 76–79.

Primus, R.J., Kellogg, C.K., 1989. Pubertal-related changes influence the developmentof environment-related social interaction in the male rat. Dev. Psychobiol. 22,633–643.

Rodgers, R.J., Cole, J.C., 1994. The elevated plus-maze: pharmacology methodologyand ethology. In: Cooper, S.J., Hendrie, C.A. (Eds.), Ethology and Psychopharma-cology. J. Wiley & Sons Ltd., Chichester, pp. 9–44.

Rodgers, R.J., Johnson, N.J.T., 1995. Factor analysis of spatiotemporal and ethologicalmeasures in the murine elevated plus-maze. Pharmacol. Biochem. Behav. 52,297–303.

Rodgers, R.J., Cao, B.J., Dalvi, A., Holmes, A., 1997. Animal models of anxiety: anethological perspective. Braz. J. Med. Biol. Res. 30 (3), 289–304.

Rosa, V.P., Vandresen, N., Calixto, A.V., Kovaleski, D.F., Faria, M.S., 2000. Temporalanalysis of the rat’s behaviour in the plus-maze: effect of midazolam. Pharmacol.Biochem. Behav. 67, 177–182.

Shiffman, H.R., Lore, R., Passafiume, J., Neeb, R., 1970. Role of vibrissae for depthperception in the rat (Rattus norvegicus). Anim. Behav. 18, 290–292.

Treit, D., Fundytus, M., 1989. Thigmotaxis as a test for anxiolytic activity in rats.Pharmacol. Biochem. 31, 959–962.

Treit, D., Menard, J., Royan, C., 1993. Anxiogenic stimuli in the elevated plus-maze.Pharmacol. Biochem. Behav. 44, 463–469.

Wallace, D.G., Gorny, B., Whishaw, I.Q., 2002. Rats can track odors, other rats, andthemselves: implications for the study of spatial behavior. Behav. Brain Res. 131,185–192.