Guilherme Delgado Martins 

 

 

 

 

 

 

Preference for reaching tactile stimulation in a territorial fish is not affected by social 

stress 

 

 

 

 

 

 

 

 

 

 

 

 

São José do Rio Preto, SP, Brasil. 

2019 



 

Guilherme Delgado Martins 

 

 

 

 

Preference for reaching tactile stimulation in a territorial fish is not affected by social stress 

 

 

 

 

 

 

 

Dissertação apresentada como parte dos requisitos para 

obtenção do título de Mestre em Biologia Animal, junto ao 

Programa de Pós-Graduação em Biologia Animal, do 

Instituto de Biociências, Letras e Ciências Exatas da 

Universidade Estadual Paulista “Júlio de Mesquita Filho”, 

Campus de São José do Rio Preto.  

 

Orientadora: Profª. Drª. Eliane Gonçalves de Freitas 

 

Financiador: CNPq, proc.: 131338/2017-0 

 

 

 

 

 

 

 

 

 

 

 

São José do Rio Preto, SP, Brasil. 

2019 



 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 



 

 

Guilherme Delgado Martins 

 

 

 

 

 

 

 

 

 

 Preference for reaching tactile stimulation in a territorial fish is not 

affected by social stress 

 

 

Dissertação apresentada como parte dos requisitos 

para obtenção do título de Mestre em Biologia 

Animal, junto ao Programa de Pós-Graduação em 

Biologia Animal, do Instituto de Biociências, Letras e 

Ciências Exatas da Universidade Estadual Paulista 

“Júlio de Mesquita Filho”, Câmpus de São José do Rio 

Preto. 

 

Financiadora: CNPq, proc.: 131338/2017-0  
 

 
 

Comissão Examinadora 

 

Profª. Drª. Eliane Gonçalves de Freitas 

UNESP – Câmpus de São José do Rio Preto 

Orientador 

 

Prof. Dr. João Luis Saraiva 

Centro de Ciências do Mar (CCMAR), Universidade do Algarve, Portugal. 

 

Dra. Caroline Marques Maia 

Instituto Gilson Volpato de Educação Científica (IGVEC) 

 

 

 

São José do Rio Preto 

30 de abril de 2019 



 

AGRADECIMENTOS 

Agradeço aos membros do laboratório pelas discussões sobre o assunto e por todo o 

suporte para a pesquisa, em especial ao Kawan Carvalho Martins e à Marcela Cesar Bolognesi 

pela ajuda com a confecção dos estimuladores tácteis e dos aparatos dos testes, e à professora 

Eliane Gonçalves-de-Freitas pela oportunidade, confiança e apoio.  

Agradeço à minha namorada Francielly por todo o amor, companheirismo, ajuda 

compreensão e carinho durante toda essa jornada, sentimentos que serão presentes para o 

resto de nossas vidas. 

Agradeço a meus pais pelo cuidado e amor dedicados desde o início de minha vida 

passando por todo meu período dentro da universidade. 

Esse trabalho foi financiado por bolsa de mestrado concedida pelo CNPq – Conselho 

Nacional de Desenvolvimento Científico e Tecnológico (auxílio concedido a GDM, 

#131338/2017-0; EGF #428296/2016-5). 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 



 

RESUMO 

O bem-estar animal é avaliado por indicadores fisiológicos e comportamentais, como nível de 

estresse, taxa de crescimento, comportamentos estereotipados e desempenho cognitivo. Uma 

abordagem recente utiliza a percepção do próprio animal como uma forma de saber se o 

mesmo se encontra em bom estado de bem-estar, por meio de testes de preferência e 

motivação, considerando que o animal está bem saúde quando está em condições por ele 

preferidas. Neste estudo, usamos essa abordagem para testar se peixes territoriais percebem a 

estimulação táctil corporal como um recurso positivo, uma vez que essa condição tem sido 

considerada uma maneira de melhorar o bem-estar em peixes. Assim, testamos se a tilápia-do-

nilo prefere e está motivada a acessar a estimulação tátil; e se a preferência aumenta após um 

estímulo negativo, como o estresse oriundo de interações agressivas (estresse social). Machos 

adultos foram isolados por oito dias em aquários contendo um estimulador táctil central 

formado por hastes plásticas verticais com cerdas de silicone, pelas quais os peixes passam 

recebendo estimulação táctil corporal. Posteriormente, os animais foram submetidos a um 

teste de preferência no qual o peixe tinha a opção de atravessar uma área com estimulação 

táctil ou uma área sem estimulação. Em seguida, os peixes passaram por um teste de 

motivação, no qual deviam superar estímulos aversivos (alta iluminação) para acessar a 

estimulação táctil. Depois disso, os animais foram pareados para estabelecer a hierarquia de 

social por meio de interações agressivas, e novamente submetidos a testes de preferência e 

motivação. Um grupo controle foi submetido aos mesmos procedimentos, mas sem o estressor 

social. Observamos que o número de atravessamentos foi maior pela área sem estimulação 

táctil, tanto nos testes de preferência quanto nos de motivação, embora algumas diferenças 

individuais tenham sido detectadas. No entanto, os peixes não evitaram a área com 

estimulação, mesmo com a opção de permanecer em áreas neutras; os animais também 

superaram a rota aversiva de alta iluminação para obter o acesso ao estimulador táctil. O 

estresse social não afetou o comportamento de preferência e motivação. Além disso, 

observamos que os peixes que atravessaram mais o estimulador se tornaram submissos, de 

modo que a estimulação táctil atuou modulando a hierarquia social reduzindo a agressividade. 

No geral, concluímos que os peixes não percebem a estimulação tátil como um alívio de 

estresse social. No entanto, como os animais não evitaram a estimulação e enfrentaram um 

estímulo aversivo para obter acesso a ela, sugerimos que a estimulação táctil pode representar 

uma condição positiva para peixes territoriais. 

Palavras-chave: bem-estar de peixes, escolha, motivação, comportamento social. 



 

ABSTRACT 

Animal welfare is evaluated by several physiological and behavioral indicators such as stress 

level, growth rate, stereotyped behavior and cognitive performance. A recent approach uses 

the perception of the animal itself as a way to evaluate if this animal is in a good welfare state, 

by preference and motivation tests, considering that the animal is in good welfare when it is in 

conditions that it choses itself. Here, we used this approach to test whether territorial fish 

perceive body tactile stimulation as a positive resource, since this condition has been 

considered a way to improve fish welfare. Thus, we tested whether the fish Nile tilapia prefers 

and are motivated to access tactile stimulation; and if the preference increases after a negative 

stimulus, such as the stress from aggressive interaction (social stress). Adult males were 

isolated for eight days in aquaria containing a central tactile stimulator device formed by 

vertical plastic sticks with silicone bristles, whose fish pass throughout, receiving body tactile 

stimulation. Afterwards, they were assigned to a preference test in which the fish had the 

option of crossing between an area with tactile stimulation and an area without stimulation. 

Then, fish went to a motivation test, in which it must overcome aversive stimuli (bright light) 

to access the tactile stimulation. Thereafter, they were paired to establish social rank by 

aggressive interactions, and subjected again to preference and motivation tests. A control 

group underwent the same procedures but without the social stressor. We observed that the 

number of crossings was higher in the area without tactile stimulation either in preference or 

motivation tests, although some individual differences were detected.  Nevertheless, the fish 

did not avoid the area with stimulation, even with the option of remaining in neutral areas; 

and also overcame the aversive route to obtain the access to the tactile stimulator. Social 

stress did not affect the preference and motivation behavior. Additionally, we observed that 

fish that crossed more through the apparatus became subordinate, thus that tactile stimulation 

modulates social rank hierarchy by reducing aggressiveness. Overall, we conclude that fish 

does not perceive tactile stimulation as a social stressor relieve. However, as fish did not 

avoid the stimulation and faced an aversive stimulus to gain access to it, we suggest that 

tactile stimulation may represent a positive condition for territorial fish.  

Keywords: fish welfare, choice, motivation, social behavior. 

 

 

 



 

Sumário 

1. INTRODUCTION ............................................................................................................. 7 

2. METHODS ........................................................................................................................ 9 

2.1. Fish housing .................................................................................................................... 9 

2.2. Strategy of study........................................................................................................... 10 

2.3. Experiencing tactile stimulation ................................................................................. 10 

2.4. Experimental Procedures ............................................................................................ 12 

2.4.1. Preference Test .......................................................................................................... 13 

2.4.2. Motivation Test ......................................................................................................... 13 

2.4.3. Social stress ................................................................................................................ 14 

2.5. Data analysis ................................................................................................................. 15 

2.5.1. Preference Index ....................................................................................................... 16 

3. ETHICAL NOTE ............................................................................................................ 17 

4. RESULTS ........................................................................................................................ 17 

4.1. Number of crossings from day 1 to 8.......................................................................... 17 

4.2. Preference test .............................................................................................................. 18 

4.3. Preference Index .......................................................................................................... 18 

4.4. Motivation test .............................................................................................................. 23 

4.5. Social rank .................................................................................................................... 25 

5. DISCUSSION .................................................................................................................. 29 

6. CONCLUSION................................................................................................................ 32 

REFERENCES .................................................................................................................... 33 

 

 

 

 

 

 

 

 



7 
 

 

1. INTRODUCTION 

Attention to animal welfare has been a frequent concern of society in recent years
 

(BROOM, 2011). Studies on fish welfare and aquaculture conditions have been increasing 

considerably
 
(ASHLEY, 2007) as long as fish are considered sentient beings

 
(VOLPATO et 

al., 2007)⁠, capable of presenting consciousness of feelings and sensations
 
(GALHARDO; 

OLIVEIRA, 2006). The most traditional animal welfare research relies on indicators of 

animal state as physiological responses, such as catecholamine and corticosteroid levels in 

blood plasma, which suggest stress levels variation
 
(HUNTINGFORD, 2006); behavioral 

responses such as the presence of abnormal behaviors
 
(HUNTINGFROD, 2006), stereotyped 

behaviors
 
(ALMAZÁN-RUEDA et al., 2004), changes in aggressive behavior

 
(BOSCOLO et 

al., 2011) and cognitive performance
 
(BRANDÃO et al., 2015). A more recent approach 

considers that a welfare condition can be inferred by the perception of animal needs and from 

the own animal perspective
 
(DUNCAN, 2006; KIRKDEN; PAJOR, 2006; NICOL et al., 

2009; GALHARDO; OLIVEIRA, 2011; FRANKS, 2019). 

An efficient way to investigate whether a particular condition confers welfare by 

animal perception is to test the choice and preference of these animals for such condition
 

(DUNCAN, 1992; VOLPATO et al., 2007; MAIA; VOLPATO, 2016). These tests are 

designed to analyze how the animal feels by functional and behavioral traits
 
(DUNCAN, 

2006) and the animal's ability to choose the absence of negative states
 
(DAWKINS, 206; 

BOISSY et al., 2007) and the presence of positive ones
 
(DUNCAN, 2002). Thus, if the 

animal presents this positive perception regarding some environment’s condition, it is 

expected that it will choose to access or remain in that condition
 
(VOLPATO et al., 2007; 

DUNCAN, 2002), for example the preference for structured environment in zebrafish (Danio 

rerio) and checker barbs (Puntius oligolepis)
 
(KISTLER et al., 2011) and the preference for 

substrate in Mozambique tilapia
 
(GALHARDO; OLIVEIRA, 2011). It is important to 

emphasize that choice and preference may represent different concepts depending on the 

context
 
(MAIA; VOLPATO, 2016). The preference-based approach requires consistency 

between the choices over time
 
(VOLPATO et al., 2007; MAIA; VOLPATO, 2016).  

As a complement to the preference tests, it is important to understand how animals are 

motivated to achieve a particular condition
 
(DUNCAN, 1992; DAWKINS, 2006; MILLOT et 

al., 2014; MAIA et al., 2017; FRANKS, 2019) according to the idea that the more important a 

condition is to the animal, the higher is the "price it is willing to pay" to reach it
 
(KIRKDEN; 



8 
 

PAJOR, 2006; GALHARDO; OLIVEIRA, 2011). The study of motivation is crucial to 

understand how animals make their decisions
 
(BERRIDGE, 2004; JENSEN; PEDERSEN, 

2008). It corresponds to the balance between the external stimuli and the internal state of the 

animal in the previous moment to a decision making
 
(KIRKDEN; PAJOR, 2006), for 

example, the choice to access a particular resource. Motivation tests can, then, indicate how 

valuable a resource is to the animals
 
(MAIA; VOLPATO, 2016), establishing a connection 

across perception, needing and preference in the analysis of welfare
 
(JENSEN; PEDERSEN, 

2008). In this context, one of the most suitable experimental designs to study motivation is the 

use of aversive obstacles or stimuli, which the animal must overcome in order to reach a 

particular resource
 
(DUNCAN, 1992; MAIA et al., 2017). The way an individual cope with 

these obstacles and stimuli indicates the importance of the condition or resource to be 

achieved by the animal
 
(GALHARDO; OLIVEIRA, 2011). Physical obstacles like push-door 

operant devices are commonly used
 
(OLSSON et al., 2002). In this model, increased costs are 

imposed by adding weights to a door and the animal has to push these heavy doors to access a 

certain resource. Psychological effort tests are also efficient in demonstrating animal 

motivation
 
(MAIA et al., 2017), for example pathways with aversive stimulus that animals 

have to cross to reach the resource. Theses stimuli can be a high light route
 
(MAIA et al., 

2017) or the presence of artificial water currents for fish
 
(SULLIVAN et al., 2016). Thus, 

different resources or attributes in the environment can be tested by preference and 

motivation. 

A recent way thought to improve welfare and increase positive state perception in 

vertebrates is the body tactile stimulation
 
(GONÇALVES-DE-FREITAS et al., 2019). In 

mammals, for example, physical interactions as body massage or touches are able to reduce 

stress
 
(FIELD et al., 2005. HERNANDEZ-REIF et al., 2007; PROBST et al., 2012); to 

relieve pain
 
(FIELD, 1998; JANE et al., 2009); to elevate serotonin levels

 
(FIELD et al., 

2005) to minimize behavioral traits of anxiety and depression
 
(FREITAS et al., 2015; 

ANTONIAZZI et al., 2017) and to prevent learning deficit
 
(DE LOS ANGELES et al., 2016). 

Furthermore, teleost fishes have been tested for this effect as well. Soares et al. (2009) were 

the first to show that tactile stimulation reduces stress in a coral reef fish which perform a 

client-cleaner interaction (Ctenochaetus striatus and Labroides dimidiatus, respectively). 

These authors observed that the clients seek for tactile stimulus promoted by the cleaners, and 

that such stimulation reduces cortisol levels of the client fish after confinement stress. 

Another study demonstrated that tactile stimulation reduces aggressive interaction in pairs of 

male Nile tilapia, a territorial species
 
(BOLOGNESI et al., 2019), although did not reduce 



9 
 

cortisol levels. In Soares et al.
 
(2009) study, tactile stimulation is part of the species’ natural 

behavior. However, Nile tilapia is a territorial fish; therefore it is not clear yet if tactile 

stimulation is perceived as a positive condition, in order to improve their welfare. In this 

context, a preference-based approach is adequate to answer this question.  

Thus, in this study we tested whether Nile tilapia freely choose tactile stimulation and 

also their motivation to access this resource. We also tested if such preference and motivation 

increase after a negative stimulus, such as the stress from aggressive interactions (social 

stress). If the perception of tactile stimulation is generalized and positive in vertebrates, we 

expected that territorial fish such as Nile tilapia would freely choose to access this resource 

and that they would be more motivated to access tactile stimulation after experiencing a 

stressful situation. Nile tilapia, Oreochromis niloticus (L.) is a very important species in 

aquaculture
 
(BARCELLOS et al., 1999. GONÇALVES-DE-FREITAS et al., 2019) and the 

knowledge of tools that promote welfare of these animals is highly relevant
 
(VOLPATO et 

al., 2009; GONÇALVES-DE-FREITAS et al., 2019). In social system of the Nile tilapia, the 

males establish a dominance hierarchy by aggressive confrontations
 

(ALVARENGA; 

VOLPATO, 1995, CARVALHO et al., 2008; BARRETO et al., 2011). The occurrence of 

these confrontations between conspecifics causes social stress, one of the main causes of 

chronic stress in Nile tilapia rearing
 
(BARCELLOS et al., 1999; BOSCOLO et al., 2011). The 

social stress arising from the establishment of the dominance hierarchy can generate serious 

damages to these animals, thus becoming an important factor of poor welfare
 
(GONÇALVES-

DE-FREITAS et al., 2019). Therefore, it is important to investigate conditions that reduce 

detrimental effects from social interaction in Nile tilapia, such as tactile stimulation.  

2. METHODS 

2.1. Fish housing 

Adult males of GIFT Nile tilapia from a commercial supplier (Ribeirão Preto, SP, 

Brazil), were kept in outdoor ponds at the IBILCE, UNESP, São José do Rio Preto, SP, 

Brazil. The fish were collected for the study and taken to the laboratory where they all were 

acclimated together for 15 days in a polyethylene water tank (ca. 500 L, 1 fish/10 L) with 

water at 28ºC and 12L:12D light regime (7:00 a.m. to 7:00 p.m.). The fish were fed with 

ration for tropical fish (28% CP, apparent satiety) once a day (10:00 a.m.). The water quality 

was maintained by biological filters with filtration of 400 L/h, and constant aeration. 

 



10 
 

2.2. Strategy of study  

Nile tilapia males were initially isolated for 8 days in an aquarium with the tactile 

stimulator apparatus on its center to allow them experience tactile stimulation. On the 9
th

 day 

they underwent a preference test with four sessions of choice for tactile stimulation and on the 

10
th

 day, a motivation test in which the fish had to cross an aversive area to achieve the 

stimulation. Then the fish were paired with another male of the experiment to set a social rank 

(dominant and subordinate) and create social stress (16 replicas). After that, the preference 

and motivation tests were repeated. A control group was subjected to the same procedure, but 

without social stress (16 replicas).  

Before isolation the animals were anesthetized by immersion in Benzocaine (0.03 g.L-

1), weighted, sized and individually identified by green elastomer (VIE tags), inserted under 2 

or 3 scales on each side of the body. The mean (± S.E.) standard length and weight of fish 

were respectively: Treatment with social stress: 10.76 cm ± 0.61 cm; 41.74 g ± 7.86 g; 

Control group: 10.42 cm ± 0.71 cm; 39.02 g ± 9.55 g. The animals were observed in glass 

aquaria (80 x 30 x 40 cm) coated with blue plastic to avoid visual contact between animals 

from neighboring aquaria, and because the blue color is less stressing for Nile tilapia (MAIA; 

VOLPATO, 2013). Video-recording was done by cameras placed above the aquaria that send 

the records to a computer in an adjacent room, and cameras placed on tripods in front of the 

aquaria. The photoperiod was set to 12L:12D (from 7:00 a.m. to 7:00 p.m.) and temperature 

to 28º C. The water quality was monitored by commercial kits and electronic devices: 

Ammonia (0.025 ppm) and Nitrite (0.125 ppm); pH (8.36 ± 0.25). 

2.3. Experiencing tactile stimulation 

We used an apparatus developed by Bolognesi et al.
 
(2019) to provide tactile 

stimulation in fish, which is made of a rectangular PVC (polyvinyl chloride) structure, filled 

with vertical plastic sticks with silicone bristles each side (Figure 1A). Silicone bristles were 

chosen because they are soft, thus avoiding mucus removing from the skin, scale loss and 

other possible injuries on the fish’s body. The apparatus was positioned in the center of the 

aquarium so that fish had to pass through bristled sticks to receive tactile stimulation (Figure 

1B). To stimulate fish to cross through the apparatus, we offered food (dry shrimps) on the 

aquarium side, opposite to the side where the fish was, using a handling feeder (Figure 1 C). 

 

 



11 
 

 

 

 

 

C)  

 

 

Figure 1. A) Photo of the apparatus for tactile stimulation positioned in the aquarium’s center. B) 

Example of a fish passing through the silicon bristles receiving the body tactile stimulation. C) Feeder. 

Before the beginning of the experiment, the fish underwent a period of habituation to 

this feeder. Here, the term "habituation" does not refer to the extinction or decrease of a 

behavior, but to the fact that the fish becomes accustomed to a new experimental condition. 

The food was the stimulus for the fish to experience initial contact with the tactile stimulator. 

The food used was dried shrimp, attached to the feeder. The fish were removed from the 

water tank and placed together, randomly in an aquarium (60x60x40 cm, ca. 140L, four fish 



12 
 

per aquarium), where they remained for 3 days under the same conditions as those used in 

acclimation, except for feeding. During this period the animals were fed with the feeders six 

times a day being three times in the morning (8:00 a.m., 9:30 a.m. and 11:00 a.m.) and three 

times in the afternoon (2:00 p.m., 3:30 p.m. and 5:00 p.m.). 

2.4. Experimental Procedures  

On the 1
st
 day of the experiment the animals were removed from the habituation 

aquaria and isolated in the experimental aquarium with the tactile apparatus. Two feeding 

sessions were held in the morning (9:00 a.m. and 11:00 a.m.) and two in the afternoon (2:00 

p.m. and 4:00 p.m.) for 8 days. All feeding sessions were video-recorded 5 minutes before, 5 

minutes during and 5 minutes after feeding. For data analysis, we considered the number of 

crossings before and after feeding. This protocol had been previously tested and validated by 

Bolognesi et al.
 
(2019). Afterwards, fish underwent preference and motivation tests, followed 

by the effect of social stress. 

On the 9
th

 day the animals were assigned to preference test with 4 sessions of choice 

(20 min each) at 9:00 a.m., 11:00 a.m., 2:00 p.m. and 4:00 p.m. On the 10
th

 day, the animals 

were assigned to 2 sessions of the motivation test (20 min each) at 9:00 a.m. and 2:00 p.m. On 

the 11
th

 day, the fish were removed from their original aquarium and paired with another one, 

also from the experiment, in a neutral aquarium for aggressive interactions (20 min; 8:30 

a.m.), thus generating social stress. After that, the animals returned to their original aquarium 

and were assigned again to the preference test (9:00 a.m. and 11:00 a.m.). The pairing was 

repeated at 1:30 pm, followed by the preference test (2:00 p.m. and 4:00 p.m.). On the 12
th

 

day the fish were removed again from their original aquarium and paired with another one in 

neutral aquarium for aggressive interactions (20 minutes; 9:30 am). After that, the animals 

returned to their original aquarium and were assigned to the motivation test (10:00 am). The 

pairing was repeated at 2:30 p.m., followed by the motivation test (3:00 p.m.).  

For the control group, the procedures and schedules were identical, except for the 

agonistic interaction. At this time, instead of pairing with another male, the fish were only 

placed in a new aquarium for 20 min so they had the same manipulation as the social stress 

treatment. Then, the animals returned to the aquariums for the tests of preference and 

motivation. 

 

 



13 
 

2.4.1. Preference Test 

In this test, the tactile apparatus had half of its area filled with silicone bristles, and the 

other half with the sticks but without the silicone bristles so that fish could choose one of 

these areas (with or without tactile stimulation) (Figure 2). During the test, the animals were 

filmed for 20 minutes and the frequency of crossings through the two areas was recorded. In 

this test we evaluated the spontaneous choice of the fish and, therefore, we did not use the 

feeder, avoiding the effect of conditioning on the preference. The test was repeated in the 

afternoon to evaluate the consistency and performed in a way that did not present any clue 

that could bias the animal's preference. In addition, the positions of the regions of the 

apparatus (with the bristles and without the bristles) were alternated between the replicates. 

 

 

Figure 2. Design of the preference test. 

2.4.2. Motivation Test 

This test was based on the study of Maia et al.
 
(2017), aiming to test the fish’s 

propensity to surpass an aversive obstacle to reach a resource. For this, a small area in one 

aquarium’s end was covered by a black fabric, creating a dark area (refugee). The fish were 

initially confined in this region through the use of transparent plastic plates. On the opposite 

side to the dark area the tactile stimulator was fixed, corresponding to the model with an area 

with the silicone bristles and another without the silicone bristles, like the preference tests. 

Between the dark area and the tactile stimulator, we made a route lightened by a LED lamp 

(900 lumens), positioned above the aquarium, thus creating an aversive obstacle for the 

animal, similar to the protocol from Maia et al.
 
(2017) for rainbow trout. Light was used 



14 
 

because it is a negative stimulus for several fish species
 
(MARCHESAN et al., 2005. 

MESQUITA et al., 2007), including Nile tilapia
 
(MAXIMINO et al., 2007).  

The "price to be paid" by the fish (motivation) was inferred from the latency for living 

the dark area and crossing the aversive route to access the stimulator (device). The animals 

were released from the dark area after 5 minutes and then filmed for 20 minutes. The latency 

to access the apparatus and the number of crossings in both areas of the apparatus were 

recorded. 

 

Figure 3. Design of the motivation test. 

2.4.3. Social stress 

We analyzed the aggressive interactions performed by each fish during the pairing 

(treatment with social stress), according to the previously described ethogram for Nile tilapia
 

(GONÇALVES-DE-FREITAS et al., 2008). Aggressive behavior was divided as attacks and 

displays. Attacks are the interactions with physical contact and higher energy expenditure (as 

nipping, mouth fight and undulation), and the displays are interactions without physical 

contact and lower energy expenditure
 
(ROS et al., 2006) (as lateral threat and circling). 

Aggressive interactions were quantified to determine the social rank of the paired fish in the 



15 
 

dominance hierarchy (dominants and subordinates). We used the dominance index to 

determine the social rank. The dominance index was calculated for each individual as the 

number of interactions emitted divided by the number of interactions emitted and received
 

(GONÇALVES-DE-FREITAS et al., 2008). This value varies from 0 to 1, and the animal 

with the index closer to 1 is considered the dominant and closer to 0 the subordinate
 

(GONÇALVES-DE-FREITAS et al., 2008). 

 

A)

 

B) 

      

Figure 4. Sequence of events during the experiment in: A) Treatment group B) Control group. 

2.5. Data analysis 

The data were tested for normality by Kolmogorov-Smirnov test and homoscedasticity 

by Fmax
 
(LEHNER, 1996). Mixed model ANOVA was used to compare between the 



16 
 

treatments (with social stress and without social stress) and inside each treatment: the number 

of crossings in the initial 8 days of contact with the tactile stimulator and in the preference and 

motivation tests; the latency of the access to the tactile stimulator in motivation tests. Data 

were also contrasted by Planned Comparisons. 

In the first 8 days of the experiment we quantified the number of crossings through the 

tactile stimulator by the fish 5 minutes before and 5 minutes after feeding to test the efficiency 

of using the apparatus over time. On the days of the preference test we quantified the number 

of crossings through the tactile stimulator in the area with and without the bristles, between 

treatments. In motivation tests we quantified the latency to access the tactile stimulator, in 

addition to the number of crossings in each areas (with and without the bristles) and also the 

frequency of aggressive interactions emitted by each fish during the pairing (treatment with 

social stress). The association between aggressive interactions and elevated plasma cortisol 

levels is already well established for Nile tilapia
 
(CORREA et al., 2003; BARRETO et al., 

2015; BOLOGNESI et al., 2019) and therefore we did not measure the cortisol levels of 

treatments. 

2.5.1. Preference Index 

We used the preference index (PI) based on the study of Maia & Volpato
 
(2016) to 

complement the analysis of the number of crossings in the preference test of each fish, 

individually. This analyze allows us to check over preferred and non-preferred items in a 

preference test, in our case, with two options of choice. The index calculation follows some 

steps that are represented in an example in Table 1. First, the frequency of crossings in each 

area (with or without tactile stimulation) was summed. Then, we calculated the areas above 

the line of cumulative-frequency so that these calculated areas increase as the preference test 

trials progress. The most recent choices may better represent what the animal really prefers 

considering its choices over time
 
(MAIA; VOLPATO, 2016; MAIA et al., 2017). Then, all 

the calculated areas are cumulative. The positive PIs represent the preferred option, 

meanwhile the negative PI values indicate the non-preferred option, and the values of PI 

represent the intensity of preference/dispreference responses
 
(MAIA; VOLPATO, 2016; 

MAIA et al., 2017). We calculated for each fish, separated PI values that include four test 

trials before manipulation and four test trials after manipulation (manipulation was: male 

pairing in a new aquarium in the treatment group or isolated in a new aquarium in the control 

group). 

 



17 
 

Table 1. Preference Index steps. Example of treatment’s group fish 1 PI values for the area with tactile 

stimulation before manipulation.  
 

Test 

period 
Frequency 

Cumulative 

frequency 
Area 

Cumulative 

area 

Expected 

area 

Area 

variation 

Preference 

Index (PI) 

1 2 2 1 1 5,25 -4,25 -4,25 

2 6 8 9 10 30,75 -20,75 -25 

3 8 16 20 30 50,75 -20,75 -45,75 

4 5 21 17,5 47,5 80,5 -33 -78,75 

 

3. ETHICAL NOTE 

This study is in accordance with the Ethical Principles adopted by the National 

Council for the Control of Animal Experimentation (CONCEA / Brazil) and was approved by 

the Committee on Ethics in Animal Use, IBILCE, UNESP, São José do Rio Preto, permit 

#171/2017.  

4. RESULTS 

4.1. Number of crossings from day 1 to 8 

A different number of crossings through the tactile stimulator was found in the 8 days 

of initial contact with the apparatus (F (7,105) = 20.57 p < 0.0001, Figure 5). The number of 

crossings in day 2 increased from day 1 (p = 0.00015) and was lower than in the remaining 

days (p < 0.0008). The number of crossings became stable from day 3 to 8 (p > 0.664). 

 

Figure 5.  Number of crossings through the tactile stimulator during the eight days of initial contact 

with the apparatus (32 replicas). ANOVA for repeated measures followed by Fisher-LSD test. 

Asterisk indicates significant differences between the days. Data are mean ± SE. 



18 
 

4.2. Preference test 

There was no significant statistical interaction for the number of crossings through 

tactile device between treatments in the preference test (with vs. without social stress; F (3,90) = 

0.71, p = 0.54, Figure 6). However, we observed differences in the number of crossings 

between the areas with and without tactile stimulation before and after manipulation (F (3,90) = 

19.44, p = 0.0001). Planned comparisons showed that fish crossed more times through the 

area without tactile stimulation than through the area with tactile stimulation in the control, 

before (F (1,30) = 7.03, p = 0.012) and after (F (1,30) = 12.17, p = 0.015) manipulation and in the 

social stress treatment before (F (1,30) = 7.53, p = 0.010) and after (F (1,30) = 12.17, p = 0.001) 

manipulation. We also observed by planned comparisons that after fighting, the number of 

crossings through the area with tactile stimulation decreased (F (1,30) = 13.34, p = 0.009) such 

as the number of crossings through the area without tactile stimulation (F (1,30) = 13.09, p = 

0.008), but was similar before (F (1,30) = 2.05, p = 0.122) and after (F (1,30) = 2.77, p = 0.106) 

manipulation in the control group. 

Figure 6. Number of crossings in preference test through the area with or without stimulation over the 

periods before and after manipulation (male pairing in a new aquarium in the treatment group or 

isolated in a new aquarium in the control group). ANOVA for repeated measures and planned 

comparisons. Asterisk indicates significant differences within treatments in the periods. Letters 

compares before and after manipulation for control and treatment groups. Similar letters indicate no 

significant differences. Data are mean ± SE. 

4.3. Preference Index 

 In the control group (16 replicas) only individuals 6, 12, 14 and 16 preferred to access 

the area with tactile stimulation before manipulation (Figure 7A), and individuals 2, 4, 12 and 



19 
 

16 preferred tactile stimulation after manipulation (Figure 7B). In the treatment group (16 

replicas) only individuals 4, 6, 8, 12, and 14 preferred to access the area with tactile 

stimulation, meanwhile the rest of the fish preferred to access the area without tactile 

stimulation (Figure 8A). In the period after manipulation only fish 8 preferred to access the 

area with tactile stimulation, meanwhile the rest of the fish preferred to access the area 

without tactile stimulation (Figure 8B). 

A) 

 



20 
 

B) 

 

Figure 7. Preference Index of control group in the period before (A) and after (B) manipulation 

(isolation in a new aquarium). We did not count fish 13 in these calculations because it did not cross 

through the tactile stimulator in one of the periods. 

 

 

 

 



21 
 

A) 

 

 

 

 

 

 

 

 



22 
 

B) 

 

Figure 8. Preference Index of treatment group in the period before (A) and after (B) manipulation 

(pairing for fighting). We did not count fish 16 in these calculations because it did not cross through 

the tactile stimulator in one of the periods. 

 

 

 

 



23 
 

4.4. Motivation test 

No difference was observed for the latency to reach the apparatus between the 

treatments (with or without fights) and over the periods before and after manipulation (F (1,26) 

= 0.34, p = 0.56, Figure 9A). As the latency varied for each fish, we analyzed the number of 

crossing per minute available to the animal to access the tactile device (20 min – latency to 

fish leave the dark area). We did not find difference between the number of crossings of 

control and treatment groups (F (1,30) = 0.14, p = 0.703, Figure 9B). Again, planned 

comparisons showed that fish crossed more times through the area without tactile stimulation 

(F (3,90) = 20.61, p = 0.0001). In the control group, the number of crossings through the area 

without tactile stimulation was higher than the area with tactile stimulation before 

manipulation (F (1,30) = 15.85, p = 0.0004). After manipulation there was no difference 

between the crossings in both areas (F (1,30) = 3.19, p = 0.083). In the treatment group the 

number of crossings through the area without tactile stimulation was higher than the area with 

tactile stimulation before manipulation (F (1,30) = 32.15, p = 0.00004) and after manipulation 

(F (1,30) = 14.70, p = 0.0006). 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 



24 
 

A)  

 

B) 

 

Figure 9. A) Latency to access the tactile stimulator in the periods before and after manipulation 

(male pairing in a new aquarium in the treatment group or isolated in a new aquarium in the control 

group) for control and treatment group in the motivation test. ANOVA for repeated measures and 

planned comparisons. B) Number of crossings through the two areas of tactile stimulator (with and 

without stimulation) during the motivation test over the periods before and after manipulation. 

ANOVA for repeated measures and planned comparisons. Asterisk indicates significant differences 

between the areas with and without tactile stimulation. Letters compares before and after manipulation 



25 
 

for control and treatment groups. Similar letters indicate no significant differences. Data are mean ± 

SE. 

4.5. Social rank 

We analyzed the number of crossings through the tactile apparatus in the initial eight 

days according to the social rank (dominant and subordinate) that the fish acquired after 

agonistic interactions in the treatment group. There was a significant interaction between the 

social rank and the days of observation (F (7,98) = 4.49, p = 0.0002, Figure 10A). Planned 

comparisons showed that the number of crossings trough the stimulator of dominant fish was 

higher than subordinate one on day 2 (F (1,14) = 6.25, p = 0.05). On day 8 we observed that the 

number of crossings of subordinate fish was higher than of dominant fish (F (1,14) = 8.22, p = 

0.012). 

In the preference test there were differences in the number of crossings between the 

areas with and without tactile stimulation (F (1,14) = 8.50, p = 0.011, Figure 10B) and between 

dominant and subordinate individuals (F (3,42) = 9.70, p = 0.0005) but no significant 

interaction between preference and social rank (F (3,42) = 0.53, p = 0.66). Planned comparisons 

showed that the number of crossings for dominant individuals was higher in the area without 

tactile stimulation before (F (1,14) = 14.64, p = 0.001) and after aggressive interaction (F (1,14) = 

5.52, p = 0.04). There were no differences between the crossings in the different areas for 

subordinate individuals, either before (F (1,14) = 1.05, p = 0.321) or after aggressive interaction 

(F (1,14) = 3.98, p = 0.065). 

In the area with tactile stimulation, planned comparisons showed that the number of 

crossings by subordinate individuals before manipulation was higher than the crossings by 

dominant individuals (F (1,14) = 4.97, p = 0.049). We observed that for dominant individuals 

the number of crossings through the area without tactile stimulation after manipulation 

decreased in comparison with the period before manipulation (F (1,14) = 5.69, p = 0.032). 

There was not difference between the crossings through the area with tactile stimulation in the 

periods before and after manipulation (F (1,14) = 1.84, p = 0.196). Analyzing the subordinate 

fish, we observed that the crossings through the area with tactile stimulation were decreased 

after manipulation in comparison with before manipulation (F (1,14) = 7.54, p = 0.015) such as 

the crossings through the area without tactile stimulation (F (1,14) = 8.51, p = 0.011). 

We did not observe difference between the latency of dominant and subordinate fish 

both, before and after manipulation (F (1,14) = 0.59, p = 0.456, Figure 11A). Regarding the 

number of crossings, we observed difference between the crossings in the area with tactile 



26 
 

stimulation and in the area without tactile stimulation (F (1,14) = 9.29, p = 0.008, Figure 11B). 

There was no difference between the number of crossings of subordinate and dominant 

individuals (F (3,42) = 1.79, p = 0.163). The number of crossings in the area without tactile 

stimulation was higher than in the area with tactile stimulation for dominant individuals 

before the manipulation (F (1,14) = 12.60, p = 0.003) and after the manipulation (F (1,14) = 9.11, 

p = 0.009). 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 



27 
 

A) 

 

B) 

 

Figure 10. A) Number of crossings through the tactile stimulator during the first eight days of contact 

with the apparatus for dominant and subordinate individuals. ANOVA for repeated measures and 

planned comparisons. Asterisk indicates significant differences between the number of crossings of 

dominant and subordinate individuals. B) Number of crossings through the two areas of tactile 

stimulator (with and without stimulation) during the preference test over the periods before and after 

manipulation (male pairing in a new aquarium in the treatment group or isolated in a new aquarium in 

the control group) dominant and subordinate individuals. ANOVA for repeated measures and planned 



28 
 

comparisons. Asterisk indicates significant differences between the areas with and without tactile 

stimulation. Letters compares before and after manipulation for dominant and subordinates. Similar 

letters indicate no significant differences. Data are mean ± SE. 

 

A) 

 

B)  

 

Figure 11. A) Latency of access to tactile stimulator in the periods before and after manipulation for 

subordinate and dominant individuals. ANOVA for repeated measures and planned comparisons. B) 

Number of crossings through the two areas of tactile stimulator (with and without stimulation) during 

the motivation test over the periods before and after manipulation (male pairing in a new aquarium in 

the treatment group or isolated in a new aquarium in the control group) for subordinate and dominant 

individuals. ANOVA for repeated measures followed by and planned comparisons. Asterisk indicates 



29 
 

significant differences between the areas with and without tactile stimulation. Letters indicates 

significant differences over the periods (before and after manipulation) for subordinate and dominant 

individuals. Data are mean ± SE. 

5. DISCUSSION 

In this study we showed that territorial fish use tactile stimulation spontaneously and 

the preference for the access varies individually. The animals went through an aversive route 

to obtain the access to tactile stimulation, and did not avoid the apparatus, thus suggesting that 

tactile stimulation is a resource of some importance. On the other hand, social stress did not 

influence the preference and motivation, showing that fish do not use tactile stimulation to 

relieve the effects of social aggressive interaction. However, we detected an effect of social 

rank, with subordinate fish accessing tactile stimulation more frequently than dominant ones, 

suggesting that tactile stimulation has a more effective influence in these animals. 

The pattern of crossings in the initial eight days of the experiment indicates that the 

tactile stimulator was efficient for providing body tactile stimulation, as previously showed by 

Bolognesi et al. (2019). Furthermore, fish gradually increased the crossings through the 

apparatus, irrespective of food presence followed by a stabilization, which means that fish 

need some time to adjust to the presence of the apparatus, as showed by Bolognesi et al.
 

(2019) as well. On the other hand, we can interpret that fish increase the access to the 

apparatus after the initial days because it is a positive condition for it
 
(DAWKINS, 2006; 

BOLOGNESI et al., 2019). Thus, the first part of this study provided an adequate experience 

of individual tactile stimulation.  

When the access to a resource that is not part of the natural range of animal´s life 

increases, probably that resource is interpreted as something good
 
(KIRKDEN; PAJOR, 2006; 

GALHARDO; OLIVEIRA, 2011), therefore we expect that the individual chooses that 

resource. However, despite the fish spontaneously crosses through the tactile device in the 

first phase, they do not prefer this condition in the preference test. Then, the higher number of 

crossings through the area without tactile stimulation can be explained by some factors. 

Natural physical contacts in Nile tilapia are mainly related to aggressive interactions 

(DAMSGARD; HUNTINGFORD, 2012)
 
than contacts such as the client-cleaner interaction 

in coral-reef species
 
(GRUTTER, 1995) or any other no aggressive contact, for example in 

shoal species
 
(HUTH; WISSEL, 1992). In this way, the artificial contact provided by the 

tactile stimulator could be interpreted by fish as a negative interaction, what can explain the 

higher number of crossings through the area without tactile stimulation. Another explanation 



30 
 

would be the fact that the stimulator apparatus could act as a visual barrier to the fish. For 

cichlids like Nile tilapia, vision plays a fundamental role in the species ecology
 
(CARLETON, 

2009) and is an important sensorial path of movement and space localization
 
(DOUGLAS; 

HAWRYSHYN, 1990), so that fish would prefer to move by an area with no restriction to the 

movements. Besides, physical barriers could represent obstacles to swimming capacity
 

(AMARAL et al., 2017) what can increase the energy expenditure e oxygen consume
 

(BOTHA et al., 2018) allowing the occurrence of negative consequences for the animal’s 

energetic balance. Yet, simply because it is easy to move by ways with no barriers.  

The question that remains here is why, then, fish do not avoid the tactile stimulation 

area if it is easier to detour the “barrier”? Besides, if the fish had the option of remaining in 

neutral areas of the aquarium, why did they cross the stimulation area? If the stimulation was 

negative, we would expect fish to avoid it. In fact, they did not. On the contrary, they were 

motivated to cross an aversive route to access the tactile stimulator. It would be possible that 

the route of high illumination created might not represent a true aversive stimulus for Nile 

tilapia. However, the quantity of lumens used in this study has already been applied to other 

fish species in motivation tests
 
(MAIA et al., 2017), and Nile tilapia is sensible to increased 

light intensity as well
 
(CARVALHO et al., 2013). In this way, the fact that the animals left a 

shaded area and then crossed to the aquarium’s extremity, where the apparatus was 

positioned, indicates that fish take some risk to reach the tactile stimulation.   

The preference indexes obtained for the preference test reinforce the general idea that 

most of the fish does not prefer to move through tactile device. However, the PI also shows 

the access to tactile stimulation varies individually, as shown for other resource types, as 

environmental color and shelter
 

(BROWNE et al., 2010; MAIA; VOLPATO, 2019). 

Therefore, we must consider such an individuality to evaluate preferences instead of 

concludes so fast that territorial fish, such as Nile tilapia, does not prefer tactile stimulation. 

Thus, considering that Nile tilapia does not avoid the apparatus, and looking to individual 

variations, we conclude that tactile stimulation is a condition that fish spontaneously reach 

for, thus meaning a positive resource for territorial fishes.   

Bolognesi et al. (2019) have already shown that tactile stimulation does not lowers 

cortisol after both social and non-social stress in Nile tilapia males. However, cortisol per se 

could not be the best indicator of positive effect of tactile stimulation. In this way, preference 

tests bring an advantage for evidencing the animal feelings and perceptions, irrespective of 

association with physiological indicators
 
(VOLPATO et al., 2007). Thus, considering the 

hypothesis that tactile stimulation would have a positive effect on relieving consequences 



31 
 

from negative stimulus, we predicted that a social stressor would cause an increase on the 

access of the animals to tactile stimulation, as previously observed by Bolognesi et al. (2019). 

They found an increase of Nile tilapia fish movement by the environment after social 

aggressive interaction, making fish pass through the tactile apparatus in the middle of the 

aquarium. However, we observed that in both the preference test and in the motivation test 

this type of stressful condition did not cause changes in the pattern of crossings through the 

apparatus. By comparing the periods before and after the manipulation, we detected a 

reduction in the number of crossings; however, this reduction occurs in the two options of 

choice, that is, areas with and without stimulation. This suggests that the social stressor in this 

study had an effect of reducing the activity of the animals in a general way, as showed by 

other authors
 
(CONTE, 2004; SCHRECK, 2016). Differences of our study from Bolognesi et 

al. (2019)’s study would be the longer time of fighting and smaller size of aquaria, thus 

putting fish in a more stressful condition. In fact, higher stress level in fish can cause a 

decrease in movement
 
(SCHRECK, 1990) swimming performance

 
(KUTTY; SUKURMAN, 

1975; LANKFORD et al., 2005) and in the animal's motivation to perform some spatial task
 

(GAIKWAD et al., 2011; WOOD et al., 2011). 

According to the latency to reach the apparatus, animals that experienced or not social 

stressor seems to have the same motivation to access the tactile stimulator. The pattern of 

crosses in the motivation test is in line with the results from preference test, reinforcing the 

idea that motivation complements the analysis of what the animal actually prefers
 
(DUNCAN, 

2002; MAIA et al., 2017; FRANKS, 2019). Again, by evaluating the PI, we observed a 

preference inversion after manipulation (aquarium exchange) in the control group, with fish 2 

and fish 4 that previously preferred to access the area without tactile stimulation, preferring 

the area with stimulation and fish 6 and fish 14 that previously preferred to access the area 

with tactile stimulation, preferring the area without stimulation. These results, besides 

reinforcing the individual variability of preference responses, suggest that the simple 

manipulation of the animals is enough to change the pattern of crossings. Regarding the 

treatment group, we observed that after aggressive interaction (pairing) fish 2, fish 4, fish 6 

and fish 14, which previously preferred to access the area with tactile stimulation, preferred 

the area without stimulation. This result reinforces the idea that the social stress seems to have 

acted in the decrease of the animal’s activity, including the crossings through the area with 

tactile stimulation. 

The association between the social rank and the number of crossings in the 8 days 

previous to preference and motivation tests demonstrates the tendency of the animals that 



32 
 

accessed more frequently the stimulation to become subordinate. In fact, tactile stimulation 

reduces the aggressiveness of Nile tilapia
 
(BOLOGNESI et al., 2019), therefore individuals 

who experienced more tactile stimulation should become less aggressive and lose the contest, 

as aggressive ability define the winner and the loser individual
 
(BOSCOLO et al., 2011). 

Another interesting fact regarding subordinate fish is that in the preference test, they equally 

accessed the areas with and without stimulation, unlike the dominant ones. This result is in 

agreement with the pattern of crossings in the initial 8 days of the fish that would become 

subordinate. However, after social stressor, subordinate and dominant showed the same 

number of crossings by the stimulator. It is known that the establishment social hierarchy in 

Nile tilapia could be stressful in a similar way either for subordinate or dominant animals
 

(CORREA et al., 2003). Therefore, after pairing, the stress levels of dominant and 

subordinate individuals could be equal, allowing a similar pattern of access to tactile 

stimulation for both. Despite this, some studies show that the effects of agonistic interactions 

can be higher in subordinate individuals in a longer period
 
(EJICK; SCHRECK, 1980) what 

could explain the decrease in crossings through the area with tactile simulation and through 

the area without tactile stimulation after the pairing for aggressive interactions in subordinate 

fish. 

In this study, we could observe that the pattern of preference and motivation responses 

by tactile stimulation is very variable for Nile tilapia. We had the purpose of testing whether 

or not the fish prefers tactile stimulation and, therefore, it was limited to two options of 

choice. In conditions in which few choices are offered to animals, we cannot conclude 

assertively that the one chosen more frequently will always be preferred, especially if it is a 

variable environment (MAIA; VOLPATO, 2019)
 
and where the options of choice can have 

gradual effects on the animal, as it is the case of tactile stimulation. In some studies, such as 

this one, preference and motivation responses are analyzed together
 
(SULLIVAN et al., 

2016). The conclusions of these analyzes can sometimes extrapolate the individual views of 

each test, showing that even if an option of choice is not accessed more often, animals may 

rather be motivated to access this option and consequently, this condition may represent a 

positive aspect from the perspective of the animal itself. 

6. CONCLUSION 

We conclude that tactile stimulation is not aversive to Nile tilapia, since animals do 

not avoid contact with the stimulator; social stress has no effect on preference and motivation, 

since fish do not increase the search for tactile stimulation after this situation; tactile 



33 
 

stimulation can be determinant in the modulation of dominance hierarchy by reducing 

aggressiveness and making individuals become subordinate; animals are motivated to access 

tactile stimulation as they face an aversive stimulus to gain such access, suggesting that 

stimulation may represent a positive condition for these animals. 

 

REFERENCES  

ALMAZÁN-RUEDA, P.; SCHRAMA, J. W.; VERRETH, J. A. Behavioural responses under 

different feeding methods and light regimes of the African catfish (Clarias gariepinus) 

juveniles. Aquaculture v. 231, p. 347–359, 2004. 

ALVARENGA, C. M. D.; VOLPATO, G. L. Agonistic profile and metabolism in alevins of 

the Nile tilapia. Physiology & Behavior v. 57, p. 75–80, 1995. 

AMARAL, S. D., et al. To swim or to jump? Passage behaviour of a potamodromous cyprinid 

over an experimental broad‐crested weir. River Research and Applications v. 34, p. 174–182, 

2017. 

ANTONIAZZI C. T et al. Tactile Stimulation During Different Developmental Periods 

Modifies Hippocampal BDNF and GR, Affecting Memory and Behavior in Adult Rats. 

Hippocampus v. 27, p. 210–220, 2017.  

ASHLEY, P. J. Fish welfare: Current issues in aquaculture. Applied Animal Behaviour 

Science v. 104, p. 199–235, 2007. 

BARCELLOS, L. J. G et al. Plasmatic levels of cortisol in the response to acute stress in Nile 

tilapia, Oreochromis niloticus (L.), previously exposed to chronic stress. Aquaculture 

Research v. 30, p. 437–444, 1999. 

BARRETO, R. E.; CARVALHO, G. G. A.; VOLPATO, G. L. The aggressive behavior of 

Nile tilapia introduced into novel environments with variation in enrichment. Zoology, v. 114, 

p. 53–57, 2011. 

BARRETO, T. N., BOSCOLO, C. N. P., GONÇALVES-DE-FREITAS, E. Homogeneously 

sized groups increase aggressive interaction and affect social stress in Thai strain Nile tilapia 

(Oreochromis niloticus). Marine and Freshwater Behaviour and Physiology v. 48, p. 309–

318, 2015. 

BERRIDGE, K.C. Motivation concepts in behavioral neuroscience. Physiology and Behavior 

v. 81, p. 79–209, 2004. 

BOISSY, A. et al. Emotions and cognition: a new approach to animal welfare. Animal 

Welfare v. 16, p. 37–43, 2007. 

BOLOGNESI, M. C., GAUY, A. C. S., GONÇALVES-DE-FREITAS, E. Effect of tactile 

stimulation in a territorial fish. Scientific Reports v. 9, p. 40, 2019. 

BOSCOLO, C. N. P., MORAIS, R. N. & GONÇALVES-DE-FREITAS, E. Same-sized fsh 

groups increase aggressive interaction of sex-reversed males Nile tilapia GIFT strain. Applied 

Animal Behaviour Science v. 135, p. 154–159, 2011. 



34 
 

BOTHA, T. L et al. A tool for determining maximum sustained swimming ability of selected 

inland fish species in an Afrotropic ecozone. Water SA. v. 44, p. 511–515, 2018. 

BRANDÃO, M. L., BRAITHWAITE, V. A, GONÇALVES-DE-FREITAS, E. solation 

impairs cognition in a social fish Applied Animal Behaviour Science v. 171, p. 204–2010, 

2015. 

BROOM, D. M. A history of animal welfare science. Acta Biotheoretica. v. 59, p. 121–137, 

2011. 

BROWNE, W. J. et al. Consistency, transitivity and inter-relationships between measures of 

choice in environmental preference tests with chickens. Behavioural Processes v. 83, p. 72–

88, 2010. 

CARLETON, K. L. Cichlid fish visual systems: mechanisms of spectral tuning. Integrative 

Zoology v. 4, p. 75-86, 2009. 

CARVALHO, T. B.; GONÇALVES-DE-FREITAS, E. Sex group composition, social 

interaction, and metabolism in the fish Nile tilapia. Brazilian Journal of Biology v. 68, p. 

807–812, 2008. 

CARVALHO, T. B.; MENDONÇA, F. Z., COSTA-FERREIRA, R. S.; GONÇALVES-DE-

FREITAS, E. The effect of increased light intensity on the aggressive behavior of the Nile 

tilapia, Oreochromis niloticus (Teleostei: Cichlidae). Zoology v. 30, p. 125–129, 2013. 

CONTE, F. S. Stress and the welfare of cultured fish. Applied Animal Behaviour Science. v. 

86, p. 205–223, 2004. 

CORREA, S. A. et al. Effect of the establishment of dominance relationships on cortisol and 

other metabolic parameters in Nile tilapia (Oreochromis niloticus). Brazilian Journal of 

Medical and Biological Research v. 36, p. 1725-1731, 2003. 

DAMSGÅRD, B.; HUNTINGFORD, F. Fighting and aggression. In: HUNTINGFORD, F., 

JOBLING, M., KADRI, S. (Eds.). Aquaculture and Behavior. Oxford: Willey-Blackwell, p. 

248-285, 2012. 

DAWKINS, M. S. Through animal eyes: what behaviour tell us. Applied Animal Behaviour 

Science v. 100, p. 4–10, 2006. 

DE LOS ANGELES, G. A. M. et al. Tactile stimulation effects on hippocampal neurogenesis 

and spatial learning and memory in prenatally stressed rats. Brain Research Bulletin v. 124, p. 

1-11, 2016. 

DOUGLAS R. H., HAWRYSHYN C. W. Behavioural studies on fish vision: an analysis of 

visual capabilities In: DOUGLAS, R. H., DJAMGOZ, M. B. A (Eds.). The visual system of 

fish. Chapman and Hall, p. 373-418, 1990.  

DUNCAN, I.J.H. Measuring Preferences and the Strength of Preferences. Symposium: Quant. 

Behav. Poult. p. 658-663 (1992). 

DUNCAN, I.J.H. Poultry welfare: science or subjectivity? Poultry Science v. 43, p. 643-652, 

2002. 

DUNCAN, I.J.H.. The changing concept of animal sentience. Applied Animal Behaviour 

Science v. 100, p. 11–19, 2006. 



35 
 

EJIKE C.; SCHRECK C. B. Stress and social hierarchy rank in coho salmon. Transactions of 

the American Fisheries Society v. 109, p. 423-426, 1980. 

FIELD, T. Massage Therapy Effects. American Psychologist v. 53, p. 1270–1281, 1998. 

FIELD, T. et al. Cortisol decreases and serotonin and dopamine increase following massage 

therapy. International Journal of Neuroscience v. 115, p. 1397–1413, 2005. 

FRANKS, B. What do animals want? Animal Welfare v. 28, p. 1-10, 2019. 

FREITAS, D. et al. Neonatal tactile stimulation decreases depression-like and anxiety-like 

behaviors and potentiates sertraline action in young rats. International Journal of 

Developmental Neuroscience. v.47, p. 192–197, 2015. 

GAIKWAD, S., et al. Acute stress disrupts performance of zebrafish in the cued and spatial 

memory tests: The utility of fish models to study stress–memory interplay. Behavioural 

Processes v. 87, p. 224–230, 2011. 

GALHARDO, L., ALMEIDA, O., OLIVEIRA, R. F. Measuring motivation in a cichlid fish: 

an adaption of the push-door paradigm Applied Animal Behaviour Science v. 130, p. 60–70, 

2011. 

GALHARDO, L., ALMEIDA, O. & OLIVEIRA, R. F. Preference for the presence of 

substrate in male cichlid fish: effects of social dominance and context. Applied Animal 

Behaviour Science v. 120, p. 224–230, 2009. 

GALHARDO, L.; OLIVEIRA, R. Bem-estar animal: um conceito legítimo para peixes? 

Revista de Etologia v. 8, p. 51–61, 2006. 

GONÇALVES-DE-FREITAS, E. et al. Social Behavior and Welfare in Nile Tilapia. Fishes v. 

4, p. 23, 2019. 

GONÇALVES-DE-FREITAS, E. et al. Effect of water renewal on dominance hierarchy of 

juvenile Nile tilapia. Applied Animal Behaviour Science v. 112, p. 187–195, 2008. 

GRUTTER, A. S. Relationship between changing rates and ectoparasite loads in coral reef 

fshes. Marine Ecology Progress Series v. 118, p. 51–58, 1995. 

HERNANDEZ-REIF, M.; DIEGO, M.; FIELD, T. Preterm infants show reduced stress 

behaviors and activity after 5 days of massage therapy. Infant Behavior and Development v. 

30, p. 557–561, 2007. 

HUNTINGFORD, F. A. et al. Current issues in fsh welfare. Journal of Fish Biology v. 68, p. 

332–372, 2006 

HUTH, A., WISSEL, C. The Simulation of the Movement of Fish Schools. Journal of 

Theoretical Biology v. 156, p. 365–385, 1992. 

JANE, S. W. et al. Effects of a Full-Body Massage on Pain Intensity, Anxiety, and 

Physiological Relaxation in Taiwanese Patients with Metastatic Bone Pain: A Pilot Study. 

Journal of Pain and Symptom Management v. 37, p. 754–763, 2009. 

JENSEN, M. B.; PEDERSEN, L. J. Using motivation tests to assess ethological needs and 

preferences Applied Animal Behaviour Science v. 113, p. 340–356, 2008. 



36 
 

KIRKDEN, R. D.; PAJOR, E. A. Using preference, motivation and aversion tests to ask 

scientific questions about animals’ feelings. Applied Animal Behaviour Science v. 100, p. 29–

47, 2006. 

KISTLER, C. et al. Preference for structured environment in zebrafish (Danio rerio) and 

checker barbs (Puntius oligolepis). Applied Animal Behaviour Science v. 135, p. 318–327, 

2011. 

KUTTY, M. N., SUKURMAN, N. Influence of upper and lower temperature extremes on the 

swimming performance of Tilapia mossambica. Transactions of the American Fisheries 

Society v. 104, p. 755–761, 1975. 

LANKFORD, S. E. et al. The cost of chronic stress: impacts of a nonhabituating stress 

response on metabolic variables and swimming performance in sturgeon. Physiological and 

Biochemical Zoology v. 78, p. 599–609, 2005. 

LEHNER, P. N. Handbook of ethological methods. United Kingdom: Cambridge University 

Press, 1996, 672p. 

MAIA, C. M. et al. Physical and psychological motivation tests of individual preferences in 

rainbow trout. Journal of Zoology v. 302, p. 108–118, 2017. 

MAIA, C. M. VOLPATO, G. L. A history-based method to estimate animal preference. 

Scientific Reports v. 6, 28328, 2016. 

MAIA, C. M. & VOLPATO, G. L. Environmental light color affects the stress response of 

Nile tilapia. Zoology v. 116, p. 64–66, 2013. 

MAIA, C. M., VOLPATO, G. What to choose when the best preference is not available: does 

the Nile tilapia follow a linear sequence of preferences? Journal of Zoology. 

10.1111/jzo.12646, 2019. 

MARCHESAN, M. et al. Behavioural effects of artificial light on fish species of commercial 

interest. Fisheries Research v. 73, p. 171-185, 2005. 

MAXIMINO, C. et al. A Comparative Analysis of the Preference for Dark Environments in 

Five Teleosts. International Journal of Comparative Psychology v. 20, p. 351-367, 2007. 

MESQUITA, F. D. et al. A preliminary study into the effectiveness of stroboscopic light as an 

aversive stimulus for fish. Applied Animal Behaviour Science v. 111, p. 402–407, 2008. 

MILLOT, S. et al. Use of conditioned place preference/avoidance tests to assess affective 

states in fish. Applied Animal Behaviour Science v. 154, p. 104–111, 2014 

NICOL, C. J. et al. Associations between welfare indicators and environmental choice in 

laying hens. Animal Behavior v. 78, p. 413–424, 2009. 

OLSSON, I. A. S, KEELING, L. J.C The Push-Door for Measuring Motivation in Hens: 

Laying Hens are Motivated to Perch at Night. Animal Welfare v. 9, p. 11–19, 2002. 

PROBST, J. K. et al. Gentle touching in early life reduces avoidance distance and slaughter 

stress in beef cattle. Applied Animal Behaviour Science v. 139, p. 42–49, 2012. 

ROS, A. F. H., BECKER, K. & OLIVEIRA, R. F. Aggressive behaviour and energy 

metabolism in a cichlid fish, Oreochromis mossambicus. Physiology & Behavior v. 89, p. 

164–170, 2006. 



37 
 

SCHRECK, C. B. Physiological, behavioral, and performance indicators of stress. American 

Fisheries Society Symposium v. 8, p. 29–37, 1990. 

SCHRECK, C. B. The Concept of Stress in Fish. Fish Physiology v. 35, p. 1–34, 2016. 

SOARES, M. C. et al. Tactile stimulation lowers stress in fish. Nature Communications v. 2, 

p. 534–535, 2011. 

SULLIVAN, M., LAWRENCE, C., BLACHE, D. Why did the fish cross the tank? 

Objectively measuring the value of enrichment for captive fish. Applied Animal Behaviour 

Science v. 174, p. 181–188, 2016. 

VOLPATO, G. L., FREITAS, E. G., CASTILHO, M. F. Insights into the concept of fish 

welfare. Diseases of Aquatic Organisms v.75, p. 165–171, 2007. 

VOLPATO, G. L. et al. Animal welfare: from concepts to reality. Oecologia brasiliensis. v. 

13, p. 5–15, 2009. 

WOOD, L. S., DESJARDINS, J. K., FERNAND, R. D. Effects of stress and motivation on 

performing a spatial task. Neurobiology of Learning and Memory v. 95, p. 277–285, 2011.