Ci a ion: Egui aun, H.; Ma inez, I.
Non-Linea Analyses o Fish
Beha iou s in Response o Aqua ic
En i onmen al Pollu an s—A Re iew.
Fishes 2023,8, 311. h ps://doi.o g/
10.3390/ ishes8060311
Academic Edi o : Youji Wang
Recei ed: 11 May 2023
Re ised: 5 June 2023
Accep ed: 9 June 2023
Published: 12 June 2023
Copy igh : © 2023 by he au ho s.
Licensee MDPI, Basel, Swi ze land.
This a icle is an open access a icle
dis ibu ed unde he e ms and
condi ions o he C ea i e Commons
A ibu ion (CC BY) license (h ps://
c ea i ecommons.o g/licenses/by/
4.0/).
ishes
Re iew
Non-Linea Analyses o Fish Beha iou s in Response o
Aqua ic En i onmen al Pollu an s—A Re iew
Ha kai z Egui aun 1,2,* and Icia Ma inez 2,3,4
1Depa men o G aphic Design & Enginee ing P ojec s, Facul y o Enginee ing in Bilbao, Uni e si y o he
Basque Coun y UPV/EHU, 48013 Bilbao, Spain
2Resea ch Cen e o Expe imen al Ma ine Biology and Bio echnology—Plen ziako i sas Es azioa (PiE),
Uni e si y o he Basque Coun y (UPV/EHU), 48620 Plen zia, Spain; icia [email p o ec ed]
3
Depa men o Zoology and Animal Cell Biology, Facul y o Science and Technology, Uni e si y o he Basque
Coun y (UPV/EHU), 48940 Leioa, Spain
4IKERBASQUE, Basque Founda ion o Science, 48009 Bilbao, Spain
*Co espondence: [email p o ec ed]
Abs ac :
Analysis o ish beha iou is an e ec i e way o indi ec ly iden i y he p esence o en i on-
men al pollu an s ha nega i ely a ec ish li e, i s p oduc ion and quali y. Moni o ing indi idual
and collec i e beha iou s p oduces la ge amoun s o non-linea da a ha equi e ailo -sui ed com-
pu a ional me hods o in e p e and manage he in o ma ion. F ac al dimension (FD) and en opy
a e wo g oups o such non-linea analysing me hods ha se e as indica o s o he complexi y (FD)
and p edic abili y (en opy) o he beha iou s. Since beha iou al complexi y and p edic abili y may
be modula ed by con aminan s, he changes in i s FD and en opy alues ha e a clea po en ial o
be embedded in a biological ea ly wa ning sys em (BEWS), which may be pa icula ly use ul in
P ecision Fish Fa ming se ings and o moni o wild popula ions. This wo k p esen s a e iew o
he e ec s o a wide ange o en i onmen al con aminan s, including oxic compounds, cleaning
and disin ec ing agen s, s imulan (ca eine), anaes he ics and an ibio ics, hea y me als (lead, cuppe ,
and me cu y), selenium, pes icides and pe sis en en i onmen al pollu an s, on he FD and en opy
alues o collec i e and indi idual beha iou al esponses o di e en ish species. All he e ised
s udies demons a e he use ulness o bo h FD and en opy o indica e he p esence o pollu an s and
unde line he need o conside ea ly changes in he end o he e olu ion o hei alues p io o
hem becoming signi ican ly di e en om he con ol alues, i.e., while i is s ill possible o iden i y
he con aminan and p ese e he heal h and in eg i y o he ish.
Keywo ds:
aquacul u e; aqua ic pollu an s; biological ea ly wa ning sys ems (BEWS); en opy; en i-
onmen al con aminan s/pollu an s; ish beha iou ; ac al dimension; p ecision ish a ming (PFF)
Key Con ibu ion:
This pape is a e iew o ecen scien i ic publica ions assessing he non-linea
analyses o ish beha iou al esponses o en i onmen al pollu an s. All he wo ks indica e he alue
o ac al dimension and en opy analyses o di e en ypes o beha iou al esponses o de ec
changes induced by con aminan s, unde lining he need o conside ea ly changes in he end o
e olu ion o hei alues p io o hem becoming signi ican ly di e en om he con ol alues, i.e.,
be o e he ish has su e ed signi ican damage.
1. In oduc ion
An es ima ed 811 million people a e hunge s icken and abou 3000 million canno
a o d heal hy die s [
1
]. Gi en he ac ha human popula ion is expec ed o g ow om
he cu en 7700 million people o 9700 million by 2050 (Uni ed Na ions Wo ld Popula ion
P ospec s h ps://popula ion.un.o g/wpp/, accessed on 11 May 2023), he need o sus-
ainably inc ease ood p oduc ion has become impe a i e. Sea ood ( esh- and seawa e
o ganisms, animals and algae) is conside ed o be a key playe o ensu e he cu en and
Fishes 2023,8, 311. h ps://doi.o g/10.3390/ ishes8060311 h ps://www.mdpi.com/jou nal/ ishes
Fishes 2023,8, 311 2 o 25
u u e sus ainable supply o heal hy and nu i ious ood [
1
]. P esen ly, sea ood p o ides
abou 17% o wo ldwide animal p o ein, bu in Asia and some A ican coun ies his alue
is o e 50% [
1
]. Addi ionally, and gi en he agile condi ion o many wild ish s ocks
and comme cial cap u e ishe ies, mos o he sea ood supply is expec ed o come om
aquacul u e p oduc ion [1].
Howe e , he aquacul u e indus y aces se ious challenges, such as he unp edic abil-
i y o p oduc ion and he need o ensu e he heal h and wel a e o he ish [
2
]. Many
o hese challenges a e caused by clima e change [
3
]; eme gen diseases; he o e use o
an ibio ics and consequen inc ease in an ibio ic esis an pa hogens [
4
]; and he inc ease
in he amoun and a ie y o human d ugs, chemical and biological aqua ic pollu an s
(i.e., see [
5
]), including he wo ldwide p esence o mic o and nano-plas ics wi h hei
associa ed con aminan load [
6
–
8
], and he challenges in ob aining aw ma e ials o eed
manu ac u e [
9
]. In addi ion, pandemics and wa s a e ad e sely a ec ing he s abili y o
he en i e p oduc ion sys em.
In esponse o hese challenges, Eu opean policy [
10
] and ish a ming indus ies [
9
]
ocused hei e o s on a ou ing esilien and g een p oduc ion sys ems. A ecen ly pub-
lished epo o he Eu opean Inno a ion Council iden i ying eme ging echnologies [
11
]
explici ly lis s En i onmen al In elligence and Moni o ing Sys ems as ele an a eas whe e
inno a ion is equi ed, and one a ge ed key a ea o u gen inno a ions comp ises Ea ly
Wa ning Tools and P ecision Fa ming. The aquacul u e indus y needs o be esilien
acco ding o he de ini ion o he e m used in he EC S a egic Fo esigh Repo [
12
].
Pa adoxically, a s ong indus y needs o be able o o ecas he quan i y and quali y o
i s p oduc ion, i.e., i mus main ain a p oduc ion sys em as p edic able and s able as
possible. In o he wo ds, as close as possible o “an equilib ium condi ion o s eady s a e o
s able oscilla ion” as de ined by Holling [
13
], and we mus keep in mind ha p oduc ion
op imiza ion and p edic abili y may no be compa ible wi h ishes’ biological esilience o
unexpec ed challenges.
1.1. P ecision Fish Fa ming
The need o inc eased and con olled ish p oduc ion led Fø e e al. [
14
] o in oduce
he concep o P ecision Fish Fa ming (PFF, also e e ed o as In elligen Aquacul u e)
whose aim is “ o apply con ol-enginee ing p inciples o ish p oduc ion, he eby imp o ing
he a me ’s abili y o moni o , con ol and documen biological p ocesses in ish a ms” [
14
].
In ensi e ish a ming includes all li e s ages: om he p oduc ion o eggs o he ma ke
size ish. The ha che y phase akes place indoo s unde comple ely con olled condi ions.
The subsequen ong owing can ake place in eci cula ing aquacul u e sys ems (RAS) o in
ou doo ponds o cages ha mus be enla ged o accommoda e he inc easing biomass as
he ish g ows. Rea ing in ou doo cages is he p e e ed me hod o seawa e species, whe e
he ish is subjec o na u al luc ua ions o en i onmen al a iables and, as men ioned
abo e, o he inc easing amoun and a ie y o pollu an s ha ac as s esso s and may
nega i ely in luence hei heal h and wel a e esul ing in a poo e quali y p oduc [
14
].
Fø e e al. [
14
] concep ualized PFF as consis ing o se e al cyclical ope a ional p ocesses
pe o med in ou phases as depic ed in Figu e 1( aken om hei pape and ep oduced
wi h hei pe mission): phase (1) is obse a ion and phase (2) is in e p e a ion o he ishes’
esponses ha esul in he basis o phase (3), ha is, decision making, upon which ac ions
a e implemen ed in phase 4. A simila concep ual amewo k has been success ully applied
o o he p oduc manu ac u ing indus ies (see e e ences in [14].
Cu en ly, mos p ac ices in ish a ming ope a ions a e pe o med manually, bu a
compe i i e indus y and pa icula ly PFF equi e online, as and accu a e me hods o
moni o ish beha iou and a ming ope a ions, such as biomass moni o ing and manage-
men o c i ical ope a ions including eeding, c owding, p esence o pa asi es (see e iew
by [14]), and assessmen o ish heal h and wel a e (see e iews [15,16]).
Fishes 2023,8, 311 3 o 25
Fishes 2023, 8, x FOR PEER REVIEW 3 o 25
managemen o c i ical ope a ions including eeding, c owding, p esence o pa asi es (see
e iew by [14]), and assessmen o ish heal h and wel a e (see e iews [15,16]).
Figu e 1. Cyclical ep esen a ion o P ecision Fish Fa ming (PFF) whe e ope a ional p ocesses a e
di ided in o ou phases: Obse e, In e p e , Decide and Ac . The inne cycle ep esen s he
p esen s a e-o - he-a in indus y, wi h manual ac ions and moni o ing, and expe ience-based
in e p e a ion and decision-making. The ou e cycle illus a es how he in oduc ion o PFF may
in luence he diffe en phases o he cycle. Figu e c edi s: And eas Myskja Lien, SINTEF Ocean.
Taken om Fø e e al. [14] wi h pe mission om he au ho s.
Bo h Fø e e al. [14] and Li e al. [17] p esen a summa y o senso s and moni o ing
me hods commonly used in he aquacul u e indus y and esea ch (Table 1). Roughly,
hey can be di ided in o hose based on he analysis o sounds/acous ics [18,19], some o
which p oduce images (i.e., [20]) and he mo e common me hods based on he analysis o
ideo images. Bo h he senso s hemsel es and hei deploymen may in luence ish
beha iou ; o example, acous ic eleme y equi es he in asi e p ocess o agging a
ce ain numbe o ish [19,21], he sona may in luence he ishes’ hea ing [22], and he use
o eco ding ins umen s handled by human o obo s ha e been shown o in luence hei
beha iou [23,24].
Table 1. Summa y o he mos common senso ypes and moni o ing me hods used in aquacul u e
p oduc ion and esea ch. Taken om Fø e e al. [14] acco ding o CC BY Licence. Fo u he
in o ma ion on addi ional applica ions o he senso s please e e o he o iginal publica ion.
Senso Type
Senso
Implemen a i
on
Animal Beha iou al Va iables In o ma ion
Le el
Sona Single-beam
sona Biomass dep h dis ibu ion wi hin beam G oup
Sona Spli -beam
sona Biomass dep h dis ibu ion Indi idual-
based g oup
Figu e 1.
Cyclical ep esen a ion o P ecision Fish Fa ming (PFF) whe e ope a ional p ocesses a e
di ided in o ou phases: Obse e, In e p e , Decide and Ac . The inne cycle ep esen s he p esen
s a e-o - he-a in indus y, wi h manual ac ions and moni o ing, and expe ience-based in e p e a ion
and decision-making. The ou e cycle illus a es how he in oduc ion o PFF may in luence he di e en
phases o he cycle. Figu e c edi s: And eas Myskja Lien, SINTEF Ocean. Taken om Fø e e al. [
14
]
wi h pe mission om he au ho s.
Bo h Fø e e al. [
14
] and Li e al. [
17
] p esen a summa y o senso s and moni o ing
me hods commonly used in he aquacul u e indus y and esea ch (Table 1). Roughly,
hey can be di ided in o hose based on he analysis o sounds/acous ics [
18
,
19
], some o
which p oduce images (i.e., [
20
]) and he mo e common me hods based on he analysis
o ideo images. Bo h he senso s hemsel es and hei deploymen may in luence ish
beha iou ; o example, acous ic eleme y equi es he in asi e p ocess o agging a
ce ain numbe o ish [
19
,
21
], he sona may in luence he ishes’ hea ing [
22
], and he
use o eco ding ins umen s handled by human o obo s ha e been shown o in luence
hei beha iou [23,24].
Table 1.
Summa y o he mos common senso ypes and moni o ing me hods used in aquacul u e
p oduc ion and esea ch. Taken om Fø e e al. [
14
] acco ding o CC BY Licence. Fo u he
in o ma ion on addi ional applica ions o he senso s please e e o he o iginal publica ion.
Senso Type Senso Implemen a ion Animal Beha iou al Va iables In o ma ion Le el
Sona Single-beam sona
Biomass dep h dis ibu ion wi hin beam
G oup
Sona Spli -beam sona Biomass dep h dis ibu ion Indi idual-based g oup
Sona Mul i-beam sona Mo emen dynamics (posi ion, and
speed) wi hin beam G oup
Hyd oacous ic Teleme y Indi idual ish ags Biomass dep h dis ibu ion Indi idual
Passi e hyd oacous ic sensing
Hyd ophone Mo emen dynamics (posi ion, and
speed) wi hin en i e cage olume G oup
Came a Su ace came a e.g., dep h, posi ion, accele a ion and
spa ial o ien a ion G oup
Came a Feeding came a (subme ged) Sound emi ed om ish popula ion,
gene al soundscape Indi idual-based g oup
Came a S e eo came a (subme ged) Su ace ac i i y (jumping/splashing) Indi idual-based g oup
Fixed senso s ( ixed came as and passi e hyd oacous ic sensing) in e e e leas wi h
he beha iou o he ish. Howe e , he o me can only eco d he ish in a ce ain egion
and he la e equi es he de elopmen o la ge da abases con aining speci ic acous ics
Fishes 2023,8, 311 4 o 25
inge p in s o each ele an pa ame e (species, sex, ype o beha iou , e c.) as well as
he possibili y o sepa a ely analysing he a ge ed and he backg ound sounds. Al hough
such da abases a e al eady being cons uc ed (i.e., h ps:// ishsounds.ne /index.js;h ps:
//web.u i.edu/gso/ esea ch/ ish-sounds/, bo h accessed on 11 May 2023), he app oach
is s ill a away om being p ac ically implemen ed o moni o ishes.
The leap om he human-based obse a ion, in e p e a ion and ope a ion o he sys-
em based on he expe ience o he ish a me o he au oma ic moni o ing and con ol
o he acili ies, equi es he handling and au oma ic in e p e a ion o he la ge amoun
o da a o which machine lea ning, a i icial in elligence algo i hms and decision sup-
po sys ems a e applied (see e iews by [
17
,
25
–
29
]). Compu e ision models ob ained
h ough he applica ion o in elligen me hods ha e been used o s udy indi idual and
collec i e ish beha iou [
28
,
30
] and o he de ec ion o pain [
31
], s ess esponses [
17
] and
unusual/abno mal beha iou s in ish schools [17,27,30,32].
1.2. Con aminan s as S esso s In luencing Fish Beha iou
The wel a e o he a med ish is cen al o all cu en aquacul u e p ac ices [
33
,
34
]
including PFF. Ko e e al. [
35
] p oposed o conside ish wel a e based on allos asis (achie -
ing s abili y h ough change) a he han he mo e classical homeos asis [
36
] which in ol es
mechanisms o keep key physiological a iables (such as blood O
2
and hea a e) wi hin
a na ow op imal ange o alues. Some well-known pa ame e s ha de e mine op imal
allos asis, heal h and wel a e, and o which he e a e sophis ica ed senso s, moni o ing
and con ol sys ems comme cially a ailable, include oxygen and ni ogen/NH4
+
concen-
a ions, empe a u e, ype o eed and eeding a e and ime [
14
,
28
,
37
]. Op imal wel a e
implies ha he animals a e kep unde condi ions as close as possible o wha is no mal
o hem, e en hough ha is e y a ely he case [
35
]. Too high and oo low allos a ic loads
cause loss o wel a e and may e en esul s in dea h. The s udy and iden i ica ion o wha
no mal beha iou o each a med species is, as well as hei op imal housing condi ions,
a e s ill in i s in ancy, in la ge pa , due o he di icul y o s udying he di e en species in
hei na u al en i onmen s. Rega ding s ess and s esso s, we will ollow he de ini ions
p oposed by Wendelaa Bonga [
38
] in which “s ess is a condi ion in which he dynamic
equilib ium o an animal is h ea ened o dis u bed as a esul o he ac ion o in insic
and/o ex insic s imuli”, and s esso s a e he ex insic s imuli.
New me hods a e being de eloped o he as sc eening o known d ugs, pes icides,
oxins, and mic oo ganisms [
39
–
41
]. Un o una ely, he e a e s ill addi ional s esso s ha
emain unknown and/o whose eal e ec is no easily measu able pa icula ly when
he ne e ec is he esul s o se e al in e ac ing chemicals in he en i onmen [
42
]. In
he las case, changes in he aqua ic o ganisms’ beha iou ha e been p oposed o se e
as Biological Ea ly Wa ning Sys ems (BEWS), o moni o he p esence o en i onmen al
con aminan s in wa e esou ces [43–45] and ish p oduc ion in aquacul u e [14,46–49].
Hea y me als, pes icides, o ganic pe sis en con aminan s and a whole ange o
human and e e ina y medicinal and ec ea ional d ugs a e ound in inc easing amoun s
in aqua ic (pa icula ly in eshwa e ) sys ems (see e iews by [
50
,
51
]). In 1983, S eele [
52
]
e ised he e ec s on ish beha iou o some wa e bo ne pollu an s (mos ly pes icides) a
concen a ions o de s o magni ude lowe han hose inducing mo bidi y and mo ali y
and e e ed o p e ious wo ks [53] p oposing ha pollu an s a ec ing basic neu ological
unc ions would be pa icula ly in e es ing, since hey would mos likely also in luence
he animal’s beha iou . Indeed, we ha e shown ha he dec ease in he Shannon en opy
(SE, [
54
,
55
]) o he collec i e beha iou o he Eu opean seabass induced by he p esence o
he neu o oxican me hyl me cu y (MeHg
+
) [
49
] and modula ed by he p esence o Se [
42
],
was e lec ed in he biochemical changes in he an ioxidan ac i i ies o hei li e s [
56
], and
likely also by he me aboli e p o ile o hei b ains (unpublished esul s using indi iduals
om he expe imen men ioned in [
49
] and shown in Figu e 2, albei no con i med by
subsequen s udies). Social s esses ha e also been shown o modula e he indi idual and
collec i e beha iou s o ishes [46].
Fishes 2023,8, 311 5 o 25
Fishes 2023, 8, x FOR PEER REVIEW 5 o 25
emain unknown and/o whose eal effec is no easily measu able pa icula ly when he
ne effec is he esul s o se e al in e ac ing chemicals in he en i onmen [42]. In he las
case, changes in he aqua ic o ganisms’ beha iou ha e been p oposed o se e as
Biological Ea ly Wa ning Sys ems (BEWS), o moni o he p esence o en i onmen al
con aminan s in wa e esou ces [43–45] and ish p oduc ion in aquacul u e [14,46–49].
Hea y me als, pes icides, o ganic pe sis en con aminan s and a whole ange o
human and e e ina y medicinal and ec ea ional d ugs a e ound in inc easing amoun s
in aqua ic (pa icula ly in eshwa e ) sys ems (see e iews by [50,51]). In 1983, S eele [52]
e ised he effec s on ish beha iou o some wa e bo ne pollu an s (mos ly pes icides) a
concen a ions o de s o magni ude lowe han hose inducing mo bidi y and mo ali y
and e e ed o p e ious wo ks [53] p oposing ha pollu an s affec ing basic neu ological
unc ions would be pa icula ly in e es ing, since hey would mos likely also in luence
he animal’s beha iou . Indeed, we ha e shown ha he dec ease in he Shannon en opy
(SE, [54,55]) o he collec i e beha iou o he Eu opean seabass induced by he p esence
o he neu o oxican me hyl me cu y (MeHg+) [49] and modula ed by he p esence o Se
[42], was e lec ed in he biochemical changes in he an ioxidan ac i i ies o hei li e s
[56], and likely also by he me aboli e p o ile o hei b ains (unpublished esul s o a
p elimina y expe imen shown in Figu e 2, albei no con i med by subsequen s udies).
Social s esses ha e also been shown o modula e he indi idual and collec i e beha iou s
o ishes [46].
Figu e 2. Pa ial leas squa es-disc iminan analysis (PLS-DA) o spec a om HR 1H-NMR
analysis om pe chlo ic acid ex ac s o Eu opean seabass b ains ha we e un ea ed (black do s),
ea ed o 6 days wi h 10 µg/L o Na2SeO3 ollowed by 2 weeks in seawa e (blue do s), p e ea ed
wi h 10 µg/L o Na2SeO3 (6 days) ollowed by 2 weeks o exposu e o MeHg+ (4 µg/L) (o ange do s)
and ish exposed o 2 weeks o 4 µg/L MeHg+ ( ed do s). Each do co esponds o one ish and he
colo co esponds o he ea men as explained abo e and in he igu e. The b oken line’s ci cles
ep esen he 95% o con iden bounds o each g oup o he same ea men . The same colo code
has been used o he samples and he ci cles. Unpublished esul s. HR H1NMR analyses, by
cou esy o D Daniel Pad o (Molecula Imaging Uni , CICBiomaGUNE, Donos ia, Spain).
Addi ional e iews documen ed he effec s o oxic agen s [57,58] and medicinal
d ugs [44] on he beha iou o diffe en ish species. Toxic subs ances al e ed locomo ion,
Figu e 2.
Pa ial leas squa es-disc iminan analysis (PLS-DA) o spec a om HR
1
H-NMR analysis
om pe chlo ic acid ex ac s o Eu opean seabass b ains ha we e un ea ed (black do s), ea ed
o 6 days wi h 10
µ
g/L o Na
2
SeO
3
ollowed by 2 weeks in seawa e (blue do s), p e ea ed wi h
10
µ
g/L o Na
2
SeO
3
(6 days) ollowed by 2 weeks o exposu e o MeHg
+
(4
µ
g/L) (o ange do s) and
ish exposed o 2 weeks o 4
µ
g/L MeHg
+
( ed do s). Each do co esponds o one ish and he colo
co esponds o he ea men as explained abo e and in he igu e. The b oken line’s ci cles ep esen
he 95% o con iden bounds o each g oup o he same ea men . The same colo code has been
used o he samples and he ci cles. Unpublished esul s. HR H
1
NMR analyses, by cou esy o
D Daniel Pad o (Molecula Imaging Uni , CICBiomaGUNE, Donos ia, Spain).
Addi ional e iews documen ed he e ec s o oxic agen s [
57
,
58
] and medicinal
d ugs [
44
] on he beha iou o di e en ish species. Toxic subs ances al e ed locomo-
ion, a oidance, escape, a ac ion and ala m eac ions, ep oduc i e and eeding be-
ha iou , pa en al ca e and he beha iou o de eloping ish [
57
] and pe sonali y and
cogni ion [
58
]. O he s, such as an idep essan s, an ihis amines, be a-blocke s, non-s e oidal
an i-in lamma o y d ugs and psychia ic, an iepilep ic and an icholines e asic d ugs in-
duced changes on boldness, agg ession, ac i i y, sociali y, eeding a e and ep oduc i e
beha iou [
44
] in a species-speci ic manne . No ably, psychia ic d ugs, an idep essan s and
an ihis amines we e ound o induce beha iou al changes a concen a ions close o hose
ound in aqua ic sys ems (ng/L o
µ
g/L) [
44
,
58
,
59
]. Changes induced in he indi idual lo-
como o ac i i y (dis ance a elled, sinuosi y and numbe o bu s s) o he zeb a ish la a
by he oxican s cyanide, oden icide, and o ganophospha e led B odin e al. [
44
] o p opose
he use o a zeb a ish-moni o ing biological sys em o he isk assessmen o d inking
wa e . Su yan o e al. [
59
] ha e ecen ly shown ha 16 ou o 20 di e en an ibio ics es ed
induced changes on swimming, explo a o y and/o locomo o beha iou s o zeb a ish ha
we e e lec ed in he FD and pe mu a ion en opy alues o some beha iou al pa ame e s
keeping a ela ionship wi h he an ibio ic’s oxici y.
Fish heal h is u he a ec ed when, along wi h he known en i onmen al pollu an s
and d ugs, we e alua e he nega i e e ec s o he mic o- and nanoplas ics, he oxic
subs ances hey con ain and elease, and hei accumula ion in o he ood webs [
8
,
60
,
61
].
Nanoplas ic inges ion h ough he ood web has been shown o damage he b ain o ishes
and induce diminished explo a o y beha iou compa ed o con ols [6].
Fishes 2023,8, 311 6 o 25
An h opogenic noise pollu ion is ano he sou ce o s ess o ishes ha in luences
he physiology and beha iou o indi iduals, causing empo a y o pe manen hea ing
loss, s ess and beha iou al eac ions o he noises [
62
] as well as changes in hei shoaling
beha iou [
63
], and e e ences he ein. In addi ion, noises made by p eda o species in
he en i onmen may also esul in a s ess esponse in bo h wild and a med ish. The
iden i ica ion o s ess ul noises will equi e passi e acous ic moni o ing de ices o iden i y,
i s ly, he no mal noises p oduced by he moni o ed species as well as he noises causing
he s ess and, subsequen ly, he noises he ish p oduce in esponse o he s esso s.
1.3. F ac al Dimension (FD) and En opy P ope ies in Biological Sys ems
Biological sys ems a e complex sys ems wi h an inhe en and a iable dynamic made
up o a la ge amoun o in o ma ion (signals) gene a ed o e ime ha con ain s ochas ic
and de e minis ic componen s [
64
]. Typically, he signals p ocessed a e ideo images
o echo sounds p oduced by obse ing he animal, bu classical obse a ion app oaches
a e no able o app aise he complexi y o he sys ems, pa icula ly in eal li e si ua ions
occu ing in complex en i onmen s wi h po en ially la ge backg ounds. The use o al e -
na i e echniques, which combine linea and non-linea app oaches, and he applica ion
o ma hema ical models and no el algo i hms, allow o cap u e mo e o he complexi y
inna e o biological sys ems [65,66].
The FD o a sys em, which is usually no an in ege numbe , is cha ac e is ic o ac al
s uc u es, i.e., hose ha show sel -simila i y [
67
]. En opy, he o he non-linea p ope y,
desc ibes he deg ee o p edic abili y/chao ici y o a sys em [
54
,
55
,
68
]. The e a e di e en
measu emen s o es ima e he FD and en opy (i.e., Rényi, Shannon, Kolmogo o en opy
among o he s) alues o a sys em.
The FD has a di ec ela ionship wi h he complexi y o he signal analysed (such as
mo emen o sound) [
69
]. F ac al analysis has been success ully applied o signal p ocess-
ing in many scien i ic ields, including he ecogni ion o pa e ns in human speech [
70
]
and walk [
71
] and in ish swimming [
72
–
74
]. In 1996, Alados e al. [
75
] ound ha he
FD o he explo a o y beha iou o goa s was lowe ed by s ess. The au ho s explained
hese esul s based on he ac ha , since s ess gene ally inc eases ene gy consump ion, i
would also lead o a educ ion in he FD (i.e., complexi y) o hei explo a o y beha iou .
In addi ion, he au ho s sugges ed in he same wo k [
75
] ha mos biological s uc u es
and beha iou pa e ns ha e e ol ed o allow he o ganism o explo e i s en i onmen
o o enhance i s ole ance o in e nal and en i onmen al changes; consequen ly, na u al
selec ion should ha e inc eased he complexi y in social and beha iou al s uc u es, i.e.,
hei FD, o maximum possible le els. S udying he collec i e beha iou o in elligen
agen s, Mann e al. [
76
] pos ula ed he causal en opic p inciple, acco ding o which agen s
in a g oup ollow beha iou al ules ha maximize he en opy o he sys em. The causal en-
opic p inciple was ound o be able o p edic many social in e ac ions in bo h human and
animal g oups. Acco ding o hose wo ks [
75
,
76
], op imally unc ional biological sys ems
should maximize bo h hei FD and en opy alues, while s esso s should diminish hem.
1.4. Aim o he Wo k
I is no cu en ly possible o know all he ele an pa ame e s ac ing as s esso s,
including en i onmen al pollu an s, o how each one o hem will, bo h indi idually
and h ough hei in e ac ions, in luence he s a us o a biological sys em (a any le el o
complexi y, be i an indi idual ish, shoal, all ish in a cage o he en i e p oduc ion o
a m) a any gi en ime. Adequa e moni o ing o he ish sys em, measu ing he ele an
a iables ha will allow he a me o ake he co ec decisions and ac ing in ime o
minimize he e ec s o he undesi able pe u bances, a e essen ial ac i i ies wi hin PFF. I is
in he cases when he s esso is ei he unknown o no egula ly measu ed, when changes
in non-linea p ope ies o he sys em, such as i s en opy and/o FD, ind hei op imal
p ac ical applica ion.
Fishes 2023,8, 311 7 o 25
A e iew o he e ec o o he s esso s (including agging/pain, numbe o ish,
posi i e emo ional con agion and ea /anxie y) on he en opy and FD alues o indi idual
and collec i e ish beha iou s has been ecen ly published by us [
46
] and will no be deal
wi h he e. The pu poses o his wo k a e (i) o gi e an o e iew o ecen wo ks epo ing
he e ec s o ele an en i onmen al con aminan s on he FD and en opy on indi idual
and collec i e beha iou al esponses o ish, and (ii) o indica e hei po en ial applica ions
o iden i y he p esence o undesi able en i onmen al con aminan s in ei he ish a ming
se ings o na u al popula ions.
2. Ma e ials and Me hods
The Web o Science and Scopus da abases we e used o bibliog aphic sea ches. The
Web o Science p o ided a much highe numbe o hi s bu mos o hem we e no di ec ly
ela ed o ou a ge ed subjec and he hi s ha we e ele an we e also iden i ied in Scopus.
The sea ch s ings and numbe o documen s ound (upda ed on he 28 Ap il 2023) a e
lis ed in Table 2. A e e iewing all he documen s, we ound only 18 dealing wi h he
exac subjec o ou s udy, namely he e ec on con aminan s on he FD and/o en opy
analyses o hei indi idual and/o collec i e beha iou s. A summa y o he main da a and
esul s om he 18 selec ed publica ions, so ed acco ding o he ype o oxican s, is shown
in Appendix A, Table A1. Table A2 o Appendix Ashows he e ec s o con aminan s on he
endencies o he FD and en opy alues o indi idual and collec i e beha iou al pa e ns.
Table 2. Sea ch s ings used in he Scopus da abase and numbe o hi s on 28 Ap il 2023.
Que y Documen s
(TITLE-ABS-KEY (en opy) AND TITLE-ABS-KEY ( ish) AND TITLE-ABS-KEY
(“biological wa ning sys em” OR bws)) 5
(TITLE-ABS-KEY (en opy) AND TITLE-ABS-KEY ( ish) AND TITLE-ABS-KEY
(“biological ea ly wa ning sys em” OR bews)) 3
(TITLE-ABS-KEY ( ish AND beha *) AND TITLE-ABS-KEY ( ac al* OR en opy)) 147
(TITLE-ABS-KEY (aquacul *) AND TITLE-ABS-KEY ( ac al* OR en opy)) 94
(TITLE-ABS-KEY (“Fish beha io*”) AND TITLE-ABS-KEY (en opy)) 12
(TITLE-ABS-KEY (“Fish beha io*”) AND TITLE-ABS-KEY ( ac al)) 10
(TITLE-ABS-KEY (“collec i e beha iou ” OR “collec i e beha io ”) AND
TITLE-ABS-KEY ( ish) AND TITLE-ABS-KEY (wel a e OR s ess* OR heal h OR disease))
25
(TITLE-ABS-KEY ((collec i e AND beha io*) AND ish) AND TITLE-ABS-KEY
(con aminan * OR wel a e OR s ess* OR heal h OR disease)) 62
3. F ac al Dimension and En opy Analyses o Fish Beha iou al Pa ame e s
Fo a p ac ical e iew on echnical issues we ecommend he e iew o [
77
]. These
au ho s add ess he challenges encoun e ed when pe o ming 2D and 3D ideo beha iou al
acking as well as he ype o compu a ional analyses (Fou ie and wa ele ans o ms,
ac al analysis and pe mu a ion en opy) and o machine lea ning echniques (mul ilaye
pe cep on, sel -o ganizing map, hidden Ma ko model and deep lea ning) wi h clea
po en ial o assis in he oxicological moni o ing o wa e quali y by BEWS.
In ou opinion, he mos comple e wo k abou he use o aqua ic o ganisms ( om
bac e ia, algae, bi al es and ish o mul ispecies sys ems) as BEWS o he con inuously
de ec ion o pollu an s in eshwa e sys ems, is he e iew by Bae and Pa k [
43
]. The
pu pose o hose wo ks was o apply BEWS o e ec i ely moni o d inking wa e , dis-
ibu ion sys ems and was ewa e e luen s in con amina ed o es o ed si es wo ldwide;
howe e , he same p inciples can be applied o moni o ing he en i onmen al condi ions
in aquacul u e se ings. Fo example, mos aqua ic o ganisms a e sensi i e o he p esence
o esidual chlo ine in chlo ina ed d inking wa e which may educe hei use ulness as
BEWS o d inking wa e , which is usually chlo ina ed. Howe e , his e y sensi i i y is
an ad an age o moni o he wa e in eshwa e aquacul u e, which mus ob iously be
chlo ine- ee. Addi ionally, al hough BEWS a e no e ec i e o de ec human pa hogens,
hey will mos likely de ec he p esence o ish pa hogens and pa asi es, and he e o e se e
as a good BEWS o moni o ish wel a e in ish p oduc ion. Many o he wo ks men ioned
Fishes 2023,8, 311 8 o 25
in he e iew ely on beha iou al moni o ing which has many ad an ages bu also ce ain
d awbacks since i is di icul o objec i ely quan i y and in e p e beha iou al da a. Such
challenge a ise due o he non-linea i y o beha iou s, a ia ion in indi idual beha iou
and he la ge amoun s o da a ob ained by con inuous moni o ing. Acco dingly, he wo k
also e iews me hods used o non-linea signal p ocessing (e.g., pe mu a ion en opy,
FD, Fou ie ans o m and wa ele ans o ma ion) and o he in e p e a ion o he da a,
including machine lea ning echniques, o di e en species. I mus always be kep in
mind ha BEWS in eg a e he esponse o all he s esso s/chemicals o which he o ganism
is exposed and does no p o ide quali a i e o quan i a i e in o ma ion abou each o he
po en ially p esen pollu an s. Consequen ly, one mus combine he implemen a ion o
BEWS wi h chemical moni o ing o iden i y and quan i y he po en ial oxican s. We do no
in end o epea he in o ma ion o hose e iews in his wo k and he eade is s ongly
encou aged o e e o hem [43,77] o u he in o ma ion.
The po en ial use o BEWS based on ac al and en opy analysis o ish beha iou al
pa ame e s has been success ully es ed on highly oxic subs ances a le hal doses [
78
],
common cleaning and disin ec ing agen s [
79
,
80
], s imulan s, an ibio ics and anaes he -
ics, [
59
,
81
,
82
], hea y me als [
42
,
49
,
83
–
87
] and pe sis en en i onmen al pollu an s and
pes icides [
87
–
90
]. Mos o he o ganisms on which hey ha e been es ed a e eshwa e
model species, such as zeb a ish and medaka, bu some wo ks use wild cap u ed ish [
82
],
o he ish species ele an o aquacul u e [
42
,
49
,
80
,
85
] and eshwa e - a med sh imps [
87
].
Thus, he selec ed s udies comp ehend all he mos ele an ypes o en i onmen al chemi-
cal pollu an s and a ele an ange o species (model ish, esh and seawa e species and
species ele an o aquacul u e).
3.1. E ec s o Exposu e o Toxic Compounds and o Cleaning and Disin ec ing Agen s
Assessmen o he exposu e o adul Japanese medaka (O yzias la ipes) o le hal doses
o po assium cyanide (KCN 10 mg/mL) and phenol (25 mg/mL) in wa e was ca ied ou
by Fukuda e al. [
78
] measu ing he 3D swimming speed and e ical posi ion o he ish,
as well as he uni o mi y o hese wo pa ame e s es ima ed by hei Shannon–Wea e
en opy [
55
]. Exposu e o bo h oxican s signi ican ly inc eased he maximum swimming
speed bu also signi ican ly dec eased hei median swimming speed. In e es ingly, a
dec ease in he en opy o he e ical posi ion o he ish ook place jus a e 10–30 min
exposu e o bo h chemicals and p io o any mo ali y. Al hough he e we e la ge indi idual
a ia ions, as i is usual in biological sys ems, he dec ease in SE o he sys em beha ed as a
use ul indica o o he p esence o he pollu an s.
A po en ial BEWS o he de ec ion o suble hal concen a ions o sodium hypochlo i e
(NaClO, om 0.001 o 0.005%) in eshwa e has been desc ibed by Nimke dphol e al. [
79
].
These au ho s used he FD (es ima ed using a c i ical exponen me hod [
91
]) o he swim-
ming ajec o y o indi idual male zeb a ish ob ained using wo ideo came as ia 3D
coo dina e compu a ion wi h pe spec i e co ec ion (3DCCPC). The biological ac i i y
o zeb a ish indeed displayed a ac al s uc u e and esponded o he p esence o he
disin ec an : he alue o FD o swimming ajec o ies ended o inc ease wi h he pH
and he alues o he FD o swimming eloci ies ended o inc ease wi h i s concen a-
ion. Al hough he pH alue always inc eased as NaClO was added and he expe imen s
we e pe o med unde con olled condi ions, he pH alue i sel a ied o a gi en NaClO
concen a ion. This was a ibu ed o possible eac ions o he NaClO wi h wa e o o he
compounds, and i would explain why he pH and beha iou al esponses do no necessa y
co ela e wi h he concen a ion o he pollu an . In addi ion, i may be easonable o as-
sume ha he e ec s caused on he ish by he pH alone may be di e en om hose caused
by he NaClO alone, e lec ing on he obse ed di e ences in he FD o he esponses o he
wo pa ame e s.
A commonly used de e gen (sodium dodecyl benzene sul ona e, SDBS) also induced
species-speci ic changes in he beha iou o h ee di e en eshwa e species: zeb a ish
(Danio e io), Japanese medaka (O yzias la ipes) and ed ca p (Cyp inus ca pio) [
80
]. The
Fishes 2023,8, 311 9 o 25
changes a ec ed hei swimming ail and speed and hei su ace beha iou . In addi ion,
he Shannon–Wea e en opy [
55
] o hese pa ame e s de ia ed om he “no m”. In
he h ee species, he en opy dec eased (i.e., hei beha iou became less no mal) wi h
inc easing concen a ions o he de e gen , bu wi h di e en sensi i i ies: zeb a ish was
he mos sensi i e species, esponding a he lowes concen a ion, ollowed by ed ca p
and medaka, which was he mos esis an .
3.2. E ec s o Exposu e o S imulan s, Anaes he ics and An ibio ics
Toxic subs ances ha e also been shown o in luence he beha iou and esponses o
aqua ic o ganisms. Ladu e al. [
81
] used ans e en opy o assess he esponse o ca eine-
ea ed zeb a ish o a eplica shoal o zeb a ish. The esul s showed ha , while he ans e
en opy was always highe om he eplica o he ali e ish ( o be expec ed, gi en ha
he eplica does no ha e he capaci y o eac o s imulus by i sel ), he di e ence was
signi ican only when he ish had been exposed o a leas 25–50 mg ca eine/mL and a
smalle dose o 5 mg/mL did no elici a esponse di e en om ha o he con ol.
Kane e al. [
82
] de eloped a ep oducible, eliable sys em, claimed o be capable o
de ec ing consis en beha iou al al e a ions o e ime caused by he exposu e o sub-le hal
doses o con aminan s and o he s esso s. Thei sys em was de eloped a e s udying
he beha iou o g oups o 12 mummichog killi ish (Fundulus he e ocli us) by compa ing
he ini ial 30 min base-line eco ding o he ish g oups p io o exposu e o 60 mg/L o
he commonly used aes he ic icaine me hanesul ona e (MS222), wi h he subsequen
eco dings o hei beha iou du ing he 90 min pos -exposu e. The exposed ish displayed
signi ican inc eases in pe cen mo emen , eloci y, dis ance om cen e, and ela i e bu s
equency while i s FD dec eased signi ican ly be ween he baseline and he subsequen
ime pe iods. Angula change and space u iliza ion we e no signi ican ly a ec ed. The
au ho s claimed ha wi h addi ional modi ica ions (including longe eco ding imes,
physiological adap a ion esponses o con aminan s and elimina ion o backg ound noise),
he sys em has he po en ial o p oduce quan i iable and au oma ed eco dings in eal
ime, sui able o de ec and quan i y beha iou al di e ences associa ed wi h exposu e o
addi ional en i onmen al pollu an s, including POPs, pes icides, ag icul u al was e, me als
and ha m ul oxins.
O high ele ance, due o he cu en gene al abuse and misuse o an ibio ics, is he wo k
o Su yan o e al. [
59
], who examined he e ec o exposu e o 100 ppb o 20 commonly used
an ibio ics (amikacin, amoxicillin, azi h omycin, ce u oxime, cip o loxacin, doxycycline,
e y h omycin, gen amycin, no loxacin, o loxacin, oxy e acycline, penicillin G po assium
sal , s ep omycin, sul ame hoxazole, sul ame hazine, sul apy idine, e acycline, ime ho-
p im, ylosin, and ancomycin) on he locomo ion complexi y o golden zeb a ish. F om
he h ee-dimensional (3D) ideo eco ding o he ish, hei swimming mo emen and
di ec ion we e analysed using idT acke [
92
] o iden i y swimming pa e ns and esponses
o he ish o he an ibio ics. The FD and en opy o he swimming pa e ns and esponses
we e calcula ed as desc ibed by Audi a e al. [
93
]. The FD alues om he amoxicillin-,
penicillin- and ylosin- ea ed g oups we e signi ican ly lowe han hose o he con ol
g oup. The en opy was also signi ican ly a ec ed in some cases: amikacin, o example,
signi ican ly inc eased he en opy o he beha iou (i.e., induced high andomness in
he mo ion), while oxy e acycline signi ican ly dec eased i ( he ish mo ed li le and
was concen a ed in a small a ea). P incipal componen analysis (PCA) and hie a chical
clus e ing analysis o he FD and en opy measu emen s we e applied o u he examine
he ela ionships be ween he e ec s o he an ibio ics and e ealed ha h ee an ibio ics
om h ee di e en classes (amoxicillin, a lac am; ime hop im, a sulphonamide; and
ylosin, a mac olide) we e clus e ed oge he and we e clea ly sepa a ed om he con ol
g oup and he o he an ibio ics. These h ee an ibio ics also had in common a much highe
e ec , no iceable e en a 1 ppb. O he 20 an ibio ics es ed, only azi h omycin, ce u oxime,
doxycycline and no loxacin did no cause al e a ions in he ish’s beha iou a he es ed
Fishes 2023,8, 311 16 o 25
Table A1. Con .
Re e ence Species Me hod Subjec o he S udy Main Findings
[80]
Zeb a ish (D. e io), Japanese medaka
(O yzias la ipes) and ed ca p
(Cyp inus ca pio).
Quan i ica ion o swimming beha iou al
changes induced upon exposu e o a
common de e gen (sodium dodecyl
benzene sul ona e, SDBS) by h ee
pa ame e s: swimming ail, swimming
speed, su ace beha iou and hei
uni o mi y assessed by hei
Shannon–Wea e en opy.
Assessmen o he uni o mi y o
indi idual ish swimming beha iou al
esponses o di e en
concen a ions o SDBS.
The en opy o he h ee species dec eased (i.e., hei
beha iou became less “no mal”) wi h inc easing
concen a ions o he de e gen , bu wi h di e en
sensi i i ies: zeb a ish was he mos sensi i e species,
esponding a he lowes concen a ion, ollowed by
ed ca p and medaka, which was he mos esis an .
S imulan s, anaes he ics and an ibio ics
[81] Zeb a ish (D. e io)The in e ac ion be ween he ish and a
eplica by ans e en opy.
E ec o ca eine on he in e ac ion
be ween indi idual zeb a ish and a
eplica o a shoal o conspeci ics.
The ans e en opy was always highe om he
eplica o he ish bu only in ish exposed o a leas
25–50 mg ca eine/mL he di e ence was signi ican .
[82]Mummichog killi ish
(Fundulus he e ocli us)
Veloci y, o al dis ance a elled, angula
change, pe cen mo emen , space
u iliza ion, and pa h complexi y (FD).
In- ank modelling o al e a ions in
indi idual ish beha iou exposed o
con aminan s and s esses o he sys em,
namely a e e ence oxican ( icane
me hanesul onae (m-aminobenzoic acid
e hyl es e me hansul ona e, MS222)
De elopmen o a emo ely con olled anspo able
sys em o de ec suble hal s ess- and
con aminan -induced beha iou in ish.
[59] Golden zeb a ish (D. e io)
3D ideo eco ding, idT acking he
ajec o ies and analysis o swimming,
explo a o y beha iou , ac al dimension
and pe mu a ion en opy o he
beha iou al da a, and PCA and
hie a chical clus e ing o
he FD and en opy da a.
Analysis o he e ec s on he beha iou
o ish in shoals o se en adul and
mixed-gende ish a e exposu e o 20
an ibio ics om eigh classes.
Only azi h omycin, ce u oxime, doxycycline and
no loxacin did no cause al e a ions in he ish
beha iou a he es ed 100 ppb. All o he an ibio ics
did cause changes in beha iou . Amoxicillin,
ime hop im, and ylosin caused al e a ions e en a
1ppb.
Hea y me als (Pb, Cu, Hg) and selenium (Se)
[83] Fa head minnows (Pimephales p omelas)
Video eco ding hei beha iou ,
calcula ing ou indi idual speci ic
ep oduc i e beha iou s and hei
ac al dimension.
E ec o he exposu e o imma u e
male– emale pai s o sub-le hal
concen a ions o Pb on hei indi idual
speci ic ep oduc i e beha iou .
Exposu e induces less complexi y (lowe FD) in he
ishes’ beha iou al sequences, bu only p io o
seconda y sexual cha ac e s being e iden .
[94]Medaka
(O. la ipes)
Video eco ding o indi idual and g oup
(n= 4) esponses (speed, Y-posi ion, s op
numbe , s op du a ion, u ning a e and
meande ing) hei mul i-laye pe cep ion
and FD. and hei upon exposu e
o Cu 1 mg/L.
E ec o he exposu e o sub-le hal
concen a ions o Cu on indi idual (MLP
and FD) and collec i e (FD) beha iou al
pa ame e s o assess hei use ulness o
de ec he con aminan .
Exposu e induces less complex beha iou and
signi ican ly dec eases he FD in bo h g oups and
indi idual beha iou s bu he esponse is mo e
consis en when analysing g oups o ish.
Fishes 2023,8, 311 17 o 25
Table A1. Con .
Re e ence Species Me hod Subjec o he S udy Main Findings
[49] Eu opean seabass (Dicen a chus lab ax).
The FD and SE o he ajec o y
ollowing he esponse o a s ochas ic
e en o ish ea ed wi h MeH+
4µg m/L o 9 days.
Changes in he shoals’ beha iou in
esponse o an e en (FD and SE) upon
exposu e o MeHg+in he wa e .
The Ka z–Cas iglioni FD and pa icula ly he SE we e
he mos sensi i e algo i hms o disc imina e he
esponses o MeHg+-con amina ed ish, indica ing a
po en ial alue o de elop a non-in asi e me hod o
he iden i ica ion and quan i ica ion o
beha iou al di e ences.
[85]Eu opean seabass
(D. lab ax)
SE o he schooling beha iou du ing
eco e y a e exposu e o MeHg+4µg
m/L o 9 days.
Quan i ica ion o he changes in he
shoals’ beha iou by i s SE, a e
exposu e o MeHg+in he wa e .
Du ing he 11 days pos -exposu e pe iod, he SE o
he con ol ish ended o inc ease, while he SE o
MeHg+- ea ed ish did no show a eco e y end.
[84] Eu opean seabass (D. lab ax)
SE o he shoaling and schooling
beha iou a e exposu e o (Na2SeO3)
10 µg/L o 6 days.
Changes in he shoals’ beha iou (by SE)
upon exposu e o sodium seleni e
(Na2SeO3) in he wa e o 6 days.
The SE o he schooling esponse o he exposed
g oup was only sligh ly lowe han ha o he
con ol g oup.
[42] Eu opean seabass (D. lab ax)
SE o he ideo eco ded ajec o y o he
shoal o ish ed wi h Se:Hg mola a ios
o 29.5, 6.6, 0.8 and 0.4 o 14 days.
Tes ing he e ec o eeding he ish
di e en mola Se:Hg a ios on he SE o
hei ajec o y.
The basal SE o ish ed wi h mola Se:Hg > 1 ended
o inc ease. The basal SE o ish ed wi h mola
Se:Hg < 1 ended o dec ease.
[86] Zeb a ish (D. e io)
Video eco ding o school beha iou al
esponses and calcula ion o he
Shannon–Wea e en opy o eigh
pa ame e s a e a 24 h exposu e o a low
dose (0.05 mg/L) o me cu ic chlo ide
(HgCl2).
Analyses o linea and non-linea
measu emen s o he collec i e
beha iou o exposed ish.
The use o eigh pa ame e s ( he en opy o
swimming speed, dep h du ing each 3 min in e al,
and changes in he sum en opy o speed, dep h,
u ning equency, dis ance and dispe sion) was
op imal o de ec low le els o he HgCl2
in 15–20 min.
[87]F eshwa e sh imps
(Mac ob achium jelskii)
Video eco ding o indi idual sh imps’
beha iou in g oups o h ee animals,
exposed o HgCl210 µg/L.
Tes ing whe he ma hema ical (linea
pa ame e s) and non-linea ( ac al,
in o ma ion en opy and mul i ac al
pa ame e s) me hods applied o ideo
acking o sh imps exposed o HgCl210
µ
g/L can adequa ely desc ibe changes in
hei locomo ion beha iou .
None o he me hods de ec ed he e ec o a 96 h
exposu e o 10 µg/L me cu ic chlo ide on ei he
linea o non-linea locomo ion pa ame e s.
Fishes 2023,8, 311 18 o 25
Table A1. Con .
Re e ence Species Me hod Subjec o he S udy Main Findings
Pes icides and pe sis en en i onmen al pollu an s
[87]F eshwa e sh imps
(Mac ob achium jelskii)
Video eco ding o indi idual sh imps’
beha iou in g oups o h ee animals,
exposed 0.15 µg/L o he
pes icide del ame h in.
Tes ing whe he ma hema ical (linea
pa ame e s) and non-linea ( ac al,
in o ma ion en opy and mul i ac al
pa ame e s) me hods applied o ideo
acking o sh imps exposed o 0.15
µg/L o he pes icide del ame h in can
adequa ely desc ibe changes in hei
locomo ion beha iou .
72 h o exposu e o 0.15
µ
g/L o del ame h in al e ed
he alues o some linea (e.g., he ack leng h) and
non-linea ( ac al dimension (box coun ing o
in o ma ion en opy) and mul i ac al analysis)
pa ame e s o hei locomo ion beha iou .
[88]Japanese medaka
(O. la ipes)
Res ing, swimming (in a s aigh line
and in ci cles) and SE o indi idual ish
placed in pai s in he anks and exposed
o ibu yl in (TBT), polychlo ina ed
biphenyls (PCBs), o a mix u e
(a 1 µg/g bw/day) o each o 3 weeks.
Changes in he beha iou o indi idual
ish e lec ed in he SE o he mean
swimming eloci y and he posi ion o
indi idual ish es ima ed om da a on
he es ing, swimming in a s aigh line
and swimming in ci cles.
PCBS induced swimming pa e ns consis en wi h
hype ac i i y and TBT inc eased he en opy
o ish’s posi ion.
[89]Japanese medaka
(O. La ipes)
Video eco ding o he swimming
ajec o ies o medaka ed (a 3% o bw)
he PCB Kanechlo -400 du ing 3 weeks
in doses up o 125 µg/g eed.
E ec s o Kanechlo -400 on he collec i e
beha iou o g oups o h ee sexually
imma u e o ish and in g oups o
mixed- ea ed and un ea ed ish.
Analysis o he ac al dimension o he
ajec o y and he ac al dimensions o
swimming eloci y and u ning angle.
Kanechlo -400 induced a sho ened schooling ime,
inc eased equency o beha iou al pa e n change
changed, disin eg a ion o schools and inc eased he
equency o collisions (hype ac i i y). When mixed
in he same g oup, Kanechlo 400-exposed medaka
in luenced he beha iou o unexposed ish in he
same school. FDs o he swimming ajec o y and
u ning angle signi ican ly inc eased bu only in he
highes PCB-exposed g oup.
Fishes 2023,8, 311 19 o 25
Table A2.
E ec s o con aminan s on he endencies o he F ac al dimension and en opy alues o indi idual o collec i e beha iou al pa e ns. FD, ac al
dimension; SE, Shannon en opy.
Re e ence Toxican FD/En opy Indi idual/Collec i e Beha iou Modi ica ion upon Exposu e
F ac al dimension analyses
[79] Suble hal NaClO and pH FD o swimming ajec o ies and FD o
swimming eloci ies. Indi idual The FD o he ajec o ies inc eased wi h pH while
he FD o he eloci ies inc eased wi h NaClO.
[89]
3 weeks exposu e o he PCB
Kanechlo -400 in he eed (3% o body
w and doses up o 125 µg/g eed).
E ec s o Kanechlo -400 on he se e al
pa ame e s o collec i e beha iou o g oups o
h ee sexually imma u e o ish. The ac al
dimension o he ajec o y and he ac al
dimensions o swimming
eloci y and u ning angle.
Collec i e (n= 3 imma u e ish)
The FDs o he swimming ajec o y and u ning
angle inc eased signi ican ly, bu only in he highes
PCB-exposed g oup.
[87]
Up o 96 h exposu e o HgCl
2
(10
µ
g/L).
T ack leng h, speed and D2P as well as he
non-linea ac al dimension, box coun ing o
in o ma ion en opy and mul i ac al
analysis me hods.
Indi idual beha iou in g oups (n= 3)
The mul i ac al na u e o locomo ion was ini ially
signi ican ly highe in HgCl2- ea ed ish. I
dec eased la e o he same alues as he con ol.
O he pa ame e s did no change.
[82] MS222 Indi idual The FD dec eased.
[83] Exposu e o 0.5 ppm Pb FD o speci ic ep oduc i e beha iou s. Indi idual beha iou in g oups (n= 2)
The FD dec eased (dec eased he complexi y o he
beha iou al ep oduc i e sequences) bu only be o e
seconda y sexual cha ac e s we e e iden .
[94] Cu, suble hal
FD (indi idual and g oup) o speed, Y-posi ion,
s op numbe , s op du a ion, u ning a e
and meande ing.
Indi idual and collec i e (n= 4)
The FD dec eased o bo h indi idual and collec i e
esponses.
The FD o g oup esponses we e less a iable han
hose o indi idual ish.
[87]Up o 96 h exposu e o del ame h in
(0.15 µg/L).
T ack leng h, speed and D2P as well as he
non-linea ac al dimension (es ima ed ia box
coun ing and in o ma ion en opy) and
mul i ac al analysis me hods.
Indi idual beha iou in g oups (n= 3) The ac al and mul i ac al dimensions o he
beha iou dec eased a e 72 h o exposu e.
F ac al and En opy analyses
[59]Acu e exposu e o 20 an ibio ics om 8
amilies.
FD and en opy o he swimming ajec o y o
he collec i e esponse. Collec i e (n= 7)
The FD dec eased upon exposu e o amoxicillin,
penicillin and ylosin ( he mos e ec i e an ibio ics).
Pe mu a ion en opy dec eased wi h oxy e acycline
and inc eased wi h amikacin.
Fishes 2023,8, 311 20 o 25
Table A2. Con .
Re e ence Toxican FD/En opy Indi idual/Collec i e Beha iou Modi ica ion upon Exposu e
[49] MeHg+4µg /L
The Higuchi FD [96], Ka z FD [97] Ka z-
Cas iglioni FD [98] and SE [54] and
Pe mu a ion en opy [99] o he ajec o y
ollowing he esponse o a s ochas ic e en o
ish in oxica ed wi h MeHg+in he wa e .
Collec i e (n= 81, 41)
Upon exposu e o MeHg
+
, he Higuchi FD su e ed a
small dec ease.; Ka z FD no change and
Ka z–Cas iglioni FD ended o inc ease; he SE
clea ly dec eased and pe mu a ion en opy showed
only a small dec ease.
En opy analyses
[81] Ca eine up o 50 mg/mL T ans e en opy be ween a zeb a ish o a
eplica shoal o zeb a ish. Indi idual
The ans e en opy om he eplica o he ali e ish
inc eased.
[88]
3 weeks exposu e o ibu yl in (TBT),
polychlo ina ed biphenyls (PCBs), o a
mix u e (a 1 µg/g bw/day o each
chemical).
The SE o he mean swimming eloci y and he
posi ion o indi idual ish es ima ed om da a
on he es ing, swimming in a s aigh line and
swimming in ci cles beha iou s o indi idual
ish.
Indi idual beha iou in g oups (n= 2) The SE o TBT- and PCB- ea ed ish inc eased, only
he o me did so signi ican ly.
[85]11 days ecupe a ion a e exposu e o
MeHg+4µg /L
SE o he con ol and MeHg+exposed ish
g oups, s essed by hal ing he wa e low
du ing he expe imen in bo h anks.
Collec i e (n= 26 and n= 19) The SE ended o inc ease in con ol ish g oup and
o dec ease sligh ly in he ea ed g oup.
[42]
14 days exposu e o eeds con aining
Se:Hg mola a ios > 1 (29.6 and 6.6) and
<1 (0.8 and 0.4).
SE o he shoaling (basal) and schooling
( esponse) beha iou s o he g oup o ish. Collec i e (n= 7)
The basal SE o ish ed wi h mola Se:Hg > 1 ended
o inc ease. The basal SE o ish ed wi h mola Se:Hg
< 1 ended o dec ease.
[86]
24 h exposu e o HgCl
2
(50
µ
g/L mg/L).
Eigh pa ame e s ( he en opy o swimming
speed, dep h du ing each 3 min in e al, and
changes in he sum en opy o speed, dep h,
u ning equency, dis ance and dispe sion)
was op imal o de ec low le els o he
con aminan in 15–20 min.
Collec i e (n= 5)
Shannon–Wea e en opy displayed s able alues
o e he p e-exposu e pe iod, a sudden signi ican
inc ease upon he addi ion o HgCl2 o 15 min and
hen a as dec ease o low alues.
[78]Le hal concen a ions o
phenol and KCN. SE o he e ical posi ion o he ish. Indi idual The SE dec eased wi h inc easing
amoun s o oxican .
[80] De e gen SDBS Shannon–Wea e en opy o swimming ail
and speed and hei su ace beha iou . Indi idual
The Shannon–Wea e en opy dec eased wi h
inc easing concen a ions o he de e gen wi h
species-speci ic sensi i i ies: zeb a ish was he mos
sensi i e o he h ee, ollowed by ed
ca p and medaka.
[84] Na2SeO3, 10 µg/L, 6 days SE o he schooling esponses o con ol and
ea ed g oups. Collec i e (n= 76) The SE o he ea ed g oup was only sligh ly lowe
han he SE o he con ol.
Fishes 2023,8, 311 21 o 25
Re e ences
1. FAO. The S a e o Wo ld Fishe ies and Aquacul u e 2022. SOFIA. Towa ds Blue T ans o ma ion; FAO: Rome, I aly, 2022.
2.
S en i o d, G.D.; S i unyalucksana, K.; Flegel, T.W.; Williams, B.; Wi hyachumna nkul, B.; I sa hi phaisa n, O.; Bass, D. New
Pa adigms o Help Sol e he Global Aquacul u e Disease C isis. PLoS Pa hog. 2017,13, e1006160. [C ossRe ] [PubMed]
3.
Su yan, R.M.; A imi su, M.L.; Cole i, H.A.; Hopc o , R.R.; Lindebe g, M.R.; Ba beaux, S.J.; Ba en, S.D.; Bu , W.J.; Bishop, M.A.;
Bodkin, J.L.; e al. Ecosys em esponse pe sis s a e a p olonged ma ine hea wa e. Sci. Rep.
2021
,11, 6235. [C ossRe ] [PubMed]
4.
San os, L.; Ramos, F. An imic obial esis ance in aquacul u e: Cu en knowledge and al e na i es o ackle he p oblem. In . J.
An imic ob. Agen s 2018,52, 135–143. [C ossRe ] [PubMed]
5. Led o d, H. An i-anxie y d ug makes i e ish mo e agg essi e. Na u e 2013,815, 814–815. [C ossRe ]
6.
Ma sson, K.; Johnson, E.V.; Malmendal, A.; Linse, S.; Hansson, L.-A.; Cede all, T. B ain damage and beha iou al diso de s in
ish induced by plas ic nanopa icles deli e ed h ough he ood chain. Sci. Rep. 2017,7, 11452. [C ossRe ] [PubMed]
7.
Alimi, O.S.; Fa ne Buda z, J.; He nandez, L.M.; Tu enkji, N. Mic oplas ics and Nanoplas ics in Aqua ic En i onmen s: Agg ega-
ion, Deposi ion, and Enhanced Con aminan T anspo . En i on. Sci. Technol. 2018,52, 1704–1724. [C ossRe ]
8. And ady, A.L. Mic oplas ics in he ma ine en i onmen . Ma . Pollu . Bull. 2011,62, 1596–1605. [C ossRe ]
9.
EATiP. A Re iew o he S a egic Resea ch and Inno a ion Agenda. Ou Vision o he Fu u e o Eu opean Aquacul u e. Eu opean
Aquacul u e Technology and Inno a ion Pla o m 2017. A ailable online: h ps://ea ip.eu/wp-con en /uploads/2018/02/
EATIP-SRIA-2017.pd (accessed on 11 May 2023).
10.
Science Ad ice o Policy by Eu opean Academies (SAPEA). A Sus ainable Food Sys em o he Eu opean Union; SAPEA: Be lin,
Ge many, 2020; A ailable online: h ps://www.sapea.in o/wp-con en /uploads/sus ainable- ood-sys em- epo .pd (accessed
on 11 May 2023)ISBN 978-3-9820301-7-3.
11.
Eu opean Commission; Eu opean Inno a ion Council and SMEs Execu i e Agency. Iden i ica ion o Eme ging Technologies
and B eak h ough Inno a ions. Lopa ka, M., Pól o a, A., Manimaa an, S., Eds.; Publica ions O ice o he Eu opean Union:
Luxenbou gh, 2022.
12.
Eu opean Commission 2020 S a egic Fo esigh Repo . Cha ing he Cou se Towa ds a Mo e Resilien Eu ope; 2020. A ailable online:
h ps://commission.eu opa.eu/sys em/ iles/2021-04/s a egic_ o esigh _ epo _2020_1_0.pd (accessed on 11 May 2023).
13. Holling, C.S. Resilience and S abili y o Ecological Sys ems. Annu. Re . Ecol. Sys . 1973,4, 1–23. [C ossRe ]
14.
Fø e, M.; F ank, K.; No on, T.; S endsen, E.; Al edsen, J.A.; Demps e , T.; Egui aun, H.; Wa son, W.; S ahl, A.; Sunde, L.M.; e al.
P ecision ish a ming: A new amewo k o imp o e p oduc ion in aquacul u e. Biosys . Eng. 2018,173, 176–193. [C ossRe ]
15.
A echa ala-Lopez, P.; Cab e a-Ál a ez, M.J.; Maia, C.M.; Sa ai a, J.L. En i onmen al en ichmen in ish aquacul u e: A e iew o
undamen al and p ac ical aspec s. Re . Aquac. 2021,14, 704–728. [C ossRe ]
16.
Ba e o, M.O.; Planellas, S.R.; Yang, Y.; Phillips, C.; Desco ich, K. Eme ging indica o s o ish wel a e in aquacul u e. Re . Aquac.
2021,14, 343–361. [C ossRe ]
17.
Li, D.; Wang, G.; Du, L.; Zheng, Y.; Wang, Z. Recen ad ances in in elligen ecogni ion me hods o ish s ess beha io . Aquac.
Eng. 2021,96, 102222. [C ossRe ]
18.
Ma ques, T.A.; Thomas, L.; Ma in, S.W.; Mellinge , D.K.; Wa d, J.A.; Mo e i, D.J.; Ha is, D.; Tyack, P.L. Es ima ing animal
popula ion densi y using passi e acous ics. Biol. Re . 2012,88, 287–309. [C ossRe ] [PubMed]
19.
Fø e, M.; S endsen, E.; Al edsen, J.; Uglem, I.; Bloeche , N.; S eie , H.; Sunde, L.; F ank, K. Using acous ic eleme y
o moni o he e ec s o c owding and delousing p ocedu es on a med A lan ic salmon (Salmo sala ). Aquacul u e
2018
,
495, 757–765. [C ossRe ]
20.
Rose, C.S.; S one , A.W.; Ma eson, K. Use o high- equency imaging sona o obse e ish beha iou nea bai ed ishing gea s.
Fish. Res. 2005,76, 291–304. [C ossRe ]
21.
Ges o, M.; Zupa, W.; Al onso, S.; Spedica o, M.T.; Lembo, G.; Ca bona a, P. Using acous ic eleme y o assess beha io al esponses
o acu e hypoxia and ammonia exposu e in a med ainbow ou o di e en compe i i e abili y. Appl. Anim. Beha . Sci.
2020
,
230, 105084. [C ossRe ]
22.
Hal o sen, M.B.; Zeddies, D.G.; Ellison, W.T.; Chicoine, D.R.; Poppe , A.N. E ec s o mid- equency ac i e sona on hea ing in
ish. J. Acous . Soc. Am. 2012,131, 599–607. [C ossRe ]
23.
K uusmaa, M.; Gkli a, R.; Tuh an, J.A.; Tu ikene, A.; Al edsen, J.A. Salmon beha iou al esponse o obo s in an aquacul u e sea
cage. R. Soc. Open Sci. 2020,7, 191220. [C ossRe ]
24.
K uusmaa, M.; Rieucau, G.; Mon oya, J.C.C.; Ma kna, R.; Handega d, N.O. Collec i e esponses o a la ge macke el school
depend on he size and speed o a obo ic ish bu no on ail mo ion. Bioinspi a ion Biomim. 2016,11, 056020. [C ossRe ]
25.
Zhao, S.; Zhang, S.; Liu, J.; Wang, H.; Zhu, J.; Li, D.; Zhao, R. Applica ion o machine lea ning in in elligen ish aquacul u e: A
e iew. Aquacul u e 2021,540, 736724. [C ossRe ]
26.
Gladju, J.; Kamalam, B.S.; Kanaga aj, A. Applica ions o da a mining and machine lea ning amewo k in aquacul u e and
ishe ies: A e iew. Sma Ag ic. Technol. 2022,2, 100061. [C ossRe ]
27.
Sabe ioon, M.; Gholizadeh, A.; Cisa , P.; Pau sina, A.; U ban, J. Applica ion o machine ision sys ems in aquacul u e wi h
emphasis on ish: S a e-o - he-a and key issues. Re . Aquac. 2016,9, 369–387. [C ossRe ]
Fishes 2023,8, 311 22 o 25
28.
Yang, L.; Liu, Y.; Yu, H.; Fang, X.; Song, L.; Li, D.; Chen, Y. Compu e Vision Models in In elligen Aquacul u e wi h Emphasis on
Fish De ec ion and Beha io Analysis: A Re iew. A ch. Compu . Me hods Eng. 2021,28, 2785–2816. [C ossRe ]
29.
Wei, Y.; Wei, Q.; An, D. In elligen moni o ing and con ol echnologies o open sea cage cul u e: A e iew. Compu . Elec on.
Ag ic. 2020,169, 105119. [C ossRe ]
30.
Sadoul, B.; Mengues, P.E.; F iggens, N.; P une , P.; Colson, V. A new me hod o measu ing g oup beha iou s o ish shoals om
eco ded ideos aken in nea aquacul u e condi ions. Aquacul u e 2014,430, 179–187. [C ossRe ]
31.
Deakin, A.G.; Spence , J.W.; Cossins, A.R.; Young, I.S.; Sneddon, L.U. Wel a e Challenges In luence he Complexi y o Mo emen :
F ac al Analysis o Beha iou in Zeb a ish. Fishes 2019,4, 8. [C ossRe ]
32.
Zhao, J.; Bao, W.; Zhang, F.; Zhu, S.; Liu, Y.; Lu, H.; Shen, M.; Ye, Z. Modi ied mo ion in luence map and ecu en neu-
al ne wo k-based moni o ing o he local unusual beha io s o ish school in in ensi e aquacul u e. Aquacul u e
2018
,
493, 165–175. [C ossRe ]
33.
Van De Vis, J.W.; Poelman, M.; Lambooij, E.; Begou , M.-L.; Pila czyk, M. Fish wel a e assu ance sys em: Ini ial s eps o se up an
e ec i e ool o sa egua d and moni o a med ish wel a e a a company le el. Fish Physiol. Biochem.
2012
,38, 243–257. [C ossRe ]
34. Ashley, P.J. Fish wel a e: Cu en issues in aquacul u e. Appl. Anim. Beha . Sci. 2007,104, 199–235. [C ossRe ]
35.
Ko e, S.M.; Oli ie , B.; Koolhaas, J.M. A new animal wel a e concep based on allos asis. Physiol. Beha .
2007
,92, 422–428. [C ossRe ]
36.
S e ling, P.; Eye , J. Allos asis: A New Pa adigm o Explain A ousal Pa hology. In Handbook o Li e S ess, Cogni ion and Heal h;
Fishe , S., Reason, J., Eds.; John Wiley & Sons L d.: Hoboken, NJ, USA, 1988; pp. 629–639. ISBN 0471912697.
37.
K is iansen, T.S.; Mada o, A.; S ien, L.H.; B acke, M.B.M.; Noble, C. Theo e ical Basis and P inciples o Wel a e Assessmen o Fa med
Fish, 1s ed.; Else ie Inc.: Ams e dam, The Ne he lands, 2020; Volume 38.
38. Bonga, S.E.W. The s ess esponse in ish. Physiol. Re . 1997,77, 591–625. [C ossRe ] [PubMed]
39.
Rapini, R.; Ma azza, G. Elec ochemical ap asenso s o con aminan s de ec ion in ood and en i onmen : Recen ad ances.
Bioelec ochemis y 2017,118, 47–61. [C ossRe ]
40.
Ejeian, F.; E edali, P.; Mansou i-Teh ani, H.-A.; Soozanipou , A.; Low, Z.-X.; Asadnia, M.; Tahe i-Ka ani, A.; Razmjou, A.
Biosenso s o was ewa e moni o ing: A e iew. Biosens. Bioelec on. 2018,118, 66–79. [C ossRe ] [PubMed]
41.
Fe nández-Baca, C.P.; Spi i o, C.M.; Bae, J.S.; Szegle es, Z.M.; Ba o , N.; Sausele, D.J.; B ooks, Y.M.; Welle , D.L.; Richa dson, R.E.
Rapid qPCR-Based Wa e Quali y Moni o ing in New Yo k S a e Rec ea ional Wa e s. F on . Wa e 2021,3, 711477. [C ossRe ]
42.
Egui aun, H.; Casque o, O.; Ma inez, I. The Shannon En opy T end o a Fish Sys em Es ima ed by a Machine Vision App oach
Seems o Re lec he Mola Se:Hg Ra io o I s Feed. En opy 2018,20, 90. [C ossRe ] [PubMed]
43.
Bae, M.-J.; Pa k, Y.-S. Biological ea ly wa ning sys em based on he esponses o aqua ic o ganisms o dis u bances: A e iew. Sci.
To al. En i on. 2014,466–467, 635–649. [C ossRe ] [PubMed]
44.
B odin, T.; Pio ano, S.; Fick, J.; Klaminde , J.; Heynen, M.; Jonsson, M. Ecological e ec s o pha maceu icals in aqua ic sys ems—
Impac s h ough beha iou al al e a ions. Philos. T ans. R. Soc. B Biol. Sci. 2014,369, 20130580. [C ossRe ] [PubMed]
45.
Wang, B.; Zhu, J.; Wang, A.; Wang, J.; Wu, Y.; Yao, W. Ea ly de ec ion o cyanide, o ganophospha e and oden icide pollu ion
based on locomo o ac i i y o zeb a ish la ae. Pee J 2021,9, e12703. [C ossRe ]
46.
Egui aun, H.; Ma inez, I. En opy and F ac al Techniques o Moni o ing Fish Beha iou and Wel a e in Aquacul u al P ecision
Fish Fa ming—A Re iew. En opy 2023,25, 559. [C ossRe ]
47.
Egui aun, H.; Izagi e, U.; Ma inez, I. A pa adigm shi in sa e sea ood p oduc ion: F om con aminan de ec ion o ish
moni o ing—Applica ion o biological wa ning sys ems o aquacul u e. T ends Food Sci. Technol. 2015,43, 104–113. [C ossRe ]
48.
Egui aun, H.; Casque o, O.; Sø ensen, A.J.; Ma inez, I. Reducing he Numbe o Indi iduals o Moni o Shoaling Fish
Sys ems—Applica ion o he Shannon En opy o Cons uc a Biological Wa ning Sys em Model. F on . Physiol.
2018
,
9, 493. [C ossRe ] [PubMed]
49.
Egui aun, H.; López-De-Ipiña, K.; Ma inez, I. Applica ion o En opy and F ac al Dimension Analyses o he Pa e n Recogni ion
o Con amina ed Fish Responses in Aquacul u e. En opy 2014,16, 6133–6151. [C ossRe ]
50.
Ga ilescu, M.; Demne o á, K.; Aamand, J.; Aga hos, S.; Fa a, F. Eme ging pollu an s in he en i onmen : P esen and u u e
challenges in biomoni o ing, ecological isks and bio emedia ion. New Bio echnol. 2015,32, 147–156. [C ossRe ]
51.
B ausch, J.M.; Conno s, K.A.; B ooks, B.W.; Rand, G.M. Human Pha maceu icals in he Aqua ic En i onmen : A Re iew o Recen
Toxicological S udies and Conside a ions o Toxici y Tes ing. Re . En i on. Con am. Toxicol. 2012,218, 1–99. [C ossRe ]
52.
S eele, C.W. Open ield explo a o y beha iou o ish: An unde u ilized ool o beha iou al oxicology. Ma . Pollu . Bull.
1983
,
14, 124–125. [C ossRe ]
53.
Klee ekope , H. E ec s o Suble hal Concen a ions o Pollu an s on he Beha io o Fish. J. Fish. Res. Boa d Can.
1976
,
33, 2036–2039.
54. Shannon, C.E. A Ma hema ical Theo y o Communica ion. Bell Sys . Tech. J. 1948,27, 379–423. [C ossRe ]
55. Shannon, C.E.; Wea e , W.W. The Ma hema ical Theo y o Communica ions; Uni e si y o Illinois P ess: U bana, IL, USA, 1963.
56.
Espino, M.; Egui aun, H.; de Ce io, O.D.; Ca e o, J.A.; E xeba ia, N.; Ma inez, I. An ioxidan Ac i i ies and Selenogene
T ansc ip ion in he Eu opean Sea Bass (Dicen a chus lab ax) Li e Depend, in a Non-linea Manne , on he Se/Hg Mola Ra io o
he Feeds. Biol. T ace Elemen Res. 2021,200, 2365–2379. [C ossRe ]
Fishes 2023,8, 311 23 o 25
57.
Dø ing, K.B. Assessmen o animal beha iou as a me hod o indica e en i onmen al oxici y. Comp. Biochem. Physiol. Pa C
Comp. Pha macol. 1991,100, 247–252. [C ossRe ] [PubMed]
58.
Jacquin, L.; Pe i jean, Q.; Cô e, J.; La aille, P.; Jean, S. E ec s o Pollu ion on Fish Beha io , Pe sonali y, and Cogni ion: Some
Resea ch Pe spec i es. F on . Ecol. E ol. 2020,8, 86. [C ossRe ]
59.
Su yan o, M.E.; Yang, C.-C.; Audi a, G.; Vasquez, R.D.; Roldan, M.J.M.; Ge , T.-R.; Hsiao, C.-D. E alua ion o Locomo ion
Complexi y in Zeb a ish a e Exposu e o Twen y An ibio ics by F ac al Dimension and En opy Analysis. An ibio ics
2022
,
11, 1059. [C ossRe ]
60.
Okoye, C.O.; Addey, C.I.; Ode inde, O.; Oko o, J.O.; Uwamungu, J.Y.; Ikechukwu, C.K.; Okeke, E.S.; Eje omedoghene, O.; Odii,
E.C. Toxic Chemicals and Pe sis en O ganic Pollu an s Associa ed wi h Mic o-and Nanoplas ics Pollu ion. Chem. Eng. J. Ad .
2022,11, 100310. [C ossRe ]
61.
Llo ca, M.; Fa é, M. Cu en Insigh s in o Po en ial E ec s o Mic o-Nanoplas ics on Human Heal h by in- i o Tes s. F on .
Toxicol. 2021,3, 752140. [C ossRe ]
62.
Weilga , L. The impac s o an h opogenic ocean noise on ce aceans and implica ions o managemen . Can. J. Zool.
2007
,85,
1091–1116. [C ossRe ]
63.
He be -Read, J.E.; K eme , L.; B uin jes, R.; Rad o d, A.N.; Ioannou, C.C. An h opogenic Noise Pollu ion om Pile-D i ing
Dis up s he S uc u e and Dynamics o Fish Shoals. P oceeding R. Soc. B 2017,84, 20171627I. [C ossRe ] [PubMed]
64.
Cos a, M.; Goldbe ge , A.L.; Peng, C.-K. Mul iscale en opy analysis o biological signals. Phys. Re . E S a . Nonlin. So Ma e .
Phys. 2005,71, 021906. [C ossRe ] [PubMed]
65. Ki ano, H. Compu a ional Sys ems Biology. Na u e 2002,420, 206–210. [C ossRe ]
66.
Spasic, S.; Sa ic, A.; Nikolic, L.; Budimi , S.; Janose ic, D.; Mi o ic, A. Applica ions o Higuchi’s F ac al Dimension in he
Analysis o Biological Signals. In P oceedings o he 2012 20 h Telecommunica ions Fo um, TELFOR, Belg ade, Se bia, 20–22
No embe 2012; pp. 639–641. [C ossRe ]
67.
Mandelb o , B. How Long Is he Coas o B i ain? S a is ical Sel -Simila i y and F ac ional Dimension. Science
1967
,156, 636–638. [C ossRe ]
68. Zmeskal, O.; Dzik, P.; Vesely, M. En opy o ac al sys ems. Compu . Ma h. Appl. 2013,66, 135–146. [C ossRe ]
69.
Ki h, K.; Sou ina, O.; Kulish, V.; Khoa, N.M. An Algo i hm o F ac al Dimension Calcula ion Based on Renyi En opy o Sho
Time Signal Analysis. In P oceedings o he 2009 7 h In e na ional Con e ence on In o ma ion, Communica ions and Signal
P ocessing, ICICS 2009, Macau, China, 8–10 Decembe 2009; pp. 1–5. [C ossRe ]
70.
Ezeiza, A.; de Ipiña, K.L.; He nández, C.; Ba oso, N. Enhancing he Fea u e Ex ac ion P ocess o Au oma ic Speech Recogni ion
wi h F ac al Dimensions. Cogn. Compu . 2012,5, 545–550. [C ossRe ]
71.
Sekine, M.; Tamu a, T.; Akay, M.; Fujimo o, T.; Togawa, T.; Fukui, Y. Disc imina ion o walking pa e ns using wa ele -based
ac al analysis. IEEE T ans. Neu al Sys . Rehabil. Eng. 2002,10, 188–196. [C ossRe ]
72. Inada, Y.; Kawachi, K. O de and Flexibili y in he Mo ion o Fish Schools. J. Theo . Biol. 2002,214, 371–387. [C ossRe ]
73.
Tikhono , D.; Ende lein, J.; Malchow, H.; Med insky, A.B. Chaos and ac als in ish school mo ion. Chaos Soli ons F ac als
2001
,
12, 277–288. [C ossRe ]
74. Tikhono , D.; Malchow, H. Chaos and ac als in ish school mo ion, II. Chaos Soli ons F ac als 2003,16, 287–289. [C ossRe ]
75.
Alados, C.L.; Escós, J.M.; Emlen, J.M. F ac al s uc u e o sequen ial beha iou pa e ns: An indica o o s ess. Anim. Beha .
1996
,
51, 437–443. [C ossRe ]
76. Mann, R.P.; Ga ne , R. The en opic basis o collec i e beha iou . J. R. Soc. In e ace 2015,12, 20150037. [C ossRe ] [PubMed]
77.
Xia, C.; Fu, L.; Liu, Z.; Liu, H.; Chen, L.; Liu, Y. Aqua ic Toxic Analysis by Moni o ing Fish Beha io Using Compu e Vision: A
Recen P og ess. J. Toxicol. 2018,2018, 2591924. [C ossRe ] [PubMed]
78.
Fukuda, S.; Kang, I.J.; Mo oishi, J.; Nakamu a, A. The applica ion o en opy o de ec ing beha io al esponses in Japanese
medaka (O yzias la ipes) exposed o di e en oxican s. En i on. Toxicol. 2010,25, 446–455. [C ossRe ] [PubMed]
79.
Nimke dphol, K.; Nakagawa, M. E ec o sodium hypochlo i e on zeb a ish swimming beha io es ima ed by ac al dimension
analysis. J. Biosci. Bioeng. 2008,105, 486–492. [C ossRe ]
80.
Zhang, Y.; Ma, J.; Zhou, S.; Ma, F. Concen a ion-dependen oxici y e ec o SDBS on swimming beha io o eshwa e ishes.
En i on. Toxicol. Pha macol. 2015,40, 77–85. [C ossRe ]
81.
Ladu, F.; Mwa o, V.; Li, J.; Mac ì, S.; Po i i, M. Acu e ca eine adminis a ion a ec s zeb a ish esponse o a obo ic s imulus.
Beha . B ain Res. 2015,289, 48–54. [C ossRe ] [PubMed]
82.
Kane, A.S.; Salie no, J.D.; Gipson, G.T.; Mol eno, T.C.; Hun e , C. A ideo-based mo emen analysis sys em o quan i y beha io al
s ess esponses o ish. Wa e Res. 2004,38, 3993–4001. [C ossRe ] [PubMed]
83.
Alados, C.L.; Webe , D.N. Lead e ec s on he p edic abili y o ep oduc i e beha io in a head minnows (pimephales p omelas):
A ma hema ical model. En i on. Toxicol. Chem. 1999,18, 2392–2399. [C ossRe ] [PubMed]
84.
Egui aun, H.; Lopez-De-Ipiña, K.; Ma inez, I. E olu ion o Shannon En opy in a Fish Sys em (Eu opean Seabass,Dicen a chus
lab ax) du ing Exposu e o Sodium Seleni e (Na
2
SeO
3
). In P oceedings o he 2nd In e na ional Elec onic Con e ence on En opy
and i s Applica ions, Online, 15–30 No embe 2015; MDPI: Basel, Swi ze land, 2015. Sci o um Elec onic Con e ence Se ies,
Session Complex Sys ems (C006). Volume 2. [C ossRe ]
85.
Egui aun, H.; López-De-Ipiña, K.; Ma inez, I. Shannon En opy in a Eu opean Seabass (Dicen a chus lab ax) Sys em du ing he
Ini ial Reco e y Pe iod a e a Sho -Te m Exposu e o Me hylme cu y. En opy 2016,18, 209. [C ossRe ]
Fishes 2023,8, 311 24 o 25
86.
Huang, Y.; Zhang, J.S.; Mi, F.J.; Zhang, G.H.; Sun, J. Moni o ing low-le el me cu y con amina ion by zeb a ish school beha io al
esponses. IOP Con . Se ies Ea h En i on. Sci. 2020,612, 012077. [C ossRe ]
87.
Teno io, B.M.; Filho, E.A.D.S.; Nei a, G.S.M.; da Sil a, V.A.; Teno io, F.D.C.A.M.; Sil a, T.D.J.D.; e Sil a, E.C.S.; Noguei a, R.D.A.
Can ac al me hods applied o ideo acking de ec he e ec s o del ame h in pes icide o me cu y on he locomo ion beha io
o sh imps? Eco oxicol. En i on. Sa . 2017,142, 243–249. [C ossRe ]
88.
Nakayama, K.; Oshima, Y.; Hi ama su, K.; Honjo, T. Al e a ion o Gene al Beha io o Male Medaka, o yzias la ipes, Exposed o
T ibu yl in and/o Polychlo ina ed Biphenyls. J. Fac. Ag ic. Kyushu Uni . 2004,49, 85–92. [C ossRe ]
89.
Nakayama, K.; Oshima, Y.; Hi ama su, K.; Shimasaki, Y.; Honjo, T. E ec s o polychlo ina ed biphenyls on he schooling beha io
o japanese medaka (O yzias la ipes). En i on. Toxicol. Chem. 2005,24, 2588–2593. [C ossRe ]
90.
Zhang, G.; Chen, L.; Chen, J.; Ren, Z.; Wang, Z.; Chon, T.-S. E idence o he S epwise Beha io al Response Model
(SBRM): The e ec s o Ca bama e Pes icides on medaka (O yzias la ipes) in an online moni o ing sys em. Chemosphe e
2012
,
87, 734–741. [C ossRe ]
91.
Nakagawa, M. A C i ical Exponen Me hod o E alua e F ac al Dimensions o Sel -A ine Da a. J. Phys. Soc. Jpn.
1993
,
62, 4233–4239. [C ossRe ]
92.
Pé ez-Escude o, A.; Vicen e-Page, J.; Hinz, R.C.; A ganda, S.; de Pola ieja, G.G. idT acke : T acking indi iduals in a g oup by
au oma ic iden i ica ion o unma ked animals. Na . Me hods 2014,11, 743–748. [C ossRe ]
93.
Audi a, G.; Su yan o, M.E.; Chen, K.H.-C.; Vasquez, R.D.; Roldan, M.J.M.; Yang, C.-C.; Hsiao, C.-D.; Huang, J.-C. Acu e and
Ch onic E ec s o Fin Ampu a ion on Beha io Pe o mance o Adul Zeb a ish in 3D Locomo ion Tes Assessed wi h F ac al
Dimension and En opy Analyses and Thei Rela ionship o Fin Regene a ion. Biology 2022,11, 969. [C ossRe ]
94.
Ji, C.W.; Lee, S.K.H.; Kwak, I.S.; Cha, E.Y.; Lee, S.K.H.; Chon, T.S. Compu a ional Analysis o Mo emen Beha io s o Medaka
(O yzias la ipes) a e he T ea men s o Coppe by Using F ac al Dimension and A i icial Neu al Ne wo ks. WIT T ans. Biomed.
Heal h 2006,10, 93–107. [C ossRe ]
95.
Higuchi, T. App oach o an i egula ime se ies on he basis o he ac al heo y. Phys. D Nonlinea Phenom.
1988
,
31, 277–283. [C ossRe ]
96. Ka z, M.J. F ac als and he analysis o wa e o ms. Compu . Biol. Med. 1988,18, 145–156. [C ossRe ] [PubMed]
97.
Cas iglioni, P. Wha is w ong in Ka z’s me hod? Commen s on: “A no e on ac al dimensions o biomedical wa e o ms”. Compu .
Biol. Med. 2010,40, 950–952. [C ossRe ]
98.
Band , C.; Pompe, B. Pe mu a ion En opy: A Na u al Complexi y Measu e o Time Se ies. Phys. Re . Le .
2002
,88, 174102.
[C ossRe ] [PubMed]
99.
Gan he , H.E.; Goudie, C.; Sunde, M.L.; Kopecky, M.J.; Wagne , P.; Oh, S.-H.; Hoeks a, W.G. Selenium: Rela ion o Dec eased
Toxici y o Me hylme cu y Added o Die s Con aining Tuna. Science 1972,175, 1122–1124. [C ossRe ]
100.
Rals on, N.V.C.; Rals on, C.R.; Raymond, L.J. Selenium Heal h Bene i Values: Upda ed C i e ia o Me cu y Risk Assessmen s.
Biol. T ace Elemen Res. 2015,171, 262–269. [C ossRe ] [PubMed]
101.
Rals on, N.V.; Raymond, L.J. Me cu y’s neu o oxici y is cha ac e ized by i s dis up ion o selenium biochemis y. Biochim. Biophys.
Ac a Gen. Subj. 2018,1862, 2405–2416. [C ossRe ]
102.
Rals on, N.V.C.; Blackwell, J.L.; Raymond, L.J. Impo ance o Mola Ra ios in Selenium-Dependen P o ec ion Agains Me hylme -
cu y Toxici y. Biol. T ace Elemen Res. 2007,119, 255–268. [C ossRe ]
103.
Yamashi a, M.; Imamu a, S.; Yamashi a, Y. Me hylme cu y and Selenium in Sea ood. Kagaku Seibu su
2012
,50, 807–817. [C ossRe ]
104.
Yamashi a, M.; Yamashi a, Y.; Suzuki, T.; Kani, Y.; Mizusawa, N.; Imamu a, S.; Takemo o, K.; Ha a, T.; Hossain, A.; Yabu, T.;
e al. Selenoneine, a No el Selenium-Con aining Compound, Media es De oxi ica ion Mechanisms agains Me hylme cu y
Accumula ion and Toxici y in Zeb a ish Emb yo. Ma . Bio echnol. 2013,15, 559–570. [C ossRe ] [PubMed]
105.
Yamashi a, Y.; Yabu, T.; Yamashi a, M. Disco e y o he s ong an ioxidan selenoneine in una and selenium edox me abolism.
Wo ld J. Biol. Chem. 2010,1, 144–150. [C ossRe ] [PubMed]
106.
Cos a, E.V.L.; Noguei a, R.D.A. Mul i ac al dimension and lacuna i y o yolk sac ascula u e a e exposu e o magne ic ield.
Mic o asc. Res. 2015,99, 1–7. [C ossRe ] [PubMed]
107.
Be nho , R.A. Me cu y Toxici y and T ea men : A Re iew o he Li e a u e. J. En i on. Public Heal h
2012
,2012, 460508. [C ossRe ]
108.
Lombana, D.A.B.; Mac ì, S.; Po i i, M. Collec i e Emo ional Con agion in Zeb a ish. F on . Beha . Neu osci.
2021
,15, 730372. [C ossRe ]
109.
Millo , S.; Bégou , M.-L.; Cha ain, B. Explo a ion beha iou and ligh esponse owa d a s imulus in h ee sea bass s ains
(Dicen a chus lab ax L.). Appl. Anim. Beha . Sci. 2009,119, 108–114. [C ossRe ]
110.
Millo , S.; Ce quei a, M.; Cas anhei a, M.-F.; Ø e li, Ø.; Oli ei a, R.F.; Ma ins, C.I.M. Beha iou al S ess Responses P edic
En i onmen al Pe cep ion in Eu opean Sea Bass (Dicen a chus lab ax). PLoS ONE 2014,9, e108800. [C ossRe ]
111.
Millo , S.; Péan, S.; Labbé, L.; Ke neis, T.; Quille , E.; Dupon -Ni e , M.; Begou , M.-L. Assessmen o Gene ic Va iabili y o Fish
Pe sonali y T ai s using Rainbow T ou Isogenic Lines. Beha . Gene . 2014,44, 383–393. [C ossRe ]
112.
Ko e , R.; Vainikka, A.; Janhunen, M.; Pii onen, J.; Hy ä inen, P. Beha io al a ia ion shows he i abili y in ju enile b own ou
Salmo u a. Beha . Ecol. Sociobiol. 2014,68, 927–934. [C ossRe ]
113.
Giacomini, A.C.V.V.; de Ab eu, M.S.; Koakoski, G.; Idalêncio, R.; Kalichak, F.; Oli ei a, T.A.; da Rosa, J.G.S.; Gusso, D.; Pia o,
A.L.; Gil Ba cellos, L.J. My s ess, ou s ess: Blun ed co isol esponse o s ess in isola ed housed zeb a ish. Physiol. Beha .
2015
,
139, 182–187. [C ossRe ] [PubMed]
Fishes 2023,8, 311 25 o 25
114.
Villegas-Ríos, D.; Réale, D.; F ei as, C.; Moland, E.; Olsen, E.M. Indi idual le el consis ency and co ela ions o ish spa ial
beha iou assessed om aqua ic animal eleme y. Anim. Beha . 2017,124, 83–94. [C ossRe ]
115.
Mikhee , V.N.; Pas e nak, A.F.; Taskinen, J. Pe sonali y In luences Risk o Pa asi ism in Fish. Dokl. Biol. Sci.
2019
,488, 141–144. [C ossRe ]
Disclaime /Publishe ’s No e:
The s a emen s, opinions and da a con ained in all publica ions a e solely hose o he indi idual
au ho (s) and con ibu o (s) and no o MDPI and/o he edi o (s). MDPI and/o he edi o (s) disclaim esponsibili y o any inju y o
people o p ope y esul ing om any ideas, me hods, ins uc ions o p oduc s e e ed o in he con en .
