Behaviour in order to evaluate the palatability of pet food in domestic cats

Full text

Applied
Animal
Behaviour
Science
159
(2014)
55–61
Contents
lists
available
at
ScienceDirect
Applied
Animal
Behaviour
Science
jou
rn
al
hom
epage
:
w
ww.elsevier.com/locate/applanim
Behaviour
in
order
to
evaluate
the
palatability
of
pet
food
in
domestic
cats
Aurélie
Becquesa,∗,
Claire
Larosea,
Céline
Barona,
Cécile
Nicerona,
Christophe
Féronb,
Patrick
Gouatb
aSPF
Diana,
ZA
du
Gohélis,
56250
Elven,
France
bLaboratoire
d’Éthologie
Expérimentale
et
Comparée
EA
4443,
Université
Paris
13,
Sorbonne
Paris
Cité,
Avenue
JB
Clément,
93430
Villetaneuse,
France
a
r
t
i
c
l
e
i
n
f
o
Article
history:
Accepted
22
July
2014
Available
online
1
August
2014
Keywords:
Cats
Pet
food
Feeding
behaviour
Palatability
a
b
s
t
r
a
c
t
Palatability
of
pet
food
has
been
mainly
assessed
by
intake
ratios.
In
the
present
study
we
have
searched
for
behavioural
clues
of
food
palatability
in
domestic
cats
Felis
catus.
Two
diets
differing
in
palatability
(Very
Palatable
Kibbles
and
Low
Palatable
Kibbles)
were
evaluated
by
a
panel
of
17
cats
using
an
automated
feeding
station
and
video
recordings.
The
cats
tested
each
diet
in
two
different
sessions,
with
only
one
diet
during
a
given
session.
A
session
lasted
for
two
consecutive
days
with
food
continuously
available
during
20
h
per
24
h
period.
At
each
of
their
visit
to
the
feeding
station,
the
quantity
of
food
eaten
by
a
cat,
the
speed
of
consumption
and
the
latency
to
eat
were
recorded.
The
behaviour
of
the
cat
was
also
analysed
for
each
visit.
All
the
cats
made
at
least
four
visits
to
the
feeding
station
during
a
24
h
period.
We
compared
the
different
quantitative
variables
between
the
two
diets
for
the
first
three
visits
and
for
the
last
visit
of
each
of
the
two
days
of
a
session.
Our
results
showed
that,
as
expected,
cats
ate
more
VPK
than
LPK.
Addressing
behavioural
patterns,
the
length
of
sniffing
was
significantly
reduced
with
VPK
on
the
two
first
visits
of
the
first
day,
suggesting
less
hesitation
in
this
situation.
Neither
the
latency
nor
the
speed
of
consumption
was
affected
by
the
palatability
of
the
kibbles.
©
2014
Elsevier
B.V.
All
rights
reserved.
1.
Introduction
Food
is
a
basic
biological
commodity
and
taste
is
a
powerful
stimulus
which
can
elicit
either
positive
or
nega-
tive
reaction.
Studies
have
shown
that
specific
behaviours
(i.e.
facial
and
body
reaction)
can
be
expressed
in
rela-
tion
to
the
taste
of
the
food.
In
rats
(Grill
and
Norgren,
1978)
and
in
primates
(Steiner
et
al.,
2001)
a
sucrose
∗Corresponding
author.
Tel.:+33
0
2
97
93
20
20;
fax:
+33
297938041.
E-mail
addresses:
abecques@diana-petfood.com,
aurelie.becques@gmail.com
(A.
Becques).
solution
elicited
tongue
protrusion
and
mouth
move-
ments
whereas
quinine
solution
elicited
gapes,
chin
rubs,
headshakes
and
forelimb
flailing.
In
humans,
sweet
foods
elicited
positive
or
hedonic
patterns
of
lip
smacking
and
tongue
protrusion,
accompanied
by
relaxation
of
muscles
of
the
middle
face,
and
an
occasional
smile.
On
the
other
hand
bitter
quinine
elicited
negative
or
aversive
gapes,
and
complex
grimaces
involving
retraction
of
the
lips,
frowning
of
the
brows
and
muscles
around
the
eyes,
and
wrinkling
of
the
nose
(Schaal
et
al.,
2000;
Steiner
et
al.,
2001).
The
domestic
cat
is,
with
the
dog,
the
most
representa-
tive
pet
in
households
(Murray
et
al.,
2010).
Their
feeding
http://dx.doi.org/10.1016/j.applanim.2014.07.003
0168-1591/©
2014
Elsevier
B.V.
All
rights
reserved.

56
A.
Becques
et
al.
/
Applied
Animal
Behaviour
Science
159
(2014)
55–61
Fig.
1.
Automated
feeding
system
(patent
Larose
2007).
Description
of
feeding
system
used:1—first
flap;
2—corridor;
3—second
flap;
4—RFID
system;
5—third
and
last
flap;
6—bowls;
7—feeding
area.
behaviour
is
unique
in
regard
to
circadian
rhythm,
cats
being
intermittent
feeders.
They
have
several
short
feed-
ing
periods,
referred
thereafter
as
meals,
throughout
the
24-h
period
and
there
is
no
direct
relationship
between
the
size
of
a
meal
and
its
timing
(Mugford,
1977;
Thorne,
1982).
Cats
may
have
11
to
16
meals
per
day
with
5
to
7
g
eaten
per
meal
(Kane
et
al.,
1981,
1987).
Cats
are
strictly
carnivore
with
specific
needs
in
nutrients;
they
are
selec-
tive
in
food
and
prefer
to
consume
freshly
killed
carcasses
like
mice,
rats,
rabbits,
birds,
lizards,
insects
and
other
ani-
mals
rather
than
carrion
(Bontempo,
2005;
Bradshaw
et
al.,
1996;
Bradshaw,
2006;
Watson,
2011).
It
is,
therefore,
a
challenge
for
industrials
to
find
the
equilibrium
between
palatability
and
the
nutritional
quality
of
a
diet
for
cats.
It
does
not
matter
how
well-formulated
a
diet
if
cats
will
not
eat
it.
To
investigate
the
role
of
palatability,
Petfood
industries
usually
use
expert
panels
of
cats
trained
to
discriminate
food
with
different
sensory
properties.
In
the
usual
test-
ing
procedure,
cats
have
the
choice
between
two
different
diets
presented
simultaneously
and
available
for
several
hours.
The
quantity
of
each
food
eaten
and
the
dynamic
of
the
consumption
of
each
food
during
the
test
are
used
as
indices
of
food
palatability.
The
aim
of
our
study
was
to
enrich
the
classical
palatability
criteria
by
looking
for
dif-
ferences
in
the
behaviour
of
cats
which
could
be
attributed
to
palatability
perception.
To
our
knowledge
only
one
study
has
examined
the
behaviour
of
cats
during
a
food
test
(Van
den
Bos
et
al.,
2000).
In
their
study,
cats
were
presented
successively
two
canned
diets
differing
in
their
level
of
palatability.
The
complete
duration
of
each
test
session
did
not
exceed
30
min
and
the
cats
were
under
their
usual
diet
the
rest
of
the
time.
The
authors
then
described
a
“hedo-
nic”
taste
reactivity
pattern
where
cats
licked
and
sniffed
the
feeding
bowl,
licked
their
lips
and
groomed
their
face,
and
an
“aversive”
taste
reactivity
pattern
where
cats
licked
and
sniffed
the
food
and
licked
their
nose.
Based
on
these
results
we
decided
to
study
the
behaviour
of
cats
at
feed-
ing
occasions
using
two
different
dry
food
diets
between
which
a
known
difference
of
palatability
has
been
estab-
lished
using
the
classical
expert
panel
evaluation.
Dry
food
is
strategic
for
petfooders
because
it
is
the
most
popu-
lar
form
of
diet
given
and
bought
by
pet
owners
(Crane
et
al.,
2000).
In
order
to
respect
the
natural
feeding
rhythm
of
cats,
each
test
lasted
20
h
with
the
diet
continuously
available.
2.
Methods
2.1.
Animals
The
experiments
were
conducted
at
Panelis,
Diana
Pet
Food
Division
expert
panels
in
palatability
measurement.
This
center
specialized
in
the
evaluation
of
food
palata-
bility
for
dogs
and
cats
is
modelled
on
the
real-life
home
environment
and
is
committed
to
the
well-being
of
pets
and
to
the
expertise
in
palatability
measurement
and
to
the
observation
of
the
feeding
behaviour
of
dogs
and
cats.
Cats
were
recruited
between
two
months
and
three
months
of
age
from
breeders
and
private
owners.
Kittens
were
always
adopted
by
two,
a
male
and
a
female
from
the
same
litter.
The
selected
kittens
were
in
good
health
and
toler-
ant
towards
both
conspecifics
and
human
contact.
From
recruitment
until
approximately
eight
months
of
age,
cats
get
accustomed
to
their
new
environment
with
other
kit-
tens
in
our
catteries
and
to
be
in
contact
with
different
persons.
They
were
trained
with
our
experimental
proce-
dures
(see
below)
and
tasted
different
types
of
food
from
economical
to
super
premium
brands.
The
population
of
cats
kept
in
the
expert
panel
is
representative
of
cat
popu-
lation
in
France
(approximately
80%
of
European
cats
and
20%
of
purebred
cats;
FACCO/TNS
SOFRES
survey,
2012)
and
balanced
in
terms
of
age
(from
8
months
to
10
years)
and
sex.
Because
pet
owners
frequently
sterilize
males
(81%)
and
females
(75%)
(FACCO/TNS
SOFRES
survey,
2012),
the
males
of
the
expert
panel
were
castrated
at
puberty
and
most
of
the
females
were
sterilized.
In
the
present
study
seventeen
adult
cats
(5
±
1.5
years
old)
have
been
used
and
six
of
them
were
purebred
cats
(one
British
Shortair,
one
Somali,
one
Chartreux
and
three
Armenian
Van).
The
cats
were
distributed
into
two
groups,
housed
in
two
separate
rooms
several
months
before
the
beginning
of
the
experi-
ment.
The
rooms
were
equivalent
in
surface
(13
m2indoor
and
outside)
and
equipment;
three
litters
per
room,
toys
and
cat
trees.
Cats
had
a
free
access
to
an
enriched
outdoor
courtyard.
In
our
study,
Room
1
contained
three
males
and
six
females
(all
sterilized);
Room
2
contained
three
steril-
ized
males
and
five
females
with
only
one
intact
female.
2.2.
Automated
feeding
system
Each
room
is
equipped
with
two
automated
feeding
stations
(Fig.
1,
Larose
et
al.,
2007).
At
each
visit
and

A.
Becques
et
al.
/
Applied
Animal
Behaviour
Science
159
(2014)
55–61
57
Table
1
List
and
description
of
the
variables
taken
into
account
in
each
session.
Variable
(unit
of
measurement)
Description
Consumption
(g)
Total
amount
of
food
eaten
Speed
of
consumption
(g
min−1)
Weight
of
kibbles
eaten
divided
by
the
time
the
cat
spent
eating
(i.e.
ingesting
and
chewing)
Latency
to
eat
(s)
Defined
as
the
latency
between
the
opening
of
the
feeding
area
flap
by
the
cat
and
the
first
eating
of
kibbles
Number
of
visits
per
session The
number
of
visits
per
day
and
the
time
of
each
visit
were
recorded
Eating
in
a
sitting
position
(%) Time
spent
to
eat
in
a
sitting
position
divided
by
the
time
spent
eating.
In
a
sitting
position
the
cat
was
resting
on
its
two
hind
legs
with
the
tail
down
and
the
two
front
legs
were
straight.
During
the
remainder
of
the
time
the
cat
was
eating
standing
up
with
its
four
legs
straight
Sniffing
food
(%)
Time
spent
sniffing
the
food
divided
by
the
time
spent
around
the
bowl.
When
the
cat
was
near
the
bowl
(i.e.
head
at
less
than
6
cm
from
the
edge
of
the
bowl)
it
was
considered
to
be
sniffing
when
its
head
was
tilted
with
the
nose
at
less
than
3
cm
from
the
bowl
Licking
(nb
min−1)
The
number
of
licks
per
min
spent
inside
the
feeding
station.
Because
the
difference
between
the
licking
of
the
lips
and
the
licking
of
the
nose
could
not
always
be
verified,
particularly
during
the
night
period,
the
global
number
of
licks
on
the
face
was
quantified
whatever
the
target
before
entering
into
the
feeding
station,
a
RFID
(radio
frequency
identification)
system
automatically
identifies
cats
by
their
own
microchip
(Metal
Process,
Montrévrain,
France)
inserted
in
their
collar.
Only
one
cat
at
a
time
can
be
inside
the
feeding
station.
The
area
becomes
free
as
soon
as
the
cat
is
detected
leaving
the
feeding
station.
Accord-
ing
to
the
feeding
habits
of
cats,
the
device
gives
a
free
access
to
the
food
every
day
from
11:15
h
am
to
07:15
h
am
(i.e.
20
h
per
24
h).
The
food
is
placed
into
two
stain-
less
steel
bowls
at
the
centre
of
the
feeding
area.
A
scale
is
placed
under
each
bowl
to
measure
the
consumption
of
a
cat
at
each
visit.
Two
feeding
stations
per
room
are
suf-
ficient
to
allow
an
easy
access
to
food
to
all
cats
of
the
group.
All
feeding
stations
are
connected
to
a
computer
system
(Dell
Inc.,
Round
Rock,
Texas,
United
States)
record-
ing
varied
data
about
cats’
visits:
time,
duration
and
food
consumption.
2.3.
Diets
The
two
diets
used
in
our
experiments
were
made
of
the
same
kind
of
kibbles
and
differed
only
in
the
coat-
ing
and
other
additions
(see
below).
To
avoid
a
kibble
and
a
batch
effect,
only
Royal
CaninTM kibbles,
as
a
kib-
ble
basis,
were
used
and
all
the
kibbles
were
coated
at
the
same
time
with
an
identical
quantity
and
nature
of
fat
(6%
of
poultry
fat).
One
half
of
the
kibbles
was
coated
with
one
palatability
enhancer
and
the
other
half
with
a
sec-
ond
one,
during
two
minutes
by
using
a
mixer
(FORBERG
International
AS)
dedicated
to
the
coating
of
kibbles.
The
food
was
then
put
into
plastic
bags
hermetically
sealed
by
a
specific
machine
and
kept
at
20 ◦C
for
two
weeks
before
consumption.
Two
different
palatability
enhancers
applied
on
the
kibbles
were
chosen:
one
very
palatable
(VPK)
was
a
super-premium
hydrolyzate
with
a
poultry
basis;
the
other
one
was
less
palatable
(LPK)
and
was
a
standard
hydrolyzate
with
a
viscera
basis.
According
to
previous
preference
tests
cats
tended
to
eat
significantly
more
VPK
than
LPK
(Diana
Pet
Food
division,
unpublished
results).
The
cats
included
in
the
present
study
have
expe-
rienced
the
two
types
of
palatability
enhancers
prior
to
the
experiment.
The
difference
between
VPK
and
LPK
is
based
only
on
the
palatability
produced
by
the
coating,
the
kibble
base
being
the
same
and
no
nutritional
differences
exist-
ing
in
terms
of
caloric
value
and
nutritional
components
(3567
kcal
kg−1).
To
further
increase
the
palatability
of
the
VPK
food,
180
g
commercially
canned
tuna
(Entire
natural
tuna
Petit
NavireTM)
were
mixed
to
the
kibbles
just
before
the
beginning
of
test
(in
the
following
proportions
by
weight:
83%
of
kibbles
and
17%
of
tuna
in
the
bowl).
Mixing
the
tuna
only
slightly
decreased
the
caloric
value
of
the
diet
(3203
kcal
kg−1).
It
was
the
first
time
cats
tested
tuna
mixed
with
their
kibbles
and
this
diet
will
be
referred
to
as
VPK
thereafter.
2.4.
Recording
procedure
In
order
to
link
specific
behaviour
to
a
specific
diet,
we
chose
to
present
only
one
diet
per
session.
A
session
lasted
two
consecutive
days.
A
first
session
with
VPK
and
a
second
with
LPK
were
planned
in
each
group.
On
each
day
the
food
was
available
to
cats
during
a
20
h
period.
These
two
sessions
were
included
within
a
testing
time
table
where
cats
were
made
to
test
different
diets
accord-
ing
to
the
requirements
of
customers.
A
group
had
two
days
of
testing
between
the
first
and
the
second
session
whereas
the
other
group
had
a
ten
day
period
of
testing
between
the
two
sessions.
The
cats
experienced
different
diets
on
a
day-by-day
basis
which
limited
the
interac-
tions
between
the
two
successive
diets.
Before
each
test,
the
area
and
the
two
stainless
steel
bowls
were
washed
and
dried.
Before
the
beginning
of
the
test,
a
video
camera
(Panasonic
Inc.,
Osaka,
Japan)
with
dual
mode
colour
and
monochrome
CCD
high
quality
was
positioned
on
the
outside
and
in
front
of
each
device
in
order
to
record
the
behaviour
of
cats
during
their
meal.
Each
camera
had
a
zoom
lens
with
manual
iris
3.5–8
mm.
To
be
able
to
record
day
and
night,
two
infrared
spotlights
(Ans-
mann,
Assamstadt,
Germany)
were
positioned
near
the
camera.
Each
camera
was
connected
to
a
computer
(Dell
Inc,
Round
Rock,
Texas,
United
States)
storing
digital
images
using
the
MPEG
Recorder
Software
(©Noldus
Information
Technology,
version
10.1,
Wageningen,
Netherlands).

58
A.
Becques
et
al.
/
Applied
Animal
Behaviour
Science
159
(2014)
55–61
Table
2
Comparison
of
each
behavioural
criteria
(Mean
±
standard
error
of
mean
[SEM])
analysed
by
visit
and
day
for
each
diet.
A
paired
t-test
was
used
to
compare
the
two
diets.
The
significant
values
are
in
bold.
Variables Day Visits VPK
LPK NP-value
Mean
±
SEM
Mean
±
SEM
Consumption
(g) D1 V1
21.1
±
2.0
3.4
±
0.9
17
<0.001
V2
9.4
±
1.3 4.3
±
1.1 17
0.006
V3
7.6
±
1.3 2.6
±
0.6 17
0.005
F
4.8
±
1.2
7.1
±
0.8
17
0.100
D2 V1
15.4
±
1.8
4.7
±
0.8
17
<0.001
V2
12.2
±
1.4
4.8
±
0.8
17
0.001
V3
11.1
±
1.8
5.7
±
0.7
17
0.013
F
9.7
±
1.3
5.8
±
0.9
17
0.021
Speed
of
consumption
(g
min−1)
D1 V1
4.8
±
0.5
3.8
±
0.7
17
0.228
V2
4.1
±
0.5
4.6
±
0.8
13
0.730
V3
4.2
±
0.6
4.5
±
0.6
12
0.928
F
3.4
±
0.5
4.3
±
0.4
12
0.061
D2 V1
5.2
±
0.7 4.2
±
0.4
15
0.210
V2
5.0
±
0.6
4.2
±
0.4
15
0.243
V3
5.3
±
0.5
4.9
±
0.9
16
0.730
F
4.9
±
0.6
3.8
±
0.4
15
0.156
Latency
to
eat
(s) D1 V1
14.9
±
1.6
14.5
±
2.7
17
0.904
V2
17.0
±
2.3 18.0
±
4.2 13
0.902
V3
17.7
±
2.5
18.6
±
2.9
12
0.696
F
22.8
±
3.4
11.2
±
0.9
12
0.001
D2 V1
13.3
±
1.8
12.4
±
1.3
15
0.602
V2
15.4
±
2.0
13.2
±
1.1
15
0.391
V3
13.1
±
1.8
21.7
±
9.5
16
0.573
F
19.3
±
3.7
12.1
±
1.3
15
0.103
Eat
in
a
sitting
position
(%)
D1 V1
63.2
±
9.7
42.6
±
9.1
17
0.145
V2
69.8
±
0.7
63.7
±
10.4
13
0.523
V3
61.9
±
9.5
49.2
±
11.7
12
0.446
F
65.6
±
10.4
79.0
±
6.6
12
0.223
D2 V1
76.8
±
8.6
60.5
±
9.8
15
0.121
V2
80.5
±
6.9
74.3
±
8.4
15
0.524
V3
73.2
±
8.6
70.5
±
8.8
16
0.808
F
77.0
±
8.5
81.6
±
6.1
15
0.635
Sniffing
food
(%) D1 V1
6.6
±
1.1
24.4
±
4.8
17
<0.001
V2
9.9
±
2.1
20.2
±
4.0
17
0.011
V3
14.2
±
4.8
30.1
±
7.4
17
0.108
F
16.5
±
5.5
10.9
±
2.7
17
0.413
D2 V1
9.6
±
3.0
12.5
±
2.9
17
0.509
V2
8.9
±
2.6
9.7
±
2.9
17
0.823
V3
9.9
±
2.5
10.8
±
3.0
17
0.830
F
9.6
±
1.6
6.1
±
1.0
17
0.027
Licking
(nb
min−1)D1 V1
5.9
±
0.8
5.7
±
0.6
17
0.894
V2
4.6
±
0.5
4.5
±
0.7
17
0.854
V3
4.7
±
0.6
4.6
±
0.7
17
0.905
F
2.5
±
0.6
4.2
±
0.4
17
0.018
D2 V1
4.8
±
0.5
5.1
±
0.6
17
0.622
V2
4.9
±
0.7
5.0
±
0.6
17
0.816
V3
4.9
±
0.6
5.0
±
0.6
17
0.901
F
3.7
±
0.5
3.6
±
0.5
17
0.789
2.5.
Data
observed
All
the
video-recordings
were
analysed
by
the
same
observer,
using
The
Observer
XT
Software
(©
Noldus
Information
Technology,
version
10.1,
Wageningen,
Netherlands).
The
variables
measured
during
each
visit
of
a
cat
to
the
feeding
station
are
listed
in
Table
1.
2.6.
Statistical
analysis
As
cats
made
at
least
four
visits
per
day,
the
first
three
visits
(V1,
V2,
V3)
and
the
last
one
(FV)
were
selected
in
order
to
describe
the
complete
feeding
period.
The
comparisons
between
the
two
diets
were
made
independently
for
each
visit
on
each
day.
A
cat
entering
in
the
feeding
station
usually
ate
some
food.
In
some
occa-
sions,
a
cat
that
has
entered
in
the
station
did
not
eat
any
food.
In
this
event
all
the
variables
linked
with
consumption
(i.e.
speed
of
consumption,
latency
to
eat,
eat
in
a
sitting
position)
cannot
be
calculated.
As
a
consequence
these
data
and
their
paired
data
(i.e.
same
cat,
same
day
and
same
visit
but
other
diet)
were
discarded
from
the
statistical
analysis.
The
mean
and
SEM
were
calculated
accordingly
and
the
sample
size
was
added
for
each
comparison
in
Table
2.
We
used
non-parametric
statistics
because
the
sam-
ple
size
was
small
(N
≤
17).
Data
were
compared
between
the
two
diets
using
permutation
tests
for
paired

A.
Becques
et
al.
/
Applied
Animal
Behaviour
Science
159
(2014)
55–61
59
samples
(StatXact,
Cytel
Software
Corporation,
Cambridge,
MA,
U.S.A.).
All
effects
were
evaluated
at
the
˛
=
0.05
level.
3.
Results
3.1.
Consumption
As
expected
from
Diana
Pet
Food
division
(unpub-
lished
results),
the
cats
ate
significantly
more
food
per
day
(80.66
g
±
4.92
g)
during
the
VPK
session
than
during
the
LPK
session
(52.91
g
±
2.66,
P
=
0.0008).
On
day
1,
the
cats
ate
significantly
more
VPK
than
LPK
during
the
first
three
visits
(Table
2).
This
difference
was
not
observed
during
the
last
visit.
On
day
2,
the
differences
of
consumption
between
the
two
diets
were
always
significant
whenever
the
visit.
No
significant
differences
were
observed
between
the
two
diets
for
the
other
two
variables
related
to
food
consump-
tion
(i.e.
speed
of
consumption
and
latency
to
eat,
Table
2).
Cats
did
significantly
fewer
visits
per
day
(8.74
±
0.44)
during
the
VPK
session
than
during
the
LPK
session
(11.44
±
0.92;
P
=
0.0002).
The
time
of
the
different
visits
did
not
differ
significantly
between
the
two
sessions
in
the
majority
of
cases
(Fig.
2).
The
only
significant
differ-
ences
occurred
on
day
2.
The
third
visit
was
earlier
for
the
LPK
than
for
the
VPK
diet
(P
=
0.0089)
whereas
the
inverse
was
true
for
the
last
visit,
earlier
for
the
VPK
than
for
the
LPK
session
(P
=
0.0281).
The
duration
between
two
visits
was
2.5
±
0.05
h
and
varied
from
six
minutes
to
more
than
eleven
hours.
3.2.
Behaviour
3.2.1.
Eating
in
a
sitting
position
The
proportion
of
eating
time
spent
in
a
sitting
position
did
not
differ
significantly
between
the
two
diets
(Table
2).
3.2.2.
Sniffing
the
food
On
day
1
the
cats
spent
more
time
sniffing
the
LPK
food
than
the
VPK
food
when
they
were
around
the
bowl.
The
difference
was
significant
only
for
the
two
first
visits.
On
the
second
day
there
were
no
more
differences
during
the
first
three
visits
whereas
during
the
last
visit
the
cats
spent
less
time
sniffing
the
LPK
food
than
the
VPK
food
when
they
were
around
the
bowl
(Table
2).
3.2.3.
Licking
behaviour
There
was
no
significant
difference
in
the
number
of
licks
in
proportion
to
the
time
spent
in
the
feeding
station
with
the
exception
of
the
last
visit
of
day
1
(Table
2),
where
more
licks
occurred
during
the
LPK
test.
4.
Discussion
In
our
experiment
the
two
diets
were
proposed
individ-
ually
to
the
cats
during
two
different
sessions.
Compared
to
the
study
by
Van
den
Bos
et
al.
(2000)
where
cats
had
a
food
access
during
30
min,
in
our
study
they
had
free
access
to
the
food
during
20
h.
This
situation
corresponded
to
the
usual
situation
of
a
domestic
cat
at
home.
Under
these
conditions
the
cats
could
display
a
stronger
affinity
for
one
food
rather
than
for
another
even
if
this
attractive-
ness
could
vary
between
visits
at
the
feeding
station.
Two
main
findings
resulted
from
the
present
study.
As
expected,
the
consumption
of
kibbles
was
higher
during
the
VPK
session
than
during
the
LPK
one.
The
dif-
ference
in
consumption
was
also
significant
for
the
first
three
visits
on
each
of
the
two
days
of
a
session.
On
the
last
visit
of
day
1,
nevertheless,
the
difference
was
not
sig-
nificant
anymore.
As
the
food
was
available
during
a
20
h
period,
one
may
think
that
a
decrease
in
the
freshness
of
the
VPK
diet
could
be
involved.
During
the
last
visit
of
day
2,
however,
the
difference
of
consumption
between
the
two
diets
was
still
observed.
If
the
attraction
of
the
diet
decreased
with
time
the
effect
should
have
been
higher
for
LPK
since
the
last
visits
occurred
later
than
during
the
VPK
session
on
day
2.
A
regulation
process
of
food
intake
might
also
have
caused
this
result.
Cats
are
able
to
reg-
ulate
their
consumption
but
this
regulation
is
a
complex
phenomenon
and
is
still
poorly
understood.
Kane
et
al.
(1981)
have
demonstrated
that
cats
offered
commercial
cat
food
with
different
caloric
content,
quickly
adjusted
the
amount
of
food
ingested
to
maintain
stable
their
calorie
intake.
Thorne
(1982)
also
used
commercial
cat
food
with
different
water
contents
and
found
evidence
for
an
energy
intake
regulation.
In
our
study
cats
ate
more
VPK
than
LPK
during
a
session
and
the
difference
of
consumption
clearly
over-compensated
the
difference
in
caloric
value
of
the
VPK
diet
(average
calorie
intake/session:
VPK
=
258.3
kcal
and
LPK
=
188.73
kcal).
If
a
regulation
of
energy
intake
might
have
caused
the
lack
of
a
significant
difference
of
consump-
tion
during
the
last
visit
of
day
1,
the
higher
palatability
of
VPK
was
clearly
demonstrated
by
the
higher
consumption
of
this
diet
during
the
two
days
of
the
session.
We
could
not
exclude,
nevertheless,
that
a
regulation
process
of
food
intake
might
have
taken
place
over
a
longer
period.
Among
the
behavioural
variables,
only
the
time
spent
sniffing
the
food
while
the
cat
was
near
the
bowl,
revealed
clear
significant
results.
The
sense
of
smell
is
powerful
and
important
in
cats
and
plays
a
major
role
in
food
selection.
Perinatal
experiences
have
been
shown
to
influence
prefer-
ence
for
a
specific
diet.
Two-day-old
kittens
tend
to
prefer
an
odour
previously
encountered
in
utero,
through
the
mother’s
diet,
rather
than
an
unknown
odour.
Later
they
prefer
to
eat
the
food
odorized
with
a
flavour
experimented
perinatally
than
a
food
odorized
with
an
unknown
flavour
(Becques
et
al.,
2009).
When
an
adult
cat
is
eating,
blowing
a
succession
of
attractive
food
odours
increased
the
dura-
tion
of
the
meal
and
the
quantity
of
food
ingested
(Mugford,
1977;
Bradshaw,
1986).
In
our
study,
the
behaviour
of
cats
differed
between
the
two
diets
in
the
proportion
of
time
spent
sniffing
during
the
two
first
visits
on
day
1.
This
dif-
ference
disappeared
for
the
other
visits
on
day
1
suggesting
that
the
novelty
of
the
situation
was
involved.
The
first
day
of
a
session
corresponded
to
a
change
in
the
diet
of
the
cats.
One
may
have
expected
that
the
novelty
of
the
diet
should
have
caused
more
sniffing.
On
the
contrary
the
cats
tended
to
sniff
more
LPK,
a
diet
that
they
have
already
experienced,
than
VPK
that
they
have
previously
experienced
but
with-
out
the
addition
of
tuna.
The
tuna
was
very
odorant
and
it
seemed
that
this
odour
was
attractive
enough
to
elicit
eat-
ing
in
a
short
lapse
of
time.
On
the
other
hand
the
longer

60
A.
Becques
et
al.
/
Applied
Animal
Behaviour
Science
159
(2014)
55–61
Fig.
2.
Comparison
of
the
time
of
the
different
visits
(V1
to
VF)
between
the
two
sessions.
The
time
of
a
visit
is
expressed
in
hours
and
corresponded
to
the
elapsed
time
between
the
beginning
of
the
test
and
the
given
visit.
Cats
(n
=
17)
had
a
free
access
to
the
food
by
using
a
feeding
station.
Black
bars:
VPK
session;
empty
bars:
LPK
session.
A
session
lasted
for
two
days
(D1
and
D2).
A
paired
t-test
was
used
to
compare
the
two
sessions.
NS:
no
significant
result,
*:
P
<
0.05, **:
P
<
0.01.
duration
of
sniffing
the
LPK
diet
may
correspond
to
a
hes-
itation
to
consume
a
less
palatable
diet.
On
the
last
visit
of
day
2,
nevertheless,
the
cats
spent
less
time
sniffing
the
LPK
than
the
VPK
diet.
In
fact,
there
was
no
major
change
in
the
sniffing
duration
of
the
VPK
diet
whereas
the
sniffing
duration
of
the
LPK
diet
reached
its
lowest
level
suggest-
ing
a
kind
of
resignation
by
the
cats.
In
conclusion,
a
low
amount
of
time
spent
sniffing
the
food
seems
therefore
to
be
a
reliable
behavioural
indicator
of
the
palatability
of
food
when
newly
proposed
to
cats.
Other
behavioural
variables
had
been
measured
but
they
did
not
differ
clearly
between
the
two
diets.
The
speed
of
consumption
in
cats
in
a
given
state
of
hunger
has
been
shown
to
be
greater
for
a
high
palatability
diet
at
least
at
the
beginning
of
a
meal
(Foucault,
1992).
In
our
experi-
ment,
neither
the
speed
of
consumption
nor
the
latency
to
eat,
differed
significantly
between
the
two
diets.
The
cats
have
access
to
food
during
20
h
every
day
with
a
constant
time
schedule.
As
a
consequence
the
cats
have
experienced
that
food
was
always
available
during
a
test
session.
At
the
beginning
of
a
test,
after
four
hours
without
any
food,
cats
were
not
very
hungry.
Our
data
revealed
that
during
a
ses-
sion
the
time
between
two
successive
meals
could
exceed
ten
hours
even
though
food
was
available,
and
some
cats
had
their
first
meal
two
hours
or
more
after
the
bowls
were
refilled.
Moreover,
being
alone
inside
the
feeding
station
prevented
any
direct
social
pressure,
and
a
cat
could
take
its
time
to
eat.
These
reasons
might
explain
the
lack
of
any
differences
in
the
speed
of
consumption
and
the
latency
to
eat.
Licking
behaviour
did
not
reveal
any
clear
differ-
ences
between
the
two
diets.
In
their
study,
Van
den
Bos
et
al.
(2000)
have
related
licking
behaviour
to
food
palata-
bility.
A
difference
was
made
between
licking
the
nose,
related
as
an
aversive
taste
reactivity
pattern,
and
licking
of
lips
which
appeared
more
after
a
more
palatable
wet
food.
We
were
unable
to
clearly
distinguish
these
two
types
of
licking
behaviour
on
videos.
This
might
have
caused
the
lack
of
a
behavioural
difference
between
the
two
diets.
A
video
recording
system,
focussed
on
the
head
of
the
cat,
would
surely
have
allowed
a
more
precise
analysis
of
the
facial
mimics
displayed
by
the
cat
when
eating
and
after.
More
details
of
the
eating
behaviour
could
also
have
been
recorded
using
a
transparent
feeding
bowl
or
a
plate
with-
out
any
edge.
Bradshaw
and
Cook
(1996)
have
demonstrated
in
their
study
that
cats
display
a
variety
of
signals
(visual,
vocal,
tactile
and
olfactory)
just
before
and
after
the
meal.
In
our
protocol
a
meal
corresponded
to
the
time
spent
inside
the
feeding
station.
In
the
future,
it
would
be
interesting
to
complete
this
study
with
observations
performed
just
before
and
just
after
the
period
spent
inside
the
feeding
sta-
tion.
Moreover,
our
subject
cats
live
in
a
social
group,
and
the
social
transmission
of
information
about
the
diet
may
also
have
an
important
influence
on
food
attractiveness
as
shown
in
the
rat
(Galef,
1993).
Finally
our
study
suggested
that
behavioural
criteria,
such
as
the
time
spent
sniffing
the
food,
can
be
used
to
assess
the
palatability
of
food.
Such
criteria
can
be
easily
implemented
by
professionals
and
by
pet
owners.
We
think
that
behavioural
criteria
might
reveal
to
be
more
relevant
than
the
simple
measurement
of
food
consumption
to
collect
the
cat’s
assessment
of
a
diet.
Acknowledgement
We
thank
Lavinia
Bruneau
for
the
revision
of
the
English
version
of
the
manuscript.
References
Becques,
A.,
Larose,
C.,
Gouat,
P.,
Serra,
J.,
2009.
Effect
of
pre-
and
postnatal
olfactogustatory
experience
on
early
preferences
at
birth
and
dietary
selection
at
weaning
in
kittens.
Chem.
Senses
35,
41–45.

A.
Becques
et
al.
/
Applied
Animal
Behaviour
Science
159
(2014)
55–61
61
Bradshaw,
J.W.S.,
1986.
Mere
exposure
reduces
cat’s
neophobia
to
unfa-
miliar
food.
Anim.
Behav.
34,
613–614.
Bradshaw,
J.W.S.,
Goodwin,
D.,
Legrand-Defrétin,
V.,
Nott,
H.M.R.,
1996.
Food
selection
by
the
domestic
cat,
an
obligate
carnivore.
Comp.
Biochem.
Physiol.,
A
114
(3),
205–209.
Bradshaw,
J.W.S.,
Cook,
S.E.,
1996.
Patterns
of
pet
cat
behaviour
at
feeding
occasions.
Appl.
Anim.
Behav.
Sci.
47,
61–74.
Bradshaw,
J.W.S.,
2006.
The
evolutionary
basis
for
the
feeing
behavior
of
domestic
dogs
(Canis
familiaris)
and
cats
(Felis
catus).
J.
Nutr.
136,
1927S–1931S.
Bontempo,
V.,
2005.
Nutrition
and
health
of
dogs
and
cats:
evolution
of
petfood.
Vet.
Res.
Commun.
29,
45–50.
Crane,
S.W.,
Griffin,
R.W.,
Peter,
R.M.,
2000.
In:
Hand,
M.S.,
Thatcher,
C.D.,
Remillard,
R.L.,
Roudebush,
P.
(Eds.),
Introduction
to
Commercial
Pet
Foods.
Chapter
3
in
Small
Animal
Clinical
Nutrition.
,
fourth
ed.
Mark
Morris
Institute,
Topeka,
KS.
Foucault,
V.,
1992.
Contribution
à
l’étude
du
comportement
alimentaire
du
chat.
Veterinary
PhD
thesis
at
ENVL
n◦73,
140
p.
Galef
Jr.,
B.G.,
1993.
Functions
of
social
learning
about
food:
a
causal
analy-
sis
of
effects
of
diet
novelty
on
preference
transmission.
Anim.
Behav.
46,
257–265.
Grill,
H.J.,
Norgren,
R.,
1978.
The
taste
reactivity
test
I.
Mimetic
responses
to
gustatory
stimuli
in
neurologically
normal
rats.
Brain
Res.
143,
263–279.
Kane,
E.,
Rogers,
Q.R.,
Morris,
J.G.,
Leung,
P.M.B.,
1981.
Feeding
behaviour
of
cat
fed
laboratory
and
commercial
diets.
Nutr.
Res.
1,
499–507.
Kane,
E.,
Leung,
P.M.B.,
Rogers,
Q.R.,
Morris,
J.G.,
1987.
Diurnal
feeding
and
drinking
patterns
of
adult
cats
as
affected
by
changes
in
the
level
of
fat
in
the
diet.
Appetite
9,
89–98.
Larose,
C.,
Letoux,
M.,
Guiller,
I.,
2007.
Improved
appetence
measure-
mentsystem.
Patent
WO
107508
A1,
27-09-2007.
Mugford,
R.A.,
1977.
External
influences
on
the
feeding
of
carnivores.
In:
Kare,
M.,
Maller,
O.
(Eds.),
The
Chemical
Senses
and
Nutrition.
Aca-
demic
Press,
New
York,
NY,
pp.
25–50.
Murray,
J.K.,
Browne,
W.J.,
Roberts,
M.A.,
Whitmarsh,
A.,
Gruffydd-Jones,
T.J.,
2010.
Number
and
ownership
profiles
of
cats
and
dogs
in
the
UK.
Vet.
Rec.
166
(6),
163–168.
Schaal,
B.,
Marlier,
L.,
Soussignan,
R.,
2000.
Human
fetuses
learn
odours
from
their
pregnant
mother’s
diet.
Chem.
Senses
25,
729–737.
Steiner,
J.E.,
Glaser,
D.,
Hawilo,
M.E.,
Berridge,
K.C.,
2001.
Comparative
expression
of
hedonic
impact:
affective
reactions
to
taste
by
human
infants
and
others
primates.
Neurosci.
Behav.
Rev.
25,
53–74.
Thorne,
C.J.,
1982.
Feeding
behavior
in
the
cat-recent
advances.
J.
Small
Anim.
Pract.
23,
555–562.
Van
den
Bos,
R.,
Meijer,
M.K.,
Spruijt,
B.M.,
2000.
Taste
reactivity
pat-
terns
in
domestic
cats
(Felis
silvestris
catus).
Appl.
Anim.
Behav.
Sci.
69,
149–168.
Watson,
T.,
2011.
L’appétence
chez
le
chat.
L’Essentiel
220,
10–12.