Content uploaded by Agata Mazzaglia
Author content
All content in this area was uploaded by Agata Mazzaglia
Content may be subject to copyright.
Volume XXIII
Number 1
2011
PAPER
36 Ital. J. Food Sci., vol. 23 - 2011
- Key words: chocolate, instrumental determination, sensory prole, Modica, spicy -
SENSORY PROFILE
OF A SPECIALTY SICILIAN CHOCOLATE
C.M. LANZA*, A. MAZZAGLIA and E. PAGLIARINI1
DOFATA, Sezione Tecnologie Agrolimentari, Università di Catania,
Via S. Soa 98, 95123 Catania, Italy
1DISTAM, Sezione Tecnologie Alimentari, Università di Milano, Via Celoria 2, 20133 Milano, Italy
*Corresponding author: cmlanza@unict.it
ABSTRACT
In this study a sensory profile definition by a trained panel and instrumental measures deter-
mination (pH, acidity, reducing sugar content) were carried out on the Sicilian chocolate known
as “Modica” that is a niche cacao product spiced (cinnamon, vanilla) so as to obtain the Protected
Geographical Indication (PGI) branding for such a product. Sensory and instrumental data were
also submitted to statistical analysis. The chocolate of Modica has shown a sensory profile differ-
ent from other industrial products; in fact it is unique for the presence of sugar crystals and its
gritty texture resulting from the manner of its production.
Ital. J. Food Sci., vol. 23 - 2011 37
INTRODUCTION
Chocolate originates from Mexico where the
Mayas, Incas and Aztecs cultivated the cacao
tree (Theobroma cacao) preparing this product
only on special occasions (COE and COE, 1996).
While we eat chocolate the pleasure centres of
our brain are activated. One typical quality of
chocolate is its melt point; it is solid at ambient
temperature but it melts in the mouth and it is
dissolved in saliva allowing a clear final assess-
ment of its texture.
Particle size distribution and ingredient com-
position of chocolate (sugar and cocoa about
70% total in a continuous fat phase) play an im-
portant role in shaping its rheological behaviour
and sensory perception (AFOAKWA et al., 2007).
In the past chocolate was seen only as a food of
sensual pleasure with negative effects on health,
however, today chocolate has been revaluated
positively, thanks to greater nutritional infor -
mation that discredits many fallacies. Despite
high lipid and sugar content, its consumption
has some beneficial effects on the human diet;
cocoa is rich in antioxidants, above all polyphe-
nols and minerals such as potassium, magnesi-
um, copper and iron, so its intake may be use-
ful in dietary deficiencies or may balance low
levels of neurotransmitters involved in the reg-
ulation of food intake (serotonin and dopamine)
(BRUINSMA, 1999). Thus it is not surprising that
chocolate has always been the most commonly
and intensely craved food in western cultures
(WEINGARTEN and ELSTON, 1990; OSMAN and
SOBAL, 2006).
Switzerland ranks first among the world’s con-
sumers of chocolate, it is a “chocolate heavy user
country” followed by Belgium and Denmark; It-
aly only ranks 12th place with a constant grow-
ing trend, in fact, the per capita consumption
passed from 3.2 kg in 1997 to 4.3 kg in 2006
(+36.4%) (BOMMEZZADRI, 2007).
In the choices of the consumer, tradition has
prevailed: as for bars, milk chocolate is always
the best seller, and there is a continuing rise in
dark chocolate sales, while white chocolate sales
are declining (BOMMEZZADRI, 2007).
Differences in the sensory characters of choco-
late can be ascribed to the use of different cocoa
types, i.e. flavour quality of chocolate usually de-
pends on the origin of the cocoa beans (JINAP et
al., 1995), variations in ingredient proportions
and in the processing methods (JACKSON, 1999)
that differ in relation to national consumer pref-
erences and producer company practices (BECK-
ETT, 2000; WHITEFIELD, 2005). In this panorama
there are niche chocolates that offer variety in
sensory characteristics. For these products it is
useful to define the standard of sensory identity
through analyzing the characteristics of locally
crafted chocolates. In this research the choco-
late of Modica has been characterized.
During the Spanish domination (1516-1713),
an Aztec recipe to prepare the “xocoatl”, cocoa
mass mixed with vanilla or cinnamon and sugar,
was transferred to the inhabitants of the county
of Modica (Sicily) as a gift of fidelity. The formu-
la of this chocolate was lost and only through
historical sources the original recipe has been
recovered. Until 1992 the chocolate of Modica
was quite unknown, it was mainly consumed at
a local level with a production of a few thousand
bars/year. Unlike other chocolate commercial
products, this type of chocolate has often been
considered as a new product (CIUFFOLETTI and
CRESTI, 2004).
Today some techniques and ingredients are
employed to prepare a primitive type of choc-
olate bar whose ancient formula may be ex-
ploited to put on the market as a new product.
Winning strategies of correct visibility regard-
ing marketing and packaging gave a new sta-
tus to the product. In fact, 300.000 bars/year
of production have entered into national dis-
tribution only in specialized shops. One of the
greatest producers of chocolate of Modica has
had a 25% growth trend in 2006 and the choc-
olate spiced with cinnamon/vanilla has been
the most sold with a 10% growth trend (BOM-
MEZZADRI, 2007).
With reference to the chocolate production
technology (roasting of cocoa beans, mixing,
conching, and tempering) the chocolate of Mod-
ica processing method is very simple.
For the production of chocolate of Modica, the
Disciplinary (2003) compiled by the Consortium
of Guardianship foresees that a mass of cocoa
heated to 45°C and mixed with sugar and spic-
es (vanilla, cinnamon, chilli) without addition
of emulsifiers is manually worked with stone
tools at a constant temperature that does not
allow the melting of sugar crystals. By subse-
quent cold tempering the cocoa butter consoli-
dates and the product is ready to be formed in
rectangular shape (www.cioccolatomodica.it).
Since there are no scientific publications
about this product and considering that the
characteristics of chocolate have been assessed
by sensory and instrumental measures (GUI-
NARD and MAZZUCCHELLI, 1999), the aim of this
study is to underline the diversity of this choc-
olate percepted by the consumers safeguard-
ing its identity. The memory of this chocolate
was lost and only through historical sources
the original recipe has been discovered. For
this ancient chocolate that is produced by a
number of craft-made producers, the definition
of the sensory profile as well as the analytical
measures of pH, titratable acidity and reduc-
ing sugar on the cinnamon and vanilla choc-
olate of Modica can constitute a fundamen-
tal step to identify the parameters that should
be included in the PGI disciplinary (Reg. CEE,
1992). This step constitutes a winning strate-
gy so that the memory becomes enterprise giv-
ing a new status to the product.
38 Ital. J. Food Sci., vol. 23 - 2011
MATERIALS AND METHODS
Sampling
Samples analyzed were: 3 craft-made choco-
lates of Modica spiced with cinnamon (Bc, DAc,
Sc), 5 craft-made chocolates of Modica spiced
with vanilla (Bv, Ev, Cv, Rv, Sv) obtained from
different producers, a craft-made chocolate of
Modica without spices (BNA), a sample of dark
chocolate by intermediate sensory character -
istics due to muscovado sugar (D), and a com-
mercial chocolate (N) (extra bitter Italian choco-
late with 72% of cocoa), indicated as Nc and Nv
in the sensory evaluation, respectively of cinna-
mon and vanilla chocolate of Modica. The choc-
olate samples were brought to room tempera-
ture (25°±0.5°C) prior to chemical and senso-
ry analyses.
Sensory evaluation
A descriptive panel of 8 judges (5 females and
3 males, aged between 25 and 40 years) select-
ed from students and university staff was uti-
lized to define the sensory profile (UNI 10957,
2003). The judges were trained in 12 sessions,
each approximately 1 h in duration, using both
commercial and Modica craft-made chocolate,
in order to develop a common vocabulary for
the description of the sensory attributes of dark
chocolate samples and to familiarize the panel-
lists with scales and procedures. Each attribute
term was extensively described and explained to
avoid any doubt about the relevant meaning. The
panel had agreed on attributes utilized includ-
ing a list of reference standards for each partic-
ipant (Table 1). These attributes correspond to
the highest intensity score on the rating scale
used, and were established according to a pre-
vious study (LANZA et al., 2004).
Experimental design
To reduce perception fatigue of the judges,
the working plan fixed the sensory evaluation of
the three cinnamon and the five vanilla choco-
late in triplicate at different times. In every ses-
sion, the set submitted to judges was composed
of four samples: the sample to evaluate and as
reference standards chocolate of Modica with-
out spices, chocolate with muscovado sugar and
commercial chocolate.
Random samples, prepared by cutting the
chocolate bar into squares were evaluated in
triplicate; this was performed by assigning to
every attribute a score between 1 (absence of cor-
responding sensation) and 9 (extremely intense)
in individual booths under incandescent white
lighting in the sensory laboratory of the DOFA-
TA Department. Within each session the design
was balanced for carry over effect among sam-
ples and session (PAGLIARINI, 2002).
Water at room temperature was used to rinse
after sample tasting. A computerized data col-
lection program was used (FIZZ Software Solu-
tions for sensory Analysis and Consumer Tests,
Biosystemes, Couternon, France).
pH and titratable acidity determination
The pH and titratable acidity were determined,
in triplicate, as described by JINAP and DIMICK
(1990): 10 g samples were pulverized in 90 mL
boiling water.
The extract was filtered and the pH was meas-
ured.
Titratable acidity was determined on the ex-
tract with 0.10 M sodium hydroxide to an end-
point of pH 8.0.
Reducing sugars
The concentration of reducing sugar was de-
termined by the Fehling titration method on the
extract after the inversion of sucrose with sul-
phuric acid (AOAC, 1990).
Statistical analysis
The sensory data for each attribute were sub-
mitted to Analysis of the Variance (ANOVA) with
samples (S), judges (J), replicates (R) and their
relevant interactions J x S, S x R, J x R as effects
by using SAS/STAT® statistical software pack-
age version 9.1 (SAS Institute Inc., Cary, USA).
The significance of these effects was tested with
the F test. The mean values were submitted to
the multiple comparison test using the proce-
dure LSD (Least Significant Difference) that en-
ables the determination of the attributes that
differentiate the samples.
Principal Component Analysis (PCA) was also
applied to sensory and instrumental means data
in order to interpret differences among choco-
late samples using THE UNSCRAMBLER® sta-
tistical software package version 9.2 (Camo As,
Trondheim, Norway).
RESULTS AND DISCUSSION
The mean values of the instrumental data are
reported in Table 2. The different values of pH
and titratable acidity among samples can be as-
cribed to the different origins of the cocoa mass
utilized by the different producers. The large
range in reducing sugar values for these prod-
ucts are indicative of the absence of a standard-
ized process for producing chocolate of Modica.
During their training period the judges pro-
duced a list of attributes that were useful to de-
fine the sensory profile. Among the terms gener-
ated, those with a percentage of elicitation great-
er than 70% are included in the evaluation card:
Colour uniformity, Bright and Presence of crys-
Ital. J. Food Sci., vol. 23 - 2011 39
Table 1 - List of sensory attributes evaluated, their definitions and corresponding standards.
Attribute Denition Reference
Appearance
Bright Perceived color of an object indicating the Dark chocolate(Perugina*®)
relationship between reected and absorbed light
Presence Amount of sugar crystals present in the surface Chocolate of Modica
of crystals of sample (appearance)
Aroma
Butter aroma Characteristic aroma of butter perceived with White chocolate (Milka®)
the sense of smell
Cocoa Characteristic aroma of cocoa perceived with Cocoa powder (Perugina)
the sense of smell
Cinnamon/ Characteristic aroma of cinnamon perceived with Dark chocolate (Perugina)
the sense of smell with 1% of cinnamon aroma
Vanilla Characteristic aroma of vanilla perceived with Dark chocolate (Perugina)
the sense of smell with 1% of vanilla aroma
Chocolate Characteristic aroma of chocolate perceived with Dark chocolate (Perugina)
the sense of smell
Taste
Sour One of the four basic tastes caused by aqueous Dark chocolate (Perugina)
solutions of acid compounds perceived with 5% of citric acid
on the tongue
Bitter One of the four basic tastes caused by aqueous Dark chocolate (Lindt**®)
solutions of bitter compounds perceived
on the tongue
Sweet One of the four basic tastes caused by aqueous Dark chocolate (Perugina)
solutions of sweet compounds perceived with 10% of sucrose
on the tongue
Flavour
Cocoa Characteristic avour of cocoa perceived Cocoa powder (Perugina)
with the swallowing
Cinnamon Characteristic avour of cinnamon perceived Dark chocolate (Perugina) with
with the swallowing 1% of cinnamon/vanilla aroma
Vanilla Characteristic avour of vanilla perceived Dark chocolate (Perugina)
with the swallowing with 1% of vanilla aroma
Chocolate Characteristic avour of chocolate perceived Dark chocolate (Perugina)
with the swallowing
Mouthfeel
Astringent Sensory perception in the oral cavity that may Dark chocolate (Perugina)
include drying sensation, and roughing with 1% of tartaric acid
of the oral tissue
Texture
Firm Strength required to compress a substance Chocolate of Modica
between the molars
Cohesive Degree of compression (between the teeth) Dark chocolate (Lindt)
obtained prior to breaking of the product
Adhesive Strength required to remove product completely Dark chocolate (Lindt)
from palate using tongue, after compression of
the sample between tongue and palate
Melting A phase change in the mouth due to the Dark chocolate (Perugina)
increasing of temperature in oral cavity
Friable Strength with which a product crumbles and akes Dark chocolate (Perugina) with
20% of cereal powder
Gritty Amount of small particles perceived in the mouth Dark chocolate (Perugina) with
when biting the sample 20% of “Pavesini”® biscuits
*dark chocolate with 60% of cocoa mass; ** dark chocolate with 99% of cocoa mass.
40 Ital. J. Food Sci., vol. 23 - 2011
Table 3 - Influence of samples (6), judges (8) and replications (3) on the nineteen descriptors for cinnamon chocolate.
F value
Attributes Samples Judges Replications SxJ SxR JxR
Colour uniformity 110.45*** 0.89 n.s. 1.90 n.s. 1.07 n.s. 1.23 n.s. 1.38 n.s.
Bright 97.10*** 1.29 n.s. 2.26 n.s. 1.12 n.s. 1.97* 0.94 n.s.
Presence of crystals 119.25*** 3.60** 0.42 n.s. 3.21*** 0.27 n.s. 0.58 n.s.
Cocoa aroma 2.52* 4.37*** 3.73* 1.67* 0.97 n.s. 0.50 n.s.
Chocolate aroma 15.60*** 1.57 n.s. 0.05 n.s. 1.65* 0.50 n.s. 0.48 n.s.
Cinnamon aroma 109.54*** 0.53 n.s. 0.42 n.s. 2.51*** 2.12* 1.40 n.s.
Butter aroma 4.12** 0.93 n.s. 0.46 n.s. 2.31** 1.04 n.s. 0.76 n.s.
Sweet 28.13*** 1.46 n.s. 0.44 n.s. 1.08 n.s. 0.42 n.s. 0.30 n.s.
Sour 7.30*** 0.68 n.s. 0.32 n.s. 3.20*** 1.44 n.s. 0.80 n.s.
Cocoa avour 0.72 n.s. 2.18* 0.62 n.s. 1.99** 1.24 n.s. 0.43 n.s.
Chocolate avour 14.51*** 3.98*** 0.73 n.s. 1.83* 1.20 n.s. 0.53 n.s.
Cinnamon avour 76.50*** 0.75 n.s. 0.36 n.s. 1.45 n.s. 1.21 n.s. 0.52 n.s.
Astringent 10.44*** 2.34* 0.39 n.s. 3.64*** 0.40 n.s. 1.43 n.s.
Firm 5.31*** 0.90 n.s. 1.33 n.s. 2.34** 0.83 n.s. 0.62 n.s.
Cohesive 8.93*** 2.68* 0.29 n.s. 1.98** 1.10 n.s. 0.52 n.s.
Adhesive 6.11*** 2.00 n.s. 1.46 n.s. 1.38 n.s. 0.49 n.s. 0.72 n.s.
Melting 2.25 n.s. 2.95** 2.09 n.s. 1.82* 0.85 n.s. 0.97 n.s.
Friable 71.12*** 4.02*** 0.09 n.s. 1.60* 0.49 n.s. 0.37 n.s.
Gritty 112.62*** 4.04*** 0.20 n.s. 3.69*** 0.54 n.s. 1.06 n.s.
*** signicant difference for p ≤ 0.001; ** signicant difference for p ≤ 0.01; * signicant difference for p ≤ 0.05; n.s. no signicant difference.
Table 2 - Instrumental means data.
Sample pH Std. dev. Tritatable acidity Std. dev. Reducing sugar Std. dev.
Meq NaOH/g g/100 g
N 5.70 ±0.18 4.30 ±0.35 28.00 ±0.05
BNA 5.92 ±0.04 2.73 ±0.02 55.44 ±3.09
D 5.51 ±0.01 4.32 ±0.00 40.97 ±0.01
Cinnamon
Bc 5.74 ±0.02 3.03 ±0.02 40.41 ±0.79
Dac 6.78 ±0.06 1.21 ±0.12 67.11 ±1.56
Sc 5.77 ±0.02 2.73 ±0.02 41.68 ±0.24
Vanilla
Bv 5.76 ±0.08 2.81 ±0.14 54.90 ±0.49
Ev 6.92 ±0.08 0.99 ±0.08 62.17 ±0.01
Cv 6.94 ±0.06 1.28 ±0.11 58.48 ±1.38
Rv 6.97 ±0.10 1.07 ±0.06 64.42 ±2.03
Sv 6.08 ±0.00 2.12 ±0.07 62.35 ±1.26
tals (appearance), Butter Aroma, Cocoa Aroma,
Chocolate Aroma (olfactive), Sour, Bitter, Sweet
(gustative), Cocoa Flavour, Chocolate Flavour
(flavour), Astringent (mouthfeel), Firm, Cohesive,
Adhesive, Melting, Friable, Gritty (texture), for
the cinnamon samples while for the vanilla choc-
olates they were classified by their vanilla aroma
and flavour, however for the vanilla samples the
judges did not elicit the attributes Colour uni-
formity, Bright, Sour, Astringent, and Friable.
The results of ANOVA for cinnamon chocolates
(Table 3) showed significant differences among
the samples for all the attributes with the ex-
ception of Cocoa flavour and Melting. The judge
effect indicated significant differences for many
attributes however the replication effect was not
significant except for Cocoa aroma attribute. The
interactions SxJ revealed significant differences
with the exception of Colour uniformity, Bright,
Sweet, Cinnamon flavour, Adhesive. The inter-
actions S x R showed a good homogeneity of the
samples during replicates, with the exception of
Bright and Cinnamon aroma, and finally, the in-
teractions J x R underlined a good reliability of
the answers furnished by the judges.
ANOVA of vanilla chocolate (Table 4) showed
significant differences among samples for all of
the attributes, except that for Cocoa aroma and
flavour and Melting. The judge effect indicated
significant differences for all attributes howev-
er the replication effect was not significant ex-
cept for the Presence of crystals attribute. The
interactions SxJ reveal significant differences,
except that for Chocolate aroma and flavour.
Ital. J. Food Sci., vol. 23 - 2011 41
The interactions S x R did not show a good ho-
mogeneity of the samples during replicates, and
finally, the interactions J x R underlined a good
reliability of the answers furnished by the judg-
es with the exception of Chocolate aroma and
Cohesive.
The mean values of the samples’ sensory data
submitted to the multiple comparison test (LSD)
were useful to determine what attributes differ-
entiate the samples (Tables 5 and 6).
With respect to the cinnamon chocolate, the
commercial sample (Nc) shows the highest mean
score as far as Color uniformity, Bright, But-
ter aroma and Astringent attributes are con-
cerned, while with respect to the vanilla samples,
the commercial sample (Nv) shows the highest
mean scores for Aroma and Chocolate Flavour,
Butter Aroma and Adhesive attributes. Even if
the samples of Modica have different rheologi-
cal characteristics (Presence of crystals, Gritty
Table 4 - Influence of samples (8), judges (8) and replications (3) on the fourteen descriptors for vanilla chocolate.
F value
Attributes Samples Judges Replications SxJ SxR JxR
Presence of crystals 131.31*** 38.64*** 4.98** 4.50*** 20.80*** 0.87 n.s.
Cocoa aroma 1.73 n.s. 5.09*** 0.71 n.s. 2.72*** 1.67 n.s. 1.35 n.s.
Chocolate aroma 22.32*** 12.29*** 1.04 n.s. 1.29 n.s. 0.80 n.s. 1.91*
Vanilla aroma 7.01*** 11.09*** 0.19 n.s. 2.58*** 2.50** 0.85 n.s.
Butter aroma 3.14** 12.80*** 1.22 n.s. 2.15*** 1.44 n.s. 0.60 n.s.
Sweet 25.03*** 27.60*** 1.64 n.s. 2.53*** 10.63*** 0.61 n.s.
Cocoa avour 1.40 n.s. 10.90*** 0.03 n.s. 2.74*** 0.99 n.s. 0.76 n.s.
Chocolate avour 10.90*** 10.38*** 0.79 n.s. 1.40 n.s. 1.16 n.s. 1.09 n.s.
Vanilla avour 9.75*** 11.82*** 1.39 n.s. 2.65*** 2.48** 0.91 n.s.
Firm 2.91** 17.98*** 0.83 n.s. 1.78** 0.62 n.s. 0.61 n.s.
Cohesive 3.33** 17.10*** 0.06 n.s. 2.17*** 0.81 n.s. 2.06*
Adhesive 5.57*** 19.51*** 0.88 n.s. 2.04** 1.25 n.s. 1.65 n.s.
Melting 1.40 n.s. 7.54*** 0.61 n.s. 2.16*** 1.92* 1.12 n.s.
Gritty 93.85*** 36.90*** 1.14 n.s. 5.26*** 10.88*** 1.14 n.s.
*** signicant difference for p ≤ 0.001; ** signicant difference for p ≤ 0.01; * signicant difference for p ≤ 0.05; n.s. no signicant difference.
Table 5 - Mean of the score among the 19 sensory attributes for cinnamon and no spicy samples (6).
Attributes Bc BNA DAc Sc Dc Nc
Color uniformity 1.60
a1
2.54
b
5.10
c
6.85
d
7.33
d
8.42
e
Bright 1.58
a
1.95
a
4.85
b
6.29
c
6.46
c
7.43
d
Presence of crystals 7.90
e
7.42
de
6.54
c
6.98
cd
4.46
b
1.00
a
Cocoa aroma 4.52
abc
5.17
c
4.06
a
4.29
ab
4.88
bc
5.07
bc
Chocolate aroma 4.10
bc
2.83
a
3.46
ab
3.90
bc
4.58
c
6.31
d
Cinnamon aroma 4.83
b
1.00
a
5.71
c
6.13
c
1.00
a
1.00
a
Butter aroma 1.85
a
2.33
a
1.92
a
1.85
a
2.33
a
3.10
b
Sweet 7.06
c
6.42
c
6.79
c
6.44
c
4.25
b
3.28
a
Sour 2.42
ab
2.88
b
2.06
a
2.73
ab
4.00
c
3.10
b
Cocoa avour 4.56 4.75 4.44 4.56 5.08 4.74
Chocolate avour 4.08
b
2.79
a
3.92
b
3.88
b
4.46
b
6.11
c
Cinnamon avour 4.42
b
1.00
a
5.52
c
6.35
d
1.00
a
1.00
a
Astringent 2.38
a
2.67
a
2.58
a
3.04
a
4.38
b
4.08
b
Firm 4.02
a
4.46
ab
5.40
c
5.50
c
5.42
c
5.07
bc
Cohesive 2.92
a
4.17
b
3.94
b
4.13
b
4.96
c
5.31
c
Adhesive 3.23
a
4.54
c
3.52
ab
4.06
abc
4.42
bc
5.57
d
Melting 4.77
4.92
5.40
5.42
5.29
5.83
Friable 6.86
d
5.63
c
6.90
d
6.71
d
3.96
b
1.39
a
Gritty 7.56
d
6.88
c
7.00
cd
6.73
c
4.33
b
1.14
a
1The values marked with different letters in the same line are signicantly different (p ≤ 0.05).
42 Ital. J. Food Sci., vol. 23 - 2011
and Friable), they are not significantly different
from commercial chocolate samples as for the
intrinsic attributes of chocolate that is Melting
and Cocoa flavour.
The next step regards the principal compo-
nent analysis of all of the sensory and instru-
mental data of the cinnamon and vanilla sam-
ples in order to identify the importance of vari-
ous attributes to discriminate among the sam-
ples obtaining a multidimensional space. Fig-
ures 1 and 2 report the principal component
score plot and the principal component load-
ing plot from chocolate sensory and analytical
data, respectively.
The variance explained by the first two prin-
cipal components was 75%. Figure 1 (score plot)
shows the position of the samples in the 3 repli-
cates. Since the distance among the replicates of
the same product is very little, it can be affirmed
that the results are reliable and that the judges
have furnished their judgments in a consistent
way. Furthermore, chocolate samples appear to
be well separated in the space.
Moving left to right along the first component
(explained variance 54%), commercial chocolate
samples (N and D) are distinct from the Modi-
ca chocolate. The second component (explained
variance 21%) distinguishes the cinnamon and
not spicy samples from the vanilla chocolate.
Principal components loading were examined
in order to identify the importance of various at-
tributes in discriminating among the samples.
Table 6 - Mean of the score among the 14 sensory attributes for vanilla and no spicy samples (8).
Attributes BNA Bv Ev Cv Dv Rv Sv Nv
Presence of crystals 7.42
e1
6.84
e
7.33
e
5.44
c
5.04
bc
4.81
b
6.31
d
1.08
a
Cocoa aroma 5.17
5.33
4.81
4.60
5.63
5.00
4.94
4.92
Chocolate aroma 2.83
a
4.60
cd
3.71
b
4.33
bc
5.00
d
4.48
cd
4.96
cd
6.58
e
Vanilla aroma 1.00
a
2.35
d
2.02
cd
2.10
cd
1.71
bc
2.25
d
2.15
cd
1.50
b
Butter aroma 2.33
a
2.38
a
2.27
a
2.56
a
2.54
a
2.29
a
2.38
a
3.54
b
Sweet 6.42
d
6.10
cd
6.42
d
5.50
b
4.29
a
5.73
bc
6.34
d
3.79
a
Cocoa avour 4.75
5.33
4.83
4.60
5.25
4.94
4.58
4.96
Chocolate avour 2.79
a
4.58
c
3.65
b
4.17
bc
4.46
c
4.54
c
4.73
c
5.54
d
Vanilla avour 1.00
a
2.19
c
1.81
bc
2.17
c
1.63
b
2.17
c
1.98
bc
1.13
a
Firm 4.46
a
5.54
c
4.81
ab
5.06
abc
5.54
c
5.46
bc
5.57
c
5.42
bc
Cohesive 4.17
ab
3.73
a
3.96
ab
4.34
b
4.13
ab
4.46
b
4.02
ab
5.17
c
Adhesive 4.54
cd
3.21
a
3.58
ab
4.04
bc
3.92
bc
4.19
bc
3.90
abc
5.13
d
Melting 4.92
5.73
5.65
5.46
5.21
5.38
5.79
5.17
Gritty 6.88
d
6.56
d
7.04
d
5.04
b
5.33
bc
4.83
b
5.65
c
1.04
a
1The values marked with different letters in the same line are signicantly different (p ≤ 0.05).
Fig. 1 - Score plot of chocolate samples.
Ital. J. Food Sci., vol. 23 - 2011 43
Figure 2 represents the plot of the sensory and
physico-chemical parameters in the plane de-
fined by the first two components.
The commercial samples (N and D) on the
left of the first PC in Fig. 1 are negatively corre-
lated to the Presence of Crystal, Friable, Gritty
and Sweet, and positively correlated to Choco-
late Aroma and Flavour, Cohesive and Astrin-
gent. The results from PCA show that the senso-
ry profile of the chocolate of Modica is very dif-
ferent from that of the other samples. In gener-
al, the second dimension makes the difference
between the chocolate of Modica and commer-
cial products.
CONCLUSIONS
The results obtained in this research show the
peculiar sensory profile of the chocolate of Modi-
ca, a fact that differentiates it from the commer-
cial dark chocolate, and principally the choco-
late of Modica’s attributes of appearance (Pres-
ence of crystals), taste (Sweet), and texture (Grit-
ty), that are a result of its production technolo-
gy. Despite its diversity, the chocolate of Modi-
ca did not lose two sensory attributes that are
typical characteristics of chocolate: melting and
cocoa flavour.
If aroma makes the difference between the cin-
namon and vanilla references, it is not clear why
some attributes (Colour uniformity, Bright, Sour,
Astringency, and Friable) of the cinnamon sam-
ples are not present in the sensory profile of va-
nilla samples. This result underlines the scarce
homogeneity among the samples examined. In
fact, each producer of the Association has “in-
Fig. 2 - Loading plot of chocolate samples.
terpreted” the original recipe, varying the ingre-
dients in their quantities and modifying some
phases of the workmanship to shorten the pro-
duction process. Therefore a follow up study is
necessary to set a range of chemico-physical and
sensory parameters to conform and standard-
ize the production, so as to achieve the identi-
fication with the PGI brand. In fact, cacao and
its products are still developing and there are
some avant-gardists with strangest tastes who
like orange, chilli, cinnamon, ginger, and even
absinthe flavoured chocolate. This study made
it possible to sort out the sensory dimensions
on an ancient chocolate helping to get relevant
and useful information to guide product devel-
opment. It is nevertheless a fact that the indus-
try and marketing must better understand the
consumer sensory development in order to sat-
isfy his needs.
REFERENCES
Afoakwa E.O., Paterson A. and Fowler M. 2007. Factors
influencing rheological and textural qualities in choc-
olate - A review. Trends in Food Science & Technolo-
gy 18:290-298.
AOAC 1990. Official Methods of Analysis of the A.O.A.C.
(15th Ed.). Association of Official Analytical Chemists,
Inc., Kenneth Helrich Ed., Arlington, VA.
Beckett S.T. 2000. The Science of Chocolate. London: Roy-
al Society of Chemistry.
Bommezzadri M. 2007. Dossier Cioccolato. Food 17(10):131-
168.
Bruinsma K. and Taren D.L. 1999. Chocolate: Food or drug?.
Journal of the American dietetic Association 99:1249-
1256.
Ciuffoletti Z. and Cresti C. 2004. Dolceamaro: storia e storie
dal cacao al cioccolato. Alinari IDEA, pag. 87. Firenze, Italy.
44 Ital. J. Food Sci., vol. 23 - 2011
Coe S.D. and Coe M.D. 1996. The true history of chocolate.
New York: Thamer and Hudson Ltd.
Guinard J.X. and Mazzucchelli R. 1999. Effects of sugar and
fat on the sensory properties of milk chocolate: descrip-
tive analysis and instrumental measurements. Journal
of the Science of Food and Agriculture 79:1331-1339.
http://www.cioccolatomodica.it
Jackson K. 1999. Recipes. In: Beckett S.T. Editor. Industri-
al chocolate manufacture and use. 3rd Ed. Oxford: Black-
well Science. p. 36-56.
Jinap S. and Dimick P.S. 1990. Acidic characteristics of fer-
mented and dried cocoa beans from different countries of
origin. Journal of Food Science 55:547-550.
Jinap S., Dimick P.S. and Hollender R. 1995. Flavour evalu-
ation of chocolate formulated from cocoa beans from dif-
ferent countries. Food Control 6(2):105-110.
Lanza C.M., Pagliarini E., Mazzaglia A. and Laureati M.
2004. Sensory characterization of the typical chocolate
of Modica. European Conference on Sensory Science of
Food and Beverages “A Sense of Identity” 26-29 Septem-
ber, Firenze, Italy.
Osman J.L. and Sobal J. 2006. Chocolate craving in Amer-
ican and Spanish individuals: Biological and cultural in-
fluences. Appetite 47:290-301.
Pagliarini E. 2002. Metodi discriminanti. In: Valutazione sen-
soriale: aspetti teorici, pratici e metodologici. 1st Ed. Mi-
lano (Italy): Hoepli. p 59.
Reg. CEE n° 2081/92.1992. Gazzetta Ufficiale, n. L 208
del 24/07.
UNI 10957 2003 Sensory analysis – Method for establish-
ing a sensory profile in foodstuffs and beverage. Milano:
Ente Nazionale Italiano di Unificazione.
Weingarten H. and Elston D. 1990. The phenomenology of
food craving. Appetite 15:231-246.
Whitefield R. 2005. Making chocolates in the factory. Lon-
don, UK: Kennedy’s Publications Ltd.
Paper received March 6, 2010 Accepted June 29, 2010
