A Multithreshold Model for Sensory Analysis

ABSTRACT:  A multithreshold Bayesian procedure involving a probit approach is presented to analyze data obtained from sensory panels. The procedure assumed that the analyzed trait follows an underlying Gaussian distribution and each panelist applied individual thresholding. The proposed method provides estimates of the effects that affect the underlying measure and also estimates the panelists' ability to discriminate between the categories. The procedure avoids the use of any form of standardization technique, and the results obtained are expressed as standard deviations of the underlying trait. As an example, the procedure was applied to previously analyzed data from an experiment involving rabbit selection, in which animals from the 7th generation were compared with those from the 21st generation. Seventh generation embryos were frozen, thawed, and implanted in 21st generation does. The control (C) and selected (S) groups were contemporary. Sensory analysis was carried on samples of the longissimus dorsi muscle. The parameters evaluated were: intensity of rabbit flavor (IRF), aniseed odor (AO), aniseed flavor (AF), liver flavor (LF), tenderness (T), juiciness (J), and fibrousness (F). There were substantial differences between the selected group and control group for IRF, AO, AF, and LF and between males and females for IRF, AF, T, J, and F, which confirmed previous results. The proposed procedure has a greater ability to detect these differences, and in addition provides information about the ability of panelists to discriminate between samples.     

Comparison between rabbit lines for sensory meat quality

Rabbits from three synthetic lines were used in the experiment. Line R was selected for growth for 24 generations. Lines V and A were selected for litter size at weaning for 30 and 33 generations, respectively. Sensory analysis was carried out on the Longissimus muscle. The parameters evaluated were: juiciness (J), hardness (H), fibrousness (F), flouriness (Fl), intensity of rabbit flavour (IRF), aniseed odour (AO), aniseed flavour (AF), liver odour (LO) and liver flavour (LF). A Bayesian analysis was performed. Line V was only 82% as juicy as line R. Line V was 18% harder and 17% more fibrous than line R. Lines A and R had the same H and F. No differences in Fl were found. Small effects were found for flavour traits. We conclude that line origin has an influence on some sensory traits determining rabbit meat tenderness.     

Changes in collagen, texture and sensory properties of meat when selecting rabbits for growth rate

The consequences of selection for growth rate and the associated decrease of maturity at slaughter in rabbits on collagen content, collagen solubility, meat texture (Warner–Bratzler shear device) and the sensory properties of the m. Longissimus were studied. Sixty rabbits from the 7th generation of a line selected for growth rate (group C) were compared to 60 rabbits from the 23rd generation of the same line (group S). Both groups were contemporarily reared and slaughtered at 2000 g. No changes on collagen content were found, but group S had a higher (5%) collagen solubility. Shear force, shear firmness and area or total work needed to cut the sample were not different between groups, and hardness evaluated in the panel test was not relevantly changed. Most of the sensory properties studied did not differ relevantly between groups. Group S had 8% less aniseed odour and 10% more juiciness.     

The effect of selection for growth rate on carcass composition and meat characteristics of rabbits

The effect of selection for growth rate on carcass composition and meat quality was assessed by comparing two groups of rabbits belonging to different generations of a selection experiment. A Bayesian approach was used. Embryos belonging to generations 3 and 4 of selection were frozen and thawed to be contemporary of animals from generation 10. A control group (C), formed from offspring of these embryos, was contemporary to offspring of generations 10 and 11 of selection, chosen at random, which constituted the selected group (S). One hundred and thirty-one contemporary rabbits were slaughtered at approximately the Spanish commercial live weight of 2 kg. Carcasses were dissected and measured according to the norms of the World Rabbit Scientific Association. An animal model including effects of genetic group (C, S) and sex, and slaughter weight as a covariate was used. S animals had a higher development of liver, kidneys and of a set of organs consisting of the thymus, trachea, oesophagus, lung and heart, relative to C. For dissectible fat, S animals had less than C: −0.31 g for scapular fat, −1.62 g for perirenal fat and −2.03 g for inguinal fat. S had a lower content (−0.39%) of dissectible fat percentage in the “Reference” carcass, indicating a lower degree of maturity at slaughter. The meat to bone ratio was not affected by selection, but the meat and bone contents of the hind leg were 3.25 and 0.71 g higher, respectively, in the C group. Selected animals had a lower water holding capacity in the raw meat (−2.10%), a higher water holding capacity in the cooked meat (2.17%), a higher cooking loss (3.31%) and a lower fat percentage in the meat of a hind leg (−0.37%). Females had more fat than males: 0.26 g for scapular fat, 1.02 g for perirenal fat, 1.10 g for inguinal fat, and 0.24% for total dissectible fat percentage of the “Reference” carcass.     

The effect of fat-enriched diets on the perirenal fat quality and sensory characteristics of meat from rabbits

Two hundred and eighty four medium sized young rabbits were fed ad libitum with two fat-enriched experimental diets (V: vegetable fat, A: animal fat) and a control diet (C). Diet had a significant effect on the ultimate pH (pHu) of the Longissimus dorsi muscle and on the colour, weight and fatty acid composition of the perirenal fat. The carcasses from diet C had significantly less perirenal fat (14.2 g) compared to the carcasses from group A and V (24.4 and 23.0 g) using carcass weight and age of the animals as covariables. Supplementing the diet with animal fat produced perirenal fat which was richer in oleic acid (41%) than in groups V (20.9%) and C (27.9%). The P:S ratio was higher in group V (2.45) compared with groups A and C (0.53 and 0.42). Diet also affected the organoleptic quality of the loin meat. The meat from groups V and C was considered to have a more aniseed flavour than that from group A. In contrast the loin meat from group A had more liver flavour. Both groups V and A were considered juicier than group C. No differences were found in hardness, chewiness and fibrousness. These results suggest than from a human health point of view, the use of a diet supplemented with animal fat is useful on the basis of the nutritional index, P:S of the perirenal fat. However the amount of fat added in the diets V and A should be less than used in this study if the carcass is sold whole in order to avoid excessive perirenal fat. From an organoleptic point of view, the greater aniseed and grass flavour of group V compared to group A may give meat of better sensory quality.     

Effect of selection for growth rate on the ageing of myofibrils, meat texture properties and the muscle proteolytic potential of m. longissimus in rabbits

The effect of selection for growth rate on the degradation of the myofibrillar proteins and on meat texture properties of rabbit longissimus muscle at two ageing times (1 and 7 days) was studied as well as its effect on the proteolytic potential of the muscle. Two groups of contemporary animals (20 rabbits per group), one selected for growth rate (S) for 14 generations and the other unselected control group (C) were compared. The control group was formed from the offspring of the embryos belonging to the 7th generation and was compared with selected animals belonging to 21st generation. Myofibrillar protein degradation was studied by SDS-PAGE electrophoresis (12.5% and 4-15% polyacrylamide gels) followed by densitometric analysis of the pherograms. Texture properties were evaluated by Warner-Bratzler (WB) test and Texture profile analysis (TPA). The activities of proteolytic enzymes calpains and cathepsins and of their inhibitors were determined in the muscle at 24h. Densitometric analysis of the pherograms of samples aged 7 days showed an extra 30kDa band and the disappearance of a band with higher molecular weight than the myosin heavy chain with respect to samples aged 24h in both groups of rabbits. TPA results showed that cohesiveness was significantly lower in meat at 7 days than at 24h (P<0.0001), whereas springiness and chewiness presented a clear tendency to be lower at 7 days than at 24h (P=0.0646 and P=0.0764, respectively). Regarding the genetic type, S animals presented higher hardness and chewiness than C rabbits. Shear firmness (WB test) was significantly (P<0.0001) higher for S group, whereas no significant differences in shear force and area were found. No significant effect (P>0.05) of ageing time was detected using WB test. Selection for growth rate did not affect the activities of proteolytic enzymes or their inhibitors.     

Changes in volatile, sensory and microbial profiles during preparation of smoked ewe cheese

BACKGROUND: Oscypek is a special type of Polish smoked ewe cheese with a unique flavour described as slightly sour, piquant, salted and smoked. In this study the volatile, sensory and microbial profiles of Oscypek cheese were analysed during its various preparation stages of curding, scalding, brining and smoking. RESULTS: The smoked ewe cheese was characterised by 54 volatile compounds belonging to nine different chemical groups (free fatty acids, esters, ketones, alcohols, aldehydes, furans/furanones, phenols, sulfur compounds, terpenes). The sensory aroma profile analysis of unsmoked and smoked cheese showed an important correlation with the analysis of volatile compounds. The microbial profile data indicated that in smoked cheese such as Oscypek the levels of selected bacteria diminished after the curding stage as a result of the subsequent scalding, brining and smoking stages. CONCLUSION: From the results it can be concluded that, although the analysed smoked cheese consisted of three groups of compounds, the first derived from biochemical reactions (free fatty acids, esters, ketones, alcohols, aldehydes, sulfur compounds), the second from smoking (furans and furanones, phenols) and the third from milk flavour (terpenes), it is the smoking process that mainly influences its typical flavour. Copyright © 2011 Society of Chemical Industry     

Flavour development in pork. Influence of flavour precursor concentrations in longissimus dorsi from pigs with different raw meat qualities

Flavour development and overall eating quality of pan-fried pork chops of longissimus dorsi from eight different raw meat qualities aged for 4 and 15 days were assessed by a trained sensory panel. The raw meat qualities were obtained through combinations of strategic feeding/fasting (control vs. low glycogen concentration), slaughter live-weight (84kg vs. 110kg), and gender (female vs. castrate). The flavour development was investigated for possible correlation with the concentrations of selected individual flavour precursors present in the raw meat: monosaccharides, IMP and degradation products, fatty acids, lactate and thiamine. Differences in precursor concentrations between the raw meat qualities were observed with feeding/fasting and ageing as the main factors with the largest influence of all experimental factors. However, the concentrations of the precursors could not explain the differences in sensory perception of the pan-fried pork chops. Overall, the differences were small.     

A novel two-stage process for the production of enzyme-modified cheese

A novel method to generate enzyme-modified cheese (EMC) was developed. Initially a range of proteolysed products were produced from flavourless curd substrate using combinations of proteinases and peptidase hydrolysed to a set level of primary proteolysis. One product was selected based on ranked sensory preference analysis and subsequently further hydrolysed to a set degree of lipolysis using various lipases to create a selection of EMC products. These EMCs were compared to a target commercial Cheddar-type EMC, deemed to have a flavour profile close to natural Cheddar cheese. The sensory profile of each EMC generated were similar and not unlike the target EMC, however, greater diversity of flavour may be achieved by reducing the extent of lipolysis and/or increasing levels of free glutamate. The process demonstrates the potential to create a wide range of enzyme-modified cheese flavours from a single substrate using combinations of enzymes under controlled conditions.     

Comparison of differences between lexicons for descriptive analysis of Cheddar cheese flavour in Ireland,New Zealand,and the United States of America

Flavour lexicons for cheese provide a way to document cheese flavour for both research and marketing. The objective of this study was to compare differences and similarities in three independently developed sensory languages for Cheddar cheese flavour at three different locations (Ireland, New Zealand, United States of America) using an international selection of Cheddar cheeses. Twelve Cheddar cheeses (four from each country) were evaluated by the three panels using the respective sensory languages. Sensory space patterns obtained by principal component analysis were consistent between the three sites indicating that the overall differentiation of the cheeses by each panel was similar. The key flavour characteristics among the cheeses were described by different attributes. Cheeses were grouped by each site by country of origin suggesting international differences in Cheddar cheese flavour. Cross-cultural differences can exist in sensory language and perception, but highly trained panels using standardized, representative languages can provide comparable results.     

Differentiation of Turkish Rakies through Headspace Solid‐Phase Microextraction and Gas Chromatography‐Mass Spectrometry Analysis

Composition of flavour compounds from several commercial raki samples was determined through headspace solid‐phase microextraction (HS‐SPME) and gas chromatography coupled to mass spectrometry (GC‐MS). The extraction of volatile compounds was performed using a 75 μm CAR/PDMS fibre. Forty three volatile compounds were identified. The major compounds were trans‐anethole (86.47–94.19%), valencene (1.15–6.77%), estragole (2.66–5.46%), and cis‐anethole (0.72–2.33%). Principal component analysis (PCA) enabled the differentiation of the raki samples thanks to compounds arising mainly from aniseed.     

Influence of exopolysaccharide‐producing cultures on the volatile profile and sensory quality of low‐fat Tulum cheese during ripening

The objective of this investigation was to compare the composition and changes in the concentration of volatiles in low‐fat and full‐fat Tulum cheeses during ripening. Tulum cheese was manufactured from low‐ or full‐fat milk using exopolysaccharide (EPS)‐producing or non‐EPS‐producing starter cultures. A total of 82 volatile compounds were identified belonging to the following chemical groups: acids (seven), esters (21), ketones (14), aldehydes (six), alcohols (14) and miscellaneous compounds (20). The relative amounts of acids, alcohols and aldehydes increased in the cheeses made with EPS‐producing cultures during 90 days of ripening. Differences were found in the volatile profile of full‐fat Tulum cheese compared with the low‐fat variant, especially after 90 days of ripening. Exopolysaccharide‐producing cultures changed the volatile profile, and the EPS‐producing cultures including Streptococcus thermophilus + Lactobacillus delbrueckii subsp. bulgaricus + Lactobacillus helveticus (LF‐EPS2) produced cheese with higher levels of methyl ketones and aldehydes than the non‐EPS cultures. In the sensory analysis, full‐fat Tulum cheeses and the cheese produced with the EPS‐producing culture containing Lb. helveticus (LF‐EPS2) were preferred by the expert panel. It was concluded that the use of EPS‐producing starter cultures in the manufacture of low‐fat Tulum cheese had the potential to improve the flavour.     

八角等提取液与壳聚糖复合物对草莓保鲜效果的影响

用八角等提取液的抑菌作用和壳聚糖的成膜性研制纯天然、无毒的草莓保鲜剂。制备八角、丁香、肉桂的乙醇提取物,并加入壳聚糖配制不同浓度的保鲜液,将采摘的草莓用保鲜液浸泡5min后捞出,每隔24h测定草莓的失重率、腐烂指数、感官品质、呼吸强度、有机酸含量、可溶性糖含量、VC含量。八角等提取液浓度为2.0%、壳聚糖浓度为1.5%的保鲜液对草莓保鲜效果较好。八角等提取液与壳聚糖复合物对草莓具有明显的保鲜效果。     

DERIVATION OF A PROFILE METHOD FOR SENSORY ANALYSIS OF BEER FLAVOUR

Flavour assessment ultimately depends on sensory methods of analysis. Profile methods are considered to offer the best current solution to the problem of describing, and as far as possible quantifying, beer flavour. The sensory characteristics of flavour must be identified and described systematically and objectively, without prejudice due to preferences. A flavour ‘vocabulary’ is needed which is applicable to all types of beer. This involves selection of significant terms which have the same meaning for different people. Any system universally applicable to all types of beer must necessarily be complex, but such a “universal” system is essential for research. It also provides a “dictionary” from which terms can be selected for simpler profiles for quality control of particular products. The steps in the development of a profile system are described. Development has depended and will continue to depend on the collaboration of a large number of individuals and groups of people within the industry. Profile systems cannot be static but must continuously evolve.     

Measurement and manipulation of aroma delivery allows control of perceived fruit flavour in low- and regular-fat milks

To investigate the relationship between aroma delivery and perception, low‐fat (LF) (0.1 g per 100 mL) and regular‐fat (RF) (3.6 g per 100 mL) milk samples were prepared with the same amount of ethyl hexanoate. Sensory analysis was carried out using around ninety panellists (four replicates each) and simultaneous in vivo aroma delivery was measured for every sample. Panellists rated the fruity flavour intensity higher in LF samples using a paired comparison method (P < 0.0001, one‐tailed test). In vivo delivery was expressed as the ratio between the maximum breath concentrations (I_max), measured in the LF and RF samples (i.e. I_max LF milk/I_max RF milk), to remove person‐to‐person variation. The sensory results were in accordance with differences in aroma delivery, with LF samples delivering nearly twice as much ethyl hexanoate as RF samples. Aroma content in LF samples was adjusted so that aroma delivery in vivo was the same for both LF and RF samples. In this case, panellists could not differentiate the samples, showing a good correlation between aroma delivery and perception.     

Use of healthier fats in biscuits (olive and sunflower oil): changing sensory features and their relation with consumers' liking

The aim of this study was to evaluate consumer acceptability of biscuits when saturated fat was replaced by olive or sunflower oil and to determine the sensory characteristics responsible for changes in acceptability. Ninety seven consumers evaluated the acceptability of six biscuit samples varying in the fat source (dairy shortening, olive oil and sunflower oil) and fat content (10.6% and 15.6%). Using a Check All That Apply question (CATA), consumers also evaluated sensory properties of biscuits. Results indicated that the replacement of saturated fat (dairy shortening) by vegetable oils had an effect on biscuit acceptability which depended on biscuit fat content. According to biscuits' acceptability data, three different clusters of consumers were identified. By using a multiple factor analysis, the relationship among sensory CATA data and acceptability of each cluster explained the different acceptability patterns of consumers. For most of consumers acceptability was related to attributes “crispy”, “easy to chew” and “biscuit flavour” which, for one group were perceived in shortening biscuits and, for another in both olive and shortening biscuits. However, for the third group of consumers, acceptability was only related to flavour attributes like “roasted flavour” or “biscuit flavour” that were perceived in vegetable oil biscuits which were the preferred biscuits while, shortening biscuits were disliked and perceived as having an “off flavour”.     

Use of generalized Procrustes analysis (GPA) to test the effects of sex and carcass weight on sensory quality evaluations of Terrincho lamb meat

Generalized Procrustes analysis (GPA) was used to test the sensory attributes of Terrincho lamb meat, a product with Protected Designation of Origin (PDO) labelling. Effects of sex and carcass weight on the following sensory attributes: toughness, juiciness, flavour intensity, odour intensity, stringiness, and level of sweetness were tested. Tests were performed on 28 females and 29 males placed within three classes of carcass weight, 3 to 9 kg. Meat sensory quality was evaluated, within five sessions, by a trained taste panel. Meat was cooked in a conventional oven, cut into pieces, and delivered to the panel for evaluation. Data were analyzed using a GPA to minimize differences between assessors. Two factors explained 72.76% of the total variability. The panel did not clearly detect sex effects, yet could differentiate meat by carcass weight as heavier animals were toughest and had more intense odour and flavour, whilst light animals were most succulent.     

Analysis of flavour compounds and prediction of sensory properties in sea buckthorn (Hippophaë rhamnoides L.) berries

The aim of this study was to investigate sugars, organic acids, flavonol glycosides (FGs), proanthocyanidins and volatiles as flavour compounds in sea buckthorn (SB) berries of five cultivars and to predict the sensory properties of berries. The profiles of flavour compounds in SB berries varied significantly among the cultivars. Total proanthocyanidins and FGs were highest in ‘Pertsik’ and ‘Raisa’, respectively. Total volatiles were highest in ‘Vorobyevskaya’ and lowest in ‘Raisa’. A previously established PLS model was used to predict the sensory properties of SB berries based on the non-volatile flavour compounds. The mouth-drying astringency can be predicted the most reliably, which has the highest regression coefficients with quinic acid, isorhamnetin-3-O-sophoroside-7-O-rhamnoside and total proanthocyanidins. Bitterness cannot be predicted using the model. ‘Pertsik’ berries were predicted to be more mouth-drying astringency and sour than those of ‘Raisa’. The research supports the cultivar selection in cultivation and industry of SB berries.