Developing a cuts-based system to improve consumer acceptability of

Meat Science 121 (2016) 216–227

Contents lists available at ScienceDirect

Meat Science journal homepage: www.elsevier.com/locate/meatsci

Developing a cuts-based system to improve consumer acceptability of pork: Impact of gender, ageing period, endpoint temperature and cooking method H.A. Channon a,b,⁎, D.N. D'Souza c, F.R. Dunshea b a b c

Australian Pork Limited, P.O. Box 4746, Kingston, ACT 2604, Australia Faculty of Veterinary and Agricultural Sciences, The University of Melbourne, Parkville, VIC 3010, Australia SunPork Solutions, P.O. Box 5950, Manly, QLD 4179, Australia

a r t i c l e

i n f o

Article history: Received 25 February 2016 Received in revised form 27 May 2016 Accepted 13 June 2016 Available online 14 June 2016 Keywords: Pork Eating quality Gender Ageing period Endpoint temperature Cuts Cooking method Consumers

a b s t r a c t The effect of gender (entire male, female and castrate), ageing period (2 or 7 days) and endpoint temperature (70 or 75 °C) on consumer perceptions of cuts from the loin (Musculus longissimus thoracis et lumborum), silverside (Musculus biceps femoris) and shoulder (Musculus triceps brachii (roast) and Musculus supraspinatus (stir fry)) when roasted or stir fried (all primals) or grilled as steaks (loin only) was investigated. Higher scores for juiciness (P = 0.035), flavour (P = 0.017), overall liking (P = 0.018), quality grade (P = 0.026) were obtained from castrates than entire males, with females intermediate. Neither ageing period nor endpoint temperature, as main effects, influenced sensory scores. Loin steaks and silverside roasts obtained lower (P b 0.001) scores for all sensory traits except aroma; scores for shoulder cuts were highest (P b 0.001). Cooking to 70 °C improved (P b 0.05) juiciness, flavor and overall liking scores of loin steaks compared with 75 °C. Different pathway interventions are required to optimize eating quality of different pork cuts and the cooking methods used to prepare them. © 2016 Elsevier Ltd. All rights reserved.

1. Introduction In Australia, no industry system has yet been developed to consistently supply high quality fresh pork cuts to consumers. In comparison, a cut x cooking method-based system, known as Meat Standards Australia, has been implemented for both beef (Polkinghorne, Watson, Thompson, & Pethick, 2008; Watson, Polkinghorne, & Thompson, 2008) and sheep meat (Pleasants, Thompson, & Pethick, 2005). Whilst many studies have investigated the effects of various production, preand post-slaughter management of pigs and carcases as well as cooking methods used on technological and eating quality of pork, additional data is needed to support the development of a flexible and nonprescriptive cuts-based eating quality system for pork (Channon, Hamilton, D'Souza, & Dunshea, 2016). The Australian pork industry has typically finished males as entire males in response to consumer demand for lean pork cuts and higher feed conversion efficiency on-farm of entire males compared with castrated males. However, the prevalence of boar taint cannot necessarily be managed through liveweight at slaughter of entire males (D'Souza et al., 2011). Issues with boar taint therefore need to be overcome to ⁎ Corresponding author. E-mail address: [email protected] (H.A. Channon).

http://dx.doi.org/10.1016/j.meatsci.2016.06.011 0309-1740/© 2016 Elsevier Ltd. All rights reserved.

increase pork consumption frequency since flavour, tenderness and juiciness are the key eating quality attributes that influence overall liking of pork (Channon, Taverner, D'Souza, & Warner, 2014; Crawford et al., 2010; Wood et al., 1996). Current data on eating quality performance of Australian pork sourced from different genders is required to both quantify whether pork from entire male carcases (slaughtered at 90– 105 kg liveweight) is as acceptable to that from female and castrated males, given the risk of boar taint and typically lower intramuscular fat levels, and enable gender recommendations to be included in any eating quality system that is developed. Channon et al. (2016), as part of a quantitative literature review, identified that ageing of pork for 3 days or more post-slaughter can increase sensory tenderness scores by N10% compared with ageing for 2 days. It is notable that, whilst the Meat Standards Australia systems for both beef and sheepmeat include ageing period as a key pathway intervention (Meat Standards Australia 2011; Meat and Livestock Australia, 2015), such benefits of ageing on pork eating quality have not been formally utilized within Australian pork supply chain arrangements as a means of delivering higher quality, with less variability, to consumers. Notably, many studies that have investigated various treatment effects on sensory quality have primarily utilized the loin (typically grilled, broiled or roasted steaks or chops of varying thicknesses), and

H.A. Channon et al. / Meat Science 121 (2016) 216–227

where different muscles were evaluated, sensory quality was either not determined (eg. Aalhus, Best, Costello, & Schaefer, 1997; Aalhus, Schaefer, Murray, & Jones, 1992; Aaslyng & Barton Gade, 2001; Aluwé et al., 2009; Barton Gade, 2008; Crawford et al., 2010; Sather, Jones, Schaefer, Colyn, & Robertson, 1997) or not directly compared (Channon et al., 2014; Wood, Nute, Fursey, & Cuthbertson, 1995). No published eating quality studies have involved pork stir fry, despite its increasing consumer popularity as a cooking method for pork (Celsius Research 2014). Improvements in juiciness and/or tenderness when pork is cooked (as a steak or as roasts) to an endpoint temperature of b68 °C have been reported (Bejerholm & Aaslyng, 2003; Ellis et al., 1998; Heymann, Hedrick, Karrasch, Eggeman, & Ellersieck, 1990; Moeller et al., 2010; Saunders, Wilkinson, & Hall, 2000; Simmons, Carr, & McKeith, 1985; Tikk et al., 2007; Wood et al., 1995) compared with pork cooked to higher endpoint temperatures. The minimum endpoint temperatures for whole cuts of meat in the USA, including pork, have been revised to 145 °F (62.7 °C), followed by a rest period of three minutes, before carving or consuming (USDA 2011). However, many Australian consumers are reluctant to cook pork to a medium-rare degree of doneness, comparable to an endpoint temperature of 64–68 °C, due to unfounded cultural perceptions that undercooking of pork may present food safety risks. Whilst inherent overcooking of pork by Australian consumers contributes to poor eating quality performance, resulting in tough, dry, chewy pork with poor flavour (Channon & Warner, 2011), the preference of Australian consumers of pork cooked to a mediumwell done to well done degree of doneness led to the use of an endpoint temperature of 75 °C for loin steaks in previous Australian studies (eg. Channon et al., 2014; D'Souza & Mullan, 2002; Moore, Mullan, & D'Souza, 2012). Additional information is therefore needed to both quantify the effect of endpoint temperature on eating quality traits of different pork cuts and support the development of consumer messaging to modify cooking practices to reduce the incidence of overcooking and enable high quality pork meals to be prepared and enjoyed by consumers. The objective of this study was to determine the effect of gender, ageing period, cut type, cooking method used and endpoint temperature to improve pork consistency and thereby reduce the fail rate of pork to b10%. 2. Methodology 2.1. Animal management The animal management procedures used in this study were approved by the Animal Ethics Committee at Rivalea Australia Pty Ltd., NSW, Australia (Project 11 M044). A total of 60 pigs (Large White x Landrace, PrimeGro™ Genetics) were sourced from a commercial piggery. At birth, piglets were injected with iron and randomly allocated to gender treatment; females, entire males or castrates. Males allocated to the castrate treatment were castrated at 1 day of age. All animals were conventionally housed indoors on concrete slatted floors. Liveweight was recorded at birth, at the commencement of the finisher phase at 15 weeks of age and on the day prior to slaughter. At 15 weeks of age, pigs were allocated on the basis of liveweight to two slaughter groups within sex (10 pigs per gender per slaughter day). Pigs were individually tattooed on the day prior to slaughter immediately after weighing to enable individual carcases to be identified on the slaughter floor and to utilise liveweight data for individual animals to determine average daily gain during the finisher period. Diet specifications fed to pigs during the finisher phase are shown in Table 1. Animals were slaughtered over two slaughter days at 21–22 weeks of age, one week apart. Animals were minimally handled on farm, during transport and at the abattoir. Pigs were housed within gender and slaughter date groups for the entire finisher period, during transport and in lairage. Following

217

Table 1 Diet specifications fed to pigs during weaner, grower and finisher phases. Diet specifications

Protein (%) DE/MJ kg Fat (%) Fibre (%) Lysine (%)

Weaner (5–15 kg liveweight)

Grower (15–60 kg liveweight)

Finisher (60–100 kg liveweight)

22.53 14.46 3.08 2.45 1.42

16.15 14.00 2.91 3.22 0.92

13.01 13.79 2.96 4.03 0.78

weighing on the day prior to slaughter, pigs were transported 300 km, for a duration of 3 h, to the abattoir, unloaded and held in lairage for 17 h prior to slaughter in their gender groups, with ad libitum access to water. Each gender group was moved separately to the point of stunning and stunned using 90% carbon dioxide (Butina, Denmark). Following evisceration, a fat sample was obtained from the belly region of each carcase for determination of and rostenone and/or skatole concentration and frozen at −20 °C. Hot carcase weight (AUS-MEAT Trim 1; AUS-MEAT, Brisbane, Australia) and fat depth at the P2 site (located 59 mm from the midline of the carcase at the last thoracic rib) using PorkScan (PorkScan Pty Ltd., Canberra, Australia) were measured and recorded for all carcases. Carcases were then dressed to AUS-MEAT Trim 13, split prior to chilling and placed in a chiller fitted with overhead fans and conventionally chilled (1–2 °C for 24 h), with the same chiller and chiller settings used on both slaughter days. Measurements of muscle pH and temperature decline post-slaughter were made in the loin (Musculus longissimus thoracis) adjacent to the P2 site at 45 min, 90 min, 3 h, 6 h and 24 h post-slaughter in both sides of each carcase using a portable pH meter (MPI, Kansas USA) fitted with a polypropylene spear-type gel electrode and a separate temperature probe (Noronix pocket probe thermometer). Cold carcase weight was obtained immediately prior to boning at 24 h post-slaughter. Whole shoulders and boneless, rind-on loin (M. longissimus thoracis et lumborum) and silverside (M. biceps femoris) were collected from both sides of each carcase. For the loin, a 2–3 cm slice was firstly removed from the caudal end and a total of four 2.5 cm thick steaks were then sliced and denuded of subcutaneous fat. A 15 cm piece for roasting followed by a 10 cm piece, for later preparation into stir fry samples, were then cut. The remaining loin piece was used to undertaken objective meat quality measurements. For the silverside, a 10 cm roasting piece was firstly cut from the cranial end, followed by another 7–10 cm piece for stir fry. Remaining muscle was used for objective meat quality measurements. The bolar blade (M. triceps brachii), and chuck tender (M. supraspinatus) were removed from the shoulder and used for roasting and stir fry, respectively. As sufficient quantities of samples for both roasting and stir fry could not be obtained from the bolar blade, the chuck tender was used to ensure that consumers evaluated stir fry from the same muscle. Sections were also retained for objective measurements. Rind and subcutaneous fat on all roast pieces was removed after cooking and roasts were then sliced by the chef for consumer evaluation. All cuts were labelled, individually vacuum packaged, boxed and frozen following ageing at 2 °C for either 2 or 7 days post-slaughter, depending on treatment allocation. Pork was transported by frozen transport to the sensory facility, sorted into their pre-allocated sensory sessions in a controlled temperature room (12 °C) two days after arrival and held at − 18 °C until required for sensory assessment. 2.2. Objective measurements Androstenone (castrates and entire males) and skatole (all genders) concentrations were determined on belly fat samples (Frontage

218

H.A. Channon et al. / Meat Science 121 (2016) 216–227

Laboratories Co., Ltd., Shanghai, China). Androstenone concentration was determined using high performance liquid chromatography with mass spectrometric detection (HPLC/MS/MS); the lower limit of quantification was 20 ng/g. Skatole was determined using high performance liquid chromatography – fluorescence detection (HPLC-FLD); the lower limit of quantification was 5.5 ng/g. Muscle pH and temperature and meat colour were determined on samples from all muscles aged for either 2 or 7 day post-slaughter. Meat colour was determined following exposure to air at room temperature for 10 min using a Konica Minolta Chromameter Model CR-400 (calibrated on a white tile) set on the L*, a* and b* system where L* denotes relative lightness (higher L* values = paler meat), a* relative redness (higher a* values = more red) and b* relative yellowness (higher b* values = more yellow), using D65 illumination and a 2° standard observer. Each loin and silverside sample was prepared into a 70 ± 5 g block for Warner-Bratzler shear force (WBSF) assessment, a 40 ± 5 g block for drip loss assessment with remaining trim reserved for analysis of intramuscular fat content. WBSF and drip loss was not determined for shoulder muscles due to the small muscle size. Samples for WBSF assessment and intramuscular fat were then labelled, individually bagged and frozen at −20 °C. WBSF samples were cooked from frozen state in a waterbath, suspended from a metal rack, preheated to either 70 or 75 °C for 30 min. Samples were cooled in iced water for 30 min, dried and weighed to determine cooking loss (expressed as a percentage of weight lost due to cooking) and stored at 4 °C for 24 h (Bouton, Harris, & Shorthose, 1971). From each sample, five 1 cm2 replicate samples were cut parallel to the orientation of muscle fibres and WBSF was measured using a Warner Bratzler shear blade fitted to a Mecmesin BFG500N gauge (Bodine Electric Company, Chicago, USA). Intramuscular fat content was analysed from a 2–3 g sample of meat using Soxhlet extraction (AOAC Method 920.39, 2000). Drip loss was determined using a modified method of Rasmussen and Andersson (1995). A sample of pork loin was cut to a 40 ± 5 g cube, weighed and weight recorded. The sample was then wrapped in 20 cm2 piece of square netting. The wrapped sample was suspended in a 200 ml plastic container and left to stand in a 4 °C chiller for 24 h, after which it was removed, gently rolled in paper towelling and reweighed to determine percentage drip loss. 2.3. Sensory evaluation Consumer sensory sessions were conducted at a central facility, with consumers recruited by an independent recruitment company. All consumers (n = 480) involved had eaten fresh pork within the past month, were aged between 18 and 65 years and must have had no meat industry experience. Each session, involving 8 consumers, was completed within 35 min. The thawing protocol used per testing day (three sensory sessions per day) was to remove the cartons from the freezer and place them into a 4 °C constant temperature room for 44 to 52 h to ensure that samples of varying sizes/dimensions were defrosted to the same temperature prior to being prepared and cooked. Samples were thawed in their sealed cartons of individual sensory sessions to ensure no confusion or mix up of samples when multiple sessions were required to be thawed at the same time. As a randomized tasting order was used for the seven samples evaluated by each consumer, samples were not cooked to order. The kitchen/preparation area was maintained at a temperature of 23 °C during the sensory sessions. Prepared samples were kept warm using 20 sealed and labelled glass Pyrex containers (World Kitchen, Rosemont, Illinois, USA) on top of heated warming plates (Cuisinart Model CWT-240 A; n = 4) preheated to 65 °C to prevent moisture and heat loss. Roasting was undertaken on all primal cuts (loin, shoulder: bolar blade and silverside) using an electric fan-forced oven (Smeg, Model SA20XMFR). Loin and silverside roasts were trimmed to similar weights and dimensions as the bolar blade roasts. The three roast pieces were

placed onto a greaseproof paper-lined roasting tray (with removable drip tray) and an oven proof label attached to the tray to identify each piece. Two trays of roasts were required for each session. A meat thermometer (a stainless steel probe with a 1 m lead connected to a digital thermometer on the outside of the oven) was inserted into the centre of each roast and used to monitor its temperature of the roast pieces throughout cooking. The temperature of the roasts was 1–3 °C before cooking commenced. Pork roasts were cooked using an oven setting of 175 °C to either a 70 °C or 75 °C final endpoint temperature after a five minute resting period. Once cooked and rested, rind and all visible fat from each roast was removed, the piece was cut in half along the grain, the cut flat surface was turned down onto the cutting board surface, both half pieces were cut into 6 ± 1 mm thick slices across the grain and all slices were then placed into a labelled Pyrex holding container. A total of 18–24 slices were obtained from each roast piece with 12 slices required for each sensory session. Three pieces of roast slices were provided to each consumer for evaluation. Loin steaks were grilled using a Silex Grill Model GTTPowersave 10.10–30 (Silex Elektrogerate GmbH, 22143 Hamburg, Germany) set at 160 °C. The temperature of the loin steaks was 5–7 °C prior to cooking. Steaks were measured for length and graded into small (≤130 mm) and large (N130 mm) length and different grilling times were then used to deliver the final endpoint temperatures of either 70 °C or 75 °C after a two minute resting period. Steaks of ≤130 mm length allocated to the 70 °C endpoint temperature treatment were cooked for 3 min 10 s and those in the 75 °C treatment cooked for 3 min 40 s. For steaks of N130 mm length, cooking times were increased by 15 s for both cooking temperatures. Steaks were brushed on each side with rice bran oil before being placed onto the grill plate. After a two minutes resting period, each steak was trimmed on all four sides to remove fat and edges and the centre piece from each steak was placed in an individual labelled Pyrex container before being provided for consumer evaluation. The stir fry cooking method was undertaken on all primal cuts. Pieces for stir frying were cut into 50 mm × 10 mm × 10 mm, each weighing about 7 g. About 15 stir fry pieces were dissected from each primal cut, placed into a labelled plastic storage container and refrigerated at 5 °C until required. The temperature of the stir fry pieces was 5– 7 °C before cooking commenced. Electric woks (Sunbeam Model WW4500D) set on setting 7 were used to cook the stir fry pieces. Five ml of rice bran oil was added to the wok and pre-heated for 30 s before the batch of 15 pieces of stir fry was cooked. Stir fry was turned every 15–20 s for even colour development during cooking. All batches of stir fry were continuously monitored for temperature throughout the cooking and resting periods using a k-type thermocouple attached to a digital thermometer due to the short resting time required to achieve the target final endpoint temperature. Each stir fry batch was immediately placed into the labelled Pyrex container after cooking, transferred to the warming plate and sealed. This process was repeated for the six batches of stir fry required for each sensory session. Three pieces of stir fry were provided to each consumer for evaluation. Consumers record their responses using the computerised sensory evaluation program, Compusense 5 Version 5.2 (Guelph, Ontario, Canada). Individual demographic information was firstly captured, including gender, household size, age, current meat purchasing behaviour, cooking and consumption habits of fresh pork. Consumers were then presented with each pork sample for evaluation on a numbered plastic plate as per the randomized sensory design. Consumers assessed the eating quality attributes of the pork samples using a continuous 0–100 line scale as per Australian Standard for Sensory Analysis (2007). Word anchors were applied at each end of the line scale: aroma liking – 0 (dislike extremely) to 100 (like extremely); tenderness – 0 (not tender) to 100 (very tender); juiciness – 0: (not juicy) to 100 (very juicy); flavour – 0 (dislike extremely) to 100 (like extremely); overall liking – 0 (dislike extremely) to 100 (like extremely). Each consumer also graded the samples for quality into one of the following categories: 1: Unsatisfactory (this was terrible, I did not

H.A. Channon et al. / Meat Science 121 (2016) 216–227

enjoy it all); 2: Below average (this was not nice, I did not enjoy it); 3: Average (this pork was nice, I somewhat enjoyed it); 4: Above average (this pork was really nice, I enjoyed it) and 5; Excellent (this pork was excellent, I really enjoyed it). Each sample was also rated on a five point scale for repurchase intention by each consumer: 1 - I definitely would not buy it; 2 - I would probably not buy it; 3 - I might buy it; 4 - I would probably buy it and 5 - I would definitely buy it. Fail rate, expressed as the percentage of consumer responses that were graded as 1 or 2 for quality grade and re-purchase intention, was determined for each treatment. 2.4. Statistical analysis Analysis of variance using Genstat 13.1 (VSN International, 2010) was used to determine the effect of gender on growth performance, carcase attributes and levels of androstenone and skatole in subcutaneous fat and the effect of gender, ageing period and endpoint temperature, and their interactions, on objective meat quality attributes. Birth weight was included as a covariate in the analysis of gender effects on growth performance and carcase measurements. A block structure of day of slaughter was used for growth performance, carcase measurements and intramuscular fat. The analysis of variance for meat quality measures of muscle pH and temperature, meat colour, drip loss, WBSF and cooking loss included a block structure of side within slaughter day to account for the allocation of sides to the ageing period x endpoint temperature treatments. Chi-square analysis was used to determine gender effects on androstenone and skatole concentrations (Genstat 16, VSN International, 2013). Analysis of variance using R (R version 2.14.0, http://www.r-project. org) was used to determine the effect of treatment factors, and their interactions, on sensory attributes. Data was analysed as a 3 × 2 × 2 × 7 factorial study: gender (3 levels), ageing period (2 levels), endpoint temperature (2 levels) and cut type x cooking method (packs; 7 levels); a total of 84 treatment combinations. Cooking effects were analysed as a 2 (cooking method; roast or stir fry) × 3 (cut; shoulder, loin or silverside) factorial + 1 (loin steak). A total of 10 sides were evaluated per treatment with four consumer tastings per cut per side (n = 40 per treatment combination). Only pork from a single gender was assessed in a session to minimize any risks associated with boar taint of pork from entire male pigs affecting consumer assessment of samples of pork from other genders. For the sensory analysis, two pigs per gender were divided into two pairs. Each pair of four sides was allocated to one of four combinations of ageing (2 or 7 day post-slaughter) and endpoint temperature (70 or 75 °C) in one of three ways to account for the 10 pairs per gender. Components of the treatments therefore occurred with different accuracy in the different strata of the analysis. Within each sensory session, the two sides (from the same gender) were cooked to the same endpoint temperature to minimize risks of cooking errors and were from different ageing period treatments. Gender and endpoint temperature effects were therefore estimated as a ‘between session’ variance. The analysis was conducted in three stages: firstly, an analysis of the 120 means corresponding to the 120 sides (60 pigs × 2 sides); secondly, as an analysis of the 1120 observations for 20 pigs corresponding to pigs of one gender and, thirdly, all three genders were involved to provide expected mean squares for each of the variance terms (n = 3360 observations). Variance components corresponding to consumer panel sessions, between cut type x cooking method packs within a carcase, between sides, between pigs and between consumers were accounted for in the analysis. Differences between sessions, variation between sides and between pigs were small in relation to differences between consumers. The variance between samples (within individual consumers) was the highest component for each sensory attribute evaluated. Regression analysis was also used (R version 2.14.0, http://www.rproject.org) to predict overall liking of pork based on tenderness,

219

aroma, juiciness and flavour and to determine whether quality grading score and re-purchase intention could be predicted from these sensory variables. Chi square analysis was used to determine treatment effects on fail rates for both quality grade and re-purchase intention. 3. Results 3.1. Objective measurements Final liveweight and hot carcase weight was not influenced by gender (Table 2). Both entire male and female carcases were leaner at the P2 site (P b 0.001) than castrates. Dressing percentage of entire male carcases was 2% lower (P b 0.001) compared with female and castrate carcases. Both androstenone and skatole levels in subcutaneous fat from entire male pigs were higher (P b 0.001) than for castrates and females. Androstenone concentrations in subcutaneous fat exceeded 1000 ng/g in 20% of entire males, with 15% of entire males having levels of both androstenone (exceeding 1000 ng/g) and skatole (exceeding 200 ng/g) present in subcutaneous fat. Gender did not influence loin muscle pH measured from 45 min to 24 h post-slaughter. Loin muscle temperature was lower (P b 0.001) for entire males than castrates at each of these time periods (Table 3). In comparison to females, lower (P b 0.001) loin temperatures were also observed in entire males from 90 min to 24 h post-slaughter. Intramuscular fat content of loin muscles from castrates was higher (P = 0.009) than females and entire males (3.46%, 2.89% and 2.47%, respectively). For the silverside, muscles from castrated had higher (P = 0.002) intramuscular fat levels compared with entire males, with females intermediate. No differences between genders was observed for intramuscular fat content of the blade muscle. Muscle pH of the loin was influenced (P b 0.001) by the interaction of gender and ageing period, with muscle pH of loins from entire males increasing by 0.06 pH units when aged for 7 days compared with 2 days (Table 4). An interaction between gender and ageing period was also observed (P = 0.044) for muscle pH of the silverside, such that ageing for 7 days resulting in a consistent reduction in muscle pH for all three genders, with greatest differences being observed for entire males. Muscle pH of the blade was not influenced by gender or ageing period. Ageing for 7 days reduced (P = 0.016) muscle pH of the chuck tender compared with ageing for 2 days. Drip loss of the loin was not influenced by either gender or ageing period, whilst ageing for 7 days reduced (P b 0.001) drip loss of the silverside by 0.65% compared to 2 day aged muscles. The main effects of gender and ageing period influenced both muscle lightness and redness values of the loin and silverside, although differences were relatively small. For the silverside, ageing for 7 days increased (P b 0.001) yellowness compared with 2 day ageing (4.38 vs. Table 2 Predicted means and standard error of the difference (s.e.d.) for effect of gender on liveweight at various stages of growth, average daily gain and carcase parameters. Entire male

Female

Castrate

Grower-finisher weight Full final liveweight (kg)

52.7 96.4

55.0 96.3

55.8 97.8

s.e.d. 1.37 2.63

0.089 0.81

P2 fat depth (mm) HCW (kg) Dressing percentage (%) Cold carcase weight (kg)

9.9 73.1 75.8 63.5

10.2 75.5 78.5 66.0

14.7 76.1 77.7 66.4

0.88 2.19 0.64 1.95

b0.001 0.39 b0.001 0.29

Androstenone (ng/g) Skatole (ng/g) Androstenone N1000 ng/g (n) Skatole N200 ng/g (n) Boar taint prevalence⁎

768 185 4/20 6/20 3/20

n.d. 45 n.d. 1/20 n.d.

108 38 0/20 0/20 0/20

131.8 23.6

P value

b0.001 b0.001 0.035 0.007 0.072

n.d. not determined. ⁎ Number of pigs with androstenone concentrations N1000 ng/g and skatole concentrations N200 ng/g.

220

H.A. Channon et al. / Meat Science 121 (2016) 216–227

3.2. Sensory measurements

Table 3 Predicted means and standard error of the difference (s.e.d.) for effect of gender on muscle pH and temperature measured at the P2 site in the loin (M. longissimus thoracis) from 45 min to 24 h post-slaughter. Entire male pH 45 min 90 min 3h 6h 24 h Temperature 45 min 90 min 3h 6h 24 h

6.37 6.25 5.88 5.85 5.73

27.1 20.0 12.7 6.8 3.6

Female 6.47 6.30 5.89 5.88 5.70

27.6 21.7 14.2 8.1 4.1

Castrate 6.50 6.28 5.91 5.84 5.74

31.7 22.8 16.1 9.6 4.0

s.e.d.

P value

0.069 0.093 0.076 0.060 0.028

0.15 0.82 0.93 0.78 0.31

0.83 0.70 0.54 0.43 0.10

b0.001 b0.001 b0.001 b0.001 b0.001

Summary statistics for the sensory traits evaluated in this study are shown in Table 6.Outcomes from the base model are presented in Table A of the Supplementary Material and least square means for the effects of all experimental treatments investigated in this study on pork sensory attributes are presented in Tables B-J in the Supplementary Material. Pork from entire male pigs obtained lower scores for juiciness (P = 0.035), flavour (P = 0.017), overall liking (P = 0.018) and quality grade (P = 0.026) than castrates, with females intermediate (Table 7). Across all cuts, the fail rate of pork differed (P = 0.005) between genders, with the fail rate of 23.0% found for entire males across all treatments; 3.9% and 5.3% higher than from females and castrates, respectively. For re-purchase intention, fail rate differences due to gender were also significant (P = 0.005); with a fail rate of 28.7% for pork from entire males, compared with 23.6% for females and 23.5% for castrates. Neither ageing period nor endpoint temperature, as main effects, influenced any of the eating quality traits evaluated in this study. Although juiciness scores improved for pork from females aged for 7 days rather than 2 days (albeit only slightly), juiciness scores of 7 day aged pork from entire males were reduced by 5.3 units compared to 2 day aged pork (Table 8). For aroma, differences were only found for cut type x cooking method (Table 9). Overall, average aroma scores for roasts across all cuts were 12.2 units lower (P b 0.001) than stir-fry. Compared with the average scores of the other cut type x cooking method combinations, aroma of the pork loin steak was influenced by final endpoint temperature, with lower (P = 0.029) aroma scores obtained when cooked to 75 °C compared with 70 °C. The interaction between cut type x cooking method significantly influenced (P b 0.001) tenderness. Across all treatments, silverside roasts were rated lowest (P b 0.001) for tenderness compared with all other cut types x cooking methods evaluated. Roast and stir fry cuts from the shoulder were rated highest (P b 0.001) for tenderness, followed by loin and silverside. For the shoulder and silverside, average tenderness scores for roasts were lower (P b 0.001) than for stir fry (by 5.2 and 7.6 units, respectively), however, for loin tenderness scores for roasts were 2.8 units higher than stir fry.

3.19, respectively; s.e.d. 0.212). Blade muscles from female carcases were darker (lower L* value; P b 0.001) and redder (higher a* value; P b 0.001) than from entire male and castrates. Muscle lightness of the chuck tender was higher (P = 0.014) from entire males than both females and castrates, whilst females produced redder (P b 0.001) and yellower (P = 0.004) from females compared with castrates and entire males. Ageing for 7 days increased redness (P b 0.05) and yellowness (P b 0.001) of both the blade and chuck tender compared to ageing for 2 days. Cooking pork to a final endpoint temperature of 75 °C, rather than to 70 °C, increased WBSF values for both the loin (40.9 vs. 36.9 N, respectively; s.e.d. 2.81; P = 0.01) and the silverside (36.7 vs. 33.4 N, sed 1.79; P = 0.013) (Table 5). Higher cooking loss from the loin was found following cooking to a final endpoint temperature of 75 °C, rather than to 70 °C (21.5 vs. 18.3%, respectively; s.e.d. 0.547; P b 0.001) and also for the silverside (22.6 vs. 20.0%, for 75 °C and 70 °C. respectively; s.e.d. 0.531; P b 0.001). Loins from entire males had a higher (P = 0.005) cooking loss compared with females and castrates. Cooking losses from the silverside increased following ageing for 7 days compared with 2 days (22.0% vs. 20.6%, respectively; s.e.d. 0.531; P = 0.008).

Table 4 Predicted means and standard error of the difference (s.e.d) for effect of gender (G) and ageing period (AP; 2 or 7 days (d)) on muscle pH and meat colour (L*, a*, b*) of the loin (M. longissimus thoracis), silverside (M. biceps femoris), blade (M. triceps brachii) and chuck tender (M. supraspinatus), Gender

Entire male

Ageing period

2d

7d

2d

7d

2d

7d

5.41 52.36 4.58 2.44 3.53 2.47

5.47 53.53 4.67 3.13 3.18

5.48 51.19 5.39 2.24 3.12 2.89

5.44 52.00 6.12 3.12 2.49

5.48 52.33 4.45 2.19 2.96 3.46

5.46 53.69 4.88 3.46 2.83

5.53 49.77 7.08 3.35 2.48 1.76

5.42 51.11 7.62 4.54 1.86

5.50 48.36 7.30 3.07 2.80 2.15

5.45 50.04 7.75 4.22 1.79

5.50 49.58 6.83 3.16 2.10 2.79

5.61 44.83 11.38 3.88 1.78

5.58 46.17 11.65 5.45

5.62 43.19 12.69 3.87 1.81

5.64 42.53 13.36 5.14

5.78 43.82 14.58 4.58

5.76 43.60 15.13 5.46

5.75 42.43 15.71 4.89

5.68 42.65 16.21 6.05

Loin Muscle pH L value a value b value Drip loss IMF % Silverside Muscle pH L value a value b value Drip loss IMF % Blade Muscle pH L value a value b value IMF% Chuck tender Muscle pH L value a value b value

Female

Castrate

s.e.d.

P value G

AP

GxAP

0.018 0.822 0.414 0.282 0.619 0.317

0.044 0.027 b0.001 0.76 0.40 0.009

0.73 0.021 0.086 b0.001 0.30

b0.001 0.89 0.54 0.34 0.85

5.45 51.09 7.47 4.39 1.77

0.022 0.752 0.411 0.300 0.289 0.283

0.99 0.04 0.44 0.37 0.20 0.002

b0.001 b0.001 0.024 b0.001 b0.001

0.044 0.95 0.94 0.99 0.26

5.68 44.23 11.14 3.62 1.77

5.58 44.61 12.16 5.41

0.048 0.813 0.559 0.287 0.220

0.49 b0.001 b0.001 0.69 0.99

0.20 0.46 0.049 b0.001

0.21 0.22 0.66 0.45

5.75 42.56 14.15 4.22

5.69 41.71 15.30 5.23

0.036 0.775 0.421 0.314

0.081 0.014 b0.001 0.004

0.016 0.53 0.003 b0.001

0.46 0.62 0.50 0.82

H.A. Channon et al. / Meat Science 121 (2016) 216–227

221

Table 5 Predicted means and standard error of the difference (s.e.d.) for effect of gender (G), ageing period (AP) and endpoint temperature (T) on Warner-Bratzler shear force (WBSF, N) and cooking loss (%) from loin (M. longissimus thoracis) and silverside (M. biceps femoris) muscles. Gender

Entire male

Ageing period

2 days

Temperature

70 °C

75 °C

70 °C

75 °C

70 °C

75 °C

70 °C

75 °C

70 °C

75 °C

70 °C

75 °C

s.e.d.

G

AP

41.9 34.8

40.5 39.6

36.7 33.2

40.8 40.0

37.9 33.5

44.0 37.3

34.7 36.3

39.3 35.8

35.3 30.1

38.8 30.7

34.5 32.5

41.9 36.4

4.31 3.21

0.71 0.052

0.25 0.29

19.8 21.1

23.8 22.1

20.2 22.8

23.9 23.9

18.0 20.6

19.5 21.6

17.9 19.5

21.1 23.3

15.8 16.7

20.6 21.4

18.0 19.4

19.8 23.3

1.52 1.44

0.005 0.058

0.29 0.008

WBSF (N) Loin Silverside Cooking loss (%) Loin Silverside

Female 7 days

Castrate

2 days

7 days

Loin steaks obtained the lowest (P b 0.001) juiciness scores followed by silverside roast, silverside stir fry, loin stir fry and roast, shoulder roast and shoulder stir fry. Pork cooked as stir fry from both the shoulder and silverside was juicier (P = 0.001) than roasts (by 5.2 and 6.5 units, respectively), with no difference in juiciness scores between roast and stir fry from the loin. Cooking loin steaks to an endpoint temperature of 75 °C reduced (P = 0.027) juiciness scores compared with cooking to 70 °C compared with the average scores of the other cut type x cooking method combinations. Cut type x cooking method interactions also influenced (P b 0.001) flavour. Overall, lowest (P b 0.001) flavour scores were obtained for silverside roasts than all other cut type x cooking methods and for each primal, flavour scores for roasts were lower than for stir fry. Flavour scores for loin steaks cooked to a 75 °C endpoint temperature were 4.2 units lower (P = 0.027) than those cooked to 70 °C whilst there was no effect of temperature on the average scores of the other cut type x cooking method combinations. Average overall liking scores across cooking method (roast and stir fry) were highest (P b 0.001) for pork from shoulders, followed by loin and silverside. Across all treatments, loin steaks and silverside roasts had lower (P b 0.001) overall liking scores than the other five cut x cooking method treatments. Across all cuts, roasts obtained lower (P b 0.001) overall liking scores than stir fry. Cooking loin steaks to an endpoint temperature of 75 °C reduced (P = 0.006) overall liking scores by 5.6 units than cooking to 70 °C compared with the average scores of the other cut type x cooking method combinations. Higher (P b 0.001) quality grade scores were obtained for the stir fry and roast from the shoulder, followed by those from the loin and silverside. Across all cut types, quality grade scores were 0.27 units higher (P b 0.001) for stir fry than for roasts. When cooked to an endpoint temperature of 75 °C, quality grade scores of loin steaks were 0.50 units lower (P = 0.01) compared with the average of the other cut x cooking method combinations, with cooking to an endpoint temperature of 70 °C reducing this difference between loin steaks and all other cut type x cooking method combinations to 0.24 units.

P value

2 days

7 days T 0.01 0.013 b0.001 b0.001

Pork from shoulders obtained higher (P b 0.001) re-purchase intention scores than those from the loin and the silverside. Average re-purchase intention scores for stir fry cuts were 0.31 units higher (P b 0.001) than roasts. Re-purchase intention scores for loin steaks were 0.48 units lower than the other cut type x cooking method treatments. Re-purchase intention scores of loin steaks cooked to an endpoint temperature of 75 °C were 0.36 units lower than when cooked to 70 °C whilst there was no effect of endpoint temperature on the average scores for the other cut type x cooking method combinations. The fail rate percentages of each cut type x cooking method taking into account gender of the pig, ageing period and endpoint temperature for quality grade and re-purchase intention scores, respectively, are shown in Table 10. Quality grade fail rates of stir fry and roasts from the shoulder met the fail rate target of b10% (5.4% and 10.0%, respectively). Across all genders, the fail rate of loin steaks increased (P b 0.001) as endpoint temperature increased from 70 °C to 75 °C from 22.9% to 37.5%, respectively. Flavour, juiciness and tenderness were highly correlated with overall liking (R = 0.905, R = 0.818 and R = 0.808, respectively) across all of the cut types x cooking method treatments evaluated, with aroma moderately correlated with the other four sensory attributes (Table 11). Correlation coefficients between sensory attributes, shear force and intramuscular fat for each cut type x cooking method evaluated in this study are presented in Table J of the supplementary material. Tenderness and juiciness were also highly correlated (R = 0.811). In this study, overall liking of pork (across all cut types x cooking methods) was influenced, in order of importance, by flavour, tenderness, juiciness and aroma: Overall liking ¼ −0:774 þ 0:618  Flavour þ 0:235  Tenderness þ 0:156  Juiciness þ 0:019  Aroma All coefficients were statistically significant (P b 0.001). Table 7 Predicted means and standard error of the difference (s.e.d.) for the main effect of gender on consumer scores for pork eating quality traits and fail rate percentage (%) (determined as percentage of consumer evaluations of 1 or 2 for quality grade and re-purchase intention). Gender

Table 6 Summary statistics for sensory attributes of cooked pork aroma, tenderness, juiciness, flavour, overall liking, quality grade and re-purchase intention of pork (n = 3360). Sensory trait⁎

Median

Mean

SD

Ist quartile

3rd quartile

Aroma Tenderness Juiciness Flavour Overall liking Quality grade Re-purchase intention

62.0 54.0 55.5 61.0 59.0 3.00 4.00

59.1 52.6 53.3 58.4 57.1 3.32 3.40

23.4 27.7 26.7 24.3 25.1 1.01 1.23

42.5 29.5 32.0 41.5 38.0 3.00 2.00

77.5 76.0 75.0 77.1 77.0 4.00 4.00

⁎ Aroma liking, flavour liking and overall liking — 0 (dislike extremely) to 100 (like extremely); Tenderness — 0 (not tender) to 100 (very tender); Juiciness — 0: (not juicy) to 100 (very juicy); Quality grade — 1 (unsatisfactory) to 5 (excellent); and, Re-purchase intention — 1 (definitely would not buy) to 5 (definitely would buy).

Sensory trait⁎

Entire male

Female

Castrate

s.e.d.

P value

Aroma Tenderness Juiciness Flavour Overall liking Quality grade Re-purchase intention Fail rate (%) Quality grade Re-purchase intention

58.0 51.3 51.6 56.6 55.2 3.23 3.30

59.4 51.9 52.9 57.9 56.6 3.34 3.43

60.0 54.5 55.5 60.8 59.5 3.39 3.47

1.39 1.63 1.46 1.41 1.45 0.057 0.072

0.35 0.12 0.035 0.017 0.018 0.026 0.058

23.0 28.7

19.1 23.6

17.1 23.5

0.005 0.005

⁎ Aroma liking, flavour liking and overall liking – 0 (dislike extremely) to 100 (like extremely); Tenderness – 0 (not tender) to 100 (very tender); Juiciness – 0: (not juicy) to 100 (very juicy); Quality grade – 1 (unsatisfactory) to 5 (excellent); and, Re-purchase intention – 1 (definitely would not buy) to 5 (definitely would buy).

222

H.A. Channon et al. / Meat Science 121 (2016) 216–227

purchase intention (P b 0.01 for all coefficients in both regression equations):

Table 8 Predicted means and standard error of the difference (s.e.d.) for the interaction between gender and ageing period for sensory juiciness (where 0 = not juicy to 100 = very juicy). Ageing period (days)

Entire male

Female

Castrate

s.e.d.

P value

2 7

54.2 48.9

51.8 54.0

55.0 56.0

0.87

0.014

Quality grade score ¼ 1:389−0:001  Aroma þ 0:003  Tenderness þ 0:004  Flavour þ 0:029  Overall liking

Linear regression analysis of quality grade score using the five sensory traits assessed in this study identified that overall liking had the largest coefficient for both quality grade score and re-

Re−purchase intention ¼ 1:159−0:002  Aroma þ 0:004  Tenderness þ 0:001  Juiciness þ 0:038 Overall liking

Table 9 Predicted means and standard error of the difference (s.e.d.) for effect of ageing period (A), endpoint temperature (T) and cut type x cooking treatment (C) on consumer scores for aroma, tenderness, juiciness, flavour, overall liking, quality grade and re-purchase intention scores of pork shoulder (roast and stir fry), loin (roast, stir fry and steak) and silverside (roast and stir fry). Cut type x cooking method Shoulder Sensory attributea

Loin

Silverside

Ageing period (d)

Temp (°C)

Roast

Stir fry

Roast

Stir fry

Steak

Roast

Stir fry

s.e.d

P value†

2 7 2 7

70 70 75 75

54.2 52.3 54.9 50.7 53.1

62.8 64.6 63.3 64.3 63.7

52.6 46.7 52.9 51.7 51.0

68.1 67.9 63.4 65.7 66.3

64.6 63.2 58.5 58.6 61.2

57.7 51.5 54.5 52.2 54.0

67.0 65.1 63.5 62.9 64.6

1.13

b

1.34

C****

66.1 61.0 66.7 62.0 64.0

70.3 72.3 65.9 67.3 68.9

57.9 53.2 55.5 56.5 55.7

52.9 55.8 52.8 49.9 52.9

47.7 40.2 38.5 39.8 41.5

37.8 36.1 40.0 40.3 38.6

48.2 47.0 45.7 44.7 46.4

1.51

C_Stk****, C_R*** C_Cut****, C_RxCut****, TC_R*

1.70

C****

65.0 63.3 64.0 65.1 64.4

71.1 72.7 66.5 67.8 69.5

56.3 50.6 54.0 51.9 53.2

52.8 54.3 50.4 51.3 52.2

44.2 41.3 35.3 35.4 39.1

46.0 41.3 44.6 47.4 43.7

52.8 55.0 50.7 46.6 51.3

1.36

C_Stk****,C_R****, C_Cut****, C_RxCut**, TC* TC_Stk*, TC_R**, ATC_R*

1.68

C****

60.5 61.3 62.8 61.7 61.6

67.5 70.0 69.9 69.2 69.1

56.8 52.8 58.4 57.4 56.4

58.9 63.5 58.8 62.2 60.9

56.0 53.8 49.6 52.9 53.1

49.2 47.8 50.8 51.0 49.7

59.2 57.5 57.9 58.1 58.2

1.25

C_Stk****,C_R****, C_Cut****, TC_Stk*

1.40

C****

60.5 61.5 66.3 63.8 63.0

69.2 72.4 70.7 72.0 71.1

56.3 53.3 58.3 58.7 56.6

56.8 60.1 58.8 60.1 58.9

52.4 49.3 44.1 48.6 48.6

46.5 43.2 47.3 49.4 46.6

56.4 53.8 53.9 54.9 54.8

1.30

C_Stk****, C_R****, C_Cut****, C_RxCut**, TC_Stk**, TC_R (P = 0.052)

1.53

C****

Aroma

Average Tenderness 2 7 2 7

70 70 75 75

Average Juiciness 2 7 2 7

70 70 75 75

Average Flavour 2 7 2 7

70 70 75 75

Average Overall liking 2 7 2 7

70 70 75 75

Average Quality grade 2 7 2 7

70 70 75 75

Average Re-purchase intention 2 7 2 7 Average

70 70 75 75

C_Stk*, C_R****, C_RxCut**, TC_Stk*, AC_R**

3.51 3.54 3.65 3.58 3.57

3.89 3.98 3.90 3.90 3.92

3.30 3.18 3.34 3.33 3.29

3.30 3.52 3.35 3.40 3.39

3.17 3.07 2.81 2.96 3.00

2.85 2.78 2.89 2.89 2.85

3.26 3.21 3.14 3.26 3.22

0.055

C_Stk****, C_R****, C_Cut****, C_RxCut**, TC_Stk**

0.065

C****

3.69 3.64 3.80 3.67 3.70

4.06 4.12 3.99 4.10 4.07

3.39 3.36 3.46 3.41 3.40

3.41 3.61 3.44 3.52 3.49

3.27 3.06 2.79 2.82 2.99

2.79 2.80 2.90 2.88 2.84

3.42 3.29 3.16 3.33 3.30

0.070

C_Stk****,C_R****,C_Cut****, C_RxCut**, TC_Stk**

0.078

C****

C_Stk – differences between loin steak and average scores of all other cut type x cooking method treatments. C_R – differences between roast and stir fry across all muscles. C_Cut – differences between cut types (shoulder, loin and silverside) excluding loin steak. C_RxCut – interaction between cut type (loin, silverside and shoulder) x cooking method (roast or stir fry). TC – interaction between all cut type x cooking methods and endpoint temperature. TC_Stk – interaction between endpoint temperature and loin steak and the average scores of all other cut type x cooking method treatments. TC_R – interaction between endpoint temperature and roast and stir fry cuts. AC_R – interaction between ageing period and roast and stir fry cuts. ATC_R – interaction between ageing period, endpoint temperature and roast and stir fry cuts. † ⁎ P b 0.05, ⁎⁎P b 0.01, ⁎⁎⁎P b 0.001, ⁎⁎⁎⁎P b 0.0001. a Aroma liking, flavour and overall liking – 0 (dislike extremely) to 100 (like extremely); tenderness – 0 (not tender) to 100 (very tender); juiciness – 0: (not juicy) to 100 (very juicy); quality grade – 1 (unsatisfactory) to 5 (excellent); re-purchase intention – 1 (definitely would not buy) to 5 (definitely would buy). b Where: C – differences between the cut type x x cooking method treatments.

H.A. Channon et al. / Meat Science 121 (2016) 216–227

223

Table 10 Percentage of consumer evaluations of pork from different cuts and cooking method, gender (entire males, females and castrates), ageing period (2 or 7 days) and endpoint temperature (70 or 75 °C) obtaining scores of 1 or 2 (fail rate) for quality grade and re-purchase intention (n = 40 evaluations per treatment per cut). Gender

Entire male

Ageing Period

2 days

Endpoint temperature

70 °C

75 °C

70 °C

75 °C

70 °C

75 °C

70 °C

75 °C

70 °C

75 °C

70 °C

75 °C

within cut

12.5 7.5 35 17.5 32.5 42.5 22.5

2.5 2.5 17.5 22.5 35 37.5 20

15 2.5 30 12.5 17.5 45 27.5

12.5 12.5 27.5 20 40 47.5 27.5

15 5 25 12.5 30 42.5 10

12.5 12.5 17.5 12.5 32.5 35 32.5

10 2.5 10 7.5 27.5 30 20

5 0 12.5 10 42.5 35 17.5

12.5 7.5 12.5 32.5 17.5 30 20

5 7.5 17.5 15 37.5 30 20

5 2.5 12.5 5 12.5 32.5 20

12.5 2.5 12.5 15 37.5 25 20

10.0 5.4 19.2 15.2 30.2 36.0 21.5 P b 0.001

20 7.5 45 25 35 50 27.5

25 40 27.5 22.5 20 12.5 32.5

27.5 42.5 20 30 25 25 17.5

40 30 30 30 35 22.5 37.5

20 12.5 10 32.5 17.5 22.5 27.5

30 30 17.5 25 25 45 7.5

17.5 17.5 22.5 22.5 12.5 10 20

35 25 22.5 22.5 22.5 42.5 27.5

15 17.5 25 22.5 22.5 25 27.5

17.5 20 45 27.5 17.5 32.5 20

35 15 27.5 20 12.5 15 27.5

15 32.5 22.5 32.5 15 27.5 25

24.8 24.2 26.3 26.7 22.5 27.5 24.4 P = 0.538

Fail rates - quality grade Shoulder (Blade) roast Shoulder (Chuck tender) stir fry Loin roast Loin stir fry Loin steak Silverside roast Silverside stir fry P value Fail rates – re-purchase intention Shoulder (Blade) roast Shoulder (Chuck tender) stir fry Loin roast Loin stir fry Loin steak Silverside roast Silverside stir fry P value

Female 7 days

As the coefficient for overall liking was larger than for all other sensory attributes, quality grade score and re-purchase intention were each predicted using overall liking alone: Quality grade score ¼ 1:38 þ 0:034

   Overall liking SD 0:53; R2 ¼ 0:65

Re−purchase intention ¼ 1:10 þ 0:040

   Overall liking SD 0:70; R2 ¼ 0:55 :

4. Discussion Flavour liking accounted for the greatest variation in overall liking, followed by tenderness and juiciness; as observed in previous Australian sensory studies involving consumers (Channon, Kerr, & Walker, 2004; Channon et al., 2014). Whilst other studies have not reported outcomes from the prediction of consumer overall liking scores using flavour liking, tenderness, juiciness and/or aroma scores, Moeller et al. (2010) showed that consumer ratings of overall liking were positively correlated with flavour liking (R = 0.79), tenderness level (R = 0.68) and juiciness level (R = 0.63), with overall liking and flavour liking most strongly correlated with likelihood of purchase (R = 0.78 and R = 0.75, respectively). High correlations between overall liking and flavour (R = 0.65) and tenderness (R = 0.60) were also reported by Wood et al. (1996) following ageing of loin steaks for 10 days, In contrast, significant positive correlations between overall acceptance and tenderness (R = 0.71) and juiciness (R = 0.60), but not for flavour, were reported by Alonso, Campo, Provincial, Roncalés, and Beltrán (2010).

Table 11 Pearson correlation coefficients between sensory attributes of aroma, tenderness, juiciness, flavour and overall liking across all treatments.

Tenderness Juiciness Flavour Overall liking

Castrate

2 days

Aroma

Tenderness

Juiciness

Flavour

0.302 0.334 0.566 0.489

0.811 0.687 0.808

0.728 0.816

0.905

7 days

Average fail

2 days

7 days

rate (%)

4.1. Gender In this study, pork from castrates was juicier and had higher flavour liking scores compared with entire males, with higher consumer scores also obtained for overall liking and quality grade score (females were intermediate for these traits). In agreement with these outcomes, pork from castrates has been shown to be juicier (Font i Furnols et al., 2009; Nold, Romans, Costello, Henson, & Libal, 1997), have less unpleasant flavour and aroma (Jeremiah, Sather, & Squires, 1999), more flavourful and more acceptable when compared against younger boars (Cliplef & Strain, 1981) and/or more preferred (Banon, Andreu, Laencina, & Garrido, 2004) compared with entire males. Whilst pork from castrated males has been reported to be juicier (Bérard, Kreuzer, & Bee, 2010; Jeremiah & Weiss, 1984; Jonsäll, Johansson, & Lundström, 2001), more tender (Bereskin, Rough, & Davey, 1978; Jonsäll et al., 2001) and more acceptable (Ngapo, Riendeau, Laberge, & Fortin, 2012) than from females, this was not observed in this study. Furthermore, in this study, entire males did not have a stronger abnormal odour or less acceptable flavour than females in contrast to Brown, Warriss, Nute, Edwards, and Knowles (1998). Jeremiah et al. (1999) showed that texture of pork was not influenced by gender when entire males, females and castrates were slaughtered at an average liveweight of 96 kg, even when higher levels of unpleasant odours and flavours were detected from pork sourced from entire males. Pauly et al. (2010) also found that tenderness did not differ between pork from entire males, castrates and immunocastrated males. As variable effects on eating quality have been reported for gender, this suggests that other factors, including the levels of androstenone and/or skatole and consumer sensitivity to these compounds, also need to be considered in addition to gender alone (Font i Furnols, 2012). In this study, intramuscular fat levels in both the pork loin and silverside from castrated males averaged 1% higher compared with entire males, which may reflect increased fat deposition during the finishing period by castrates (Bonneau, 1998). As the gender effect on intramuscular fat content was relatively large in this study, this may contribute, in part, to explaining eating quality improvements associated with castrated males (Banon et al., 2004). Despite the higher intramuscular fat levels in loin and silverside muscles from castrated pigs compared with entire males in this study, sensory tenderness was not influenced by gender, in contrast to other studies (Font i Furnols et al., 2009; Nold et al., 1997); perhaps reflective of weak correlations between intramuscular fat and sensory tenderness (Huff-Lonergan et al., 2002).

224

H.A. Channon et al. / Meat Science 121 (2016) 216–227

Despite the 15% incidence of boar taint in entire males in this study (based on the thresholds levels of N 1000 ng/g androstenone and 200 ng/g skatole described by Hennessy et al. (1997)), the lack of gender effect on aroma may be explained by consumers not being exposed to volatiles released during cooking aroma as they were presented with pre-cooked samples held at 60 °C until required and/or pork cuts from each gender were evaluated in different sessions. Although pork from castrates was more acceptable and the average fail rate for quality grade and re-purchase intention was 5.3% and 6.2% lower, respectively, than from entire males, recommendations to include castrates into an eating quality system for pork will need to be carefully considered by the Australian pork industry due to castrated males being less efficient converters of feed to gain than other genders (subsequently producing carcases with higher levels of subcutaneous fat) as well as welfare issues associated with castration. Studies involving immunocastrated males are required to both ascertain the impact of various pathway factors as investigated in this study and establish whether immunocastrated males may be included into any eating quality system developed for pork; particularly given that immunization of entire male pigs against gonadotrophin releasing factor (GnRF) can effectively minimize risks associated with boar taint. 4.2. Ageing period Positive effects have previously been reported on sensory attributes of pork following ageing for 4 days (Rosenvold et al., 2002), 7 days (Channon, Baud, Kerr, & Walker, 2003; Channon et al., 2004; Channon et al., 2014; Moore et al., 2012; Taylor, Nute, & Warkup, 1995) and 10–12 days post-slaughter (Taylor et al., 1995; Wood et al., 1996) compared with ageing for 1–2 days. In this study, ageing period, as a main effect, did not influence consumer sensory scores or fail rates of any of the cut type x cooking method treatments that were investigated. It was expected that ageing would result in a significant, positive shift in eating quality acceptability, given that Wood et al. (1996) showed a greater improvement in tenderness following ageing for 10 days postslaughter, compared with 1 day ageing, than genotype (Duroc vs. Large White) and plane of nutrition (ad libitum vs. 80% ad libitum). In contrast to this study, where fail rates of 30.8% and 29.1% were found for loin steaks aged for 2 and 7 days, respectively, Channon et al. (2014) reported that ageing improved the fail rate from 26.3% for 2 day aged loin steaks to 18.0% for 7 day aged steaks, respectively. The lack of an ageing effect was surprising given previous studies conducted with this genetic line that demonstrated a positive effect on eating quality due to ageing in vacuum packaging for 7 days (Channon et al., 2003; Channon et al., 2014). It was not considered that the lack of effect of ageing period was influenced by the rate of muscle pH and temperature decline in the loin muscle from 45 min to 24 h as it may be regarded as normal (Warner, Kauffman, & Russell, 1993). It may be that ageing for 7 days may not have been long enough to result in consistent improvements in sensory quality for all cuts investigated. This is despite Dransfield, Jones, and MacFie (1980-81) showing that improvements in tenderness of pork are rapid during the first 1– 2 days post-slaughter and then continue at a slower pace to plateau at around 6 days post-slaughter. It is not known whether this is a widespread issue in Australia, limited to this supply chain or may have been a one-off event. It may be that Australia's closed herd together with selection pressure for pigs with increased rates of protein accretion, may have reduced proteolytic enzyme activity (limiting their effect on weakening of the myofibrillar lattice post-slaughter), influenced muscle fibre composition and/or density. Further work will be needed to understand those mechanisms that are influencing the ability of Australian pork to age as well as validate additional interventions to improve overall eating quality performance of pork. The severity of this issue for the Australian pork industry cannot be understated, as reflected by the inclusion of a minimum five-day ageing period in the Meat Standards Australia system for both beef and sheepmeat before

it can be sold to the consumer and requirements for 14 days and 21 days for particular beef cuts to achieve a higher grade (Meat and Livestock Australia, 2011; Meat and Livestock Australia, 2015). 4.3. Cut type, cooking method and effect of endpoint temperature 4.3.1. Roasting As silverside and loin roasts, irrespective of cooking temperature used, did not achieve the fail rate target of b 10% for either quality grade or re-purchase intention, further work is required to identify and implement effective interventions to improve their eating quality performance. This finding presents significant issues for industry, given that the silverside muscle is included in product specifications for fresh leg roasts sold at the retail level. Using beef semitendinosus, Christensen, Purslow, and Larsen (2000) considered that increased breaking strength of muscle fibres with increasing temperature may be due to heat denaturation of actin and other sarcoplasmic proteins which denature at 70–80 °C and 65 °C, respectively. The connective tissue content of the silverside has been shown to be higher than the loin (Boutten, Brazier, Morche, Morel, & Vendeuvre, 2000; Therkildsen et al., 2002) and higher concentrations of stable crosslinks per unit collagen have been found in muscles with higher total collagen content (Light, Champion, Voyle, & Bailey, 1985). Unfortunately, no assessment of collagen content and/or solubility was undertaken in this study so whilst it may be postulated that the proportion of heat-stable collagen crosslinks relative to heat-labile collagen crosslinks in the endomysium and perimysium, collagen concentration per unit width and/or overall collagen content may have been higher in silverside muscles compared with the loin, this cannot be confirmed as a potential explanation for the lower sensory scores for the silverside compared with the loin. It may be that different cooking instructions need to be communicated to consumers for cuts prepared from muscles with higher connective tissue contents than those with lower levels in order to optimize eating quality performance. Whilst Stanton and Light (1990) showed, in beef, that endomysial collagen may be damaged by proteolytic enzymes, particularly cathepsins, during ageing leading to weakening of the collagen matrix, it may be suggested that such effects in this study would have been minor due to the observed lack of ageing effect on sensory attributes of pork, regardless of cut type. The supplementation of pigs with up to 80 g/kg soy lecithin has been shown to affect collagen stability of the loin by reducing hardness, chewiness and/or cohesiveness values when objectively assessed (Akit et al., 2014; D'Souza, Mullan, Pethick, Pluske, & Dunshea, 2012; D'Souza et al., 2015), hydroxyproline levels (Akit et al., 2014), and gene expression of procollagen (Akit et al., 2011). Further studies are required to determine whether feeding soy lecithin to finisher pigs may provide opportunities to improve the eating quality performance of pork muscles with inherently high collagen content. All roasts in this study were cooked with subcutaneous fat and rind left on because Australian consumers typically prepare pork roasts in this way to also produce ‘crackling’ of the rind. It may be that differences in juiciness scores may have resulted from fat migration from subcutaneous fat remaining on the three primals when roasted, as the thickness of this fat layer is influenced by differences in anatomical locations from where the cuts were obtained as well as differences in fat deposition between genders. Improved juiciness of the blade roast may have also been attributable to lipid migration from melted intermuscular fat. As average intramuscular fat levels in the blade were 0.7–1.7 units lower than those of the loin, intramuscular fat was not considered to explain juiciness differences between the three primals. It may also be that flavour of roast cuts from the different primals may be influenced by variation in fatty acid composition of the different adipose depots in the animal. Unfortunately, fat samples were not retained to quantify this. Interestingly, Heymann et al. (1990) also found that the fatty acid composition of pork cooked to an endpoint temperature of 71.1 °C and higher contained more C14, C16:1, C18:1, C18:2 and C20 fatty acids

H.A. Channon et al. / Meat Science 121 (2016) 216–227

than that cooked to 65.6 °C. These higher levels of fatty acids in roasts cooked to higher temperatures was proposed by Heymann et al. (1990) to be due to the high level of retention and concentration of these fatty acids during cooking. Although cooking loss were higher when cooked in a water bath to 75 °C compared with 70 °C from both the loin and silverside (when prepared for WB shear force determination), juiciness scores for roasts cooked to 75 °C were less sensitive to endpoint temperature compared with stir frying or grilling. Wood et al. (1995) also showed that the effect of endpoint temperature (from 65 to 80 °C) on juiciness was less in roasts than in grilled steaks and suggested that this may have been due to the greater surface area of steaks. As tenderness and juiciness of roasts was not affected by endpoint temperature and flavour scores were improved at 80 °C, Wood et al. (1995) concluded that cooking to a well done degree of doneness by UK consumers would be acceptable. Heymann et al. (1990) reported that roasts sourced from the shoulder (blade), loin and boneless leg (containing the semimembranosus and adductor muscles) cooked to 65.6 or 71.1 °C were juicier than those cooked to 76.7 or 82.2 °C, but roasts cooked to 82.2 °C obtained higher sensory scores than those cooked to 65.6 or 71.1 °C. Siemens et al. (1990) reported that consumers preferred loin roasts cooked to 71.1 °C than those cooked to 76.7 °C whilst Zondagh, Holmes, Rowe, and Schrumpf (1986) found that juiciness of pork loin roasts declined with increasing temperature from 77 °C to 95 °C in an electric oven. Lower juiciness scores are considered to result from increased water being expelled from muscle due to myofibrillar protein denaturation and structural changes in muscle related to collagen shrinkage at higher endpoint temperatures (Davey & Gilbert, 1974). In this study, lower shear force values were found for pork loin and silverside cooked to 70 °C compared with 75 °C – similar observations for WB shear force when pork was cooked to 71.1 °C have reported by Siemens et al. (1990) (compared with 76.7 °C) and Crawford et al. (2010) (compared with 79 °C). As endpoint temperature, and moisture loss increased, increased percentage protein and lipid content of roasts (Heymann et al., 1990) and loin chops (Siemens et al., 1990) has been reported. Our data suggest that pork roasts may be cooked up to an endpoint temperature of 75 °C in a fan-forced oven set at 175 °C, as little differences in sensory scores or fail rates of roasts due to endpoint temperature (70 or 75 °C) were observed. Heymann et al. (1990) concluded that ‘the optimum endpoint temperature for fresh pork roasts should be at least 71.1°C and should not exceed 76.7°C, with 71.1°C being the most desirable endpoint temperature’. Saunders et al. (2000) recommended that large roasts should be cooked ‘to a moderately low endpoint temperature of 70–75 °C in a moderate to high oven (to reduce cooking times) set at 180–190 °C. 4.3.2. Grilling Compared with the average scores of the other cut types x cooking methods evaluated, higher juiciness, flavour, aroma and overall liking scores and lower fail rates were obtained for grilled loin steaks cooked to 70 °C, compared with 75 °C. Different cut types sourced from the same muscles therefore need to be cooked differently to optimize their eating quality (Wood et al., 1995). Saunders et al. (2000) reported that loin chops tended to become tougher as endpoint temperature increased due to loss of moisture and concluded that pork loin chops should be ‘cooked to as low an endpoint as possible of 65–70 °C in a moderate oven at 160 °C’. Boles, Parrish, Skaggs, and Christian (1991) also showed that grilling loin steaks to an endpoint temperature of 77 °C reduced both tenderness and juiciness compared with 71 °C. Siemens et al. (1990) also showed that loin chops were preferred when cooked to 71.1 °C compared to those cooked to 76.6 °C. Simmons et al. (1985) reported that as endpoint temperature of pork loin chops (cooked as either grilled or roasted) increased (60, 70 or 80 °C), moisture content, juiciness and tenderness scores were reduced and cooking loss increased. Prestat, Jensen, McKeith, and Brewer (2002), in a study evaluating moisture enhancement, cooking method

225

and endpoint temperature, found that grilling or frying loin chops to 80 °C rather than 70 °C reduced juiciness of non-enhanced chops, with flavour not influenced by endpoint temperature. Interestingly, Wood et al. (1995) found that grilling pork to an endpoint temperature of 80 °C increased flavour intensity compared with 72.5 °C, which may have been due to the increased concentration of Maillard reaction products on the cooked surface. Wood et al. (1995) concluded that, for steaks, the endpoint temperature of 72.5 °C resulted in pork that was more tender and juicier than when cooked to 80 °C and more flavourful steaks cooked to 65 °C. In a study comparing cooking method and endpoint temperature (65, 75 or 80 °C) of roasts (cooked in an oven at 90 °C) and steaks from the loin and silverside, Bejerholm and Aaslyng (2003) recommended cooking to an endpoint temperature of 75 °C to discriminate between meat samples in terms of its overall sensory properties. 4.3.3. Stir fry The higher scores for tenderness, juiciness and flavour of stir fried cuts, compared with roasted cuts, may reflect the shorter time needed to cook the product to the target endpoint temperature, maximizing juiciness and tenderness. In this study, slices presented to consumers from roasted pieces were trimmed however, for both stir fry and grill, outer cooked surfaces remained on pieces presented for consumer evaluation – which may have also influenced consumer perceptions of flavour. Lower sensory scores observed for stir fry when cooked to 75 °C compared with 70 °C, may reflect higher moisture losses perhaps due to the greater surface area exposed to heat. In this study, rigorous attention was made when cooking stir fry and it is recognized that this situation may not necessarily be repeatable in the home, as there may be a relatively high risk of overcooking stir fry pork pieces particularly if pork is cooked at the commencement of the preparation of a stir fried meal and remains in the pan rather than adding it once the other meal components are cooked. On a cut level, stir fry samples obtained from the chuck tender located in the shoulder performed best, with an overall fail rate of 5.4%. Despite this, as 24.2% of evaluations for this cut still rated either 1 or 2 for re-purchase intention, further effort is required to influence consumer purchasing behaviour for pork. 5. Conclusions Eating quality differences between the seven different cut type and cooking method treatments investigated in this study exceeded those of the other pathway interventions imposed. It was demonstrated that Australians should cook loin steaks to an endpoint temperature of 70 °C rather than 75 °C. It was also verified that eating quality assessments of one cut cannot be reliably used to determine effects of pathway interventions on eating quality consistency of other cuts, when cooked using different methods. Further work is therefore required to identify and verify alternate interventions that could be commercially implemented in order to consistently deliver high quality Australian pork with fail rates of b10% to consumers, in addition to determining whether the lack of ageing observed in this study is a significant issue. Sensory quality differences between muscles, cooking method used and endpoint temperature identified in this study indicate that further work is required to better understand the size of potential interactions between muscle, cut type, cooking method used and final endpoint temperature. This work, particularly including quality grades and repurchase intention, will provide opportunities to quantify the extent to which eating quality can be positively shifted by the implementation of particular pathway interventions by pork supply chain partners. The boar taint incidence of 15%, higher fail rates and lower scores for juiciness, flavour and overall liking of pork from entire males compared to females and castrates in this study indicates that the merit of including entire males in a pathway system focused on delivering pork of guaranteed high eating quality to consumers is questionable. Additional studies that quantify eating quality performance of pork cuts from

226

H.A. Channon et al. / Meat Science 121 (2016) 216–227

immunocastrated males compared with females are needed to provide industry with alternative options for the management of male pigs and enable pork to be produced that meets quality specifications as stipulated by an eating quality assurance system. This information will be used as part of the development of an eating quality model to deliver consistent, high quality pork to consumers through the implementation of effective pathway interventions by Australian pork supply chains. Acknowledgements Funding provided by the High Integrity Australian Pork Co-operative Research Centre is gratefully acknowledged (3 A-103). The authors also gratefully acknowledge Dr. Richard Jarrett, Jessica Tan and Karen McNaughton from the South Australian Research and Development Institute, Waite Campus, South Australia for biometrical services and support and for the co-ordination and conduct of the consumer sensory sessions. Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.meatsci.2016.06.011. References Aalhus, J. L., Schaefer, A. L., Murray, A. C., & Jones, S. D. M. (1992). The effect of ractopamine on myofibre distribution and morphology and their relation to meat quality in swine. Meat Science, 31, 397–409. Aalhus, J. L., Best, D. R., Costello, F., & Schaefer, A. L. (1997). The effects of porcine somatotropin on muscle fibre morphology and meat quality of pigs of known stress susceptibility. Meat Science, 45, 283–295. Aaslyng, M. D., & Barton Gade, P. (2001). Low stress pre-slaughter handling: Effect of lairage time on the meat quality of pork. Meat Science, 57, 87–92. Akit, H., Collins, C. L., Fahri, F. T., Hung, T. Y., D'Souza, D. N., Leury, B. J., & Dunshea, F. R. (2011). Dietary lecithin alters the expression of genes involved in skeletal muscle collagen synthesis and degradation. In R. J. van Barneveld (Ed.), Manipulating pig production XIII (pp. 63). Werribee: Australasian Pig Science Association. Akit, H., Collins, C. L., Fahri, F. T., Hung, A. T., D'Souza, D. N., Leury, B. J., & Dunshea, F. R. (2014). Dietary lecithin improves dressing percentage and decreases chewiness in the longissimus muscle in finisher gilts. Meat Science, 96, 1147–1151. Alonso, V., Campo, M. M., Provincial, L., Roncalés, P., & Beltrán, J. A. (2010). Effect of protein level in commercial diets on pork meat quality. Meat Science, 85, 7–14. Aluwé, M., Millet, S., Nijs, G., Tuyttens, F. A. M., Verheyden, K., De Brabander, H. F., ... Van Oeckel, M. J. (2009). Absence of an effect of dietary fibre or clinoptilolite on boar taint in entire male pigs fed practical diets. Meat Science, 82, 346–352. Australian Standard for Sensory Analysis AS2542 (2007). Part 2.3: Specific methods – guidelines for the use of quantitative response scales (rating) AS2542.2.3. . Banon, S., Andreu, C., Laencina, J., & Garrido, M. D. (2004). Fresh and eating pork quality from entire versus castrate heavy males. Food Quality and Preference, 15, 293–300. Barton Gade, P. (2008). Effect of rearing system and mixing at loading on transport and lairage behaviour and meat quality: Comparison of outdoor and conventionally raised pigs. Animal, 2, 902–911. Bejerholm, C., & Aaslyng, M. D. (2003). The influence of cooking technique and core temperature on results of a sensory analysis of pork – depending on the raw meat quality. Food Quality and Preference, 15, 19–30. Bérard, J., Kreuzer, M., & Bee, G. (2010). In large litters birth weight and gender is decisive for growth performance but less for carcass and pork quality traits. Meat Science, 86, 845–851. Bereskin, B., Rough, D. K., & Davey, R. J. (1978). Some factors affecting the evaluation of pork quality. Journal of Animal Science, 47, 389–397. Boles, J. A., Parrish, F. C., Jr., Skaggs, C. L., & Christian, L. L. (1991). Effect of porcine somatotropin, stress susceptibility, and final end point of cooking on the sensory, physical, and chemical properties of pork loin chops. Journal of Animal Science, 69, 2865–2870. Bonneau, M. D. (1998). Use of entire males for pig meat in the European Union. Meat Science, 49(Supplement 1), S257–S272. Bouton, P. E., Harris, P. V., & Shorthose, W. R. (1971). Effect of ultimate pH upon the water holding capacity and tenderness of mutton. Journal of Food Science, 36, 435–439. Boutten, B., Brazier, M., Morche, N., Morel, A., & Vendeuvre, J. L. (2000). Effects of animal and muscle characteristics on collagen and consequences for ham production. Meat Science, 55, 233–238. Brown, S. N., Warriss, P. D., Nute, G. R., Edwards, J. E., & Knowles, T. G. (1998). Meat quality in pigs subjected to minimal preslaughter stress. Meat Science, 49, 257–265. Channon, H. A., & Warner, R. D. (2011). Delivering consistent quality Australian pork to consumers – a systems approach. In R. J. van Barneveld (Ed.), Manipulating pig production XIII (pp. 262–293). Werribee: Australasian Pig Science Association. Channon, H. A., Baud, S. R., Kerr, M. G., & Walker, P. J. (2003). Effect of low voltage electrical stimulation of pig carcasses and ageing on sensory attributes of fresh pork. Meat Science, 65, 1315–1324.

Channon, H. A., Kerr, M. G., & Walker, P. J. (2004). Effect of Duroc content, sex and ageing period on meat and eating quality attributes of pork loin. Meat Science, 66, 881–888. Channon, H. A., Taverner, M. R., D'Souza, D. N., & Warner, R. D. (2014). Aitchbone hanging and ageing period are additive factors influencing pork eating quality. Meat Science, 96, 581–590. Channon, H. A., Hamilton, A. J., D'Souza, D. N., & Dunshea, F. R. (2016). Estimating the impact of various pathway parameters on tenderness, flavour and juiciness of pork using Monte Carlo simulation methods. Meat Science, 116, 58–66. Christensen, M., Purslow, P. P., & Larsen, L. M. (2000). The effect of cooking temperature on mechanical properties of whole meat, single muscle fibres and perimysial connective tissue. Meat Science, 55, 301–307. Cliplef, R. L., & Strain, J. H. (1981). Relationship of sex odor to panel acceptability and certain other organoleptic characteristics of pork chops. Canadian Journal of Animal Science, 61, 45–52. Crawford, S. M., Moeller, S. J., Zerby, H. N., Irvin, K. M., Kuber, P. S., Velleman, S. G., & Leeds, T. D. (2010). Effects of cooked temperature on pork tenderness and relationships among muscle physiology and pork quality traits in loins from Landrace and Berkshire swine. Meat Science, 84, 607–612. D'Souza, D. N., Blake, B., Williams, I., Mullan, B., Pethick, D., & Dunshea, F. (2015). Dietary lecithin supplementation can improve the quality of the M. Longissimus thoracis. Animals, 5, 1180–1191. Davey, G. L., & Gilbert, K. V. (1974). Temperature dependent cooking toughness in beef. Journal of Science of Food and Agriculture, 25, 931–938. Dransfield, E., Jones, R., & MacFie, H. (1980-81). Tenderising in M. longissimus dorsi of beef, veal, rabbit, lamb and pork. Meat Science, 5, 139–147. D'Souza, D. N., & Mullan, B. P. (2002). The effect of genotype, sex and management strategy on the eating quality of pork. Meat Science, 60, 95–101. D'Souza, D. N., Dunshea, F. R., Hewitt, R. J. E., Luxford, B. G., Meaney, D., Schwenke, F., ... van Barneveld, R. J. (2011). High boar taint risk in entire male carcases. In R. J. van Barneveld (Ed.), Manipulating pig production XIII (pp. 259). Werribee, Australia: Australasian Pig Science Association. D'Souza, D. N., Mullan, B. P., Pethick, D. W., Pluske, J. R., & Dunshea, F. R. (2012). Nutritional strategies affect carcass and pork quality but have no effect on intramuscular fat content of pork. Animal Production Science, 52, 276–282. Ellis, M., Brewer, M. S., Sutton, D. S., Lan, H. Y., Johnson, R. C., & McKeith, F. K. (1998). Aging and cooking effects on sensory traits of pork from pigs of different breed lines. Journal of Muscle Foods, 9, 281–291. Font i Furnols, M. (2012). Consumer studies on sensory acceptability of boar taint: A review. Meat Science, 92, 319–329. Font i Furnols, M., González, J., Gispert, M., Oliver, M. A., Hortós, M., Pérez, J., ... Guerrero, L. (2009). Sensory characterization of meat from pigs vaccinated against gonadotropin releasing factor compared to meat from surgically castrated, entire male and female pigs. Meat Science, 83, 438–442. Hennessy, D. P., Colantoni, C., Dunshea, F. R., Howard, K., Jackson, P., Long, K., ... Walker, J. (1997). Elimination of boar taint: A boar vaccine for male pigs. Manipulating pig production VI (pp. 143) (Werribee, Victoria). Heymann, H., Hedrick, H. B., Karrasch, M. A., Eggeman, M. K., & Ellersieck, M. R. (1990). Sensory and chemical characteristics of fresh pork roasts cooked to different endpoint temperatures. Journal of Food Science, 55, 613–617. Huff-Lonergan, E., Baas, T. J., Malek, M., Dekkers, J. C., Prusa, K., & Rothschild, M. F. (2002). Correlations among selected pork quality traits. Journal of Animal Science, 80, 617–627. Jeremiah, L. E., & Weiss, G. M. (1984). The effects of slaughter weight and sex on the cooking losses from and palatability attributes of pork loin chops. Canadian Journal of Animal Science, 64, 39–43. Jeremiah, L. E., Sather, A. P., & Squires, E. J. (1999). Gender and diet influences on pork palatability and consumer acceptance. I. Flavor and texture profiles and consumer acceptance. Journal of Muscle Foods, 10, 305–316. Jonsäll, A., Johansson, L., & Lundström, K. (2001). Sensory quality and cooking loss of ham muscle (M. biceps femoris) from pigs reared indoors and outdoors. Meat Science, 57, 245–250. Light, N., Champion, A. E., Voyle, C., & Bailey, A. J. (1985). The role of epimysial, perimysial and endomysial collagen in determining texture in six bovine muscles. Meat Science, 13, 137–149. Meat and Livestock Australia (2011). Tips and tools. Meat Standards Australia beef information kit. How ageing affects beef eating quality (pp. 31–32). Meat and Livestock Australia (Available at http://www.mla.com.au/Marketing-beef-and-lamb/MeatStandards-Australia/MSA-beef/Grading [Verified May 2016]). Meat and Livestock Australia (2015). Tips and tools. Meat Standards Australia sheepmeat information kit (pp. 11) (Available at: http://www.mla.com.au/Marketing-beef-andlamb/Meat-Standards-Australia/MSA-sheepmeat [Verified May 2016]). Moeller, S. J., Miller, R. K., Edwards, K. K., Zerby, H. N., Logan, K. E., Aldredge, T. L., ... BoxSteffensmeier, J. M. (2010). Consumer perceptions of pork eating quality as affected by pork quality attributes and end-point cooked temperature. Meat Science, 84, 14–22. Moore, K. L., Mullan, B. P., & D'Souza, D. N. (2012). The interaction between ractopamine supplementation, porcine somatotropin and moisture infusion on pork quality. Meat Science, 92, 125–131. Ngapo, T. M., Riendeau, L., Laberge, C., & Fortin, J. (2012). “chilled” pork — Part II. Consumer perception of sensory quality. Meat Science, 92, 338–345. Nold, R. A., Romans, J. R., Costello, W. J., Henson, J. A., & Libal, G. W. (1997). Sensory characteristics and carcass traits of boars, barrows, and gilts fed high- or adequate-protein diets and slaughtered at 100 or 110 kilograms. Journal of Animal Science, 75, 2641–2651. Pauly, C., Spring-Staehli, P., O'Doherty, J. V., Kragten, S. A., Dubois, S., Messadene, J., & Bee, G. (2010). The effects of method of castration, rearing condition and diet on sensory quality of pork assessed by a trained panel. Meat Science, 86, 498–504.

H.A. Channon et al. / Meat Science 121 (2016) 216–227 Pleasants, A. B., Thompson, J. M., & Pethick, D. W. (2005). A model relating a function of tenderness, juiciness, flavour and overall liking to the eating quality of sheep meat. Australian Journal of Experimental Agriculture, 45, 483–489. Polkinghorne, R., Watson, R., Thompson, J. M., & Pethick, D. W. (2008). Current usage and future development of the Meat Standards Australian (MSA) grading system. Australian Journal of Experimental Agriculture, 48, 1459–1464. Prestat, C., Jensen, J., McKeith, F. K., & Brewer, M. S. (2002). Cooking method and endpoint temperature effects on sensory and color characteristics of pumped pork loin chops. Meat Science, 60, 395–400. Rasmussen, A. J., & Andersson, M. (1995). New method for determination of drip loss in pork muscles. Proceedings of 42nd International Congress of Meat Science and Technology. 2-6 September 1995 (pp. 286–287). Norway: Lillehammer. Rosenvold, K., Lærke, H. N., Jensen, S. K., Karlsson, A. H., Lundström, K., & Andersen, H. J. (2002). Manipulation of critical quality indicators and attributes in pork through vitamin E supplementation, muscle glycogen reducing finishing feeding and preslaughter stress. Meat Science, 62, 485–496. Sather, A. P., Jones, S. D. M., Schaefer, A. L., Colyn, J., & Robertson, W. M. (1997). Feedlot performance, carcass composition and meat quality of free-range reared pigs. Canadian Journal of Animal Science, 77, 225–232. Saunders, A., Wilkinson, B. H. P., & Hall, C. (2000). Optimisation of cooking procedures for pork (UNZ 4/1386). Final report to Australian Pork Limited, Canberra ACT. Siemens, A. L., Heymann, H., Hedrick, H. B., Karrasch, M. A., Ellersieck, M., & Eggeman, M. K. (1990). Effects of external fat cover, bone removal and endpoint cooking temperature on sensory attributes and composition of pork center loin chops and roasts. Journal of Sensory Studies, 4, 179–188. Simmons, S. L., Carr, T. R., & McKeith, F. K. (1985). Effect of internal temperature and thickness on palatability of pork loin chops. Journal of Food Science, 50, 313–315.

227

Stanton, C., & Light, N. (1990). The effects of conditioning on meat collagen: Part 3—Evidence for proteolytic damage to endomysial collagen after conditioning. Meat Science, 27, 41–54. Taylor, A. A., Nute, G. R., & Warkup, C. C. (1995). The effect of chilling, electrical stimulation and conditioning on pork eating quality. Meat Science, 39, 339–347. Therkildsen, M., Riis, B., Karlsson, A., Kristensen, L., Ertbjerg, P., Purslow, P. P., ... Oksbjerg, N. (2002). Compensatory growth response in pigs, muscle protein turn-over and meat texture: Effects of restriction/realimentation period. Animal Science, 75, 367–377. Tikk, K., Tikk, M., Aaslyng, M. D., Karlsson, A. H., Lindahl, G., & Andersen, H. J. (2007). Significance of fat supplemented diets on pork quality - connections between specific fatty acids and sensory attributes of pork. Meat Science, 77, 275–286. Warner, R. D., Kauffman, R. G., & Russell, R. L. (1993). Quality attributes of major porcine muscles: A comparison with the Longissimus Lumborum. Meat Science, 33, 359–372. Watson, R., Polkinghorne, R., & Thompson, J. M. (2008). Development of the Meat Standards Australia (MSA) prediction model for beef palatability. Australian Journal of Experimental Agriculture, 48, 1368–1379. Wood, J. D., Nute, G. R., Fursey, G. A. J., & Cuthbertson, A. (1995). The effect of cooking conditions on the eating quality of pork. Meat Science, 40, 127–135. Wood, J. D., Brown, S. N., Nute, G. R., Whittington, F. M., Perry, A. M., Johnson, S. P., & Enser, M. (1996). Effects of breed, feed level and conditioning time on the tenderness of pork. Meat Science, 44, 105–112. Zondagh, I. B., Holmes, Z. A., Rowe, K., & Schrumpf, D. E. (1986). Prediction of pork and lamb meat quality characteristics. Journal of Food Science, 51, 40–46.