Full Paper
Descriptive Sensory Analysis of Japanese Jidori Chicken (Choshu-Kurokashiwa) Thigh Meat: Sensory Attributes Distinguishing it from Broiler Chickens
2025 年 62 巻 論文ID: 2025025
詳細
2025 年 62 巻 論文ID: 2025025
This study aimed to characterize the sensory attributes of Japanese jidori thigh meat, such as texture, odor, and flavor/taste, and to identify the instrumental parameters that distinguish it from commercial broiler chickens. Six chicken thigh meats were used: one jidori-niku (Choshu-Kurokashiwa, 14-week old) and five broilers—three Ross 308 (7, 7, 9-week old), one Hubbard ColorPac (10-week old), and one Hubbard RedBro (10-week old)—from different producers. Descriptive sensory analysis was conducted with a trained panel to establish a comprehensive sensory lexicon and quantify attribute intensities of thigh samples cooked in a steam convection oven at 185°C. Instrumental assessments included cooking loss, shear force value, pH, inosine-5’-monophosphate, free amino acid content, and fatty acid composition. Choshu-Kurokashiwa meat exhibited greater “springiness” and reduced “tenderness” (p < 0.001), along with higher shear force value (p < 0.001) than broiler meat. Choshu-Kurokashiwa meat received also the highest sensory scores for “meaty odor,” “meaty flavor,” “umami taste,” and “aftertaste intensity.” While Choshu-Kurokashiwa meat had lower free amino acid levels, it was enriched in arachidonic acid (p < 0.05). Principal component analysis revealed a clear separation between Choshu-Kurokashiwa meat and the five broilers in terms of “springiness,” “tenderness,” “meaty flavor,” “light taste,” “umami taste,” and “aftertaste intensity.” These results show that Choshu-Kurokashiwa meat possesses distinct texture and flavor/taste characteristics compared to commercial broiler meat.
In Japan, where broilers account for over 90% of the chicken meat market share[1], jidori-niku (Japanese indigenous chicken meat) plays an essential role in maintaining diversity. Jidori-niku is certified by Japan’s Ministry of Agriculture, Forestry, and Fisheries under Japanese Agriculture Standard 844 (JAS844)[2], which defines a product based on genetic characteristics, feeding duration and conditions, and stocking densities. Hence, jidori-niku produced according to JAS844 has established itself as a distinct category within the Japanese chicken meat market.
Consumers are willing to pay a premium for jidori-niku in anticipation of a distinctive culinary experience[3]. Abe et al. reported that consumers associated “jidori-niku-like qualities” with attributes clearly distinct from general palatability in chicken meat[4]. Similarly, Hikichi et al. indicated that consumers expected jidori-niku to deliver exceptional texture, taste, and flavor[5]. To meet consumer expectations, jidori producers and breeders must possess deep knowledge of jidori-niku characteristics.
Descriptive analysis, a systematic method for objectively defining food attributes, is widely recognized as an effective means for the sensory characterization of livestock products. It employs trained panels to objectively quantify sensory properties through structured sessions of lexicon generation, capturing the sample’s overall characteristics and bridging subjective perceptions with measurable parameters[6]. Previous studies have applied extensively descriptive analysis to characterize sensory properties across chicken meat products[7,8,9,10,11]. For example, Sasaki et al. used it to identify “springiness” as a defining textural trait of jidori-niku[12]. Nevertheless, it has rarely been employed to fully characterize the odor and flavor/taste of jidori-niku, resulting in incomplete sensory profiles.
At present, there is no exact lexicon describing jidori-niku’s sensory qualities. Matsuishi et al. identified breed-specific aromas in Nagoya-Cochin chickens, but did not qualitatively characterize these aromas[13]. The author previously reported significantly higher sensory intensities for certain attributes in jidori thigh meat compared to broiler chickens[14]. However, the attributes used in the sensory evaluation might be inadequate for fully describing jidori-niku, because the study did not include sensory lexicon generation sessions. To clarify which attributes are perceived as distinctive in jidori-niku, sensory evaluation including lexicon generation by a trained panel is required.
In addition, when making sensory evaluations, it is essential to replicate real-world consumption contexts. In particular, previous studies have focused on taste, but many have been inadequate for this reason. For instance, the author’s previous study compared broths from jidori-niku and broiler meat, and demonstrated a stronger sour taste in jidori breast meat[15]. However, the study used a cooking method not generally employed by consumers. Sensory evaluation should employ cooked samples using a method as similar as possible to that used in the real world.
Furthermore, most studies have attempted to elucidate the sensory properties of jidori-niku indirectly via physicochemical properties, such as free amino acid content and fatty acid composition[16,17,18,19]. While these instrumental analyses provide valuable chemical insights, they cannot fully capture the complex sensory experiences perceived by consumers when eating chicken meat. A comprehensive sensory profile of jidori-niku requires in-depth understanding that combines the results of direct sensory evaluation as well as indirect instrumental analysis.
This study aimed to thoroughly characterize the sensory attributes of jidori-niku, examining its texture, odor, and flavor/taste. To this end, I conducted a descriptive sensory analysis using a trained panel, established a sensory lexicon, and quantified the attribute intensities of jidori-niku and broiler chickens. Choshu-Kurokashiwa jidori thigh meat was compared to that of five broiler breeds selected to represent diverse genetic backgrounds and feeding practices. This ensured that the unique traits distinguishing jidori-niku from commercial broilers were robustly identified. To mimic a realistic consumption context, I adopted oven-cooked thigh meat for descriptive analysis. Measurements of cooking loss, shear force value, pH, free amino acid and inosine-5’-monophosphate (IMP) contents, and fatty acid composition were also performed. By integrating sensory and instrumental data, this study identified the characteristics that distinguish jidori-niku from commercial broiler meat, offering actionable insights for breeders and marketers.
Meat samples comprising the thigh and drumstick were obtained from one JAS844-fed jidori chicken breed (Choshu-Kurokashiwa, (Shamo × (Kurokashiwa × Rhode Island Red)) × White Plymouth Rock) × Rhode Island Red) and five commercial broiler samples differing in age and breed type (Table 1). Information about the samples was obtained by interviewing chicken processors. None of the samples were frozen until obtained from slaughter. Choshu-Kurokashiwa and broiler-A leg meat samples were obtained from chickens slaughtered in the same processing facility on the same day. The other broilers (B–E) were sourced from different processors. Each sample was weighed, vacuum-packed in polyethylene bags, and stored at -30°C until analysis.
| Jidori-niku | Broiler-A | Broiler-B | Broiler-C | Broiler-D | Broiler-E | |
| Breed | Choshu-Kurokashiwa | Ross 308 | Ross 308 | Ross 308 | Hubbard ColorPac | Hubbard Redbro |
| Slaughter age (weeks) | 14 | 7 | 7 | 9 | 10 | 10 |
| Sex | Male | not identified | not identified | not identified | not identified | not identified |
| Weight of leg meat (g)1 | 353±30 | 261±39 | 277±31 | 337±57 | 368±57 | 248±47 |
1 n= 18 for Choshu-Kurokashiwa; n= 18 for broiler-A; n= 28 for broiler-B; n= 24 for broiler-C; n= 11 for broiler-D; n= 17 for broiler-E.
For sensory evaluation, chicken legs were thawed in a refrigerator at 1°C for 24 h. Legs were separated at the knee joint, and only the thigh was used for analysis. Four standardized meat pieces (25 × 15 × 30 mm) were excised from the skinned thigh of each sample, as described by Aoya et al.[20] with some modifications. Preliminary observations indicated some variation in lipid and connective tissue distribution, depending on the cutting location. To minimize the effect of variability, each panelist consistently evaluated samples taken from the same anatomical location within one session. Samples were heated in a steam convection oven (SSC-20SC; Maruzen, Tokyo, Japan) at 185°C for 5 min while exposed to steam, and then for another 5 min after changing positions to ensure uniform heating. This method was determined after prior testing of cooking conditions that allowed for the core temperature of the samples to reach 75°C. Cooked samples were then placed into polypropylene cups, covered with polyethylene terephthalate lids, and held in a Styrofoam container maintained at 60°C until evaluation. The samples were evaluated within 90 min of cooking.
General procedures for sensory evaluationDescriptive sensory analysis was conducted using a screened and trained panel, following the methodology of Lawless and Heymann[6]. Sensory evaluations were performed under fluorescent lighting with 2 m spacing between panelists. All sensory procedures complied fully with the relevant national regulations, institutional policies, and ethical principles outlined in the Declaration of Helsinki.
Screening of candidatesSensory panel screening was performed in two substages: an initial recognition test for five basic tastes, and a discrimination test for odor characteristics[21]. Twenty candidates were recruited from AKIKAWA FOODS & FARMS Co. Ltd. (Yamaguchi-shi, Yamaguchi, Japan). Four panel members (three females and one male, aged 24–47 years) who correctly answered all screening questions were selected.
TrainingThe training session relied on the same six chicken thigh meat sample types used also for the evaluation sessions (Table 1). The panelists developed sensory attributes, definitions, and references during seven training sessions, each lasting 30 min. In the initial session, panelists received chicken thigh samples and selected appropriate sensory attributes from a Japanese lexicon containing 42 candidate attributes prepared by researchers based on previous studies[22,23]. In the second session, the panelists refined the initial list of candidate attributes. From the third to the sixth sessions, the panelists were presented with samples and potential reference foods to clearly define the attributes and assign suitable references. Adjustments to the lexicon continued until the panelists reached a consensus. The final lexicon consisted of 12 attributes for odor, texture, and flavor/taste (Table 2). “Odor” and “flavor” were defined as orthonasal and retronasal odor, respectively. Panelists first evaluated the odor and then ate the samples to assess texture and flavor/taste. During the final training session, cooked meat samples and selected references were provided under conditions identical to those in the actual testing. The panelists evaluated the samples and provided feedback to confirm the suitability of their attributes and references.
| Attribute | Japanese | Definition and reference1 | |
| Odor2 | |||
| Rancid-like odor | abura-kusai | Oxidized rapeseed oil-like odor | |
| Meaty odor | niku-rashii | Meaty odor of seared red beef meat without charring | |
| Pan-fried chicken-like odor | koubashii-niku | Odor of pan-fried broiler chicken thigh with crispy skin | |
| Texture | |||
| Springiness | danryoku-sei | The force and degree of recovery from a deforming: weakest = silken tofu, strongest = gummy candy | |
| Plump | fukkura | A texture characterized by the meat being softly swollen while maintaining its structure, exhibiting a gentle resistance when compressed: weakest = dried squid, strongest = grilled mackerel | |
| Tenderness | yawarakai | The force required to deform a sample during the first to second masticatory cycles: weakest = hardtack biscuit, strongest = overcooked udon noodles | |
| Juiciness | jūshī | Amount of fluids released from the surface and body of a sample: weakest = sugar cookie, strongest = mandarin orange | |
| Flavor/taste | |||
| Meaty flavor | niku-rashii | Meaty flavor of seared red beef meat without charring | |
| Light taste | tanpaku-na | Mild and light taste, like boiled chicken pectoralis minor muscle | |
| Fatty taste | aburappoi | Perception of lipids coating the oral cavity taste, like fried chicken skin | |
| Umami taste | umami | Umami taste of five basic tastes, like monosodium glutamate | |
| Aftertaste intensity | ato-aji | Taste in the mouth after swallowing, like Gouda cheese. | |
1 References and attribute descriptions were selected by consensus of the trained panel.
2 "Odor" and "flavor" mean orthonasal and retronasal odor, respectively.
Evaluations were conducted over nine sessions. Four trained panelists participated in the evaluations, with each panelist subsequently attending seven, eight, eight, and nine sessions, resulting in 32 individual assessments. In each evaluation session, the panelists assessed five chicken samples: Choshu-Kurokashiwa, broiler-A, broiler-B, broiler-C, and either broiler-D (five sessions) or broiler-E (four sessions). Due to limited sample availability, broilers-D and -E could not be included in all sessions. All sessions began at 1:30 or 2:30 p.m., wherein the panelists received instructions regarding the evaluation procedures, sensory attributes, and their definitions. The service order was determined using a Latin square design. Sensory attributes were scored using a 15-cm unstructured line scale anchored with verbal attributes: odor and flavor/taste attributes ranged from “not perceived” (left) to “extremely strong” (right), while textural attributes ranged from “extremely weak” (left) to “extremely strong” (right). The panelists marked their perceptions on the line, and attribute intensities were quantified by measuring the distance from the left anchor to the panelist’s mark.
Instrumental analysisThigh meat samples separated from those used for sensory evaluations were analyzed instrumentally (n = 8 for Choshu-Kurokashiwa, broilers-A, -B, and -C; n = 4 for broiler-D; and n = 7 for broiler-E). The cooking loss and shear force value of the biceps femoris muscle, a major muscle of the thigh that allows for uniform sampling, were measured using a method[24] common in meat quality assessment in Japan[25,26,27]. The biceps femoris muscle was excised from the thigh and cut into standardized pieces measuring 2.0 × 5.0 × 1.5 cm. After initial weighing, samples were vacuum-sealed in polypropylene bags and cooked for 60 min in a 70°C water bath (SWB-25; AS ONE, Osaka, Japan) to ensure precise and uniform temperature control. The difference in the weight of each sample before and after cooking was expressed as the percentage of cooking loss. Shear force value was determined using a Warner-Bratzler shear device (G-R Manufacturing, Manhattan, KS, USA). After cutting the cooked samples into strips of 1.0 × 4.0 × 1.0 cm, the strips were sheared perpendicular to the muscle fibers and the peak shear force (kg/cm2) was recorded.
The remaining thigh meat was used for pH, IMP, free amino acid contents, and fatty acid composition analyses. Meat was skinned and cut into small pieces using a knife. Three grams of each sample were homogenized in 20 mL of dH2O for 60 s using a homogenizer (IKA-ULTRA-TURRAX T25; IKA-Labortechnik, Staufen im Breigslau, Germany). The pH of the homogenates was measured (F72; Horiba, Kyoto, Japan). The homogenate was mixed with 10 mL of 10% (w/v) sulfosalicylic acid and centrifuged at 10,000 × g for 15 min and 4°C (Himac CR22F; Hitachi Koki, Tokyo, Japan). The resulting supernatant was combined with 5 mL n-hexane and mixed for 30 s to remove lipids. The aqueous phase was used for both IMP and free amino acid analyses.
For IMP analysis, 10 mL of the aqueous phase was adjusted to pH 6.8 with NaOH, diluted to 25 mL with dH2O, and filtered through a 0.45-µm membrane filter (DISMIC 13HP045AN; Toyo Roshi Kaisha, Tokyo, Japan). IMP content was determined by HPLC (LC20AD; Shimadzu, Kyoto, Japan) with an STR ODS II column (4.6 × 150 mm; Shinwa Chemical Industries, Kyoto, Japan). The mobile phase consisted of 100 mM phosphate buffer containing 2% acetonitrile (pH 6.8), the flow rate was 1.0 mL/min, and the oven temperature was 40°C. Quantification was performed using an IMP standard (Sigma-Aldrich, St. Louis, MO, USA).
For free amino acid analysis, another 10-mL aliquot of the aqueous phase was adjusted to pH 2.7, diluted, and filtered using the same method as for IMP. The filtrate was separated on a Shimadzu LC-20 Prominence liquid chromatography system equipped with a Shim-Pack Amino-Li type column (6.0 × 100 mm; Shimadzu) and Shim-Pack ISC-30/S 0504 Li pre-column (4.0 × 50 mm; Shimadzu). A lithium citrate buffer system was used for gradient elution: from pH 2.2 (0.15 N lithium citrate, 7% methyl cellosolve) to pH 10.0 (0.3 N lithium citrate), at a flow rate of 0.6 mL/min and a column temperature of 39°C. Detection was performed using a Shimadzu RF-535 spectrofluorometric detector (excitation wavelength, 350 nm; emission wavelength, 450 nm), followed by derivatization with o-phthalaldehyde and N-acetyl-l-cysteine. A mixed amino acid standard solution (Wako Pure Chemical, Osaka, Japan) was used as the external standard.
For fatty acid composition analysis, total lipids were extracted using a modified Folch method[28]. One gram of minced meat was combined with 12 mL of chloroform: methanol (2:1, v/v) and heated in boiling water for 30 min. After cooling, 1 mL dH2O was added, and the mixture was centrifuged at 2,400 × g for 10 min (Himac CR22F). Approximately 2 mL of the chloroform layer was collected and evaporated under nitrogen at 50°C using a Reacti-Therm heating unit and Reacti-Vap evaporator (Thermo Fisher Scientific, Waltham, MA, USA). To remove non-lipid-contaminating compounds and water, 3 mL petroleum ether and approximately 100 mg Na2SO4 were added to the test tubes, which were then vortexed. The supernatant was transferred to another test tube and evaporated under a gentle nitrogen flow (50°C) until it was dry again.
Approximately 20 mg of extracted lipid was subjected to base-catalyzed transesterification with 0.5 M sodium methoxide in anhydrous methanol at 75°C for 30 min in a water bath (TR-3A; AS ONE). The resulting fatty acid methyl esters (FAMEs) were extracted with n-hexane, washed with water, centrifuged at 2,400 × g for 10 min, and stored at 1°C until analysis.
FAME composition was analyzed using a GC-2010 gas chromatograph (Shimadzu) equipped with a DB-23 capillary column (60 m × 0.32 mm internal diameter; J&W Scientific, Folsom, CA, USA). Operating conditions included a detector temperature of 250°C, an injector temperature of 200°C, a split ratio of 20:1, and He as carrier gas. FAMEs were identified by comparison with a standard mixture (Supelco 37 Component FAME Mix; Sigma-Aldrich).
Statistical analysisAll statistical analyses were performed in R version 3.6.1[29]. Descriptive sensory scores were analyzed using a linear mixed model implemented via the “lme4” package[30]. Chicken samples, serving order, and meat cut-out location were used as fixed effects, whereas sensory panelists were used as random effects. Instrumental measurements were analyzed using one-way analysis of variance. Where applicable, multiple comparisons were conducted using the Tukey–Kramer test via the “multcomp” package[31]. The significance level was set at p < 0.05.
Principal component analysis (PCA) was applied to visualize the relationships between chicken samples and sensory or instrumental variables. PCA was performed using the “FactomineR” package[32].
Table 3 summarizes the descriptive sensory characteristics of the six chicken thigh meat samples tested. The trained panel identified significant differences (p < 0.05) in “meaty odor,” “springiness,” “tenderness,” “meaty flavor,” “light taste,” “umami taste,” and “aftertaste intensity”. Multiple comparisons confirmed the distinct texture profile of Choshu-Kurokashiwa, which exhibited significantly higher “springiness” and lower “tenderness” than all other samples. Choshu-Kurokashiwa displayed also the strongest “meaty odor,” “meaty flavor,” “umami taste,” and “aftertaste intensity”, as well as the weakest “light taste” among the samples with significant differences.
| Attribute | Choshu-Kurokashiwa | Broiler-A | Broiler-B | Broiler-C | Broiler-D | Broiler-E | p - value | |
| Odor | ||||||||
| Rancid-like odor | 7.5±1.4 | 7.7±1.4 | 7.4±1.4 | 8.6±1.4 | 8.4±1.4 | 6.8±1.5 | 0.205 | |
| Meaty odor | 7.0±2.4a | 5.7±2.4b | 6.5±2.4ab | 6.2±2.4ab | 5.7±2.4ab | 7.0±2.4ab | 0.025 | |
| Pan-fried chicken-like odor | 8.8±1.0 | 8.5±1.0 | 8.3±1.0 | 9.0±1.0 | 8.4±1.1 | 9.1±1.1 | 0.705 | |
| Texture | ||||||||
| Springiness | 10.9±0.6a | 7.9±0.6b | 8.0±0.6b | 7.9±0.6b | 8.6±0.7b | 8.2±0.8b | <0.001 | |
| Plump | 7.8±0.5 | 8.7±0.5 | 8.9±0.5 | 8.7±0.5 | 8.3±0.6 | 8.2±0.7 | 0.412 | |
| Tenderness | 4.8±0.5b | 7.7±0.5a | 8.5±0.5a | 7.5±0.5a | 7.7±0.7a | 8.1±0.8a | <0.001 | |
| Juiciness | 7.8±0.7 | 6.7±0.7 | 6.7±0.7 | 6.6±0.7 | 8.2±0.8 | 7.9±0.9 | 0.122 | |
| Flavor / Taste | ||||||||
| Meaty flavor | 7.5±2.2a | 5.4±2.2b | 6.0±2.2ab | 5.4±2.2b | 5.5±2.2b | 6.4±2.2ab | <0.001 | |
| Light taste | 4.4±0.6b | 7.5±0.6a | 6.6±0.6a | 7.5±0.6a | 6.9±0.7ab | 5.5±0.9ab | <0.001 | |
| Fatty taste | 7.9±0.7 | 6.4±0.7 | 7.1±0.7 | 6.5±0.7 | 8.1±0.8 | 8.0±0.9 | 0.092 | |
| Umami taste | 8.9±1.1 | 7.4±1.1 | 7.6±1.1 | 7.8±1.1 | 7.5±1.1 | 8.7±1.2 | 0.05 | |
| Aftertaste intensity | 9.0±0.8a | 7.2±0.8b | 7.8±0.8ab | 6.9±0.8b | 7.5±0.8ab | 8.4±0.9ab | 0.007 | |
a–b Values with different superscripts within a row differ significantly (p < 0.05).
1 Odor and flavor were defined orthonasal and retronasal odor, respectively. Sensory attributes were scored using a 15 cm unstructured line scale anchored with verbal attributes: odor and flavor/taste attributes ranged from "not perceived" (left) to "extremely strong" (right), while textural attributes ranged from "extremely weak" (right) to "extremely strong" (left). Values are expressed as least-square means ± standard error of the mean.
Table 4 presents the instrumental characteristics of the six chicken thigh meat samples. Significant differences in cooking loss and shear force value were observed between the samples (p < 0.01). Choshu-Kurokashiwa exhibited the second lowest cooking loss after broiler-E, and the highest shear force value. Although significant differences in pH were observed among samples, this was not a defining characteristic of Choshu-Kurokashiwa. IMP content did not differ significantly among samples (p > 0.1). Notably, Choshu-Kurokashiwa had the lowest concentration of various free amino acids relative to the other samples. Its fatty acid composition displayed unique features, including lower palmitoleic acid (C16:1) and higher arachidonic acid (C20:4n-6) compared to broilers. For other fatty acids, Choshu-Kurokashiwa showed intermediate levels; whereas broilers-D and -E were rich in polyunsaturated fatty acids.
| Choshu-Kurokashiwa | Broiler-A | Broiler-B | Broiler-C | Broiler-D | Broiler-E | p - value | ||
| n | 8 | 8 | 8 | 8 | 4 | 7 | ||
| Cooking loss (%)2 | 25.4±1.0b | 29.8±0.9a | 27.3±1.2ab | 29.7±1.0a | 26.1±0.3ab | 24.8±1.3b | <0.001 | |
| Shear force value (kg/cm2) | 1.1±0.2a | 0.8±0.1ab | 0.5±0.1b | 0.4±0.0b | 0.8±0.0ab | 0.5±0.1b | 0.003 | |
| pH | 6.32±0.03bc | 6.35±0.05ac | 6.49±0.05a | 6.27±0.02c | 6.34±0.03ac | 6.44±0.03ab | 0.003 | |
| IMP content (mg/100 g) | 133.6±6.7 | 142.1±8.4 | 141.6±7.0 | 128.2±10.5 | 159.9±14.0 | 128.6±15.4 | 0.474 | |
| Free amino acid content (mg/100 g) | ||||||||
| Asp | 14.8±1.5 | 17.3±1.4 | 19.4±1.4 | 16.5±1.4 | 12.7±2.9 | 17.8±1.5 | 0.198 | |
| Thr | 21.0±2.1 | 21.5±1.9 | 25.0±1.9 | 19.7±1.9 | 23.2±3.8 | 27.4±2.0 | 0.012 | |
| Ser | 25.0±1.9 | 31.1±1.8 | 28.3±1.8 | 30.7±1.8 | 20.2±3.6 | 27.2±1.9 | 0.058 | |
| Asn | 9.7±1.2b | 13.3±1.2ab | 14.8±1.2a | 11.9±1.1ab | 13.9±2.3ab | 12.8±1.2ab | 0.046 | |
| Glu | 50.8±4.6 | 46.8±4.4 | 53.0±4.4 | 54.9±4.3 | 47.5±8.6 | 53.2±4.5 | 0.36 | |
| Gln | 94.9±7.0bc | 131.4±6.6a | 111.8±6.6ab | 101.3±6.4bc | 78.5±13.0bc | 75.6±6.8c | <0.001 | |
| Pro | 5.9±0.9b | 10.8±0.9a | 10.4±0.9a | 9.0±0.9ab | 11.5±1.8ab | 11.0±0.9a | 0.004 | |
| Gly | 18.1±1.7b | 24.0±1.6ab | 25.5±1.6a | 23.4±1.6ab | 24.5±3.2ab | 21.8±1.7ab | 0.113 | |
| Ala | 23.6±2.5b | 36.5±2.4a | 37.2±2.4a | 34.1±2.3a | 35.9±4.7ab | 33.9±2.4ab | 0.011 | |
| Val | 7.0±1.1b | 8.9±1.1ab | 10.8±1.1ab | 9.4±1.0ab | 11.8±2.1ab | 12.2±1.1a | 0.031 | |
| Met | 3.0±0.4 | 3.7±0.4 | 4.2±0.4 | 3.6±0.4 | 5.3±0.7 | 4.0±0.4 | 0.126 | |
| Ile | 4.4±0.7b | 5.7±0.6ab | 6.7±0.6ab | 5.7±0.6ab | 6.1±1.2ab | 7.4±0.6a | 0.034 | |
| Leu | 8.8±1.3b | 11.9±1.2ab | 13.9±1.2a | 12.0±1.2ab | 12.3±2.4ab | 14.4±1.2a | 0.039 | |
| Tyr | 6.0±1.0b | 9.4±0.9ab | 11.2±0.9a | 9.9±0.9ab | 11.1±1.9ab | 11.3±1.0a | 0.006 | |
| Phe | 4.6±0.8b | 7.1±0.7ab | 8.0±0.7a | 6.9±0.7ab | 6.4±1.4ab | 8.3±0.7a | 0.021 | |
| His | 3.1±0.4b | 3.1±0.4b | 3.7±0.4ab | 3.4±0.4ab | 3.7±0.8ab | 5.1±0.4a | 0.013 | |
| Lys | 12.9±2.1b | 17.2±2.0ab | 19.5±2.0ab | 16.1±2.0ab | 22.1±3.9ab | 23.3±2.1a | 0.021 | |
| Arg | 17.4±1.1 | 16.2±1.0 | 16.3±1.0 | 16.8±1.0 | 18.5±2.0 | 20.6±1.1 | 0.057 | |
| Total | 331.0±25.6 | 415.9±24.2 | 419.6±24.2 | 385.2±23.7 | 361.4±47.8 | 387.2±25.2 | 0.243 | |
| Fatty acid compositon (%) | ||||||||
| Myristic acid (C14:0) | 0.68±0.02b | 0.69±0.01b | 0.83±0.03a | 0.58±0.02c | 0.65±0.02bc | 0.58±0.02c | <0.001 | |
| Myristoleic acid C14:1) | 0.13±0.02a | 0.18±0.01b | 0.16±0.01ab | 0.15±0.01ab | 0.12±0.00ab | 0.14±0.01ab | 0.025 | |
| Palmitic acid (C16:0) | 23.5±0.3 | 23.3±0.3 | 23.5±0.1 | 24.6±0.5 | 23.4±0.5 | 24.0±0.3 | 0.07 | |
| Palmitoleic acid (C16:1) | 3.91±0.45b | 5.02±0.28ab | 4.35±0.31ab | 5.84±0.45a | 4.78±0.21ab | 4.95±0.40ab | 0.021 | |
| Stearic acid (C18:0) | 8.08±0.46 | 7.35±0.54 | 8.08±0.26 | 7.40±0.26 | 7.09±0.28 | 6.90±0.33 | 0.232 | |
| Oleic acid (C18:1) | 42.7±0.8bc | 46.8±0.6a | 44.6±0.6ac | 40.8±1.0bd | 38.5±1.1cd | 38.8±1.1d | <0.001 | |
| Linoleic acid (C18:2n-6) | 17.3±0.8ac | 13.8±0.5c | 15.8±0.6c | 17.0±1.3bc | 21.6±1.0ab | 20.8±0.8a | <0.001 | |
| α-Linolenic acid (C18:3 n-3α) | 1.00±0.05bc | 0.86±0.03c | 0.91±0.04c | 1.17±0.06b | 1.50±0.04a | 1.19±0.04b | 0.004 | |
| γ-Linolenic acid (C18:3 n-6γ) | 0.19±0.02ab | 0.18±0.01ab | 0.16±0.01b | 0.23±0.02a | 0.26±0.01a | 0.21±0.01ab | <0.001 | |
| Arachidic acid (C20:0) | 0.08±0.00a | 0.07±0.00ab | 0.08±0.00a | 0.07±0.00b | 0.07±0.00ab | 0.07±0.00b | 0.002 | |
| Eicosadienoic acid (C20:2 n-6) | 0.20±0.02 | 0.16±0.01 | 0.19±0.02 | 0.18±0.02 | 0.18±0.01 | 0.19±0.02 | 0.679 | |
| Arachidonic acid (C20:4n-6) | 2.25±0.21a | 1.64±0.15ab | 1.39±0.12b | 1.96±0.21ab | 1.79±0.13ab | 2.16±0.20ab | 0.015 | |
| Saturated fatty acid | 32.4±0.4 | 31.4±0.7 | 32.4±0.3 | 32.6±0.4 | 31.2±0.2 | 31.6±0.6 | 0.251 | |
| Monounsaturated fatty acid | 46.7±1.2bc | 52.0±0.8a | 49.1±0.9ab | 46.8±1.3bc | 43.4±0.9bc | 43.9±1.5c | <0.001 | |
| Polyunsaturated fatty acid | 20.9±1.0ab | 16.6±0.5b | 18.5±0.7b | 20.6±1.5ab | 25.4±1.1a | 24.5±1.1a | <0.001 | |
a–d Values with different superscripts within a row differ significantly (p < 0.05).
1 Values are expressed as mean ± standard error of the mean.
2 To measure cooking loss and shear force value, the biceps femoris muscle was used.
Descriptive and instrumental data were subjected to PCA (Fig. 1), which revealed clear differences between Choshu-Kurokashiwa and broilers. The first four principal components (PCs) explained 95.6% of total variance, with PC1 and PC2 collectively accounting for 64.8%. This indicated that most of the original variability was captured by the components.
Principal components analysis (PCA) of descriptive sensory and instrumental characteristics of chicken thigh meat samples. The axes indicate the principal components PC1 and PC2. (A) Sample loadings. (B) Eigenvectors of each characteristic. MUFA: monounsaturated fatty acid, PUFA: polyunsaturated fatty acid.
As shown in Fig. 1A, Choshu-Kurokashiwa exhibited the most negative coordinate on PC1 (-8.62), clealy separating it from other samples, which clustered between -1.06 and 3.91. Accordingly, PC1 reflected key differences in the sensory and instrumental characteristics between Choshu-Kurokashiwa and broiler chickens.
The eigenvectors of each variable are shown in Fig. 1B. PC1 was strongly associated with texture attributes, exhibiting high negative loadings for “springiness” (-0.94) and positive loadings for “plump” (0.93) and “tenderness” (0.89), thereby explaining the extremely negative PC1 score of Choshu-Kurokashiwa. A similarly strong negative correlation with PC1 (-0.67) was observed for the shear force value. These results confirm the variation in texture captured by PC1, particularly the greater “springiness” and reduced “tenderness” and “plump” values of Choshu-Kurokashiwa.
PC1 correlated with several flavor/taste attributes, as well as with many free amino acids and C20:4n-6 composition. “Meaty flavor” (-0.88), “umami taste” (-0.85), and “aftertaste intensity” (-0.84) showed strong negative loadings, consistent with Choshu-Kurokashiwa’s high sensory scores in these attributes. Choshu-Kurokashiwa’s elevated C20:4n-6 composition (2.25%) had a moderate negative loading (-0.82); whereas most free amino acids exhibited strong positive loadings on PC1, more typical of broiler characteristics.
For PC2, Choshu-Kurokashiwa showed a negative score (-3.16), which was close to those of broiler-A (-4.25), broiler-B (-1.28), and broiler-C (-1.37). PC2 correlated positively with linoleic acid (C18:2n-6, 0.92), Arg (0.90), polyunsaturated fatty acids (0.89), His (0.87), Lys (0.87), and Val (0.86), and negatively with Gln (-0.85), oleic acid (C18:1, -0.85), and monounsaturated fatty acids (-0.84). PC2 did not appear to play a substantial role in distinguishing Choshu-Kurokashiwa from broiler samples.
Using descriptive analysis, a lexicon was developed to characterize the odor, texture, and flavor/taste of Choshu-Kurokashiwa and broiler thigh meat. This lexicon revealed that Choshu-Kurokashiwa possessed distinct sensory attributes for texture, flavor/taste (Table 3), which were confirmed by PCA (Fig. 1). A clear separation was observed between Choshu-Kurokashiwa and the five broiler samples along PC1, with key contributions from “springiness,” “tenderness,” “plump,” “meaty flavor,” “umami taste,” “aftertaste intensity,” and “light taste.”
By integrating descriptive sensory analysis with instrumental data, this study confirmed the critical role of texture in defining Choshu-Kurokashiwa’s sensory uniqueness. Producers, breeders, and researchers have long recognized its distinct texture[12], whereas consumers associate jidori-niku with its firmness and chewiness[4,5]. The findings of the present study support these perceptions: Choshu-Kurokashiwa exhibited significantly higher “springiness” and lower “tenderness” (i.e., increased firmness) compared to broilers (Table 3). These results were validated by measuring shear force value of meat samples (Table 4), confirming that Choshu-Kurokashiwa aligned with consumer expectations for a firmer texture. While the sensory evaluation employed oven cooking to mimic consumer practices, the water bath method was adopted based on its widespread use in prior studies to examine meat physicochemical properties. The differences in meat cooking methods may influence the results; thus, integration of sensory and physicochemical analyses should be interpreted as exploratory.
Descriptive analysis also identified “meaty flavor”—referenced against seared red beef meat without charring—as a defining property of Choshu-Kurokashiwa (Tables 2 and 3). In contrast, the “pan-fried chicken-like odor”—referenced against pan-fried broiler chicken thigh with crispy skin—did not differ significantly across samples. This suggests that Choshu-Kurokashiwa exhibits a distinctive flavor profile reminiscent of beef and differs from that of conventional broilers. A Previous study has reported a breed-specific odor in Nagoya-Cochin chickens[13]. The present study is the first to show that “meaty flavor” may define a breed-specific sensory profile in Choshu-Kurokashiwa.
PCA identified three divergent taste attributes along PC1: “umami taste” and “aftertaste intensity” aligned with Choshu-Kurokashiwa, while “light taste” associated with broilers. Because “aftertaste intensity” and “light taste” are near opposites, an inverse relationship between these two characteristics was not surprising. Interestingly, umami components (Glu, Asp, and IMP) that bind to umami receptors did not correlate with umami intensity. This is in line with the author’s previous report, whereby a stronger “umami taste” of Choshu-Kurokashiwa was not accompanied by higher levels of these umami components[14].
Unique texture represents a potential driver of enhanced “meaty flavor,” “umami taste,” and “aftertaste intensity” of Choshu-Kurokashiwa thigh meat. Choshu-Kurokashiwa meat has been shown to have distinctly different texture characteristics from broilers in both the previous research and the present study. In the authors’ previous report, Choshu-Kurokashiwa meat demonstrated strong chewiness[14]. Chewiness plays an important role in taste/flavor perception by stimulating saliva secretion, thereby solubilizing the taste and aroma-active components in foods[33,34]. In addition, chewiness promotes contact between food and oral taste buds, enhancing taste receptivity[33]. A previous report on the relationships among the sensory attributes of chicken breast meat showed that flavor intensity correlated highly and positively with chewiness[9]. In the present study, Choshu-Kurokashiwa likely required more chewing than broilers, increasing the release of aromatic components to intensify “meaty flavor” and that of umami components to intensify “umami taste”. The enhanced mixing of water- and lipid-soluble components with saliva also made the “aftertaste” more perceptible even after swallowing. Sensory comparison of minced versus intact Choshu-Kurokashiwa and broiler meat would identify the contribution of texture to taste/flavor characteristics.
Elevated C20:4n-6 composition constitutes another candidate factor for the stronger “meaty flavor,” “umami taste,” and “aftertaste intensity” of Choshu-Kurokashiwa thigh meat. C20:4n-6 was positioned negatively on PC1 in PCA, along with the above characteristics. C20:4n-6 in meat is subjected to self-oxidation, forming aldehydes responsible for the “meaty flavor”[35]. In addition, the C20:4n-6 in chicken meat is associated with its palatability. There is no molecular or chemical evidence of C20:4n-6 binding to umami receptors, or that it strengthens the binding of umami components to receptors. However, dietary C20:4n-6 supplementation in broilers and Hinai-jidori increases C20:4n-6 content in meat, intensifying their “umami taste” and “aftertaste”[36,37]. In the present study, Choshu-Kurokashiwa contained higher levels of C20:4n-6, which may have resulted in stronger “meaty flavor,” “umami taste,” and “aftertaste intensity” compared to broilers. In spite of the correlation between C20:4n-6 and strong “meaty flavor,” “umami taste,” and “aftertaste intensity”, further validation is required to demonstrate a causative relationship. Sensory evaluation and GC/MS analysis of aroma compounds in defatted and non-defatted meat can be used to assess the effect of C20:4n-6 on the flavor/taste of jidori-niku.
The present study has several limitations. First, it focused on a single jidori breed (Choshu-Kurokashiwa), which limits the generalizability of the findings. Approximately 50 jidori chicken brands have been developed locally across Japan[38]. Although all jidori chickens were fed according to JAS844 guidelines, variations in genetics, diet, rearing practices, and rearing periods among jidori brands may have influenced their sensory profiles. Yoshida et al. reported a stronger “umami taste” in broiler chicken thigh meat than in jidori-niku, which contradicts the results of the present study[39]. Future studies should include multiple jidori varieties raised under controlled conditions to clarify the effects of genetics, feeding system, and production setting on the sensory characteristics. Second, broilers-D and -E underwent fewer evaluations than other chicken samples. Thus, caution is required when interpreting the results, because differences in the number of evaluations may lead to bias.
In conclusion, this study identified the distinctive sensory attributes that differentiate Choshu-Kurokashiwa from commercial broiler chickens. The sensory lexicon developed herein allows producers and marketers to communicate the unique qualities of Choshu-Kurokashiwa more precisely, helping align consumer expectations with product characteristics. Accurate labeling and communication may improve consumer satisfaction and support repeat purchasing behavior[40,41]. Consumers seek in jidori-niku sensory characteristics that differ from those of commercial broilers[4,5]. Thus, research on consumer preferences for “jidori-niku”, not generic “chicken meat,” is required. To identify key drivers of consumer preference for jidori-niku, I will apply external preference mapping, previously used in livestock research[42,43,44], integrating consumer hedonic ratings, descriptive analysis data from a trained panel, and statistical modeling.
This study was supported by Yamaguchi Prefecture (2024). I would like to extend my heartfelt appreciation to Ms. Hitomi Umemura, Ms. Mika Ogane, and Ms. Riho Sueyoshi at AKIKAWA FOODS & FARMS Co., Ltd. for their invaluable contributions to the data collection process. I sincerely thank all the sensory panelists at AKIKAWA FOODS & FARMS Co., Ltd.
The author confirms sole responsibility for the study conception and design, data collection, analysis and interpretation of the results, as well as manuscript preparation.
The author declares no conflict of interest.
