Product collection took place at 2 commercial harvest facilities in midwestern United States. Carcasses that presented dark-colored lean were visually identified and tagged as they were presented for carcass grading (23–38 h postmortem). Once away from the grading rail, DC carcasses were further evaluated by USDA Agriculture Marketing Service graders and then allocated to 1 of 3 treatment groups based on visual DC severity, with the ribeye presenting either shady, moderate, or moderately severe shading (n = 8). Each treatment group was determined using common USDA separation of dark cutters as follows: shady = 1/4 dark, moderate = 1/2 dark, moderately severe = 3/4 dark, and full = 100% dark. Carcasses with normal BCR colored lean were also selected and tagged to serve as the experimental control (n = 8). The Oklahoma State University (OSU) research team collected carcass data for each identified carcass, including marbling score, lean maturity, skeletal maturity, ribeye area (REA), preliminary fat thickness, adjusted fat thickness, hot carcass weight (HCW), and sex. All cattle were grain finished, A-skeletal maturity, spray-chilled, and predominantly black hided. Beef strip loins (Institutional Meat Purchase Specifications #180) from respective previously tagged carcasses were also tagged to maintain traceability during final processing and to utilize their unique identifications throughout analysis.

Strip loins were transported on ice to the Robert M. Kerr Food and Agricultural Products Center at OSU (Stillwater, OK). Upon arrival, pH was assessed for each loin to confirm that they were categorized appropriately and then bisected, where each half was randomly assigned to a wet-aged period of either 21 or 39 d and vacuum packaged (Walton’s Vacuum Pouch; 16 × 20 pouches; 4-mL thickness) using a Sipromac vacuum sealer (Siprovac 650A Double Chamber Vacuum Sealer; Sipromc; Drummondville, Canada) and placed in dark storage at 4°C. Packages were checked regularly throughout aging.

Upon the completion of its respective aging period, each loin half was sliced into 2.54-cm thick steaks from the most anterior end utilizing a meat slicer (Model SG13; Globe, Dayton, OH). Three steaks from each loin half were randomly assigned to Warner-Bratzler shear force (WBSF), trained sensory panels, and consumer sensory panels. Steaks were then vacuum packaged and placed in cardboard totes in frozen storage (−21°C) until further analysis.

pH was measured in triplicate at 3 random locations in each loin upon initial processing and packaging utilizing a pH probe (Handheld HI 99163; probe FC232; Hanna Instruments). pH was measured in triplicate again when loins were sliced at the conclusion of each respective aging period. One face steak from the most anterior end was used to determine percent protein, fat, and moisture following grinding each steak using a table-top grinder (Big Bite Grinder, 4.5 mm, fine grind, LEM) and pressing into a 140-mm sample cup. An AOAC-approved near-infrared spectrophotometer was then utilized to evaluate sample composition (FoodScan Lab Analyzer, Foss, NIRsystem Inc.; Slangerupgrade, Denmark).

Steaks were thawed for 24 h at 4°C and were then cooked to an internal temperature of 71°C using a Rational^® oven (Model SCC WE 102G, Rational AG Landsberg am Lech; Germany) at 204°C with 0% humidity. Internal temperature was monitored utilizing the manufacturer temperature probe installed in the oven, and peak temperatures were measured when removed from the oven using a meat thermometer (Thermopen mk4; Salt Lake City, UT) and recorded. After cooking, steaks were cooled at 4°C for 18 h. Steaks were then cored following American Meat Science Association (AMSA, 2016) guidelines with 6 cores (1.27-cm in diameter) collected by hand from each steak taken parallel to the longitudinal muscle fiber orientation. An Instron Universal Testing Machine (Model 5943; Instron Corporation; Norwood, MA) compatible with Bluehill 3 software was used with a standard WBSF blade, and the crosshead speed was set at 250 mm/min. The maximum load (kgF) was recorded for each core, and the average of the cores taken for each sample were used for analysis.

After thawing for 24 h at 4°C, steaks designated for WBSF analysis were weighed while still in the package. Steaks were then removed from the package, patted dry with a paper towel, and weighed again to record raw steak weight. Each package and identification tag were also patted dry with a paper towel, weighed, and recorded. Each weight collected was utilized to calculate thaw loss. Steaks were then cooked using the same methods described for WBSF, and peak temperatures were recorded. Cooking loss was then determined using the raw steak weights and the cooked steak weights, which were recorded after peak temperature was reached and a 5-min rest time was allowed.

Trained sensory panel evaluation was conducted using an 8-member panel evaluating palatability and sensory attributes. The panelists were trained to recognize beef flavors and aromas according to the AMSA Sensory Evaluation Guidelines (AMSA, 2016). Panelists were trained in a total of eight 30-min sessions within 2 wk prior to panels. Eight samples were evaluated by panelists during each panel session. Steaks were thawed and cooked following the same procedures utilized for WBSF, and peak temperatures were recorded. After cooking, steaks were cut into 1.0-cm³ cubes. Two pieces of each sample were placed in sample cups labeled with a random 3-digit code. Sample cup temperature was maintained utilizing a commercial catering warmer (Model PS-1220-15, Food Warming Equipment Co.; Crystal Lake, IL). Sample order was determined ahead of the panel and was set up in a manner in which no panelists were consuming the same samples at the same time to avoid bias.

Panelists were provided deionized water and unsalted crackers to cleanse their palate between each sample. Samples were evaluated under red lighting to combat sample bias. Samples were evaluated using an 8-point scale for initial juiciness, sustained juiciness, tenderness, connective tissue as well as attributes of the following: beef identity, brown/roasted, sour, metallic, bloody/serumy, fat-like, umami, and rancid (1 = extremely dry, 8 = extremely juicy; 1 = extremely tough, 8 = extremely tender; 1 = no connective tissue, 8 = abundant connective tissue; 1 = not detectable, 8 = extremely intense).

Untrained discriminating panelists (n = 122) were recruited in 2 locations: a technology center (Oklahoma City, OK) and the OSU campus (Stillwater, OK). Panelists were volunteers who consisted of culinary professionals and students being trained in culinary arts as well as consumers from the general population of Stillwater, Oklahoma. Panels were hosted at both locations to serve both sets of volunteers, and a total of 8 panel sessions took place. All panel sessions included up to 16 panelists and lasted approximately 40 min. Panelists were each provided deionized water and unsalted crackers to cleanse their palate, a napkin, expectorant cup, and a toothpick to utilize when evaluating samples. All samples (n = 8) were served to each panelist at once, hence the order in which each panelist consumed the samples was at their own individual discretion and therefore eliminated any panelist-to-panelist and sample-to-sample bias. Samples were prepared in the same manner as previously outlined for trained sensory panels. After cooking, steaks were cut into 1.0-cm³ cubes. Two pieces of each sample were placed in sample cups labeled with a random 3-digit code. Sample cup temperature was maintained utilizing a commercial catering warmer.

Verbal instructions were provided at once to all consumers before beginning the electronic survey and ballot (Qualtrics Software; Provo, UT) to be completed using the individual’s own smartphone or other electronic device. The survey included a study consent form, approved by the OSU Institutional Review Board, a short demographic survey, and ballots for all 8 samples. The demographics survey collected the following information: age, gender, ethnicity, education level, household income, working status, knowledge of the meat industry, whether meat was regularly purchased, whether beef specifically was regularly purchased, and meat purchasing habits. The meat purchasing habits questions included how often beef was consumed in the home and in a restaurant/fast-food chain and whether they regularly purchase organic, traditional, aged, grass-fed, or locally sourced beef. If participants answered that they did purchase meat and beef, they were then asked what the primary factor influencing their purchasing decision was with the following choices to select from: 1) availability, 2) color, 3) price, 4) number of steaks in package, 5) USDA grade (marbling), 6) weight, or 7) other. If participants answered that they did not purchase meat, they were instead asked to identify their primary reasoning with the following to select from: 1) animal welfare, 2) ethical reasons, 3) environmental reasons, 4) price, 5) religious reasons, or 6) other.

Samples were evaluated using a 7-point scale for initial and sustained juiciness, tenderness, and connective tissue. A 5-point scale was utilized for attributes of beef identity, brown/roasted, sour, metallic, bloody/serumy, fat-like, umami, and rancid (1 = extremely dry, 7 = extremely juicy; 1 = extremely tough, 7 = extremely tender; 1 = no connective tissue, 7 = abundant connective tissue; 1 = not detectable, 5 = extremely intense).

A split-plot design was utilized to evaluate the impact of wet aging and DC severity on sensory and palatability attributes of DC strip loins. Within the whole plot, a randomized complete block design was utilized in which loin served as the experimental unit and an equal number of loins (n = 8/treatment) were utilized across treatments. Each loin half within the aging period (21 or 39 d) served as the subplot. The fixed effects included DC severity, wet-aged period length, and the resulting interaction. The panel served as random effect when evaluating trained and consumer panel data. The least-squares means were determined using the PROC GLIMMIX procedure of SAS (SAS 9.4; SAS Inst.; Cary, NC) and were considered significant at P < .05. In addition, Kenward-Roger was utilized throughout all analyses for denominator degrees of freedom. Using the PDIFF option, least-squares means were separated, and the LINES statement was used to generate superscripts when overall F-tests indicated significant differences. Survey demographics were analyzed using the PROC FREQ procedure of SAS.

Simple averages were calculated for each treatment group for HCW, marbling score, REA measurement, and 12th rib fat thickness (Table 1). Carcasses with normal BCR colored lean in the ribeye had the lowest average marbling score; however, they were all within 50° of marbling overall. Normal BCR also maintained the lowest fat thickness and heaviest carcass weights. Shady, moderate, and moderately severe DC carcasses had an average marbling score of USDA. Small, moderate DC carcasses possessed the largest average REA.

In agreement with previous studies conducted by Grayson et al. (2016) and Denzer (2020), normal BCR colored loins had a lower pH (P < .05) compared to all DC severity group loins (Table 2). In addition, shady DC loins had a higher (P < .05) pH than normal loins. Moderately severe and moderate DC loins had the highest (P < .05) ultimate pH, with an average of 6.31 and 6.25, respectively. Shady DC loins had the next greatest (P < .05) pH with an average of 5.91, while normal BCR colored loins had the lowest pH with an average of 5.60 (P < .05). Grayson et al. (2016) reported similar findings for differences in pH among atypical DC classes. Proximate analyses conducted utilizing face steaks from each strip loin are displayed in Table 3. There were no significant differences (P > .05) in protein, moisture, and fat between all classes of DC loins and the control. Denzer (2020) reported no differences in normal and DC strip loins for moisture and fat, and Smith et al. (2021) found that normal and DC loins possessed similar protein content. Additionally, Smith et al. (2021) found that DC loins exhibit greater moisture and lower fat percentages compared to normal-pH loins. This is attributed to the higher water-holding capacity of DC beef, which results from the lack of muscle fiber shrinkage at higher pH levels.

No differences were observed for thaw loss among normal BCR colored controls or DC steaks for aging or treatment effects (P > .05; Table 3). Moreover, there was no difference for cooking loss for aging or treatment effects (P > .05). In contrast to the current findings, Holdstock et al. (2014) reported that DC steaks with elevated pH values had the least cook loss, followed by atypical DC steaks having the next lowest, and normal-pH steaks resulting in the greatest amount of cook loss. The differences observed in previous studies are to be expected, as the continued elevation in pH causes an increased water-holding capacity and therefore reduced excess moisture release. However, in the present study, there was not a difference in the percentage of moisture among treatments thus yielding no differences in thaw loss.

There was a significant interaction (P < .05) observed for treatment × aging period for WBSF values between control and all DC severity groups (Table 4). Normal BCR colored and shady DC steaks aged for 21 d were comparable (P > .05) in shear force value but were more tender (P < .05) than moderate DC steaks. In addition, moderately severe DC steaks aged for 21d had WBSF values that were similar to normal BCR colored, shady, and moderate DC steaks (P > .05). The control and all DC severity groups that were aged 39 d had similar WBSF values (P > .05). While shear force values from BCR colored controls and varying DC severity groups did not follow a trend, the average shear force values for all treatment groups met USDA certified very tender threshold value requirements (ASTM, 2011). Additionally, previous literature indicates that as postmortem aging time increases, proteolysis and myofibrillar protein breakdown continue, therefore increased aging time should result in a product with increased tenderness (Davey and Gilbert, 1969; Holman et al., 2019; Ramanathan et al., 2020).

No differences in tenderness, juiciness, or connective tissue were observed for wet-aged period (P > .05; Table 5). Trained panelists did not detect any difference in initial and sustained juiciness and connective tissue between normal and all DC severity classes (P > .05). These results are supported by findings of Wulf et al. (2002) and Holdstock et al. (2014), as no differences in juiciness were observed between normal colored and DC steaks. In contrast to the WBSF results, there was a significant effect determined by trained panelists for tenderness among control/BCR colored and DC loins. Steaks from DC loins classified as moderately severe had greater tenderness ratings than those classified as shady and control loins (P < .05). Additionally, DC loins classified as moderate had greater (P < .05) tenderness ratings than normal BCR loins and similar (P > .05) tenderness to both shady and moderately severe DC loins. Shady DC and normal loins also had similar tenderness ratings (P > .05). Grayson et al. (2016) noted variability in tenderness among DC loins and found that steaks from DC loins with an ultimate pH of 6.1 to 6.4 were less tender than steaks from normal loins and steaks from DC loins with a pH ranging from 6.6 to 6.9 based on slice shear force values. Moreover, following the trend reported in the present study, Grayson et al. (2016) determined that samples from the DC group with the highest pH values (pH 6.9) were the most tender according to trained panelists. Samples from DC loins with a pH ranging between 6.4 and 6.6 were assigned the next highest values, followed by normal (pH 5.6) loins, and DC samples with an average pH 6.1 were rated the least tender (Grayson et al., 2016).

Among the 8 attributes panelists evaluated (Table 6), the only notable differences detected were for metallic (P < .05). The wet-aged period had no impact on any of the attributes evaluated (P > .05). All DC classes had less intense metallic compared to normal colored steaks (P < .05). Denzer (2020) found that normal-pH steaks also possessed a stronger metallic flavor than DC steaks. Normal-pH steaks evaluated by Grayson et al. (2016) possessed a stronger metallic flavor compared to DC steaks with pH values ranging from 6.6 to 6.9, according to trained panelists. In contrast, Yancey et al. (2005) noted no difference in metallic flavor for normal-pH and DC beef steaks. No differences were noted for beef identity by trained panelists in the current study (P > .05), which is supported by the results of Grayson et al. (2016), who found no differences in beef flavor among normal-pH steaks and DC steaks ranging in pH from 6.1 to 6.9. However, Yancey et al. (2005) found that normal-pH steaks had a significantly higher rating for beef flavor in comparison to DC steaks. Normal-pH steaks and atypical DC steaks with a pH below 5.8 were also found to have a more intense beef flavor compared to DC steaks with a pH above 6.0 in a study conducted by Holdstock et al. (2014). The strong metallic attribute could be overpowering or masking the beef identity in normal-pH beef and therefore could be responsible for the lack of difference observed among normal colored, normal-pH samples, and DC samples.

The brown/roasted attribute was similar between control BCR steaks and all DC classes in the current work (P > .05); however, Yancey et al. (2005) and Grayson et al. (2016) found that DC steaks with elevated pH values had less brown/roasted flavor compared to normal-pH steaks. It could be theorized that because normal BCR colored steaks have a greater amount of remaining glucose, a reducing sugar that plays a critical role in the Maillard reaction, that upon cooking a richer flavor is imparted and a brown color is produced. This could be responsible for the increased brown/roasted flavor observed in previous studies. There were no significant sour attributes noted in the present study between normal colored and DC steaks (P > .05), but Yancey et al. (2005) and Grayson et al. (2016) reported DC steaks having a more intense sour flavor. The authors attributed it to either lactic acid-producing bacteria forming as aging periods increased or a greater amount of lactic acid inherently present in normal-pH steaks, respectively. As in the current study, Grayson et al. (2016) determined no difference in bloody/serumy flavor for normal-pH and DC steaks; however, Yancey et al. (2005) found that DC steaks from B-maturity carcasses had a stronger bloody/serumy flavor compared to normal-pH steaks derived from carcasses of any maturity.

In contrast to our nonsignificant difference (P > .05) in fat-like, Grayson et al. (2016) observed that fat-like flavor increased with higher pH levels. Steaks with an average pH of 6.9 exhibited the highest fat-like flavor, while normal-pH steaks had the lowest. Umami flavor was found to be more intense in normal-pH steaks than steaks with a pH ranging between 6.6 and 6.9 (Grayson et al., 2016), contrasting the results of the current study in which umami was similar for normal BCR colored and DC steaks (P > .05). This could be based on the average pH being 6.0 across all 3 severity classes, which is significantly lower than those with umami flavor detected by Grayson et al. (2016). No difference in rancidity was observed in the present study for normal colored and DC steaks (P > .05), and the results reported by Wulf et al. (2002) agree with the current findings. Conversely, Yancey et al. (2005) and Grayson et al. (2016) reported a more intense rancid flavor in steaks with higher pH values compared to normal-pH steaks and that DC steaks with a pH ranging from 6.6 to 6.9 had higher scores for rancid flavor compared to DC steaks with pH values ranging from 6.1 to 6.4 and normal-pH steaks. Previous research has indicated that an increased pH can produce an environment favorable of myoglobin stabilization and prevent oxidative changes (Nerimetla et al., 2017; Ramanathan et al., 2019), which could support the results of the present study in which no differences were observed for rancid flavor among normal colored and DC steaks, as rancid flavor is most typically correlated with lipid oxidation.

Demographic results for the consumer preference survey are presented in Table 7. A total of 122 respondents participated in the survey. Participant gender was categorized into 3 groups: female (50.8%), male (47.5%), and prefer not to specify (1.6%). The US Census Bureau (2020) reported that 50.4% of the population are female. The majority of respondents ranged in age from 36 to 55 y of age (52.5%). Caucasian (71.3%) was the most frequently selected ethnicity, which closely resembles US Census (2020) results that show 75.5%. When asked about the highest level of education they had completed, more than half of the participants possessed either a 4-y or advanced degree (53.3%) in comparison to the population average of 34.3% reported by the US Census Bureau (2020). The majority of participants (69.7%) reported that they are employed full time. Additionally, half of the respondents reported that they had an annual household income ranging from $75 000 to $100 000 or greater (52.5%). The average household income was identified to be $75 149 based on the US Census (2020).

Survey results on consumer panel participant knowledge of the meat industry, beef purchasing preferences, and beef purchasing habits are presented in Table 7. When asked about their general knowledge of the meat industry, 64.8% selected that they were either slightly or somewhat knowledgeable. When questioned whether the respondent was the household’s primary shopper, 54.1% answered that they were. Upon asking participants if they purchased meat, 95.9% responded with yes. When asked how often the panelist purchases beef, the 2 most prevalent answers were once per week (38.1%) and once every 2 wk (35.4%). When prompted about their beef consumption frequency at home, 2 or 3 times per week were the top answers, both selected at 31.2%. When indicating how often they eat beef at a fast-food chain or restaurant, 51% of respondents answered once per week.

122 total consumer panelists participated in sample evaluation (Tables 8 and 9). No significant differences were observed for initial juiciness, tenderness, or connective tissue for wet-aged period (P > .05). These results closely align with those previously discussed by trained panelists as well as supporting documentation (Wulf et al., 2002; Holdstock et al., 2014; Denzer, 2020). However, consumers distinguished that steaks from loins aged for 39 d had increased sustained juiciness in comparison to steaks from loins aged 21 d (P < .05). While it has been well documented that trained panelists determine no differences in juiciness between normal-pH controls and DC treatments (Wulf et al., 2002; Holdstock et al., 2014; Denzer, 2020), consumers have not previously taken note of differences in juiciness. To our knowledge, no studies have been published including consumer feedback when evaluating palatability and sensory attributes of DC beef aged different lengths of time.

Similar to the aging main effects, there were no observed differences (P > .05) for initial juiciness and connective tissue for treatment, but there was a significant treatment effect for sustained juiciness. Moderately severe DC had greater (P < .05) sustained juiciness compared to shady DC and control BCR colored steaks. Moderately severe and moderate DC were similar (P > .05) for sustained juiciness scores and moderate DC, shady DC, and control BCR colored steaks were also comparable. Additionally, a difference was observed for tenderness that mirrored the results of the differences observed for sustained juiciness. Consumers observed that moderately severe DC steaks had increased tenderness scores compared to shady DC and control BCR colored steaks (P < .05). Moderately severe and moderate DC were similar in tenderness (P > .05) and moderate DC, shady DC, and normal colored steaks were similar (P > .05). These results are supported by a recent study comparing Canadian B4 (DC beef) derived from Canada AAA, AA, and A carcasses (comparable to USDA choice, select, and standard) to normal Canada AAA, AA, and A carcasses (Leighton et al., 2023). The DC beef was further split into 2 categories (moderately dark and dark) based on ribeye color intensity (Leighton et al., 2023). It was determined by consumers that B4DKAAA were both more tender and juicier compared to B4MDAAA and NAA (Leighton et al., 2023). Moreover, while their scores were from trained panelists, Grayson et al. (2016) reported the same trend for tenderness among DC and normal-pH steaks as was conveyed by the consumer panel in the present study. Again, due to the increase in pH of each DC category, particularly in comparison to normal-pH steaks, we would anticipate that bonds of water and protein are stronger at elevated pH values and consequently have a higher water-holding capacity. We would therefore expect DC samples to be more tender and juicier than normal colored, pH control samples. We hypothesize that since all steaks were very tender, consumers were unable to detect a clear difference and therefore found the samples to be highly acceptable for both tenderness and juiciness.

There were no differences observed for aging period or DC severity among any of the attributes evaluated (P > .05). Viljoen et al. (2002) likewise reported that consumer panelists did not distinguish any difference in flavor or overall acceptability comparing normal-pH and DC steaks. The authors did note, however, that females preferred normal-pH compared to DC steaks in contrast to males (Viljoen et al., 2002). Additionally, in support of current results, Leighton et al. (2023) identified that B4DKAA and B4MDAA steaks had similar flavor acceptability to NAA steaks according to consumer panelists.