Developing Indigenous Plant‐Enriched Yogurt to Enhance Iron Solubility—A Collaboration With Senegalese Women Farmers

ABSTRACT

This study utilized collaborations with Senegalese women farmers to investigate the impact of lactic acid bacteria (LAB) fermentation of sorghum, baobab, and milk on nutritional and sensory qualities for plant‐enriched yogurt. Using a simplex lattice mixture design, five samples of varying weight ratios of sorghum/baobab (17.5/32.5–47.5/2.5, w.b.) were prepared by mixing sorghum/baobab blend (50 g, w.b.), milk (500 mL), Lactobacillus bulgaricus species and Streptococcus thermophilus species, and fermenting at 40°C for 9 h. After only fermenting milk, one Control was prepared by adding sorghum/baobab (32.5/17.5). The samples were evaluated for pH, texture, LAB count, and iron solubility after simulated digestion. A total of 125 panelists, including 32 Africans, evaluated the samples flavored with bananas after fermentation, using a nine‐point hedonic scale and check‐all‐that‐apply (CATA) method. Data were analyzed using variance analysis, least significant difference test, agglomerative hierarchical clustering, and Fisher's exact test. All samples had a pH below 4.6 and a LAB count of 10⁸, except for sample 17.5/32.5 (10⁴), likely due to baobab's acidity hindering LAB growth. Fermenting sorghum, baobab, and milk together increased iron solubility by 64.1% compared to the Control. From the CATA results, sweet flavor and creamy mouthfeel increased overall liking, while tart flavor and gritty mouthfeel decreased liking. Cluster analysis showed that Africans favored samples with both less and more baobab equally (40/10, 25/25, overall liking of 6.8), while Caucasians preferred less baobab (40/10, overall liking of 4.8). Both clusters disfavored the Control (overall liking 6.0 and 4.1), which had the lowest instrumental texture consistency.

1. Introduction

In 2024, 295 million people in West Africa were food insecure, comprising nearly two‐thirds of the population, an increase from 44% in 2019 to 63% in 2024 (FAO et al. ²⁰²⁵). Undernutrition primarily stems from food insecurity, which disrupts access to sufficient and nutritious food (FAO et al. ²⁰²⁴). West Africa has about 5% of the world's population but 16.5% of its people were undernourished in 2024, up from 11% in 2019, highlighting a sharp increase over the period (FAO et al. ²⁰²⁵). In Senegal, a 2025 study reported that 34.6% of adolescents aged 10–19 years were iron deficient, with prevalence rising to 46.8% among girls aged 15–19 years (Faye et al. 2025). About 30% of Senegal's population experienced moderate or severe food insecurity in 2022, with prevalence rising to 37% in rural areas; in Kaffrine, the level was 41%, exceeding the national average (ANSD ²⁰²⁴). In the Kaffrine region of Senegal, smallholder women farmers stated, “we engage in agriculture growing crops like sorghum, cowpea, peanuts, and others because we wish to remain in our community and cultivate our land” (Ndangane women farmers in Kaffrine region, personal communication, 2023).

Sorghum (Sorghum bicolor L.) was the sixth most cultivated commodity in Western Africa in 2023. By region, Africa produced the most sorghum, accounting for 45% of global sorghum production, and Senegal contributed about 445,000 tons, roughly 1% of the world's production (FAOSTAT ²⁰²³). Sorghum serves as a source of calories and nutrients and an essential food component in rural regions in Africa (Adebo 2020). Sorghum is a source of nonheme iron, and the bioavailability of iron contained in sorghum can be enhanced through fermentation (Kruger et al. 2012). Foods from fermented sorghum have a long history of solid cultural affinity to people of African descent. The drought‐resistant cereal crop (sorghum) has huge potential for future utilization, especially with the effects of climate change and the resulting decrease in water supply (Mwamahonje et al. 2024). While sorghum demonstrates the resilience of cereals, baobab can contribute to the diversification of food sources.

African baobab (Adansonia digitata L., Malvaceae) holds crucial importance as an indigenous fruit, contributing to improving nutrition, food security, and revenue creation for Africa's rural populations (Muthai et al. 2017). The baobab fruit pulp, which is naturally dry at harvest, is nutrient‐dense, with high quantities of vitamin C, dietary fiber (DF), calcium (Ca), and potassium (K) (Asogwa et al. 2021). Despite the valuable nutritional benefits of baobab pulp, the fruit has yet to be fully utilized to its potential (Offiah and Falade 2023). Processing methods like fermentation can increase the utilization of baobab by diversifying its applications in food systems.

Fermentation is one of the oldest forms of processing and preserving food in Africa. It can diversify milk products, break down lactose, and enhance the digestibility and functionality of underutilized crops (Agyei et al. 2020). Fermenting milk with lactic acid bacteria (LAB) can produce diverse products like yogurt, cheese, and cream (Lucey 2004). Lactic acid fermentation can benefit underutilized crops such as sorghum and baobab. In sorghum, it improves flour functionality (Elkhalifa et al. 2005), reduces phytates, and promotes ferric iron formation, both of which enhance iron bioavailability (Kruger et al. 2012; Scheers et al. ²⁰¹⁶). In baobab, fermentation drives beneficial biochemical changes in the pulp, and supports yogurt production with better nutrition and acceptability (Zumunta and Umar 2020; Dauda et al. ²⁰²³). Yogurt is a popular dairy product worldwide with a unique texture, flavor, and health benefits (Li et al. 2021). Conventional yogurt is produced by fermenting cow milk with starter cultures containing LAB until the pH drops below 4.6 and the final LAB count reaches 7 log10 cfu/g (Montemurro et al. 2021).

Studies have been published on yogurt enriched with baobab (Aluko 2017) and sorghum (Oliveira et al. 2020), but no studies have yet examined the cofermentation of milk, baobab, and sorghum to develop a plant‐enriched yogurt. It is hypothesized that controlled LAB fermentation of sorghum flour, baobab powder, and liquid milk will enhance the nutrition profile and sensory liking compared to adding cooked sorghum and baobab powder after fermenting milk into yogurt. This study investigates the impact of fermenting sorghum and baobab with milk on the nutritional profile and sensory quality of plant‐enriched yogurt. This research would enhance the nutritional value and taste of plant‐enriched yogurt by exploring the effects of fermenting sorghum and baobab with milk, providing foundational knowledge to innovate novel fermented food with indigenous ingredients to support local food sovereignty.

2. Materials and Methods

2.1. Materials

Meadow Gold whole milk (Dairy Farmers of America, Inc., El Paso, TX, USA), sorghum flour (Bob's Red Mill Natural Foods, Milwaukie, OR, USA), and bananas were procured from Bozeman‐area grocery stores. Baobab fruit pulp powder was purchased from Anthony's Goods (Tarzana, CA, USA), and YO‐MIX 495 LYO 250 DCU (DANISCO France SAS, Vinay, France) yogurt starter culture containing Streptococcus thermophilus and Lactobacillus delbrueckii subsp. bulgaricus was donated by Dairy Connection (Madison, WI, USA). High‐performance liquid chromatography (HPLC) grade water (225294) and sodium bicarbonate (218048) were from Thermo Fisher Scientific Inc. (Fair Lawn, NJ, USA); hydrochloric acid, 6 M, was from Lab Alley LLC. (Austin, TX, USA). Bile (from bovine and ovine, B8381), pancreatin (from porcine pancreas, P3292), and pepsin (from porcine gastric mucosa, P7000) were from Millipore Sigma (St Louis, MO, USA).

2.2. Methods

2.2.1. Experimental Design

Ten MSU‐Bozeman campus staff and students participated in preliminary sensory tests to establish desirable yogurt formulations. A two‐component simplex lattice mixture design was employed using Minitab 20.0 (Minitab LLC., State College, PA, USA) to investigate the effects of varying weight ratios of sorghum/baobab (ranging from 17.5/32.5 to 47.5/2.5, w.b.) on the nutritional, microbial, texture, and sensory qualities of the plant‐enriched yogurt. This design encompassed the examination of sorghum flour (X) and baobab powder (Y) as primary components in five experimental runs (Table 1). Sample codes are expressed as X/Y, where X represents the grams of sorghum, and Y represents the grams of baobab used in the blend. A Control was also made by adding precooked and dried sorghum and raw baobab powder (32.5/17.5, w.b.) into fermented yogurt, with the ratio selected after preliminary sensory tests.

TABLE 1.

A two‐component simplex lattice mixture design, pH values of the yogurt samples immediately after inoculation and 9 h of fermentation.

Samples (X/Y i )	Sorghum flour (X) (g, w.b. iii )	Baobab powder (Y) (g, w.b. iii )	pH at 0 h ii	pH at 9 h ii
Control	32.5	17.5	6.91 ± 0.01a	3.99 ± 0.04d
17.5/32.5	17.5	32.5	4.09 ± 0.03f	4.06 ± 0.01c
25/25	25.0	25.0	4.47 ± 0.00e	4.39 ± 0.02b
32.5/17.5	32.5	17.5	4.97 ± 0.03d	4.56 ± 0.03a
40/10	40.0	10.0	5.85 ± 0.43c	4.59 ± 0.02a
47.5/2.5	47.5	2.5	6.33 ± 0.02b	4.55 ± 0.02a

ⁱ Sample codes are expressed as X/Y, where X and Y represent the grams of sorghum and baobab used in the blend, respectively.

ⁱⁱ For the same column, values (means ± SD) followed by the same letter are not significantly different based on Fisher's least significant difference test (α = 0.05).

ⁱⁱⁱ w.b. denotes wet basis.

2.2.2. Yogurt Sample Preparation

Sorghum flour (300 g) was precooked in water (1:3 w/v, w.b.) on a Max Burton Induction Cooktop (Aervoe Industries Inc., Nashville, TN, USA) for 8 min to 80°C and held at 80°C for 2 min, with continuous manual stirring. The precooked sorghum flour was then dried in a convection oven (G.S. Blodgett Corp., Essex Junction, VT, USA) at 95°C for 1 h and milled with the pastry setting into flour in a Wondermill Grain Grinder (Grote Molen Inc., Pocatello, ID, USA).

Six yogurt samples were prepared using the method outlined by Allen et al. (2021) with slight modifications (Table 1). Milk was heated on a Max Burton Induction Cooktop (Aervoe Industries Inc.) for 6 min to 85°C and held for 30 s at 85°C while continuously stirring with a spatula. The milk was then cooled in ice water to 45°C, measured with a Thermapen MK4 Thermometer (ThermoWorks, American Fork, UT, USA), and mixed with the baobab fruit pulp powder and precooked sorghum flour and for 1 min using a wooden spatula and an Immersion Blender (Vita‐Mix Corp., Olmsted Township, OH, USA). The mixture was inoculated with 0.37 DCU of yogurt starter culture at 40°C and fermented in a Metro C5 3 Series Proofing Cabinet (Intermetro Industries Corp., Wilkes‐Barre, PA, USA) at 40°C for 9 h until a pH at or below 4.59 was achieved. The Control sample (32.5/17.5, w.b.), as shown in Table 1, was prepared by fermenting milk alone before adding the precooked sorghum and baobab fruit pulp powder. The yogurt samples were stored between 1°C and 4°C for 24–96 h before further analysis.

2.2.3. Physicochemical Analyses

The moisture contents of the sorghum flour (before and after the precooking treatment) and the baobab fruit pulp were measured using a Mettler Toledo HC103 Moisture Analyzer (Mettler‐Toledo LLC., Columbus, OH, USA) at 140°C and 105°C, respectively. The pH of the yogurt samples was determined at 31°C using a Fisher Scientific Accumet AE150 pH Meter (Thermo Fisher Scientific Inc.) with an Orion 8165BNWP Ross Sure‐Flow Epoxy Electrode Probe (Thermo Fisher Scientific Inc., Waltham, MA, USA) and an Accumet 13‐620‐21 Automatic Temperature Compensation Probe (Thermo Fisher Scientific Inc., Waltham, MA, USA).

Soluble, insoluble, and total DF, and vitamin C of the yogurt samples were determined by NP Analytical Laboratories (St. Louis, MO, USA) using the Association of Official Analytical Chemists (AOAC) International (2006), 922.06, and 952.20 methods, respectively.

2.2.4. Texture Analysis

The texture of the yogurt samples was determined using a TA.XT Plus C Texture Analyzer (Texture Technologies Corp., South Hamilton, MA, USA) equipped with a TA‐94 Back Extrusion Rig (A/BE), including a 38‐mm disk probe, a 64‐mm diameter sample container, and a 5‐kg load cell. A back extrusion method was performed using Exponent 7.0.6.0 Software (Texture Technologies Corp., South Hamilton, MA, USA; Stable Micro Systems Ltd. ²⁰¹⁸). Eighty‐five grams of the yogurt sample at 5–8°C was loaded into the sample container, briefly stirred by hand with a spoon for 10 s, then compressed by the probe at 1 mm/s for 25 mm, and the probe return distance of 30 mm. Six measurement replicates were recorded. The force–time curve was used to determine firmness (maximum on the positive curve), consistency (area under the positive curve), cohesiveness (maximum on the negative curve), and work of cohesion (area under the negative curve; Stable Micro Systems Ltd. ²⁰¹⁸).

2.2.5. Microbial Analysis

The yogurt's LAB count was determined using the ISO 15214:1998 method by the Standardized Biofilm Methods Laboratory at the Center for Biofilm Engineering, Montana State University (Bozeman, MT, USA).

2.2.6. Iron Solubility Analysis

The iron solubility assay was conducted using the method described by Jovaní et al. (2001) and Silva et al. (2017) with some adjustments. Fifty grams of each sample were mixed in 20 mL of HPLC‐grade water and homogenized for 5 min at 2000 rpm using a Fisherbrand 850 Homogenizer (Thermo Fisher Scientific Inc., Waltham, MA, USA). The pH of the mixtures was raised to 2.0 by adding 6 M HCl. Three grams of pepsin solution (1.6 g pepsin in 10 mL of 0.1 M HCl) was introduced into the mixture, shaken in a Thermo Scientific Precision SWB 15 Shaking Water Bath (Thermo Fisher Scientific Inc., Waltham, MA, USA) at 37°C for 2 h to simulate the gastric digestion phase. Afterward, to halt gastric digestion, it was submerged in ice water for 10 min. Then, with a dropwise addition of 1 M NaHCO₃, the pH of the mixture was raised to 5.

A 25 g solution of pancreatin‐bile salt mixture (0.4% pancreatin (w/v) and 2.5% bile extract (w/v) in 0.1 M NaHCO₃) was introduced into the mixture, then shaken in water bath at 37°C for 2 h to simulate enteric digestion. Afterward, it was submerged in ice water for 10 min to halt the enteric digestion simulation. The pH was raised to 7.2 by adding 1.0 M NaHCO₃. The digested mixtures were centrifuged at 3000 × g at 4°C for 30 min using an accuSpin 24C clinical centrifuge (Thermo Fisher Scientific, Suzhou, Jiangsu, PR China). The supernatant was heated in a Panasonic Genius Sensor 1200 W Microwave Oven (Thermo Fisher Scientific Inc., Newington, NH, USA) for an 85% reduction in volume. Total soluble iron concentrations were determined in the supernatant by NP Analytical Laboratories (St. Louis, MO, USA) using the 999.10 and 968.08 methods described by AOAC International (2006).

2.2.7. Consumer Sensory Analysis

The sensory tests approved by the Montana State University Institution Review Board (IRB; 2023‐1066‐EXEMPT) were carried out in the Hannon Culinary Arts Classroom on the MSU‐Bozeman campus with 125 participants aged 18 or older. Informed consent was obtained from each panelist before their participation. Six samples (Table 1, 20 g w.b. per sample) were presented to each panelist, along with a cup of room‐temperature water for mouth rinsing between samples. Panelists used their cellphones to evaluate the samples on RedJade Software (RedJade Sensory Solutions LLC., Pleasant Hill, CA, USA). The test was approximately 30 min.

Each sample was evaluated for overall liking and liking in appearance, flavor, and mouthfeel on a nine‐point hedonic scale anchored with “Dislike extremely” (1), “Neither like nor dislike” (5), to “Like extremely” (9) (Lawless and Heymann 2010). We employed a check‐all‐that‐apply (CATA) method described by Jaeger et al. (2019) to evaluate the sensory attributes of the flavored samples. For flavoring, 10% freshly mashed banana was prepared using a Vitamix immersion blender and incorporated into the yogurt prior to sensory evaluation. Panelists were asked to have a second sample spoonful before checking the CATA terms. The terms included in the CATA questions encompassed a range of attributes in the modalities of flavor, mouthfeel, and aftertaste established during the preliminary sensory tests by a Culinary Research Chef. During sensory evaluation, the panelists were given 23 CATA terms to select from. The terms for flavor (FL) attributes included sweet, astringent, cereal, sour, tangy, creamy, banana, tart and milky; for aftertaste (AT), nutty, sour, earthy, sweet, tangy, and astringent; and for mouthfeel (MF), smooth, whipped, light, creamy, gritty, thin, thick, and chalky. The panelists were also asked to specify other terms not provided. To visualize the relationships between sensory attributes and the samples, and to explore which term impacted overall liking, a principal coordinate analysis (PCoA) plot was generated.

2.2.8. Statistical Analysis

R version 4.4.0 (R Core Team ²⁰²⁴; R Foundation for Statistical Computing, Vienna, Austria) was used to analyze all the data. The means of the yogurt samples were compared using a one‐way analysis of variance (ANOVA) and pairwise using Fisher's least significant difference test (α = 0.05). Using panelists’ sensory response data, agglomerative hierarchical clustering was utilized to identify clusters of panelists with similar response patterns. Fisher's exact test was then used to compare the demographic distributions across the resulting clusters to the pooled population (α = 0.05).

3. Results and Discussion

3.1. Physicochemical

The moisture contents of the sorghum flour before and after the precooking and drying were 10.36 ± 0.17% and 3.09 ± 0.06% (w.b.), respectively. The baobab fruit pulp powder had a moisture content of 9.55 ± 0.34% (w.b.).

The pH in yogurt is crucial because it regulates the fermentation endpoint, ensuring proper texture, flavor, and safety by preventing spoilage and whey separation (Liu et al. 2024). At 0 h of fermentation (immediately after inoculating with LAB), the pH of the samples decreased noticeably with increasing levels of baobab, with sample 17.5/32.5, characterized by the highest baobab powder content, exhibiting the lowest pH of 4.09 ± 0.03 (Table 1). The lower pH in samples with higher levels of baobab powder can be attributed to the inherent acidity of baobab powder. Tembo et al. (2017) reported a pH of 3.11 ± 0.01 for fresh baobab pulp due to the presence of organic acids like ascorbic, malic, tartaric, and citric acids. Sorghum flour has a pH of 6.2 (Hugo et al. 2003). The acidity of baobab powder played a considerable role in influencing the pH of the yogurt samples.

After 9 h of fermentation, the yogurt samples had pH values ranging from 3.99 ± 0.04 to 4.59 ± 0.02 (Table 1). According to the U.S. Food and Drug Administration, 21 CFR §131.200 (2021), yogurt must have a pH of 4.6 or lower when measured on the finished product within 24 h after filling. The pH values of the yogurts in this study were higher than those of the yogurt‐like samples made by blending milk, baobab, and sucrose, with pH ranging from 3.41 to 3.82 (Eke et al. 2013). The difference highlights the distinct pH dynamics resulting from the specific composition of the yogurt formulations, emphasizing the impact of baobab powder and sorghum flour on acidity levels.

DF is a nondigestible carbohydrate that, based on solubility in water, is divided into two classes: soluble DF, consisting of pectin, mucilage, and gum, and insoluble DF, consisting of lignin, cellulose, and part of hemicellulose (Bader Ul Ain et al. 2019). Consumption of DF may prevent immune‐related diseases, increase intestinal immune barrier function, and exert health‐promoting effects on the body (Beukema et al. 2020). Including fiber‐rich foods in daily consumption is recommended to support glycemic Control in diabetic patients and regulate energy intake and satiety (Duranti 2006). DF cannot be digested directly by human enzymes, but gut microbes can ferment it in the colon to produce fatty acids beneficial for health (Fu et al. 2022). Lactic acid fermentation is important for DF as it can break down complex polysaccharides, modifying the fiber before consumption (Wang et al. 2021).

The yogurt samples had total dietary fiber (TDF) contents ranging from < 1.0 ± 0.00% to 3.5 ± 0.19%, with soluble dietary fiber (SDF) ranging from < 0.05 ± 0.00% to 2.33 ± 0.13% and insoluble dietary fiber (IDF) from 0.68 ± 0.04% to 1.18 ± 0.08%. Baobab and sorghum are the sources of DF in the samples, as shown in Table 2. There was a detectable difference (p < 0.05) in TDF, SDF, and IDF across the samples due to varying ratios of sorghum and baobab in the blend. The Control, 32.5/17.5, and 25/25 samples showed no detectable differences in TDF or SDF; however, IDF differed detectably between the Control and samples 32.5/17.5 and 25/25. Which shows that unfermented sorghum and baobab had an impact on the IDF, influenced by sorghum's greater insoluble fiber fraction (Bader Ul Ain et al. 2019), while there was slight or no impact on TDF and SDF.

TABLE 2.

Total, soluble, insoluble dietary fibers (TDF, SDF, IDF), and vitamin C of the yogurt samples ⁱⁱ .

Sample (X/Y i )	TDF % (w.b.) iii	SDF % (w.b.) iii	IDF % (w.b.) iii	Vitamin C (mg/100 g)	Total soluble iron % (w.b.) iii
Control	2.08 ± 0.08b	1.25 ± 0.06b	0.83 ± 0.07c	0.09 ± 0.00a	7.02 ± 0.26a
17.5/32.5	3.5 ± 0.19a	2.33 ± 0.13a	1.18 ± 0.08a	0.04 ± 0.00b	– iv
25/25	2.45 ± 0.68b	1.46 ± 0.53b	0.99 ± 0.14b	0.04 ± 0.01b	– iv
32.5/17.5	2.2 ± 0.1b	1.24 ± 0.09b	0.97 ± 0.03b	0.02 ± 0.00c	11.52 ± 5.16a
40/10	1.38 ± 0.08c	0.57 ± 0.07c	0.78 ± 0.04cd	0.02 ± 0.00c	– iv
47.5/2.5	< 1.0 ± 0.00c	< 0.5 ± 0.00c	0.68 ± 0.04d	0.02 ± 0.00c	– iv

ⁱ Sample codes are expressed as X/Y, where X represents the grams of sorghum and Y represents the grams of baobab used in the blend.

ⁱⁱ For the same column, values (means ± SD) followed by the same letter are not significantly different based on Fisher's least significant difference test (α = 0.05).

ⁱⁱⁱ w.b. denotes wet basis.

^iv “–” denotes sample was not tested.

Sorghum flour contains an average TDF content of 8.16 g/100 g (USDA ²⁰²³), while baobab pulp powder contains 56.62 g/100 g (Monteiro et al. 2022). The TDF content varied across the different sorghum/baobab ratios as samples 17.5/32.5 and 25/25, with the highest amounts of baobab having higher values (3.5% and 2.45%) when compared to samples 40/10 and 47.5/2.5, with higher sorghum content having the lowest TDF values (1.38 and < 1.00). Garzón et al. (2024) reported a TDF content of 8.6 g/100 g when 200 g of white sorghum flour and skim milk powder were fermented with LAB. The higher amount of TDF reported could be attributed to the use of a higher quantity of sorghum flour in the study, when compared to 47.5 g, which was the highest amount in the yogurt samples. Aluko (2017) reported a crude fiber of 0.68% and 0.93% for yogurt enriched with 30% and 40% baobab pulp powder, respectively, which was slightly lower than our sample 17.5/32.5 which contains 1.18% IDF; our samples had a higher value as it also contains sorghum which could have increase the fiber content. Dauda et al. (2023) studied yogurt made from LAB fermentation of cow milk (99.15%) and baobab powder (0.85%), reporting a crude fiber content of 0.09, which was lower than our study and the study reported by Aluko (2017). The low crude fiber content can be attributed to the lower amount of baobab added to the milk before fermentation to obtain yogurt.

The highest SDF and IDF contents (2.33 ± 0.13% and 1.18 ± 0.08) were found in sample 17.5/32.5, which had a detectable difference (p < 0.05) in comparison to the Control and other samples, as it also contains the highest amount of baobab. As the baobab ratio decreased, the SDF and IDF decreased across the samples. However, there was no detectable difference between the Control and the sample with an equivalent ratio on the SDF and IDF, highlighting that fermentation had no impact on these samples. The increasing trend in SDF as the baobab increases aligns with previous studies that have reported baobab pulp powder to have a high soluble fiber content due to pectin concentration (Foltz et al. 2021). According to Bader Ul Ain et al. (2019), sorghum contains a higher fraction of IDF of 5.03 g while SDF had 1.68 g/100 g, which also explains why the samples with increased sorghum content had higher IDF values. The differences between the samples could indicate that baobab contributed more soluble fiber while sorghum provided more of the insoluble fiber fraction.

The vitamin C content of the yogurt varied across the samples, ranging from 0.02 ± 0.00 to 0.09 ± 0.00 mg/100 g, with the Control recording the highest value and showing a detectable difference (p < 0.05) among other samples as shown in Table 2. Baobab pulp powder contains vitamin C content ranging from 163.8 to 284.26 mg/100 g depending on the harvested origin (Monteiro et al. 2022), possessing a higher content of six to ten times that of an orange. The vitamin C content of the yogurt samples decreased as the baobab powder content decreased. Notably, sample 32.5/17.5 had a vitamin C value of 0.02 mg/100 g, which was very low when compared to the Control value of 0.09 mg/100 g that had similar sorghum/baobab amounts but was not fermented.

The reduction in vitamin C content could be attributed to the processing conditions during the fermentation step. Znamirowska et al. (2021) reported a reduction of 8% in the vitamin C content of fermented milk enriched with 30 ± 3 mg/100 g of vitamin C (ascorbic acid). The author noted that light, iron, and oxygen could have affected the products during storage, which could be the reason for the loss of vitamin C content in milk and dairy products. The 9 h of heat during fermentation may have influenced the stability of vitamin C in the yogurt samples. Though there was a reduction of vitamin C during the processing and storage of the yogurt samples, the study by Sreeramaiah et al. (2007) has shown that the presence of vitamin C can enhance iron solubility in food. The low vitamin C concentrations observed across all samples highlight a key limitation of the current processing method. Although baobab pulp is naturally rich in vitamin C, its instability to heat and oxygen likely contributed to substantial degradation during the fermentation processing and storage. Future research should prioritize optimizing processing conditions to retain vitamin C from the baobab in the samples. Hence, the reason for fermenting all the ingredients together when compared to the Control. The addition of baobab, which contains high vitamin C content, to dairy products can increase nutritional value. However, the unstable nature of vitamin C (Cunha et al. 2024) poses a challenge to the developed product.

The study did not assess physicochemical and microbiological changes over the typical 21–28 days yogurt shelf life; we acknowledge that the absence of a shelf‐life evaluation limits the practical applicability of the findings. There is a limited cold chain supply system in Senegal, especially in rural areas where our partnering farmers and schools are located. Our partnering farmers indicated short shelf life of yogurt of 1–3 days is more commonly practiced in traditional, smaller markets, due to the unpredictable electricity supply. Therefore, this study investigated the yogurt for up to 96 h to reflect the shorter practical storage period for the market of our partnering communities. Future work should include analyses of titratable acidity, sensory, instrumental flavor profile, texture, and nutrition over a standard storage period of 28‐day shelf‐life study to ensure practical applicability and provide a more complete physicochemical profile.

3.2. Microbial Analysis

The LAB count varied among the yogurt samples, influenced by the proportion of sorghum and baobab, and whether they were fermented together with milk or added unfermented as in the Control, ranging from 5.02 × 10⁴ to 6.53 × 10⁸ CFU/g (Figure 1). The LAB count is important for substantiating claims of “live and active cultures,” since US regulations require ≥ 10⁷ CFU/g at manufacture (USDA ²⁰²¹), while the voluntary International Dairy Food Association, Live & Active Cultures (LAC) seal requires ≥ 10⁸ CFU/g (IDFA ²⁰²⁴). The Control, which is a mixture between fermented yogurt and unfermented sorghum and baobab, had the highest LAB count among all samples (p < 0.05). Samples 47.5/2.5 and 40/10 also had higher LAB counts of 10⁸ CFU/g compared to other treatment samples (p < 0.05), suggesting that a higher level of sorghum flour supported LAB growth, potentially due to the fermentable carbohydrates in sorghum (Garzón et al. 2024).

FIGURE 1.

Bar charts showing lactic acid bacteria (LAB) count for the yogurt samples. Sample codes are expressed as X/Y, where X represents the grams of sorghum and Y represents the grams of baobab used in the blend. Control (32.5/17.5 added after fermentation of milk to yogurt). Same letter are not significantly different based on Fisher's least significant difference test (α = 0.05).

In contrast, samples 17.5/32.5 and 25/25, with the highest levels baobab powder having no detectable differences (p < 0.05) from each other, had the lowest LAB count of 10⁴ CFU/g among all samples. The lower LAB counts can be attributed to the acidic nature of baobab, containing organic acids (Tembo et al. 2017), this effect is evident in the lower initial pH values observed at 0 h (as low as 4.09) which, despite its nutritional benefits, may have likely hindered the growth of LAB. Zumunta and Umar (2020) reported a LAB count of 10⁴ CFU/mL in yogurts made with 9 and 15 h of LAB fermentation of milk and baobab, showing a similar impact of baobab on the final product. Although both the Control and sample 32.5/17.5 contained the same sorghum‐to‐baobab ratio (32.5 g and 17.5 g, respectively), their treatments differed in that the Control incorporated unfermented sorghum and baobab into yogurt after fermentation, whereas 32.5/17.5 was cofermented with milk. The Control showed the highest LAB count (10⁸ CFU/g), while 32.5/17.5 reached only 10⁷ CFU/g, suggesting that the method of incorporation, rather than ingredient proportion alone, strongly influenced microbial survival. The reduced LAB count in baobab‐rich samples can be mitigated through future research exploring strategies such as selecting acid‐tolerant LAB strains, adjusting the initial pH of the matrix, or modifying fermentation time. However, this study used a commercially available strain to ensure our partnering farmers can obtain and follow the developed recipes.

3.3. Instrumental Texture Analysis

The texture properties of the plant‐enriched yogurts are presented in Table 3. Instrumental texture analysis is critical for product development as it provides reproducible and quantifiable data. Though it gives weak predictions to sensory perception, it can be used as an objective measurement of the yogurt, complementing sensorial texture perception (Hutchings et al. 2024).

TABLE 3.

Instrumental texture of the yogurt samples ⁱⁱ .

Sample (X/Y i )	Firmness (g)	Consistency (g.s)	Cohesiveness (g)	Cohesion (g.s)
Control	138.79 ± 40.79c	1623.92 ± 556.01b	−117.98 ± 29.52a	−157.12 ± 68a
17.5/32.5	180.16 ± 28.81bc	1938.18 ± 134.07b	−154.51 ± 53.03a	−194.07 ± 10.52a
25/25	158.54 ± 7.26c	2019.41 ± 141.12b	−108.89 ± 10.17a	−202.62 ± 15.06a
32.5/17.5	152.35 ± 7.16c	2138.34 ± 146.9b	−102.08 ± 7.02a	−220.17 ± 18.67a
40/10	238.31 ± 10.96ab	3309.47 ± 267.49a	−190.17 ± 11.27a	−394.91 ± 28.13c
47.5/2.5	296.84 ± 94.26a	2207.55 ± 546.06b	−313.26 ± 112.47b	−301.39 ± 74.16b

ⁱ Sample codes are expressed as X/Y, where X represents the grams of sorghum and Y represents the grams of baobab used in the blend.

ⁱⁱ For the same column, values (means ± SD) followed by the same letter are not significantly different based on Fisher's least significant difference test (α = 0.05).

The firmness of the yogurt ranged from 138.79 ± 40.79 g to 296.84 ± 94.26 g for the samples. The firmness of yogurt is defined as the force necessary to achieve a specific deformation (Kose et al. 2018). Consequently, a higher firmness value in the sample indicates firmer yogurt. Firmness increased as the sorghum content increased, with sample 47.25/2.5 having the highest value (296.84 ± 94.26) showing a detectable difference (p < 0.05) from the Control, while sample 17.5/32.5 (180.16 ± 28.81) was higher than sample 25/25 (158.54 ± 7.26) but had no detectable difference with the Control. Khalifa and Ibrahim (2015) reported that adding starch as a stabilizer increased the water‐binding capacity and viscosity of camel milk yogurt. Their report aligns with this study's finding that the sorghum‐rich yogurt containing more starch had enhanced structural integrity, thus the highest firmness value.

The textural consistency of yogurt samples ranged from 1623.92 ± 556.01 g.s in the Control to 3309.47 ± 267.49 g.s in the 40/10 sample. Consistency in yogurt is the degree or resistance to flow, describing its thickness (Aportela‐Palacios et al. ²⁰⁰⁵). Consistency values increased with higher sorghum content with the Control (1623.92 ± 556.01 g.s) having the lowest value, and sample 40/10 (3309.47 ± 267.49 g.s) and sample 47.5/2.5 (2207.55 ± 546.06) the highest; however, sample 40/10 showed a higher value than sample 47.5/2.5, likely due to an optimal ratio of sorghum to baobab that enhanced water‐binding and gel network formation. Aside from sample 40/10, there was no detectable difference (p < 0.05) between the Control and other samples. Lorusso et al. (2018) reported an increase in viscosity as the quinoa flour quantity increased in the novel quinoa/dairy milk yogurt. Their report aligns with the present study showing that cereal‐based additives enhance yogurt viscosity. The fiber source in wheat bran added to the yogurt served as a resistance to flow in the yogurt and increased the consistency and viscosity as the fiber percentage increased (Aportela‐Palacios et al. ²⁰⁰⁵). Sorghum flour, as a source of DF, as discussed, has been shown to enhance the consistency of yogurt.

Cohesiveness values ranged from −102.08 ± 7.02 g to −313.26 ± 112.47 g, with the most negative value recorded in sample 47.5/2.5. Cohesiveness is defined as the degree a substance can undergo deformation while maintaining its structure, indicating the strength of internal bonds (Mudgil et al. 2017). The negative values indicate resistance to deformation, which increased with sorghum content, with sample 47.5/2.5 (−313.26 ± 112.47) showing the most negative value and a detectable difference (p < 0.05) from the other samples. Baobab contains pectin and soluble fibers, which may have mitigated the loss of gel structure and prevented whey separation (Lucey 2004), by enhancing gel stability at intermediate levels. Increasing the baobab had no detectable impact on yogurt cohesiveness, although sample 17.5/32.5 (154.51 ± 53.03) showed slightly higher values. Overall, the results suggest that sorghum primarily contributed to greater internal binding strength, while baobab moderated structural stability without detectably impacting cohesiveness. The trend was likewise reflected in the cohesion results.

The cohesion values ranged from −157.12 ± 68.0 g.s to −394.91 ± 28.13 g.s, with the most negative values recorded in samples 40/10 and 47.5/2.5 (Table 3). The Control exhibits the highest work of cohesion value, indicating greater resistance to withdrawal compared to the other yogurt samples, as cohesion reflects the force between particles that unites the substance (Edsman et al. 2015). Samples 47.5/2.5 and 40/10 had the lowest cohesion, suggesting that these yogurt samples’ structure are weakest and break apart more efficiently. There was a detectable difference (p < 0.05) between these samples with the Control and other samples. Our findings highlight the dual impact of combining precooked sorghum flour and baobab powder on the yogurt texture. While moderate inclusion of sorghum (25/25 and 32.5/17.5) improves consistency and firmness without compromising cohesion, a higher amount of sorghum (47.5/2.5) reduces structural integrity.

3.4. Iron Solubility

Fermentation increased the total soluble iron content, with sample 32.5/17.5 having 11.52 ± 5.16% (w/w, w.b.) compared to 7.02 ± 0.26% (w/w, w.b.) in the Control with the same amounts of sorghum and baobab but unfermented (Table 2). The increase in soluble iron content is consistent with the concept of bioaccessibility, which refers to the portion of a nutrient that is liberated from the food matrix during digestion, making it potentially available for uptake and utilization by the body (Sulaiman et al. 2021). There was no detectable difference (p < 0.05) between the Control and sample 32.5/17.5, however the difference in soluble iron content in the two samples may suggest that fermentation effectively enhanced the solubility of iron. Iron solubility increases with lactic fermentation, as evidenced by the degradation of antinutritional factors, such as phytate, through organic acid production, which enhances mineral solubility (Sreeramaiah et al. 2007; Scheers et al. ²⁰¹⁶).

The ascorbic acid in baobab may reduce the inhibitory impact of phytate in sorghum, further enhancing the release of iron in the samples (Piskin et al. 2022). Sorghum and baobab both contributed to the total iron content of the samples, with baobab providing a highly bioavailable source and sorghum contributing additional iron that was further released through fermentation. Although calcium may temporarily reduce iron absorption, long‐term evidence suggests that the body adapts to maintain iron homeostasis (Piskin et al. 2022), so including milk as a raw material, which is accessible to women farmers and helps reduce waste, may not compromise overall iron bioaccesibility. Traditionally, cooked grains such as sorghum and maize, along with baobab pulp, have been incorporated into yogurt‐like products in West Africa, enhancing both the nutritional quality and cultural relevance (Padonou et al. 2023). The results showed that when sorghum, baobab, and milk were cofermented, the process enhanced the iron solubility from the cereal and fruit, unlike when the milk was fermented into yogurt, and then mixed with baobab and precooked sorghum.

The standard deviation (± 5.16%) indicates variability among replicates, which may be attributed to sample handling, nonuniform fermentation kinetics, inherent biological variability, or analytical sensitivity limits (Bland et al. 2010). There is a need for sensitive quantification methods, using instruments capable of detecting small differences, to help reduce replicate variation. Several factors may have contributed to the variability, including sample handling during preparation and pH adjustment during gastric and enteric digestion, which may also alter mineral solubility. The variability limits the statistical robustness of the sample comparisons; the results indicate a possible but not statistically definitive enhancement from cofermentation. Future improvements, such as the selection of iron‐free reagents, could help limit variability. However, the presence of soluble iron in fermented samples represents a nutritional improvement over the unfermented Control. Studies using in vitro digestion models have demonstrated that soluble iron does not always translate into increased uptake or ferritin formation (Scheers et al. 2016). Although the increased soluble iron does not guarantee enhanced bioavailability, it shows potential for improved iron uptake and absorption. Hence, methodological refinement is needed to reduce variability, and further assessment using in vivo digestion is required.

3.5. Sensory Evaluations of Plant‐Enriched Yogurts

The consumer sensory acceptance test included the overall liking, liking of appearance, flavor, and mouthfeel of the yogurt samples. The overall liking scores ranged from 4.7 ± 2.1 to 5.5 ± 1.9 on a nine‐point hedonic scale (Figure 2). Samples 40/10 and 17.5/32.5 received the highest liking scores and was detectably different from the Control (p < 0.05). In contrast, the Control received the lowest overall liking score, indicating that the panelists did not prefer unfermented sorghum/baobab in the yogurt, though it had no detectable difference from samples 25/25, 32.5/17.5, and 47.5/2.5. The sample containing 20% baobab (40/10) achieved the highest flavor and overall liking scores in our study. A study on yogurt enriched with 0%–40% baobab showed that consumers preferred just the sample with 10% baobab, having the highest overall liking score (Aluko 2017). Our results therefore suggest that cofermentation with sorghum and milk may allow higher baobab levels to be incorporated while maintaining consumer preference. Both sample 40/10, containing a higher proportion of sorghum, and sample 17.5/32.5, with the highest proportion of baobab, received the highest overall liking scores, indicating that consumers accepted contrasting formulations. The outcome reflects how different subgroups of consumers evaluated the yogurt. Cluster analysis showed a detectable difference in ethnicity among the panelists, thereby explaining the contrast, with certain groups preferring samples higher in sorghum while others favored those higher in baobab (Section 3.6).

FIGURE 2.

Grouped bar charts showing sensory liking of the yogurt samples. Sample codes are expressed as X/Y, where X represents the grams of sorghum and Y represents the grams of baobab used in the blend. Control (32.5/17.5 added after fermentation of milk to yogurt). Same letter are not significantly different based on Fisher's least significant difference test (α = 0.05).

Appearance liking scores were highest for the Control (6.1 ± 1.5) and sample 40/10 (6.1 ± 1.6), and lowest for sample 47.5/2.5 (5.7 ± 1.9), which is the only sample with a detectable difference (p < 0.05) from the Control (Figure 2). The addition of unfermented sorghum and baobab to the fermented yogurt for the Control exhibited a visibly textured surface that panelists may have associated the appearance with creaminess and product quality as shown in Figure S1. While sample 47.5/2.5 appeared lighter and less dense, indicating that higher sorghum levels with minimal baobab reduced the visual appeal. Samples 17.5/32.5 and 25/25, with higher baobab levels appeared noticeably thick and darker in coloration, whereas sample 47.5/2.5 with highest sorghum was thin and lighter, and was consequently perceived less favorably in terms of appearance. Appearance scores were similar across samples, except 47.5/2.5, which was rated lower than the Control and 40/10, indicating less distinct visual than flavor evaluations.

The Control had the lowest flavor liking score (4.6 ± 2.3), which was noticeably lower (p < 0.05) than that of sample 40/10 (5.5 ± 1.9), which had the highest liking score. The Control's low flavor liking score could be attributed to the unfermented sorghum and baobab, which made the sample more sour and tart than the other samples. Only samples 40/10 and 47.5/2.5, which recorded the highest scores for milky flavor (0.464 and 0.432, respectively) in the CATA attributes (Table 4), were detectably different in overall flavor liking from the Control, which was instead characterized by the highest tart (0.536) and sour flavor (0.656) values. The results show that the panelists favored samples containing more sorghum and disfavored samples containing more baobab or unfermented baobab in the case of the Control sample, which gave the yogurt a tart note.

TABLE 4.

Multiple pairwise comparisons of CATA attributes between the yogurt samples ⁱ .

Attribute ii	Control	17.5/32.5	25/25	32.5/17.5	40/10	47.5/2.5	Mean impact—overall liking iii
Smooth_MF	0.464a	0.352ab	0.320ab	0.424ab	0.296b	0.352ab	0.867*
Creamy_MF	0.416ab	0.504ab	0.456ab	0.520ab	0.560a	0.384b	1.186*
Thin_MF	0.472a	0.056b	0.040b	0.088b	0.048b	0.528a	−0.495*
Gritty_MF	0.312a	0.384a	0.408a	0.312a	0.440a	0.416a	−1.042*
Chalky_MF	0.400b	0.640a	0.592a	0.488ab	0.592a	0.504ab	−0.818*
Light_MF	0.416a	0.176b	0.128b	0.104b	0.072b	0.472a	0.206
Thick_MF	0.112b	0.504a	0.488a	0.584a	0.616a	0.128b	0.451*
Whipped_MF	0.128b	0.176ab	0.264a	0.280a	0.224ab	0.104b	0.756
Sweet_FL	0.136b	0.352a	0.328a	0.328a	0.304a	0.256ab	1.460*
Sour_FL	0.656a	0.368bc	0.208cd	0.168d	0.320bcd	0.432b	−1.135*
Creamy_FL	0.200c	0.360abc	0.440ab	0.488a	0.504a	0.304bc	0.981*
Tangy_FL	0.544a	0.424ab	0.208c	0.208c	0.304bc	0.376b	−0.481*
Banana_FL	0.432b	0.640a	0.640a	0.584ab	0.632a	0.520ab	0.412*
Tart_FL	0.536a	0.328b	0.256bc	0.112c	0.296b	0.376b	−0.743*
Cereal_FL	0.072b	0.144ab	0.168ab	0.208a	0.168ab	0.152ab	0.608
Milky_FL	0.224b	0.360ab	0.392a	0.400a	0.464a	0.432a	1.004*
Nutty_AT	0.120b	0.240ab	0.288a	0.312a	0.240ab	0.216ab	0.439*
Earthy_AT	0.200c	0.320abc	0.400ab	0.432a	0.272bc	0.280bc	−0.245
Sweet_AT	0.144a	0.232a	0.272a	0.264a	0.256a	0.208a	1.479*
Sour_AT	0.632a	0.352bc	0.232c	0.248bc	0.368bc	0.400b	−1.028*
Tangy_AT	0.432a	0.416a	0.304a	0.144b	0.392a	0.304a	0.197

ⁱ For the same row, values followed by the same letter(s) are not significantly different based on Sheskin's critical difference test (α = 0.05).

ⁱⁱ Flavor (FL), mouthfeel (MF), and aftertaste (AT) yogurt attributes.

ⁱⁱⁱ Values with * indicate significant mean impact of CATA attributes based on a two‐sample t‐test (α = 0.05).

There was no detectable difference in the mouthfeel liking scores among the samples (p < 0.05). This could be attributed to the fact that they all contained sorghum and baobab. Although the Control contained unfermented sorghum and baobab; the results show that fermentation had little or no impact on the mouthfeel liking of the yogurt. Greis et al. (2022) reported that yogurts may share similar mouthfeel profiles despite differing rheological properties, highlighting that sensory evaluations alone may not capture subtle physicochemical differences.

3.6. Check‐All‐That‐Apply

The CATA analysis was used to profile how the panelists perceived the yogurt's sensory features. Cruz et al. (2013) reported that CATA analysis performs well in describing yogurts’ sensory attributes, as the terms used are preselected based on the samples. In the PcoA plot of the CATA and acceptance data (Figure 3), the first two dimensions explained 72.8% (Dim1) and 16.5% (Dim2) of the variation, which indicates that the plot captured most of the information. Sample 40/10 was located in the positive region of Dim2, with the highest scores for creamy and milky flavor, along with thick and creamy mouthfeel (Table 4), which aligned with its highest overall liking score (5.5). Notably, sample 17.5/32.5 was associated with tangy, tart and sour flavor attributes, reminiscent of a high baobab content. Oludara and Bamidele (2019) have reported similar attributes on yogurt stabilized with baobab pulp, to have a tart and sour taste because of the presence of organic acid in baobab.

FIGURE 3.

Principal coordinate analysis (PCoA) biplot showing check‐all‐that‐apply attributes of the yogurt samples. Sample codes are expressed as X/Y, where X represents the grams of sorghum and Y represents the grams of baobab used in the blend. Control (32.5/17.5 added after fermentation of milk to yogurt). CATA responses across flavor (FL), mouthfeel (MF), and aftertaste (AT) attributes.

The Control was positioned on the negative side of Dim1 (Figure 3), associated with sour aftertaste and astringent flavor attributes, and a thin texture. This sample corresponded with lower hedonic scores, indicating reduced panelist acceptance. Studies have shown that yogurt deemed too thin is judged by consumers and given lower overall liking scores (Grygorczyk et al. 2013). The CATA analysis revealed detectable differences (p < 0.05) among the yogurt samples in terms of mouthfeel, flavor, and aftertaste attributes (Table 4). Several attributes showed a detectable mean impact on overall liking, highlighting their influence in shaping the panelists’ preferences (Figure 4).

FIGURE 4.

Diverging bar chart showing mean impact scores for the key attributes of the yogurt samples. Flavor (FL), mouthfeel (MF), and aftertaste (AT) yogurt attributes.

Creamy (1.186) and smooth (0.867) mouthfeel were positively associated with overall liking, with creamy mouthfeel exerting the strongest impact (p < 0.05) compared to other mouthfeel attributes such as gritty (−1.042) and chalky (−0.818), which were negatively associated with liking (Table 4). The Control, which contained unfermented sorghum, and sample 47.5/2.5, with the highest sorghum proportion, both had detectable difference for light (0.416 and 0.472) and thin (0.472 and 0.528) mouthfeel attributes compared to samples 17.5/32.5, 25/25, 32.5/17.5, and 40/10 (Table 4). These results suggest that higher sorghum content, whether fermented or unfermented, contributes to a lighter and thinner texture. Consistent with the Control, instrumental texture lower firmness (138.79 ± 40.79) and consistency (1623.92 ± 556.01) values. A gritty mouthfeel was prevalent across all samples, negatively impacting overall liking (−1.042; Table 4). Although the sorghum flour was precooked and milled before fermentation, the fiber particle size and hydration properties may still have contributed to the gritty perception. There is also a possibility to explore products with more sorghum, some grittiness, but actually enjoyed by the Africans, as shown from the cluster's overall liking in Figure 4. Further studies should investigate finer milling and extended hydration to reduce particle size and enhance water absorption, thereby minimizing grittiness in the samples.

In contrast, the baobab‐rich samples (17.5/32.5, 25/25, 32.5/17.5, and 40/10) exhibited higher thick mouthfeel scores (0.504–0.616) than both the Control (0.112) and 47.5/2.5 (0.128), indicating that baobab addition enhanced perceived thickness. Sample 40/10 with higher firmness and consistency values (238.31 g and 3309.47 g.s) was perceived by panelists as thicker in mouthfeel (0.616) with a detectable difference (p < 0.05) from the thick mouthfeel attribute (0.112) of the Control sample. Gritty mouthfeel had no detectable difference among the samples, as they all contained sorghum (Table 4). However, gritty mouthfeel had a strong negative mean impact on liking (−1.042), indicating that its presence consistently reduced consumer acceptance. These findings align with previous studies that have shown grain‐based yogurt is attributed to grittiness and chalkiness, which reduces its overall liking (Ma et al. 2024).

Sweet flavor had the highest positive impact on the overall liking (1.460) for the flavor attributes, with all the samples scoring higher than the Control (Table 4). Samples 25/25 and 40/10, which had the highest overall liking scores, also recorded higher flavor values for creamy (0.440 and 0.504) and milky (0.400 and 0.464) than the Control with a detectable difference (p < 0.05). This indicates that these attributes were important drivers of preference in our study, consistent with evidence that a creamy and milky flavor is positively associated with consumer liking in plant‐based milk (Jaeger et al. 2024). Sour and tangy flavors were negatively associated with liking (−1.135 and −0.481, respectively), with the Control showing the strongest sour notes.

Sweet aftertaste had the strongest positive impact on overall liking (1.479) among all attributes, although no detectable differences were observed across samples (Table 4). Nutty and sweet aftertastes positively contributed to the overall liking, while sour aftertaste strongly negatively influenced panelist preference in the Control and sample 17.5/32.5. Overall, sample 40/10 produced the most desirable sensory qualities, showing that optimized sorghum and baobab ratios enhance creaminess and sweetness while reducing sourness and chalkiness. It is important to note that most CATA attributes did not differ detectably among the cofermented samples, indicating that their sensory profiles were broadly similar. However, the discussion of specific attributes that differed detectably from the Control highlighted general sensory results rather than sample‐specific differences.

3.7. Cluster Analysis of Consumers and Sensory Evaluation

The panelists were categorized into two clusters as shown in Table 5. There was a detectable difference in the ethnicity from the pooled percentage based on Fisher's exact test (p < 0.05). Cluster 1 had 75.9% Caucasians, while Cluster 2 had a greater representation (52.4%) of Black or African panelists. Notably, these Black participants were Africans who were international students studying in the United States, rather than African American citizens, which may account for cultural influences on preference patterns observed between the clusters.

TABLE 5.

Cluster analysis of the panelists using agglomerative hierarchical clustering ⁱⁱ .

Demographic grouping i		Total pool	Cluster 1	Cluster 2
Gender	Man	43	26	17
			(31.33%)	(40.48%)
	Woman	81	57	24
			(68.67%)	(57.14%)
	Nonbinary/third gender	1	0	1
			(0.00%)	(2.38%)
Age	18–24	51	38	13
			(45.78%)	(30.95%)
	25–34	46	23	23
			(27.71%)	(54.76%)
	35–74	28	22	6
			(26.51%)	(14.29%)
Ethnicity	White	81	63	18
			(75.90%)*	(42.86%)*
	Black or African	32	10	22
			(12.05%)*	(52.38%)*
	Other	12	10	2
			(12.05%)*	(4.76%)*
Yogurt consumption	Daily	11	8	3
			(9.64%)	(7.14%)
	4–6 times a week	20	15	5
			(18.07%)	(11.90%)
	2–3 times a week	33	22	11
			(26.51%)	(26.19%)
	Once a week	42	23	19
			(27.71%)	(45.23%)
	Other	10	7	3
			(8.43%)	(7.14%)
	Never	9	8	1
			(9.64%)	(2.38%)

ⁱ The results are displayed as the number of individuals and the corresponding percentage relative to the total in the cluster.

ⁱⁱ The percentage followed by an * indicates a significant difference from the pooled percentage based on the Fisher's exact test (α = 0.05).

The overall liking scores of the yogurts by Cluster 2 ranged from 6.0 ± 1.7 to 7.0 ± 1.5 (Figure 5). This cluster equally favored the samples with varying sorghum/baobab ratios (40/10, 25/25, 17.5/32.5, with overall liking 6.8, 6.8, and 7.0), potentially attributed to the familiarity of the mostly West African panelists’ palates with sorghum and baobab, commonly consumed in African countries. The earthy aftertaste was highest in sample 32.5/17.5 (0.432) but lower in 40/10 (0.272), suggesting that Cluster 2 panelists were less favorable toward the heavier earthy aftertaste, which may have contributed to their reduced liking of 32.5/17.5. The CATA attributes result showed that Africans preferred higher quantities of either sorghum flour or baobab powder. For Cluster 1, the tart, earthy, and tangy attributes are weaker drivers of acceptance than for Cluster 2. Overall, the tart and tangy attributes of baobab, familiar to African consumers, act as positive drivers for the panelists. On the other hand, Cluster 1, with a higher representation of Caucasians, showed overall liking scores ranging from 4.1 ± 1.6 to 4.8 ± 1.7. This cluster favored the sample with less baobab (40/10, overall liking 4.8), possibly due to their taste buds being less accustomed to baobab, as unfamiliarity with novel foods can reduce acceptance, while greater familiarity enhances liking (Xiao et al. 2025). The influence of ethnic background on the sensory perception of the yogurt could explain the variation across clusters. Nevertheless, both clusters disfavored Control, with overall liking scores of 4.1 and 6.0 for Clusters 1 and 2, respectively. This underscores the impact of fermentation on improving the sensory quality of plant‐enriched yogurts.

FIGURE 5.

Grouped bar charts showing overall liking of the yogurt samples by Cluster 1 and Cluster 2 panelists. Sample codes are expressed as X/Y, where X represents the grams of sorghum and Y represents the grams of baobab used in the blend. Control (32.5/17.5 added after fermentation of milk to yogurt). Cluster 1 (mostly Caucasians), Cluster 2 (mostly Black or African). Same letter are not significantly different based on Fisher's least significant difference test (α = 0.05).

4. Conclusion

The plant‐enriched yogurt samples demonstrate that controlled cofermenting sorghum flour and baobab powder with dairy milk effectively enhanced iron solubility, the nutritional and sensory qualities of the yogurt while supporting microbial viability. Variations in sorghum and baobab ratios influenced product characteristics, with higher sorghum levels improving firmness, consistency, and LAB growth. In comparison, higher baobab levels increased soluble fiber and vitamin C content but reduced microbial counts due to acidity. Notably, fermentation enhanced iron solubility, and consumer sensory testing revealed that samples with balanced ratios of sorghum and baobab (specifically 40/10 and 25/25) increased overall liking scores, characterized by creaminess, and reduced chalkiness. Exploring changes in volatile and nonvolatile compounds can provide insights into flavor profiles. Future research on flavor profile analyses could enhance the understanding of flavor development in yogurt samples.

The findings highlight the potential of plant‐enriched yogurt as a culturally relevant, nutrient‐rich dairy product with enhanced biofunctional properties that could address iron deficiency and dietary quality challenges in West Africa and beyond. By utilizing indigenous crops through innovative food processing, this work supports both nutritional security and the livelihoods of smallholder women farmers. Future studies should incorporate community‐based sensory evaluation in communities in West Africa that traditionally consume sorghum and baobab fermented foods to inform better product optimization. Future research should also optimize processing to preserve vitamin C stability, evaluate the in vivo bioavailability of nutrients, and expand community‐based sensory testing in African contexts, thereby advancing the role of plant‐enriched fermented dairy in sustainable African food systems.

Author Contributions

Chidimma Ifeh: conceptualization, methodology, investigation, writing – original draft, writing – review and editing, project administration, data curation, visualization. Edwin Allan: writing – review and editing, methodology, visualization. Aliou Ndiaye: conceptualization, writing – review and editing. Mary P. Miles: writing – review and editing, supervision, funding acquisition, investigation, methodology, conceptualization. Wan‐Yuan Kuo: conceptualization, investigation, funding acquisition, writing – review and editing, visualization, supervision, methodology.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Supporting Information Figure S1: jfds70978‐sup‐0001‐FigureS1.docx