henhaobio
Sign in

Current Research in Food ScienceBioengineeringPeer-reviewed journal article

Biaxial testing and sensory texture evaluation of plant-based and animal deli meat

Skyler R. St. PierreiD, Lauren Somersille SibleyiD, Steven TraniD, Vy TraniD, Ethan C. DarwiniD, Ellen Kuhl AiD

PDFDOIPreprintCodeData
In-depth full-text digest available · Jump to digest ↓

Abstract

Animal agriculture is one of the largest contributors to global carbon emissions. Plant-based meats offer a sustainable alternative to animal meat; yet, people are reluctant to switch their diets and spending habits, in large due to the taste and texture of plant-based meats. Deli meat is a convenient form of protein commonly used in sandwiches, yet little is known about its material or sensory properties. Here we performed biaxial testing with multiple different stretch ratios of four plant-based and four animal deli meats and fit accurate material models to the resulting stress-stretch data. Strikingly, the plant-based products, turkey, ham, deli, and prosciutto, with stiffnesses of 378 {+/-} 15 kPa, 343 {+/-} 62 kPa, 213 {+/-} 25 kPa, and 113 {+/-} 56 kPa, were more than twice as stiff as their animal counterparts, turkey, chicken, ham, and prosciutto, with 134 {+/-} 46 kPa, 117 {+/-} 17 kPa, 117 {+/-} 21 kPa, and 49 {+/-} 21 kPa. In a complementary sensory texture survey, n = 18 participants were able to correlate the physical stiffness with the sensory brittleness, with Spearmans correlation coefficient of{rho} = 0.857 and p = 0.011, but not with the sensory softness or hardness. Notably, the participants perceived all four plant-based products as less fibrous, less moist, and less meaty than the four animal products. Our study confirms the common belief that plant-based products struggle to meet the physical and sensory signature of animal deli meats. We anticipate that integrating rigorous mechanical testing, physics-based modeling, and sensory texture surveys could shape the path towards designing delicious, nutritious, and environmentally friendly meats that mimic the texture and mouthfeel of animal products and are healthy for people and for the planet. Data and code are freely available at https://github.com/LivingMatterLab/CANN.

In-depth summary

Claude Opus 5.5 · xhigh effort · 8 Nov 2026 · from the full text of version 1 (25 Feb 2025) · CC BY 4.0

Plant-based deli slices are over twice as stiff as animal deli meats under biaxial tension; stiffness tracks brittleness rated by 18 tasters (ρ = 0.857).

Texture is a major barrier to consumer uptake of plant-based meat, yet thin deli slices are ill-suited to texture profile analysis and no prior work had jointly characterized the mechanics and sensory texture of plant-based versus animal deli meats. The study applies biaxial extension, suited to thin sheets, plus hyperelastic and neural-network model fitting and a consumer survey, giving the first constitutive models for deli meat.

Key results

  1. ShownAs a group, the plant-based deli products were reported to be over twice as stiff as the animal products. The three Tofurky products were the stiffest, animal prosciutto was the softest, and plant prosciutto had stiffness comparable to animal turkey, chicken and ham.

    Stiffness E (= 3µ), n = 8 samples per product: plant turkey 378 ± 15, plant ham 343 ± 62, plant deli 213 ± 25, plant prosciutto 113 ± 56 kPa vs animal turkey 134 ± 46, chicken 117 ± 17, ham 117 ± 21, prosciutto 49 ± 21 kPa. · Abstract; 3. Results; Current plant-based deli meats fail to mimic the…

  2. ShownTasters rated all four animal meats as more fibrous, moister and meatier than all four plant-based products. Plant prosciutto was the only plant product that ranked among the animal meats for softness, fattiness and brittleness.

    5-point Likert ratings from n = 18 participants on twelve features; one-way ANOVA (p < 0.05) found significant between-product differences for every feature except viscous, springy and sticky. · 3. Results; Current plant-based deli meats fail to mimic the…

  3. SuggestedMeasured stiffness correlated with perceived brittleness but not with perceived softness or hardness. This differs from the authors' earlier work on three-dimensional minced meats, where stiffness tracked softness inversely.

    Spearman correlations across products: stiffness vs brittleness ρ = 0.857, p = 0.011; stiffness vs fattiness ρ = 0.810, p = 0.022 (described as inverse); softness vs moistness ρ = 0.929, p = 0.002. These were the only significant pairs among stiffness and nine ANOVA-significant features. · 3. Results; Physical stiffness is correlated with sensory…

  4. ShownAmong the hyperelastic models, Mooney Rivlin fit best overall, the two-term models found by the neural network came second, and neo Hooke fit worst. The differences between the three were small.

    Largest R^2 gap was 0.1 (plant prosciutto: neo Hooke R^2 = 0.36 vs Mooney Rivlin R^2 = 0.46); animal chicken fit best and plant prosciutto worst; four meats averaged R^2 above 0.75. · 3. Results; We discovered the first constitutive model for…

  5. ShownThe L1-regularized constitutive neural network chose two-term models from a small subset of its candidate terms. For animal prosciutto it recovered exactly the Mooney Rivlin model; for animal ham it chose exponential terms in both invariants.

    Network has eight candidate terms, 16 weights and 28 possible two-term models; the linear first-invariant term (I1-3) was selected for six of eight meats, paired with its exponential form in four of those; only four distinct terms were ever chosen. · 2.7. Automated model discovery; 3. Results; We discovered the first constitutive model for…

Methods, data and reproducibility

Deli slices (n = 8 per product) were stretched on a CellScale BioTester 5000 in five biaxial modes; averaged stress-stretch curves were fit with neo Hooke and Mooney Rivlin hyperelastic models and an L1-regularized constitutive neural network, and stiffness was rank-correlated with a twelve-feature Likert texture survey.

  • Plant-based deli slices: Tofurky oven roasted, hickory smoked and smoked ham style, plus Mia prosciutto style; wheat gluten is the main protein · n = four products; n = 8 samples each
  • Animal deli slices: Hillshire Farm turkey breast, chicken breast and black forest ham, plus Columbus prosciutto (pork and salt only) · turkey, chicken, pig (pork) · n = four products; n = 8 samples each
  • Untrained consumer panel that tasted bite-sized samples of all eight products and completed neophobia and meat-attachment questionnaires (Stanford IRB-75418) · human · n = n = 18 participants
Methods & tools
CellScale BioTester 5000 biaxial tester, neo Hooke hyperelastic model, Mooney Rivlin hyperelastic model, Constitutive neural network with L1 regularization, Adam optimizer, Least squares parameter fitting, Food Neophobia Survey, Meat Attachment Questionnaire, Sensory Texture Survey (5-point Likert), One-way ANOVA, Spearman's rank correlation
Software & environment
The Python scripts for model fitting are in the repository; the paper gives no package versions, dependency list or container.
Reproducibility gaps
The printed tables shorten the data to eleven points per loading mode (extended data are in the repository). Detailed participant demographics are withheld for confidentiality.
Limitations (5)
  • The sensory panel was small (n = 18) and untrained. The authors say this may explain mismatches such as animal prosciutto being rated harder and less soft despite being less stiff, and call for larger studies. · stated by the authors
  • For four of the eight meats, one- and two-term hyperelastic models could fit only some loading modes at once. The authors call for tests at larger stretches and other stretch rates, and possibly viscoelastic models. · stated by the authors
  • Because deformation through the slice thickness could not be measured, the authors assumed the meats were perfectly incompressible in their constitutive modeling. · stated by the authors
  • Each Spearman correlation uses only eight product means, tested against nine features with no reported multiple-comparison correction. The fattiness link is called inverse but reported as ρ = 0.810, and the methods mention ten samples instead of n = 8. · noted in this summary
  • Each product comes from one commercial brand, and three of the four plant products share a Tofurky wheat gluten and tofu base. The plant-versus-animal contrasts may reflect these specific formulations rather than plant-based deli meat in general. · noted in this summary

Read this if you design plant-based meat analogs or study food mechanics and want biaxial stress-stretch data, fitted hyperelastic models, and sensory texture comparisons for thin deli slices.

Citations
10
FWCI
2.45
Field percentile
top 11%

Reproducibility ledger

Nobody has reported trying to reproduce this paper yet. Did you replicate an experiment or run the code? Your report saves the next lab a week.

Sign in to file a reproducibility report.

Related papers

Similarity Map
SelectedSame fieldCross-field
THISPierre '24Dunne '25Pierre '26Pierre '23Vervenne '25Pierre '24
82% semantic matchBioengineering
Skyler R. St. Pierre, Ethan C. Darwin, Divya Adil+9 more·npj Science of Food·Nov 15, 2024· 43 citations