How a lab can convert their sensory panel data to the use of a Texture Analyser

Converting sensory panel data to the use of a Texture Analyser is a process that can significantly streamline quality control and product development in food science. Here's a step-by-step procedure for how a lab can make this transition:

Step 1: Identify key sensory attributes and context for measurement

First, establish the context for measurement—Are you benchmarking against an existing product? Seeking to match a “gold standard”? Or aiming to achieve a specific mouthfeel or functional property?

Next, review your existing sensory panel data to identify the most important texture-related attributes for your product. These might include properties like hardness, crispness, chewiness, or adhesiveness.

Step 2: Select appropriate Texture Analyser tests

Choose Texture Analyser tests that best simulate the sensory evaluation process which will make interpretation of your results easier:

• For hardness, it is likely that you will use a compression/penetration test
• For crispness, consider a snap or fracture test
• For chewiness, a two-cycle compression test (Texture Profile Analysis) may be suitable
• For adhesiveness, use a probe withdrawal test

Step 3: Develop test protocols

Create standardised protocols for each Texture Analyser test, including:

• Sample preparation methods – this will be key to the repeatability of your results.  Standardise sample size, shape, temperature, and storage to minimise sources of error.
• Test parameters (speed, distance, probe type) – this will be an iterative process of choosing a pre-test speed, test speed, test distance and trigger force to optimise the settings to obtain the most repeatable results. The approach speed, the amount of force the Texture Analyser is looking to find for the sample surface before collecting data and the degree of deformation will be based on your sample size, probe size and sample hardness. Check that the Texture Analyser test that you have developed can differentiate between samples that you know are different.
• Number of replicates – Any decision regarding texture analysis results should be based on as many replicates as there is time to gather.  There will be a sweet spot between the time that is available to run as many tests as possible and how many are required for statistical reasons.
• Data analysis procedures – You will need to ensure that you are measuring the correct parameters from your graph profiles that are meaningful to your sample.  Stable Micro Systems and their distributors are experts at this and are always happy to help.

If the method that you have developed doesn’t reveal useful or repeatable data you will need to adjust the sample preparation, probe, test type of settings.  This is the heart of the iterative process.

• Review results and systematically tweak one parameter at a time if needed. 
• Optimise until repeatable, discriminative, and relevant results are obtained for your texture attribute(s).
• Confirm repeatability (multiple runs, multiple users if possible).
• Validate with known controls or reference materials.

Step 4: Conduct parallel testing

Perform both sensory panel evaluations and Texture Analyser tests on the same set of samples. This step is crucial for establishing correlations.

Step 5: Statistical analysis

Use statistical methods to analyse the relationship between sensory scores and instrumental measurements:

• Perform correlation analysis
• Consider using principal component analysis (PCA) to identify key relationships in the dataset
• Use regression analysis to model sensory attributes from instrumental parameters

Step 6: Develop predictive models

Based on the statistical analysis, create models that predict sensory attributes from Texture Analyser measurements. These models will likely be a combination of different instrumental parameters.

Step 7: Validate the models

Test the predictive models on a new set of samples to ensure their accuracy and reliability.

Step 8: Establish acceptance criteria

Define acceptable ranges for Texture Analyser measurements that correspond to desired sensory attributes. These will serve as your new quality control standards.

Step 9: Train personnel

Provide training to lab personnel on how to use the Texture Analyser, interpret results, and relate them to sensory attributes.

Step 10: Implement and monitor

Begin using the Texture Analyser for routine quality control, while periodically conducting sensory panels to ensure the instrumental measurements continue to align with sensory perceptions.

Step 11: Continuous improvement

Regularly review and refine your models and procedures as you gather more data and encounter new product variations.

By following this procedure, a lab can effectively transition from relying solely on sensory panels to using a Texture Analyser for more objective, consistent, and efficient texture analysis. Remember that while instrumental analysis offers many advantages, it's still valuable to conduct occasional sensory evaluations to ensure the Texture Analyser results continue to reflect true sensory experiences.