Addressing methodological challenges in sensory science for a deeper understanding of consumer testing context

Albiol Tapia, Marta

Permalink

https://hdl.handle.net/2142/116081 Copy

Description

Title

Addressing methodological challenges in sensory science for a deeper understanding of consumer testing context

Author(s)

Albiol Tapia, Marta

Issue Date

2022-07-13

Director of Research (if dissertation) or Advisor (if thesis)

Lee, Soo Yeun

Doctoral Committee Chair(s)

Schmidt, Shelly J

Committee Member(s)

• Pepino de Gruev, Marta Yanina
• Lee, Youngsoo

Department of Study

Food Science & Human Nutrition

Discipline

Food Science & Human Nutrition

Degree Granting Institution

University of Illinois at Urbana-Champaign

Degree Name

Ph.D.

Degree Level

Dissertation

Keyword(s)

• sensory
• consumer
• context
• environmental setting
• hedonic contrast
• outlier
• product information
• remote testing
• online testing
• engagement
• consumption frequency
• liking
• acceptance

Abstract

Sensory evaluation of food relies heavily on eating context. Based on literature findings, consumer testing methodologies need additional studies to further understand the effect of different contextual factors in food evaluations. This research addresses how context effect, created from differences in environmental setting and product information, affects consumer’s acceptability of tea and cola. Environmental setting differences, created by testing in a sensory laboratory and a study commons, did not appear to have a large influence on hedonic scores. However, tea samples showed significantly higher appearance liking ratings in the sensory booth location, likely due to lighting conditions. Product information differences, created by serving the same samples with or without product name and image, varied largely depending on the product category and brand, where well-established cola brands received higher ratings when product information was provided. The laboratory sensory booth setting provided higher panelist sensory engagement overall, as consumers reported this location to assist more in their evaluation of samples. Next, the effect of large differences across samples served in the same evaluative set in a consumer acceptance test design was investigated. In this study, tea and cola products were used to determine the effect of including a significantly different sample, namely an outlier, in a set of samples that are similar in liking. An additional consumer test with orange juice products was conducted to determine the effects of an outlier being more liked or less liked compared to the other samples, and the outlier being served randomly versus fixed at a given presentation position. Results showed that including a less-liked outlier in an evaluative set did not create a context effect; thus, the liking ratings of other samples were not compressed towards the high end of the hedonic scale. However, including an outlier in the first serving position resulted in a higher discrimination of other samples. Fixing an outlier in the last serving position resulted in significantly lower scores for less-liked outliers, but no difference for more-liked outliers. Overall, this study showed that contrast effect is minimal when created by the presence of an outlier in an evaluative set, where liking scores display a rather absolute behavior. In the context of the COVID-19 pandemic, it became challenging for sensory scientists to conduct in-person sensory tests, particularly large central location tests. Given that sensory literature comparing central location and home use tests shows no clear consensus about how each methodology affects sample ratings and panelist engagement, the next phase of this research was designed to address this challenge. Three methods of remote consumer testing were compared in terms of sample acceptance, sensory engagement, and method practicality. Participants rated chocolate-chip cookie products on a 9-pt hedonic scale in each of three methods: 1) a live (synchronous) Zoom session, 2) an asynchronous video-guided session, and 3) a fully written protocol session. Results showed no significant differences in sample liking pattern across the methods used. Engagement scores approached the limit of significance for the Active Involvement dimension, indicating panelists were least likely to feel distracted, zoned out or lose interest in the written protocol method. There were no significant differences in the time spent on the test by the panelists across the three methods. Overall, asynchronous methods showed to be most suitable in terms of the convenience of the time of day at which the tests were completed, but showed no significant differences in other aspects of method practicality. Recruitment for consumer tests can sometimes be time-consuming and costly, especially when there are strict requirements regarding consumption frequency of the product tested. Therefore, the last phase of this study aimed at comparing sensory acceptability of two types of products: a familiar, generally liked and frequently consumed product (chocolate-chip cookies), and a niche, generally disliked and infrequently consumed product (black licorice), in subjects with different levels of consumption frequency. When subjects were grouped into frequent and infrequent consumers, a significant difference in liking means was observed for the niche product, but not for the familiar product. Cluster analysis supported this fact by showing that a segment of consumers rate samples differently based on consumption frequency in the niche product, but not in the familiar product. Analysis of variance showed that liking ratings remained similar as less frequent consumers were progressively added to the analysis in the cookie test, but differences were observed in the licorice ratings when infrequent consumers were added. Major differences in frequency of consumption were observed between subjects who indicated recognizing at least one licorice product and those who indicated they did not. In the cookie test, the distribution between those two groups was similar. Findings support the possibility of making screening criteria in consumer tests of familiar products more lax, as some subjects may not consume certain products frequently but may be qualified to indicate their sensory acceptance of them.

Graduation Semester

2022-08

Type of Resource

Thesis

Copyright and License Information

Copyright 2022 Marta Albiol Tapia

Owning Collections