Comparison of a descriptive analysis and instrumental measurements (electronic nose and electronic tongue) for the sensory profiling of Korean fermented soybean paste (doenjang)

Abstract

The objective of this study was to investigate the feasibility of using an electronic nose (e‐nose) and electronic tongue (e‐tongue) in the sensory profiling of doenjang, a fermented soybean paste. The descriptive analysis of doenjang by trained panelists was compared with that assessed using the e‐nose and e‐tongue. Eighteen aroma/taste/flavor attributes developed by trained panels from 14 doenjang samples were compared with the results from e‐nose and e‐tongue measurements. The e‐nose detected 11 volatile compounds, while the e‐tongue measured the intensities of 5 basic tastes. The correlation between the descriptive results and instrument measurements (e‐nose and e‐tongue) demonstrated that the e‐nose results were not correlated with most of the odors and flavors of doenjang, whereas the e‐tongue was effective in estimating taste intensities for sweetness (r = .798), umami (r = .821), and saltiness (r = .786). However, low correlations between the results from a trained panel and from the e‐tongue were found for sourness and bitterness. Practical applications The use of instruments to measure volatile compounds and taste attributes may not reflect the overall sensory characteristics of food products. The results of an e‐nose and e‐tongue test were not correlated with the sensory attributes identified by the trained panel. These findings may be due to interaction effects among volatile compounds and/or tastes and also might not fully consider the threshold of each volatile compound. Therefore, it is necessary to account for discrepancies between current technologies for mechanical sensory measurements and human perception studies in the evaluation of food products that have complex odor/tastes/flavors.

Identification of sensory characteristics that drive consumer preferences of commercially mass‐produced doenjang in Korea

“…A stronger brown color was significantly stronger in M7, and TS8–10, which had traditional doenjang characteristics ( p < .05); this is in accordance with the results of previous studies (Jung et al, ; Kim & Lee, ; Kim et al, ). It has been reported that doenjang manufactured through a modified method typically has a lighter color than TSD, mainly because modified methods involve the inoculation of A. oryzae , which results in a lighter color (Kim et al, ).…”

“…High intensities of alcohol aromatics have been noted in doenjang in two cases: (a) when alcohol is artificially added to doenjang during manufacturing to minimize the growth of undesirable microorganisms during storage and distribution, and (b) when alcohol naturally forms during doenjang fermentation, particularly in those doenjang samples with a high carbohydrate and low salt content (Chiou, Ferng, & Beuchat, ; Kim et al, ; Park, Lee, Kim, & Lee, ). The typical method of manufacturing commercial doenjang involves the addition of ethyl alcohol to increase the length of preservation; hence, many previous studies have reported a high level of alcohol aromatics in doenjang made using a modified method (Jung et al, ; Kim et al, ). In the present study, high intensities of alcohol aromatics were observed in TS8 seemed that the company added ethyl alcohol for preservation like modified method.…”

“…The eight panelists, who were all females aged 25–45 years, had intensive training from the Korea Food Research Institute (KFRI; Seongnam‐si, Korea) in assessing doenjang products prior to this study (Jung et al, ; Kim et al, ). Training for 10 commercial doenjang sample assessments was conducted in 30‐min sessions held twice a day for 8 days.…”

“…Training for 10 commercial doenjang sample assessments was conducted in 30‐min sessions held twice a day for 8 days. During the training sessions, the panelists reviewed the previously published sensory lexicon for describing doenjang (Jung et al, ), and if necessary, modified this with reference products. A total of 31 attributes for odor, taste, and flavor, as well as mouthfeel, were finalized: three appearance‐related, eight odor‐related, five taste‐related, five flavor‐related, five mouthfeel‐related, and five aftertaste‐related attributes.…”

“…The doenjang samples were prepared in the same manner as that used in previous studies (Jung et al, ; Kim et al, ). In brief, doenjang samples were prepared 1 day prior to training or evaluation, and stored at 4 ± 1°C in a refrigerator.…”