Archives of Design Research

Historically, the primary role of product packaging was protective, safeguarding its contents from potential harm with aesthetics being essentially an afterthought (Schifferstein et al., 2022). However, as the commercial environment evolved, so did the function of packaging. It began to play a critical role in product accessibility and logistical efficiency and had a greater influence over product image (Hurley et al., 2017). This shift caused packaging to not only protect products but also communicate brand messages to potential buyers (Yam et al., 2005; Orth & Malkewitz, 2008; Lydekaityte & Tambo, 2020).

Packaging transcended its essential protective function with changing market dynamics, becoming instrumental in establishing and differentiating brand identities (Chuang & Ou, 2001). Today, businesses recognize packaging’s ability to shape the consumer experience. Many consumers regard aesthetic appeal as equally important, if not more so, than the protective utility of packaging (Boz et al., 2020).

Among the various visual elements of packaging, such as structure, typography, and graphics, color has the greatest influence on consumer perceptions (Ares & Deliza, 2010; Spence, 2016; Mutsikiwa et al., 2013). Color often forms the basis for a consumer’s initial impression of a product (Kuo et al., 2021). Color has a strong influence over commerce (Labrecque & Milne, 2011; Krishna et al., 2017). Beyond attracting attention, color is intertwined with product narratives and brand values, communicating a brand’s identity and trustworthiness. In today’s competitive marketplace, where packaging aesthetics are pivotal to consumer decisions, visual elements have become integral to product marketing strategies. As a result, many brands integrate strategic design elements into their packaging that are designed to leave a lasting impact on consumers.

Packaging design shapes initial consumer interactions with the product and their subsequent perceptions, so color is not merely an aesthetic choice. It is a strategic tool that significantly influences product categorization, brand imagery, emotional connection with associated adjectives, perceived quality and value, and brand identity (Chuang & Ou, 2001; Garber & Hyatt, 2003; Bottomley & Doyle, 2006). Colors can elicit diverse emotional responses from consumers, from admiration to, at times, displeasure. The role of packaging color has a profound impact on brand communication and purchase decisions (Gabbas et al., 2021).

As a brand’s visual language evolves, the use of associated adjectives to evoke imagery becomes more pronounced. These associated adjectives enhance the emotional and symbolic nuances of colors, increasing the likelihood that the target audience will receive the intended message (Sutopo, 2023). When the association between a color and its associated adjectives is strong, consumers consistently interpret the meaning of packaging colors as intended (Mogaji, 2021). For instance, red evokes vigor, while blue suggests calm and reliability (Schwarz, 2022). Given that certain colors are linked with specific associated adjectives, such as red with passion and blue with trustworthiness, leveraging these colors empowers brands to communicate their core values more effectively, such as being innovative or reliable (Jin et al., 2019). However, these associated adjectives must be genuinely manifested in the products. Discrepancies can cloud consumer understanding, potentially diminishing the product’s appeal (Millis & King, 2001).

The importance of the relationship between colors and associated adjectives to branding cannot be understated. Packaging color palettes must be meticulously selected to reflect the brand’s essence and distinctive features and increase its emotional appeal. Color combinations are crucial ways that brands can create a unique identity in crowded marketplaces (Gabbas et al., 2021). To harness the power of the relationship between colors and associated adjectives, brands’ color palettes must be properly chosen to cause with associated adjectives, ensuring that consumer emotional reactions are powerful and targeted emotional reactions in consumers. While an exhaustive discussion on color combination dynamics, such as the merits of three-color combinations, is beyond the scope of this study, using the correct colors to communicate with associated adjectives remains is fundamental to visual storytelling and consumer engagement.

Colors are characterized by three foundational attributes: hue, value, and chroma (Gabbas et al., 2021; Pereira, 2021). Understanding these attributes allows for a more nuanced exploration of color combinations in design contexts. This study analyzes how certain color combinations evoke images through associated adjectives.

Color combinations can be used in design to evoke specific emotions and communicate brand narratives. Combining multiple colors changes consumers’ emotional responses from what they would have with individual colors (Chuang & Ou, 2001). For instance, while green is typically associated with eco-friendliness, its perception can be changed by changing the color’s value and chroma (Guo et al., 2020). Historically, individual colors have been tied to broad emotional responses and perceptions (Clarke & Costall, 2008; Hanada, 2018). However, juxtaposing two colors can heighten these emotional responses. For example, while red and black signify aggression, red and yellow signify warmth and red and blue can induce feelings of unrest (Elliot & Maier, 2012). Established associations, like the combination of green and yellow’s link with cleanliness, have been strategically employed in soap packaging (Bottomley & Doyle, 2006).

The interplay of three colors can be used to produce a wider variety of emotional responses than fewer colors. Three-color combinations can enrich the intricacy of emotional responses and better communicate visual narratives (Gupta, 2015). Combining three colors does not have a mere additive effect, instead producing different emotional responses than the individual colors would alone. Three-color combinations can have diverse cultural associations, making them exceptionally advantageous for cross-cultural branding efforts (Wang et al., 2022). Given these unique attributes, there is a compelling need to study further and understand how combinations of three colors work together. This versatility helps three-color combinations better communicate associated adjectives that are crucial parts of brand stories to diverse audiences. Three-color combinations can increase the efficacy of associative communication (Beaird et al., 2020).

Three-color combinations excel at evoking nuanced emotions and portraying intricate brand stories because of the vast number of possible combinations. Meticulously selected three-color combinations with synergy balance simplicity with the ability to communicate complex meaning, so emerging as a potent tool in visual communication. This study’s hypotheses were as follows:

Hypothesis 1. The hue, value, and chroma of laundry detergent packaging colors will have different effects on consumers’ perceptions of images evoked through associated adjectives.

Hypothesis 2. The container, label, and cap of laundry detergent packaging will have different effects on consumers’ perceptions of images evoked through associated adjectives.

Hypothesis 3. The value and chroma of laundry detergent packaging will have different effects on consumers’ perceptions of images evoked through associated adjectives when the hue is kept constant.

This study investigated laundry detergents because they are low-involvement products widely utilized by people from most demographics and purchase decisions about them are often influenced by scent or brand image. The laundry detergents selected for analysis were the 10 best-selling laundry detergents (Bigdata finance platform, n.d., retrieved 2022; Nielsen Korean Click, n.d., retrieved 2022).

Laundry detergent packaging designs were categorized and analyzed based on their color, label, graphics, and typography, which is the system used by Mutsikiwa & Marumbwa (2013) and Boz et al. (2020). This study examined packaging color due to its profound impact on consumers’ initial package perceptions, so instead of defining colors as primary, secondary, or accent, this study categorized them according to whether they were on the container, label, or cap. This categorization better reflected the characteristics of laundry detergent packaging. Images sourced from brand websites revealed distinct patterns in how packaging parts were colored. Containers and caps typically utilized a single color, while labels displayed a mix of colors primarily for branding and detailing. Although the labels featured many colors, colors from minor details or texts might not significantly impact the overall perception of the product. Therefore, the primary colors, specifically those that constituted more than 70% of the label’s colored area, were analyzed. The dominant colors in label design are crucial in shaping the overarching impression and brand image, making this approach essential for capturing core insights. Other visual elements were excluded from this study.

The colors’ CMYK and RGB values were determined by Korean Standard Color Analysis. The Wallkill color tool was then used to translate these values to the Munsell color system, which defines colors by their hue, value, and chroma. The hues that were analyzed were those of the Munsell system: white (N10), black (N0), red (R), yellow-red (YR), yellow (Y), green-yellow (GY), green (G), blue-green (BG), blue (B), purple-blue (PB), purple (P), and red-purple (RP). Colors occupying areas less than or equal to 5% of the container, label, and cap’s colored area were excluded from the analysis.

Examination of current laundry detergent packaging design trends revealed a spectrum of characteristics (Fig. 1 and 2). Laundry detergents have various attributes, such as emphasizing functional solid properties and being natural and eco-friendly. Various color combinations were used to incorporate different hues, values, chromas, and complementary colors. Colors were distributed from 10GY to 10PB, with B and PB hues being prevalent. Colors tended to have lower values and higher chromas, especially when the product emphasized its functionality. Detergents that utilized colors situated centrally on the color wheel, namely those between 5Y and 10G, predominantly communicated eco-friendliness. These colors tended to have higher values and lower chromas. YG and B hues used for functional products like soap communicate functionality (Bottomley & Doyle, 2006). There were also a variety of adjectives used on the packaging. Descriptors like “strong” and “fresh” were more common on products that communicated functionality. In contrast, adjectives associated with nature were more common than those that communicated eco-friendliness. This study was conducted on colors and their associated adjectives to better understand how brands can communicate through their packaging.

Figure 1 Analysis of visual elements of detergents packaging design

Figure 2 Color distribution of detergents: Hue, Value, Chroma

The analyzed packaging of current laundry detergent brands revealed 38 distinct adjectives prominent in advertising and promotions. These adjectives, often featured in taglines or as integral parts of brand names, highlight their consistent use and significance in the industry. Given their current synergy in conveying brand messages, we utilized these adjectives to understand the images they evoke when paired with certain three-color combinations. Among these adjectives, only those frequently used and consistently appearing across top brands were shortlisted for our experiment. While elements like typography and design structure influence the consumer experience, our study focuses on color’s role in communicating these images.

The adjectives, related to the study’s theme, were refined to ensure their semantic accuracy and relevance. Particular attention was paid to ensuring that English translations retained the nuances of their original Korean meanings. Those not aligned with our research objectives, such as “suitable” and “appropriate,” were excluded. A professor specializing in Korean language and literature was consulted to finally select the adjectives that were contextually relevant to the Korean market. The final list of adjectives is “strong,” “mild,” “safe,” “fresh,” “fragrant,” “natural,” and “clean.”

Survey stimuli were developed by applying three-color combinations to laundry detergent packaging designs with one color each for the container, label, and cap. Rather than using standard color swatches, a tangible representation was used that was similar to actual product designs. Single colors were used to control for other influences, so elements that might affect the colors, like shadows, were removed. To focus solely on color, typography and layout were excluded from the stimuli.

The stimuli were created using the 11 Munsell color system hues (Table 1). A total of 30 stimuli were created with the same shape and size but different colors (Fig. 3).

Figure 3 An example of Stimuli

The survey was conducted online in a controlled setting with participants from various age groups. Participants observed 30 stimuli individually and were asked to rate how vividly the associated adjectives evoked specific images for them using a 7-point Likert scale where 1 means “strongly disagree” and 7 means “strongly agree.” A total of 173 respondents participated in the survey (Table 2). The resulting data was analyzed using IBM SPSS Statistics 26.0 software. Within our analysis framework, we deconstructed color combinations into hue, value, and chroma and statistically determined which combinations best evoked specific adjectives. This methodology was designed to deeply understand the influence of individual color components on the imagery and associations. The reliability analysis showed a Cronbach’s alpha value of .780, indicating satisfactory internal consistency.

Chi-squared tests were conducted to determine whether the strength of consumer perceptions of associated adjectives varied by hue. The responses were categorized as follows: 1–3 was recoded to 1, 4 was recoded to 2, and 5–7 was recoded to 3, representing weak, neutral, and strong perceptions.

Strong

Analysis of the “strong” in relation to the hues of the container, label, and cap revealed significant associations for all hues (containers: Χ² = 448.014, df = 20, p < .000, λ = .111; labels: Χ² = 879.326, df = 18, p < .000, λ = .207; caps: Χ² = 348.198, df = 16, p < .000, λ = .095) (Tables 3–5). Container hues of YR (86.7%), R (78.0%), and Y (57.2%); label hues of N0 (81.9%), YR (78.6%), R (65.3%), and Y (60.4%); and cap hues of R (60.1%), N0 (55.9%), and RP (55.9%) were most strongly associated with the “strong” image.

The hues of the labels (λ = .207) were most likely to produce a “strong” image, followed in descending order of influence by those of the containers (λ = .111) and caps (λ = .095).

Mild

Analysis of the “mild” in relation to the hues of the container, label, and cap revealed significant associations for all hues (containers: Χ² = 696.618, df = 20, p < .000, λ = .152; labels: Χ² = 937.291, df = 18, p < .000, λ = .206; caps: Χ² = 643.213, df = 16, p < .000, λ = .125) (Tables 6–8). Container hues GY (72.3%), B (51.4%), and N10 (51.4%); label hues G (75.7%), GY (61.6%), and B (56.6%); and cap hues GY (85.5%) and N10 (54.2%) were most strongly associated with the “mild” image.

The hues of the labels (λ = .206) were most likely to produce a “mild” image, followed in descending order of influence by those of the containers (λ = .152) and caps (λ = .125).

Safe

Analysis of the “safe” in relation to the hues of the container, label, and cap revealed significant associations for all hues (containers: Χ² = 429.374, df = 20, p < .000, λ = .170; labels: Χ² = 549.192, df = 18, p < .000, λ = .213; caps: Χ² = 399.669, df = 16, p < .000, λ = .061) (Tables 9–11). Container hues GY (71.1%), N10 (53.3%), and BG (53.2%); label hues GY (64.2%), YR (52.3%), and N10 (50.1%); and cap hues RP (59.5%) and GY (56.3%) were most strongly associated with the “safe” image.

The hues of the labels (λ = .213) most strongly communicated the “safe” image, followed in descending order of influence by those of the containers (λ = .170) and caps (λ = .061).

Fragrant

Analysis of the “fragrant” in relation to the hues of the container, label, and cap revealed significant associations for all hues (containers: Χ² = 538.972, df = 20, p < .000, λ = .096; labels: Χ² = 628.571, df = 18, p < .000, λ = .104; caps: Χ² = 399.669, df = 16, p < .000, λ = .061) (Tables 12–14). Container hues PB (55.5%), GY (54.9%), and N10 (47.8%); label hues G (61.8%), GY (52.0%), and YR (48.6%); and cap hues RP (69.4%) and GY (58.4%) were most strongly associated with the “fragrant” image.

The hues of the labels (λ = .104) most strongly communicated the “fragrant” image, followed in descending order of influence by those of the containers (λ = .096) and caps (λ = .061).

Fresh

Analysis of the “fresh” in relation to the hues of the container, label, and cap revealed significant associations for all hues (containers: Χ² = 829.623, df = 20, p < .000, λ = .222; labels: Χ² = 900.678, df = 18, p < .000, λ = .239; caps: Χ² = 632.390, df = 16, p < .000, λ = .166) (Tables 15–17). Container hues B (59.2%), GY (57.8%), BG (55.5%), and N10 (52.3%); label hues PB (57.2%), B (56.6%), YR (48.8%), and N10 (45.9%); and cap hues GY (70.5%) and RP (63.0%) were most strongly associated with the “fresh” image.

The hues of the labels (λ = .239) most strongly communicated the “fresh” image, followed in descending order of influence by those of the containers (λ = .222) and caps (λ = .166).

Natural

Analysis of the “natural” in relation to the hues of the container, label, and cap revealed significant associations for all hues (containers: Χ² = 447.354, df = 20, p < .000, λ = .035; labels: Χ² = 601.819, df = 18, p < .000, λ = .099; caps: Χ² = 404.172, df = 16, p < .000, λ = .026) (Tables 18–20). Container hues GY (68.2%) and N10 (44.3%); label hues G (63.6%), GY (60.4%), and B (52.6%); and cap hues GY (79.8%), G (48.6%), and RP (42.8%) were most strongly associated with the “natural” image.

The hues of the labels (λ = .099) most strongly communicated the “natural” image, followed in descending order of influence by those of the containers (λ = .035) and caps (λ = .026).

Clean

Analysis of the “clean” in relation to the hues of the container, label, and cap revealed significant associations for all hues (containers: Χ² = 867.456, df = 20, p < .000, λ = .215; labels: Χ² = 772.814, df = 18, p < .000, λ = .234; caps: Χ² = 722.667, df = 16, p < .000, λ = .183) (Tables 21^-23). Container hues GY (68.2%), B (57.1%), BG (53.2%), and N10 (45.0%); label hues B (71.1%), N10 (68.0%), YR (59.9%), and PB (56.6%); and cap hues GY (87.9%), G (64.4%), and RP (55.1%) were most strongly associated with the “clean” image.

The hues of the labels (λ = .234) most strongly communicated the “clean” image, followed in descending order of influence by those of the containers (λ = .215) and caps (λ = .183).

Multiple regression analysis was conducted to determine whether the strength of consumer perceptions of associated adjectives varied by value.

Strong

Analysis on the influence of the values of the container, label, and cap on the “strong” revealed an overall statistical significance (F = 371.897, df = 3, p < .05) (Table 24). Container (t = -21.217, p < .05), label (t = -27.205, p < .05), and cap value (t = -4.695, p < .05) were all individually statistically significantly correlated with the “strong” image.

The values of the labels (β = -.359) were most likely to produce a “strong” image, followed in order of decreasing strength by those of the containers (β = -.276) and caps (β = -.061). Specifically, lower the values of the container, the label, and the cap enhances the perception of “strong” image.

Mild

Analysis on the influence of the values of the container, label, and cap on the “mild” revealed an overall statistical significance (F = 444.368, df = 3, p < .05) (Table 25). Container (t = 23.940, p < .05), label (t = 27.758, p < .05), and cap values (t = 7.868, p < .05) were all individually statistically significantly correlated with the “mild” image.

The values of the labels (β = .360) were most likely to produce a “mild” image, followed in order of decreasing strength by those of the containers (β = .306) and caps (β = .009). Specifically, higher the values of the containers, the labels, and the caps enhances the perception of “mild” image.

Safe

Analysis on the influence of the values of the container, label, and cap on the “safe” revealed an overall statistical significance (F = 210.761, df = 3, p < .05) (Table 26). Container (t = 15.686, p < .05), label (t = 18.535, p < .05), and cap values (t = 7.187, p < .05) were all individually statistically significantly correlated with the “safe” image.

The values of the labels (β = .255) most strongly communicated the “safe” image, followed in order of decreasing strength by those of the containers (β = .212) and caps (β = .097). Specifically, higher the values of the containers, the labels, and the caps enhances the perception of “safe” image.

Fragrant

Analysis on the influence of the values of the container, label, and cap on the “fragrant” revealed an overall statistical significance (F = 234.144, df = 3, p < .05) (Table 27). Container (t = 15.019, p < .05), label (t = 21.969, p < .05), and cap values (t = 5.087, p < .05) were all individually statistically significantly correlated with the “fragrant” image.

The values of the labels (β = .300) most strongly communicated the “fragrant” image, followed in order of decreasing strength by those of the containers (β = .202) and caps (β = .068). Specifically, higher the values of the containers, the labels, and the caps enhances the perception of “fragrant” image.

Fresh

Analysis on the influence of the values of the container, label, and cap on the “fresh” revealed an overall statistical (F = 250.890, df = 3, p < .05) (Table 28). Container (t = 16.669, p < .05), label (t = 20.554, p < .05), and cap values (t = 7.836, p < .05) were all individually statistically significantly correlated with the “fresh” image.

The values of the labels (β = .280) most strongly communicated the “fresh” image, followed in order of decreasing strength by those of the containers (β = .223) and caps (β = .104). Specifically, higher the values of the containers, the labels, and the caps enhances the perception of “fresh” image.

Natural

Analysis on the influence of the values of the container, label, and cap on the “natural” revealed an overall statistical significance (F = 309.504, df = 3, p < .05) (Table 29). Container (t = 19.249, p < .05), label (t = 24.617, p < .05), and cap values (t = 4.840, p < .05) were all individually statistically significantly correlated with the “natural” image.

The values of the labels (β = .330) most strongly communicated the “natural” image, followed in order of decreasing strength by those of the containers (β = .254) and caps (β = .064). Specifically, higher the values of the containers, the labels, and the caps enhances the perception of “natural” image.

Clean

Analysis on the influence of the values of the container, label, and cap on the “clean” revealed an overall statistical significance (F = 236.584, df = 3, p < .05) (Table 30). Container (t = 16.694, p < .05), label (t = 20.080, p < .05), and cap values (t = 6.907, p < .05) were all individually statistically significantly correlated with the “clean” image.

The values of the labels (β = .274) most strongly communicated the “clean” image, followed in order of decreasing strength by those of the containers (β = .224) and caps (β = .092). Specifically, higher the values of the containers, the labels, and the caps enhances the perception of “clean” image.

Multiple regression analysis was conducted to determine whether the strength of consumer perceptions of associated adjectives varied by chroma.

Strong

Analysis on the influence of the chromas of the container, label, and cap on the “strong” revealed an overall statistical significance (F = 139.129, df = 3, p < .05) (Table 31). Container chroma (t = 19.028, p < .05) was individually statistically significantly correlated with “strong” image while label (t = 1.535, p > .05) and cap chroma (t = .806, p > .05) were not.

The chromas of the containers, labels, and caps were analyzed for their influence on the perception of the “strong” image. It was found that the container’s chroma (β=.285) has the most pronounced impact. Specifically, a higher the chromas of the container enhances the perception of “strong” image.

Mild

Analysis on the influence of the chromas of the container, label, and cap on the “mild” revealed an overall statistical significance (F = 165.208, df = 3, p < .05) (Table 32). Container (t = -18.529, p < .05), label (t = -2.146, p < .05), and cap color (t = 6.147, p < .05) were all individually statistically significantly correlated with the “mild” image.

The chromas of the containers (β = -.276) were most likely to produce a “mild” image, followed in order of decreasing strength by those of the labels (β = -.033) and caps (β = .087). Specifically, lower the chromas of the container, the label, and the cap enhances the perception of “mild” image.

Safe

Analysis on the influence of the chromas of the container, label, and cap on the “safe” revealed an overall statistical significance (F = 73.581, df = 3, p < .05) (Table 33). Container (t = -9.026, p < .05) and cap chroma (t = 7.689, p < .05) were individually statistically significantly correlated with the “safe” image while label chroma was not (t = .438, p > .05).

The chromas of the containers (β = -.138) most strongly communicated the “safe” image, followed in order of decreasing strength by those of the caps (β = .112). Specifically, lower the chromas of the container combined with higher the chromas of cap enhances the perception of “safe” image.

Fragrant

Analysis on the influence of the chromas of the container, label, and cap on the “fragrant” revealed an overall statistical significance (F = 148.046, df = 3, p < .05) (Table 34). Container (t = -13.108, p < .05) and cap chromas (t = 11.145, p < .05) were individually statistically significantly correlated with the “fragrant” image while label chroma was not (t = -.204, p > .05).

The chromas of the containers (β = -.196) most strongly communicated the “fragrant” image, followed in order of decreasing strength by those of the caps (β = -.159). Specifically, lower the chromas of the container combined with higher the chromas of the label and the cap enhances the perception of “fragrant” image.

Fresh

Analysis on the influence of the chromas of the container, label, and cap on the “fresh” revealed an overall statistical significance (F = 164.620, df = 3, p < .05) (Table 35). Container (t = -9.896, p < .05), label (t = 3.193, p < .05), and cap chromas (t = 13.344, p < .05) were all individually statistically significantly correlated with the “fresh” image.

The chromas of the containers (β = -.147) most strongly communicated the “fresh” image, followed in order of decreasing strength by those of the labels (β = .049) and the caps (β = .190). Specifically, a lower chroma of the container combined with higher chromas of the label and cap enhances the perception of the “fresh” image.

Natural

Analysis on the influence of the chromas of the container, label, and cap on the “natural” revealed an overall statistical significance (F = 119.949, df = 3, p < .05) (Table 36). Container (t = -17.087, p < .05), label (t = -2.33, p < .05), and cap chromas (t = 2.640, p < .05) were all individually statistically significantly correlated with the “natural” image.

The chromas of the containers (β = -.257) most strongly communicated the “natural” image, followed in order of decreasing strength by those of the labels (β = -.036) and the caps (β =.038). Specifically, lower the chromas of the container, and the label combined with higher the chromas of the cap enhances the perception of “natural” image.

Clean

Analysis on the influence of the chromas of the container, label, and cap on the “clean” revealed an overall statistical significance (F = 153.034, df = 3, p < .05) (Table 37). Container (t = -9.529, p < .05), label (t = 2.597, p < .05), and cap chromas (t = 13.244, p < .05) were all individually statistically significantly correlated with the “clean” image.

The chromas of the containers (β = -.142) most strongly communicated the “clean” image, followed in order of decreasing strength by those of the labels (β = .040) and the caps (β = -.189). Specifically, lower the chromas of the container and the cap combined with higher the chromas of the label enhances the perception of “clean” image.

Sections 4.1–4.3 described how color hues, values, and chromas affected the strength of consumers’ perceptions of associated adjectives. To increase this study’s practical relevance and applicability, the results are organized in this paper by associated adjectives (Table 38). For hue and value, associated adjectives are best communicated through the label, followed by the container and then the cap, but for chroma, it is the container, then the label, and then the cap (Table 39).

The analysis by color combination showed that “strong” image was created with values of N0, R, Y, and YR; containers with values and higher chromas of R, YR, and Y; and caps with lower values of R, N0, and RP. “Mild” image was created with labels with higher values and lower chromas of G, GY, and B; containers with higher values and chromas of GY, N10, and B; and caps with higher values and chromas of GY and N10. “Safe” image was created with labels with higher values of GY, YR, and N10; containers of higher values and lower chromas of GY, N10, or BG; and caps with higher values and chromas of RP and GY.

“Fragrant” image was created with labels with higher values of G, RP, GY, and YR; containers with higher values and lower chromas of PB, GY, and N10; and caps with higher values and chromas of RP and GY. “Fresh” image was occurred with labels with higher values and chromas of PB, B, YR, and N10; containers with higher values and lower chromas of B, GY, and BG; and caps with higher values and chromas of GY and RP. “Natural” image was occurred with labels with higher values and lower chromas of GY and N10; containers with higher values and lower chromas of G, GY, and B; and caps with higher values and chromas of GY, G, and RP. “Clean” image was created with labels with higher values and chromas of B, N10, YR, and PB; containers with higher values and lower chromas of GY, BG, B, and N10; and caps with higher values and lower chromas of GY, G, and RP.

A comparative analysis was conducted to determine whether the image conveyed by associated adjectives differed based on occupational background (Table 40). Participants with a design background were more attuned to projected brand image with heightened sensitivity to the “mild” (3.78), “safe” (4.29), “natural” (3.55), and “fresh” (3.91) image that differentiate a product from its competitors. Those without a design background had heightened sensitivity to functional image, particularly “fragrant” (3.76) and “clean” (4.48). Except for the “strong” image, all of these differences were statistically significant (p < .001).