*Corresponding Author GAINING INSIGHT INTO CONSUMERS' CHOICE OF PREFERRED WORDS Lea Cok* Faculty of Advanced Social Studies, Slovenia lea.cok@gmail.com Jani Ivan T oroš GEA College – Faculty of Entrepreneurship, Slovenia jani.ivan.toros@gea-college.si Jan Nalesnik GEA College – Faculty of Entrepreneurship, Slovenia jan.nalesnik@gea-college.si Abstract This research utilizes the MaxDiff method to categorize associations with the term trust and evaluate its effectiveness in advertising messages. Statistical analysis reveals significant differences among various word linked to term trust. Adjectives like reliable, loyal, and caring emerge as the most prevalent, reflecting respondent preferences. Furthermore, 80.8% of respondents prefer adjectives over nouns in associating with trust. By employing the MaxDiff methodology, this study offers valuable insights into nuanced perceptions of trust in advertising, facilitating the development of more impactful messaging strategies. Key Words Trust; advertising credibility; preferences; adjectives; nouns. Advances in Business-Related Scientific Research Journal, Volume 15, No. 1, 2024 104 INTRODUCTION Advertising, as defined by Kotler & Armstrong (2007), is a compensated form of displaying and advancing ideas, products, or services by a known benefactor. The authors argue that the success of advertising is influenced by various factors, including advertising objectives (informing, persuading, or reminder advertising), advertising budget, and advertising strategy. The latter comprises two major elements: crafting advertising messages and choosing advertising media. This article specifically focuses on persuasive advertising and the element of crafting advertising messages. It emphasizes the choice of words and their categorization, aiming to improve the effectiveness of advertising messages and reduce the negative impact of deceptive advertising. The occurrence of distrust in promotional content is increasing and gaining prevalence, which harms not only retailers and suppliers but also their customers. We maintain that numerous goods and offerings are capable of enriching people's way of living. Through this paper, we examine the occurrence of consumer distrust and its impact on consumer habits, exploring consumer preference words using the MaxDiff method. As mentioned, the effectiveness of communication between advertisers and their recipients (consumers) depends on many factors; however, this study focuses on the word choice approach. We are interested in determining whether there is a statistical difference among specific words that can be linked to the term trust. According to Emodi (2011), advertising is an important element of communication, with the crucial aspect being the attainment of understanding of the meaning. Agarwal & Agarwal (2023) pointed out in their article that "advertisements are designed to convey messages and create associations between products, brands, and desired meanings." Through the use of language, visual elements, and symbols, advertisers strategically manipulate to evoke certain emotions, values, and desires in consumers. Semantics plays a key role in advertising as it is responsible for conveying meaning and creating impactful messages. It helps advertisers communicate effectively with their target audience and influence their perceptions, attitudes, and behavior. Semantics is the study of the literal meaning conveyed by words and phrases. It links 'sign' with the 'signifier' (=differential mark) and the 'signified'. The signifier represents the tangible expression of a sign, like a word, sound, or image, whereas the signified refers to the idea or meaning linked to the signifier (Chandler, 1994). Simply put, it makes sentences meaningful in language by determining the object or idea that each word represents (Agarwal and Agarwal, 2023). Vasiloaia (2020) emphasizes in her article the importance that the advertiser, even if the meaning is incredibly ambiguous, incorporates their background and a range of implied meanings and associations into the work. When faced with pictures and words, our minds begin to connect all the signifiers within the construct with our personal experiences, knowledge, emotions, and feelings. Therefore, even if the advertiser of such a construct Advances in Business-Related Scientific Research Journal, Volume 15, No. 1, 2024 105 never intended it to have any meaning, it is practically impossible to exclude any associations for the viewer. Cortés de los Ríos (2002) discusses a semiotic advertising approach in which advertisers amplify the meaning of a concept by incorporating another. Metaphorical language in advertising creates images and associations to stimulate cognitive processes. Advertisers strive to find the most impactful metaphor, i.e., a term that resonates most strongly with the promoted company, considering the reader's physical and cultural background. Rajaobelina et al. (2019) note the influence of relational advertising as a factor in attitude formation in experiential advertising. They term this phenomenon as value-expression appeal, as suggested by Johar and Sirgy (1991). This principle revolves around how the portrayal of the product or brand user influences consumers' identification with their self-image, encompassing actual, ideal, social, and ideal social self-images. Considering the theory presented by Johar and Sirgy, we can agree that the relationship between an advertisement and a consumer is established when the consumer identifies with the message. Could we say that identification can also be reflected in the choice and perception of words and how this is reflected in consumer behavior? Consumer behavior is both reasonable and unreasonable. The neoclassical economic theory perceives the consumer as a reasonable thinker whose actions are logical given the constraints of limited information and cognitive capacity. As a result, consumers tend to narrow down their choices to a select few prominent alternatives and evaluate them using a limited set of criteria (Schmid, 2004). The decision-making solution meets the predetermined goal for the intended purpose. However, this raises the question of whether the decision will ultimately satisfy the consumer. The paradox of choice is a concept thoroughly researched by Schwartz (2004), where individuals feel overwhelmed by an abundance of choices, leading them to analyze and compare various options due to the fear of making the wrong decision. Because of this overwhelming occurrence, sometimes they do not decide at all. Possibly, this could be indicative of the necessity for rapid decision-making, as we face thousands of decisions daily, potentially exhausting our cognitive resources to process the most optimal choices, which might also involve irrational factors. The most recent economic and marketing literature particularly emphasizes the experiential selection of services and products, advocating for the utilization of intangible aspects associated with the emotional value perceived by customers in experiential marketing (Gentile et al., 2007). For example, Rajaobelina et al. (2019) examined the effect of heightened credibility within a banking environment through experiential advertising, which is shaped by five aspects: cognitive, emotional, sensory, relational, and behavioral. The goal is to assess whether enhancing banks' credibility via an experiential advertising approach can effectively restore consumer trust. The suggested conceptual framework comprising five aspects of experiential advertising indicates a favorable influence on advertising credibility. Findings suggest that the configuration of experiential advertising corresponds to instability in the credibility of advertising messages. Advances in Business-Related Scientific Research Journal, Volume 15, No. 1, 2024 106 In addition to disseminating information about their offerings, advertising also demonstrates an appreciation for potential customers, who are more likely to resonate with advertising messages if they evoke emotional responses. Are there specific words that can impact consumer behavior? We are curious to know if there are specific words related to trust that Slovenian consumers are more responsive to in advertising messages and what those could be. At this point, we should mention prototype categorization, which is a kind of combination of the theory of categorization founded by Eleanor Rosch in the 1980s and Lakoff's cognitive semantics (1988). Ahmed (2020), in his article, described two basic principles of prototype theory that guide the development of categories: (1) the principle of cognitive economy and (2) the principle of the perceptual structure of the world. These two principles combine to create the human categorization system (Ahmed, 2020). Interestingly, prototype theory has also been extended to the study of the meaning of many words by describing them as a complete set of characteristics (attributes). None of these characteristics is individually necessary or sufficient to determine the meaning of a word. Individual words in a language share certain common characteristics or attributes, but none of them is necessarily present in all cases for a word to belong to a particular category or have a particular meaning. A concept is unified by what certain philosophers refer to as a family resemblance structure (Ahmed, 2020). It is thus a term often used in the context of prototypical semantics, which emphasizes that members of a category are related to each other by shared characteristics, much as members of a family share relatedness but not necessarily identity. Lee and Potter (2020) investigated how listeners respond to negative, neutral, and positive words in audio advertising. In addition to measuring intervals between heartbeats, they used facial electromyography (EMG) as a physiological indicator of emotional response to gauge how words impact the dynamic interaction between the listener and the message. They recommended advertisers choose words that align with the emotional tone of the message, as they concurrently explored discrepancies between words and messages, where negative or neutral words appeared in messages of a positive nature. Individuals were less successful in recognizing target words, indicating the influence of emotional incongruity between the text and the overall tone of the message on the ability to remember and recognize words. This suggests that emotionally charged words, contrary to the general tone, disrupt cognitive processing and reduce memory efficiency. The brain attempts to adapt by limiting its processing to avoid conflicting or contradictory information. This response may act as a way to reduce discomfort caused by dissonant or conflicting information in the message. Do our brains respond similarly to deceptive messages? RESEARCH PROBLEM Advances in Business-Related Scientific Research Journal, Volume 15, No. 1, 2024 107 Darke & Ritchie (2007) pointed out in their research that deceptive advertising leads to a loss of trust, which subsequently impacts individuals' receptiveness to future advertising efforts from the same source as well as from other sources. Initial deceitfulness fosters a negative perception of advertising and marketing in general, thus eroding the credibility of subsequent advertising endeavors (Darke and Ritchie, 2007). Similarly, Newell et al. (2015) validated their hypotheses in their research, revealing that increased perceptions of deceitfulness are linked with diminished perceptions of a company's credibility. Respondents in the research showed less favorable reactions to the advertisement, held less positive opinions about the promoted brand, and consequently, expressed reduced intentions to buy the promoted product when suspicion of deception arose. Furthermore, the research highlighted that the perception of deception is sufficient to trigger negative emotions toward the advertisement, regardless of its actual deceptive nature. As a result, marketing and advertising managers should proceed with caution when crafting campaigns (Newell et al., 2015). Another aspect highlighted in this research is the acknowledgment of deceptive advertising, as outlined in Article 38 of the Consumer Protection Act, where greater emphasis is placed on the fact that deception can create economic insecurity among competitors, with consumers being almost unmentioned. They are mentioned in Article 37, which defines indecent advertising: 'Indecent advertising of goods, services, or digital content means advertising that contains elements that are offensive or could be offensive to consumers, readers, listeners, and viewers, or elements that contradict morality.' (Consumer Protection Act, 2022). Regarding this subject, we encountered an article where Xie et al. (2015) conducted two experiments aimed at distinguishing the impact of harm on consumer attitudes towards the brand and purchase intent in the case of perceived deception. The study also delved into how anticipated harm could affect a consumer's reaction to the 'diagnosis of perceived deception'. This pertains to the significance of potential deception as consumers assess advertised products and brands. The article highlights that the impact of the extent of anticipated harm resulting from deception has received less attention compared to earlier research in behavioral advertising, which primarily examined how consumers could be misled by the assertions and implications of advertising messages. The findings suggest that the significance of perceived deception can impact anticipated harm. In other words, the adverse impact of perceived deception may escalate when the expected outcomes of deception are consequences of deception are viewed as more serious. Jakomin et al. (2022) conducted a study examining how retailers promote environmentally friendly packaging for green products offered in online stores using the MaxDiff method. According to the findings, less than a third of participants express trust in manufacturer-provided information regarding green products, ene though that two-thirds of respondents are willing to pay a premium for green products. Is this outcome solely attributed to insufficient Advances in Business-Related Scientific Research Journal, Volume 15, No. 1, 2024 108 advertising of green products in online stores, or does it also reflect broader skepticism towards advertising messages in general? Interestingly, in the study by Pollay and Mittal (1993), a similar proportion of respondents (38%) rated advertising messages as truthful and informative. Are there specific language choices that could enhance the credibility of advertising messages, particularly in situations where there is no intention to deceive and the sole aim is to promote products or services? While numerous factors contribute to credibility, could we potentially mitigate the prevailing resistance towards advertising messages by catering to consumer preferences in terms of word choice? To explore the process of transitioning from distrust to trust, it is imperative to establish a comprehensive understanding of the underlying concept of trust itself. According to the Cambridge Dictionary online, the term trust is defined as "to believe that someone is good and honest and will not harm you, or that something is safe and reliable" (Trust, n.d.). Gambetta (2000) claims that trust is one of those states that cannot be intentionally triggered, either concerning oneself or others. Indeed, trust cannot be logically or intentionally cultivated, as the act of trying to do so undermines the very state one seeks to establish. Trust often emerges as a natural consequence, usually stemming from familiarity and friendship, both of which entail a degree of mutual understanding and concern for each other's well-being. This suggests that trust may arise as a by-product of moral and religious principles advocating honesty and mutual affection (Gambetta, 2000). Even though trust cannot be intentionally triggered, it may be understood by exploring words linked with it. METHODS Maximum Difference Scaling, also referred to as "best-worst scaling," is a research technique utilized to assess the relative preference variances among various elements or attributes—in our study, words—to ascertain which words respondents primarily associate with the concept of trust. MaxDiff can significantly contribute to examining purchasing motivations, identifying desired benefits, and evaluating the responsiveness of benefits to marketing strategies, particularly in scenarios where individuals within preexisting groups exhibit similarities that make it difficult to differentiate from one another (Cohen, 2003). The MaxDiff model functions under the premise that respondents evaluate each possible pair within a given subset and choose the pair with the most significant difference as the best-worst or maximum difference pair. Consequently, MaxDiff can be viewed as a more efficient method for collecting paired comparison data. We decided to choose 6 adjectives and 6 nouns. The selected adjectives were synonyms of the term trust (WordNet 2.1 database), while the nouns were chosen based on the conceptual model of the five dimensions by Rajaobelina et al. (2019), who explored how these dimensions of experiential advertising influence credibility (cognitive, emotional, sensory, Advances in Business-Related Scientific Research Journal, Volume 15, No. 1, 2024 109 relational, and behavioral). We anticipated that most respondents could relate to body parts. Additionally, Cortés de los Ríos (2002) suggests that advertisers should consider the physical and cultural backgrounds of their audience when designing advertisements. We focused on body parts and their functions in the choice of nouns: pulse (heart), vision (eyes), sense (intellect), touch (hand), step (leg), and inspiration (lungs), which also takes into account the structure of family resemblance (Ahmed, 2020). Moreover, Pulvermüller, Härle, and Hummel (2000, 2001) measured in their articles the EEG brain response to verbs referring to actions performed by different body parts. Not only were different areas in the cerebral cortex activated—for example, the motor area was activated in the case of the verb "to walk," while the perisylvian area was activated for the verb "to talk" (Pulvermüller, Härle, and Hummel, 2000)—but also the processing speed of such verbs varied (Pulvermüller, Härle, and Hummel, 2001). Although the described studies measured brain responses to verbs, we considered it crucial to choose nouns, as we were interested in which of them were most linked with the term trust. Ebaid (2018) delves into the realm of adjectives, asserting that they are essential components of linguistic constructs, primarily serving to characterize, depict, and alter nouns or pronouns. His study focuses on the prevalent and pervasive utilization of adjectives, highlighting their potency as persuasive devices, sometimes capable of standing alone without accompanying nouns. For the choice of adjectives, we used synonyms of the term trust: reliable (zanesljiv), genuine (pristen), propulsive (prodoren), consistent (dosleden), loyal (zvest), caring (skrben). We are interested in whether adjectives will be chosen more often than nouns. Please note that these are translations, which were the closest approximations of the semantic content in our estimation. For example, we were not measuring the word authentic (slo. avtentičen), but the Slovenian word pristen is so authentic that it has no literal translation in English. "Real" might be the only substitution; however, the Slovenian translation for pravi = real, thus we choose authentic = pristen. Table 1: Distribution of adjectives and nouns & their translation Adjetives Nouns Slovenian English Slovenian English zanesljiv reliable utrip pulse pristen authentic vizija vision prodoren persuasive razum intellect dosleden consistent prijem grip zvest loyal korak step skrben caring inspiracija inspiration Advances in Business-Related Scientific Research Journal, Volume 15, No. 1, 2024 110 RESEARCH QUESTIONS AND HYPOTHESIS The article by Prihatini (2020) focuses on describing word associations and their role in understanding language constituents in language learning. The author claims that word association presents a correlation between a word and other words stemming from semantic relationships, prompting our first research question: Are some words more frequently linked with the term trust by respondents? (RQ1). Even though links are unique semantic connections made by individuals in hypothesis 1, we propose that there is a statistical significance between words linked with the term trust, addressing RQ1. As we mentioned, the study by Ebaid (2018) is focused on emphasizing the features and meanings associated with products. In his article, Ebaid (2018) highlights that adjectives are sometimes used independently to name products, although adjectives typically accompany nouns. Based on the findings, a second research question could be formulated as follows: In the set of word links, is the selection of adjectives for the term trust more common? (RQ2). Taking into account the result of the mentioned research provides us with a basis for the second hypothesis: Adjectives are statistically significantly more appropriate than nouns. Respondents indicate their most and least preferred links with the term trust in the questionnaire, with all 12 words (6 adjectives and nouns) represented with equal frequency. DATA ANALYSIS The survey was generated utilizing Sawtooth Software SSI Web (Sawtooth Software, Inc., USA) and processed with the MaxDiff module. It consisted of two parts. In the first part, respondents selected words that are linked with the term trust. In the second part, they first indicated their gender, followed by their age group (up to 20 years, 20 to 40 years, over 40 years). Utilizing multiple items for scaling facilitates researchers, especially those with limited statistical expertise, to conduct sophisticated analyses. The item estimates are straightforward to interpret, as they are placed on a comprehensive scale ranging from 0 to 100 points, totalling 100" (Jakomin et al., 2022). The results derived from the "Analysis/Max Diff Scores" function in the Sawtooth Software program were delivered in both raw and rescaled formats. These scores are direct outcomes of the Hierarchical Bayes analysis. The software then presents the data in an Excel table, including counts and scores. The software also provides a 95% confidence interval, indicating the level of certainty with which specific estimates for the displayed parameters are provided. This implies that if the experiment were replicated multiple times with new random samples on each occasion, the true population average would be within the calculated confidence interval in 95% of the experiments. Advances in Business-Related Scientific Research Journal, Volume 15, No. 1, 2024 111 In simpler words, we can have 95% confidence that the genuine population average lies within the 95% confidence interval. To test H1, we compared the means between the most and least frequently selected adjectives using a Paired Samples t-test. This test evaluates the means of two measurements obtained from the same group, offering insights into whether there exists a significant statistical difference between the two datasets. It produces a p-value, with a value below 0.05 suggesting a less than 5% probability that the outcome is random. The t-test is employed to compare the means of the two groups. To test H2, we compared the means with a Paired Samples t-test of two related sample groups: adjectives and nouns. Data Retrieval: Commands first read data from the scores in the Excel file generated by the software, with the first row containing column names. We set a minimum valid value percentage at 95%. T-Test Calculation: We then conducted a t-test comparing the means between the: most and least frequently chosen adjectives, between the most frequently chosen adjective and noun, and the means of adjectives and nouns. The t-test determines whether there are statistically significant differences between the means of these two groups, with significance determined at a 95% confidence level. Handling Missing Data: Instructions also include the use of the /MISSING=ANALYSIS option, indicating that the t-test will consider any missing values when calculating the results. FINDINGS In a sample of 138 respondents, consisting of 47% males and 53% females, participants were required to select the best and worst words linked to the term trust from a set of 4 words presented in 8 displays. This measurement assessed preferences among 6 adjectives and 6 nouns, which were equally represented on average. The result shows that the words reliable (20.6 %) and loyal (19%) received the highest average ratings and are therefore likely the most strongly linked with the word trust based on participants' opinions. The result simultaneously serves as the answer to our research question (RQ1): Some words are more frequently linked with the term trust by respondents. Words such as propulsive (1.1%), step (1.5%), pulse (1.8%), and touch (2.3%) received lower ratings and are therefore less linked with trust in this context. Table 2: Results of rescaled scores Advances in Business-Related Scientific Research Journal, Volume 15, No. 1, 2024 112 From the graph above and below (which we added for better representativeness of Table 2), we can discern a pattern in which adjectives (5 out of 6) have the strongest preferences, followed by all 6 nouns except for one adjective that deviates from the pattern. The reason likely lies in the choice of preferences, which we may not have accurately anticipated. Table 3: Frequency of chosen words to term trust (n = 138) 20,6 19 15,2 12,6 12,3 8,3 2,7 2,6 2,3 1,8 1,5 1,1 0, 5,5 11, 16,5 22, reliable loyal caring consistent genuine sense Inspiration vision touch pulse step persuasive Percentage [% ] Average Advances in Business-Related Scientific Research Journal, Volume 15, No. 1, 2024 113 Table 4: Results of paired samples t-test between the adjective reliable and the adjective propulsive Paired Samples Statistics Mean N Std. Deviation Std. Error Mean Pair 1 Zanesljiv 20,601 138 2,701 ,230 Prodoren 1,109 138 1,062 ,090 Paired Samples Correlations N Correlation Sig. Pair 1 Zanesljiv & Prodoren 138 -,173 ,043 Paired Samples Test Paired Differences t df Sig. (2- tailed) Mean Std. Deviation Std. Error Mean 95% Confidence Interval of the Difference Lower Upper Pair 1 Zanesljiv - Prodoren 19,493 3,057 ,260 18,980 20,005 75,177 138 ,000 However, considering that we compared the means of two adjectives, we cannot definitively confirm the presence of differences among words linked to the term trust. This exclusion occurs because nouns (besides verbs, etc., which are not included in the scope of this study) are not considered here. Therefore, we subsequently conducted a statistical comparison between the most frequently chosen adjective reliable and the most frequently chosen noun sense, obtained from the calculation in the Sawtooth Software MaxDiff program. We analyzed the SPSS program and found, through the calculation of the t-test, that there is a statistically significant difference between these two words, specifically at p <0.02. In the table below, we see that we can confirm statistical significance, indicating a statistically significant difference in the selection of words linked with the term trust, supported by the standard deviation value (SD = 6.57). Table 5: Results of paired samples t-test between adjective reliable and the noun sense Paired Samples Statistics Mean N Std. Deviation Std. Error Mean Pair 1 Zanesljiv 20,601 138 2,701 ,230 Razum 8,263 138 5,500 ,468 Paired Samples Correlations N Correlation Sig. Pair 1 Zanesljiv & Razum 138 -,192 ,024 Advances in Business-Related Scientific Research Journal, Volume 15, No. 1, 2024 114 Paired Samples Test Paired Differences t df Sig. (2- tailed) Mean Std. Deviation Std. Error Mean 95% Confidence Interval of the Difference Lower Upper Pair 1 Zanesljiv - Razum 12,339 6,550 ,555 11,240 13,440 22,212 138 ,000 Considering that we compared both adjectives with each other and adjectives with nouns, and in both cases, the p-value did not exceed 0.05, we can confirm that there is a statistical difference between words linked to the term trust (H1). Among the 12 possible words that were displayed equally, 80.8% of 138 respondents chose adjectives as a more relevant link to the term trust. The result answers our second research question (RQ2): In the set of word links, the selection of adjectives for the term trust is more common. To test H2, where we claim that adjectives are statistically significantly more appropriate than nouns, we compared the frequencies of average ratings of nouns and average ratings of adjectives, obtained from calculations in the Sawtooth Software MaxDiff program. We analyzed the SPSS program and found, through the calculation of the t-test, that there is also a statistically significant difference between nouns and adjectives, at the level of p < 0.05, as seen in Table 6. Table 6: Results of paired samples t-test between means of adjectives and nouns Paired Samples Statistics Mean N Std. Deviation Std. Error Mean Pair 1 PRID-avr 13,47 139 1,66 ,14 SAM-avr 3,20 139 1,66 ,14 Paired Samples Correlations N Correlation Sig. Pair 1 PRID-avr & SAM-avr 139 -1,000 ,000 Paired Samples Test Paired Differences t df Sig. (2- tailed) Mean Std. Deviation Std. Error Mean 95% Confidence Interval of the Difference Lower Upper Pair 1 PRID-avr - SAM- avr 10,27 3,32 ,28 9,71 10,83 36,50 138 ,000 DISCUSSION Advances in Business-Related Scientific Research Journal, Volume 15, No. 1, 2024 115 This study investigates the notion of distrust and its implications for consumer behavior, with a specific focus on word preferences in advertising messages using methodologies, particularly MaxDiff, in an innovative approach. In our study, we focused on word preferences within the research questions: "Some words are more frequently linked with the term trust by respondents" (RQ1) and "In the set of word links, the selection of adjectives for the term trust is more common" (RQ2), to possibly understand consumer word preferences better. In a study conducted by Jakomin et al. (2022), measurements were made using the MaxDiff method, which revealed that fewer than one-third of those surveyed expressed confidence in the information provided by manufacturers regarding their products (in their case, green products). In the methodology section, MaxDiff was presented as a tool for discerning the word preferences of participants and identifying which words are associated with the concept of trust. We asked 138 respondents (n=138) which word out of 12 possible words (6 adjectives and 6 nouns) they chose as the closest link to the term trust. It turned out that the adjective "reliable" accounts for 20.6% of all attributes (words). It is followed by the adjectives "loyal" (19%) and "caring" (15.6%). To measure H1, which assesses statistical significance between words linked with the term trust, we compared the average values of two adjectives: the most frequently chosen "reliable" and the least chosen "propulsive" with a t-test. For additional confirmation, we also conducted a measurement between the most frequently chosen adjective "reliable" and the most frequently chosen noun "sense" to avoid excluding nouns. The result of p < 0.04 (for the comparison between "reliable" and "propulsive") and p < 0.02 (for the comparison between "reliable" and "sense") means that in 95% of trials, the true population mean value would lie within the confidence interval that has been calculated. We found that 80.2% of the total preferences are represented by adjectives. We confirmed with a t-test and a standard deviation value (SD=3.33, p < 0.02) that there is a statistical difference between adjectives and nouns, thus confirming H2, that adjectives are statistically significantly more appropriate than nouns. Our findings show that certain words, particularly adjectives like "zanesljiv" (reliable), "zvest" (loyal), and "skrben" (caring), are more strongly linked to the term trust in the minds of respondents. This highlights the significant role that word choice plays in influencing consumer perceptions and trust in advertising messages. The fact that adjectives received the highest preference and that statistical tests confirmed their significance compared to nouns underscores the importance of selecting the right words when crafting advertising messages. These insights suggest that words are indeed an important element in building trust with consumers and can have a significant impact on their perceptions and preferences. We must mention that respondents were provided with a limited set of words from which they could choose, and they did not have the option for open-ended responses. We are aware that there may be more suitable Advances in Business-Related Scientific Research Journal, Volume 15, No. 1, 2024 116 words that were not measured. For example, instead of "propulsive" (1.1%), which turned out to be the least frequently chosen word, even though it is an adjective, it was rated as less preferable than nouns. This study explores the occurrence of specific words associated with the concept of trust to gauge consumer preferences in language. By pinpointing words that have a meaningful impact on consumers, marketers, and advertisers can tailor their messages effectively, leading to increased credibility and deeper connections with their intended audience. Ultimately, the insights gained from this research can improve the efficacy of advertising tactics, offering long-term benefits to both consumers and advertisers. In conclusion, the decline in trust within advertising not only impacts marketers but also impacts consumers' capacity to make knowledgeable decisions regarding products and services that could enrich their lives. This research paper delves into the critical issue of credibility within advertising messages, which has increased due to growing consumer distrust. We proposed the use of the MaxDiff method to analyze word preferences as attributes to enhance the effectiveness of advertising messages, as we measure the links between 12 words to the term trust. One of the limitations evident in this research is the limited number of words available as options for possible connections to the term trust. For easier statistical processing and an innovative approach, we selected the MaxDiff module, which does not allow open-ended responses. It is also worth mentioning that the word propulsive stands out within the word set, indicating a very atypical link. The link to the term trust seems inappropriate, for which perhaps a better alternative would be persuasive. Another important limitation to mention is the automatic exclusion of individuals who are not computer literate. The survey was conducted online, meaning that only those with internet access and computer literacy could participate. This represents a convenience sample, which may affect the representativeness of our sample. Not all potential participants had equal opportunities to take part, potentially biasing our sample towards individuals who are more technologically adept and have internet access. As a result, the overall validity of the results may be affected. At the same time, we selected adjectives using the simple synonymy principle, while the nouns were selected based on prototypical categorization (Ahmed, 2020) while simultaneously taking into account the reader's cultural and physical experience (Cortés de los Ríos, 2022) and the influence of the five dimensions that affect credibility (Rajaobelina et al., 2019). Perhaps we should opt for one or the other approach here, which means that it might be better to also use simple synonyms for the nouns, such as credibility, trustworthiness, reputation, commitment, loyalty, and dedication. The aim of utilizing quantitative data in this study is to acquire objective, numerical findings concerning consumer preferences in advertising. The emphasis lies not so much on understanding the reasons behind consumers' preferences for specific words and phrases, but rather on identifying which words possess greater persuasive influence. Advances in Business-Related Scientific Research Journal, Volume 15, No. 1, 2024 117 Considering that word preferences are subjective and that the concept of trust is influenced by cultural factors, utilizing the MaxDiff method, similar to what we conducted in our study in the Slovenian language, could be implemented in different countries. By conducting cross-cultural studies, researchers could uncover cultural nuances and differences in the interpretation of trust-related words, providing valuable insights for global marketing strategies. Future research could utilize the MaxDiff method to delve deeper into the connections of trust across sex and age groups, examining how perceptions of trust and word preferences vary among these demographics. Researchers could design surveys tailored to specific sex and age groups, presenting participants with sets of words related to trust and asking them to choose the most and least preferable words. By analyzing the responses from different demographic groups, researchers could gain insights into how trust and word preferences differ. Additionally, exploring the long-term effects and correlations of trust and word preferences on consumer behavior and brand loyalty could offer deeper insights into advertising strategies. Longitudinal studies tracking changes in consumer perceptions and behaviors over time in response to different advertising messages and word choices could provide valuable data on the effectiveness of various marketing approaches. For example, a longitudinal study could track a sample of consumers over several years, periodically measuring their levels of trust in a particular brand and their word preferences in advertising messages. Researchers could collect data on consumer behavior, such as purchase intentions, brand loyalty, and actual purchase decisions, at multiple time points throughout the study. Throughout the study period, participants could be exposed to different advertising campaigns featuring varying word choices related to trust. By analyzing changes in consumer perceptions and behaviors over time in response to these advertising messages, researchers could assess the long- term effects and correlations between trust, word preferences, and consumer behavior. For instance, they might find that consumers who consistently encounter advertising messages with trustworthy word choices are more likely to exhibit higher levels of brand loyalty and make repeat purchases over time compared to those exposed to less trustworthy messaging. Conversely, they could identify that changes in word preferences in advertising are associated with shifts in consumer trust levels and subsequent changes in brand loyalty and purchase behavior. Similarly, we see this method as useful in brand name selection, in cases where there is doubt and there is a desire to learn how people respond to a particular name, or when alternative options are very close and pose a challenge in decision-making. All options are presented in the MaxDiff questionnaire, where respondents are asked: "Which name do you find as most suitable for a company that deals with..." and they choose the best and worst representative. Here, MaxDiff can serve as an answer to "What if?" Advances in Business-Related Scientific Research Journal, Volume 15, No. 1, 2024 118 Finally, it is worth mentioning Sweetser (1987), who analyzed the word "lie" in the English language (which is somewhat the opposite of the term trust), stating that "prototypical semantics treat the meaning of words as determined by central or prototypical use, rather than categorical boundaries. This allows clear definitions for words with unclear boundaries of use. We determine the best example of word use and expect real-world examples to more or less correspond to this best example, but not completely or not at all." REFERENCES Chandler, D. (1994). Semiotics for Beginners: Glossary of Key Terms. http://visual-memory.co.uk/daniel/Documents/S4B/sem-gloss.htmL Cohen, S. H. (2003). Maximum Difference Scaling: Improved Measures of Importance and Preference for Segmentation. www.sawtoothsoftware.com Cortés de los Ríos, M. E. (2002). Cognitive semantics and axiology: a new proposal to study metaphor in economics advertising discourse. Revista Alicantina de Estudios Ingleses, 15, 39. https://doi.org/10.14198/RAEI.2002.15.03 Darke, P. R., & Ritchie, R. J. B. (2007). The defensive consumer: Advertising deception, defensive processing, and distrust. Journal of Marketing Research, 44(1), 114–127. https://doi.org/10.1509/jmkr.44.1.114 Ebaid, H. A. (2018). Adjectives as Persuasive Tools: The Case of Product Naming. Open Journal of Modern Linguistics, 8(6), 262–293. https://doi.org/10.4236/ojml.2018.86022 Gentile, C., Spiller, N., & Noci, G. (2007). How to Sustain the Customer Experience:. An Overview of Experience Components that Co-create Value With the Customer. European Management Journal, 25(5), 395–410. https://doi.org/10.1016/j.emj.2007.08.005 Jakomin, E., Novak, I., Štefanič, J., Toroš, J., & Elesini, U. S. (2022). Use of MaxDiff method in selecting green packaging attributes that influence purchase decisions in online shops. Packaging Technology and Science, 35(12), 879–892. https://doi.org/10.1002/pts.2684 Kotler, P., & Armstrong, G. (2008). Principles of Marketing. Pearson Prentice Hall. Upper Saddle River, NJ. Lakoff, George (1988). Cognitive semantics. In Umberto Eco (ed.), Meaning and Mental Representations. Bloomington: Indiana University Press. pp. 119--154. Lee, S., & Potter, R. F. (2020). The Impact of Emotional Words on Listeners ’Emotional and Cognitive Responses in the Context of Advertisements. Communication Research, 47(8), 1155–1180. https://doi.org/10.1177/0093650218765523 Newell, S. J., Goldsmith, R. E., & Banzhaf, E. J. (2015). The Effect of Misleading Environmental Claims on Consumer Perceptions of Advertisements. Journal, 6(2), 48–60. https://doi.org/10.1080/10696679.1998.11501795 Pollay, R. W., & Mittal, B. (1993). Here’s the Beef: Factors, Determinants, and Segments in Consumer Criticism of Advertising. Journal of Marketing, 57(3), 99–114. https://doi.org/10.2307/1251857 Prihatini, A. (2020). Word Association and Its Function at the Constituent Understanding on the Language Learning. KnE Social Sciences. https://doi.org/10.18502/KSS.V4I4.6461 Pulvermüller F, Härle M, Hummel F. (2000) Neurophysiological distinction of verb categories. Neuroreport, 11(12), 2789–93. Pulvermüller, F. et al. (2001) Walking or talking? Behavioral and neurophysiological correlates of action verb processing. Brain Lang, 78, 143–168 Rajaobelina, L., Lacroix, C., & St-Onge, A. (2019). Does experiential advertising impact credibility? International Journal of Bank Marketing, 37(1), 171–191. https://doi.org/10.1108/IJBM-01-2017-0021 Schmid, A. A. 2004. Conflict and cooperation. Malden: Blackwell. Schwartz, B. (2004). The Paradox Of Choice: Why More Is Less. The Paradox Of Choice: Why More Is Less. https://works.swarthmore.edu/fac-psychology/198 Advances in Business-Related Scientific Research Journal, Volume 15, No. 1, 2024 119 Sweetser, E. E. (1987). The definition of "lie": An examination of the folk models underlying a semantic prototype. In D. Holland & N. Quinn (Eds.), Cultural models in language and thought (pp. 43–66). Cambridge University Press. https://doi.org/10.1017/CBO9780511607660.003 Vasiloaia, M. (2020). Semantic Ambiguity in Advertising Language. Economy Transdisciplinarity Cognition, 23(2/2020), 95–99. Xie, G. X., Madrigal, R., & Boush, D. M. (2015). Disentangling the Effects of Perceived Deception and Anticipated Harm on Consumer Responses to Deceptive Advertising. Journal of Business Ethics, 129(2), 281–293. https://doi.org/10.1007/s10551-014-2155-2