The Influence of Monochromatic Illustrations on the Attention to Public Health Messages: An Eye-Tracking Study

Abstract

Past research suggests that images can be a useful tool in attracting people’s attention and improving public health communication. This is especially important for the effective transmission of health-related messages to the youth, who should be warned about risks of alcohol and tobacco use. This study explores the application of eye-tracking methodology to investigate the influence of monochromatic illustrations on youths’ visual attention to digital ads, with a particular emphasis on those presenting harmful effects of alcohol and tobacco consumption. Four digital ads were designed for the purpose of the experiment. Two ad topics (alcohol versus tobacco) and two ad contents (text-only versus text with a monochromatic illustration) were used as variables in the digital ad design. Analyses of eye fixations and viewing time revealed that more students noticed the digital ads with the illustration than the text-only ads. Furthermore, we found that the ads with illustrations had a longer viewing time and a higher number of fixations, regardless of the ad topic. The findings highlight the effectiveness of eye-tracking technology in examining the role of illustrations in digital ads, focusing on a better perception of addiction prevention campaigns whose message is targeted primarily toward young people.

1. Introduction

In the current digital era, online content plays a major role in the optimization of visual advertising, especially for young target audiences. As new media continue to evolve, the creative use of digital images has expanded. Previous research suggests that images can be a useful tool in attracting people’s attention and improving marketing communication. Images serve more than just a visual embellishment; sometimes, they are key components that can significantly enhance the effectiveness of marketing messages, particularly in the domain of public health communication. For campaigns that require seriousness and directness, such as addiction prevention, the use of a black monochromatic palette can amplify the emotional effect of the advertised content. Additionally, monochromatic imagery can stand out from the typical colorful marketing content, thereby emphasizing its uniqueness in a mass of information and attracting more attention and the target group’s awareness. With the development of sophisticated technologies for the analysis of graphic content such as eye tracking, researchers are equipped to directly measure how digital ads attract attention and how successfully they convey visual messages. As a result, they are able to conduct precise image analyses and adjust its visual parameters to maximize viewer engagement.

For example, a recent study [1] examined visual attention to ad features and found that images of people attracted greater attention than other ad elements. Focusing on the ad format, another research [2] revealed that a native advertisement attracted more attention than a banner advertisement. More recently, an eye-tracking experiment was conducted by Chen-Sankey et al. [3], who analyzed visual attention metrics in the context of e-cigarette advertisement and showed that the participants directed attention differently to various ad features. All of these studies have predominantly focused on digital advertising and photos, while monochromatic illustrations conveying public health warnings have not been sufficiently evaluated in terms of objective direct measurements. Digital ads and corresponding photos can offer defined parameters for effectiveness analyses, making them popular for both academic and commercial studies. On the other hand, illustrations frequently present more challenges for quantitative assessment due to their specific structure. The illustration format may exhibit different potential for attracting attention, calling for a more in-depth investigation. Therefore, in this study, eye tracking was employed to examine the effects of monochromatic illustrations on the visual processing of public health warnings. This is especially important for the effective transmission of health-related messages to the youth, who should be warned about risks of alcohol and tobacco use since both are harmful and potentially addictive [4].

2. Problem Statement

According to the 2030 Agenda for Sustainable Development [5], one of the Sustainable Development Goals is to ensure health and promote wellbeing for people of all ages (Goal 3). In that context, the youth require careful consideration because supporting their wellbeing sets the foundations for a generation that can drive progress of society, economic stability, and sustainable development. Under the goal of wellbeing promotion, the Agenda for Sustainable Development addresses prevention of substance abuse—including harmful consumption of alcohol (Goal 3.5) and strengthening the capacity of all countries for the early warning and management of global health risks (Goal 3.d). Within that framework, public health campaigns can achieve a great amount in addressing the importance of the youth’s healthy lifestyle on the national level. Thus, this study seeks to explore whether digital advertising with a specific, yet simple, type of illustration can be effective in promoting anti-alcohol and anti-smoking behavior among young Croatian people.

In Croatia, alcohol use is a public health concern, so national strategy aims to reduce alcohol-related harm by concentrating on consumption among youth [6], with a special focus on prevention [7] and changes in drinking culture by promoting non-drinking as a socially accepted behavior [8]. Particularly alarming are the findings of a study on the role of social marketing in the prevention of alcohol abuse [9], which indicated that alcohol consumption is a very common lifestyle among Croatian students. The concern is not limited to drinking only, but extends to smoking. A recent study on students in southern Croatia [10] revealed that there is a lack of knowledge about the harmful effects of tobacco, which emphasizes the need for better prevention in this area also. Therefore, our study covers both anti-alcohol and anti-smoking messages as a component of the support to the public health authorities in tailored initiatives aimed at the young population and their healthy lifestyle [11,12].

Previous studies investigated whether specific forms of warning messages regarding tobacco harm can be effectively communicated through packaging [13,14,15,16]. These studies examined the perception of warning messages in conditions where a harmful product is already packed and offered to a potential consumer. In an effort to emphasize the importance of prevention and warning about product harm in earlier stages (before a young person is in direct contact with the product), the present study is focused on ads as a medium for transmitting preventive messages. Nowadays, digital advertising is the most rapidly expanding type of promotion [17]. It can quite successfully target message recipients, ranging from a general to a more specific audience. In current advertising-related targeting, more focus is on the sentiment of target groups instead of just their demographic data [18], and this is also the case for persuasive communication in community-based public health campaigns. For example, a study on public health advertisements confirmed that emotion-based advertisements increased willingness to donate more than rational-based ones [19]. Even in situations in which people search for health information of a more sensitive nature, such as topics on psychotic disorders, digital ads can be an effective method for encouraging help seeking in the mental health community [20].

Still, there are studies that suggest that people sometimes pay less attention to digital ads than other content elements regularly incorporated into a webpage design. In a study [21], Sørum reported that the participants noticed the ads but their attention was not maintained on the ads for a long time. This implies that, in order to understand the complexities of people’s visual attention in a digital environment, it is essential to consider the visual perception of an ad as a process unfolded in two primary stages. The first stage refers to the ability of the ad to attract attention, and the subsequent stage refers to its ability to maintain the attention of viewers. A powerful objective measure of both attention stages is viewers’ eye movements. The effectiveness of eye movement analyses for understanding aspects of viewer behavior is recognized in a wide range of research disciplines [22], and marketing is not an exception [23,24,25,26,27]. Therefore, to explore the potential of the deliberated selection of visual elements in public health marketing, this study employed eye-tracking technology to investigate viewers’ visual attention to digital ads that convey health-related messages. More specifically, our objectives were to examine whether digital ads with illustrations promoting anti-alcohol and anti-smoking behavior can attract the attention of Croatian students, and to understand which types of ad designs are most likely to maintain their attention.

Hypotheses

Building upon the defined research objectives, this chapter introduces a set of hypotheses that served as framework that guided a systematic design of the ads in our study, testing their attention-grabbing properties.

The foundation of our work lies in the acknowledgment that design quality is an important aspect of visual advertising that influences visual perception. Past research provides strong evidence for this. In a study on health advertisements [28], the authors selected the ads that varied in three levels of design quality (i.e., low, mid, high) and examined whether they gain different levels of participants’ attention. They found that advertisements of low levels of design quality were viewed for a shorter amount of time than those of higher levels of design quality. A study on print versions of ads [29] showed that highly rated advertisements attracted more attention compared to those that were poorly rated. Judging the quality of design often involves considering the attractiveness of its elements. In a study on content elements of online banner ads, it was found that the image was the most attractive element [30]. The positive effect of images is supported by results of past research that examined the ability of visual ads to capture visual attention [31,32]. The superiority of pictorial content was also demonstrated in a study in the field of health-related advertising. O‘Malley et al. [33] used eye-tracking technology to measure the attention toward text-only, image-only, and image–text ads promoting osteoporosis prevention. They found that the combination of text and images was more effective at attracting the participants’ visual attention than the image-only and text-only ad formats. Past research suggests that graphic images included in cigarette warning labels on advertising improved recall of the warning and associated health risks [34]. They did so by attracting and maintaining the visual attention of smokers. Images used as an addition to text warnings on cigarette packaging may be particularly effective for youth populations, as suggested by Casado-Aranda et al. [35].

Therefore, we set three hypotheses as follows. The first hypothesis refers to the effect of images (particularly illustrations) on the ad’s ability to attract youth visual attention, while the second hypothesis refers to attention maintenance. Finally, the third hypothesis refers to the effect of illustrations on the ad’s general noticeability. All the hypotheses focus on the visual attention variables because the attention is necessary for further processing [36] and greater visual attention may increase the possibility that young people will comprehend and remember the viewed information more easily.

H1: 

A digital ad with text and an illustration attracts the attention of young viewers faster than a digital ad with no illustration.

H2: 

A digital ad with text and an illustration maintains the attention of young viewers for longer than a digital ad with no illustration.

H3: 

Young participants notice a digital ad with text and an illustration more than a digital ad with no illustration.

To test the hypotheses, we employed an eye-tracking experiment. Eye tracking is considered as a suitable tool for consumer neuroscience research [37]. It allows a researcher to obtain quantitative data by recording viewers’ eye movements during advertising exposition, and provides an insight into the viewers’ underlying cognitive processes [35]. Hence, understanding eye movement behavior can be particularly useful in improving the persuasive quality of digital ads in public health promotion [27], especially of those that target young generations and gain their attention in today’s media-saturated environment.

3. Materials and Methods

3.1. Stimuli Design

We designed four digital ads that had different topics (alcohol vs. smoking) and different content elements (text only vs. text with an illustration). Other design elements and visual attributes were consistent in all the ads, such as size, color, layout, and alignment. Different messages were used for different ad topics, but both consisted of two words that were presented with the same typeface and in the same type size. Translated in English, the text “Alcohol harms.” was used for the anti-alcohol message, and the text “Stop smoking.” was used for the anti-smoking message. Both ads contained a brief description of health-related issues that were directly associated with the topic. The brief description did not exceed more than three lines of text, similar to the conditions in previous studies [33]. The number of characters in the brief description ranged from 91 to 100, and the word count ranged from 18 to 19 words. Each ad took up 12.65% of the webpage area. The webpage was designed as a static page with a neutral color palette [38] and generic layout to reduce the effects of the color saturation and surrounding on the perception of the ad. Since the target group in this study was students, all textual content for the webpage was taken from the website of the University of Zagreb.

The results from human–computer interaction research suggest that banner blindness is strongest for digital banners placed on the right side of the screen [39]. A study on teenagers’ perception of advertisements [40] showed that ads in the left column have a positive effect on visual attention. Therefore, the ads in our study were positioned on the left part of the webpage.

The illustrations used in those digital ads that contained the combination of text and illustration were originally created by one of the authors (M.M.), who made them particularly for the purpose of the study. The anti-alcohol ad contained an illustration with a person in a bottle while the anti-smoking ad contained an illustration with a person in an ashtray. In creating the illustrations, the goal was to present dramatic, almost shocking visual representations of negative consequences arising from alcohol abuse and cigarette smoking. The designer sought to apply the same graphic style to both illustrations and to make their level of visual complexity as similar as possible. Both illustrations were horizontally oriented. Since previous research [30] suggests that attention to digital ads may vary depending on the viewers’ familiarity with the advertised brand, the digital ads in our study contained an imaginary brand presented through a simple generic logo.

3.2. Participants

There were 58 participants in the study, of which 47 were females. Given that the study deals with messages that target a young population, age was a criterion for participant selection. Their age ranged from 20 to 25 years (M = 22.5, SD = 1.71). All had normal or corrected vision and reported no visual or neurological disorders. All signed informed consent.

3.3. Equipment

The experiment took place in a controlled, quiet testing lab at the Faculty of Graphic Arts, University of Zagreb, designed to minimize external distractions and reduce background noise. Distinct light sources were removed and the test materials were presented directly in front of the participants to focus attention. The participants’ eye movements were recorded by Tobii Eye Tracker X60 (Tobii Technology, Stockholm, Sweden), with a sampling rate of 60 Hz and an accuracy of 0.5 degrees. The digital ads were presented on a computer display (model LEN L1900pA, Lenovo, Morrisville, NC, USA) with a screen resolution of 1280 × 1024 pixels. Tobii Studio 3.2.1 software was used for the presentation of stimuli.

3.4. Procedure

Each student participated in an eye-tracking test individually. After a successful 5-point calibration, the experiment started. Each trial started by displaying a central fixation dot on the computer screen. Then, a webpage with one of the ads was presented for 10 s, and then the same procedure was carried out for a webpage with another ad. Each student was exposed to just one version of the ad for each topic. The participants were pre-informed of the time constraints for viewing the webpage. They were instructed to view the page freely, without a specific task, since a recent study showed that attention to digital ads may be influenced by a goal-oriented process [41]. Order of stimuli presentation is also noted to have an impact on people’s responses [25]. Therefore, the order of ad presentations was randomized by selecting the Counterbalance mode in the presentation software (Tobii Studio 3.2.1), which was enabled for all materials in the session. The software arranged the order according to a Latin square method.

3.5. Dependent Measures

One of the benefits of the eye-tracking software is that it allows the experimenter to predefine areas (or regions) of interest [42]. The eye-tracking data in our experiment were calculated for the following areas of interest: area of the digital ad, area of the illustration on the ad, and area of the slogan. An example of one digital ad can be seen in Figure 1. For the digital ads with no illustration, only two areas of interest were analyzed (i.e., area of the digital ad and area of the slogan). We analyzed the measures of visual attention that were calculated in similar studies, which employed vision-based tracking systems [43,44,45,46]. In order to test the hypotheses, we categorized these measures into two groups, with one related to attention attraction and one related to attention maintenance.

For the first one (attention attraction), two variables were calculated: time to first fixation (TFF) and fixations before (FB). TFF is the time spent from the stimulus onset until the particular area of interest was fixated on for the first time [47]. A shorter time usually indicates that the visual element is more effective in attracting the attention. FB refers to the number of fixations that the participant directed toward other visual elements on the webpage before noticing an area of interest.

For the second group of measures (attention maintenance), three variables were calculated: total fixation duration (TFD), fixation count (FC), and total visit duration (TVD). TFD is the total time that the participant spent fixating on visual content. FC is the number of times that the participant fixated on the area of interest. TVD is the sum of visit durations that the participant spent looking at the area of interest.

For the third hypothesis, which refers to a general ad’s noticeability, a binary variable was used, also known as hit ratio [48]. If a participant fixated on the ad at least once, the value was 1. If a participant did not fixate on the ad at all, the value was 0.

4. Results

The participants’ fixations were first analyzed by checking the attention maps generated by the eye-tracking software (Figure 2). Attention maps in the form of heatmaps are a common way to present a visualization of the density of viewers’ eye-tracking data. Different color hues represent different amounts of participants’ fixations. Longer fixations recorded on the stimulus are presented by a red color, followed by yellow and green colors. By comparing the regions of the webpage marked in red (Figure 2), a higher prevalence of the red color can be noticed in the ads containing an illustration than on the ads with no illustration, indicating a longer average fixation duration on the illustrated ads.

The eye-tracking data were then statistically analyzed with IBM SPSS 23. A two-factor repeated measures analysis of variance (ANOVA) was carried out to examine whether ad topic (alcohol vs. smoking) and ad content (text only vs. text with an illustration) significantly influenced the participants’ eye-tracking measures. The results showed that ad topic had no influence on any of the measures of visual attention (for both attention attraction and attention maintenance), and this applies to all the examined areas of interest. Ad content had no statistically significant effect on the attention attraction variables (TFF, FB), p > 0.05. However, ad content significantly influenced attention maintenance variables. The ads with an illustration had a longer TFD (M = 2.19, SD = 0.18) than the ads with no illustration ((M = 1.78, SD = 1.18), F(1.57) = 4.69, p < 0.0). Furthermore, the illustrated ads had a significantly higher FC (M = 10.36, SD = 0.78) than the ads with no illustration ((M = 8.53, SD = 0.84), F(1.57) = 5.92, p < 0.05), and TVD was also longer for the illustrated digital ads (M = 2.76, SD = 0.23) compared to the ads with no illustration ((M = 2.13, SD = 0.22), F(1.57) = 6.59, p < 0.05). Figure 3 shows the mean values for each measure.

By analyzing the eye-tracking data recorded only in the area of the illustration (

Table 1. Descriptive statistics and comparisons between two ad topics for all dependent variables (note: area of interest was the illustration).

Ad Topic	Statistic	TFF	FB	TFD	FC	TVD
Alcohol	Mean (SD)	4.03 (0.45)	9.71 (1.21)	1.20 (0.15)	5.43 (0.61)	1.35 (0.17)
Smoking	Mean (SD)	3.98 (0.44)	7.68 (0.82)	1.21 (0.16)	5.28 (0.63)	1.41 (0.18)
ANOVA results	F-value	0.12	2.14	1.74	0.43	0.72
	p-value	0.91	0.15	0.19	0.084	0.79

), we found no significant influence of the illustration’s visual representation (man in the bottle vs. man in the ashtray) on any of the variables (all ps > 0.05).

McNemar’s tests were used to examine the influence of ad topic and ad content on the general noticeability. Figure 4 shows the distribution of participants depending on whether they noticed the ad. The only statistically significant difference was found between the ads containing an illustration and those without, indicating that more participants noticed the ads with text and an illustration (96.6%) than the ads with text only (89.7%); p < 0.05.

5. Discussion

The goal of this study was to determine whether adding an illustration in digital ads that promote anti-alcohol and anti-smoking behavior can make the ads more attention grabbing. Thus, we compared the viewers’ eye-tracking data for the ads with an illustration and those with text only. The main finding was that the inclusion of the illustration in the ads had a beneficial influence on the variables that measured how long the ads maintain the participants’ attention for. This confirms hypothesis H2 and suggests that adding pictorial content to health-related advertising has the potential to hold the youth’s focus in a digital environment. This is not surprising since it is reasonable to expect that the visual richness of an ad maintains longer visual attention [49]. The result is consistent with prior research [32], which demonstrated the effect of ad content on participants’ perception and showed that a higher level of attention was recorded among the participants exposed to pictorial ads than those exposed to text ads. This is also in line with the results reported by O’Malley et al. [33], who found that an ad with an image and text was more effective at attracting participants’ attention than an ad with text only. This could be of particular importance for the effectiveness of public health promotions, which should not only be easily noticed by young people but also remembered well. As reported in a study on banner advertisement exposure [50], a longer duration of fixation on an ad can lead to better memory performance. Thus, our results contribute to a more general perspective provided by prior works [51] that showed that presentation style in ads can greatly influence whether the message will be seen and remembered. High values for two measures recorded on illustrated ads in our study (i.e., TFD and FC) could also be considered as an indicator of the ad’s visual interest. According to findings of earlier studies in advertising, total fixation time can be a significant predictor of ad clicking [41], and longer fixations are also associated with an increased depth of ad processing [23]. Additionally, more fixations correlate with high memory [52,53]. Taken together, these results suggest that illustrations in digital ads can assist marketing practitioners in developing a digital ad that is more noticeable and, consequently, more memorable. Still, this does not necessarily apply to all formats of advertising, such as television commercials. A neuroscience-based research study on online advertising [54] showed that television commercials that had a longer visit duration were less recalled. Since our results refer to a static digital ad (as well as to a static webpage), this suggests that the level of content dynamics likely influences attention and recall performances. Therefore, using an illustration for improving the attention-grabbing properties of an ad does not guarantee good memorability for all types of advertising.

Additionally, we made an analysis of the ads’ noticeability on a more general level, in which we compared the frequency of the people who detected the ad and those who did not. More participants noticed the ads with text and an illustration than the ads with no illustration, which supports hypothesis H3. This is in line with previous findings on the strong effect of the pictorial content in ads on attention [55]. Still, caution is needed when generalizing these results to diverse ad categories. For example, results from the field of tourism advertising indicated that ads containing text without an image attracted the participants’ attention more effectively [56]. This implies that relying just on imagery in digital ads is not enough. Specific contextual characteristics of the promotional message, such as field of advertising and target audience, should also be taken into account.

Regarding the ability of a digital ad to attract attention, our results showed no effect of ad content on the variables that measured how fast the ads attracted the participants’ attention, which refutes hypothesis H1. This outcome deviates from other studies upon which we based our assumptions, so it could be ascribed to the dominance of the textual component. As Gómez-Carmona et al. [57] state, ads with negative text can activate the viewer’s emotional system faster. Thus, it seems that the verbal aspect of the digital ads in our study surpassed the pictorial aspect.

Although they were not included in the hypotheses, additional conclusions regarding the ad’s pictorial content can be drawn from our results. The students in our experiment were exposed to two different illustrations (one for each topic), and the variations in the illustration did not influence their visual attention. This could be explained by the position of the illustration, which was centrally located in both versions of the ad. Since the central part of the composition has been identified as the most effective location in terms of attracting the viewers’ interest in previous research [58], it is possible that placing the illustrations in the ad center had a more profound influence on attention than their motifs. However, a previous study on online health messages [59] indicated that the total fixation duration was influenced by different illustration types. Also, a study on a tourist hotel’s marketing images [60] showed that different images led to different levels of customers’ attention, which suggests that the lack of an image content effect should not be expected in all types of marketing materials.

Although our results may not be representative enough to apply these findings to broader demographic groups, they still indicate how a simple form of an ad image, such as a monochromatic illustration, can be effective in the context of digital advertising. Still, some of the discrepancies with other neuromarketing studies highlight the need for further investigation of more specific illustration attributes to comprehensively understand how illustrated content enhances the visual interest of a digital ad. In a larger context of public health promotion, the results of this study may contribute to the ongoing efforts to promote wellbeing for the youth and strengthen the prevention of harmful use of alcohol and tobacco among Croatian students.

6. Conclusions

This study explored how a combination of text and image can be effective in promoting anti-alcohol and anti-smoking behavior among young people through digital media. The objective was to obtain direct insight into the visual attention of Croatian students when they view digital ads signaling alcohol and tobacco harm and to compare the amount of attention attracted by the text-only and the text–illustration ad format. Based on the findings, it can be concluded that the presence of an illustration in the ads improved the level of visual interest to some extent. Although it did not reduce the time needed for ad detection, it did increase the amount of attention that students dedicated to the ad. It also increased the possibility of spotting the ad on the webpage. This study contributes to the existing body of knowledge on the visual perception of digital ads by giving further evidence toward the influence of images on people’s visual processing.

The findings may be of interest for those engaged in public health marketing and those who are developing efficient marketing approaches for the promotion of youth wellbeing, supporting Goal 3 of the 2030 Agenda for Sustainable Development. Regarding theoretical implications, this study contributes to research-based knowledge on youths’ visual stimulation, suggesting that monochromatic illustrations can be particularly effective in retaining their attention on important messages. The findings could also contribute to the development of decision-making authorities’ guidelines or regulations that promote the use of specific visual techniques to enhance public health initiatives.

However, this study still has some limitations. Our sample was not representative enough to analyze the potential influence of gender on the eye-tracking outcomes. Furthermore, certain aspects of the ad effectiveness, such as comprehension or recall, were not addressed in our investigation. Investigating further hypotheses on the long-term effects on memory from ad exposure could provide a more comprehensive understanding of the effectiveness of the monochromatic illustrations. Future studies would benefit from the inclusion of the variable of whether the participants themselves have a precedent of alcohol and tobacco intake. A broader potential of illustrated anti-alcohol and anti-smoking messages in current non-commercial marketing practices, such as the power to encourage changes in youth behavior, is also worthy of a detailed examination. This calls for teams of researchers and specialists from different domains, like graphic design, marketing, and neurosciences. It is highly likely that cross-disciplinary collaboration among these fields could lead to better persuasive performances of digital ads in promoting a healthy lifestyle for the youth.