The effect of augmented reality versus traditional advertising: a comparison between neurophysiological and self-reported measures

The market for augmented reality (hereafter, AR), which can be defined as the integration of digitally created media within an existing real environment (Azuma et al., 2001), is projected to increase from roughly 5.91 billion in 2018 to more than 198 billion in 2025 (Statista, 2019). Companies around the world have been increasingly using AR as an advertising tool to showcase their products and services in retail stores. Examples of AR applications that might simplify consumers’ decision-making are Ikea Place, which enables customers to place the company’s furniture wherever they imagine it in their homes (Joseph, 2017), and Lacoste LCST AR, which allows customers to virtually try on shoes (Arthur, 2014).

The past literature on advertising suggests that ad-induced arousal strongly influences consumers’ intention to buy the products shown in the ad (Sanbonmatsu & Kardes, 1988; Singh & Churchill Jr, 1987). However, measuring arousal and its impact on customers’ behavioral intentions is a major challenge for research in interactive advertising (Belanche et al., 2017; Pozharliev et al., 2017). Specifically, past empirical evidence on AR applications in marketing communication indicates that AR advertising can enhance different aspects of consumers’ affective responses (Poushneh & Vasquez-Parraga, 2017; Yim et al., 2017). However, previous studies on emotions elicited by consumer experience with AR advertising typically relied solely on retrospective self-reported measures, and hence reported representations of emotions rather than emotions themselves, which in turn may lead to missing important relationships between emotions, advertising type, and purchase outcomes (Jessen et al., 2020; Park & Yoo, 2020; Poncin & Mimoun, 2014; van Esch et al., 2019). Consumer neuroscience methods offer more direct access to the unconscious emotional process that affect customer responses to advertising (Dmochowski et al., 2014; Falk et al., 2012; Venkatraman et al., 2015) and thus may help us uncover relationships that other methods would miss. Specifically, we expect neurophysiological, but not self-reported, measures of emotion to mediate the relationship between advertising-type AR (versus traditional) and purchase outcomes. Therefore, the main objective of this paper is to study whether retrospective self-reported measure of arousal versus real-time physiological measure of arousal differ in their prediction regarding the behavioral intentions for a product advertised through AR (versus print advertisement). This question is of critical importance for companies that are looking to assess the impact of AR advertising on consumer emotions and whether these emotions drive WTP for the product shown in the ad.

The present research offers two major contributions to advancing extant scientific knowledge. First, we contribute to the literature on consumers’ responses to AR. In particular, while previous work has investigated issues such as the impact of responsiveness and playfulness of AR on attitude toward a product (Park & Yoo, 2020), the effect of discomfort with AR on attitude toward the brand (van Esch et al., 2019), or the influence of customer engagement with AR on anticipated satisfaction (Jessen et al., 2020), our research investigates the intensity of consumers’ emotional responses as a possible mediator of the impact of AR (versus print) advertising on consumers’ behavioral intentions. Moreover, building on the processing fluency model, we show that AR (versus print) advertising enhances the ease with which consumers process product-related information, which in turn leads to consumers having more positive emotional responses toward the product/brand. In doing so, we identify a novel theoretical mechanism—processing fluency—underlying consumers’ enhanced emotional responses toward AR advertising (Lee & Labroo, 2004).

Second, we contribute to the growing body of literature that seeks to understand how unconscious emotional processes affect online experiences and customer responses to advertising (Dmochowski et al., 2014; Falk et al., 2012; Venkatraman et al., 2015). This stream suggests that neurophysiological measures of emotions can predict future consumer behavior above and beyond traditional self-reported metrics (Berns & Moore, 2012; Boksem & Smidts, 2015; Pozharliev et al., 2015). On this basis, we introduce the idea that a neurophysiological measure of emotions—namely physiological arousal—can potentially offer unique information on consumers’ behavioral intentions for a product experienced through AR, which would be unobtainable with self-reported measures. In particular, we use consumer neuroscience (namely, galvanic skin response; hereafter, GSR) measures to examine whether the underlying physiological arousal (versus the self-reported affect intensity) will drive customer willingness to pay (hereafter, WTP) for a product advertised through AR versus print advertising. In doing so, our research helps clarify the methodological strengths of neurophysiological measures for research in interactive advertising. Specifically, our results show that neurophysiological measures of emotions provide a better way of capturing consumers’ responses to AR compared to self-reported measures.

Our research might also offer practical implications for marketing communication managers and retailers looking for new ways to emotionally engage customers, and particularly for those managers interested in empirical evidence supporting their decision to invest in AR. Specifically, we show that companies that are looking to leverage AR should use physiological arousal to measure the impact of AR advertising on customer behavioral intention for the product(s) shown in the ad.

One key finding is that neurophysiological measures can be more effective in unveiling the dynamic nature of emotions and their impact on consumer intentions toward products experienced through AR advertising (Berns & Moore, 2012; Boksem & Smidts, 2015).

According to the processing fluency model, advertising exposures enhances the ease with which consumers process product/brand-related information; this increased fluency leads to consumers having more favorable responses toward the product/brand. Although the effect of processing fluency on affective responses is well-documented (Lee & Labroo, 2004; Reber et al., 1998), there are no previous studies that examine whether the ability of AR (versus print) advertising to enhance consumers’ emotional responses is driven by the medium’s superior conceptual and/or perceptual fluency. Thus, the objective of study 2 is to examine the role that processing fluency plays in enhancing consumer emotional responses to AR advertising.

This research has shown that the higher WTP for consumers exposed to AR (as opposed to standard print advertising) is driven by consumers’ physiological arousal, but not by their self-reported affect intensity. Overall, we believe these results might offer relevant theoretical and managerial insights, as well as generate questions about customers’ emotional responses to AR advertising that future studies might address.

The limitations of our work can serve as ideas for future research. First, our findings are limited to one particular digital platform, namely Ikea Place, where most of the products that customers see are much bigger than their viewing screen size (e.g., bed, table, sofa). It would be interesting to see if our results are confirmed with small-sized products (e.g., watches, diamond rings). After all, using AR to present those products may generate more realistic product visualization and thus affect customers’ physiological responses to the digital platform and/or the advertised product or service (Yim et al., 2017).

Second, our unconscious measures of customer experience with AR are limited to individual physiological responses (GSR). GSR allows us to understand the impact of arousal and emotional intensity in the relationship between AR and purchase outcomes, but it does not allow us to assess more nuanced emotional responses. The application of other consumer neuroscience methods and metrics could produce further insights into the affective, cognitive, and social processes that affect customers’ responses to and behavior with AR-based technologies. For example, a combination of GSR and heart rate variability (HRV) will allow us to assess both the intensity and valance of the emotional responses to AR-based technologies. The electroencephalography (EEG) method offers another neurophysiological approach to assess the valance of the emotional responses to AR-based advertising. For example, Vecchiato et al. (2010) found greater left (versus right) frontal brain activity for commercials that consumers particularly liked (versus disliked). Thus, the use of EEG frontal asymmetry, in combination with GSR and HRV, could provide further relevant insights into customers’ emotional responses and the unconscious drivers of behavioral intentions with AR-based technologies.

Third, our findings are limited to one particular product category, namely home furniture (e.g., bed, table, sofa). Future research could, therefore, compare how emotional responses toward AR advertising might differentiate across different products and how this is tied to consumer behavior. Specifically, previous research suggests that the relative level and influence of affective responses (e.g., pleasure and arousal) to a product differ significantly, depending on whether the product is hedonic or functional in nature. Specifically, felt arousal during the product experience is higher for the hedonic product than for the functional product, and emotional responses were significant drivers of consumers’ behavioral intentions toward said product (Kempf, 1999).

Finally, the active traditional advertising condition asked participants to write a prompt imagining where and how they would organize the piece of furniture in a room of their house/apartment. Previous research suggests that mental simulation is important for harnessing one’s imagination (Taylor et al., 1998). Specifically, mental simulation plays a key role in problem solving and emotion regulation because it enables consumers to envision possibilities and develop plans for bringing those possibilities about (Taylor et al., 1998). Therefore, future research could look at the role of mental simulation in problem solving and/or emotional responses to AR-based advertising. For example, one could formulate the imagination instructions as a mental simulation by identifying a specific goal, such as one relating to the décor or comfort, and the steps needed to achieve this goal. Moreover, future work could provide further theoretical convergence to our findings. For example, scholars could study the competitive value of fluency in AR (versus traditional) advertising and whether there is anything traditional advertising can do to “catch up” with AR (Reber et al., 1998).