When brands come alive: An exploration of the effects of experience and agency as the building blocks for a humanized brand

(1)

When brands come alive:

An exploration of the effects of experience and agency as

the building blocks for a humanized brand.

Nathalie Huijser

Student number: 10587012

Master’s Thesis

Graduate School of Communication

Master track: Persuasive Communication

Supervisor: Dr. S. F. Bernritter

(2)

Contents

Abstract………...p. 02 1. Introduction………....p. 02

2. Theoretical Background……….…p. 05 2.1 Brand anthropomorphism……….…p. 05 2.2 Creating a humanized brand..……….……….….p. 08 3. Methodology………...p. 12 3.1 Participants and design………...p. 12 3.2 Procedure………..…p. 13 3.3 Measures……….……..p. 15 4. Results……….……p. 16 4.1 Randomization checks……….….p. 16 4.2 Correlation analyses……….….p. 17 4.3 Manipulation checks……….……p. 18 4.4 Main analyses……….…..p. 19 4.5 Explorative additional analyses…….……….. p. 23 5. Conclusion………..p. 23 5.1 Discussion……….…p. 23 5.2 Limitations………....p. 26 5.3 Implications and suggestions for future research……….….p. 28 6. References……….……..p. 31 7. Appendix………...p. 36

(3)

Abstract

This study investigated if a humanized brand, created by attributing the inherently human capacities for experience and agency to that brand, positively affects

consumers’ purchase intentions, and if this effect is mediated by consumers’ tendency to anthropomorphize the brand. Building upon research into the dimensions of mind perception, perceived agency in a brand was expected to strengthen the effect of perceived experience on the tendency to anthropomorphize, which in turn was expected to affect participants’ purchase intentions. These assumptions were tested through an online 2x2 between-subjects experiment. No evidence for the expected moderated mediation effect was found (for which this study elaborately provides possible explanations, such as the manner in which experience and agency were operationalized). Building upon the results from the performed analyses and previous literature, suggestions for future research are provided. The present study contributes to existing literature as it explores the building blocks for creating a human brand, which has received minimal scientific attention. Moreover, the discussed limitations and suggestions could assist future studies in examining the antecedents and effects of perceived human presence in brands.

1. Introduction

“Wherefore men say that the Gods have a king, because they themselves either are or were in ancient times under the rule of a king. For they imagine, not only the forms of the Gods, but their ways of life to be like their own.” – Xenophanes of Colophon.

While a brand is traditionally defined as “a type of product manufactured by a particular company under a particular name,” (Oxford Dictionaries, n. d.) research

(4)

suggests that brands can encompass much more. In the eyes of the consumer, brands can be perceived as human beings, with human traits, behaviours and mind

(Puzakova, Kwak, & Rocereto, 2009). As a result, consumers relate to brands in similar ways as they do to people (Fournier, 1998). The tendency to perceive human presence in nonhuman agents, such as brands, has been termed anthropomorphism (Fournier, 1998). Xenophanes’ description of this phenomenon encompasses its’ complexity: anthropomorphism does not refer to a mere humane depiction, but rather to the state of perceiving a nonhuman entity as a complete human (Sider, 1994).

Consumers’ tendency to anthropomorphize brands has been found to be very beneficial for those brands in several ways. For one, brands that are presented as having human-like features are more positively evaluated than brands that lack such features (Aggarwal & McGill, 2007). Brands that are perceived as human have even been found to inspire brand love amongst consumers (Rauschnabel & Ahuvia, 2014). The positive feelings consumers experience towards brands they anthropomorphize often result in willingness to spend more money than they would for similar, non-anthropomorphized brands (Puzakova et al., 2011). Following these findings, creating a brand that motivates consumers to anthropomorphize the brand seems to be the ideal branding strategy. However, existing research on effective ways of inducing people to anthropomorphize brands is limited. One study showed that priming people with a text written in a human tone of voice (first person language and casual tone), and afterwards presenting them with a product that was congruent with a human schema (a car that appeared to be smiling), was an effective way inducing the tendency to anthropomorphize a brand (Aggarwal & McGill, 2007). Other studies have efficiently induced this tendency by asking people to imagine that the brand had come to life as a person (Aggarwal & McGill, 2011; Rai & Diermeier, 2015). However, the need for

(5)

extensive priming makes the described processes quite difficult to operationalize through marketing efforts in a non-experimental setting.

It could thus be concluded that a more feasible way of inducing people to anthropomorphize a brand is needed. While perceived humanity in brands carries positive effects, the existing effective methods to create such brands are hardly applicable to real-world marketing efforts. As described, anthropomorphism is not a mere humane depiction; it requires making inferences about unobservable specific human characteristics (Epley, Waytz, Akalis & Cacioppo, 2008). To create a human brand, one should thus know how people perceive humanity, in order to bestow this upon a brand.

According to Gray, Gray & Wegner (2007), people perceive minds along the dimensions of agency and experience. Experience is the capacity to feel pain, pleasure and emotions, whereas agency describes one’s capacity to act, plan and exert self-control (Gray et al., 2007). The capacities for experience and agency are inherently human characteristics, which separate humans from robots and animals, respectively (Henkel, Boegershausen, Hoegg, Aquino, & Lemmink, 2018; Zlotowski, Strasser, & Bartneck, 2014). When others are viewed as incapable of these capacities, they are regarded as nonhuman (Waytz, Epley, & Cacioppo, 2010; Złotowski et al., 2014).

Since the agency and experience dimensions closely relate to what is perceived as human, it is the present study’s expectation that these capacities form the building blocks for creating a human brand. The present study investigates if humanizing a brand, achieved by attributing capacities for experience and agency to that brand, positively affects consumers’ intentions of purchasing that brand, and if this effect is mediated by consumers’ tendency to anthropomorphize the brand.

(6)

The present study aims to contribute to the existing literature on human presence in brands by investigating how a human brand can be created. While the positive effects of consumers’ tendency to anthropomorphize brands have been extensively researched, limited literature exists on antecedents of this tendency. Moreover, the existing literature mainly focuses on antecedents that are difficult to manipulate. This study intends to fill the research gap by exploring if a human brand can be created through branding efforts, by attributing experiential and agentic mental capacities to that brand. Uncovering how the tendency to anthropomorphize can be manipulated would naturally be highly relevant from a managerial perspective as well. If brand managers know how to create a human brand through their branding strategy, their firm could benefit from increased popularity and sales.

2. Theoretical Background

2.1 Brand anthropomorphism

Anthropomorphism has been described as “the tendency to imbue the real or imagined behaviour of nonhuman agents with humanlike characteristics, motivations, intentions, or emotions” (Epley, Waytz & Cacioppo, 2007, p. 864). The word derives from the ancient Greek words anthrōp (meaning “human being”) and morphos (“form”) (Merriam Webster Online, n. d.). As a term, anthropomorphism was first coined by the Greek philosopher Xenophanes, when he remarked that people project their own physical and cultural characteristics upon their gods (Eisenstadt, 1974). Since then, the concept has been a topic of debate among philosophers (Puzakova, Kwak & Rocereto, 2009). More recently, anthropomorphism has become a

(7)

in human terms through their marketing efforts (Aaker, 1997; Aggarwal & McGill, 2007).

Research on brand anthropomorphism was initially focused on brand personalities: the attribution of a specific set of human characteristics to a brand (Aaker, 1997; Aaker & Fournier, 1995). Aaker (1997) found that consumers might perceive a brand to be dependable, honest, or even charming. She identified five dimensions of brand personality: sincerity, excitement, competence, sophistication, and ruggedness. These dimensions, also referred to as the “Big Five,” each encompass several representative traits (Aaker, 1997). The attribution of human personalities to brands was termed anismism by Aaker. Whilst animism and anthropomorphism have often been used interchangeably in marketing research, they refer to different

definitions (Puzakova et al., 2009). Animism has been described as simply attributing life to the nonliving (Epley et al., 2007), but animate life is not uniquely human (Epley et al., 2008). Anthropomorphism defines the more complex state of viewing nonhuman entities as complete humans (Epley et al., 2007; Puzakova et al., 2009); it requires making inferences about uniquely human characteristics that are not directly observable (Epley et al., 2008).

Anthropomorphized brands have been defined as “brands perceived by consumers as actual human beings with various emotional states, mind, soul, and conscious behaviours that can act as prominent members of social ties” (Puzakova, 2009, pp. 413-414). This description thus encompasses the more complex definition of anthropomorphism as described, where consumers view brands as complete

humans. This phenomenon has been found to be highly beneficial for firms. Research suggests that anthropomorphized brands make consumers’ brand experience more authentic (Morhart et al., 2015). Furthermore, consumers tend to trust, care for, and

(8)

forgive anthropomorphized brands sooner than non-anthropomorphized brands (Puzakova, Kwak & Andreas, 2011). Puzakova and colleagues (2011) suggest that these effects occur because the relationship the consumer experiences with the anthropomorphized brand is similar to their relationships with friends; whom are more easily forgiven than acquaintances, for instance. Another positive effect of brand anthropomorphism is consumers’ increased motivation to interact with the brand, according to Aggarwal and McGill (2011). Their findings even suggest that consumers tend to assimilate their behaviour to the anthropomorphized brand, in case it involves a liked brand (Aggarwal and McGill, 2011). Overall, consumers have been found to evaluate anthropomorphized brands more positively than

non-anthropomorphized brands (Aggarwal & McGill, 2007). Anthropomorphized brands have even been found to inspire brand love amongst consumers (Rauschnabel & Ahuvia, 2014). These positive feelings that consumers experience towards

anthropomorphized brands often result in willingness to spend more money than they would for similar, non-anthropomorphized brands (Puzakova et al., 2011)

Research thus suggests that perceived humanity in brands results in positive outcomes for the firms owning such brands. Quite some research exists on specific consumer characteristics that make it more or less likely for consumers to

anthropomorphize a brand, such as effectance motivation and sociality motivation

(Epley et al., 2007; Puzakova et al., 2009; Waytz, Epley, & Cacioppo, 2010; Waytz, Morewedge, Epley, Monteleone, Gao, & Cacioppo, 2010). However, such

characteristics differ per individual and are difficult to induce via marketing efforts. The following section explores the creation of a human brand through a more controllable tactic: branding.

(9)

2.2 Creating a humanized brand

To create a brand that is perceived as human by consumers, one should understand what constitutes humanity. Gray, Gray and Wegner (2007) explored the dimensions of mind perception and found that people perceive minds along the dimensions of experience and agency. The agency and experience dimensions relate to the capacities of having intentions and abilities, regardless of their valence (Gray et al., 2007; Rai & Diermeier, 2015). Experience relates to moral patiency, which is the capacity to be the target of another’s right or wrong actions. Agency is related to moral agency, which is the capacity to do right or wrong (Gray et al., 2007; Gray & Wegner, 2009). Experience is defined by the following capacities: hunger, fear, pain, pleasure, rage, desire, pride, embarrassment, consciousness and having a personality (Gray et al., 2007). The capacity for experience corresponds to Human Nature, since its’ defining capacities separate humans from robots (Henkel et al., 2018; Zlotowski et al, 2014). Agency is defined by the following capacities: self-control, morality, memory, emotion recognition, planning, communication, and thought (Gray et al., 2007). The capacity for agency has also been referred to as being Uniquely Human, since it encompasses the characteristics that separate us from animals (Henkel et al., 2018; Zlotowski et al, 2014). The agency and experience dimensions thus closely relate to what is perceived as human. This becomes especially clear when one considers the effects of the perceived absence of experience and/or agency in others.

When human agents are regarded incapable of experience and agency, they are seen as nonhuman objects; a process that has also been termed dehumanization (Waytz et al., 2010; Złotowski et al., 2014). As a result, those incapable of experience are viewed as robotic, cold and cruel, and those regarded incapable of agency are seen as animalistic and subservient (Gray & Wegner, 2011).

(10)

In line with these findings, it seems reasonable to assume that perceived experiential and agentic mental capacities closely relate to perceived humanity. Since perceived humanity in a brand has been shown to have many positive outcomes (Aggarwal & McGill, 2007; Rauschnabel & Ahuvia, 2014), resulting in willingness to spend more money than they would for similar, non-anthropomorphized brands (Puzakova et al., 2011), the following hypotheses are proposed:

H1a: Exposure to a brand that is perceived to have a high experiential mental state will result in higher purchase intentions for that brand compared to exposure to a brand that is perceived to have a low experiential mental state.

H1b: Exposure to a brand that is perceived to have a high agentic mental state will result in higher purchase intentions for that brand compared to exposure to a brand that is perceived to have a low agentic mental state.

Agency is considered to be the capacity that separates us from animals (Henkel et al., 2018; Zlotowski et al, 2014) but not from machines, since machines can be designed to possess agentic capacities. This is also reflected in the capacities that people ascribe to robots, as they are perceived to be capable of agency (Gray & Wegner, 2012). However, robots are not perceived to be capable of experience; Gray and Wegner (2012) found that humanlike robots become unnerving to people when they are associated with the capacities of feeling hunger, fear, or other emotions. According to the authors, this feeling of unease suggests that experience, rather than agency, is seen as fundamental to humans (Gray & Wegner, 2012). This finding is corroborated by Rai and Diermeier (2015), who discovered that people generally associate companies with agency, but regard them as incapable of having experiential

(11)

mental states. Experience thus appears to be integral to our conception of human nature

Following these findings, the present study expects that perceived high experiential mental capacities in brands positively affect people’s tendency to anthropomorphize those brands, compared to brands that are perceived to have low experiential mental capacities. In line with these expectations, this study theorizes the following:

H2: Exposure to a brand that is perceived to have a high experiential mental state will result in a higher tendency to anthropomorphize that brand, compared to exposure to a brand that is perceived to have a low experiential mental state has.

As discussed, experience is defined by the capability to feel hunger, fear, pain, pleasure, rage, desire, pride, embarrassment, and the capacity for consciousness and for having a personality. Agency is defined by the capacities for self-control,

morality, memory, emotion recognition, planning, communication, and thought (Gray et al., 2007). Experience refers to the capacity to sense and feel, and agency to the capacity to plan and act (Waytz et al., 2010). Agency is thus the ability to act upon experiential states of mind; someone could be perceived as being capable of feeling desire, hunger or embarrassment, but planning, thought and memory are needed to act upon those feelings. Therefore, it is expected that perceived agency in a brand

strengthens the effects of experience on consumers’ tendency to anthropomorphize that brand. Moreover, since agency is the Uniquely Human capacity that separates humans from animals (Henkel et al., 2018; Zlotowski et al, 2014), it is expected that perceived agency in a brand also has a direct effect on consumers’ tendency to anthropomorphize that brand.

(12)

It is thus hypothesized that the effect described in H2 is moderated by agency, but also that agency has a direct, positive influence on brand

anthropomorphism. We expect that a) brands that are perceived to have both a high experiential and agentic mental state induce the highest likeliness of brand

anthropomorphism, and that b) perceived agency in a brand directly influences people’s tendency to anthropomorphize that brand. The following hypotheses are proposed:

H3: The relationship described in H2 is moderated by agency, in that the relationship between experience and the tendency to anthropomorphize will be stronger when exposed to a brand that is perceived to have a high agentic mental state, than when exposed to a brand that is perceived to have a low agentic mental state.

H4: Exposure to a brand that is perceived to have a high agentic mental state will result in a higher tendency to anthropomorphize that brand, compared to exposure to a brand that is perceived to have a low agentic mental state.

The first section of this chapter discusses theory on the relationship between brand anthropomorphism and brand attitude: studies have found that consumers evaluate anthropomorphized brands more positively than their

non-anthropomorphized counterparts (Aggarwal & McGill, 2007; Puzakova et al., 2011; Rauschnabel & Ahuvia, 2014). Aggarwal and McGill (2007) argue that brand anthropomorphism mediates the relationship between of exposure to a humanized, branded product and consumers’ brand attitudes. This positive effect of brand

anthropomorphism has, in turn, been found to positively affect consumers willingness to pay: consumers are prepared to spend a higher sum of money for products from anthropomorphized brands than they would for similar products from a

(13)

non-anthropomorphized brand (Puzakova et al., 2011). Based upon these findings, it seems theoretically grounded to assume that brand anthropomorphism has a direct, positive effect on purchase intention. Adding this expectation to the previously discussed assumptions, a moderated mediation effect is proposed: It is expected that the effect of experience on purchase intention is mediated by brand

anthropomorphism, and that the effect of experience on brand anthropomorphism is moderated by agency.

H5: The relationship between experience and purchase intention (H1) will be mediated by brand anthropomorphism, in that the relationship between experience and brand anthropomorphism, moderated by agency (H3), will affect purchase intentions, in such that an increased tendency to anthropomorphize will result in higher purchase intentions.

Figure 1.

The conceptual model.

3. Methodology

3.1 Participants and design

To test the hypotheses, an experiment with a 2 (high vs. low agency) x 2 (high vs. low experience) between-subjects design was conducted. Participation in the

(14)

experiment was secured via convenience sampling. A total of 219 respondents were recruited to participate, of which 34 (15.5%) failed to complete the survey and were consequently excluded from any further analyses. The final sample consisted of N = 185 participants (125 female), with an average age of 27.37 years (SD = 9,86). The youngest participant was 18 years old, and the oldest participant was 71 years. A total of 150 participants (81.1%) were Dutch, and 29 participants (15.7%) were non-Dutch Europeans. The remaining 6 participants (3.2%) had a non-European nationality.

3.2 Procedure

The in Qualtrics created survey started with a general introduction about the procedure, after which anonymity, confidential processing of the gathered data, and the right to withdraw consent were guaranteed. Next, participants were told that they would be exposed to a description of a new brand, on which they would have to provide their impressions. After reading this information, participants were randomly assigned to one of the four experimental conditions.

Participants in all conditions were exposed to a text describing the Australian bicycle brand REID Cycles, which is not sold in the Netherlands. This brand was selected to ensure that participants were not familiar with the brand, and therefore did not have pre-existing attitudes towards the brand. After reading the description of the brand’s mission statement and product range, participants were exposed to the experience and agency manipulation. Participants read a short text describing which qualities consumers ascribe to REID Cycles in their reviews, and/or on which domains they indicate that there is room for improvement. In the conditions where REID Cycles is positioned as a brand with a high experiential mental state,

(15)

“ambitious,” and “a brand with character”. These characteristics were selected from the defining capacities of experience (Gray et al., 2007) and adapted to fit the

manipulation material. In the conditions where REID Cycles is positioned as a brand with a highly agentic mental state, participants could read that consumers described the brand as “moral,” “good at planning,” and “communicative.” These characteristics were selected from the defining capacities of agency (Gray et al., 2007), and have been slightly adapted to fit the manipulation material. In the conditions where REID Cycles was positioned as having a low experiential mental state, the reviews indicated that there was room for improvement on the three experiential characteristics. In the low agency conditions, participants read that there was room for improvement on the three agentic characteristics. This approach was based on Bernritter, Verlegh and Smit (2016). For the manipulation material, see appendix figures 1 through 4.

After exposure to the manipulation material, participants’ purchase intention, brand attitude and their tendency to anthropomorphize the brand were measured. Following these questions, participants were exposed to the manipulation material once more, after which they were asked to what extent they perceived the brand as capable of all defining capacities of both agency and experience (Gray et al., 2007). Lastly, several control measures followed: participants were asked to indicate their bike use frequency, gender, nationality, age and effectance motivation. The latter was taken into account since effectance motivation has been shown to affect the tendency of anthropomorphizing nonhuman entities (Epley et al., 2007; Puzakova et al., 2009; Waytz, Epley, & Cacioppo, 2010; Waytz, Morewedge, Epley, Monteleone, Gao, & Cacioppo, 2010). The survey ended with an expression of gratitude for participation and participants were once more informed on how to withdraw consent if so desired.

(16)

3.3 Measures

Manipulation check:

Experience and agency: To verify whether the manipulation of experience and agency was successful, the experience- and agency-scales (Gray et al., 2007) were administered. Participants were requested to indicate, on a 7-point Likert scale ranging from totally incapable (1) to perfectly capable (7), to what extent they perceived the brand to be capable of all defining capacities of both agency and experience. The original scale for experience consisted of 11 items and the original scale for agency consisted of 7 items. However, to fit the present study’s design, in which three items of both scales were operationalized in the manipulation, 3-item scales for experience (α = .86) and agency (α = .87) were created to serve as the manipulation check.

Dependent variable:

Purchase intention: The purchase intention measure consisted of one 7-point Likert item, ranging from extremely unlikely (1) to extremely likely (7): “If I were in the market to buy a bike a this time, I would consider buying a bike at REID Cycles.” The question was taken from the studies of Chapman and Aylesworth (1999) and Kervyn, Fiske and Malone (2012).

Mediator:

Brand anthropomorphism: The degree to which participants

anthropomorphized the brand was measured on a 7-point Likert scale ranging from very untrue of what I believe (1) to very true of what I believe (7): the brand has intentions, a mind of its’ own, desires, consciousness, and the ability to experience

(17)

emotions (Waytz et al., 2010). The five-item scale proved to be reliable with Cronbach’s Alpha of .86.

Control variables:

Brand attitude: Participants’ brand attitude was measured on a 7-point Likert scale ranging from completely disagree (1) to completely agree (7): The brand is good, trustworthy, respectable, favorable, and of high quality (Van Noort &

Willemsen, 2012). The five-item scale was reliable with a Cronbach’s Alpha of .86. Effectance motivation: Effectance motivation was measured using the Desire for Control scale, which was developed by Burger and Cooper (1979), following Epley and colleagues (2008), who argue that the desire for control is equal to effectance motivation. The Desire for Control scale originally includes twenty statements (Burger & Cooper, 1979), amongst which a lot of duplicates of one another. In order to not strain the participants too much, the scale was reduced to eleven items. Participants were asked to indicate the extent to which the 11 statements applied to them on a 5-point Likert scale, ranging from The statement doesn’t apply to me at all (1) to The statement always applies to me (5). Examples of the items were: “I prefer a job where I have a lot of control over what I do and when I do it.”, or: “I would prefer to be a leader rather than a follower.” However, because the scale turned out to be non-reliable (α = .32), effectance motivation was excluded from any further analyses.

4. Results

4.1 Randomization checks

To investigate if participants’ gender, nationality, and bike use were equally distributed among the experimental conditions, several Chi-square tests were run. The

(18)

first Chi-square test, with the experimental condition as the independent-, and gender as the dependent variable, showed that there were no significant differences in gender distributions across conditions, χ2 (3) = 2.30, p = .513. The second Chi-square test, with the experimental condition as the independent-, and nationality as the dependent variable, revealed no significant differences in nationality distributions across

conditions, χ2 (6) = 7.72, p = .259. The last Chi-square test, with the experimental condition as the independent-, and bike use as the dependent variable, revealed no significant differences in bike use distributions across conditions, χ2 (3) = 8.49, p = .205. Hence, the equal distribution of participants’ gender, nationality, and bike use across conditions could be assumed.

To investigate if participants’ age was equally distributed among the

experimental groups, a one-way ANOVA was run with the experimental condition as the independent-, and age as the dependent variable. The analysis revealed that there were no significant age differences across the experimental conditions, F(3,181) = 1.64, p = .182. The equal distribution of participants’ age across conditions was thus assumed as well.

4.2 Correlation analyses

To check if any of the control variables affected the dependent variable, purchase intention, a correlation analysis was run. The results revealed a significant positive weak relationship between gender and purchase intention, r = .15 p = .045. This means that women had a higher intent of purchasing the product than men did. Moreover, a significant, positive moderate relationship between brand attitude and purchase intention was found, r = .40, p = <.001. As brand attitude increases,

(19)

between purchase intention and age (r = -.04, p = .614), nationality (r = .04, p = .550), and bike use (r = <.01, p = .961). Hence, only gender and brand attitude were

controlled for in the analyses involving the dependent variable purchase intention. A second correlation analysis was run to check if any of the control variables correlated with the mediator, brand anthropomorphism. The results revealed a

significant positive weak relationship between gender and brand anthropomorphism, r = .22 p = .003. This means that women had a higher tendency to anthropomorphize the brand than men did. Moreover, a significant, positive moderate relationship between brand attitude and brand anthropomorphism was found, r = .42, p = <.001. The results indicated no significant relationships between brand anthropomorphism and age (r = .08, p = .244), nationality (r = -.03, p = .665), and bike use (r = -.01, p = .887). Gender and brand attitude were thus the only control variables taken into account in the analyses involving brand anthropomorphism.

4.3 Manipulation checks

To investigate if the manipulation of experience and agency was successful, two independent samples t-tests were run. The first independent samples t-test were used to test the effectiveness of the experience manipulation. The results revealed that those in the high experience conditions perceived the brand as having a significantly higher experiential mental state (M = 5.47 SD = 0.98) than those participants in the low experience conditions (M = 4.79 SD = 1.36), t (196.43) = 3.90, p = <.001, 95% CI [.33, .1.02]. Hence, the experience manipulation proved to be successful.

The second independent samples t-test was used to test if the agency manipulation was successful. Results indicated that those in the high agency

(20)

= 5.29, SD = 1.04) than participants in the low experience conditions (M = 4.36, SD = 1.23), t(183) = 5.59, p = <001, 95% CI [.61, 1.27]. The agency manipulation thus proved to be successful as well.

4.4 Main analyses

To test hypotheses 1a and 1b, a two-way ANOVA was conducted, with experience and agency as independent variables, and purchase intention as dependent variable. Gender and brand attitude were taken into account by adding them to the analysis as covariates. Results showed that the model was significant, F (5,197) = 7.40, p = <.001, 𝑅2 = .171. Hypothesis 1a stated that exposure to a brand that is perceived to have a high experiential mental state would result in higher purchase intention than exposure to brands that are perceived to have a low experiential mental state. The findings revealed that purchase intention was not significantly higher for those exposed to a brand that is perceived to be highly experiential (M = 5.13, SD = 1.36) than for those exposed to a brand that is perceived to have low experiential capacities (M = 5.05, SD = 1.28), F (1,197) = 1.23, p = .269, η² = .007, hereby rejecting hypothesis 1a. Hypothesis 1b stated that exposure to a brand that is perceived to have a high agentic mental state would result in higher purchase intention than exposure to brands that are perceived to have a low agentic mental state. The results of the ANOVA revealed that purchase intention was not

significantly higher for those exposed to a brand that is perceived to be highly agentic (M = 5.24, SD = 1.23) than for those exposed to a brand that is perceived to have low agentic capacities (M = 4.95, SD = 1.36), F (1,197) = 0.66, p = .416, η² = .004. As a result, hypothesis 1b was also rejected.

(21)

To test hypotheses 2, 3 and 4, a two-way ANOVA was conducted, with agency and experience as independent variables and brand anthropomorphism as dependent variable. Gender and brand attitude were incorporated in the analysis as covariates. Results showed that the model was significant, F (5,197) = 8.44, p = <.001, 𝑅2 = .191. Hypothesis 2 stated that exposure to a brand that is perceived to have a high experiential mental state would result in a higher tendency to

anthropomorphize those brands, compared to exposure to a brand that is perceived to have a low experiential mental state. The results indicated that brand

anthropomorphism was not significantly higher amongst those exposed to a brand that is perceived to be highly experiential (M = 4.84, SD = 0.98) than for those exposed to a brand that is perceived to have low experiential capacities (M = 4.86, SD = 1.03), F(1,197) = 0.29, p = .593, η² = .002, hereby rejecting hypothesis 2. Hypothesis 3 stated that the relationship described in H2 would be moderated by agency, in that the relationship between experience and the tendency to anthropomorphize would be stronger when exposed to a brand that is perceived to have a high agentic mental state, than when exposed to a brand that is perceived to have a low agentic mental state. The results showed no significant interaction effect between experience and agency, F(1,197) = 0.24, p = .623, η² = .001, meaning that agency does not moderate the effect of experience on brand anthropomorphism. Hypothesis 3 is thus rejected. Finally, hypothesis 4 stated that exposure to a brand that is perceived to have a high agentic mental state would result in a higher tendency to anthropomorphize that brand, compared to exposure to a brand that is perceived to have a low agentic mental state. Results of the ANOVA showed that brand anthropomorphism was not

significantly higher amongst those exposed to a brand that is perceived to be highly agentic (M = 4.88, SD = 1.06) than for those exposed to a brand that is perceived to

(22)

have low agentic capacities (M = 4.82, SD = 0.94), F(1,197) = 0.23, p = .623, η² = .001, hereby rejecting hypothesis 4.

To investigate this present study’s conceptual model (Figure 2) and hypothesis 5, a simple mediation (model 4) and a moderated mediation regression analysis

(model 7) were run using Process Macro (Hayes, 2013). Gender and brand attitude were incorporated in the analysis as covariates.

Figure 2.

The conceptual model including the unstandardized coefficients found in the results.

It was hypothesized that the relationship between experience and purchase intentions (H1) is mediated by brand anthropomorphism: the relationship between experience and brand anthropomorphism, moderated by agency (H3), affects purchase intentions, in such that an increased tendency to anthropomorphize results in a higher purchase intention.

The mediation regression analyses showed that brand anthropomorphism did not mediate the effect of experience on purchase intentions, b = .01, SE = .021, 95% CI [-.023, .063]. The moderated mediation regression analysis showed that the

expected moderated mediation effect was non-significant, b = -.01, SE = .045, 95% CI [-.139, .051], hereby rejecting hypothesis 5. All the results can be viewed in Table 1.

(23)

Table 1.

Results from the moderated mediation regression analysis.

Consequent

Brand Anthropomorphism (M) Purchase Intention (Y)

Antecedent b SE p b SE p Experience (X) 0.14 .190 .465 0.19 .180 .282 Agency (W) <0.01 .189 .993 Experience (X) x Agency (W) - 0.13 .270 .623 Brand Anthropomorphism (M) 0.11 .099 .282 Brand Evaluation 0.51 .090 <.001 0.61 .129 <.001 Gender 0.21 .151 .169 0.04 .201 .834 Constant 1.80 .486 <.001 1.20 .659 .070 𝑅2 = .437 𝑅2 = .414 F (5,179) = 8.44, p = <.001 F (4,180) = 9.29, p = <.001

(24)

4.5 Explorative additional analyses

Since the expected effects of experience and agency on purchase intention might have been overambitious after exposure to a single brand description,

additional analyses were performed with brand attitude as the dependent variable. A correlation analysis showed significant correlations between brand attitude and gender (r = .30, p = <.001), and between brand attitude and age (r = -.28, p = <.001). Gender and age were thus incorporated as covariates in the additional analyses.

To investigate the direct, separate effects of agency and experience on brand attitude, a two-way ANOVA was run, with agency and experience as independent variables and brand attitude as the dependent variable. Results showed that the model was significant, F (5,197) = 7.33 p = <.001, 𝑅2 = .170. The results showed that brand attitude was non-significantly affected by both experience, F (1,197) = 1.85, p = .176, η² = .010, and agency, F (1,197) = 3.72, p = .055, η² = .020.

To explore the effect of brand anthropomorphism on brand attitude, a mediation analysis was run using Process Macro (model 4). The results showed that brand anthropomorphism has a positive, significant influence on brand attitude, b = .323, SE = .051, p = <.001. As brand anthropomorphism increased, brand attitude increased as well.

5. Conclusion 5.1 Discussion

The main goal of the present study was to investigate if humanizing a brand, achieved by attributing capacities for experience and agency to that brand, would positively affect consumers’ intentions of purchasing that brand. This study

(25)

it was expected that perceived experience and agency in a brand would induce a tendency in consumers to anthropomorphize that brand, which in turn would positively affect their purchase intentions. The expected effects were formulated in five hypotheses, which were elaborately tested. From an exploratory standpoint, additional analyses were run with brand evaluation as the dependent variable.

Contrary to previous research that suggested that consumers are willing to spend more money on brands that are perceived as human compared to brands that are not (Puzakova et al., 2011), the present study found that agency and experience do not positively affect purchase intentions, as was hypothesized in H1a and H1b. The findings were, however, geared in the right direction. The lack of significant findings might be due to the manner in which experience and agency were manipulated, and to specific characteristics of the present study’s sample. These possibilities will be further explored in the limitations section.

Following research proclaiming the fundamentality of experience with regards to our perception of humans (Gray & Wegner, 2012; Rai & Diermeier, 2015). it was hypothesized in H2 that perceived experience in a brand would result in a higher tendency to anthropomorphize that brand. Moreover, considering the fact that agency is the ability to act upon experiential states of mind (Waytz et al., 2010), a moderating effect of agency on the relationship between experience and brand anthropomorphism was expected (H3). Lastly, a direct effect of agency on brand anthropomorphism (H4) was expected, since the capacity for agency is intertwined with humanity as well (Henkel et al., 2018; Zlotowski et al., 2014). The results showed, however, that none of these hypotheses were supported by the data. The lack of statistically significant findings might also be due to the manner in which experience and agency were manipulated, to be discussed in the limitations section. Moreover, research has shown

(26)

that the tendency to anthropomorphize is often due to consumer characteristics, which could have affected the results. Several studies point to the effects of consumers’ sociality- and effectance motivation (Epley et al., 2007; Puzakova et al., 2009; Waytz, Epley, & Cacioppo, 2010; Waytz, Morewedge, Epley, Monteleone, Gao, & Cacioppo, 2010). Sociality motivation refers to one’s need for social contact and affiliation

(Epley et al., 2007). Especially when people lack social connection with other human beings, they are more likely to anthropomorphize nonhuman agents (Waytz, Epley, & Cacioppo, 2010). Effectance motivation is the desire to understand, predict, and control one’s environment. When encountering an incomprehensible agent, such as a new brand, people with high effectance motivation often attribute the familiar concept of humanity to that agent, to gain some understanding (Waytz, Epley, & Cacioppo, 2010). While this study set out to control for effectance motivation, the variable could not be taken into account in the analyses because the scale was non-reliable.

Since consumers have been shown to evaluate anthropomorphized brands more positively than non-anthropomorphized brands (Aggarwal & McGill, 2007; Rauschnabel & Ahuvia, 2014), and to be willing to pay more for an

anthropomorphized brands’ products (Puzakova et al., 2011), this study expected that an increased tendency to anthropomorphize a brand would result in a higher intention of purchasing that brand. As experience and agency were expected to influence brand anthropomorphism, a moderated mediation effect was hypothesized in H5a & b. The obtained results were not in line with the present study’s expectations, as brand anthropomorphism was not found to significantly affect purchase intentions. The expected moderated mediation effect was non-significant as well. While the effect of brand anthropomorphism on purchase intention was not statistically significant, the additional analyses showed that brand anthropomorphism did significantly and

(27)

positively influence participants’ brand attitudes. The non-significant effect of brand anthropomorphism on purchase intention may be explained by the possibility that purchase intention was not the optimal way of assessing the effect of brand anthropomorphism in the context of this study. Brand attitude might have been a better measure, since the anthropomorphized brands’ products can be considered high-involvement products. While product involvement varies across consumers (Ling-Yee, 1997; Miller & Marks, 1996), product characteristics have been shown to largely determine the distinction between high- and low-involvement products (Laurent & Kapferer, 1985). Durable products such as electronics and cars are

textbook examples of high-involvement products, because the possibility of making a wrong purchase decision carries financial risk (Laurent & Kapferer, 1985; Gu, Park & Kanona, 2012). As a result, consumers engage in extensive information research before deciding to purchase such a high-involvement product, as opposed to the manner in which low involvement purchase decisions are made (Clarke & Belk, 1978; Gu et al., 2012). Hence, the participants in this study might not have intended to buy a new, likely-to-be expensive bicycle after viewing merely one piece of information about the brand, even if they evaluated the brand positively. The differences found between participants’ tendencies to anthropomorphize the brand, while not due to the experience and agency manipulation, could be explained by differences in

participants’ levels of effectance- and sociality motivation, as discussed above.

5.2 Limitations

The present study incorporated a manipulation check in the survey, instead of pre-testing the manipulation material. While the manipulation proved to be

(28)

brand between the experimental conditions were relatively small. One could argue that experience and agency should have been operationalized differently. Experience and agency were manipulated by listing characteristics from both dimensions as either descriptive of, or lacking with, the bicycle brand. When participants indicated which characteristics they perceived to be descriptive of the brand, they might have simply remembered which characteristics were mentioned in the brand description, instead of actually regarding the brand as “moral” or “cheerful,” for instance. If this was indeed the case, it seems reasonable to assume that the manipulation did not inspire

participants to anthropomorphize the brand. The way in which this study manipulated experience and agency might thus explain the fact that the results did not reach statistical significance.

As discussed above, several studies have asserted that levels of sociality- and effectance motivation influence people’s tendency to anthropomorphize (Epley et al., 2007; Puzakova et al., 2009; Waytz, Epley, & Cacioppo, 2010; Waytz, Morewedge, Epley, Monteleone, Gao, & Cacioppo, 2010). This study set out to control for

participants’ level of effectance motivation, but this variable had to be excluded from further analyses due to scale unreliability. This study could and perhaps should have controlled for sociality motivation as well. In order to keep the duration of the survey acceptable and not strain participants too much, the decision was made not to measure participants’ levels of sociality motivation. However, this data might have provided valuable insight into why participants’ levels of brand anthropomorphism diverged.

The fact that the present study’s convenience sample consisted largely of university students living in Amsterdam could have affected the generalizability of the results. As students generally have less to spend than employed adults do, students might be less inclined to buy a new bike. The fact that many of the participants were

(29)

living in Amsterdam could also have affected the results. Since the bike-theft rate in the Netherlands is highest in the city of Amsterdam (Bos, 2016), one could argue that its’ inhabitants are less keen on buying a new bike than inhabitants of other cities might be.

5.3 Implications & recommendations for future research

Considering the fact that the results of the present study’s main analyses were statistically non-significant, it seems unwise to make any definite claims about the effects of agency and experience on brand anthropomorphism and/or purchase intention. This study does, however, add to the existing literature on the positive effects perceived human presence in brands on brand attitude (Aggarwal & McGill, 2007; Morhart et al., 2015). The additional analyses showed that the tendency to anthropomorphize a brand positively and significantly affects brand attitude. While this study did not uncover how a human brand could be created through branding efforts, the results confirm the need for research in this area. As discussed above, the non-significant effects of experience and agency on brand anthropomorphism might be due to the manner in which these variables were manipulated. Based on the discussed literature on the relationship between the dimensions of mind perception and humanity (Gray & Wegner, 2012; Henkel et al., 2018; Rai & Diermeier, 2015; Zlotowski et al., 2014), it seems theoretically grounded to assume that perceived experience and agency in a brand affect consumers’ tendency to anthropomorphize that brand. Future studies should explore if a different, perhaps subtler or even more blatant, manipulation of agency and experience in a brand could induce the tendency to anthropomorphize. Earlier research showed how a brand could be

(30)

person (Aggarwal & McGill, 2011; Rai & Diermeier, 2015). As the present study argued, this approach is very difficult to operationalize through marketing efforts. However, future studies might consider combining this tactic with positioning a brand as having a highly experiential and agentic mental state. For instance, a commercial featuring a spokesperson who implicitly communicates having experiential and agentic capacities might induce people to imagine the brand as a person. Further research is needed to explore these possibilities.

Regarding future research into the manipulation of antecedents of brand anthropomorphism, this study also recommends taking consumers’ level of involvement with the brands’ products into account. If the brand offers

high-involvement products, brand attitude might be a better variable to measure the effects of brand anthropomorphism than purchase intention, as consumers tend to engage in an extensive information search before considering purchasing a high-involvement product (Clarke & Belk, 1978; Gu et al., 2012). Future studies should also measure participants’ levels of effectance- and sociality motivation when studying how brand anthropomorphism could be manipulated. These consumer characteristics could shed light on any differences found between participants, which are not due to the

manipulation.

The fact that the present study did not find a significant relationship between the dimensions of mind perception on the one hand, and brand anthropomorphism and/or purchase intention on the other, does not mean that no such relationship exists. As Lakens, Haans and Koole (2012) argued, the results of one study are not reliable; those of three similar studies are. Non-significant findings in one study do not necessarily mean that the expected effects will also be non-significant in a similar other study (Lakens et al., 2012). Applying the suggested recommendations, future

(31)

studies should similarly explore the antecedents and the effects of humanizing a brand to develop a more definite understanding of the phenomenon, an endeavour from which both the worlds of science and marketing could benefit.

(32)

6. References

Aaker, J. L. (1997). Dimensions of brand personality. Journal of marketing research, 34(3), 347-356.

Aaker, J. L., & Fournier, S. (1995). A brand as a character, a partner and a person: Three perspectives on the question of brand personality. Advances in Consumer Research, 22, 391–395.

Aaker, J. L., Garbinsky, E. N., & Vohs, K. D. (2012). Cultivating admiration in brands: Warmth, competence, and landing in the “golden quadrant”. Journal

of Consumer Psychology, 22(2), 191-194.

Aggarwal, Pankaj and Ann L. McGill (2007). “Is This Car Smiling at Me? Schema Congruity as Basis for Evaluating Anthropomorphized Products.” Journal of Consumer Research, 34(4), 468-479.

Aggarwal, P., & McGill, A. L. (2011). When brands seem human, do humans act like brands? Automatic behavioral priming effects of brand

anthropomorphism. Journal of Consumer Research, 39(2), 307-323.

Anthropomorphic. (n.d.). In Merriam Webster Online. Retrieved March 12, 2018, from https://www.merriam-webster.com/dictionary/anthropomorphic.

Bernritter, S. F., Verlegh, P. W., & Smit, E. G. (2016). Why nonprofits are easier to endorse on social media: The roles of warmth and brand symbolism. Journal

of Interactive Marketing, 33, 27-42.

Bos, M. (2018, May 16). Steeds minder aangifte van fietsendiefstal in Amsterdam. Het Parool. Retrieved from: https://www.parool.nl/amsterdam/steeds-minder-aangifte-van-fietsendiefstal-in-amsterdam~a4598312/

(33)

Brand. (n.d.) In Oxford Dictionaries. Retrieved March 25, 2018, from

https://en.oxforddictionaries.com/definition/brand

Burger, J. M., & Cooper, H. M. (1979). The desirability of control. Motivation and

emotion, 3(4), 381-393.

Chapman, K. J., & Aylesworth, A. (1999). Riding the coat-tails of a positive review: rave reviews and attitude transfer. Journal of Consumer Marketing, 16(5), 418-440.

Clarke, K., & Belk, R. W. (1979). The effects of product involvement and task

definition on anticipated consumer effort. Advances in Consumer Research, 5, 313-318.

Epley, N., Akalis, S., Waytz, A., & Cacioppo, J. T. (2008). Creating social connection through inferential reproduction: Loneliness and perceived agency in gadgets, gods, and greyhounds. Psychological science, 19(2), 114-120.

Epley, N., Waytz, A., & Cacioppo, J. T. (2007). On seeing human: a three-factor theory of anthropomorphism. Psychological review, 114(4), 864-886.

Epley, N., Waytz, A., Akalis, S., & Cacioppo, J. T. (2008). When we need a human: Motivational determinants of anthropomorphism. Social cognition, 26(2), 143- 155.

Eisenstadt, M. (1974). Xenophanes' proposed reform of Greek religion. Hermes, 102(2), 142-150.

Gray, H. M., Gray, K., & Wegner, D. M. (2007). Dimensions of mind perception.

Science, 315(5812), 619-619.

Gray, K., & Wegner, D. M. (2009). Moral typecasting: divergent perceptions of moral agents and moral patients. Journal of personality and social

(34)

Gray, K., & Wegner, D. M. (2011). Morality takes two: Dyadic morality and mind perception. The social psychology of morality: Exploring the causes of good

and evil, 96, 109-127.

Gray, K., & Wegner, D. M. (2012). Feeling robots and human zombies: Mind perception and the uncanny valley. Cognition, 125(1), 125-130.

Gu, B., Park, J., & Konana, P. (2012). The impact of external word-

of-mouth sources on retailer sales of high-involvement products. Information

Systems Research, 23(1), 182-196.

Hayes, A. F. (2013). Introduction to mediation, moderation, and conditional process

analysis: A regression-based approach. Guilford Press.

Henkel, A. P., Boegershausen, J., Hoegg, J., Aquino, K., & Lemmink, J. (2018). Discounting humanity: When consumers are price conscious, employees appear less human. Journal of Consumer Psychology, 28(2), 272-292.

Kervyn, N., Fiske, S. T., & Malone, C. (2012). Brands as intentional agents framework: How perceived intentions and ability can map brand perception. Journal of Consumer Psychology, 22(2), 166-176.

Lakens, D., Haans, A., & Koole, S. L. (2012). Één onderzoek is géén onderzoek: het belang van replicaties voor de psychologische wetenschap. De

Psycholoog, 47(9), 11-18.

Laurent, G., & Kapferer, J. N. (1985). Measuring consumer involvement profiles. Journal of marketing research, 22(2), 41-53.

Ling-Yee, L. (1997). Effect of collectivist orientation and ecological attitude on actual environmental commitment: The moderating role of consumer demographics and product involvement. Journal of international consumer marketing, 9(4), 31-53.

(35)

Miller, D. W., & Marks, L. J. (1996). The moderating effects of enduring involvement on imagery-evoking advertisements. American Marketing Association, 121, 121-128.

Morhart, F., Malär, L., Guèvremont, A., Girardin, F., & Grohmann, B. (2015). Brand authenticity: An integrative framework and measurement scale. Journal of

Consumer Psychology, 25(2), 200-218.

Portal, S., Abratt, R., & Bendixen, M. (2018). Building a human brand: Brand anthropomorphism unravelled. Business Horizons. 61(3), 367-374.

Puzakova, M., Kwak, H., & Andras, T. L. (2011). Brand anthropomorphization: a homocentric knowledge activation perspective. Advances in Consumer

Research, 36(1), 413-420.

Puzakova, M., Kwak, H., & Rocereto, J. (2009). Pushing the envelope of brand and personality: Antecedents and moderators of anthropomorphized brands. Advances in Consumer Research, 36, 413—420.

Rai, T. S., & Diermeier, D. (2015). Corporations are cyborgs: Organizations elicit anger but not sympathy when they can think but cannot feel. Organizational

Behavior and Human Decision Processes, 126, 18-26.

Rauschnabel, P. A., & Ahuvia, A. C. (2014). You’re so lovable: Anthropomorphism and brand love. Journal of Brand Management, 21(5), 372-395.

Sider, D. (1994). Xenophanes of Colophon, Fragments: A Text and Translation with a Commentary. American Journal of Philology, 115(3), 457-461.

Van Noort, G., & Willemsen, L. M. (2012). Online damage control: The effects of proactive versus reactive webcare interventions in consumer-generated and brand-generated platforms. Journal of Interactive Marketing, 26(3), 131-140.

(36)

Waytz, A., Epley, N., & Cacioppo, J. T. (2010). Social cognition unbound: Insights into anthropomorphism and dehumanization. Current Directions in

Psychological Science, 19(1), 58-62.

Waytz, A., Morewedge, C. K., Epley, N., Monteleone, G., Gao, J. H., & Cacioppo, J. T. (2010). Making sense by making sentient: effectance motivation increases anthropomorphism. Journal of personality and social psychology, 99(3), 410- 435.

Złotowski, J., Strasser, E., & Bartneck, C. (2014). Dimensions of

Anthropomorphism: from Humanness to Humanlikeness. Proceedings of the

2014 ACM/IEEE international conference on Human-robot interaction (pp.

(37)

7. Appendix

Experimental condition # 1: High experience, high agency

REID Cycles is a new brand that provides quality bikes at competitive prices. The owner, James Reid, created the brand in 2017 as a response to the high price tags on decent bikes. He wanted to make them affordable for everyone, and succeeded.

REID Cycles provides bikes for all types of riders. Their wide selection of bikes involves foldable bikes and e-bikes to commute to work with ease. When you’re in need of a bike to work out, REID can provide you with mountain bikes, fat-wheel bikes and all-road cross bikes. Is biking your all-time means of transportation? Then your ultimate option is probably one of the REID cruisers, single-speeds or vintage bikes. Check out all bikes on the REID Cycles website: www.reidcycles.com, or stop by one of the stores.

In their reviews about REID Cycles, consumers in general describe REID Cycles to be:

- Joyous - Ambitious

- A brand with character - Moral

- Good at planning - Communicative

(38)

Experimental condition # 2: High experience, low agency

- Joyous - Ambitious

- A brand with character

But they also indicate that there is room for improvement in the domains of: - Morality

- Planning

(39)

Experimental condition # 3: Low experience, high agency

- Moral

- Good at planning - Communicative

But they also indicate that there is room for improvement in the domains of: - Joy

- Ambition

(40)

Experimental condition # 4: Low experience, Low agency

In their reviews about REID Cycles, consumers in general describe that there is room for improvement in the domains of:

- Joy - Ambition

- Having a personal character - Morality

- Planning