1. Introduction
For the majority of existing businesses, progressive growth is stimulated by the successful marketing of products and/or services. Consumers are driven to purchase brands or products they like, and avoid those they don’t. As a result, marketers are constantly looking to modify consumer’s attitudes towards their brands or products. Currently, the most utilised method of changing consumer attitudes is evaluative conditioning [
1]. This process involves the repeated pairing of a conditioned stimulus (CS; e.g., a brand name) with an unconditioned stimulus (US; e.g., affective sound or picture). Eventually, the repeated pairings result in a transfer of affect from the US to the CS. A well-known example of evaluative conditioning in advertising, are the “Just Do It” commercials run by Nike. Throughout each commercial, the Nike brand name (CS) is repeatedly presented together with images of individuals working hard to achieve greatness (US). It is assumed that purchasing Nike products will allow the consumer to reach their potential more easily and more comfortably. As a result, liking of the brand is thought to increase. Although research suggests that evaluative conditioning can lead to an increase in brand awareness [
2,
3], sales [
4,
5], and brand positivity [
6], there is debate as to whether it can reliably change well established attitudes towards familiar brands [
6,
7].
Marketing literature constantly stresses the importance of brand identity on the consumer’s perception and decision making process [
8]. Extant research shows that consumers use brand names as an indicator of performance and quality [
9]. Such literature even goes so far as to posit that brand perceptions can influence sensory processing (e.g., taste) [
10,
11]. Findings such as these stress the importance of branding. In a market where competitors can easily copy product characteristics, it is essential that a strong brand identity and personality are formed in order to build brand equity [
12]. Understanding the importance of a strong brand, many companies have implemented conditioning paradigms in attempts to increase positive brand perceptions. The most common forms of conditioning involve businesses using heuristic stimuli such as music, celebrities, happy and fun scenes, and bright colors in order to persuade potential consumers that the product being offered is beneficial.
According to Stammerjohan et al. [
13], highly familiar brands are characterized by well-established, relatively stable attitudes that are somewhat resistant to the effects of advertising. In the majority of cases, research suggests that changing attitudes towards mature brands is unsuccessful [
14,
15,
16]. Latent inhibition (LI) is the process whereby people’s attitudes towards a stimulus are resistant to change as a result of previous exposure to that stimulus. Null findings as a result of latent inhibition are common within marketing literature [
7,
17,
18]. Such findings are intriguing given that brand name attitudes are entirely learned, highly semantic [
18], and can be derived and shaped without the individual actually having any direct experience with the brand [
19,
20]. This paper suggests that many of the abovementioned null findings occur as a result of an overreliance on explicit measures, the preferred option of many marketing firms.
Whilst the majority of marketing research sees the inclusion of explicit measures of attitudes, very few acknowledge the role of implicit processes [
21] and thus, fail to utilize implicit measures. Explicit attitudes are said to be contemplative, formulated through reasoning [
21], and are usually measured in terms of fully conscious responses such as ticking boxes, pressing buttons, or giving verbal responses. In contrast, implicit attitudes are associations that are automatically activated in the presence of relevant stimuli without any conscious awareness of evaluation [
22]. These attitude aspects require more sophisticated methodological approaches (see later). The distinction between these two types of attitudes arose after constant reports suggested that people either cannot, or do not want to fully explain their preferences [
23,
24].
Attitude research that is seen to compare implicit and explicit attitudes constantly reveal discrepant findings [
25,
26,
27,
28]. For instance, Walla et al. found that when eating ice cream, chocolate or yoghurt, although participants had no stated preference for particular food items, implicit responses showed that ice cream was preferred [
28]. Similar findings were presented by Grahl et al., who reported that specific bottle shapes can elicit a non-conscious change whilst explicit ratings remain constant [
26]. More specifically, although both males and females stated that one particular bottle (out of three) was the least attractive, physiological responses (startle reflex) suggested that this bottle elicited a negative affect in only male brains.
Neuroscience is able to offer an alternative explanation to latent inhibition regarding the discrepant findings between implicit and explicit attitudes. According to Walla, Brenner and Koller [
29] contemplation and reasoning inevitably give rise to cognitive pollution. In more simplistic terms, the more conscious/cognitive processing that takes place during the evaluation of a stimulus, the less reliable/more polluted the response becomes, with respect to raw affective processing [
30]. Given these discrepancies, it is proposed that cognitive pollution is an alternative explanation to the null findings where latent inhibition has regularly been presented in psychological studies.
Given all these findings, it is imperative to consider alternative means of assessing brand attitude that complete traditional approaches. As a result, the rationale for the present study is to assess well-established liked and disliked brand attitudes using a multidimensional approach. In doing so, the sensitivity of implicit and explicit measures will be utilized to detect baseline attitude in addition to potential changes in attitudes as a result of evaluative conditioning.
Within consumer contexts, of the implicit measures currently utilized, the Implicit Association Test (IAT) is arguably the most popular measure of consumer attitudes and whilst the use of the IAT has primarily been within psychological studies, its application within consumer contexts is growing [
31]. The IAT assumes that there is an associated network whereby concepts of memories, attitudes and valence are all intrinsically linked [
32,
33]. It is inferred that the faster a participant is able to respond to a stimulus (or stimulus pair), the stronger the implicit association. Whilst the IAT seems to be a promising measure of implicit attitudes, it has been met with substantial criticisms regarding its susceptibility to cognitive processes [
34,
35]. In sum, such criticisms have raised serious concerns regarding the IATs legitimacy of an exclusively implicit measure of attitudes. Only recently, a four-step model has been published to enhance the IAT [
36], which might help future studies, but still the IAT should be used with great caution.
Electroencephalography (EEG) is an emerging technique used to assess the affective components related to implicit attitude. It has been proposed that EEG is capable of distinguishing between positive (approach) and negative (avoidance) affect [
37,
38]. The asymmetry model proposed by Davidson et al. [
39] suggests that greater relative left frontal EEG activity is associated with the processing of positive/approach affects, whereas greater relative right frontal EEG activity is associated with the processing of negative/avoidance affects. Within marketing contexts, Ravaja et al found that a reduction in the price of certain products elicited relatively greater left frontal activation. In addition, the occurrence of greater left frontal activation was associated with an increased likelihood for participants to purchase goods [
39].
With regard to EEG, asymmetry effects aren’t the only means of investigating attitudes. Of interest to the present study is one of the most empirically valid EEG approach as an index of motivation and affect, the late positive potential (LPP) [
40]. The LPP is a late, positive event related potential (ERP) component that becomes evident at around 300ms following stimulus onset and can exceed 5000ms. The LPP has been extensively used within the literature and as a result, has received psychometric endorsement which revealed that the LPP demonstrated good to excellent reliability as a measure of emotion/affective processing [
40]. Stimuli that are seen to be more affective and thus, more motivationally relevant are said to evoke the largest LPPs [
40]. In addition, regardless of valence, LPP effects are reported to be greater in the right hemisphere than the left [
41].
In the present study a multi-session, evaluative conditioning paradigm (varying numbers of conditioning rounds) was utilized with the aim of modifying participant’s attitudes towards their most liked and disliked brand names (established brand attitudes). Participants were initially presented with well-established brand names and their baseline attitudes were recorded via self-report. Subsequent visits to the laboratory aimed to modify participant’s brand attitudes, and to investigate subsequent changes using both subjective (self-report) and objective measures (IAT and EEG). We hypothesized that the different measures are differently sensitive to evaluative conditioning effects on brand attitude, and thus provide a more detailed understanding of brand attitude in general and of how to best detect conditioning effects in general.
5. Conclusions
The current research adds to the vast amount of research that promotes the inclusion of implicit measures to assess brand attitude with a strong emphasis on EEG. Had EEG not been included in the present study, the effects of conditioning would have been completely absent. Through the inclusion of recording brain potential changes, we propose that the effects of evaluative conditioning occur at different levels of information processing and it appears as though traditional approaches lack the required sensitivity to competently and accurately assess consumer attitudes to this extent.
In an applied sense, our results reveal exciting findings that suggest advertisers should not run the same ad for too long. The ERPs presented within the current paper indicated that although a significant degree of evaluative conditioning occurred after all three CS:US exposures, conditioning effects were most prominent after only a single pairing. This finding supports those presented by several authors, who report that a single conditioning trial will elicit a large effect, whereas subsequent conditioning trials will elicit smaller effects until a maximum is reached [
6,
18,
62]. Given that the majority of the abovementioned authors relied on fictitious brands, more research in this area needs to be considered before any more robust conclusions can be drawn.
Not only does extant literature suggest that conditioning effects diminish, but it has also been reported that over exposure to a single ad can in fact result in negative conditioning effects [
63]. Furthermore, Sweldens, van Osselaer and Janiszewski report that different conditioning procedures might encourage different learning processes [
20]. With this in mind, businesses may gain benefit from utilising different mediums when creating ad campaigns. Although this paper only focused on sounds as the conditioning medium, visual stimuli [
64,
65], olfactory stimuli [
66], and gustatory stimuli [
67] have all been used to condition positive affect. In principle, these current results emphasise the need of marketers to move away from traditional self-report measures and begin to address the issue explicit types of measures inherently possess.
In addition to the abovementioned findings, our results promote the idea that liked brands may be more resistant to the effects of evaluative conditioning. Although not directly linked to our study, the use of shock advertising may provide some support for this theory. Unlike typical advertising techniques, shock advertising aims at deliberately startling and offending its audience [
68,
69]. Although one would assume such advertising techniques would deter consumers from engaging with products, research has suggested that shock advertising can have positive effects on attention, memory, and behaviour [
70]. Given our ERP findings that suggest the robust nature of well-established liked brands, it is suggested that businesses possessing such brands may be able to employ more controversial forms of marketing techniques without risking damage to their brands.
In sum, further research within this area should focus on existing brands that are well-known to participants. This method will allow for the most comprehensive understanding of how brand attitudes are formed and modified. It may also be important to note that businesses may benefit from not rely on a single ad campaign for extended periods, but instead change campaign and consider focusing on different mediums by which to present their brands.