What counts as reward differs substantially depending on individual preferences. Branding can elicit robust differences in preferences for consumer products despite their highly similar appearance and may therefore provide an ideal example of truly subjective preference in that it is largely independent of objective stimulus characteristics. Indeed, branding can be viewed as the assignment of value and meaning to often quite mundane and interchangeable products [e.g., [1
]]. Current theories of reward processing have paid increasing attention to such cultural influences on choice behavior. Our aim is to expand these theories by examining the modulatory impact of subjective brand preferences on neural activity. Furthermore, the external validity of our findings is enhanced by the high relevance of brands in everyday life.
Most previous research on the neural representation of reward has focused on the manipulation of reward according to an objectively quantifiable scale without therefore having to consider individual differences in preferences. There is no doubt that people have a general preference for larger rather than smaller amounts of money. Many human imaging studies have made explicit use of various degrees of monetary incentive value as means to manipulate reward intensity, and have reported several neural regions that adapt their activity according to the changes in reward intensity. O'Doherty et al., for example, reported stronger recruitment of the medial orbitofrontal cortex (OFC) upon gaining higher compared with smaller amounts of money in a two-alternative choice task [2
]. In another study, Breiter et al. identified the sublenticular extended amygdala (SLEA), the Nucleus accumbens (NAcc) and the hypothalamus as showing a scaled neural response in correspondence with the value of received monetary gain [3
]. In addition, neural activation patterns in the SLEA and in the OFC reflected the value of the potential rewards in the period in which participants anticipated the outcome. A later study by Knutson et al. found a dissociation of neural circuits involved in different aspects of reward: While the ventral striatum (incl. NAcc) was strongly active during the anticipation of monetary reward, the mesial prefrontal cortex (PFC), the parietal cortex and the posterior cingulum were active following feedback to the participants of having successfully obtained a reward [4
]. During reward anticipation, the NAcc activity was positively correlated with the magnitude of the monetary reward.
However, even the rewarding value of money may be influenced by context effects. Counterfactual reasoning, for example, refers to people's tendency to compare their choices with the outcome of alternative courses of action. Winning Sfr. 5 in gambling most certainly evokes some degree of satisfaction, whereas winning Sfr. 5 while knowing that one could have won Sfr. 10 had one chosen differently evokes regret or disappointment [5
]. The neural underpinnings of this effect have recently been investigated [6
]. The concept of delayed discounting, concerning the point in time when a reward is delivered, is a further example of the effect of contextual information on the perceived value of a certain reward [9
]. Animal studies strengthen this finding [13
]. Counterfactual reasoning and delayed discounting, however, reflect general effects that hold across subjects, meaning that the contextual information has a similar impact on each subject.
Food, on the other hand, might be a universal primary reinforcer but people greatly differ in their taste preferences. Similarly, interindividual variance in the attractiveness of a reward also characterizes branded consumer goods. The neural representation of brand preferences has recently received considerable attention. McClure, for instance, reported that participants show stronger hemodynamic responses in reward-related brain regions (dorsolateral prefrontal cortex (DLPFC), hippocampus, midbrain) when receiving a small amount of a soft drink pre-cued by a picture of a Coca-Cola can rather than by a circle of light or a picture of a Pepsi can [15
]. Schaefer and Rotte reported stronger activity in the medial prefrontal cortex (MPFC) and precuneus when participants were presented with logos of luxury and sports car brands compared with pragmatic, more economic car brands [16
]. In a study by Deppe et al., participants were asked to imagine choosing between pairs of brands [17
]. The authors reported reduced neural activity in regions associated with working memory and reasoning, and increased neural activity in regions related to emotion processing when presenting the most popular brand in terms of the market share as compared with less popular brands. Based on their findings, the authors postulate a winner-take-all
effect of a person's favorite brand on neural activation, an effect that would partially contradict the graded response to different amounts of monetary rewards.
To our knowledge, however, none of the available studies on brand preferences used participants' stated preferences as a means to specifically varying the subjectively perceived attractiveness of the presented brands. Furthermore, previous brain imaging studies of brand preferences did not clearly differentiate between the period of reward anticipation and that of reward receipt. This distinction between anticipatory (wanting
) and evaluative (liking
) components has already been proposed by Berridge on the basis of animal studies [18
], and evidence from human studies using monetary reward supports this concept [3
]. Thus, the available studies on brand preferences may have confounded motivational with evaluative components of reward processing. Finally, the use of more than two preference categories is a necessary precondition to unequivocally determining any modulatory influence of brand preference on neural activity patterns. It may well be that, similar to monetary rewards, brand-associated neural activity increases monotonically with the strength of the individual preference for a particular brand.
To address these issues, we developed a wheel-of-fortune game that allowed for the differentiation between an anticipation period (spinning of the wheel; wishing for a positive outcome) and an outcome period (processing the game outcome). Chocolate bars of three different brands could be won. By using chocolate bars as rewarding stimuli we introduced a product category with relatively homogeneous pricing to avoid the coupling of reward intensity with monetary value, which may be neurally processed in a different way. Established market research instruments were used prior to the fMRI experiment to determine participants' individual brand preferences. Based on the results of these instruments, brands that differed in subjective attractiveness were selected individually for each participant and used as stimuli in an fMRI experiment. During the experiment, brands were represented by their logos. However, real chocolate bars were given to the participants after the experiment.
The primary aim of our study was to explore whether there are neural structures that modulate their activation according to the subjective preferences for the chocolate bar brands that the participants played for (e.g., higher activity in case of more preferred compared to less preferred chocolate brands). Additionally, the design allowed for investigating the suggested dissociation between an anticipatory reward component (game outcome unknown, wanting) and an evaluative reward component (evaluation of game outcome, liking). This dissociation is important for understanding buying behavior, since anticipation and evaluation are associated with different facets of a brand: (a) motivational, action-relevant characteristics, and (b) emotional or cognitive evaluative aspects.