Tipping in restaurants is an everyday-life situation, which most people are familiar with. Tipping can be defined as an economic activity – an exchange between the consumer and the service provider by which consumers show their gratitude and grant the server an additional payment for the service they are providing (Lynn & Grassman, 1990). Although tipping is a global phenomenon, it is more common in some countries such as the United States, where a substantial amount of the servers’ incomes are relying on tips. Typically the tip is calculated as a percentage of the total bill, and nowadays in the United States, a 15%-20% tip is expected in “sit-down restaurants”, although in practice this may slightly vary depending on the satisfaction of the consumer (Lynn, 2006). As much as tipping is a common practice, the behavioral aspect of it causes a lot of controversy. One of the biggest unresolved questions by economists regarding tipping is the fact that a person willingly chooses to pay extra without being obliged to do so. This, in turn, violates the standard economic model, as the person is deriving no utility from the practice (Azar, 2005). In addition, the economist’s perspective claims that consumers should always act rationally regarding economic decision-making, cf. the theory of rational choice. However, since its assumptions are being violated and maximum utility is not being obtained, it can be argued that consumer’s decision-making regarding tipping is irrational, which often poses challenges for economic models in making correct predictions for such behavior. Despite the limitations of the economic view in explaining and predicting tipping behavior, the behavior itself has significant economic implications. One of such implications concerns the minimum wage policies for tipped employees, which deviates from other employees (Azar, 2010). If, for example, tipped workers would receive a relatively high minimum wage then their total income may exceed well over the average total income of that of the non-serving staff members when tips are included, creating an unequal distribution (Azar, 2005). Apart from having an impact on the minimum wage policies and laws, tips also have implication for compensation schemes, management strategies, taxes, and the monitoring of employees and restaurant quality (Azar, 2005). These economic consequences highlight the importance of understanding tipping behavior. In doing so, theoretical frameworks can be determined to make predictions by which economic effects can be exploited. In an attempt to understand the underlying motivations that govern tipping behavior, one might tend towards the domains of psychology and neuroscience. This paper tries to answer the question as to why customers engage in tipping behavior while those individuals are fundamentally not obligated to do so. The remainder of this paper will explore two possible motivational explanations for this phenomenon, including the review of relevant psychological frameworks and neuroscientific evidence to substantiate its claims. Namely, tipping to avoid negative response and tipping as an altruistic act. In addition, we will also present a research proposal for further investigation and finalize with the discussion. The main purpose of this paper is to extend on the existing literature on tipping behavior by providing possibly explanations that govern the underlying motivation, in an attempt to contribute to the development of a more complete theoretical framework regarding this topic.
Tipping to avoid guilt and embarrassment
Human behavior is strongly influenced by social context. Socially acceptable behavior is judged by the emotions that arise from the actions the individuals make. These socially bounded emotions are called the self-conscious emotions, because the individual experiences them in response to the appraisal of one’s own behavior perceived by others (Tracy & Robbins, 2007). If the social group evaluates the individual's action as counternormative, the individual is likely to experience negative self-conscious emotions such as guilt and embarrassment. These emotions are also known as the moral emotions, which inhibit behavior against social standards (Azar, 2005; Takahashi, Yahata, Koeda, Matsuda, Asai, & Okubo, 2004). Leaving a tip can be seen as a behavior that requires judgment about the social context. In countries where tipping is considered a social norm, leaving the server without a tip violates the said norm. Thus, refraining from tipping is expected to cause a negative response in the server and feelings of guilt and embarrassment in the consumer (Conlin, Lynn, & O’Donoghue, 2003). It can therefore be assumed that individuals will tip to avoid the negative emotional response of not tipping.
There exists studies that supports this idea. Early research has demonstrated an inverse power function between the amount of tip and the dining group size – when the group size increases, the tipping amount decreases (Freeman, Walker, Borden, & Latane, 1975). This study collected data from 396 groups of restaurant diners whereby the group size ranged from one to seventeen persons. They found an inverse power function with individuals tipping an average of 20% of the final bill whereas groups greater than six persons tipped only 13% on average. The inverse power function is closely related to the theory of diffusion of responsibility, which posits that an individual is less likely to take responsibility for action in the presence of others (Latané & Darley, 1970). To expand on this interpretation it can be presumed that the perception of anonymity also plays a role in addition to diffusion of responsibility. Meaning that the sense of diffusion of responsibility and anonymity during a dinner in presence of a larger group will alleviate the social pressure. This, in turn, may give rise for the motivation to tip a substantially lesser amount or to not tip at all without experiencing disutility from unfair behavior, that is, feelings of guilt and embarrassment caused by refraining from tipping. In contrast, individual consumers dining alone or in smaller groups are exposed to a higher social pressure for the internalized norm and conform to avoid the negative emotions associated with not tipping.
A study by Azar (2010) provided evidence to support the idea that avoidance of negative emotions plays a significant role in understanding what motivates consumers to tip. In this study, participants from the U.S and Israel had to fill out a questionnaire regarding reasons to tip in a restaurant. Of the American respondents, 60% indicated that their reason to tip was to avoid the feeling of guilt as a results of not tipping. Similarly, 44% mentioned the avoidance of embarrassment as one of their reasons. Social norm was mentioned as the most common (84%) reason for Americans to tip. As established before, avoidance of guilt and embarrassment are directly linked to the social norm aspect of tipping. Thus, it can be inferred from these results that majority of Americans tip to avoid the negative emotional response when refraining from tipping.
The findings of survey studies offer crucial psychological evidence in explaining tipping as a behavior to avoid embarrassment and guilt caused by violating the social norm of tipping. However, for better understanding, it is necessary to investigate the topic in terms of neuroscience. Brain areas that could be involved in the guilt-embarrassment-avoidance explanation of tipping are the medial prefrontal cortex (mPFC) and the orbitofrontal cortex (OFC). Research exists about the key role of mPFC in social cognition, that is, an umbrella term for processes such as understanding mental states of others and self, social scripts and interpersonal norms (Van Overwalle, 2009). Similarly, one of the main functions of the OFC is integrating information about reward, punishment and current emotional states into decision-making. This, in turn, is needed for correctly interpreting complex social and emotional behavior (Kringelbach & Rolls, 2004). As there are no neuroscientific studies specifically focused on the link between tipping behavior and the avoidance of negative self-conscious emotions, we base our assumptions on evidence from empirical studies that applied functional magnetic resonance imaging (fMRI) to investigated the emotions of guilt and embarrassment, and their link to social norm violation.
In an fMRI-study by Berthoz, Armony, Blair and Dolan (2002), the subjects were instructed to read sentences while in the scanner, that included normal social situations, embarrassing situations and situations involving violation of a social norm. The brain areas activated during sentences regarding intentional and unintentional social norm violations were the mPFC and the orbitofrontal cortex. In addition, the embarrassing situation condition yielded similar activation in the brain as the social norm violation condition. These findings suggest that the underlying brain mechanism would be the same for social norm violation and the feeling of embarrassment. This connection links to the possible thought process behind tipping, as refraining from tipping represents the social norm violation and embarrassment the outcome of such behavior.
Relating to above-mentioned, a study by Takahashi et al. (2004) yielded similar findings. Instead of concentrating on social norm violation, the aim of their research was to localize emotions associated with social cognition, namely guilt and embarrassment. The participants had to read sentences containing neutral, embarrassing and guilty material while in the fMRI-scanner. The results showed that viewing embarrassing and guilty material activated the mPFC. This finding is in line with the results of Berthoz et al. (2002), although no activation was found in the OFC. Moreover, an important aspect of both of these studies is that only the mere thoughts of embarrassing and guilty situations were strong enough to evoke activation in the brain areas that usually respond to the actual individual experience. In tipping context, we presume that an individual, who is making a decision about tipping will go through a conscious or unconscious evaluation process about the outcomes of their decision. If this process truly involves evaluation of the outcome of not tipping, which is the negative response of guilt and embarrassment, we can then detect activation in the mPFC and the OFC of the individual’s brain during the decision-making.
To sum up, tipping can be seen as a type of internalized social norm that we are likely to conform to. But what exactly motivates the conformity to the tipping norm is the fear of the outcome of not tipping; the guilt and embarrassment as the result of the norm violation. As mentioned by one of the studies, the majority of Americans consider the avoidance of these emotions as their main reason to tip in a restaurant. However, avoidance of the negative response is expected to be more prominent with individuals or small dining groups, as with larger groups the diffusion of responsibility and the sense of anonymity would diminish the effect of social norm violation. Furthermore, brain imaging studies not only promote the connection between embarrassment, guilt and social norm violation, but also the idea of an evaluation process of different tipping outcomes, which will eventually lead to the actual tipping decision.
Tipping as an altruistic act Another possible explanation for the motivation to engage in tipping behavior is that it might be an act of empathic concern. In this case, the customer genuinely values the welfare of others and the norm of tipping (or social norms in general), making the act of tipping an altruistic behavior. The conceptual model of Batson (1990) proclaims that altruism is a prosocial behavior that occurs in direct response to the emotional experience of empathy (McCamant, 2006). Hereby can altruism be described as the nature of behavior that is not apparently self-centered, while empathy refers to the intrinsic emotional response that allows one person to identify with the psychology of another (McCamant, 2006). Thus, Batson coined the empathy-altruism hypothesis (Batson, 1990). To illustrate the supportive notion of this hypothesis in terms of tipping behavior, consider the following example: a customer visiting a restaurant during busy working hours might empathize with a laborious server and consequently tip a certain amount with the motivational goal to increase the servers’ welfare. In short, in scope of this paper, altruism can be interpreted in terms of a behavioral definition as the costly act to confer economic and psychological benefits on other individuals (Fehr & Fishbacher, 2003). In addition to increasing other individuals’ welfare, the altruistic act may also derive a intrinsic reward in the form of personal gratification in the actor himself (Persson & Kajonius, 2016). An extensive amount of psychological research attribute this gain of intrinsic rewards as a self-centered motivation, which makes an altruistic act not truly altruistic in itself (Andreoni, 1990; Avolio & Locke, 2003; Maner & gailliot, 2007). However, we believe that altruistic behavior is evoked by the emphatic desire to increase someone’s welfare, without the conscious expectation of a reciprocal relationship whereby the other individual is obligated to provide a service in return. The intrinsic reward of an altruistic deed is thus mediated by others welfare and therefore inherently altruistic (Warneken & Tomasello, 2009). Research conducted by Batson et al. (1991) supports this view. During their research, three experiments with 252 college students where conducted to test whether empathy evokes altruistic motivation (empathy-altruism hypothesis) or egoistic motivation (empathy-joy hypothesis, counterpart of the former hypothesis). Each experiment gave the participant the opportunity to help a victim or needy person in order to increase their initial welfare, including the choice to get update information about their condition afterwards. All of the experiments results were patterned according to the empathy-altruism hypothesis. For someone to sympathize with another individual it is required that one understands the mental states of others. Several of such mental states include beliefs, intentions, emotions and desires. If the motivation to tip is an inherently altruistic act, then the tipper must be well aware of the mental states of others. This idea is closely related to the theory of mind (ToM), which is the ability to attribute mental states to oneself and others (Premack & Woodruff, 1978). ToM plays an important role in social interactions, social behavior and decision making, as it is utilized to analyze, infer and judge others’ behavior (Gweon & Saxe, 2013). A customer at a restaurant may thus acknowledge the states of mind of the restaurant employees, inferring that those individuals are also emotionally invested in social norms, such as tipping, and would highly appreciate it to receive a tip as this act would increase both their emotional and economical welfare. Therefore, the motivation to tip may, again, be an altruistic act. Much of the research in supporting ToM is conducted on young children and infants using the false-belief paradigm, as they still need to develop the understanding of emotions and belief systems (Wellman, 1992; Wellman, Cross & Watson, 2003; Perner & Lang, 1999) This, in turn, makes it easier to investigate ToM as the young children show very distinct social behavioral patterns, since they can not yet correctly infer the mental states of other with their undeveloped emotions and belief systems. However, to provide more credible evidence with respect to the relationship of ToM in tipping behavior in adults, neuroscience could possibly provide an answer. Neuroimaging studies have reported two distinct brain regions possibly involved in altruistic behaviour; the posterior superior temporal cortex (pSTC) and the ventral striatum. Specifically the right pSTC promotes the perception of agency, a significant function underlying capacities related to altruistic behaviour, such as empathy and ToM (Tankersley, Stowe, & Huettel, 2007). The striatum, on the other hand, is a critical component of the brain’s reward system (Harbaugh, Mayr, & Burghart, 2007).
Tankersley et al. (2007) tested the role of pSTC in altruistic behaviour by running two fMRI-experiments in which the participants had to watch or play a simple reaction-time game to win money for themselves or for a charity. The increased neural response in the right pSTC was found in both studies to be positively correlated with high scores of self-reported altruistic tendency of action perception in the game. Additionally, the researchers showed in the second experiment that pSTC activity was exclusively and significantly related to high altruism ratings, in contrast to other measures such as personality and impulsivity. Specifically, the right temporo-parietal junction (rTPJ) seems highly selectively involved in representing the beliefs of others, which reflects ToM (Saxe & Powell, 2006). The rTPJ is part of the right pSTC, making it also closely related to altruistic and empathic behavior. Saxe & Kanwisher (2003) found that rTPJ activation was higher when participants red stories about the character’s state of mind compared to neutral stories, concluding that the rTPJ has a significant role in the understanding and reasoning about the content of other individuals mental states. Also, false belief tasks in adults show the involvement of the rTPJ compared to control tasks, meaning that the rTPJ has an important role in ToM during adult life, instead of merely being a control system for the emergence of ToM in children, which was first hypothesized (Saxe, Schultz & Jiang, 2007). Moreover, an fMRI-study by Harbaugh, Mayr and Burghart (2007) focused on investigating the role of brain’s reward system on ‘pure altruism’ in charitable donations. The subjects played a dictator game, in which they had to donate money to a local food bank. The donation was either done voluntarily or as a mandatory tax-transfer. The participants could also rate whether they accepted or rejected the donation. In addition, the subjects also received money transfers themselves. The researchers showed that the acceptance rate of both types of monetary payoffs to the charity were positively correlated with an increased activity in the ventral striatum. The ventral striatum was also activated when the subjects themselves received money. These results were explained by the idea that people who valued the charity’s payoffs in the same way as their own, were more likely to accept the donation. Thus, altruistic actions are experienced rewarding to oneself in the same way as if one would receive something rewarding. The common neural activation of striatum in both cases supports this claim. In summary, the motivation to engage in tipping behavior may be an altruistic act in an attempt to increase an individual’s psychological and economical welfare after empathizing with the concerning individual. In doing so, the tipper himself experience a sense of personal gratification. Neurologically this could correlate with activation in the pSTC, specifically the rTPJ, which reflects altruism, empathy and ToM, and activation in the striatum, which reflects intrinsic reward after increasing someone’s welfare. Taking into account the psychological and neuroscientific findings in light of the two possible explanations regarding the motivation to tip, we have developed two hypotheses. First, we propose that tipping can be explained by people’s tendency to avoid the emotions of embarrassment and guilt that would result from violating the internalized social norm of tipping. The mental representation of the violation and its outcomes are expected to have a positive relationship between the increased activity in the mPFC and the OFC. Second, tipping can be explained by people’s altruistic tendencies to increase others welfare, consequently, this will evoke an intrinsic reward. This may have a positive relationship with activity in the striatum, pSTC and rTPJ.
Method
This experiment should involve 45 U.S. citizens aged 25 to 50, with a middle-class background, and who are familiar with being customers in a sit-down restaurant. Only participants without prior neurological disorders or physical conditions requiring metal implants can be admitted to the study. Informed written consent from all participants will be obtained beforehand.
The data will be acquired with an fMRI-scanner. We will be measuring the blood-oxygen-level-dependent (BOLD) signal in the participant’s brain while they perform a decision-making task related to restaurant tipping. Through this, we aim to locate which brain areas are active during the decision-making process. Before the participant enters the testing area, he or she will be given a short briefing regarding safety regulations. Then the participant will be comfortably positioned in the fMRI-scanner and few test images are taken to ensure a good quality of the scans. The stimuli will be displayed on a monitor, which the participant can see via 45° angled mirror positioned above the head coil of the scanner. The participants also receive a simple remote control with four buttons (A, B, C and D), which enables them to communicate during the experiment.
Before entering the fMRI, participants will be explained that they will be presented different choice-problems regarding the restaurant bill they receive from the server. The participant is instructed to make his/her decision in a way they would during a similar real-life scenario. The actual task inside the fMRI begins with a short video (1-2 minutes) depicting an average service situation in a restaurant. The aim of this video is to prime a realistic context for the decision-making situation regarding tipping. The participant sees the server from a first-person perspective, taking orders and serving food and drinks in a busy restaurant. At the end of the video, the server approaches the participant and gives the bill. After the video ends, there appears a white screen for 2 seconds before the trials start.
There will be 20 trials, each having a different total bill size amount. The trials start with the bill presented to the participant for 5s. After that, there appears a choice-problem with four options, A, B, C and D, for how much the participant is willing to tip. Each option is based on a fixed percentage of the total bill size, 5%, 15%, 20% and 30% respectively. The participant has 10s to indicate their answer by pressing the corresponding button on the remote control, with a 4s white screen in between trials. Total bill size will vary for each trial, therefore the options in each trial will show different tipping amounts, depending on the given bill size, while each option reflects the same percentages over all trials. The fixed percentages are chosen for a specific reason. Research shows that the average tipping amount in the U.S. is roughly 15-20% (Maynard & Mupandawana, 2009; Lynn, 2006; Lynn & McCall, 2016). Thus, option B reflects the lower end of the average tipping amount, while option C reflects the higher end. On the other hand, option A reflects 10% below the lower end; and option D reflects 10% above the higher end. This way patterns could be observed for people with different motivation. People who essentially don’t want to tip have the tendency to tip a low amount, up to the lower end of the average (option A and B). In contrast, people who sincerely do wish to tip will be inclined to tip at the higher end of the average or an amount even higher (option C and D). In Addition, the BOLD response in the brain will be measured during the entire decision-making task. Statistical analyses include an ANOVA test and correlations. The F-statistic and corresponding p-value should be derived from comparing differences in the decision-making task by means of ANOVA, to indicate any significant differences among the options. Furthermore, correlations between the decision-making task and the fMRI results should indicate if there are associations between brain activation and specific behavioral patterns in the decision-making task.
Expected Results We expect participant that tip to avoid negative response to have a tendency for options A and B. In contrast, we expect participant that tip for altruistic reasons to have a tendency towards options C and D. Due to this difference we expect to obtain a significant p-value from the ANOVA analyses that explains the two different motivations to tip by showing dissociation (fig. 1). Moreover, we expect brain activity in the pSTC, rTPJ and pSTC to correlate positively with tipping amount; the higher the tip, the more activity in the corresponding regions. We also expect the reverse to be true for brain activity in the mPFC and the OFC and tipping amounts, namely a negative correlate. We expect the correlation r to be semi-strong with roughly R2 = .3 for the explained variance (fig. 2).
Figure 1. An example of the expected distribution for the decision-making task. As shown, participants that want to avoid negative response tip from the lower end of the average determined tipping amount and below, for the altruistic participant the reverse is true, showing clear dissociation. Note: due to the exemplary nature of this image it is not presented according to APA format, nor are there values being depicted.
Figure 2. An example of the expected correlation for tipping amount and the rTPJ. Results for the pSTC and striatum are expected to show a similar correlation, while participant motivated to avoid negative response should show a negative correlate. Note: due to the exemplary nature of this image it is not presented according to APA format and values along the axis do not reflect the accurate numbers derived from actual data.
Discussion
In this research proposal, we investigated the motivation behind tipping behavior in restaurants. We proposed two possible explanations for this behavior. First, we hypothesized that tipping could be explained by people’s tendency to avoid the feelings of guilt and embarrassment that would result from violating the internalized social norm of tipping. Specifically, we argued that brain activity in mPFC and OFC would correlate with these behavioral patterns. The mPFC and OFC activation correlated with understanding social behavior in terms of social norms and localizing emotions related to social cognition, namely guilt and embarrassment. Furthermore, we hypothesized that tipping can be seen as an altruistic tendency to increase others welfare and, as a consequence, the intrinsic reward of the tipper himself. We argued that this behavioral pattern activates the striatum, rTPJ and pSTC. Striatum activity would reflect intrinsic reward, while the rTPJ and pSTC would be involved in empathizing with other individuals and altruism. To test these hypotheses we proposed an fMRI experiment. More elaborate interpretations of the hypotheses depend on the results of the actual experiment. However, we will provide a concise interpretation in light of our expected results. According to the hypothesis of avoiding negative response we expect participants to opt for choices A and B most of the time during the experiment. Customers who simply conform to the internalized norm of tipping in an attempt to avoid negative responses do essentially not want to tip in the first place. Therefore they will have the tendency to tip to a substantially lesser degree than the average tipping amount of roughly 15-20%, or at the lower end of the average; their tipping amount is greater than zero so they are not being perceived as someone that violates the norm. Meanwhile, the amount is also lower than the average determined tipping amount to avoid further unwanted monetary disutility. Moreover, the larger the potential dining group with those individuals, the lesser the amount they will tip. This is due to diffusion of responsibility and the perception of anonymity; disengagement from responsibility and the social norm will be seen as an opportunity to avoid disutility from tipping without experiencing feelings of guilt and embarrassed to the fullest extent. In essence, tipping to avoid negative response can be seen as a selfish act.
Regarding the altruism hypothesis, the reversed pattern can be expected to be shown in the results. Namely the tendency to opt for choices C and D in most cases. This pattern can be observed because altruistic customers are sincerely engaged in the social norm and truly wish to increase a server's economic and psychological welfare. Therefore, the altruistic customer is more likely to tip an amount approximately to the tipping average of 15-20% and is also much more inclined to tip an amount greater than this average. Also, the altruistic customer will derive a sense of intrinsic reward for increasing a server’s welfare, which is expected to correspond with the striatum. According to theory of mind the altruistic customer sympathizes with the server, which should correlate with the rTPJ, which in turn is part of the pSTC. Consequently, the motivation to tip according to altruistic tendencies can be seen as a selfless act. In conclusion, the interpretations regarding the hypotheses, as discussed above, should follow after obtaining results similar to our expectations. Results showing dissociative patterns, in combination with the theoretical frameworks discussed in the review, serves as evidence for our claims that the motivation to engage in tipping behavior in sit-down restaurants can be explained in terms of avoidance of negative responses, which is essentially a selfish act, and in terms of altruism, which is essentially a selfless act. Also, that these acts are being modulated by distinct neurological mechanisms, as discussed.
Limitations
Some degree of caution should be taken into account when interpreting these expectations. Although the definition of altruism has been clearly defined in the context of tipping behavior during the review, more interpretations of this complex emotion are possible. For example, it could indeed be argued that the ‘altruistic’ act of tipping is merely driven by the intrinsic reward one might experience, making it essentially a selfish act instead. In that case, the underlying motivations of both hypotheses are similar, only to be achieved through different neurological mechanisms. The possibility for different interpretations for the concept of altruism could signify that the explanations for the underlying motivation to tip are not mutually exclusive, regardless of the experimental outcomes.
Moreover, even when the hypotheses hold true, many other factors are involved in actual tipping. For example, group size, bill size, restaurant and service quality, and server characteristics. These different factors may alter the underlying motivation to tip accordingly. To determine a complete theoretical framework regarding tipping behavior, factors like these must be taken into consideration. It is also important to mention that participants are possibly affected by the causes and chance bias. With the expectation that, depending on the motivation, the participants will be inclined to either choose between A/B or C/D, they might get suspicious after constantly selecting the same answers during the twenty trials. Therefore they might start choosing different answers just to ‘mix things up’, this will distort the data and consequently causes outliers in the results. With respects to brain selectivity, it is possible that fMRI data among the two possible motivational explanations show overlap. Specifically, the OFC and mPFC are involved in many different brain processes. One of such processes is the evaluation system, which is also required for altruistic behavior and thus there may be overlap in activation. Lastly, this proposal is susceptible to generalization and validity problems. On one hand, participants will not suffer real monetary disutility from tipping during this experiment, suggesting that the results may not be valid in real-life scenarios. On the other hand, this proposal is designed mostly in the context of the U.S., since the norm of tipping in the U.S. is widely accepted and the average tipping amount (15-20%) used in our proposal is based on studies from the U.S. this, in fact, puts constraints on the generalizability other than the U.S. population.
Future research To overcome some of the limitations of the currently proposed research, further research should take the following suggestions into consideration. First, it is required to further specify and define the negative response hypothesis including selective brain areas that correspond with it. Second, to overcome the causes and chance bias, counterbalancing has to be taken into consideration. Meaning that for different trials the A, B, C and D options should rotate among the percentage base they reflect. For example, A = 30%, B = 20%, C = 15% and D = 5% for certain trials. In addition to this, more options to choose from could be taking into account to derive a more accurate picture of tipping behavior. However, these implications will consequently complicate the statistical analyses. Third, more variables have to be taken into consideration when examining tipping behavior (e.g. group size, bill size, the quality of the restaurant and server characteristics). On a similar note, to obtain more credibility one could compare the data found from actual restaurant customer surveys or use self-reports in addition to the proposed experiment. Fourth, different videos with distinct contextual content could be shown at the start of the fMRI to prime different motivations for tipping. In this way, different behavioral patterns could occur and be explored. And finally, the fMRI machine is not a natural environment and that could make it uncomfortable for some participants, the machine is noisy and evidence show that it might have a negative impact on cognition (Winkielman & Pashler, 2009). To resolve these issues it could be beneficial to use the portable EEG in a natural restaurant environment instead.