How individual needs influence motivation effects: a neuroscientific study on McClelland’s need theory

Work motivation refers to factors that energize, direct and maintain employee behaviour over time (Steers et al. 2004). In their overview, Steers et al. (2004) and Latham and Pinder (2005) summarize three major directions of motivation theory: first, content theories aim at identifying factors positively associated with motivation. Major content theories include, for instance, Maslow’s (1943) theory of hierarchical needs, Herzberg et al.’s (1959) motivator-hygiene theory, and McClelland’s (1987) need theory. Second, process theories focus on the processes underlying work motivation and regard motivation from a dynamic perspective. Some of the most prominent process theories are, for instance, goal-setting theory (Locke and Latham 2002), social-cognition theory (Bandura 1977), or expectancy theory (Vroom 1964). Third, justice theories add a sociological perspective to work motivation with the premise that fair procedures in an organizational context enhance motivation (Latham and Pinder 2005).

Although an exact understanding of motivation continues to evolve, most theoretical approaches argue that motivation emerges from both internal factors of an individual that drive action and external factors of the environment that stimulate action (Locke and Latham 2004). This also applies to McClelland’s (1987) theory of needs, an approach that is considered a content theory of motivation. In essence, McClelland’s (1987) theory of needs suggests that each employee has several, often competing needs and that motivation of employees results from their attempt to fulfil these needs. Ideally, the work environment offers external factors (e.g. rewards) that promise need-fulfilment. In other words, when external factors (rewards) interact with internal factors (needs) work motivation is observed (Beckmann and Heckhausen 2008; Schneider and Schmalt 2000).

Motivation is closely linked to the neurotransmitter dopamine and the dopaminergic reward system. This reward system is a network of neurons that communicate through the neurotransmitter dopamine. When exposed to rewards, the brain responds by increasing its activity. It sends a signal announcing this reward to a particular part of the midbrain, the ventral tegmental area (VTA). The VTA then releases the neurotransmitter dopamine into the nucleus accumbens (NAcc), but also into the septum, the amygdala and the prefrontal cortex. These regions are connected by a so-called pleasure or reward bundle often referred to as ‘reward circuitry’. This reward circuitry is said to supply most of the necessary motivation of human behaviour. Importantly, it is the neural circuit of interest in this study as it responds to rewards (for further information on the reward circuitry see Haber and Knutson 2010).

A particularly critical component of the reward circuitry is the striatum. The striatum is a subcortical structure of the forebrain that is often divided into a dorsal (describing an upper section of the brain) and a ventral (describing a lower brain section) part. The dorsal striatum contains the caudate nucleus and putamen, while the ventral striatum contains the NAcc. Functionally, the striatum coordinates multiple aspects of cognition, including reward perception, motivation, reinforcement learning or voluntary movement (Haber and Knutson 2010).

Neuroimaging studies have demonstrated that the striatum is activated by a large spectrum of different rewards. For example, Spreckelmeyer et al. (2009) reported that monetary gain in a decision task and smiling faces representing social reward lead to activations in the striatum, especially in the NAcc. Kohls et al. (2013) found activations in the NAcc, caudate nucleus and putamen when social approval was anticipated, whereas Mitterschiffthaler et al. (2007) reported increased activation in the ventral and dorsal striatum when listening to cheering music like Strauß’s ‘Radetzky March’. More recently, Lacey et al. (2011) even showed that artistic photographs activate the ventral striatum, and also imagining eating chocolate increases brain activation in the aforementioned areas (Small et al. 2001). Enhanced activation in the caudate nucleus, putamen and the NAcc were further related to many other rewards, such as erotic films (Karama et al. 2002), pictures of high-calorie-food in obese people (Rothemund et al. 2007), pictures of one’s own children (Leibenluft et al. 2004), or photos of alcoholic drinks presented to alcoholics (Braus et al. 2001).

Taken together, the ventral and dorsal striatum are key structures in the reward circuit. As suggested by the findings above, they are involved in processing a wide variety of rewards. In fact, meta-analyses of neural studies on rewards suggest that the striatum is activated by rewards irrespective of their modality (e.g. Sescousse et al. 2013b). Thus, it is also considered a reward modality-independent brain structure. Additionally, however, rewards have been shown to activate areas dependent on their modality, for which reason reward modality-dependent brain structures are described. With regard to such modality-dependence it has been described that the amygdala was more robustly activated by, for instance, erotic rewards versus money or food rewards. Similarly, specific parts of the insula were more likely to be triggered by food rewards compared to money and erotic rewards (Sescousse et al. 2013b).

Despite these findings it is important to keep in mind that except for money, neural studies on rewards rarely include rewards that are relevant to the management context. This study takes a step toward closing this gap as it solely examines rewards commonly used in management.

The overall aim of this study is to validate McClelland’s (1985) need theory using neuroscientific methods and thereby fostering foundations of work motivation. To fulfil this aim two hypotheses are derived. First, we examine how far different management rewards are perceived as comparably rewarding. By doing so McClelland’s core assumption that different external motivation factors may contribute to work motivation will be validated. Investigating this matter further provides a neural basis for our second hypothesis, examining whether this neural basis changes once individual needs are considered.

In this study, the following three contemporary management rewards are represented: (1) high income, (2) respectful leadership and (3) prestigious company cars. According to von Rosenstiel (1975), these rewards represent the reward categories ‘direct financial rewards’, ‘organizational rewards’ and ‘indirect financial rewards’, respectively. As behavioural results suggest that motivation triggered by both, financial and non-financial rewards increases performance—assuming that these rewards have a comparable rewarding effect (Cerasoli et al. 2014)—it is expected that rewards representing three different reward categories show effects that overlap to a certain extent. However, since Cerasoli et al. (2014) additionally demonstrate that motivation triggered by financial and non-financial rewards differently impacts the quality and quantity of performance, it is further expected that there is also a unique effect tied to rewards from different reward categories. The latter expectation is further supported by findings from de Gieter and Hofmans (2015), who show that there are differences in the effects of financial, material and psychological rewards.

As outlined in the previous section neural findings support the assumption that different rewards have both, overlapping and specific effects. In fact, neural findings show that different rewards, such as food, money or erotic pictures, activate reward modality-independent brain areas, yet they also depict reward modality-dependent activations. Combining behavioural and neural findings and further building on Sescousse et al. (2013b) and Kohls et al. (2013), we assume that there is a substantial overlap in the ventral and dorsal striatum, which represent the modality-independent reward areas, when experiencing high income, respectful leadership and prestigious company cars. However, also in line with behavioural results and also building upon Sescousse et al. (2013b) and Kohls et al. (2013) we further argue that these three rewards too depict modality-dependent activation patterns that are not shared with other rewards. Concluding, the following hypothesis is stated:

The second hypothesis refers to empirical findings suggesting that work motivation is based on an interaction of needs (internal factors) and rewards (external factors; e.g. Cerasoli et al. 2014). Building on these findings we test a key assumption of the need theory and examine on a neural level whether a closer match between an employee’s needs and the rewards that offer need fulfilment result in more rewarding effects. According to McClelland’s (1985) need theory, higher work motivation should be observed when a reward meets a corresponding need. Initial findings support this assumption, yet primarily for rewards outside the management context. For instance, when rewards match a person’s need level then this person experiences more positive affect (Brunstein et al. 1998), refers to fewer psychosomatic complaints (Baumann et al. 2005), and reports an increased feeling of flow (Schüler and Brandstätter 2013). Moreover, need–reward match affects the budget goal commitment of managers (Sandalgaard et al. 2011) and enhances trust as well as trustworthiness (Fehr and List 2004). Since the benefits of a high match between a person’s needs and the rewards presented by an environment relate to higher brain activations of the reward circuitry (primarily the striatal area), the following hypothesis is stated:

In total, 44 (29 females) healthy MBA students with a mean age of 25 years (SD = 2.26) participated in this study. Because handedness, vision, and neurological illness may systematically impact brain activations (e.g. Knecht et al. 2000), all students were right-handed, with normal or corrected-to-normal vision and not a single person reported a history of neurological illness. Participants were screened for exclusion criteria (e.g. metal in the body, physical impairment, pregnancy, psychosis) and all provided written informed consent as part of the university’s ethics committee protocol.

Contrary to the usual practice in experimental economics, the study presented in this paper was not incentivized in the sense that the compensation of subjects depended on their decisions in the experiment. Instead we followed the common neuroscientific conventions of paying a fixed compensation to the participants (€ 15). This decision was necessary, because we could have only made a monetary compensation for the money treatment, as it would not have been possible to provide an actual company car or a respectful leader as a reward in the other treatments. As incentivizing only the money treatment while using hypothetical rewards in the other two treatments would have introduced a potential confounder, we implemented all three rewards hypothetically. Nevertheless, we consider this as a potential limitation and we refer to it in the Limitation section.

The study consisted of two main parts: (1) a pre-scanning part and (2) a scanning part. At the beginning of the pre-scanning part participants were told a cover story to help them establish an employee role. As is common in neuroscientific experiments, the cover story was approved by the university’s ethical protocol. They were told that they had the chance for an internship and that depending on their task performance in the MR-scanner they would be compensated with different rewards in the internship. In the pre-scanning part, participants also received information about the task they had to complete in the scanner—which in fact was only a distraction task—and about the hypothetical rewards they could receive. The distraction task and the rewards are described in Sects. 3.2.1 and 3.2.2, respectively.

In the scanning part, an experiment with four different treatments (i.e. three reward treatments and one control treatment) was performed. An event-related fMRI-design¹ was conducted for all four treatments to account for the neural response to every single reward and to the control treatment. In each treatment, participants followed the exact same procedure. The MRI procedure is described in Sect. 3.2.4 in more detail.

This section provides an overview of the behavioural questionnaires used and gives information on the MRI-data acquisition and data analysis. Since information on the acquisition and analysis may sound very technical to readers less familiar with neuroscientific experiments, it is split. First, to make MRI-data acquisition and data analysis available for a larger readership, information on those two issues is presented in a broader sense where background information is added to make the technical information more accessible. Second, the “Appendix” offers additional technical details and displays information on the data acquisition and data analysis in a way that is commonly used when reporting to a readership highly familiar with MRI studies.

In order to test whether the participants differed regarding their number of correctly solved tasks and whether this number of correctly solved tasks varies across the three rewards, a repeated-measures ANOVA with rewards (Leader, Car, Money, and Arrow as control) and scores (Correct vs. Incorrect) was conducted. As intended by the study design, the ANOVA showed a significant main effect for the scores, indicating that more tasks were solved correctly [F(1,43) = 69.99, p < .01, η _p ²  = .62; M_Correct = 28.85, SD_Correct = 3.26 vs. M_Incorrect = 20.93, SD_Incorrect = 3.15]. There was neither a main effect for rewards nor a significant interaction between the reward treatments/control treatment and the scores. Thus, the participants performed equally well across the different reward treatments and the control treatment.

For the reaction time (RT), again a repeated-measures ANOVA with reward (Leader, Car, Money, and Arrow as control) and scores (Correct vs. Incorrect) was performed. In this case, the ANOVA showed a main effect for scores (F(1,43) = 23.05, p < .01, η _p ²  = .35). In sum, correct answers were given more quickly than incorrect ones (M_Correct = 572.86, SD_Correct = 171.86 vs. M_Incorrect = 633.76, SD_Incorrect = 183.27). Neither a significant main effect for rewards nor a significant interaction between the rewards/control treatment and the scores was found. Thus, study participants responded equally fast across the different rewards and the control treatment.

In order to identify reward modality-dependent activation patterns for processing the three rewards of this study we first analysed the simple T-contrasts of each reward treatment as well as the control treatment (see Method section). Table 2 summarizes significant activation clusters that were found to be activated after multiple comparison correction for the ‘Leader+ > Leader−’ contrast of the leader treatment, the ‘Money+ > Money−’ contrast of the money treatment, and the ‘Car+ > Car−’ contrast of the car treatment. Results depicted in Table 2 show that when potential employees anticipate working for a respectful leader, receiving high income or driving a prestigious company car, then mainly subcortical and cortical nodes of their reward circuit are elevated. As expected, the opposing contrasts ‘Leader− > Leader+’, ‘Money− > Money+’ and ‘Car− > Car+’– did not result in significant activations. Therefore, when employees anticipate working for a non-respectful leader, receiving low income, and driving a less prestigious company car then no significant neural activation clusters were observed. Table 2 further summarizes the activation clusters for the control treatment (see the ‘Arrow+ > Arrow−’ contrast in Table 2). Again, the ‘Arrow− > Arrow+’ contrast did not result in any significant activations.

As mentioned in the Method section, the Leader+, Money+ and Car+ treatment were always presented in the MR-scanner when participants solved a distraction task correctly. Thus, the simple T-contrasts shown in Table 2 represent both activations that occur from processing a particular reward and activations that occur from pleasure when solving a task correctly. In order to check whether there is neural activation that can be exclusively traced back to the rewards, the neural activations from the control treatment—which represented only activations related to the pleasure of solving a task correctly—was subtracted from the neural activations in each reward treatment.

Consequently, the simple T-contrasts of each reward treatment displayed in Table 2 were contrasted to the simple T-contrasts of the control treatment (see for example Stanley 2016). This should result in activation patterns that are specific to the particular rewards but that do not include activations related to the pleasure of solving a task correctly. In that vein the following contrasts were examined: (‘Leader+ > Leader−’ > ‘Arrow+ > Arrow−’), (‘Money+ > Money−’ > ‘Arrow+ > Arrow−’), and (“Car+ > Car−’ > ‘Arrow+ > Arrow−’). Table 3 and Fig. 2 depict the activations that occurred when study participants anticipated (a) working for a respectful leader, (b) receiving a high income, and (c) driving a prestigious company car, while controlling for the confounder (i.e. the pleasure when solving a task correctly).

With respect to the respectful leader reward, analysing the contrast of the simple T-contrasts (i.e. ‘Leader+ > Leader−’ > ‘Arrow+ > Arrow−’) resulted in activations within the putamen, thalamus and supplementary motor area, with these areas being more active when participants imagined working for a respectful leader as a reward (see Fig. 2 red areas). In a similar vein, when studying the financial reward, the ‘Money+ > Money−’ > ‘Arrow+ > Arrow−’ contrast resulted in activations within the left and right calcarine, subsuming the middle occipital gyrus (see Fig. 2 green area). Finally, when studying the company car reward, the ‘Car+ > Car−’ > ‘Arrow+ > Arrow−’ contrast did not lead to significant activation clusters.

With respect to hypothesis 1, Table 3 and Fig. 2 show that a respectful leader, high income and a prestigious company car are perceived as rewards, even after controlling for the pleasure of solving a task correctly. Moreover, these management-related rewards activate subcortical and cortical nodes of the reward circuitry (e.g. putamen) that are also activated by more basic rewards such as foods or erotic pictures.

The conjunction was analysed using the software SPM (minimum T-statistic; Friston et al. 2005), in which the conjunction null hypothesis test was used (Nichols et al. 2005). This test requires significant regions in all tested treatments (i.e. a logical ‘AND’ conjunction). For the analysis, the ‘Leader+ > Leader−’, ‘Car+ > Car−’, and ‘Money+ > Money−’ contrasts were submitted to the conjunction (corrected for multiple comparisons; FWE-corrected at p < .05, voxel-extent threshold k > 49).

Results showed that all three rewards evoked activations in the calcarine, the lentiform nucleus subsuming the putamen, and the caudate nucleus (see Table 4 and Fig. 3). The findings provide some evidence for a common neural system involved in anticipating management-related rewards, involving the calcarine and the bilateral putamen.

Consequently, when findings from the conjunction analysis (Table 4 and Fig. 3) and the contrast analysis (Table 3 and Fig. 2) are considered jointly, it can be seen that on the one hand the rewards respectful leader, high income, and prestigious company car activate similar, thus modality-independent brain areas. However, on the other hand, each of these rewards also activates distinct, modality-dependent brain areas (e.g. supplementary motor area for the reward respectful leadership). Consequently, H1 is supported.

With respect to H2, we studied whether rewards that closely match an employee’s need result in stronger neural activations of the reward circuitry than rewards that match this need to a lesser extent. Therefore, the need levels of each participant were correlated with their brain activations that occurred when this participant anticipated the corresponding rewards.

As described in the Method section, each reward was assigned to one of McClelland’s (1991) needs, whereby the respectful leader represented an affiliation-oriented reward, the high income represented an achievement-oriented reward and the company car represented a power-related reward. The participants’ need levels were measured using the PV questionnaire (McClelland 1991). Consequently, each participant was characterized by three continuous scores representing the person’s need for affiliation, achievement and power. These three continuous need levels were correlated with the per cent signal change values that were extracted for the regions of interest (see Method section).

Figure 4 summarizes all correlational results. Figure 4a depicts that the higher the study participants’ need for affiliation the stronger their activations in the NAcc (left/right), caudate nucleus (right) and putamen (left/right) when they anticipated working for a respectful leader (r_{NAcc_l} = .39, p < .01; r_{NAcc_r} = .40, p < .01; r_{Caud_r} = .34, p < .01; r_{Put_l} = .33, p < .05; r_{Put_r} = .33, p < .05). In contrast to these results, the participants’ need for achievement and power was not significantly related to the neural activations that occurred when they anticipated working for a respectful leader.

Similar results were found for the high income. As can be seen in Fig. 4b participants’ need for achievement was positively linked to their neural activations in the NAcc (left), caudate nucleus (right), and putamen (left/right) that occurred when the participants anticipated receiving high income (r_{NAcc_l} = .33, p < .05; r_{Caud_r} = .31, p < .05; r_{Put_l} = .40, p < .01; r_{Put_r} = .33, p < .05). Importantly, none of the remaining need levels was significantly related to the per cent signal change values.

Finally, Fig. 4c shows that the higher the persons’ need for power the stronger their activations in the nucleus accumbens (left) and the putamen (left/right) when they anticipated driving a prestigious company car (r_{NAcc_l} = .25, p < .05; r_{Put_l} = .34, p < .01; r_{Put_r} = .32, p < .05). The need for achievement and affiliation were not significantly linked to the neural activations of the car treatment.

With respect to hypothesis 2, results from Fig. 4 show that the better a particular reward matches a person’s need level the higher will be the activations in reward-sensitive ROIs (i.e. the NAcc, caudate and putamen). This finding holds true for all investigated rewards and needs: if a person had a high need for affiliation, the neural activations were stronger when the affiliation-oriented reward was offered. If a person had a high need for achievement, the neural activations were stronger when the achievement-oriented reward was offered. If a person had a high need for power, the neural activations were stronger when the power-oriented reward was offered. Consequently, H2 is supported by the current findings.

In line with hypothesis 1 it is concluded that the rewards high income, respectful leadership, and prestigious company car share a common reward modality-independent neural basis but also trigger reward modality-dependent activations. With respect to reward modality-independent activations we demonstrated that they lead to increased activations in the calcarine and in two areas that are commonly part of the reward circuitry—the lentiform nucleus with the putamen and the caudate nucleus. This holds particularly true for the rewards high income and respectful leadership, as the neural activations remain even after controlling for activations that result from the pleasure of solving a task correctly. This finding of reward modality-independent activations suggests that different rewards impact the same brain areas in a similar way. Therefore, it appears as if the rewards high income, respectful leadership, and prestigious company car are to some extent interchangeable on the neural level despite the circumstance that they trigger different needs of McClelland’s (1957) theory.

Our findings are mostly concordant with studies investigating the overlap between social and monetary rewards. For example, Izuma et al. (2008) reported that individuals activated the putamen and caudate when they received monetary rewards. Additionally, the authors reported that the very same brain areas were activated when individuals were offered the opportunity to achieve a good reputation, whereby gaining a good reputation was seen as a non-monetary reward. Lin et al. (2012) further showed that the putamen and the caudate are both activated when individuals process a smiling face and financial benefits. In a similar vein Sescousse et al. (2013a) showed that the promise of monetary rewards elicits neural activations in the very same brain areas (i.e. the ventral striatum including the putamen and the caudate nucleus) as erotic pictures, whereby the latter represented non-monetary rewards. Taken together, findings from previous studies and results from the current study provide evidence for the idea that the neural representations for financial rewards and non-financial rewards partly overlap and are, on a neural level, to some extent interchangeable.

Reward modality-dependent brain activations were found in particular for the superior frontal gyrus or the inferior temporal gyrus, which was only activated when either the reward respectful leadership or high income was used. The prospect of high income evoked neural activations in the middle occipital gyrus after controlling for activations that resulted from the pleasure of answering correctly. In light of previous findings our results indicate that anticipating higher income activates brain regions that are typically linked to positive arousal, processing (pleasant) visual stimuli, reward, and value of probabilistic rewards. Neuroimaging findings, for instance, have revealed that watching pleasant pictures and words leads to greater activation in the middle occipital gyrus than watching neutral ones (Mourão-Miranda et al. 2003; Rotshtein et al. 2001). In addition, the middle occipital gyrus is activated when emotions associated with autobiographical memories are regulated (Kross et al. 2009) or when rewards of high value are expected (e.g. Marsh et al. 2007).

Regarding the reward respectful leadership, the participants’ putamen, thalamus and the supplementary motor area were activated after controlling for the pleasure of answering correctly. Based on earlier findings our study suggests that a respectful leadership style activates core components of the reward-related neurocircuitry. According to Liu et al. (2011), the activated brain areas are common parts of the neural network underlying reward processing. Kohls et al. (2013), for instance, found activations in the putamen and thalamus when social approval was anticipated, whereas Smith et al. (2010) reported activations in these two brain areas when research participants felt rewarded after looking at attractive faces. Parallel to our findings, Hu et al. (2015) recently showed that activations in both the putamen and the thalamus occurred when individuals were rewarded by a smile from another person. Additionally, Pfeiffer et al. (2014) reported that putamen and thalamus activations were found when participants interacted with another person in an experiment, while these activations were not found when participants interacted with a computer.

With respect to the third reward of this study, results show that the prestigious company car elevates the inferior temporal gyrus, the putamen plus the caudate nucleus, the calcarine and the posterior cingulum. After adjusting the participants’ brain activations for activations that occurred from solving a task correctly, the whole-brain analysis did not result in significant activation clusters. Therefore, the findings need to be interpreted with caution. However, the unadjusted results indicate that the chance of driving a prestigious company car primarily evokes activations in the inferior temporal gyrus. Previous research suggests that this area is associated with aesthetics, rewards and personal involvement. Schaefer et al. (2015) demonstrated, for example, that the inferior temporal gyrus was more strongly activated when individuals were more deeply involved in tasks. The calcarine and posterior cingulate is often triggered when processing aesthetic art (i.e.; Lutz et al. 2013; Tsukiura and Cabeza 2011) and the putamen is associated with rewards (Kohls et al. 2013).

With regard to hypothesis 2 it was argued that rewards that closely match a person’s needs result in stronger neural activation of the reward circuitry than rewards that match these needs to a lesser extent. Our findings are in full support of this assumption and show that when individual needs closely relate to the offered rewards these rewards are perceived as more rewarding. In detail, if a person with a high need for achievement was rewarded with high income then this person showed stronger activation of the key parts in the reward circuitry compared to receiving the reward prestigious company car or respectful leadership. A corresponding pattern occurred in persons with a high need for affiliation when they had the chance to work for a respectful leader and for persons with a high need for power when they were rewarded with a prestigious company car.

Our region of interest analysis showed that three brain areas of the reward circuitry—the nucleus accumbens, putamen and caudate nucleus—are more activated when rewards closely match a person’s need level. This is particularly true for the rewards high income and respectful leadership. Based on these findings it is suggested that the nucleus accumbens, putamen and caudate nucleus are sensitive to processing close matchings between needs and rewards. These results are in line with findings by Schultheiss et al. (2008), who reported that individuals with a high need for power show stronger activation in brain structures involved in motivation (e.g. the striatum) when watching need-congruent pictures. In contrast, individuals with weak power needs did not display such strong neural activations. A similar result was found for the need for affiliation, albeit not in a management or work context. Quirin et al. (2013) found that the need for affiliation predicted activity in the putamen only when study participants watched video clips with a strong reference to the affiliation motive (i.e. love stories). Thus, activity in the putamen was sensitive to affiliation need-congruent stimuli. With respect to the need for achievement, Mizuno et al. (2008) showed that the putamen of study participants was more strongly activated when they received feedback on their task performance compared to when they received performance-contingent pay.

Our findings support a personality-based approach to work motivation and suggest considering the needs of employees when setting up effective reward systems. We provide neural support for the assumption that work motivation results from an interaction of internal factors (e.g. needs) and external factors (e.g. rewards). As this is—to the best of our knowledge—the first study that investigates the neural basis of different management-related rewards and their match to McClelland’s needs, our study may be used to strengthen and expand motivation theory. More particularly, we uniquely validate McClelland’s (1985) theory as we were able to demonstrate the importance of the matching between McClelland’s needs and congruent rewards on a neural level.

Our findings might also be used to refresh the discussion on conscious and unconscious needs or motives, as was required by Steers et al. (2004) in their statement on the future of work motivation theory. While individuals are aware of their conscious needs and are able to express them verbally (e.g. in a questionnaire), they lack access to or awareness of their unconscious needs. Consequently, unconscious needs or motives are difficult to measure and neuroscientific techniques may offer new methods to approach them in a more valid way. By using new methodological approaches the long-standing discussion on whether conscious and unconscious needs are related could gain new impulse. Additionally, our study provides findings that could be referred to when comparing the neural activations of rewards matched with conscious needs to the activations of the same rewards matched with unconscious needs.

What managerial implications can we draw from our results? From a practical point of view, our findings suggest that not only high income but also a respectful leader and a prestigious company car are perceived as rewarding. Moreover, these rewards are, on a neural level, to some extent interchangeable as they activate mutual brain areas. That is, a prestigious company car or respectful leadership may substitute a pay rise within a reward system to some extent. This finding could therefore be used to encourage certain diversity in the use of management rewards. In today’s work environment money is undoubtedly the most common reward (Gunkel et al. 2009) but our results suggest that other types of rewards might be similarly effective. Therefore, we suggest extending the role of non-monetary rewards at the workplace (e.g. a good leader, a sound working atmosphere or other non-monetary benefits) because a higher income or financial bonuses might be too short-sighted in many cases.

We also find that personality-based rewards are more rewarding, and thus are presumed to impact on motivation to a greater extent. Consequently, organizations should consider their employees’ needs when setting up reward systems. According to our findings, individuals with a high need for achievement, affiliation or power are more effectively rewarded by a closely corresponding reward. Rewards that do not closely match an employee’s needs may also be rewarding but, according to our results, certainly to a lesser extent. Therefore, reward systems that consider the importance of this match provide the chance of having a higher impact on the employee’s motivation and, thus, might also be more efficient.

Finally, as our findings suggest that there is potential benefit for organizations in knowing the predominant needs of their employees, organizations might sensitize leaders to determine information about these individual needs. Moreover, leaders could adapt their leadership style so that employees are motivated to express their predominant needs. Either way, using individual needs for motivational purposes requires being informed about them.

In this study certain limitations should be considered. Practical restrictions forced us to hypothetically reward the study participants and therefore we only provided a fixed pay-off. We are aware of the fact that economic experiments are usually fully incentivized, while study designs commonly used in social neurosciences do not necessarily deem this. With respect to our treatments (money, respectful leader, company car) we decided to incentivize none of these treatments to avoid potential confounding effects when only a monetary compensation could have been provided. Consequently, all three rewards were presented hypothetically. However, results should not be biased except if the hypothetical rewarding differentially impacts on neural activations within the three reward treatments (e.g. Falk et al. 2008). In this regard, there is evidence that supports the validity of using hypothetical rewards in experiments (e.g. Madden et al. 2003, 2004), which is why no such differential bias was expected for the current study. Nevertheless, we acknowledge that this might limit the validity of our study and it is certainly worth also studying the neurophysiological foundation of rewards that are actually given to study participants in future studies. Secondly, we did not investigate cultural differences regarding rewards. Future research might address this issue and study the impact of cultural preferences regarding rewards and the match between need and rewards. Thirdly, the participants of our study were MBA students. Even though they had work experience it might be valuable to conduct a future study on employees, as the relevance of rewards may depend on the years of work experience. Fourthly, due to the fact that in the current study participants had to stay in the MR-scanner for approximately 45 min, fatigue effects are possible. Thus, the length of the experiment might have influenced neurological findings. Moreover, as in most experimental studies, the question remains open as to whether the results of this study hold for real-life settings outside the laboratory (i.e. the study’s external validity needs to be proven). However, studies looking into the external validity of subjects’ behaviour in the laboratory usually show promising results (e.g. Benz and Meier 2008; Rustagi et al. 2010). Finally, this study was limited to investigating certain needs and rewards. Thus, the generalizability of current findings to other needs or rewards might be limited and should be made with caution. Future studies could, for example, include needs from other theories (e.g. Maslow 1943) and could further replicate the results with additional rewards.