The role of timing and uncertainty on employee reward : are findings from neuroscience applicable to an organisational setting?

(1)

The role of timing and uncertainty on

employee reward

Are findings from neuroscience applicable to an organisational setting?

Author: Erwin Havekotte Student ID: 10326928 Place: Oxford

Date: May 2015

U n i v e r s i t y o f A m s t e r d a m | F a c u l t y o f E c o n o m i c s & B u s i n e s s | M a s t e r t h e s i s M S c B u s i n e s s S t u d i e s S p e c i a l i z a t i o n : M a n a g i n g & L e a d i n g P e o p l e | A c a d e m i c y e a r : 2 0 1 4 - 2 0 1 5 | S u p e r v i s o r : M a a r t e n d e H a a s

(2)

Statement of originality

This document is written by Erwin Havekotte who declares to take full responsibility for the contents of this document. I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it. The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.

(3)

(4)

Abstract

The goal of this thesis was to examine if findings from the field of neuroscience can contribute to current reward theories. Neuroscientific research has provided robust proof that timing and probability have an effect on the perceived value of a reward. Another neuroscientific experiment has shown that equity and fairness also influence the perceived value of a reward. In this thesis I discuss how these findings fit in the current reward theory and if these findings can be applied in an organisational setting. I proposed a conceptual model that translates the simplified setup used in the neuroscientific experiments to an organisational context. Based on the model and assumptions from literature, interviews were conducted with individuals that receive part of their compensation as a variable reward. The findings of this exploration indicate that both timing and uncertainty have an impact on the perceived value of the reward, and thus its motivating effect. The key difference lies in the reward mechanisms that are specific for short and long term variable reward. I argue that

reducing the time and uncertainty between effort and performance assessment is the best way to optimise the value of the reward without having to increase the magnitude. Next to this, findings show that fairness of the performance assessment has an impact on the perceived value of reward.

Keywords:

(5)

List of figures and tables

Figures

Figure 1. Dopamine activity during a pavlovian test setup with variable delay ... 7

Figure 2. Dopamine activity response on reward probability ... 9

Figure 3. Subjective ratings on reward distribution experiment ... 11

Figure 4. Conceptual model for reward comparison. ... 13

Tables

Table 1. List of demographics of the interviewees ... 19

(7)

1

1. Introduction

If people are good only because they fear punishment, and hope for reward, then we are a sorry lot indeed.

– Albert Einstein.

Is reward really the best way to motivate individuals? Motivation – and more specifically motivation in an organisational context – has been a widely discussed topic over the past decades. Organisations use variable pay to incentivise their employees to increase productivity and behave in a desirable way. Locke et al. (1980) stated that “Money is the crucial incentive... no other incentive or motivational technique comes even close to money with respect to its instrumental value” (p. 379). While reward might not be the only incentive that motivates individuals, it can be argued to be one of the most powerful ones.

The idea that reward leads to motivation – which in return leads to increased productivity – is an appealing one. It is therefore no surprise that the majority of companies make use of variable pay or bonus incentive systems. Over 70 percent of organisations in the Netherlands claim to use some form of variable pay for at least their senior management (Ramakers & Schraade, 2013). A slightly smaller proportion of companies apply variable pay for some of their non-executive

employees in the organisation. Even the start of the financial crisis in 2008 did not have a big impact on how variable pay systems within organisations are used today. A recently published report by the Dutch general employers’ association AWVN (Ramakers & Schraade, 2013) shows that 69 percent of the organisations did not make any changes to their pay structure, despite the negative stance on bonus systems in the public debate. In fact, the opposite seems to be true. In the year 2013 a sum of 13 billion euro was spent in the Netherlands on variable pay, an increase of 1.4 billion euro over the year 2012 (CBS, 2015).Based on these numbers it can be concluded that variable pay is widely used and seen as valuable by many companies.

The reason that many of these organisations choose for variable pay is that it is thought to bring competitive advantage (Lawler, 2000). In the literature there is overwhelming evidence that introduction of individual variable pay systems leads to substantial increase in productivity (Locke et al., 1980, Guzzo et al., 1985, Jenkins et al., 1998, Judiesch, 1994). Therefore, we can conclude that many companies use variable pay and these incentives have been proven to lead to substantial increases in productivity. The mechanisms behind the increase in productivity after introducing variable pay are less clear and the workings of variable pay schemes are subject to much discussion. Some researchers even suggest that external rewards can lead to negative effects on employees’

(8)

2 intrinsic motivation (Deci, 1971, Lepper et al., 1973).

The aim of this research is to explore the workings of reward and performance by taking findings from neuroscientific research about reward processing and applying them to existing theory of organisational behaviour and reward. Can these findings about primal brain responses to reward be applied in an organisational context? By taking this new perspective this thesis aims to find evidence that might connect these fields of science and make recommendations for future research. This thesis starts with an overview of current reward theory, followed by findings from neuroscience on reward processing. The points where theory and findings from these fields overlap or conflict will form the foundation for my research questions. The neuroscientific findings have to be translated to an actual organisational environment to make a comparison possible. This will be done in the form of a conceptual model that allows a comparison between the neuroscientific findings and a pay for performance situation. This model will be used to test a set of research questions by interviewing individuals in the Netherlands that are subject to a variable pay system. Because of the explorative nature of this research the interviews are meant to test if the assumptions and direction of reasoning put forward are sound. Finally the implications for both existing theory and

neuroscience will be discussed and a proposal will be made for future research to build on this exploration.

(9)

3

2. Theory and research questions

2.1. Reward and performance: overview of theory

What are the mechanisms that explain the relationship between reward and performance? There have been many attempts to improve our understanding of the processes that link monetary reward to improved performance. Psychological and economic theories that offer explanations for both the positive and negative effects of reward on performance include the expectancy theory, equity and justice theories and self-determination theory (Deci & Ryan, 2000, Bartol & Locke, 2000, Gerhart & Rynes, 2003). In this section these theories will be discussed in more detail.

Expectancy theory is based on the idea that individuals choose effort levels which they believe will lead to valued outcomes (Vroom, 1964). An individual chooses the level of effort with the greatest motivational force. According to the theory this motivational force is a function of

expectancy, the likelihood that a particular outcome is the result of that action, and the attractiveness of the outcome to the individual (Klein, 1991). Expectancy theory defines two conditions that must be met in order to see the effect of reward on motivation. First, employees must value the offered reward that is linked to performance. The second requirement is that the employee must perceive that increased effort will lead to better performance, and thus to a better reward. To summarise, the reward needs to be large enough to warrant the effort by the employee and should be perceived as fair in respect to the effort input. Cadsby et al. (2007) described the contrast between a pure fixed-salary system and variable pay from an expectancy theory perspective and found that contrary to fixed salary, variable pay creates a very direct and explicit link between performance and outcome. The link they propose suggests that increased effort will lead to improved performance outcomes, and the increased performance outcomes will lead to a higher reward (effort -> performance; performance-> reward).

Justice and equality theories deal with the comparison of oneself with others. Equity occurs when a person perceives that the ratio of his or her inputs versus the outcomes are equal to the ratio of input and outcomes of others. Inequity happens when a person experiences that his ratio is not equal to that of others (Adams, 1965). People expect to get a fair return for their contributions, this concept is also known as the equity norm. People compare their inputs and outputs to others to determine the fairness of their returns when compared to others. This concept is also known as social comparison. In a work situation this means that when employees feel themselves to be in an inequitable situation, they will look for a way to reduce the inequity between them and their colleagues. The results could be that they distort the inputs and outcomes in their own minds (cognitive distortion), by altering their effort on their inputs and outputs, or they could end up leaving the organisation (Carrell and Dittrich, 1978). In order for pay for performance to work in an

(10)

4 organizational context, the employees need to perceive reward inequality as the outcome of

justifiable differences in performance, capabilities or output.

The self-determination theory (Deci & Ryan, 2000) is an approach to human motivation and personality that deals with intrinsic and extrinsic motivations. An important aspect of the theory is the degree to which the values and regulations of the requested behaviour are internalised and integrated into the subject. Internalisation is to which extent a person ‘takes in’ a value or regulation. Integration is the further transformation of internalisation, which makes the value or regulation feel as it was their own. The result is that this value or regulation will emanate from this person’s sense of self. Deci and Ryan identify three needs for intrinsic motivation and personality integration, which are the need for competence, relatedness and autonomy. Competence is the need to be in control of one’s outcomes and achieve mastery in their work. Relatedness is the need to interact with others and share connections with others. Autonomy is the need to be in control of your life and the extent to which an individual can make choices that are in harmony with their personal sense of self. All three basic psychological needs need to be satisfied for an individual’s well-being and health, but the salience of the three needs can be different between the individual’s culture, personal situation or personal experience. When all three needs are met, it is thought to lead to improved intrinsic

motivation, which in turn leads to greater effort and performance. Furthermore, Deci and Ryan argue that reward can be perceived to be controlling when it is used to persuade or pressure someone into acting in a way that this person out of free will, might not have done. The more internalised or integrated the behaviour requested by the rewarding party is, the less restrictive it is seen. The most remarkable finding by Deci and Ryan was that the introduction of extrinsic reward could decreases intrinsic motivation if it is seen as restrictive or goes against the basic psychological needs. This means that reward in itself is not always motivating but is dependent on how well a subject internalises the goals that he or she is rewarded for.

In addition, Rynes et al. (2005) propose that most ideas about pay and motivation can be captured in two general processes, namely the ‘incentive effect’ and the ‘sorting effect’. The incentive effect looks at the potential of pay for performance to impact the performance of employees when all other attributes are seen as a constant. The sorting effect deals with how

different forms of pay for performance can lead to different kind of people applying and staying at an organisation (Gerhart & Milkovich 1992, Lazear 1986). This effect can work both ways, meaning that people can self-select by applying or staying at an organisation that applies a certain pay for

performance scheme or that organizations select certain people based on their culture or payment strategy. At this moment there is more study and theoretical base for the incentive effect (Cadsby et al. 2007), whereas the evidence for the sorting effect is less abundant. The sorting effect involves both employer and employee, in combination with the less solid theoretical base, which makes it

(11)

5 harder to explore this specific kind of behaviour. For these reasons this thesis will focus on the incentive effect, but will revisit the sorting effect when convincing proof for self-selection can be found.

2.2. Dopamine as an explanation on how reward motivates people: neuroscientific findings Dopamine is a neurotransmitter in the brain. Research in neuroscience has shown that dopamine has significant effects on a broad spectrum of cognition and behaviour, including motor control, learning, attention, motivation, decision-making and mood regulation. Mid-brain dopamine neurons also play a major role in reward information processing (Wise, 2004). This has been shown by experiments that have been first tested on test animals such as rats and nonhuman primates. Abler et al. (2006) used a technique called functional magnetic resonance imaging (fMRI) to show that the humans perceive and interpret reward probability and prediction error in a similar way as their test monkeys, suggesting that their dopamine system works similarly. This would infer that some findings from tests with animals can be extrapolated to human behaviour.

People make decisions that are guided by the predictions they make about future events and by comparing their past predictions with the actual outcomes of their decisions. This principle has been the base of many learning theories (Bolles, 1972). Dopamine plays an important role in reinforcement and rewarding that has been shown to contribute to the human learning experience. Experiments with primates (Schultz, 2000) have demonstrated that mid-brain dopamine neurons show activity when there is a difference in predicted and received reward, also known as prediction error signal. When a reward is higher than was predicted, the activity of dopamine is increased (positive prediction error). Similarly, when the reward fails to meet the expected result the activity is lowered (negative prediction model). If the reward matches the prediction the activity of dopamine shows no changes. This model is better known as the reward prediction error model (Glimcher, 2011).

Dopamine experiments have shown that the perceived value of a reward is determined by at least three factors, namely magnitude, probability and timing (Abler et al., 2006). Magnitude is the relative value of the reward. The higher the value of the reward, the stronger the dopamine activity. This activity signals that individuals are driven towards the behaviour that leads to the highest payoff. Timing and probability are closely linked to each other. A delay in time when receiving a reward has a negative impact on the perceived value of this reward. This can be explained by a standard economic model that assumes that the value of the future reward is discounted because of the risk involved in waiting for it (Samuelson, 1937). This means that a reward that is received earlier in time is seen as more valuable and vice versa. Probability deals with the risks that are involved with getting the

(12)

6 reward. If the chances of getting a reward decrease, then the perceived value decreases as well (Tobler et al., 2005). As with magnitude, this guides individuals to choose for a reward that is most likely to be awarded. While magnitude is the easiest and most straightforward way to increase the perceived value, and therefore performance and motivation, probability and timing are also relevant. Next, these two aspects are discussed more in depth.

2.2.1. Timing

As discussed earlier, timing determines the subjective value of reward. In a classical experiment, a small reward that is available after a short period is compared with a larger reward that is only available in the more distant future. This experiment has shown that not only animals, but also humans seem to prefer the smaller reward when presented with this choice (Rodriguez and Logue, 1998). This has led to the idea that the value of a reward diminishes with time. This

mechanism works through the decreasing probability of receiving the reward. A reward in the future has a lower probability to materialise and therefore carries risk with it.

Kobayashi & Schultz (2008) set up experiments to test the effects of timing. They wanted to see what the effects were of delaying reward on reward preferences by using two monkeys. In the first test they set up a conditioned impulse in the form of a visual stimulus that signalled the magnitude of the reward and the delay of the reward. They measured the dopamine levels at the time of the reward stimulus and at the time of the reception of the reward itself. They found that increasing the delay for the reward had an enhancing effect on the level of dopamine activity at the moment of reward, while the dopamine measured at the time of the reward impulse showed a reversed reaction. This phenomenon can be explained by the reward prediction error model. A reward impulse that predicts a reward with a short delay is perceived as more valuable than a reward impulse that predicts a reward with a longer delay (Glimcher, 2011). When a reward is received after a short delay, the dopamine reaction is neutral because the reward matches the prediction. When the reward impulse signals a long delay for the reward, then the dopamine levels signal a weaker response to the stimulus. It is when the reward is received that the dopamine neurons are strongly active. In fact, Kobayashi & Schultz (2008) found that the reaction of the dopamine neurons to a delayed reward is quite similar to getting an unexpected reward (without an impulse). This is consistent with the error prediction model that discounts rewards that have long delays and therefore lower probability. This effect is visualised by Figure 1. It shows that the reward stimulus dopamine response was in general smaller for rewards with a longer delay and the reward dopamine response was in general larger for rewards with a longer response. A long delay shows similar

(13)

7 relationship between the reward impulse and the reward.

Figure 1. Dopamine activity during a pavlovian test setup with variable delay. Dopamine activity is aligned to

stimulus (left) and reward (right) for different delay conditions (2, 4, 8 and 16 seconds). The black ticks on the top show times a neuronal impulse is measured. The stimulus response was in general smaller for rewards with a longer delay. The panel labelled ‘free reward’ corresponds to the condition where the reward was given without a prediction signal; this is why only the reward can be shown. The panel labelled ‘no reward’ corresponds to a stimulus that predicted no reward; this is why only the impulse can be shown. Note. From “Influence of reward delays on responses of dopamine neurons” by S. Kobayashi & W. Schultz, 2008, The

Journal of Neuroscience,28(31), p. 7841

In their second experiment, Kobayashi & Schultz (2008) added the distinction between a small reward and a larger reward (four times larger). They found that the large reward that was delayed by 16 seconds had similar dopamine neuron activation as the smaller reward delayed by 2 seconds. The test monkeys had an equal preference for these options. This suggests that dopamine responds not only to the timing, but also to the magnitude of the reward. Another important finding was that the effect of delaying reward seemed to affect a large reward less than a small reward. This means that if the reward is big enough, individuals are willing to wait longer for this reward.

(14)

8

2.2.2. Probability

The previous section gave an overview of how dopamine establishes the subjective value of reward by integrating the salience of future outcomes. A delay between stimulus and reward decreases the probability of receiving the reward and decreases the perceived value of this reward. In other words, probability alters the perceived worth of the reward.

A study by Fiorillo et al. (2003) reports an experiment that aimed to measure the effects of different probabilities of receiving a reward. The experiment had a similar setup as the experiments used to measure the effects of delays in timing. Two monkeys received a conditioned impulse that signalled the reward and the delay of the reward. In this test both the magnitude and the delay of the reward were constant, namely 1-mL of water and a 1s delay. The monkeys were only rewarded with a probability (P) of 0.5 after the reward impulse. This means that the monkeys only got a reward in half of the experiments, but no reward in the other half. The experiment was repeated with a reward probability of 0, 0.25, 0.75 and 1. When the reward was fully predictable (P = 1) then the dopamine neurons show activity at the time of the reward impulse, however there is almost no activity when receiving the reward. This finding is consistent with the earlier experiments. When the probability of receiving the reward decreased, something different happens with dopamine. A low probability leads to a low dopamine impulse at the moment of the reward impulse, but to a high dopamine impulse when receiving the reward. These effects are illustrated in Figure 2. The results are similar to the results of Figure 1. When the probability of receiving a reward diminishes then the dopamine level at the time of the reward stimulus diminishes and the dopamine level of the reward increases. The similarities of these experimental results could be seen as a confirmation that delay in time and probability have a similar effect on the perceived value of the reward.

(15)

9

Figure 2. Dopamine activity response on reward probability. Dopamine activity is aligned to stimulus (left) and

reward (right) for different reward probabilities (P = 0, 0.25, 0.5, 0.75 and 1). The black ticks on the bottom show times a neuronal impulse is measured. Note. From “Discrete coding of reward probability and uncertainty by dopamine neurons.” by C.D. Fiorillo, P.N. Tobler & W. Schultz, 2003, Science, 299(5614), 1899.

However, there is something noteworthy about the findings in Figure 2. Based on the prediction error model one would expect to find the dopamine activity to shift from reward impulse to reward when the probability decreases. Instead, the figure shows the dopamine activity for the reward impulse does not decrease as fast as you would expect based on the prediction error model. Even when the probability of receiving a reward is low or non-existent, dopamine activity can be measured at the moment of the reward signal. Fiorillo et al. (2003) argue that reward uncertainty in itself might contain properties that are both rewarding and reinforcing. This could be an explanation

(16)

10 for the results in Figure 2 that contradict the prediction error. Fiorillo et al. (2003) explore this

phenomenon by comparing this effect with gambling. Gambling – by design – implies a reward uncertainty, and more often than not the losses tend to be higher than the gains. Economic theory and the reward prediction error model do not offer an explanation to why any individual would want to engage in the behaviour of gambling. Despite the mostly negative outcomes, gambling behaviour is widespread in many cultures. A possible reason for this could be that uncertainty rewarding behaviour can be advantageous in natural settings. Unlike a test environment or in a casino, the probabilities are not fixed in a natural environment. Because probabilities are not fixed they contain learnings or incentives for those who participate. Uncertainty in a natural environment can motivate an individual to learn and recognise what actions they should take, and will increase the accuracy of reward prediction in future situations. It is expected that an individual will exert more effort or adapt better to increase the probability of reward. Based on the results it can be concluded that reward uncertainty contains properties that are rewarding and reinforcing. Even though there are possible explanations for this phenomenon, there is no conclusive evidence yet how these mechanics work exactly.

2.2.3. Equity

A recent neuroscientific experiment on equity by Cappelen et al. (2014) was published around the same time of setting up this research. This unfortunately meant that equity theory was not originally included in the scope of the research. Nevertheless, considering the relevance of the work by Cappelen et al. (2014), it is worthwhile to consider their findings.

To measure the effects of equity on reward, Cappelen et al. (2014) set up a test in which candidates did simple administrative tasks for a duration of either 30, 60 or 90 minutes. The

candidates were told that their earnings would be based on a regular hourly wage of 500 Norwegian Kroner (~60 Euro) but a random process could interfere. This interference meant that their final payment could be different from their projected earnings. In the next step of the experiment they matched up three groups of individuals: Those that worked 30 and 90 minutes, those that both worked 60 minutes and those that worked 90 and 30 minutes. All of the groups worked for 120 minutes combined and had a combined earnings of 1000 Kroner. The subjects knew who they were paired up with, so there was absolute clarity on the amount of work and effort the other person put in compared to them. Then the subjects were placed in an MR scanner and were presented with different payment distributions. For all the different distributions they asked the subject to give a subjective rating from very bad (-5) to very good (+5), but to test these subjective ratings they also measured blood-oxygen-level–dependent (BOLD) responses in the striatum. The striatum is seen as

(17)

11 the part of the brain that has a key part in the emotional circuitry and is thought to play an important role in motivating and regulating behaviour (Kompus et al., 2012). Furthermore, the striatum has been associated with social preferences and moral choices. The results of the subjective ratings are shown in Figure 3.

Figure 3. Subjective ratings on reward distribution experiment. The three graphs show the mean and SE of the

subjective rating for 51 possible distributions of income. The graphs correspond to three groups of participants, those that worked 30, 60 and 90 minutes. These participants were matched up with candidates that worked 90, 60 and 30 minutes, respectively. The vertical line in the graphs represents the earning condition. Note. From “Equity theory and fair inequality: A neuroeconomic study.” by A.W. Cappelen, T. Eichele, K. Hugdahl, K. Specht, E. Ø. Sørensen, & B. Tungodden, 2014, Proceedings of the National Academy of Sciences, 111(43), 15370

In the graph we can see that the participants had expected earnings of 250, 500, and 750 Kroner. The graph shows how participants evaluate a given income distribution difference between the expected and actual earnings. Subjective ratings seem to flatten out and in some cases drop when the participant’s share of the total income is a lot higher than their expected earnings. The subjective ratings and the BOLD readings were compared and were found to be significantly correlated. These finding are some of the first neuroscientific proofs that individuals accept income inequalities if they correspond to differences in work effort. If individuals receive less than what they

(18)

12 believe they are entitled to, they get a strong negative association. Interestingly, individuals receiving more than they think they are entitled to will lead to flatten out the subjective rating of the reward, and in some cases even causes the subjective rating of the reward to decline.

2.3. Linking existing theories with findings from dopamine research

How do the findings and advancements in neuroscience fit into the existing reward theories that were discussed earlier: expectancy theory, equity theory and self-determination theory? In this section the overlap, differences and contradictions between neuroscientific results and the existing organisational theories will be discussed. This will be the basis to define the research questions for my study.

Out of the three described theories, the expectancy theory is the best aligned with the findings from the dopamine experiments. The expectancy theory is based on the idea that individuals choose effort levels which they believe will lead to valued outcomes. In other words, individuals will choose to put effort into options that are deemed the most rewarding. Dopamine experiments have shown that the perceived value of reward is determined by at least three factors, namely magnitude, probability and timing (M, P and T, respectively) (Abler et al., 2006). Dopamine helps to signal the value of reward and in this way helps individuals to choose for the most valuable option. The underlying mechanism of both theories is that it leads individuals to choose for the most valuable reward. The perceived value of reward plays a major role in both the experiment findings and the expectancy theory. This makes it a good basis to explore if the experiment findings can be applied in an organisational context. To aid in this exploration I propose a conceptual model that helps

translating experimental findings to an organisational context.

To allow comparison between the two theories I break them both down into simple steps. In case of the neuroscientific experiments we can distinguish between two steps: the reward impulse, which is a signal that announces the reward, and the receiving of the reward. The expectancy theory states that increased effort will lead to improved performance outcomes and the increased

performance outcomes will lead to a higher reward (Cadsby et al., 2007). Here I distinguish between the following three steps: effort, performance assessment and actual reward. Since I am interested in the effects of probability and timing I need to determine where timing and probably play a role in the steps that I just introduced. In the experiments this is quite straightforward as these factors can only play a role between receiving a stimulus and actually receiving a reward. In the expectancy theory this could happen at two moments, namely between effort and performance assessment or between the performance assessment and reward. I expect to find that M, P and T are uncertain at the

moment of the effort but once the performance is assessed and converted into a reward decision this uncertainty around M, P and T is largely dissolved. An alternate explanation is that M, P and T

(19)

13 remain uncertain between performance assessment and reward, because the reward has not yet been received by the individual. These two explanations are visualised in the conceptual model for reward comparison in Figure 4.

Figure 4. Conceptual model for reward comparison. The goal of this model is to allow comparison between the

experiment findings and theory. The experiments consisted of a stimulus, followed by the receiving of a reward. The expectancy theory states that effort -> performance and performance -> reward. As discussed earlier the perceived value of reward is decided by at least magnitude, probability and timing (M, P and T, respectively).

Apart from the similarities, there is a contradiction between one of the experiments and the expectancy theory. This concerns the suggestion that having reward uncertainty in a natural

environment can actually be motivating to increase efforts and learning. Uncertainty is not always a reason to avoid situations or withdraw effort, but can work as a motivator and learning stimulant. In the expectancy theory one would expect an individual to pursue another goal that has a higher reward expectation.

The equity theory is also well aligned with the dopamine findings. It relies on evaluating input and outputs. The experiment with income distributions by Cappelen et al. (2014) used the equity theory to test how individuals would react to reward differences in situations with different work effort. The results of this experiments show that individuals generally accept reward differences if they understand differences in work effort. If an individual received more reward than his or her effort would justify then the subjective rating would hit a plateau shortly after a fair distribution was reached. In some cases the subjective ratings dropped when the distribution differences grew too big. This shows that individuals base the subjective rating of their reward on comparing their efforts and reward with their peers. There are some limitations to this experiment, as it did not take factors such as experience, skill or relative output into consideration.

Experiment stimulus and reward

Stimulus

Reward

Expectancy theory

Effort

Performance Assessment

Reward

M x P x T

(20)

14 Based on the equity and the findings from this experiment I expect to find that not only the M, P and T of the reward have an influence on the perceived value of reward. In addition, the inputs in terms of effort and assessed performance of peers, as well as the received rewards of these peers play a role in determining how valuable a reward is perceived to be. This means that a reward is seen as less valuable when peers get the same reward for less effort or those peers get more reward for the same effort. Another implication is that there are limits to the effectiveness of reward when an individual rates the ratio of effort versus reward as unfair when compared to his or her peers. The self-determination theory is harder to compare to the findings in the dopamine

experiments. The self-determination theory states that basic psychological needs need to be satisfied for an individual’s well-being and health. The self-determination theory also looks at how well the rewarded behaviour is internalised by the individual. Both the psychological needs and the level of internalisation are not part of any current neuroscientific experiments. Even though I am unable to test the theory against experiment outcomes I still choose to include the theory in this thesis. The outcomes of the other comparisons can be tested against the self-determination theory and it might aid in posing rival theories.

2.4. Research questions

Earlier I discussed how the findings from dopamine experiments could be applied to an organisational setting by matching it with current reward theory. The research questions of this study are deduced from combining these two fields of research.

The conceptual model proposed in Figure 4 is based on the assumption that there is reward uncertainty at some moment in time and that the value of the reward is determined by at least three factors: magnitude, probability and timing. In the experiments this starts at the moment of the reward impulse and ends when the subjects receives the reward. To apply this in an organisational setting I will use the expectancy theory that states that effort leads to results and the assessment of these results leads to reward to explore if a comparison can be made. This means that there are two moments in time that the reward uncertainty could occur, which is between effort and performance assessment and between performance assessment and reward. At the moment an individual decides to put effort into his or her work there is uncertainty if the effort will lead to the desired outcome. This uncertainty is mostly taken away at the moment that the performance is assessed and transformed into a reward decision. This is also the moment that a reward prediction error can occur. For these reasons I expect to find that individuals experience the moment of performance assessment, the moment that effort is translated into reward and also the moment that

(21)

15 of being rewarded. An alternate explanation would be that even though performance assessment took place there is still reward uncertainty because it has not been received yet. This would mean that individuals will only feel rewarded at the moment that the reward is received. The following research question will be used to explore the model.

Research question 1: Do employees experience that they are rewarded at the moment of receiving performance feedback or at the moment of receiving their reward?

The next subject to explore is the impact of delay and probability on the perceived value of the reward. The outcomes from the dopamine experiments suggest that delaying the reward or decreasing the probability of receiving the reward leads to a lower perceived value of this reward. This leads to the idea that individuals with a short delay between effort and performance feedback would experience their reward to be more valuable than individuals with a longer delay between effort and performance feedback. To be able to test this I have defined a short delay as receiving variable pay on a monthly or quarterly basis and long delay as receiving variable pay on an annual basis. I also expect that individuals that have a low probability of receiving their reward would experience their reward to be less valuable when compared to individuals with a high probability to receive their reward.

There are multiple ways to determine the perceived value of the reward. The simplest way is to ask how the reward is perceived by the interviewees. Another way is to look at the motivating qualities of the reward. Cadsby et al. (2007) stated that increased effort will lead to improved performance outcomes, and the increased performance outcomes will lead to a higher reward. A reward that is perceived to be more valuable will motivate employees to exert more effort. This means that looking at the behaviour can give information about how valuable the reward is. To see if these links exist I will not only explore if employees alter their behaviour to get a reward (a forward link), but also if they can recall specific behaviour that led to a reward they received in the past. Based on the outcomes of the experiments I expect the rewards with a short delay and low reward uncertainty to be seen as more valuable, and thus more motivating. To explore if timing and probability affect the perceived value of reward in an organisational context I will pose the following research questions.

(22)

16

Research question 2: Do employees who receive feedback shortly after their effort perceive their reward as more valuable than employees who receive feedback with a longer delay?

Research question 3: Do employees perceive their reward as more valuable when there is little or no uncertainty between effort and performance, compared to higher uncertainty?

The last research question I would like to explore is based on the experiment by Cappelen et al. (2014) that tested how individuals respond to reward differences in situations with different work effort. It showed that the fairness between input and received reward affect the perceived value of this reward. Even though my interview questionnaire was not set up to explore this dimension, I still hope to find some evidence of the effect of fairness on the perceived value of reward. The

exploration will be based on how individuals perceive the fairness of their reward compared to the effort they had to deliver. It also includes how individuals position themselves compared to their peers. I therefore explore the following research question.

(23)

17

3. Methodology

3.1. Setup of the study

As an exploratory study, the goal of this thesis is to match two streams of science and to explore whether advancements in the field of neuroscience could fit with the existing motivational theories. Rather than testing hypotheses, the goal is to explore broader research questions to discover if any relationships can be found. Based on the earlier discussion and comparison of organisational behaviour literature and neuroscience findings, four distinct questions were posed earlier. To answer and further explore these questions, a list of topics and required information was drafted. The first research question required input on how performance was being measured, how the rating process worked, what emotions reward gave the interviewees. Furthermore, it was important to learn if a subject experienced different emotions at the moment of performance rating and the moment of receiving their reward. The second and third research question required

knowledge about the specifics of the reward system, the measures of performance and how performance is assessed. I also needed to know how the subjects rated the fairness of their reward, how they experienced the link between their effort and their reward; how they felt about their performance assessment and receiving reward and finally how confident interviewees were in receiving reward. All this input was required to determine their perceived value of the reward. The last question was initially not included in the design for the questionnaire, but the questionnaire already included a question on the fairness of the reward in terms of input and reward. In hindsight there were enough data collected to make exploring this question possible.

This resulted in the following items that required input: measure of performance, rating of performance, perceived influence on attaining targets and reward, details of the reward systems, perceived rewards for good performance, fairness of reward, link of reward to effort through effort or performance assessment both forward and backwards, emotions when rated and emotions when rewarded and reliance on reward.

In order to investigate these items I chose the method of semi-structured interviews. This method allowed for collecting comparable qualitative data, while still leaving room for the interviewees to express their feelings in their own terms and leaving the option open to give additional information. The items were converted to 16 questions that were structurally asked each interviewee. For example, to explore the item fairness, the interviewees were asked the following question: “What do you think of your reward when compared to your performance?” The

questionnaire was concluded with an open question asking the interviewee to add any thoughts or additions they had about their reward and performance. The full questionnaire is attached in appendix B.1. Questionnaire.

(24)

18

3.2. Participants

Participants were recruited from November 2014 until January 2015. To select participants, social media (LinkedIn and Facebook) and my personal network were used. The inclusion criteria were that all individuals were full-time employed and had variable pay as a performance incentive in their current employment. Another important criterion was that the target group received a variable reward that was at least partially based on their individual performance. Because of the particular interest in the temporal effect on the perceived value of the reward, an equal selection was made of interviewees that had a short interval between performance and reward and a long interval between performance and reward. As discussed earlier, the definition I use for a short interval is a monthly or a quarterly interval, and the definition of a long interval as an annual interval.

Guest et al. (2006) suggest that it takes 12 interviews in a homogenous group for saturation to be reached. This is why the initial target for the amount of interviews was set at 12 respondents, 6 for short term reward respondents and 6 for long term reward respondents. After 12 interviews there was an evaluation of the interview data to see if it required more interviewees or that there was indeed a visible saturation. I concluded that there were signs of saturation, especially for the group of short term reward interviewees. One of the interviews had to be discarded because the interviewee received a guaranteed minimal bonus amount that was not contingent upon

performance. This interview was replaced by another interview to complete the required 12 respondents.

The interviewee group consisted of nine males and three females. The youngest participant was 25 years old and the oldest 42 years, with an average age of 31 years. The average in-role tenure was 18 months. When looking at the differences between the group of interviewees with short term reward and long term reward there is one difference that stands out. The average tenure for the group with short term reward is under a year where this is over 2 years for the group with long term reward. The demographic profile of all interviewees is added in Table 1.

(25)

19

Table 1. List of demographics of the interviewees

Interviewee Reward interval Gender Age

Tenure at company (in years) Tenure in role (in years) Interview method

1 Annual Male 32 6 3 Face-to-Face

2 Annual Male 27 1 1 Skype

3 Annual Male 27 1,8 1,8 Skype

4 Quarterly Male 31 1,2 1,2 Telephone

5 Monthly Male 25 0,2 0,2 Skype

8 Annual Male 33 3 3 Skype

9 Monthly Female 38 0,8 0,8 Telephone

10 Monthly Female 42 10 1,5 Telephone

11 Annual Male 30 6,5 0,3 Skype

12 Annual Female 42 15 3,5 Skype

3.3. Interviews

All interviews were conducted by me as an interviewer as a one-to-one semi-structured interview. One interview was taken in person, eight interviews were taken by Skype (a popular voice-over-IP service that allows digital video based interaction) and the last three interviews were taken by telephone. The preferred way for interviewing was initially face-to-face interviews, since this way of interviewing approaches the ‘gold standard’ (McCoyd and Kerson, 2006) for qualitative research as close as possible. Because of time and location restrictions of both candidates and me, this

preference shifted to using Skype. Skype allows a good emulation for face-to-face interviews as it provides a report with the candidate and gives visual cues that most closely resemble a face-to-face interview. Skype allows for the opportunity to evaluate visible parts of responses (Sullivan, 2012). There are some potential drawbacks of Skype when compared to face-to-face interviews. Some researchers suggest that the ‘relative anonymity of online interactions and the lack of a shared social network online’ could potentially lead to interviewees presenting themselves more as their desired self instead of their authentic self when compared to face-to-face interviews (Bargh et al., 2002, Ellison et al., 2006). If neither face-to-face interviews nor Skype interviews were an option, the interviews were taken by telephone. Even though some studies claim that there are no significant differences between a face-to-face interview and a telephone interview (Sturges & Hanrahan, 2004) I found the information captured through Skype richer than just telephone alone due to the earlier mentioned possibility of seeing the interviewee and the possibility to collect visual cues. All

interviews lasted around 20-25 minutes. The interviews were recorded with a digital sound recorder and transcribed verbatim shortly after the interview.

(26)

20

3.4. Methods

After transcribing all interviews and adding researcher’s notes, a first wave of data reduction took place. Because the questionnaire was based on a specific set of items, the initial coding could closely follow these pre-set items. Nevertheless, other code arose from the data as well. The coding was done according to the ‘In-vivo’ method to best capture the exact words and sayings used by the interviewees. The items were assigned to main themes and were clustered if possible. I interpreted and translated quotes and sayings to the best of my knowledge. For example “And the two-and-a-halfth way is that eh people look around and eh form an image of colleagues. And that is also in some way a part of the rating that is given”, which was interpreted as “some of the performance rating is subjective, one or more colleagues could be involved in your performance rating, and there is some uncertainty about the weighing of performance measures”. After clustering all coded items, the coding scheme was used to see if other topics that would fit well with the research question could be distilled. Finally, the coding scheme was analysed for any emerging patterns within the remaining data. All transcribed and coded interviews are added in Appendix B.2 Coded Interviews.

(27)

21

4. Results

The results section is an overview of the outcomes of the interviews. The results will be displayed for each research question separately; where possible the results are supported with quotes from the interviewees.

4.1. Research question 1: Do employees experience that they are rewarded at the moment of receiving performance feedback or at the moment of receiving their reward?

Before I started with the interviews, I made some assumptions about how the dopamine experiment findings could be translated to an organisational context by using a conceptual model (Figure 4). The model predicts that the moments of the reward signal and of the reward in the experiments will match with the effort and performance assessment steps described in the

expectancy theory. One of the questions in the interview was specifically aimed to get feedback on the timing of reward. The interviewees were asked if they experience a difference in terms of emotion between getting their formal feedback and them receiving a reward. Almost all of the interviewees answered this question in accordance with my initial expectation, which was that they experience being rewarded at the moment of receiving performance feedback. Here are some of the answers to the question about feeling any difference in emotion between the performance feedback and the receiving of the reward:

“No, you know. When I actually have the confirmation about the money, the rest is no longer interesting to me at that moment. I don’t even look at my bank account when I have received it. So, not a moment.” (Interviewee 2, Male, Age 27)

“While the amount itself, well that’s just an amount. For me that’s just a bit more to spend, but it does not instil the idea on me that I performed well or poorly.” (Interviewee 3, Male, Age 27)

“Yes, you know you will receive it (the reward). It doesn’t really matter, because it will come. No, the feeling is the same to me.” (Interviewee 11, Male, Age 30)

“Receiving the salary is not the most exciting part for me. I enjoy writing big orders more. And of course, when I receive my salary I think “look, this is what I have been working for!”. But the moment of writing that order gives me a bigger high.” (Interviewee 9, Female, Age 38)

(28)

22

“I don’t have strong feelings at the moment it is transferred to my bank account, that’s really strange.” (Interviewee 6, Male, age 26)

An alternate explanation proposes that interviewees would only experience getting

rewarded at the moment of receiving the reward. This explanation assumes that until the moment of actually receiving reward there is reward uncertainty. This uncertainty is only taken away at the moment of receiving the reward. Because magnitude, probability and timing determine the value of the reward, this would mean that a long period between performance assessment and reward would lead to a lower perceived value in this alternate explanation. There were a few interviewees that specifically mentioned that the time between performance assessment and reward was long. This gave me the opportunity to ask them how they experienced the delay between performance feedback and receiving reward and if it affected the perceived value of their reward. Both

interviewees indicated that once their employer stated that they will receive a reward, they have a solid trust in receiving this reward:

“I would not say that just because there are two months in between (the rating and payment) makes me feel any less triggered by it.” (Interviewee 3, Male, Age 27)

“There is a pretty long time in between receiving my payment, so in that sense you are not even that much concerned with money. At that moment you just have a good feeling about making money and in the long term it (the payment) will come.” (Interviewee 8, Male, age 26)

Finally there was one interviewee that stated that he felt stronger emotion from the actual receiving of the reward and not at the moment of performance feedback.

“But in the end it is nicer to see that money on your bank account. Both give me a good feeling, only I would feel that having that money would make me happier than my boss telling me I’ve done well”. (Interviewee 4, Male, Age 31)

These statements confirm the expectations I had when proposing the conceptual model. The answers gave me enough confidence in the conceptual model to state that individuals experience the moment of performance feedback or performance assessment as the moment they are rewarded. All of the interviewees were confident that they would receive their reward, even if the actual payment would be months later. A delay between performance assessment and reward seemed to have little effect on the perceived value of the reward. I will use this knowledge to explore the other research

(29)

23 questions. This brings us to the next two questions:

4.2. Research question 2: Do roles with short delay or direct feedback after performance perceive their reward as more valuable than roles with a longer delay?

4.3. Research question 3: Do employees perceive their reward as more valuable when there is little or no uncertainty between effort and performance, compared to higher uncertainty?

These two questions are bundled because the outcomes are expected to result from a similar mechanism. Timing has an influence on perceived value because the value of future reward can be discounted because of the risk involved in waiting for it. A low probability of receiving a reward can also lead individuals to discount the value of this reward because of the risks of not receiving the reward. Increasing the time between effort and performance feedback, or decreasing the probability of receiving a reward, increases the likelihood of a reward prediction error. A bigger reward

prediction error will result in a lower perceived value of the reward and is less motivating to exert effort to attain this reward. This leaves me to explore if delay does indeed alter the perceived value of reward, and if probability alters the perceived value of reward and if these two factors have a similar working.

Before the interviews the expectation was that individuals with a long reward interval would see their reward as less valuable when compared to individuals that have a short reward interval. The value of reward will also influence the extent to which it motivates individuals. I expected to see that a reward that is seen as more valuable will is more likely to change an individual’s behaviour. The results from the interviews indicate that individuals with an annual variable reward indeed see their reward as less valuable when compared to individuals that receive a monthly or quarterly variable reward. Five out of six interviewees with an annual variable reward explicitly said that their variable reward does not motivate them much. Here is an overview of quotes by individuals with an annual reward incentive:

“But it (bonus) is not an important stick or carrot for me.” (Interviewee 1, Male, Age 32)

“So I think not, although it is nice to be rewarded for your effort and achieving your targets, it’s nice to be rewarded in a way. But it’s not the case that if I would not get this reward I would put in any less effort.” (Interviewee 11, Male, Age 30)

(30)

24

“And the variable reward, I would rank it maybe at the third or fourth place when it comes to motivating me to work.” (Interviewee 3, Male, Age 27) In contrast, here are some quotes by individuals with a monthly reward incentive:

“Yes, absolutely. You just work with a bigger task focus. Look at your time better: “does this task pay off?” This makes that I’m very much triggered to constantly check if I’m doing the right things.” (Interviewee 10, Female, Age 42)

“Yes, definitely. That (reward) plays a constant role in our behaviour, even though no one will admit this to a team leader” (Interviewee 6, Male, Age 26)

“To have such a month is big motivator to go on and see if I can achieve this result again. It is an extra motivator, but I don’t think it’s my main drive.” (Interviewee 7, Male, Age 26)

The interviewees were asked if they thought about the effects of their actions on the reward they would receive while working. The question was meant to check if the reward guided their actions and perhaps even altered their behaviour. The results show that interviewees with a long term reward were less guided by the reward than the interviewees with a short term reward. All six of the interviewees with a short term reward said that they thought about the impact of their actions on their reward. Some of them mentioned that their behaviour was especially affected at the end of the month. Out of the interviewees with a long term reward only two said that they thought about the impact of their actions on their reward. The other four said that they did not think about the reward during their everyday work. This solidifies the evidence that employees with short delay or direct feedback after performance perceive their reward as more valuable than roles with a longer delay.

There seems to be good evidence from the interviews that confirm the effect of time on perceived reward. When taking a closer look though, there is more than just the delay in time between effort and performance feedback that affects the perceived value. I found that the designs of annual pay for performance schemes hold a lot more reward uncertainty by design when

compared to monthly and quarterly ones. Some of the quotes from the interviewees with a long term reward illustrate this well:

“Other than that, it (the performance rating) is mostly an emotional thing by my manager. What is his gut feeling about me?” (Interviewee 12, Female, Age 42)

(31)

25

“I received three percent last year, I also received three percent this year. I have the feeling I have worked a lot harder this year than I did last year. So yea, I can’t really trace back how they assess it very clearly.” (Interviewee 3, Male, Age 27)

“And now at my current employer it is more complicated, because I have less influence on it. I don’t see it as a given that I will receive my variable reward. In that sense it’s really something extra for me. You just need a bit of luck.” (Interviewee 8, Male, Age 33)

I also found that many of the interviewees with a long term reward have targets that are less clearly defined, are broader in scope, span a longer period of time, rely on the performance of the company or others, and are more often subject to their manager’s subjective rating when compared to the interviewees with a short term reward. These differences imply that the long term reward systems contain more uncertainty between effort and performance assessment and have a higher likelihood of a reward prediction error when compared to short term reward systems.

Next, I would like to discuss reward uncertainty. The effects of the reward probability on the perceived reward value are illustrated well by one interview. The interviewee specifically made a distinction between the targets that he could directly influence and the targets that required input from others or were linked to company performance. This distinction had a big influence on the probability of the reward materialising and the perceived fairness of the reward. This interviewee said:

“Yes, but only for two out of my four tasks that can lead to my bonus. Those are directly related (to my performance). In my case these are customer visits and, for example write a business plan or market research. Those are things I do myself and need no input from others for. If I perform these tasks, I directly get rewarded for them.” (Interviewee 4, Male, Age 31)

At the end of the interview the same interviewee wanted to add something to the interview and said:

“So in my case I can do visits, but things like end of year profits make me very reliant on the company performance. And I have to say, that for me, I think that our bonus structure is not right.” (Interviewee 4, Male, Age 31)

The interviewee experienced that he could not directly exert influence on some of his targets, which decreased the predictability of receiving the reward considerably. This also increased

(32)

26 the chance of a reward prediction error, which in return leads to a lower perceived value of the reward. Even though the interviewees were not directly asked for their salary or height of the reward, some interviewees still gave an indication of their reward. This specific interviewee’s reward was clearly above the average rewards for all interviewees. Even though this was the case, the interviewee rated his reward as being low.

“Because of this 7500 euro gross, you still need *pause* you won’t ever achieve this. Because the demands that are set, they are unattainable. The goals for turnover are set pretty high. In practice you might get only half of it. If you look at the time and effort that you need to put in, then it feels kind of low.” (Interviewee 4, Male, Age 31)

An interesting distinction between those with a short and long term reward is how they got rewarded or felt that they were rewarded. One of the questions in the interview was: “How does your organisation reward good performance?”. Out of the six interviewees with a short term reward, all six answered that good performance was rewarded with money. Just one interviewee mentioned that good performance lead to improved secondary labour conditions like a nicer car and phone. The interviewees with a long term reward almost without exception were more elaborate in their

description of how good performance was rewarded and mentioned factors other than monetary reward. The answers ranged from perks, career opportunities, flexible reward schemes that allowed the interviewee to choose for education, more holiday or reward, compliments and more freedom in their daily work. This could infer that employees with a longer term reward scheme have a less direct relation between good performance and variable reward when compared to those with a short term reward.

The research questions were answered well by the findings from the interviews. Employees with a short delay or direct performance feedback after their effort perceive their reward as more valuable, and thus motivating, than employees with a longer delay. Also employees with little or no uncertainty between effort and performance feedback perceive their reward as more valuable, and thus motivating, than roles with higher uncertainty. Other than the effect of delay and uncertainty, I also found that other attributes that are locked inside the mechanism of long and short term reward systems may influence the perceived value.

(33)

27

4.4. Research question 4: Do differences in equity alter the perceived value of the reward?

Almost all interviewees mentioned that they compare their current rewards with reward at previous companies, rewards received in the past, with other companies’ reward and even with reward at companies that are not in the same line of work or sector. Interviewees also compared the input and output of their colleagues to determine the value of their reward. Here are some quotes that show that equity and fairness matter a great deal:

“I would be disappointed if it (the bonus) turned out to be less than one month salary. Alright, if I would have worked in another sector I would have had a thirteenth month without any of this fuzz.” (Interviewee 1, Male, Age 32)

“Sometimes I looked around and saw other people that crafted their targets in a clever way, but I felt they had done a lousy job. In those cases it annoyed me that they received money that I felt they did not deserve.” (Interviewee 12, Female, Age 42)

“That (the reward at competitors) is a lot higher than I could achieve when measured over the course of a year. But looking at the current market, it makes me feel that the amount I can earn extra is really reasonable.” (Interviewee 6, Male, age 26)

“But I think we can’t complain, because the bonuses here are better than those at my previous employer.” (Interviewee 7, Male, age 26)

“There are colleagues that are considerably worse off than I am and there are colleagues that are better off. The colleagues that are considerably worse off than I am do not always perform considerable worse than I do. Colleagues that are better off don’t always perform better than I do.” (Interviewee 1, Male, Age 32)

The answers indicate that not only the absolute magnitude of the reward influences the perceived value of the reward but perhaps equally the magnitude of the reward compared to past experiences, colleagues’ reward and the rewards paid by other companies. One interviewee was quite specific that this did not only apply to the effort and reward, but also to the fairness of the performance assessment:

(34)

28

“In this company it does not matter if you are male or female, young or old. If you get good results, and results are measured by turnover, than you will be rewarded. In my opinion this is the most honest performance reward there is. Normally people say that someone is likable, and therefore this person gets a bonus.” (Interviewee 9, Female, Age 38)

“I’m a woman, and you achieve a certain performance level. But often you see that even when you perform equally that companies discriminate. And this also applies to age of course. With network marketing this is not the case.” (Interviewee 9, Female, Age 38)

During the interviews I also found some evidence that indicates that interviewees see reward as a justification for their past effort. When the interviewees received a reward, they linked it back to the increased effort it took them to earn this reward. Here are some quotes that show these

thoughts:

“If I have the feeling that the rating is just, let’s say half a month salary, than I’m okay with it. If I have the feeling that the amount of effort was extraordinary, I want to be paid extraordinary as well.” (Interviewee 2, Male, Age 27)

“And those are of course the good months, that’s what it is all about. If you do well at that moment, it will get rewarded well.” (Interviewee 7, Male, Age 26)

“It’s a great feeling, because you know you have done something extra to achieve it.” (Interviewee 10, Female, Age 42)

“It gives you the feeling that all the late hours were not in vain and they did have a purpose.” (Interviewee 4, Male, Age 31)

Coming back to the research question, I can conclude that equity indeed alters the perceived value of reward. Individuals compare their effort and reward outcome with their direct environment, but also with other companies and their past employments. Individuals also use reward as a

justification for putting in more effort if the increased effort leads to more reward. Furthermore, I found no differences in interviewees with a short term or long term reward. This would suggest that equity affects the perceived reward regardless of the reward system. Finally there was one

The role of timing and uncertainty on employee reward : are findings from neuroscience applicable to an organisational setting?