Priming Productivity

(1)

Priming Productivity

The Effect of Coffee Smell and Goal State on Productivity

Master Thesis

by

Joriël David Johannes Koops

University of Groningen

Faculty of Economics and Business

MSc Marketing Management

Thesis final version

June 26

th

, 2017

President Kennedyplantsoen 35 1079 SK, Amsterdam The Netherlands +31 646198236 Student number: 1878522 jorielkoops@gmail.com j.d.j.koops@student.rug.nl

(2)

1 Abstract

(3)

2

1. Introduction

Priming is one of the most studied phenomenon in recent social cognition research (Fujita & Trope, 2014), and entails the neurological process “by which exposure to certain cues alters behavior without the person being aware of the impact of the cue on their behavior” (Bargh, 1992). This is further supported by research by Kahneman (2011), who shows that these subtle environmental cues (primes) often influence people when making quick or intuitive decisions, and that unconsciously this influences many of our everyday decision making. This effect happens, because the exposure to a certain cue activates the particular representations or associations in someone’s memory, before engaging in a certain activity or task (King et al., 2016). Because these subtle cues, in the rest of the study referred to as primes, influence people unconsciously in their everyday life, they constitute a very interesting and especially useful research topic within the field of marketing and psychology. Such primes can be used to nudge people towards the purchase of certain products, as research has for example found that customers in wine stores tend to purchase more French wines when French music is playing, and more German wine when German music is playing (Kahneman, 2011). However, it can also be used to create behavioral changes; a study by Bargh et al. (1996) found for example that people tend to walk slower after they have been primed with words associated with elderly people, while Aarts & Dijksterhuis (2003) found that being primed with the concept of a library could make people speak softer. Especially these behavioral changes can have serious consequences in people’s lives, more serious than something as trivial as drinking French or German wine with dinner. In recent studies, serious health consequences of priming have been identified: Chambaron et. al (2015) have found that people tend to eat more energy dense food (higher caloric value) when primed with a sweet odor, which means that people can be unconsciously influenced to consume more unhealthy food. Next to that, King et. al (2016) have found that hand hygiene in hospitals can be significantly improved by spreading a clean, citrus smell in the bathrooms, as this causes people to be much more likely to wash their hands after using the toilet. The topic of priming thus extends beyond the scope of selling products and can have a serious impact on many everyday situations that people find themselves in.

(5)

4

wine; the likelihood that one purchases high caloric food; the likelihood to fill in a specific word in an association task with objects related to the originally primed object (Kahneman, 2011). So far, the research regarding more general effects on people’s overall cognitive functioning is rather limited. Fitzsimons et. al (2008) have, however, written a paper about this: in their study regarding the effects of brand associations on the creativity of participants, they found that being primed with creativity actually invokes creativity. A related topic to creativity, productivity, has however not been studied as of yet by means of priming. This therefore constitutes a gap in the existing priming literature, and specifically raises the question of what the impact of productivity priming on the actual productivity levels of people is. This is a relevant question for businesses in general, because productivity is one of the major causes of better business performance (Prakash et. al, 2017); higher productivity in a business setting, can significantly improve the performance of the overall business.

Furthermore, the use of visual primes within the topic of priming (e.g. visual semantics, as described by Hoedemaker & Gordon (2017), or eye-tracking, as described by Van der Laan et. al (2017)) is the most common method to induce and measure subconscious behavioral effects, even though a prime can consist of “any sensory modality (e.g., visual, auditory, olfactory)” (Chambaron et al., 2015). Other forms of priming, such as olfactory priming (smell), are currently vastly underrepresented in the existing priming literature; when conducting a simple search for “visual prime” in the database of PsychArticles, 19 results are found, while a search for “olfactory prime” gives only 2.1_{Searches in other databases (PsychInfo, PsychBooks, PsychCritique) give}

similar results and therefore point to a clear conclusion: olfactory priming is currently very much underrepresented in the field of behavioral studies and priming, while the processes set in motion by smell may persist even after adaptation, which makes smell a suitable sensory modality for priming purposes (Smeets & Dijksterhuis, 2014).

The conclusion that can be drawn from the above is that both olfactory and productivity are underrepresented in the existing scientific literature regarding priming, and thus form interesting avenues for priming research. This study will aim to contribute to the, as of now fairly limited, scientific knowledge regarding olfactory priming, by looking primarily at the effects of this on productivity.

(6)

5

The olfactory prime chosen for this research is coffee. The choice for coffee is made, as it is the most used psychostimulant in western society (James, 1997), and research has shown that is effective in boosting productivity (Kamimori et al., 2015). The combination of these factors makes coffee a suitable prime, as familiarity is high, even among non-consumers, and consumption has a measurable effect on productivity, an understudied concept within the priming literature. The combination of both factors will constitute one of the main research avenues within this study: the effects of coffee smell priming on productivity levels. This is reflected in the following hypothesis:

H1a: Participants primed with the smell of coffee perform better at a productivity task.

Next to the effects of olfactory priming, this research will also look at the influence of the participants’ goal state on their productivity. Chartrand et al. (2008) define goals as desired end states, or achievements. Shah, Friedman & Kruglanski (2002) have written a paper about goal attainment and state that when “one’s full attention is paid to the focal goal at hand” all alternative pursuits will be put aside, meaning that the person is solely concerned with attaining the focal goal. They integrate their thoughts about goal focus and relevancy in the theory of “Goal Shielding” and state that “the inhibition of alternative goals is generally indispensable for effective self-regulation”; people are therefore only able to really exercise self-control when they have a single goal on which they have to focus. They do however nuance the seemingly black-or-white goal attainment theory, by pointing out that people may have different commitment to their focal goals; the more committed to a specific goal, the greater the inhibition to the alternative goals will be and the more mental resources are allocated to the focal goal.

(7)

6

have multiple alternative goals on their mind, that it will “withdraw attentional resources from the focal goal”, rendering the attainment of the goal less likely.

Before participants actually attain their focal goal, they will be anticipating task-specific elements and their focus will primarily be on that specific goal, thereby subconsciously decreasing the amount of attention paid to alternative goals (Chartrand et al., 2008). This state in which people are focused on a specific goal, in which little attention is paid to alternative goals, is referred to as a state of goal anticipation. The ability to anticipate goals of future actions is an important part of the interaction of humans with their environment (Von Hofsten, 2004). The expectation in this research is that people that are in a state of goal anticipation are more susceptible to the priming with coffee smell, and it is expected to increase their overall productivity, in particular when people are not specifically focused on the task at hand. This is hypothesized as following:

H1: Participants primed with the smell of coffee are more productive given that they are in

a goal anticipation state rather than goal attainment state.

When people actually fulfill their goal, after goal anticipation, they will reach goal attainment. In this state of goal attainment, the original goal has been accomplished and the focus is no longer on this specific purpose, freeing up mental resources to pursue other interests (Chartrand et al., 2008). The expectation in this research is that people that have already achieved their initial focal goal are able to focus on the matter at hand, but are not trying their best to perform optimally anymore, as they have already achieved what they wanted to achieve. Based on these assumptions, the following hypotheses are tested in this paper:

H1b: Participants in the goal anticipation state are more productive than those that are in

the goal attainment state.

(8)

7

The effect of priming with the notion of coffee is hypothesized to have a positive effect on the productivity and the creativity of people, as it is expected that participants have subconscious associations between coffee and the concepts of productivity and creativity. De Dreu et al. (2008) and Baas et al. (2008) describe a distinction within the subject of creativity, that creativity consists of three items: fluency, originality and flexibility. Fluency is hereby a measure of creative productivity and entails the number of ideas, insights or solutions that someone comes up with. Moreover, originality is a measure that refers to the uniqueness of the ideas, insights or solutions that are generated, and flexibility describes how many different cognitive categories and perspectives are touched upon. As the specific interest is on the productive element of creativity in this research, and fluency is a quantitative measure of creativity (Runco & Acer, 2012), the expectation is that participants that are primed with the smell of coffee will perform better at creativity related productivity:

H2a: Participants primed with the smell of coffee are more productive when conducting

creative tasks than those that are not primed.

Furthermore, the same effect as before is expected to be true for creativity related productivity, as it is for productivity in a math related task, with people in a goal anticipation state performing better than those in a goal attainment state:

H2b: Participants in the goal anticipation state are more productive when conducting

creative tasks than those that are in a goal attainment state.

These two sub-hypotheses regarding creativity related productivity form part of the overall hypothesis H2:

H2: Participants primed with the smell of coffee are more productive in a creative

association task, given that they are in a goal anticipation state rather than a goal attainment state.

These main hypotheses (H1 and H2) together form the main research question is used in this

research:

(9)

8

(10)

9

are therefore expected to be stronger for people scoring high on BAS, and lower for people scoring high on BIS (and vice versa). These expectations are given in H3 and H4:

H3: BIS score negatively moderates the effect of coffee priming on attained productivity

scores.

H4: BAS score positively moderates the effect of coffee priming on attained productivity

scores.

Another possible influence on the results, is the regularity of coffee consumption of the participants. The expectation is hereby that regular coffee consumption positively moderates the effects of coffee smell priming on productivity. Regular coffee consumers are assumed to have a stronger conceptual fluency with regard to coffee, and it is therefore more likely that they will associate coffee with productivity, as familiarity is important for the ease of processing (Fennis & Stroebe, 2010). Although the latter assumption may be primarily unconscious, there will also be tested whether the outspoken association of participants influences the effect on productivity. These two assumed effects can be described in the following hypothesis:

H5: The effect of priming on productivity is positively moderated by the frequency of coffee

consumption and coffee associations.

(11)

10

2. Methodology

Research design and participants

In order to test the proposed hypotheses, an experiment in a controlled setting has been conducted. This study has taken place in the research lab of the University of Groningen, where participants were put in individual, isolated booths, and answered a number of questions and took part in a number of tasks by means of the online research tool Qualtrics. The setting was truly a controlled experimental setting, as the only differing environmental variable between the participants was whether they were primed with the smell of coffee or not.

The selection of participants for the study was random, but consisted (almost) exclusively of students of the University of Groningen. They were able to enroll for the study themselves, without interference and/or selection by the researcher, and participation was completely voluntarily. Ahead of their enrollment, the participants were only exposed to communication in the form of advertisement for the study itself in general terms. Participants were to believe that they were about to participate in a study to measure their problem solving abilities and that they would receive €8 in basic compensation, and that they had the opportunity to earn up to an additional €7.50 depending on performance. The purpose of the study was communicated in this manner, to prevent participants from gaining upfront knowledge about the purpose of the study, which might have inhibited actual results.

In total, 211 participants took part in the study and provided usable data for this research. Of these participants, 123 were female and 87 were male, while one was undefined. Of these participants, 98 people were part of the primed experiment group, while 113 people were part of the control group. In order to determine the effects of priming with coffee, a 2 (coffee smell vs. control) x 2 (goal anticipation vs. goal attainment) research design was used.

Research procedure

(12)

11

participants were aware of the upcoming procedure and to reinforce the goal-orientation of the participants, by clearly stating that they were able to earn money during one of the tasks.

After this initial information, participants filled in basic demographic information such as age, gender and educational level. After that, the procedure differed slightly per goal state: in the goal anticipation condition participants were then asked to notify the researcher to come to the booth, at which moment the priming would take place by means of the researcher entering with a freshly made cup of coffee, and in the control group the researcher entered holding a cup of water. If the participants were in the attainment condition instead, they would complete the money earning task

before notifying the researcher. Once notified and inside the booth, the researcher would fill in a

couple of fields on the screen of the participant, mostly irrelevant to the study itself, while putting the cup in front of the respondent, to induce the initial priming effect by means of smell. The forms the researcher filled in included the participants ID, important in case the proper procedure was not followed an the participants had to be excluded afterwards from the data analysis, and some time-filling forms such as a made-up code, the date and day of participation; this enabled the researcher to take some time, in order for the smell to spread in the booth.

Then, the participants would start with the two productivity tasks (see Appendix C). The participants in the anticipation condition would continue afterwards with the money earning task, while the participants in the goal attainment condition had already completed this task at the beginning.

After these tasks, participants were asked to answer a slightly modified/modernized version of the Carver & White (1994) questionnaire to measure BIS/BAS levels of the participants, on a 7-point Likert scale. The questions asked in this part of the survey can be found in appendix D, but also in the original research by Carver & White (1994).

(13)

12 Independent Variables

Prime vs. Control

The prime used in this research is coffee: it is a substance that is well suited to prime productivity as, again, it is “the most widely used psychostimulant in western society” (James, 1997) and is effective in boosting productivity (Kamimori et al., 2015). Furthermore, coffee is widely and frequently consumed in the western world and even non-consumers of coffee will be familiar with the psycho stimulating effects of coffee and be able to recognize the smell, which makes it a suitable smell for olfactory priming purposes regarding productivity and thus suitable for usage in this study (ECF Coffee, 2016).

During the research, the prime was first induced by the researcher, by means of a freshly brewed cup of coffee. The researcher took at least 30 seconds in the booth with every participant, to make sure that the coffee smell could spread enough to be (subconsciously) noticed. A pre-test, prior to the conduction of the actual research, showed that 30 seconds was a proper amount of time to create a recognizable, but not too obvious, coffee smell. Because the coffee smell was spread in a ‘casual’ manner, by simply giving the impression that the researcher was consuming a cup of coffee himself at that moment, the coffee smell should not have alarmed the participants of the intentions of the researcher. This is important, as research by Wegener & Petty (1995) has found that when people become aware of the intended prime and its possible biasing effect, they often try to counter the intended prime effect; a premises that is further supported by Janiszweski & Wyer Jr. (2014). For this reason, there will be a lot of emphasis on the tasks, and specifically the money earning task, to ensure that the participants are not aware of the ongoing manipulation with coffee smell. To ensure that the only salient difference between both conditions is the coffee smell, the researcher in the control group will enter the booth holding a cup of water. This in order to keep the condition of ‘researcher walking in holding a beverage’ constant, while only changing the smell.

(14)

13

recall, one was placed in the middle, and the other two were placed last, to benefit from the ‘recency’ effect to increase recall (Fennis & Stroebe, 2010). These placements should result in the strongest effect of prime reinforcement. The other brands in the list were ‘neutral’ brands, for mineral water, ice tea and tea; products hypothesized to have little or relatively weak brand and product associations in people. The control group saw ten ‘neutral’ beverage brands, thus excluding the three coffee brands.

Goal Anticipation vs. Goal Attainment

Van der Laan et. al (2017) state that relevant goals can be activated in people, by making people strive to a certain goal, which therefore means that goal relevancy is not something that is necessarily present in participants of an experiment, but can be evoked. In this research, the repeated and emphasized notice that participants would be able to earn money at a certain point during the research, was meant to create a state of goal-orientation. In such state, participants are solely focused on the attainment of that specific focal goal, whereby all other pursuits are put aside, as “the inhibition of alternative goals is generally indispensable for effective self-regulation” (Shah et al., 2002).

For the goal anticipation condition, the goal-orientated focus on the third task created a state of goal anticipation during the first two tasks. As participants were expected to be actively invested in the performance of the first two tasks, due to their goal focus on the third task, it gave a good opportunity to analyze the effect of priming on processes of less conscious effort (productivity). In the goal attainment condition, the participant is actually able to achieve the set goal during the first task, and takes part in the remaining two tasks while in this state of goal attainment.

The money earning task was not meant as a measure of productivity in this research, but it was merely a way of inducing the state of goal anticipation or goal attainment among participants. In this task they were asked to find as many words as possible in a word-finder (appendix C). Their score on this task is straightforward, as answers are either in the puzzle or they are not. The puzzle was shown on the screen, and they had to write down the words they found on a piece of paper, which was handed to them by the researcher at the start of the research. Participants received €0,30 for each answer that they found in the puzzle, with a limit of 25 words:

(15)

14 Hereby, f is the amount of correct words found. Moderators

Behavioral Inhibition System (BIS) & Behavioral Activation System (BAS)

During this research, participants were focused primarily on performing optimally during the money-earning task. Because this situation of performing-on-the-spot may have different effects on participants, the BIS and BAS score of every participant was measured at the end of the experiment. BIS score indicates how much people tend to focus on negative consequences and how to avoid them, while BAS score indicates how much people tend to seek positive outcomes and rewards. BIS and BAS are thus opposite processes: people scoring high on BIS focus primarily on possible negative outcomes, while people scoring high on BAS focus primarily on possible positive outcomes (Carver & White, 1994). Except for some slight adjustments (modernizations) in wording, the BIS/BAS test used comes literally from Carver & White’s research.

Coffee Associations

Another hypothesized moderator, is the association that people have with the concept of coffee. This is measured by a simple question in the questionnaire, where participants were asked to indicate which association they have with coffee, whereby they could choose from ‘creativity/productivity’, ‘socializing’ and ‘other, namely: …”. The answers to this question were later transformed into a dummy variable, to compare the relevant association (creativity/productivity) with everything else.

Coffee Consumption

(16)

15 Dependent Variables

Productivity

In the first task, participants tried to solve as many basic math questions as possible in a limited time, a method established by Matarazzo et al. (2010) and Pelham et al. (2007) as a measure of productivity (see appendix C). Productivity is here thus defined as the amount of math problems that a person can solve within a given time. The more productive a participant is, the more math problems that person is able to solve. It is a proper method to measure productivity, as the amount of solved problems very much depends on the focus and effort put in. Despite this, basic math proficiency may cause differences between individuals, but the sample size of x participants is large enough to correct for these individual differences and the results can therefore still be perceived as reliable.

The total productivity score of the participants was determined by the following formula:

Productivity Score = 25 – e – w (2)

Whereby 25 is the maximum number of questions that can be answered, e is the amount of empty/blank questions, and w the amount of wrongly answered questions. What remains is therefore the amount of correct answers that every participant has given, which is their productivity score. An example of a question the participants had to answer, is the following: “Please solve: 8 – 4a + 2b = ? Whereby a=2, b=3”.

Creativity-related productivity

(17)

16

asked participants to come up with as many uses of everyday objects (e.g. brick, newspaper) as possible within a certain amount of time. In this research has been chosen to restrict the amount of answers to five per category, to prevent participants from spending too much time coming up with many answers for one category, and to use five different object.

Before the participants started with this creativity task, the coffee prime was reinforced with the participants of the prime group: they were asked to state familiarity with ten beverage brands, among which four were recognizable coffee brands. The other brands in the list were ‘neutral’ brands, of mineral water, ice tea and tea; products hypothesized to have little or relatively weak brand and product associations in people. The control group would see only ten ‘neutral’ beverage brands, excluding the coffee brands.

The participants’ total creativity score in this task was determined by the following formula:

Creativity related productivity score = a – u (3)

(18)

17

3. Results

Dataset - descriptions and exclusions

The initial post-experiment dataset consisted of 232 responses: 112 participants were in the prime group, 120 in the control group. There is a slight imbalance between both groups, because more ‘test’ responses were recorded for the prime group than for the control group; all these tests responses have been excluded from the dataset, as they are not relevant for the research outcome. Furthermore, 21 respondents have been excluded because they did not follow the proper procedure of the experiment: they failed to notify the researcher at the first moment, in which the priming should have occurred. Participants failing to do so have been excluded from both the prime and the control group, as the simple fact of the interruption may still cause uncontrollable side effects in the control group, and the missing of the coffee prime renders them useless for the prime group. After these exclusions, the dataset consisted of 211 responses (see appendix A, table 23). Of these 211 valid responses, 123 were given by women, while 87 were given by men.

With regards to the main experiment conditions (prime/control & goal state), 98 out of 210 participants were in the prime condition, while the remaining 113 were in the control condition. Furthermore, 110 were in the goal state ‘attainment’ and 101 ‘anticipation’. For clarity purposes, this is shown below in table 1, and general descriptives of the data are shown in table 2.

Factor/variable Number Mean Standard

deviation BIS score 210 4.676 .966 BAS score 210 5.343 .745 Coffee Association (1: productivity/creativity, 0: other) 210 .529 .500

Regular coffee drinkers (1:yes, 0:no) 210 .490 .501 Math questions correctly answered 209 14.202 5.422

Math percentage correct 209 .907 .117

Creativity task – amount of answers 211 13.490 7.549

Table 2 – descriptives of main variables Condition Number Prime 98 Control 113 Anticipation 101 Attainment 110

(19)

18

H1: Participants primed with the smell of coffee are more productive when conducting basic math

tasks, given that they are in a goal anticipation state rather than goal attainment state.

In order to study whether the hypothesis is statistically true, a univariate ANOVA analysis is conducted with the variables “Prime/Control”, “Goal State” (anticipation/attainment) and “Math Productivity”. In the process of studying the main hypothesized effect of H1, two more broad, general sub-hypotheses are analyzed as well, to get a better understanding of potential direct effects of the two main independent variables on productivity. The hypotheses that are tested are given in the results in such way that the effects are shown first, after which the analysis outcomes of H1 with interaction effect is given:

H1a: Participants primed with the smell of coffee are more productive than those that are

not primed.

The (2x2) ANOVA testing the hypothesis, with outcome F(1,209)=.323, p-value=.570 (p>.10), shows that being primed with the smell of coffee (M=14.429, SD=5.405) does not significantly increase participant’s productivity in the form of correct answers on a basic math task, when compared with those participants in the control condition (M=14.000, SD=5.460). The results are furthermore shown in table 3.

Based on this analysis, the conclusion can be drawn that there is insufficient evidence to accept H1a.

H1b: Participants in the goal anticipation state are more productive when conducting basic

math tasks than those that are in the goal attainment state.

The ANOVA analysis furthermore shows that there is no significant direct effect of goal state on the productivity of the participants. The output of F(1,209)=.749, p-value=.388 shows that there is no significant difference between the participants in the attainment condition (M=14.509, SD=.517) and those in the anticipation condition (M=13.859, SD=.545). The output of this analysis is shown in table 4.

Factor F-value P-value

Prime vs Control .323 .570

(20)

19

The conclusion here too can be drawn that there is insufficient evidence to accept H1b.

Despite goal state not having a significant impact on productivity, there is in fact a mildly significant difference in the accuracy of participants answers on the math task, whereby accuracy is measured as “amount of correct answers / total amount of answers”, and therefore measures the percentage of correctly given answers. The output of an ANOVA analysis gives that F(1,209)=3.236 and p-value=.074, showing that there is a significant difference between those participants in the attainment condition and those in the anticipation condition. Hereby, those in the attainment condition (M=.921, SD=.100) scored higher than those in the anticipation condition (M=.892, SD=.132).

Therefore, despite that participants in different goal states do not differ regarding productivity (amount of correct answers), they do in fact differ in the accuracy of their given answers (percentage of total answers correct), with a mildly significant result. The results are given in table 5.

Factor/Variable F-value P-value Goal State

(Attainment vs Anticipation)

3.236 .074

Table 5 – Effects of priming on productivity

Although no direct effect was found between both main conditions (prime & goal state), the possibility still exists that there is an interaction effect. For this purpose the output from the 2x2 ANOVA was studied, with inclusion of the interaction variable between the prime condition and the goal state. This analysis gives the following output: F(1,209)=0.02, p-value=.964, showing that there is no interaction effect between the prime condition and goal state with regard to productivity in a basic math task. Therefore, the outcomes of prime x attainment (M=14.725, SD=5.568), prime x anticipation (M=14.106, SD=5.383), control x attainment (M=14.322, SD=5.447) and control x

Goal State (Attainment vs Anticipation)

.749 .388

(21)

20

anticipation (M=13.635, SD=5.387) have no significant differences. The results are given in tables 6 and 7, and visualized in figure 1.

Figure 1 - Visualization of prime x goal state effect on productivity

Based on the above analysis, the conclusion can be drawn that there is insufficient evidence to support H1, that participants primed with coffee in the anticipation state are more productive. Next

to that, priming and goal state had no impact on the productivity of participants either. However, those in a state of goal attainment were significantly more accurate in their answers than those in the goal anticipation state.

H2: Participants primed with the smell of coffee are more productive in a creative association task,

given that they are in a goal anticipation state rather than a goal attainment state.

In order to test this hypothesis, another univariate ANOVA analysis has been conducted with the conditions prime and goal state, and here their effects on the productivity of the participants in a

Prime x Goal State .02 .964 Table 6 – Effects of priming and goal state on productivity

Factor Goal State Mean Standard Deviation Prime Attainment Anticipation 14.725 5.568 14.106 5.383 Control Attainment Anticipation 14.322 5.447 13.635 5.387

Table 7 – Mean and SD of goal states on productivity

0 5 10 15 20

Prime x Attainment Prime x Anticipation Control x Attainment Control x Anticipation

(22)

21

creativity task has been measured. Again, just as with H1, first the direct effects of the two main

factors – prime & goal state – are analyzed, after which the interaction, testing H2, is analyzed.

H2a: Participants primed with the smell of coffee are more productive when conducting

creative tasks than those that are not primed.

The univariate ANOVA testing the hypothesis, gives the outcome F(1,209)=.009, p-value=.925, and shows that being primed with the smell of coffee (M=13.541, SD=.764) does not have a significant influence on the creativity related productivity of the participants, as the amount of different answers is not significantly different from the control condition (M=13.442, SD=.712). The results are given in table 8.

The conclusion from the above analysis is that there is insufficient evidence to accept H2a.

H2b: Participants in the goal anticipation state are more productive when conducting

creative tasks than those that are in a goal attainment state.

The ANOVA analysis testing the effect of goal state on the creativity related productivity of participants, gives the following outcome: F(1,209)=.004, p-value=.951. This shows that goal state does not significantly influences creativity related productivity, as attainment (M=13.473, SD=.722) and anticipation (M=13.505, SD=.753) conditions are not significantly different. These values are also given in table 9.

The conclusion, here too, can be drawn that there is insufficient evidence to accept H1b.

Besides the fact that no direct effect was found between both main conditions and creativity related productivity, there was also no interaction effect found of both conditions. The ANOVA with inclusion of the interaction variable between the prime condition and the goal state gives the following output: F(1,209)=.184, p-value=.668. This shows that there is no interaction effect between the prime condition and goal state with regard to creativity related productivity, and that prime x attainment (M=13.294, SD=8.629), prime x anticipation (M=13.809, SD=7.580), control

Prime vs Control .009 .925 Table 8 – Effects of priming on creativity related productivity

Goal State (Attainment vs Anticipation)

.004 .951

(23)

22

x attainment (M=13.627, SD=7.572) and control x anticipation (M=13.241, SD=6.539) are not significantly different. The results are shown in table 10 and visualized in figure 2.

Based on the above analysis, the conclusion can be drawn that there is insufficient evidence to support H2, that participants primed with coffee in the anticipation state are more productive in

creativity related tasks. Moreover, priming and goal state had no significant influence either.

H3: BIS score negatively moderates the effect of coffee priming on attained productivity scores.

In order to analyze the effects of BIS scores on participants performance, the participants were divided in categories of “above” and “below” median regarding BIS. For this purpose, the median of the overall dataset was calculated, which gave the result of: 4.71; all participants with a BIS score of <4.71 were then coded “0” for below, while all participants with a score of >4.71 were coded “1” for above.2

With this new median BIS variable, two univariate 2x2x2 ANOVA analyses has been conducted, including both two main conditions and BIS (prime condition, goal state & BIS median score).

2_{Both for BIS and BAS: because of a cut-off point dividing the sample in two, some of the participants were right in} the middle, and have been excluded from the BIS and BAS analyses.

Prime x Goal State .184 .668

Table 10 – Effects of priming and goal state on creativity related productivity

0 5 10 15 20

Prime x Attainment Prime x Anticipation Control x Attainment Control x Anticipation

Prime x Goal State on Creativity Related Productivity

(24)

23

The first of the two tests measures the effects on productivity, while the second measures the creativity related productivity.

The first, direct, analysis of BIS on productivity, shows no direct effect: the ANOVA shows with F(1,186)=.001, p-value=.979 that there is no significant difference between those participants with high BIS (M=14.672, SD=.527) and those with low BIS (M=14.652, SD=.566). Next to this result, there is also no interaction effect between BIS and goal state, with F(1,186)=.338, p-value=.562. However, there is in fact a strong significant effect found between the prime condition and BIS score: the ANOVA shows with F(1,186)=5.221 and p-value=.023 that the interaction between both factors influences productivity. This effect can be seen in table 11.

Hereby, participants that were in the control condition scored better on productivity when they had a lower BIS score (M=15.326, SD=.770), and scored lower when they had a high BIS score (M=13.579, SD=.713). However, in the primed group, the effects were reversed: participants that were primed and had a high score on BIS scored better on productivity (M=15.765, SD=.777), while those primed with a low BIS scored lower on productivity (M=13.977, SD=.830). These values can be found in table 12.

Furthermore, another 2x2x2 ANOVA analysis has been conducted to measure the influence of BIS on creativity related productivity. In contrast with the first measure of productivity, a significant effect of BIS score on creativity related productivity was found in this analysis: F(1,186)=2.778, p-value=.097. Hereby, the higher the BIS score of a participant, the higher their productivity on a creative task was. There was here, however, no interaction effect found between BIS and either of the two main conditions. The interaction of BIS and prime gave the output F(1,186)=.216,

p-Factor F-value P-value

BIS Median .001 .979

Prime * BIS Median 5.221 .023 Goal State * BIS Median .338 .562 Prime * Goal State * BIS Median .617 .433

Table 11 – Effects of BIS on productivity

Factor BIS Median Mean SD

Prime Below Above 13.977 .830 15.765 .777 Control Below Above 15.326 .770 13.579 .713

(25)

24

value=.643. showing that there was no significant difference between prime x low BIS (M=11.881, SD=1.229), prime x high BIS (M=14.323, SD=1.150), control x low BIS (M=12.565, SD=1.140) and control x low BIS (M=13.942, SD=1.055). Next to that, the output of BIS and goal state showed with F(1,186)=2.111, p-value=.148 that there was no significant difference either between attainment x low BIS (M=11.461, SD=1.217), attainment x high BIS (M=15.035, SD=1.045), anticipation x low BIS (M=12.985, SD=1.153) and anticipation x high BIS (M=13.230, SD=1.160). No three-way interaction effect was found, with F(1,186)=.482, p-value=.488. These results are shown in table 13.

The conclusion regarding H3 can be drawn, that BIS score does not negatively moderate

productivity scores, but BIS scores by themselves do influence creativity related productivity, and some interaction effects between BIS and priming do exist.

H4: BAS score positively moderates the effect of coffee priming on attained productivity scores.

To test whether BIS and BAS are indeed more or less opposites of one another, as theory suggests, a correlation analysis has been conducted between BIS and BAS scores. The outcome is that there is no significant correlation between both variables, with a correlation of -.043, p-value=.531. Therefore, both BIS and BAS have been analyzed as separate factors.

Similarly to the BIS scores, so too have the BAS scores been divided by their median, to create a group coded 0 to indicate those participants below the median, while the group above the median is coded 1. The median score for the BAS scores, which is the average of the three subcategories “Reward Responsiveness”, “Fun Seeking” and “Drive”, amounts to 5.40, which also functions as the cut-off point between both groups.

Another 2x2x2 ANOVA analysis has been conducted, again with the two main conditions, but this time with BAS median instead of BIS median. The analysis shows that BAS score does not significantly influence productivity, with F(1,186)=.201 and p-value=.654. Next to that, there were

BIS Median 2.778 .097

Prime * BIS Median .216 .643 Goal State * BIS Median 2.111 .148 Prime * Goal State * BIS Median .482 .488

(26)

25

also no interaction effects found between BAS and the two main conditions: BAS and prime were insignificant with F(1,186)=1.841, value=.177, and BAS and goal state with F(1,186)=.336, p-value=.563. There was, however, a significant interaction effect between prime, goal state and BAS score, with F(1,186)=4.169, p-value=.043. The means of the analyses show, that in the control group, participants in the attainment condition with a low BAS score, scored better than average (M=15.458, SD=5.823), while those with a high BAS score were less productive (M=13.607, SD=5.216). On the contrary, participants in the primed group in the attainment condition, scored worse than average when they had a low BAS (M=12.889, SD=4.143), while they scored higher than average when they had a high BAS (M=16.346, SD=5.549). In the anticipation group there was not much deviation from the overall mean of M=14.613, SD=5.216, as shown by a one-sample t-test: anticipation (M=13.859, SD=5.362, p=.650) did not differ significantly from the mean, while attainment (M=14.509, SD= 5.481, p=.089) did. This indicates that the effect mainly appears to be happening in the attainment condition. The results and the corresponding means (also of the anticipation condition) can be found in table 14 and table 15.

In contrast, there were in fact multiple significant results at the 10% level, with regard to creativity related productivity. BAS score itself did not yield any significant results, with F(1,186)=.464, p-value=.497. However, the interaction between prime and BAS did yield a significant result, with F(1,186)=3.853, p-value=.051, and the interaction between goal state and BAS resulted in

BAS Median .201 .654

Prime * BAS Median 1.841 .177 Goal State * BAS Median .336 .563 Prime * Goal State * BAS 4.169 .043

Table 14 – Effects of BAS on productivity

Factor Goal State BAS Median Mean SD

Prime Attainment Anticipation Below Above Below Above 12.889 4.143 16.346 5.549 15.037 4.784 14.400 5.902 Control Attainment Anticipation Below Above Below Above 15.458 5.823 13.607 5.216 14.087 4.944 14.520 5.197

(27)

26

F(1,186)=3.086, p-value=.081. Hereby, participants in the prime condition scored higher when they had a low BAS score (M=14.296, SD=1.165) compared with a high BAS (M=11.265, SD=1.241), while participants in the control condition scored higher when they had a high BAS score (M=14.014, SD=1.053) compared with a low BAS score (M=12.545, SD=1.117). Furthermore, participants in the attainment condition scored higher when they had a high BAS score (M=14.080, SD=1.043) compared to those with a low BAS score (M=12.847, SD=1.194), while participants in the anticipation condition scored higher when they had a low BAS score (M=13.994, SD=1.086) compared with those with a high BAS score (M=11.200, SD=1.250). Finally, there was a significant interaction effect between prime, goal state and BAS score, with F(1,186)=2.955, p-value=.087; hereby, specifically participants in the prime group, who were in the anticipation state differed substantially from the rest: those with high BAS scored very (!) poorly (M=8.800, SD=1.977), while those with a low BAS (M=15.815, SD=1.473) scored above average (M=13.333, SD=7.707). These results are given below in tables 16 and 17.

Regarding H4, the conclusion can be drawn that there is no moderating effect of BAS score on the

effects of priming on productivity, but the interaction does positively moderate the effect on creativity related productivity.

BAS Median .464 .497

Prime * BAS Median 3.853 .051 Goal State * BAS Median 3.086 .081 Prime * Goal State * BAS 2.955 .087

Table 16 – Effects of BAS on creativity related productivity

Factor Goal State BAS Median Mean SD Prime Attainment Anticipation Below 12.778 9.903 Above 13.731 8.771 Below 15.815 7.158 Above 8.800 6.889 Control Attainment Anticipation Below 12.917 7.077 Above 14.429 7.681 Below 12.174 6.401 Above 13.600 7.077

(28)

27

H5: The effect of priming on productivity positively moderated by the frequency of coffee

consumption and coffee associations.

To study this hypothesis, again two 2x2x2 ANOVA analyses have been conducted. The first studies the effects of the frequency of coffee consumption, whereby the original data is transformed to a dummy variable: those participants that drink at least one cup of coffee per day are coded as “1”, regular coffee drinkers, while the participants that do not drink at least one cup per day are coded “0”, non-regular coffee drinkers. Coffee associations have also been coded into 0’s and 1’s, whereby participants that associated coffee with productivity were coded 1, while those participants that associate coffee with something else, are coded 0.

(29)

28

Furthermore, no significant results were found when looking at creativity related productivity either. Coffee consumption gave as output F(1,210)=1.414, p-value=.236 with regular coffee consumers M=. The interaction between prime and coffee consumption gave as output F(1,210)=.002, p-value=.965, showing there was no significant difference between the categories prime x regular (M=12.933, SD=1.105), prime x non-regular (M=14.146, SD=1.082, control x regular (M=12.738, SD=1.041) and control x non-regular (M=14.044, SD=1.007). Finally, the interaction between goal state and coffee consumption showed that there was no significant different between attainment * regular (M=12.681, SD=1.052), attainment * non-regular (M=14.202, SD=1.017), anticipation * regular (M=12.990, SD=1.095) and anticipation * non-regular (M=13.988, SD=1.073), with output of F(1,210)=.061, p-value=.805. No three-way interaction was found, with F(1,209)=.866, p-value=.353. These results can be seen in table 19.

Besides regular coffee consumption, the effects of coffee associations on productivity are tested. The outcomes show, again, that there is no significant effect directly, and neither through a two-way interaction. The analysis of coffee associations gave as output F(1,209)=.009, p-value=.926, showing that there is no difference regarding productivity for those that associate coffee with productivity/creativity (M=14.132, SD=.515) compared to those that associate it with something else (M=14.202, SD=.543). Furthermore, prime and coffee associations as interaction also gave

Regular Coffee Consumption .001 .973 Prime * Regular Coffee

Consumption

.142 .707

Goal State * Regular Coffee Consumption

.646 .422

Prime * Goal State * Regular Coffee Consumption

3.267 .072

Table 18 – Effects of regular coffee consumption on productivity

Regular Coffee Consumption 1.414 .236 Prime * Regular Coffee

Consumption

.002 .965

Goal State * Regular Coffee Consumption

.061 .805

Prime * Goal State * Regular Coffee Consumption

.866 .353

(30)

29

an insignificant result, with F(1,209)=2.444, p-value=.120 showing that prime x productivity/creativity association (M=13.808, SD=.746), prime * other associations (M=15.048, SD=.796), control x productivity/creativity association (M=14.457, SD=.710) and control x other associations (M=13.356, SD=.739) are not significantly different from one another. Next to that, goal state and coffee associations resulted in F(1,209)=.476, p-value=.491, furthermore showing that attainment x productivity/creativity association (M=14.687, SD=.710), attainment x other associations (M=14.241, SD=.747), anticipation x productivity/creativity association (M=13.577, SD=.746) and anticipation x other associations (M=14.163, SD=.789). Finally, there is in fact a significant three-way interaction effect found between all three factors, with output F(1,209)=5.633, p-value=.019. In this three-way interaction, the results most notably differing from the overall mean (M=14.201, SD=5.422) are, again, in the attainment condition. In the control and attainment condition those participants that associate coffee with productivity/creativity score above average (M=15.875, SD=.951), while those that associate it with something else score lower (M=12.481, SD=1.035). Furthermore, in the prime and attainment condition, those participants that associate coffee with productivity/creativity score lower than average (M=13.500, SD=13.500), while those that associate it with something else score above average (M=16.000, SD=1.076). In the anticipation condition very little deviation occurs from the mean. The results can be found in table 20, and the complete table of values in Apendix A table 25.

The same analysis on creativity related productivity, gives a similar result. However, coffee associations is here almost significant at the 10% level, with F(1,210)=2.499, p-value=.116, while the three-way interaction is no longer significant. Participants that associate coffee with creativity/productivity, scored higher than those that associate coffee with something else. However, the result falls just outside the 10% significance level and can thus not be used as ‘proof’ about an existing relationship. Furthermore, the interaction effect of prime and coffee association was not significant, with F(1,210)=.010, p-value=.919, while the interaction of goal state and

Coffee Associations .009 .926 Prime * Coffee Association 2.444 .120 Goal State * Coffee Association .476 .491 Prime * Goal State * Coffee Association 5.633 .019

(31)

30

coffee association gave F(1,210)=.001, p-value=.980. Finally, the three-way interaction was not found to be significant as well, with outcome F(1,210)=.544, p-value=.462. These results can be found in table 21, and the values in Apendix B table 26.

Concluding on H5, coffee consumption did not have a moderating effect on productivity, or

creativity related productivity. Neither did coffee associations, although there was a near significant direct effect of coffee associations on creativity related productivity, whereby a creativity/productivity association lead to higher scores than other associations did. Also, a three-way interaction effect was found with regard to productivity, but not for creativity related productivity.

Additional results

Additionally, analyses were conducted to test whether one of the goal states, or one of the moderators, had a significant effect on the money that participants earned during the experiment. When factoring in the BAS and BIS scores of participants in as well, three of the four factors were directly significant with p<.10. Prime is significant with F(1,186)=3.773, p-value=.054, goal state with F(1,186)=2.939, p-value=.088, and BAS with F(1,186)=5.010, p-value=.027. Only BIS is not significant, with F(1,186)=1.025, p-value=.313. Participants in the prime condition earned more money than those in the control condition, those in the anticipation condition earned more than those in the attainment condition, and finally those with a high BAS score earned more than those with a low BAS score. These results are shown in table 22.

Coffee Associations 2.499 .116 Prime * Coffee Association .010 .919 Goal State * Coffee Association .001 .980 Prime * Goal State * Coffee Associations .544 .462

Table 21 – Effects of coffee association on creativity related productivity

Prime vs Control 3.773 .054 Goal State (Attainment vs

Anticipation)

2.939 .088 BAS (High vs Low) 5.010 .027 BIS (High vs Low) 1.025 .313

(32)

31

4. Discussion

In this research, there was no direct effect found of either of the two main conditions, prime and goal state, on productivity scores, nor was an effect found of the two conditions on creativity related productivity. Moreover, there is no interaction effect between the two main conditions for either measure of productivity. However, a result not hypothesized but arguably part of the quality of performance was found: participants in the attainment condition did make significantly (p=.074) less mistakes during the math productivity task. A possible explanation for this, is that participants in the anticipation phase were primarily focused on the upcoming money earning task, while the participants in the attainment phase knew they had already completed the money earning task, and were now fully able to focus on the task at hand as they were less distracted. Another, additional, explanation might be that the participants in the attainment phase were more focused, because they already completed a task and were already in a focused state of mind because of the first task, while the participants in the anticipation phase started with this task and may have been less focused going in.

Although intuitively BIS and BAS reflect contrary effects and therefore it appears logical that both are correlated, but in line with the study by Carver & White (1994) and Heubeck et al. (1998) in the research was found that BIS and BAS are not significantly correlated, and therefore seem to be independent of one another (p=.531). This implies that people who are afraid to fail at things, may still be motivated to go after things and to try to get positive results, for which reason both concepts have been analyzed separately in this research.

(33)

32

control group, the opposite effect happens: participants with a high BIS perform better. There does not seem a logical explanation for this, which means that some unexplained effect is taking place, or that it is down to coincidence. Then, with regard to creativity related productivity, a slightly significant effect of BIS is identified (p=.097). Hereby, participants that scored higher on BIS, scored better on the task. This result is in line with the findings by Kim & Kwon (2017), who use a form of merged BIS/BAS scaling, and have found that on a BAS/BIS scale participants scoring high on BAS (low BIS) with regard to reward responsiveness scored worse on creativity, and those with a low BAS (high BIS) score better on creativity. This explains the found effect that participants with high BIS performed better on the creativity task.

With regard to productivity, BAS turned out to be insignificant directly, and in interaction with the two main factors. However, an odd result was found when interacting BAS with both main factors: the three way interaction actually influenced the results in the attainment state (p=.043). The effect found here is, again, in line with research by Germans & Kring (2000), who state that participants with high BAS perform better in response to positive stimuli than those with low BAS; the coffee is hereby the positive stimulus (James, 1997). The idea that coffee reinforces the effects of BIS and BAS seems to be confirmed in this research, and is in line with the research by Germans & Kring (2000) and Carver & White (1994). In the three-way interaction in this research, participants who were in the control and the attainment condition, with a low BAS score, performed better than average, while those in the same conditions with a high BAS score performed below average. On the other hand, participants in the prime and attainment condition scored below average when they had a low BAS, while they performed better when they had a high BAS. All the effects that differ from the average here, happen in the attainment condition. A possible explanation for the effects in the control condition is that those participants with a low BAS score are less goal-focused on the money earning task and are better able to focus on the current task at hand, while those with high BAS scores are satisfied that they have completed the goal-task and are less focused on the remaining tasks, and might see the other tasks as less interesting, or perhaps even annoying, side activities.

(34)

33

that BIS and BAS have reversed effects for productivity and creativity related productivity; high BIS scores seem to hinder productivity but boost creativity related productivity, while high BAS scores seem to boost productivity, but hinder creativity related productivity (Germans & Kring, 2000; Kim & Kwon, 2017). For the prime group these effects are reinforced and pronounced, while for the control group these effects appear more random. Moreover, another interaction effect with regard to BAS was found; between goal state and BAS. Participants in the attainment condition scored significantly better when they had a high BAS, while those in the anticipation condition scored significantly better when they had a low BAS. This found effect is in line with research by Carver & White (1994), who concluded that people in an anticipation goal state perform better in a laboratory setting when exposed to positive stimuli.

(35)

34

5. Conclusion

There are no direct effects of the main conditions on either measure of productivity, and neither do BIS and BAS seem to have much influence on productivity in a normal (control) situation. However, in the prime condition the expectations of BIS and BAS are confirmed with the theory, in the sense that people with low BIS and those with high BAS performed significantly better on math related productivity. The explanation for this is that coffee priming reinforced and enhanced underlying emotions and beliefs: if people were unsure of themselves, coffee may have increased those feelings; if people were sure of themselves, coffee may have increased those feelings as well. With creativity related productivity these effects seem to a certain extent reversed: through interaction effects, participants with high BAS scores performed worse on the creativity task, while those with high BIS scores performed better. Contrary to expectations, no effect of coffee consumption or coffee association was found on productivity or creativity related productivity levels in this study.

Somewhat in contrast with the above, the hypotheses that primed participants and those with a high BAS score performed significantly better were confirmed for the money earning task. Also, participants in a goal anticipation state performed better than those in a goal attainment state, which was also hypothesized. Most of the hypothesized effects in this paper therefore seem true, but only when participants are fully committed to the task at hand, and the effects are not found as hypothesized when participants might not be fully committed to the task at hand, because their mind may be on other things, such as the upcoming or already completed money earning task instead. It might therefore still be true that priming with coffee smell can make people more productive, when they are fully focused and while conducting certain tasks. This makes it a potential contributor to for example business performance. However, further research is required to assess whether this effect is also true for more everyday business tasks, rather than for tasks that are short and lead to high and immediate rewards. If during further research is found that coffee actually increases productivity in the workplace, it is advisable to purposely spread a subtle coffee smell around, or to place a coffee machine at the center of the workplace.

(36)

35

replicate the study with a more equal demographical distribution, and to test whether there is a difference between students and people that are (full-time) employed. Perhaps the association of coffee and productivity is stronger among those employed and of a higher age group, as coffee consumption has been a habit for a longer period of time, and the notion of coffee could internally be stronger associated to actual productivity performance in the workplace. Moreover, the participants of the study came from many different countries and regions. Coffee consumption differs by country and by culture, and so may the associations with coffee and productivity effects. For this reason it is recommended that the study is replicated within a certain geographical region, in which coffee consumption and associations are more homogeneous.

(37)

36 References

Aarts, H., & Dijksterhuis, A. (2003). The silence of the library: environment, situational norm, and social behavior.Journal of personality and social psychology,84(1), 18.

Baas, M., De Dreu, C. K., & Nijstad, B. A. (2008). A meta-analysis of 25 years of mood-creativity research: Hedonic tone, activation, or regulatory focus?.Psychological

bulletin,134(6), 779.

Bargh, J. A. (1992). The ecology of automaticity: Toward establishing the conditions needed to produce automatic processing effects.The American journal of psychology, 181-199.

Bargh, J. A., Chen, M., & Burrows, L. (1996). Automaticity of social behavior: Direct effects of trait construct and stereotype activation on action.Journal of personality and social

psychology,71(2), 230.

Bélanger, J. J., Lafreniere, M. A. K., Vallerand, R. J., & Kruglanski, A. W. (2013). When

passion makes the heart grow colder: The role of passion in alternative goal suppression. Journal

of Personality and Social Psychology, 104(1), 126-147.

Carver, C. S., & White, T. L. (1994). Behavioral inhibition, behavioral activation, and affective responses to impending reward and punishment: The BIS/BAS Scales. Journal of personality

and social psychology, 67(2), 319.

Chambaron, S., Chisin, Q., Chabanet, C., Issanchou, S., & Brand, G. (2015). Impact of olfactory and auditory priming on the attraction to foods with high energy density. Appetite, 95, 74-80.

Chartrand, T. L., Huber, J., Shiv, B., & Tanner, R. J. (2008). Nonconscious goals and consumer choice.Journal of Consumer Research,35(2), 189-201.

De Dreu, C. K., Baas, M., & Nijstad, B. A. (2008). Hedonic tone and activation level in the mood-creativity link: toward a dual pathway to creativity model.Journal of personality and

(38)

37

Depue, R. A., & Zald, D. H. (1993). Biological and environmental processes in nonpsychotic psychopathology: A neurobehavioral perspective. In Costello, Charles G. (Ed), Basic issues in

psychopathology, (pp. 127-237). New York, NY, US: Guilford Press.

European Coffee Federation. (February 2016). Coffee Market Overview February 2016. Retrieved from http://www.ecf-coffee.org/publications/coffee-facts-and-figures.

Fennis, B. M., & Stroebe, W. (2010). The psychology of advertising. Hove, East Sussex: Psychology Press.

Fitzsimons, G. M., Chartrand, T. L., & Fitzsimons, G. J. (2008). Automatic effects of brand exposure on motivated behavior: how apple makes you “think different”. Journal of consumer

research, 35(1), 21-35.

Fujita, K., & Trope, Y. (2014). Structured versus unstructured regulation: On procedural mindsets and the mechanisms of priming effects. Social Cognition, 32, 68-87.

Gable, S. L., Reis, H. T., & Elliot, A. J. (2000). Behavioral activation and inhibition in everyday life.Journal of personality and social psychology,78(6), 1135-1149.

Germans, M. K., & Kring, A. M. (2000). Hedonic deficit in anhedonia: support for the role of approach motivation.Personality and Individual Differences,28(4), 659-672.

Gilhooly, K.J., Fioratou, E., Anthony, S.H., & Wynn, V. (2007) Divergent thinking: strategies and executive involvement in generating novel uses for familiar objects. British Journal of

Psychology, 98(4), 611-625.

Gray, J. A. (1978). The 1977 Myers lecture: The neuropsychology of anxiety. British Journal of

Psychology, 69, 417–434.

Gray, J. A. (1981). A critique of Eysenck’s theory of personality. New York, NY: Springer.

Gredebäck, G., & Kochukhova, O. (2010). Goal anticipation during action observation is influenced by synonymous action capabilities, a puzzling developmental study. Experimental

Brain Research, 202(2), 493-497.’

(39)

38

Higgins, E. T. (1997). Beyond pleasure and pain.American psychologist,52(12), 1280.

Hoedemaker, R. S., & Gordon, P. C. (hi2017). The Onset and Time Course of Semantic Priming During Rapid Recognition of Visual Words. Journal of experimental psychology. Human

perception and performance, 1-22.

James, J. E. (1997).Understanding caffeine: A biobehavioral analysis. Sage Publications, Inc.

Janiszewski, C., & Wyer Jr., R.S. (2014). Content and process priming: A review. Journal of

Consumer Psychology 24(1), 96-118.

Kahneman, D. (2011).Thinking, fast and slow. New York: Farrar, Straus and Giroux.

Kamimori, G. H., McLellan, T. M., Tate, C. M., Voss, D. M., Niro, P., & Lieberman, H. R. (2015). Caffeine improves reaction time, vigilance and logical reasoning during extended periods with restricted opportunities for sleep.Psychopharmacology,232(12), 2031-2042.

Karremans, J. C., Stroebe, W., & Claus, J. (2006). Beyond Vicary’s fantasies: The impact of subliminal priming and brand choice. Journal of Experimental Social Psychology, 42(6), 792-798.

Kim, B. N., & Kwon, S. M. (2017). The link between hypomania risk and creativity: The role of heightened behavioral activation system (BAS) sensitivity.Journal of Affective Disorders,215,

9-14.

King, D., Vlaev, I., Everett-Thomas, R., Fitzpatrick, M., Darzi, A., & Birnbach, D. J. (2016). “Priming” hand hygiene compliance in clinical environments. Health Psychology, 35(1), 96.

Marguc, J., Van Kleef, G. A., & Förster, J. (2015). Welcome interferences: dealing with obstacles promotes creative thought in goal pursuit. Creativity and Innovation

Management, 24(2), 207-216.

Matarazzo, Kristina L., Amanda M. Durik, and Molly L. Delaney. "The effect of humorous instructional materials on interest in a math task."Motivation and emotion34.3 (2010): 293-305.

Priming Productivity