Festinger's Insufficient Rewards, Cognitive Dissonance, and Motivation



Often credited as being the father of cognitive dissonance, Leon Festinger (1961) sought to explain motivation in the absence or near-absence of rewards. He theorized that organisms experience dissonance, or feelings of discomfort, when two cognitions or perceptions are in direct conflict. In an attempt to reconcile these uneasy feelings and tension, the organism will either reflect on the past in a different light so that the conflict seems less painful, or it will view the reward as having more value than originally thought.

When rewards were completely absent or obviously lacking in value (partial rewards), Festinger (1961) observed the organisms still performing the tasks. This is contrary to reinforcement theory, where rewards are used to promote an activity and extinction used to eliminate an activity (Skinner, 1938). Therefore, Festinger theorized that in order for the organisms to decrease dissonance, they must actually begin to take some form of enjoyment in the activity itself, rather than the insufficient reward. In order to study his theory, Festinger created a series of trials using rats. He hypothesized that the rats receiving a reward only occasionally would continue to perform the task longer than the rats receiving a reward every time the task was completed, once the reward was eliminated. In other words, once behavior extinction was attempted, by eliminating the reward 100% of the time for all rats, Festinger expected that the rats that received the infrequent reward would continue to perform the task longer than the rats which always received the reward. This would mean that the rats receiving the partial reward decreased their dissonance by taking some enjoyment in performing the task itself.

In a more modern study, authors Garaus, Furtmuller, and Guttel (2016) investigated the influence that small or insufficient rewards have upon autonomous learning and motivation in online students. This builds upon the theory of cognitive dissonance by attempting to discover and explain why students who receive smaller rewards increase their intrinsic motivation to learn. Garaus et al don’t minimize the effects that large rewards can have upon high academic performance, but their goal was to determine the influence of large and small rewards upon intrinsic and extrinsic motivations.

According to the theory of cognitive dissonance, small rewards should create a disparity between cost-benefits (Garaus et al, 2016). If the reward isn’t worth the trouble, then the individual must rectify the dissonance by internalizing the reward, or learning to enjoy the act of learning itself. The obverse is that while large rewards may also motivate a person to a high-level of academic achievement, it would only be for extrinsic motivation. This means that removing the large reward should also decrease the motivation to learn. Therefore, the authors hypothesized that small rewards increase autonomous motivation to learn among online students (Garaus et al, 2016).


The classic study performed by Festinger examined an observed behavior found in other experiments investigating cognitive dissonance (Aronson & Mills, 1959). As Festinger’s experiment is easy to replicate, ran through multiple trials, and the results reinforce previous research, it should be considered a reliable study. Likewise, it has a high level of validity since the results seem to measure what is attempting to be ascertained. There is a high correlation between the more infrequent the rewards and the amount of trials until the extinction of the behavior. Although many other studies have demonstrated the external validity of Festinger’s partial rewards experiment, his study itself lacks generalizability. At the time, it was unclear whether or not rats possess the same cognitive abilities as humans in regard to rewards and behavior. Just because rats continued a behavior in the absence of rewards doesn’t mean that all organisms will go through a similar cognitive process.

Garaus et al (2016) used over 1,300 students to increase their validity by having a rather large sample size. The researchers used two groups, each with approximately the same number of participants – a control group receiving no rewards and an experimental group receiving a very small reward. These groups represent the two conditions of the independent variable. Those in the small reward group received a maximum of .70 points based on the number of homework questions answered correctly, whereas the control group received no additional points for completed homework. The dependent variable was the participant’s score on a preparatory assignment. The results confirmed the partial rewards theory in which students who received the extra credit on homework improved their preparatory scores from 40.5% to 46.6%.

By creating a pilot study (Garaus et al, 2016), the researchers increased the internal validity of their experiment. Likewise, with a higher internal validity, especially with real-world practicality, the generalizability of their findings (external validity) is increased. Additionally, by creating a team to determine a proper insufficient reward system, having a larger sample size with significant correlations, and finding similar results to other studies shows that their study has reliable conclusions.

By using these methods, both studies have attempted to maximize the validity and reliability of results. Likewise, both investigated prior research and constructed models to produce results that can be further built upon. After all, if the results can’t be replicated or are found to be unreliable, then the research performed may have been meaningless. Also, studies revealing large problems in research methodologies, such as those discovered in the Stanford Prison Experiment (Haney, Banks, & Zimbardo, 1973), are forever remembered for the issues brought to light rather than the research performed.

Theories and Past Research

Most studies performed investigating cognitive dissonance involved giving the subjects paltry rewards or an expectation that an event would almost assuredly occur. Each study noticed to some varying degree that the subjects exposed to these conditions, as opposed to full rewards, experienced higher resistance to the extinction of a particular behavior (Festinger, 1961). For instance, an experiment conducted by Aronson and Mills (1959) demonstrated how people looking for group inclusion viewed the reward of inclusion as being more valuable depending how much pain or embarrassment they had to endure. Those who gained easy entrance into the group viewed the group as boring, whereas those who had a more difficult time gaining entry found the group to be much more exciting. Being that the reward was inadequate to the goal of inclusion, it was theorized that the subjects experiencing the more difficult time actually took some pleasure in the struggle to attain group inclusion.

Studies performed by Crum, Brown, and Bitterman (1951) and also Scott and Wike (1956), showed that rats had a greater resistance to a behavior’s extinction when their rewards were delayed (a form of partial reward). Furthermore, as demonstrated by studies performed by Wike and McNemara (1957) and Fehrer (1956), the more trials run with delays or partial rewards, the more resistance to extinction the rats possessed. In essence, the more pain, effort, and time an organism endures, the more likely it is to experience resistance to extinction, presumably due to mitigating dissonance. Again, Festinger (1961) theorized that an organism will decrease dissonance by taking some pleasure in performing the activity itself rather than the reward it might earn.

If autonomous motivation leads to deeper processing and better performance (Gagne & Deci, 2005), how does a modern academic environment increase this effect? Since many institutions now offer online curriculum, face-to-face interactions and other rewards are difficult to implement (Garaus et al, 2016). This means that students receiving an online education would be greatly benefitted by being intrinsically motivated so that the act of learning itself is the reward. Or, as authors Ryan and Deci (2000) discuss in contrast, students who are externally motivated to learn do so to achieve external rewards.

Building on the theory of cognitive dissonance (Festinger, 1957), small rewards should coerce an individual to rationalize past behavior by developing attitudes compatible with that behavior and evolving favorable attitudes toward the task instead of the rewards (Pfeffer & Lawler, 1980). In other words, when rewards are seen as insufficient, it can lead to increased intrinsic motivation. Therefore, something as simple as positive verbal feedback can augment autonomous motivation (Deci, 1971).

Changed Over Time

The discipline of social psychology attempts to understand how individuals behave and respond to their environments, especially how they see themselves in the world around them. Festinger clearly established differences in behavior between receiving full rewards and partial rewards, revising previously understood motivations, such as reinforcement theory (Skinner, 1938). Similarly, Garaus et al updated the theory of cognitive dissonance in relation to small rewards to fit a modern-day purpose. Educational theories that increase learning and autonomous motivations are constantly changing to incorporate new understanding. Both studies improved upon prior research and the modern study went a step further in providing an everyday context to some, that of an online education.

In order to create a modernized study with real-world applications, Garaus et al (2016) used input from a team of professionals with various disciplines to develop their insufficient reward system – something Festinger (1961) didn’t have to do dealing with rats. Additionally, the modern experiment used a pilot study to increase internal validity (Garaus et al, 2016). Festinger’s approach to dealing with simpler organisms, rats, required much less in terms of preparation. He was only trying to answer a previously observed caveat of the reward-behavior relationship.

Where Festinger demonstrated how the process of cognitive dissonance works in regard to partial rewards, Garaus et al updated that knowledge by using it in a modern context. Furthermore, researchers Garaus et al had to think in much different terms while performing their study on human participants in an online setting, something Festinger may have never even considered a possibility. Yet, each study had a similar goal, to try and change behavior to increase the likelihood the behavior will continue even in the absence of extrinsic rewards.

Both studies used experimental designs, including control groups. However, it must be noted that the modern study performed by Garaus et al (2016) was a quasi-experimental design due their inability to randomly assign participants to the different conditions. Following in the tradition of scientific research, experimental designs have more control, increasing both their validity and reliability.

Human Behavior

Different Time Periods

Previously, it was thought that rewarding a particular behavior helped to solidify it in an organism (Skinner, 1938). However, Festinger’s research demonstrates quite well that partial rewards increase dissonance so that an organism has to justify this by either viewing the task as enjoyable itself or viewing the reward as greater than it actually is. Since the rats received their reward first during the extinction trials (being fed beforehand), no reward was even present to prompt them to continue their prior behavior. This is why those receiving the rewards most frequently during the initial trials were also the first to experience extinction of their behavior. Those infrequently receiving the reward continued the task far longer. This lends evidence to the theory that the rats must have experienced some level of dissonance and began to take pleasure in the task rather than the reward to lessen their dissonance.

In improving upon academic performance by introducing small rewards, authors Garaus et al add to the body of knowledge that dissonance creates behavioral changes that are more resistant to extinction and beneficial to intrinsic motivation. This empirical support lends credence to the theory that in the presence of insufficient rewards, people will often develop enjoyment for the activity itself, causing an increase in intrinsic motivation.

These studies both reinforce the theory of cognitive dissonance and its effects upon behavior. The results coincide with other studies performed on the topic of partial and insufficient rewards. While Festinger’s study examined the varying degree of partial rewards and strength of the ensuing behavior, Garaus et al found how to use those principles in practice to foster increased motivation and autonomy in online learning. In other words, both advanced the theories, understanding, and practicality of social psychology with their experiments.

As discussed, both studies follow scientific experimentation in attempting to increase validity and reliability. Other than the quasi-experimental aspect of Garaus et al (2016), which was only due to the inability of random assignment, both studies built upon prior reliable and valid research to expand upon the understanding of the time. Both studies can easily be replicated to determine any problems with reliability, and the variables are easy enough to measure so there is little ambiguity. In fact, the correlated reward-behavior paradigm fits expected results so well that this too should be considered an additional check for validity and reliability.

Results and Conclusions

Festinger’s (1961) results aligned perfectly with the previous research performed. The more infrequent or partial the reward, the longer it took for the rats to change their behavior. Likewise, the longer the reward was delayed, the longer the task continued. Festinger’s theory of cognitive dissonance explains these results quite well and has huge impacts upon social psychology.

Likewise, the results found by Garaus et al (2016) aligned well with what was expected. Homework assignments completed by the small rewards group were nearly four times higher than the control group (Garaus et al, 2016) even though the extra credit was negligible. Furthermore, performance on preparatory exercises increased from 40.5% of correct answers to 46.6%. In fact, every aspect investigated showed a significant positive correlation between the small rewards group and better performance. These results further demonstrate the effects of dissonance and partial or insufficient rewards upon changing behavior.

Through Festinger’s literature review and experiments, he made evident that partial rewards are a much bigger influence on a continued behavior than full rewards. Furthermore, he showed that the more infrequent or paltry the reward, the more the behavior would linger – a direct correlation. This not only comports well with his theory of cognitive dissonance, but provides tangible evidence for the effects dissonance has upon changing behavior. When an organism has to experience pain, suffering, embarrassment, or any other form of stress, it will often contrast that with the value of the reward received. In order to lessen the feelings associated with dissonance, the organism will either take some form of enjoyment in the activity itself, or place greater value on the reward than initially thought. The experiments and research performed by Festinger exemplify this theory with incredible accuracy.

These implications can transform much of the academic world. As online learning becomes more prevalent, institutions and professors strive to find ways to increase interest, minimize procrastination, bolster autonomy, and improve overall performance. While large extrinsic rewards have also been shown to increase academic performance, they undermine intrinsic motivation (Deci, Koestner, & Ryan, 1999). This means that the person is working towards the reward and not the act itself. In the absence of a large reward, the motivation to perform well is also lost. Yet, small rewards demonstrate that the act of learning is internalized, promoting higher autonomy, better performance, and continued enjoyment in learning.

Research Trends

Social psychology attempts to ascertain individual motivations behind behavior and attitudes, so it is critically important to understand why people (and other organisms) perform contrarily to what is expected when experiencing dissonance. In order to test and demonstrate this phenomenon, Festinger (1961) created multiple situations where different groups of rats received a reward (food) upon reaching their food box. However, each group received the reward to varying degrees. Some received the reward 100% of the time, others 33%, 50%, or 67% of the time. Then, Festinger added another component to a different series of testing – delayed rewards. And finally, to test the severity of effects each reward system had upon behavior, Festinger discontinued the rewards entirely, feeding the rats before the trials took place, to observe how long it took for behavior extinction.

Building upon Festinger’s partial rewards research, Garaus et al (2016) used these findings in a present-day context. If online students are intrinsically motivated and more autonomous in their learning, this would add immense benefits and understanding to the academic world. In order to test their theory of small rewards increasing autonomous motivation to learn, authors Garaus et al (2016) worked with other researchers, professors, and students to develop a reward system that would seem insufficient to most people.

This recent study had an added benefit over many others in that the researchers were able to perform a real-world study in a natural setting, rather than using questionnaires and self-reports. This trend reversal makes this recent study rare in modern times. Additionally, respondent truthfulness avoids being a factor, as it is in many modern studies, because of the use of a real-world environment. Furthermore, a control group was tested first without having any knowledge of an extra credit reward system, as is the norm in most present-day research.

Much to the interest of every psychological discipline is how behavior is influenced by rewards. Festinger’s (1961) insufficient rewards study using rats attempted to explain a previously observed behavior contradictory to the various existing reward theories. Where Festinger built upon prior research to correct an observed error in existing theories, Garaus et al (2016) built upon previous research to put insufficient rewards theory into practice. In research, it is important to increase our knowledge base and understanding of psychological theories and principles, but it is also important that our environments are continuously updated to incorporate these new understandings. This is the fundamental difference between these two studies and their context within all of the psychological fields.

Influence on Society

Benefit Society

As previously discussed, Festinger’s research has had monumental impacts over the years. Not only did his study significantly correlate partial rewards to behavioral extinction resistance, but he also found that the smaller or more infrequent the reward, the stronger the resistance (Festinger, 1961). This study has been a catalyst for other research investigating cognitive dissonance, the influence of rewards on behavior, and motivation.

The benefits to social wellbeing in the Garaus et al (2016) study are quite evident. Building on the theory of cognitive dissonance and the effects of small rewards on behavior, the authors directly applied these findings to a very modern, practical context – online education. The results of this study show that it is possible to manufacture intrinsic benefits from the use of partial rewards. In other words, due to this study, we have evidence that rewards such as grades provide an extrinsic benefit to learning, but once they are removed so too is a motivational component. However, by using partial rewards, even in the absence of grades, the intrinsic motivation to learn is retained.

The benefits to society are the same as the impacts to each individual, if used in a positive, productive manner. Not only would someone benefit personally from the findings of these partial-reward studies, specifically with education, but that person would also enjoy in the benefits of a more educated, motivated society. For instance, if monetary rewards had less impact on behavior and were replaced with intrinsic motivations, we might not be having such serious debates on climate change, overpopulation, food and water scarcity, national security, and more. Not only would society benefit from the increased desire to do more for its own sake, but each of us as individuals would also reap the benefits of a more intrinsically motivated populous.

Negative Impact

Neither the classical study nor the modern study have any appearance of negative impacts to participants, future research, or society as a whole. In fact, Festinger’s (1961) original research on partial rewards took extra measures to ensure the rats were taken care of and fed appropriately. Likewise, Garaus et al (2016) enlisted a team of professionals to assist in the setup and implementation of their study not only to increase validity and reliability, but also to safeguard the wellbeing of students and their academic performance.

If there is any adverse impact realized from either study, it is in how the findings are used or implemented in society. For instance, if the use of partial rewards is known to increase intrinsic motivation and enjoyment in the task rather than the reward, this could potentially be used immorally by corporations looking to exploit workers. After all, the findings suggest that the less someone is paid, the more that person will take delight in the job. Of course there are more variables present in performing a job, but nonetheless, a malevolent company might see these results as a reason to keep wages low or even decrease them.

Everyday Life

When discussing everyday applications and influences of these studies, it is difficult to limit it to a few words. It seems like this is similar to asking, now that we understand gravity due to Newton, how does this impact life? Well, how does it not impact life, might be the more appropriate question. Just as Festinger probably never conceived of the internet when performing his rat studies, we probably cannot conceive of the possibilities inherent in behavioral modification. Garaus et al (2016) increased intrinsic motivation to learn, which itself could revolutionize education, especially pertaining to an online world. What other intrinsic motivations could be enhanced? Imagine a world where people worked, learned, helped others, cared for the environment, etc., not because they felt compelled to, but because they truly valued those acts for themselves.

The purpose of psychological research, specifically in regard to social psychology, is to ascertain why people act the way they do and to predict future behavior. If it is possible to alter behavior in a more positive, productive manner, it seems that could have a monumental impact on society. Festinger’s prior research on cognitive dissonance has provided an immense framework to build upon, including his follow-up study investigating partial rewards on rats (Festinger, 1961). Furthermore, modern researchers can expand upon these findings to improve upon education, as seen in the Garaus et al (2016) study, among other possibilities.

Social Change

Where Festinger advanced the knowledge base of psychology for its time, Garaus et al used those advancements to change behavior in a real-world setting. Through both experiments, among many more, there is clear evidence that behavior can be changed through use of a partial reward system. Although this could possibly be used for nefarious means, it also can be used to create a deeper understanding in those it benefits, as demonstrated by Garaus et al (2016). If autonomous learning can be promoted and increased due to these theories, so can many other positive behavioral changes. If a highly educated and motivated population is desired (which would seem to be a good thing), then those same principles should be applicable to other behaviors such as employee motivation, driving behavior, and volunteer work, among many more.

Professional Experience

In considering some practical uses incorporating the findings of Festinger and Garaus et al, at least one has been in practice for decades in elementary schools across the country. As students work diligently on assignments and practice their newly acquired skills and learnings, their hard work is often rewarded with a simple sticker, quite often in the form of a gold star. Some teachers go a step further and hand write a small smiley face on a student’s well-done assignment. When hours of work are potentially met with a paltry reward, the actual reward becomes the task itself. So, in practice, schools have been using Festinger’s partial reward theory for many years, and in essence, teaching children to enjoy learning itself, rather than the receipt of an insufficient reward.


It is without question that Festinger looked to expand upon his theory of cognitive dissonance while also explaining problems noticed in earlier research, such as reinforcement theory. Like most researchers, his work built upon and corrected the preceding experimentation. Most if not all psychological disciplines deal with why people behave in certain ways and how to predict that behavior accurately. Festinger noticed an exception to traditional reward theories and performed his research with the goal of filling that knowledge gap.

Authors Garaus et al performed their quasi-experiment with similar intentions. Although there was no perceived exception to the theory of cognitive dissonance or partial rewards theory, the researchers attempted to use prior findings to discover modern contexts and implementations. This is one of the purposes of any research – to update our knowledge base to fit our modern world. As scientists explore more of our universe, human behavior, cognitive abilities, etc., the working models of our understanding will also have to be updated. For instance, the theory of gravity, as proposed by Newton, had a flaw that no one could explain regarding the bending of space due to gravity (Siegfried, 2015). Two centuries later, Albert Einstein developed his theory of general relativity which precisely explained gravity’s effect upon spacetime, light waves, and all matter.

It is the very essence of scientific research, be it general relativity or partial rewards, to build upon prior knowledge and correct exceptions. This is the influential nature of all scientific research. Regarding partial rewards theory, when new information arises from developing technologies (as it did with the internet) or from contradictions in results, it should be assumed that future researchers will work to better understand how previous studies fit into their modern society. The scientific method itself is a process of determining gaps in our knowledge and working to fill those gaps with sound evidence that better corresponds with the modern world.



Aronson, E., & Mills, J. (1959). The effect of severity of initiation on liking for a group. The Journal of Abnormal and Social Psychology, 59(2), 177-181. doi:10.1037/h0047195

Crum, J., Brown, W. L., & Bitterman, M. E. (1951). The effect of partial and delayed reinforcement on resistance to extinction. The American Journal of Psychology, 64(2), 228-237. doi:10.2307/1418669

Deci, E. L. (1971). Effects of externally mediated rewards on intrinsic motivation. Journal of Personality and Social Psychology, 18(1), 105-115. doi:10.1037/h0030644

Deci, E. L., Koestner, R., & Ryan, R. M. (1999). A meta-analytic review of experiments examining the effects of extrinsic rewards on intrinsic motivation. Psychological Bulletin, 125(6), 627-668. doi:10.1037/0033-2909.125.6.627

Fehrer, E. (1956). Effects of amount of reinforcement and of pre- and postreinforcement delays on learning and extinction. Journal of Experimental Psychology, 52(3), 167-176. doi:10.1037/h0045052

Festinger, L. (1957). A theory of cognitive dissonance. Evanston, IL: Row, Peterson

Festinger, L. (1961). The psychological effects of insufficient rewards. American Psychologist, 16(1), 1-11. doi:10.1037/h0045112

Gagné, M., & Deci, E. L. (2005). Self-determination theory and work motivation. Journal of Organizational Behavior, 26(4), 331-362. doi:10.1002/job.322

Garaus, C., Furtmuller, G., & Guttel, W. H. (2016). The hidden power of small rewards: the effects of insufficient external rewards on autonomous motivation to learn. Academy of Management Learning & Education, 15(1), 45-59. doi:10.5465/amle.2012.0284

Haney, C., Banks, C., & Zimbardo, P. (1973). Interpersonal dynamics in a simulated prison. International Journal of Criminology and Penology, 1, 69-97.

Pfeffer, J., & Lawler, J. (1980). Effects of job alternatives, extrinsic rewards, and behavioral commitment on attitude toward the organization: A field test of the insufficient justification paradigm. Administrative Science Quarterly, 25(1), 38-56. doi:10.2307/2392225

Ryan, R. M., & Deci, E. L. (2000). Intrinsic and extrinsic motivations: Classic definitions and new directions. Contemporary Educational Psychology, 25(1), 54-67. doi:10.1006/ceps.1999.1020

Scott, E. D., & Wike, E. L. (1956). The effect of partially delayed reinforcement and trial-distribution on the extinction of an instrumental response. The American Journal of Psychology, 69(2), 264-268. doi:10.2307/1418156

Siegfried, T. (2015). Einstein's genius changed science's perception of gravity. Science News, 188(8).

Skinner, B. F. (1938). The behavior of organisms; an experimental analysis. New York, London: D. Appleton-Century Company, Incorporated.

Wike, E. L., & Mcnamara, H. J. (1957). The effects of percentage of partially delayed reinforcement on the acquisition and extinction of an instrumental response. Journal of Comparative and Physiological Psychology, 50(4), 348-351. doi:10.1037/h0043517

32 views0 comments