Hard Work Pays off How Effort Justification and the Illusion of Control Intersect to Change the Perceived Success of Pure Chance Outcomes

Hard Work Pays Off: How Effort Justification and the Illusion of Control Intersect to Change the Perceived Success of Pure Chance Outcomes A thesis submitted in partial fulfillment of the requirements for graduation with Honors in Psychology. 2023 ii Table of Contents Acknowledgements ............................................................................................................ iv Abstract ............................................................................................................................... v Hard Work Pays Off: How Effort Justification and the Illusion of Control Intersect to Change the Perceived Success of Pure Chance Outcomes ................................................. 1 Illusion of Control ........................................................................................................... 2 Effort Justification .......................................................................................................... 5 Possible Interaction Between the Illusion of Control and Effort Justification ............... 8 Participants .................................................................................................................... 10 Materials ....................................................................................................................... 11 Procedure ...................................................................................................................... 13 Results ............................................................................................................................... 15 Discussion ......................................................................................................................... 18 Illusion of Control ......................................................................................................... 18 Effort Justification ........................................................................................................ 21 Interaction ..................................................................................................................... 23 Future Research ............................................................................................................ 24 Conclusions ................................................................................................................... 24 Real World Implications ............................................................................................... 25 References ......................................................................................................................... 27 iii Acknowledgements iv Abstract The illusion of control refers to the tendency for people to falsely believe that their actions affect a chance outcome. Effort justification refers to the inclination for people to prefer identical reinforcements that follow high effort tasks compared to low effort tasks. The combination of these two phenomena has been scarcely examined in past research. Evolutionary theory suggests potential for a novel relationship between the two concepts. The present study employs a 2 x 2 between-subjects design to examine this possibility. Healthy adult participants were either administered an easy or difficult set of anagram puzzles as an effort manipulation. They were subsequently either assigned raffle ticket numbers or allowed to choose custom ticket numbers as an illusion of control manipulation. Participants were then asked what probability they thought that they had of winning the raffle. Participants in the high effort and chosen numbers conditions were expected to perceive a high probability of winning compared to the low effort and assigned numbers conditions respectively, replicating past research. In addition, the condition that combined high effort and chosen numbers was expected to produce an interaction, with an elevated perceived probability of winning. Results indicated that all three hypotheses were non-significant; past research was not replicated, and effort justification and the illusion of control did not interact. These results provided implications for how to continue with future research on probability-based decision making. Keywords: effort justification, illusion of control, perceived probability v Hard Work Pays Off: How Effort Justification and the Illusion of Control Intersect to Change the Perceived Success of Pure Chance Outcomes Humans often succumb to logical mistakes when making decisions. The pioneering research of Daniel Kahneman and Amos Tversky has provided evidence for many mechanisms of cognition that can produce such errors. These researchers challenged the prevailing consensus among the psychological community that people are usually rational, and that any departure from rationality can be attributed to emotional influences (Kahneman, 2011). There are in fact many cognitive factors that can cause irrational decisions and judgments regardless of a person’s emotional state. These factors often induce people to disregard relevant statistical evidence in favor of a simple or familiar conclusion when they make a decision. Heuristics are widely studied mental shortcuts that function to facilitate decision making and make choices less cognitively demanding, but can often mislead people (Kahneman, 2011). For example, the availability heuristic prompts people to prioritize information that is readily available (Tversky & Kahneman, 1973). This tendency, however, can create a bias for an option purely because it is prevalent in a person’s mind. Availability is extremely influential in society and is especially apparent in media coverage. People often overestimate the frequency of a certain event, such as violent crime, based on how often it appears on the news. Therefore, broadcasters tend to utilize sensationalism and focus on such events more than they truly occur. The result is a consistent overestimation of the frequency of some events (and underestimation of others) based on availability. Heuristics are certainly a primary tool used in cognition to make judgements, but judgements are also affected by other factors, such as skill (Kahneman, 2011). Albert Bandura described outcome expectancy as the expectation that certain antecedent behaviors (skills) will lead to specific outcomes. Additionally, he asserted that a selfefficacy expectation refers to a person’s confidence that they can perform such antecedent behaviors, such as believing one can learn to play an instrument (with an outcome of impressing others) without having any prior experience (Bandura, 1977). These two concepts, outcome expectancy and self-efficacy expectancy, are perceptions that may not accurately reflect objective reality: people often overestimate or underestimate how much an outcome is influenced by their behavior. This notion constitutes another interesting way that people may be misled by their own cognition when making decisions: inaccurate estimations of the effect that skill factors have on an outcome. Illusion of Control One such decision making error involving skill perception is the illusion of control, which is the overestimation of the influence that personal behavior has over outcomes of complete chance (Langer, 1975). When evaluating pure chance outcomes, humans often irrationally change their perception of probability based on irrelevant factors. For instance, an individual waiting for an elevator door to close might repeatedly push the “door close'' button, even though in many elevators this button is disabled. Rather than pressing the button because of an experienced contingency between buttonpressing and door closure, people continue this action because the behavior seems relevant, and increases the feeling of being in control (Clark & Wohl, 2021). 2 Similarly, people do not always act rationally when forming judgments about the effect that their behavior has on an outcome. Even in a situation where the probability of an outcome is unchangeable, participants’ actions still affect their perceived evaluation of chance (Langer, 1975). This phenomenon means that people can fail to respond differently to controllable and uncontrollable situations. Specifically, factors that impact skill-oriented tasks can manifest similarly in tasks of pure chance, where skill is irrelevant. Martinez et al. (2011) found that participants who played a French roulette computer game after watching previous players win a large amount of money increased risk taking in subsequent trials. Additionally, they found that the sense of control over winning was greater when participants were aware of the previous gamblers’ winnings, creating the incorrect notion that the game was a skill-based task, demonstrating an illusion of control. People are also more likely to irrationally increase their estimation of winning probability based on superstitions. For example, Davis et al. (2000) looked at illusory factors that determined players bet types and sizes in the casino game of craps. They observed that players would talk to the dice asking for certain lucky numbers, blow on the dice or stack the dice on top of each other with the digits of the desired number facing up, among other superstitious rituals. These behaviors were motivated by the drive to personally influence the outcome. Superstitious behaviors often demonstrate an illusion of control as people believe that they have more control over outcomes than they do. In the same study, researchers looked at the effects of active and passive participation on betting behaviors in casino games (Davis et al., 2000). Players who got to throw the dice, otherwise known as shooters (active participants), differed in multiple 3 ways from non-shooters (passive participants), as they not only felt more confident in the outcomes, but also wagered more money. Similarly, Ladouceur et al. (1987) examined the impact of active participation in the casino game of roulette, where either the croupier (passive) or the player (active) would spin the ball and then the players would make their bets either before or after the spin began. They found that subjects indicated greater levels of confidence in their success and would bet more money when spinning the balls themselves. Additionally, the subjects would bet more money when they placed their bets before rather than after the spin, suggesting a sense of greater perceived control. According to Langer (1975), the more skill-related factors (choice, stimulus familiarity, active involvement, and competition) present in a game of chance, the more an illusion of control is likely to occur. In her study examining choice, Langer had participants split into two experimental conditions. Participants in the choice condition were allowed to pick a lottery ticket from options that had different football players on the tickets, while those in the no choice condition were assigned a random ticket. As a dependent measure, participants were then asked what price they would be willing to sell their ticket for. Participants who chose between multiple options valued their tickets much more than participants who were assigned a ticket. A theoretical perspective for the illusion of control is that it functions as a mechanism to enhance self-esteem by prompting people to give themselves credit for successes (Bradley, 1978). Under this framework, the amount of time that people devote to obtaining a desired outcome should be an important factor in determining how they perceive their control over the event; higher commitment should require more justification to maintain self-esteem. Another factor that may influence outcome 4 confidence is the extent to which participants can draw a causal relationship between their own actions and an outcome, as opposed to an external cause (Yarritu et al., 2014). For both factors, participants’ self-esteem is thought to be enhanced by greater involvement in a task since they must interpret an impending positive result as a consequence of their behavior. Yarritu et al. (2014) provided evidence that participants with more personal involvement elicited a stronger illusion of control than participants with less personal involvement. The present study aims to investigate if this same theory holds with effort, which may have a similar effect on self-esteem as involvement. Effort Justification The illusion of control is not the only factor that can affect the perceived value of a prize. When receiving a reward for completing a task, participants tend to assign a higher value for identical rewards that follow a high effort task than those that follow a low effort task (Festinger, 1957). This tendency is thought to happen because assigning a greater value to a reward justifies the greater effort that was needed to obtain it. Festinger’s cognitive dissonance theory refers to the underlying psychological tension that is created when peoples’ behavior is not consistent with their thoughts and beliefs. Cognitive dissonance may arise when great effort is expended for a relatively small reward, in that the behavior (high effort) does not justify the meager payoff. The resulting dissonance can be reduced or eliminated by enhancing the perceived value of the reward. Thus, effort justification is a way in which individuals reduce cognitive dissonance by placing a higher value on a reward that takes more effort to earn (Zentall, 2010). Interestingly, this theoretical explanation of effort suggests many similar mechanisms as Bradley’s self-esteem theory used to explain the effect that personal involvement has on the illusion of control 5 (Bradley, 1978). These theoretical similarities further encourage the potential for effort to affect the illusion of control in a similar way as personal involvement. Rosenfeld et al. (1984) used Festinger's work on cognitive dissonance as a theory for the mechanisms behind effort justification. They conducted an experiment in which subjects were either in a low effort or high effort number circling task. The experiment was conducted by an experimenter who was either high in attractiveness (friendly and pleasant) or low in attractiveness (rude and unpleasant). They expected ratings of the task to be affected by both effort justification (higher ratings for the high effort task) and impression management (higher ratings in the presence of an attractive experimenter). They found that dissonance was relevant: when individuals expended more effort, they justified the effort by increasing their rating of the task. Experimenter attractiveness had no overall effect, but moderated effort justification. Effort justification only occurred for participants in the presence of the interpersonally unattractive experimenter. Those in the attractive experimenter conditions could attribute their increased effort to the experimenter’s presence and did not need to rely on their enjoyment of the task. Thus, effort justification may not occur in all scenarios, and only occur when other factors (such as experimenter attractiveness) cannot be used to reduce dissonance. Past studies suggest that non-human animals are also affected by effort justification. This finding indicates that cognitive dissonance might be a misleading term to account for effort justification. Lydall et al. (2010) reported that rats in a high effort condition found sucrose more palatable than rats in a low effort condition. This effect likely occurred because rats in the high effort condition worked harder for the same sucrose. This study displays that similarly to humans, rats put more value on rewards 6 when they expend more effort for them. It is important to note that because effort justification is not unique to humans, it may not require complex cognitive processes. This notion suggests that effort justification might have a broader and more fundamental evolutionary value. Zentall (2010), for example, suggests that effort justification could be caused by a general behavior principle, such as within-trial contrast. This concept refers to the relative difference between the greater enjoyment of a reward and the lesser enjoyment of a task to earn the reward. High effort tasks typically have a greater disparity of enjoyment than low effort tasks, so the subsequent reward is perceived as better. Zentall’s (2010) within trial contrast theory supports an evolutionary explanation for effort justification because it takes a broader range of animal behavior into account beyond complex cognitive functions that may be exclusive to humans. Whatever the cause, it is clear that participants increase the perceived value of an award when a task is more effortful. This effect happens in everyday life as well. People increase the value of rewards such as coupons or lottery tickets if they are linked to high effort. Kim and Song (2020) gave participants a coupon following either no antecedent activity, a lower effort survey, or a higher effort game. They reported that those who performed an antecedent behavior to earn a coupon placed a higher value on the reward than those who did not complete an activity. Additionally, the higher effort game group indicated greater value for the same reward compared to the lower effort survey group, suggesting that participants demonstrated effort justification. Furthermore, Kim and Kim (2008) asked participants to imagine that they were going to buy a lottery ticket and pick its numbers. Participants in a high effort group were asked to write down every number between 1 and 60 in ascending order and choose five of those numbers. Participants in a 7 low effort group were simply instructed to choose five numbers between 1 and 60 without writing the rest out. Participants were then asked how willing they were to sell their lottery ticket to someone else. Those in the high effort condition were less likely to sell their ticket, signifying an effort justification effect. This finding suggests that consumer evaluations and price judgments are higher following high effort. Possible Interaction Between the Illusion of Control and Effort Justification Both the illusion of control and effort justification can affect the way people interpret random events. Predictions for how these two factors interact may come from an evolutionary perspective on the illusion of control. Humans and animals can both benefit from perceived control in a situation (Perlmuter & Monty, 1977). Seligman & Maier (1975) found that dogs demonstrate learned helplessness in situations where they have no control over an unpleasant outcome. That helplessness persists later, even when control over the situation returns. This concept of learned helplessness demonstrates an advantage for individuals who continue to believe their behavior affects a situation (as they do when demonstrating an illusion of control). Thus, the illusion of control could serve as a protective mechanism against potentially harmful behavioral phenomena such as learned helplessness. This theory encourages an evolutionary framework because developing an illusion of control seems to increase an organism’s chance of survival in some scenarios. For instance, an animal that continues to believe it has control over finding food (even if it is largely governed by chance) is less likely to starve. Glass and Singer (1972) documented that people who were given a choice to terminate an unpleasant noise attempted to solve an unsolvable puzzle more times on average than participants who weren’t given the choice to terminate that unpleasant noise. This finding 8 suggests that a perceived sense of control (in this case over termination of the noise) may encourage greater effort and persistence. The present study aims to investigate if these effects may be related. Specifically, greater effort is predicted to enhance perceived sense of control. A connection between the illusion of control and effort justification is also consistent with some theoretical frameworks. Energization theory asserts that a willingness to put effort into a task is a function of the necessity of the outcome as well as the likelihood of attaining the outcome. While trying to test this theory, Biner and Hua (1995) unexpectedly noticed that participants’ need state affected the manipulation check on instrumentality, or chance of winning. Participants in a high need state estimated that they had a greater chance of winning than participants in a low need state. The researchers offered a theory for this discovery that incorporated the illusion of control. Biner and Hua (1995) speculated that participants in a high need state may be more focused on task completion than participants in a low need state, causing them to believe that skill may affect the outcome more than it does. This theoretical explanation was the impetus behind a study conducted by Biner et al. (1995), which found that food-deprived subjects thought a pure chance lottery involved more skill than food-satiated subjects, suggesting that a greater need state prompts a more intense illusion of control. If these theoretical explanations of need state are true, then the illusion of control might be enhanced by effort in much the same way as need state, as effort underlies many survival factors. Participants that put more effort into a task will necessarily expend either physical or cognitive energy and be drawn toward a higher need state. The present study aims to investigate if effort can be manipulated experimentally to interact with the 9 illusion of control, as opposed to relying on a pre-existing need state component. More generally, the present study seeks to determine if evolutionary theory can be supported as an explanation for the illusion of control and effort justification as opposed to complex cognitive functions such as cognitive dissonance. Past studies suggest that the illusion of control can be amplified or dampened by preceding variables (Biner et al., 1996; Yarritu et al., 2014). The present study aimed to investigate a potential interaction between effort justification and the illusion of control. Participants completed either a high-effort or a low-effort task, and were then awarded either raffle tickets with assigned numbers or raffle tickets with numbers that they picked themselves. A significant main effect was expected to be replicated for the illusion of control, such that participants who exhibited an illusion of control by being able to choose raffle tickets would perceive a higher chance of winning a raffle than participants who were not provided the option to choose their raffle tickets (Hypothesis 1). Another main effect was expected for effort justification, such that participants who completed a high effort task would report a higher probability of winning a raffle than those who completed a low effort task (Hypothesis 2). Finally, a significant interaction was expected between the illusion of control and effort justification conditions, such that the illusion of control effect was expected to be further enhanced by the high effort task (Hypothesis 3). Method Participants Adult Whitman students, aged 18-24, were recruited through convenience sampling. Initially, 159 participants were recruited via email through the psychology 110 subject pool listserv, psychology majors listserv, and student listservs. Participants 10 taking psychology 110 were offered course credits for participating in the study. Those not in the psychology 110 class were automatically put into a random raffle to win a $25 Starbucks gift card. One participant was randomly selected once data was done being collected. There were 19 incomplete responses, four participants outside of the age range, seven participants who were excluded because of spurious responses to the dependent measure (returning either 0 or 100 percent on a probability assessment on which neither of those options were possible), and one participant who was excluded as an outlier with a response beyond 2.5 standard deviations from the mean. In total, 32 participants were excluded, leaving N = 127 as the final sample size. This participant total aligns with the estimate obtained using an a priori G*Power analysis, assuming a small to medium effect size of f = 0.4 (Faul et al., 2009). Participants were randomly assigned to one of four conditions, leaving 38 in the low effort choice condition, 29 in the low effort given condition, 27 in the high effort choice condition, and 33 in the high effort given condition following participant exclusion. Responses to the demographic questionnaire indicated that 79 participants identified as female, 39 identified as male, seven identified as non-binary/third gender, and two identified as other. Additionally, 93 participants racially identified as white, 14 as Asian, nine as multiracial, seven as Latinx, two as black, and two as other. Materials The procedure was administered virtually through a Qualtrics survey. The low effort word task condition consisted of 10 easy word anagrams (easy anagrams: lalb = ball, norb = born, yaaw = away, bluc = club, leph = help, moeh = home, fels = self, nair = 11 rain, nawt = want, vome = move), while the high effort word task condition consisted of five easy word anagrams from the preceding list and five nearly impossible word anagrams (difficult anagrams: diyamg = digamy, wopneg = gowpen, hamyap = mayhap, styrt = tryst, hethy = hythe). Easy anagrams were incorporated into the high effort task to reduce confusion and frustration in participants that could lead to attrition. Five difficult and seven easy anagram word tasks from Gignac and Wong’s (2018) study, along with three easy anagram word tasks from Ventura’s (2013) study were selected to compose the 15 total anagrams utilized across the two conditions. The illusion of control was manipulated through two modes of awarding raffle tickets, consistent with methods from past illusion of control research that used lottery tickets (Langer, 1975). In the chosen numbers condition, participants were instructed to manually input four custom numbers between 1 and 99 for three different raffle tickets. The assigned number condition, however, was composed of three raffle tickets each containing four predetermined numbers between 1 and 99 that were randomized prior to study participation. The dependent variable, perceived probability of winning the raffle, was collected using a sliding scale ranging from 0 to 100 percent. Past research has utilized a perceived confidence scale ranging from 1-10 for winning a lottery to determine if the way participants perceive a chance outcome is consistent with the outcome’s actual probability (Langer, 1975). In addition, past studies that implement a monetary dependent measure, such as how much money participants would trade their tickets for, assume that placing a higher monetary value on a ticket rationally means there is a greater believed chance of winning (Langer, 1975). The present study, therefore, chose to implement the 12 0-100 percentage scale to directly measure perceived percent chance of winning as opposed to value, as well as to obtain a more precise measurement than a simple 1-10 scale. Procedure The experiment utilized a 2 x 2 fully between-groups design. After completing the informed consent form and agreeing to participate in the study, each participant was randomly assigned to either the high effort or low effort word task condition. Participants in the high effort condition were given five easy and five difficult anagrams in alternating order. Participants in the low effort condition were presented with 10 easy anagrams. Participants in both effort conditions were given 60 s per anagram and were required to spend at least 15 s on each one before submitting an answer to prevent a wide time disparity between the two conditions. If 60 s elapsed without an answer, the procedure moved automatically on to the next item. Regardless of a submitted response being correct, incorrect, or left blank, participants moved on to the next question without feedback. After completing the anagram task, participants were notified that they had concluded the word task study to conceal the true purpose of the experiment. As a manipulation check, participants were asked to indicate on a five-point Likert scale ranging from extremely difficult to extremely easy how difficult they found the anagrams to ensure that effort was effectively manipulated. They were then told that they earned raffle tickets for completing the study, and that they were specifically given between one and six tickets based on word task performance. Every participant actually received three tickets regardless of task performance. Participants from each effort 13 condition were again randomly assigned to either the chosen number condition or the assigned number condition. Participants in the chosen number condition were instructed to choose four numbers between 1 and 99 for each of the three raffle tickets. Participants in the assigned number condition were allotted four predetermined numbers between 1 and 99 for each of the three raffle tickets. Participants in both conditions were asked to examine and record their ticket numbers so that each group was actively exposed to their number sets. Additionally, the given condition was required to remain on the page for 60 s to ensure they received a similar exposure length as the chosen number condition. Next, all participants were administered the dependent measure. After receiving their tickets, they were informed that roughly 80-120 participants were expected to also earn between one and six raffle tickets based on performance. Participants were instructed to indicate their perceived chance of winning the raffle by sliding a marker on a percentage scale from 0 to 100. As a manipulation check, they were then asked to indicate on a five-point Likert scale ranging from no control whatsoever to definitely has control how much control they felt over the outcome of the raffle. This question was added to ensure that the group that chose their raffle ticket numbers felt that their behavior had more of an impact on the raffle outcome than the given condition. Finally, participants were asked to select their fatigue level on a five-question multiple choice scale ranging from way more tired than before to way less tired than before. This question was used to ensure that participants in the high effort word task condition did not feel more tired from the survey than participants in the low effort word task condition because of differing time investments, ensuring that effort is truly 14 manipulated as opposed to greater personal involvement. Finally, participants were debriefed and given the option to withdraw their data if they found it necessary. Results The dependent variable, perceived probability, was calculated in each condition by averaging scores on a sliding scale ranging from 0 to 100 of participants’ believed percent chance of winning. The two independent variables were effort, split into the high effort anagram task or the low effort anagram task, and choice, split into the choose raffle numbers condition or the given raffle numbers condition. Results were examined with a 2 x 2 ANOVA and significance was determined at the α = .05 level. Table 1 Average Perceived Probability of Winning Raffle Dependent on Choice and Effort Conditions Choose Numbers Given Numbers All choice conditions High Effort Anagrams 11.5 (14.7) 15.7 (19.5) 13.8 (17.5) Low Effort Anagrams 12.5 (14.8) 11.0 (13.6) 11.9 (14.2) All anagram conditions 12.1 (14.7) 13.5 (17.0) 12.8 (15.8) Note. Perceived probability was measured using a sliding scale ranging from 0-100. The mean and standard deviation for each condition is reported in the table as M (SD). Regardless of the effort conditions, there was no illusion of control effect, as can be seen in the column totals in Table 1. The choice conditions (M = 12.1, SD = 14.7) scored low compared to the given conditions (M = 13.5, SD = 17.0), which is the opposite of the predicted direction in Hypothesis 1. This difference, 1.4, was not statistically significant, F(1, 123) = 0.223, p = .638, partial η2 = 0.002. The control 15 manipulation check suggested that the two conditions did not differ in believed sense of control. The choice condition (M = 1.52, SD = 0.89) scored similarly to the given condition (M = 1.58, SD = 0.92). This difference, 0.06, was not statistically significant t(125) = -0.360, p = .719, Cohen’s d = -0.064, suggesting that participants’ perception of control over the outcome of the raffle was not successfully manipulated. Regardless of the choice conditions, there was no effort justification effect. Row totals in Table 1 show that the high effort condition (M = 13.8, SD = 17.5) scored high compared to the low effort condition (M = 11.9, SD = 14.2), which was consistent with the predicted direction in Hypothesis 2. This difference, 1.9, was not statistically significant, F(1, 123) = 0.412, p = .522, partial η2 = 0.003. The word task difficulty manipulation check suggested that those in the low effort anagram condition (M = 1.36, SD = 0.64) reported the task as less challenging compared to those in the high effort condition (M = 4.15, SD = 0.61), suggesting that effort was successfully manipulated. This difference, 2.79, was statistically significant t(125) = -25.1, p < .001, Cohen’s d = 4.46. The tiredness manipulation check suggests that tiredness in the low effort condition (M = 3.12, SD = 0.54) was not significantly different than tiredness in the high effort condition (M = 3.05, SD = 0.57). This result suggests that participants in the high effort condition were not more tired at the end of the survey than participants in the low effort condition because of differing time commitments, t(125) = 0.709, p = .342, Cohen’s d = 0.126. Contrary to Hypothesis 3, there was no significant interaction between the illusion of control and effort justification, F(1, 123) = 0.992, p = .321, partial η2 = 0.008. 16 Individual group means can be seen in Figure 1. This result suggests that the high effort condition did not enhance the illusion of control effect. Figure 1 Average Perceived Probability of Winning Reported by Participants in Each Condition Note. Striped bars correspond to the low effort conditions and shaded bars correspond to the high effort conditions. Error bars display +/- 1 standard error. 17 Discussion None of the three hypotheses was supported. There was no evidence of an illusion of control effect, such that participants in the choice condition did not perceive a high chance of winning the raffle compared to participants in the given condition. There was no significant effort justification effect, such that those in the high effort anagram condition did not report a significantly high perceived percentage chance of winning the raffle compared to those in the low effort condition. Lastly, there was no significant interaction between the illusion of control and effort justification. Illusion of Control Contrary to past research and to Hypothesis 1, there was no significant main effect for the illusion of control. Those in the given condition actually had a slightly higher perceived chance of winning than those in the choice condition, but the difference was not significant. This lack of significance was further supported by our manipulation check, which suggested that the illusion of control was not successfully manipulated. Participants in the choice and given conditions did not significantly differ in their believed control over the raffle outcome. One reason that control may not have been successfully manipulated could be the use of perceived probability of winning as the dependent variable instead of monetary value. The current study had participants report their expected chance of winning the raffle using a sliding scale. In Langer’s (1975) study, participants were given physical lottery tickets in person and then asked how much money they would be willing to give up those tickets for. Although the measures are very similar and using percentage as a 18 measure was based on the various measures in Langer’s study, monetary value may be necessary to effectively measure the illusion of control. Probability assessment as a measure may only be viable after a monetary value measure is already administered, as placing a monetary value on a certain item may play a unique role in inducing an illusion of control. An assumption in past research is that when participants value a lottery ticket more, it is because they think the ticket has a greater chance of winning. The present study brings this premise into question. Participants may value the lottery tickets for reasons other than believed lottery winning percentage. They may perceive greater value based on sentimental factors such as the numbers or pictures presented on it or the person who gave it to them. It is also possible that receiving a material ticket in person rather than receiving a digital and less tangible ticket could impact how much a participant values the ticket. Participants in the current study did not receive a tangible ticket because it was conducted online, whereas in Langer’s study, participants received a physical item with notable characteristics (images of football players). Additionally, Bradely (1978) and Yarritu et al. (2014), when taken in tandem, suggest that self-esteem may underlie the illusion of control (particularly in situations of greater personal involvement). This finding provides further speculation for how the hands-off nature of the current study may have lacked necessary illusion of control components. Without a physical ticket, participants may have felt less involved in the task, and therefore may not have needed to preserve self-esteem with the illusion of control. Another possible reason that a main effect was absent for the illusion of control is that the study involved very low stakes. It is possible that participants did not take the dependent variable seriously, leading to a less reliable measure. In support of this 19 possibility is that multiple participants responded with either 0 or 100 percent, which are not possible probability estimates. Even after removing such impossible responses, along with a large outlier (a participant who reported an 80% believed chance of winning), it is still difficult to tell if the remaining participants took this portion of the study seriously. Without experimenter supervision, participants may have had less of an incentive to answer this question to the best of their abilities. This limitation may be avoided once again by using monetary value as the dependent variable instead of estimated probability of winning, since money tends to be a powerful incentive to take a task seriously (Vohs et al., 2008). In addition, conducting this study in person with physical lottery tickets may also motivate greater gravitas through experimenter supervision and a tangible reward. Another limitation was sampling, as the lopsided demographics of our sample may have skewed results. Specifically, the majority of participants identified as female, who are 2.3 times less at risk of gambling exposure than men (Carneiro et al., 2019). Since the illusion of control is closely connected to gambling behavior, this disparity may have decreased the strength of the illusion of control manipulation. Additionally, past research has indicated that men have reported a very high propensity to start gambling in their 20s whereas women do not elicit a similar inclination until their 30s. This difference suggests that gambling behaviors are much more apparent in college aged men compared to college aged women (Carneiro et al., 2019). Since participants in our study were primarily female identifying college students between the ages of 18-24, these differing demographic tendencies may have been very influential in the present study. 20 Effort Justification Also contrary to past research and Hypothesis 2, there was no significant main effect for effort justification. Although this relationship was in the hypothesized direction, participants in the high effort condition did not value their tickets significantly more than those in the low effort condition. One strength in the study was the manipulation check for effort justification as it suggested that on average, participants in the high effort condition felt their task was harder than participants in the low effort condition. Additionally, the fatigue manipulation check indicated that participants in the high effort condition did not feel significantly more fatigued compared to participants in the low effort condition. If fatigue had been higher in one group than the other, it could have indicated a systematic difference between the high and low effort groups caused by a factor other than effort. Specifically, spending far more time on the high effort task (wherein time commitment is a common method of manipulating personal involvement) would likely elicit greater fatigue. This lack of a systematic difference in fatigue across the two conditions instills confidence that effort was truly manipulated and not personal involvement, a factor that could be caused by a greater average time commitment and signaled by greater fatigue. Therefore, this manipulation check suggests that the 15 s minimum response requirement in the word task prompts was successful in ensuring that participants in the high and low effort conditions did not vary significantly in time commitment. Overall, based on the manipulation checks, effort seemed to be successfully manipulated, although it did not significantly affect the dependent variable. Another potential reason for the absence of an effort justification effect could be an inaccurate effect size prediction. Although the assumed small to medium effect size of f = 0.4 obtained through G*Power was used to determine our target sample size, the true 21 effect size may have been much smaller (Faul et al., 2009). Therefore, since the effort relationship was in the hypothesized direction, it is possible that a larger sample size could render significant results. The actual obtained effect size was 0.092, much smaller than predicted. A post-hoc power analysis using G*Power estimated that 4,322 participants would have been needed to obtain a significant result. Assuming that these two speculations about the lack of significance are accurate, future researchers may be faced with a difficult dilemma. Effort justification may be more effectively manipulated in person. At the same time, effort justification may have a very low effect size that would be far more efficient to obtain with an online study. Unless an in-person study would substantially increase the effect size, it may require an infeasible number of in-person observations to obtain an effect. Therefore, it would be useful for researchers conducting similar studies in the future to explore ways in which effort justification can be manipulated more robustly so not as many participants are required. Another interesting perspective, perhaps contrary to the previous paragraph’s assertions, is that an online study may be beneficial for manipulating effort justification under the cognitive dissonance framework. Rosenfeld et al. (1984) found that effort justification only took place while in the presence of an experimenter that was low in attractiveness (rude and unpleasant), meaning that an experimenter high in attractiveness (friendly and pleasant) represented adequate justification for a participant to expend more effort. In the current study, there were fewer external factors that could stand in as an alternate explanation for greater effort because of the simple nature of an online study. This notion suggests that any increased perception of reward value could reasonably be 22 attributed to effort justification. Despite this intriguing theoretical advantage, however, the present study did not seem to benefit from manipulating effort online. Interaction Lastly, contrary to Hypothesis 3, there was no significant interaction between the illusion of control and effort justification. The illusion of control was not further enhanced by the high effort anagram task. One implication of this finding is that the illusion of control and effort justification may be independent from each other. The lack of replicated effort justification and illusion of control main effects, however, makes it difficult to draw any meaningful conclusions about the result. Therefore, any implication pulled from the interaction result is extremely limited. Despite this constraint, it is interesting to explore what theoretical implications a null interaction effect would suggest, given (counterfactually) significant main effects of the illusion of control and effort justification. Under this hypothetical, evolutionary theory would not be supported as a function that relates the two effects. The illusion of control, specifically, would not necessarily be rooted as a protective mechanism after all. Additionally, this result would provide support for the theoretical perspectives of Festinger (1957) and Rosenfeld et al. (1984), whose work on cognitive dissonance theory led them to propose that an inconsistency in beliefs and behavior is the source of effort justification. Furthermore, this finding would not support Zentall’s (2010) evolutionary perspective, which depicts effort justification as a general behavior principle without the need of complex mental processes. 23 Future Research Subsequent research focused on decision making, perception, and uncertainty should continue to analyze possible interactions between the illusion of control and effort justification. Future researchers must find a way to effectively manipulate the illusion of control and to create a more robust effort justification effect without requiring a surplus of participants. This objective could be accomplished by manipulating the dependent variable in more consequential ways than perceived probability of winning, such as offering a monetary exchange for tickets. Simultaneously, however, future research should validate the assumption that greater monetary value derives from a greater believed chance of winning. It could also be important to conduct a similar study with a wider array of demographics, such as outside of the college age range and with more variance in gender and ethnicity. Age could be used to compare different groups to determine if heightened gambling tendencies in older adults affect how strongly the illusion of control and effort justification impact decision making. Additionally, more focus on differences in gambling behaviors between ethnicities could lead to generalizable data and potentially reveal if there are prevalent differences in gambling and risk-taking behaviors between certain cultures. Conclusions Human decision making can be flawed in certain situations, leading people to make choices that are not rational or in their best interest. Cognitive and behavioral processes can cause a person’s perception of a reward or consequence to change based on irrelevant factors, such as changes in antecedent task effort or perceived control. The interaction between the illusion of control and effort justification on human behavior is 24 not fully understood. The present study was unable to fully examine these two phenomena in tandem, but provided ample insights on each of these topics independently. The lack of an illusion of control effect suggests that monetary incentives may be necessary for the efficacy of a dependent measure. At the same time, past research may have overestimated the ability for monetary value to indicate perceived chance of winning. Furthermore, effort justification research has not fully established a leading theory for its impact on outcome perception. It is important that studies with more robust manipulations are conducted to examine this concept’s causes. Real World Implications While the present study did not successfully manipulate the illusion of control, its findings offer valuable insights for future research and real-world scenarios. Specifically, the distinction between monetary value and perceived probability as measures suggests some real-world implications for gambling addiction. Typically, irrational behavior in gambling is thought of through the lens of the illusion of control, superstition, and misguided probability assessment. Casinos and lotteries, however, may be able to influence the value that a person places on a bet or ticket with semantic tools such as iconic football players and lucky numbers. These implications further dismantle the concept of rational behavior, and could indicate an additional factor in games of chance that may be separated from probability entirely. Continuing to examine effort justification and the illusion of control with updated techniques to strengthen manipulations may also provide real world insights for lotteries. For example, an interesting phenomenon is that lower socioeconomic status individuals tend to spend more money on the lottery than their higher socioeconomic status 25 counterparts (Beckart, 2012). If subsequent studies with stronger manipulations support the present study’s third hypothesis, ample insight could be provided that contributes to an explanation for this regressive trend. Lower income individuals often work harder to earn money, therefore making their money more treasured. This additional effort may enhance the effect that the illusion of control has on people with a lower socioeconomic status who purchase and apply numbers to a lottery ticket. While speculative, applications such as those articulated above have the potential to uncover the intricate behavioral and cognitive processes involved in everyday life. Such valuable applications of psychological science can only happen by further exploring fundamental cognitive processes that may lead people’s beliefs and decisions astray. 26 References Bandura, A. (1977). Self-efficacy: Toward a unifying theory of behavioral change. Psychological Review, 84(2), 191-215. https://doi.org/10.1037/0033295X.84.2.191 Beckart, J., & Lutter, M. (2012). Why the poor play the lottery: Sociological approaches to explaining class-based lottery play. Sociology, 47(6), 1152-1170. https://doi.org/10.1177/0038038512457854 Biner, P. M., Angle, S. T., Park, J. H., Mellinger, A. E., & Barber, B. C. (1995). Need state and the illusion of control. Personality and Social Psychology Bulletin, 21(9), 899–907. https://doi.org/10.1177/0146167295219004 Biner, P. M., & Hua, D. M. (1995). Determinants of the magnitude of goal valence: The interactive effects of need, instrumentality, and the difficulty of goal attainment. Basic and Applied Social Psychology, 16(1-2), 53–74. https://doi.org/10.1207/s15324834basp1601&2_4 Bradley, G. W. (1978). Self-serving biases in the attribution process: A reexamination of the fact or fiction question. Journal of Personality and Social Psychology, 36(1), 56–71. https://doi.10.1037/0022-3514.36.1.56 Carneiro, E., Tavares, H., Sanches, M., Pinsky, I., Caetano, R., Zaleski, M., & Laranjeria, R. (2019). Gender differences in gambling exposure and at-risk gambling behavior. Journal of Gambling Studies, 36(2), 445–457. https://doi.org/10.1007/s10899-019-09884-7 27 Clark, L., & Wohl, M. J. A. (2022). Langer's illusion of control and the cognitive model of disordered gambling. Addiction, 117(4), 1146-1151. https://doi.org/10.1111/add.15649 Davis, D., Sundahl, I., & Lesbo, M. (2000). Illusory personal control as a determinant of bet size and type in casino craps games. Journal of Applied Social Psychology, 30(6), 1224-1242. https://doi.org/10.1111/j.1559-1816.2000.tb02518.x Faul, F., Erdfelder, E., Buchner, A., & Lang, A. G. (2009). Statistical power analyses using G*Power 3.1: Tests for correlation and regression analyses. Behavior Research Methods, 41, 1149-1160. Festinger, L. (1957). A theory of cognitive dissonance. Stanford, CA: Stanford University Press. Gignac, G. E., & Wong, K. K. (2018). A psychometric examination of the anagram persistence task: More than two unsolvable anagrams may not be better. Assessment, 27(6), 1198–1212. https://doi.org/10.1177/1073191118789260 Glass, D. C., & Singer, J. E. (1972). Behavioral aftereffects of unpredictable and uncontrollable aversive events. American Scientist, 60(4), 457–465. Kahneman, D. (2011). Thinking, fast and slow. New York City, NY: Fararr, Straus and Giroux. Kim, J., & Kim, J. E. (2008). “Effort-as-information”: The impact of decision-related effort on subsequent evaluation and price judgment. Association For Consumer Research, 35(1), 934-935. https://www.acrwebsite.org/volumes/13313/volumes/v35/NA-35 28 Kim, J. H., & Song, H. (2020). Effort justification for fun activities? The effect of location-based mobile coupons using games. Journal of Retailing and Consumer Services, 54. https://doi.org/10.1016/j.jretconser.2019.102029 Ladouceur, R., Mayrand, M., & Tourigny, Y. (1987). Risk-taking behavior in gamblers and non-gamblers during prolonged exposure. Journal of Gambling Studies, 3, 115-122. https://doi.org/10.1007/BF01043450 Langer, E. J. (1975). The illusion of control. Journal of Personality and Social Psychology, 32(2), 311–328. https://doi.org/10.1037/0022-3514.32.2.311 Lydall, E. S., Gilmour, G., & Dwyer, D. M. (2010). Rats place greater value on rewards produced by high effort: An animal analogue of the “effort justification” effect. Journal of Experimental Social Psychology, 46(6), 1134-1137. https://doi.org/10.1016/j.jesp.2010.05.011 Maier, S. F., & Seligman, M. E. (1976). Learned helplessness: Theory and evidence. Journal of Experimental Psychology: General, 105(1), 3–46. https://doi.org/10.1037/0096-3445.105.1.3 Martinez, F., Le Floch, V., & Gaffié, B. (2011). Reports of wins and risk taking: An investigation of the mediating effect of the illusion of control. Journal of Gambling Studies, 27, 271–285. https://doi.org/10.1007/s10899-010-9204-2 Perlmuter, L. C., & Monty, R. A. (1977). The importance of perceived control: Fact or fantasy? Experiments with both humans and animals indicate that the mere illusion of control significantly improves performance in a variety of situations. American Scientist, 65(6), 759-765. https://www.jstor.org/stable/27848177 29 Rosenfeld, P., Giacalone, R. A., & Tedeschi, J. T. (1984). Cognitive dissonance and impressions management explanations for effort justification. Personality & Social Psychology Bulletin, 10(3), 394-401. https://doi.org/10.1177/0146167284103007 Tversky, A., & Kahneman, D. (1973). Availability: A heuristic for judging frequency and probability. Cognitive Psychology, 5(2), 207-232. https://doi.org/10.1016/00100285(73)90033-9 Ventura, M., Shute, V., & Zhao, W. (2013). The relationship between video game use and a performance-based measure of persistence. Computers & Education, 60(1), 52-58. https://doi.org/10.1016/j.compedu.2012.07.003 Vohs, K. D., Mead, N. L., & Goode, M. R. (2008) Merely activating the concept of money changes personal and interpersonal behavior. Current Directions in Psychological Science, 17(3), 208-212. https://doi.org/10.1111/j.14678721.2008.00576.x Wortman, C. B., & Brehm, J. W. (1975). Responses to uncontrollable outcomes: An integration of reactance theory and the learned helplessness model. Advances in Experimental Social Psychology, 8, 277-336. http://doi.org/10.1016/s00652601(08)60253-1 Yarritu, I., Matute, H., & Vadillo, M. A. (2014). Illusion of control: The role of personal involvement. Experimental Psychology, 61(1), 38–47. https://doi.org/10.1027/1618-3169/a000225 30 Zentall, T. R. (2010). Justification of effort by humans and pigeons: Cognitive dissonance or contrast? Current Directions in Psychological Science, 19(5), 296-300. https://doi.org/10.1177/096372141038338 31