Abstract: Persistence was studied by examining di!culty and individual di"erences in mindset. Participants engaged in solving #ve geometric tracing puzzles presented in an increasingly di!cult order, the #rst three solvable, the fourth unsolvable, and the #$h solvable. %e task was to trace each line of the puzzle without retracing any lines or li$ing the marker o" the page. Assessed using Dweck’s (2000) Mindset Questionnaire (#xed or growth), participants who believed intelligence is adaptable persisted longer on the unsolvable puzzle before giving up than their #xed mindset counterparts. Results also suggest that growth mindset individuals have greater persistence following failure than their #xed mindset counterparts. %is study highlights the importance of studying the combinations and interactions of many characteristics to explain persistence. Aisthesis Volume 11, 202018 !e E"ects of Di#culty and Individual Di"erences in Mindset on Persistence by Kate M. Murner and Eric E. Hessler Some individuals are better at persisting than others, especially when situations are or become more di!cult. De#ned as the ability to continue action despite challenges, persistence is used to measure broad constructs, such as individual tolerance for frustration (Glass et al., 1969), self- control (Baumeister et al., 2007), and ego-depletion1 (Baumeister et al., 1998). Few studies research what contributes to persistence, but rather use persistence to study other constructs. Individual di"erences such as working memory capacity, the ability to hold and manipulate recent information, personality traits (extraversion or introversion), and experience (novel or expert) are of interest when studying persistence. Contrary to previous self-control literature, Lurquin (2013) found that an individual cognitive ability alone (working memory capacity) does not fully explain persistence. Rather, research reveals that persistence is a"ected by the interaction of characteristics such as motivation and task di!culty (Feather, 1961) and working memory capacity and explanatory style2 (Lurquin, 2013). %e purpose of the current 1 Ego depletion: According to Baumeister et al. (1998), “%e core idea behind ego depletion is that the self ’s acts of volition draw on some limited resource, akin to strength or energy and that, therefore, one act of volition will have a detrimental impact on subsequent volition.” 2 Explanatory style refers to “the adoption of a positive or negative attribution to an event” (Feiring et al., 2002). research is to examine persistence itself, speci#cally by examining the e"ects of having a growth mindset in di!cult situations. Measures of persistence in the literature have predominantly used di!cult or unsolvable tasks (Dvorak & Simons, 2009; Baumeister et al., 1998). Glass, Singer, and Friedman (1969) posed a set of unsolvable geometric tracing puzzles to participants suggesting they trace each line without li$ing their pen or retracing any line twice. Participants were unaware these puzzles were unsolvable. Adjusting previous methods, Lurquin (2013) presented participants with several geometric tracing puzzles ordered by increasing di!culty, of which one was unsolvable. Importantly, the unsolvable puzzle was not last, but fourth, to examine participants’ responses following failure. While Lurquin studied working memory capacity and explanatory style to explain persistence, the current study used di!culty (puzzle type) as an independent variable. Self-control is an important construct to de#ne, as it has an important in&uence on persistence and its relationship to di!culty. In de#nition, it is the conscious, deliberate control of the self that in&uences future success and happiness (Baumeister et al., 1998). Self-control includes making decisions, such as avoiding situations, making conscious choices, and managing feelings. Previous research suggests that participants perform worse on second tasks of self-control a$er exerting energy on a !e E"ects of Di#culty and Individual Di"erences in Mindset on Persistence Aisthesis Volume 11, 202019 previous task (Baumeister et al., 1998). With this knowledge established, we expected previous task exertion to in&uence persistence. However, in this study, we sought to examine how the individual characteristic mindset and task di!culty interact to encourage better self-control recovery and even longer persistence. According to Dweck (2000), mindset—which is o$en described as “#xed” or “growth”—refers to the beliefs we have regarding the nature of our abilities and characteristics. Armor and Taylor (2003) revealed a relationship between mindset and self- regulation, a form of self-control. %is relationship suggests the importance of mindset as a variable due to its correlation with individual behavior and self- control. It is suggested that individuals with a growth mindset will persist longer than those with a #xed mindset. Based on persistence literature, the following hypotheses were developed: H1: Individuals with a growth mindset will persist longer than those with a #xed mindset on the unsolvable puzzle. H2: Individuals with a growth mindset will solve the #$h solvable puzzle faster following failure on the unsolvable one. Method Participants %irty-six participants (23 women, 13 men; 18-22 years old) received partial course credit to participate. Participants were primarily White (86.1%). Other racial identities represented included Asian (8.3%) and Black/African American (5.6%). All participants were treated in accordance with the ethical principles of the American Psychological Association. Persistence Task As in Lurquin (2013), participants engaged in solving #ve geometric tracing puzzles: four solvable (Puzzles 1, 2, 3, and 5) and one unsolvable (Puzzle 4). Participants were under the impression that all puzzles were solvable. %e #ve puzzles were placed individually in clear plastic sheets in a three-ring binder. As in Vohs et al. (2008), participants were encouraged to take the task seriously and were told that performance was linked to cognitive ability and future life success. Participants traced each line of a puzzle using a dry erase marker without li$ing the marker from the page or retracing any line they had already drawn. While solving Puzzles 1–5 in order, participants were allowed to attempt each puzzle an unlimited number of times or skip any puzzle. Attempts were counted by recording the number of times a participant erased and began a puzzle again. Skips were identi#ed when a participant &ipped a page in the binder prior to solving a puzzle correctly. Once a puzzle was #nished or skipped, participants could not return to it. %e main measures of persistence were time and attempts on Puzzle 4, although data were collected for each individual puzzle. Time was measured in minutes, calculated from the moment a participant’s marker touched a page to the time they &ipped to the next puzzle. A$er Puzzle 5, participants noti#ed the researcher if they #nished or gave up. Di#culty: Puzzle Type In Lurquin (2013), participants rated how di!cult each puzzle looked prior to attempting them. Results indicated that Puzzles 3, 4, and 5 were perceived as more di!cult than Puzzles 1 and 2. Puzzle 4, which was actually unsolvable, was not perceived as more di!cult than Puzzles 3 and 5. Following Lurquin’s #ndings, we told participants that the puzzles were organized by di!culty, from easiest (Puzzle 1) to most challenging (Puzzle 5). We call this di!culty variable Puzzle Type. As a repeated- measures variable used in statistical analyses, Puzzle Type included the levels Unsolvable (Puzzle 4) and Solvable (Puzzle 5). Persistence and Recovery from Failure %e term recovery from failure describes the participant’s self-control behavior following unsuccessful attempts on Puzzle 4. Our self-control behavior of interest included the time to complete Puzzle 5. Mindset Individual variations in mindset were assessed using the Dweck Mindset Instrument (Dweck, 2000). Participants rated their agreement with statements on a Likert scale (1 = strongly agree to 6 = strongly disagree). %e 16-item measure identi#ed a mindset score on each of two subscales: Intelligence Mindset and Talent Mindset, including statements that were #xed or growth mindset oriented. A !e E"ects of Di#culty and Individual Di"erences in Mindset on Persistence Aisthesis Volume 11, 202020 sample statement from a #xed intelligence mindset is “you have a certain amount of intelligence and you can’t really do much to change it.” An example representing a growth talent mindset is “you can always substantially change how much talent you have.” Some #xed items were phrased negatively so participants were not consistently agreeing or disagreeing with statements. Scores closer to six indicate more of a growth mindset while scores closer to one indicate more of a #xed mindset (Dweck, 2000). In a previous study, internal consistency was high (Cronbach’s ' = .82 to .97; Dweck et al., 1995). In preliminary analyses, mindset scores were treated as continuous variables. In analyses of variance (ANOVAs), levels of mindset were de#ned using median splits: less versus more intelligence growth mindset and less versus more talent growth mindset. Mood %e Positive and Negative A"ect Schedule (PANAS; Watson et al., 1988), a 20-item survey, was used to measure positive and negative a"ect before the puzzle solving task. Participants described their feelings of current emotional states; for example, “upset” or “active” on a Likert scale (1 = very slight or not at all to 5 = extremely). In a previous study, internal consistency was high (Cronbach’s ' = .84 to .90; Watson et al., 1988). %is measure was administered to ensure there was no correlation between mood and mindset type. Procedure In a randomized order, participants completed the basic demographics, mood, and mindset assessments. While engaged in solving the #ve geometric tracing puzzles, the participants’ hands were videotaped and coded by time spent and number of attempts made on each puzzle. Researchers recorded whether each puzzle was solved correctly or skipped. Results Mindset %e mean score on the Intelligence Mindset subscale was 4.24 (SD = 1.05). %e mean score on the Talent Mindset subscale was 4.38 (SD = 0.74). PANAS Positive a"ect scores ranged from 23.76 to 28.30 (M = 26.03, SD = 1.12), and negative a"ect scores ranged from 12.43 to 15.74 (M = 14.08, SD = 0.82). All correlations between scores on the intelligence mindset subscale, the talent mindset subscale, positive a"ect, and negative a"ect were not signi#cant. Time and Attempts %e average total time in minutes spent on all #ve puzzles was 21.13 (SD = 11.12). %e time averages calculated for Puzzles 1-5 were: 0.91 (SD = 0.92), 2.48 (SD =1.68), 2.19 (SD = 1.46), 12.26 (SD = 10.27), and 3.29 (SD = 3.32). %e average attempts on all #ve puzzles was 31.33 (SD = 21.47). %e attempt averages calculated for Puzzles 1-5 were: 1.92 (SD = 1.27), 3.22 (SD = 2.33), 2.58 (SD = 1.93), 19.39 (SD = 20.25), and 4.22 (SD = 2.96). Solved Correctly and Skips Researchers recorded whether puzzles were solved correctly or incorrectly. %e number of skips were also recorded. Neither variable was correlated with mindset. Persistence Before conducting ANOVAs, we adopted a 2 (mindset: #xed, growth) x 2 (puzzle type: solvable and unsolvable) mixed factorial design with mindset as an independent-groups variable and puzzle type as a repeated-measures variable. Cases in which participants skipped Puzzle 5 were excluded from the two-way mixed ANOVAs. Analyses were conducted examining the e"ects of mindset and puzzle type on time and attempts. %ere was a signi#cant main e"ect of puzzle type, F(1, 21) = 26.89, p < .001, (2 = .56, and a marginally signi#cant main e"ect of intelligence mindset, F(1, 21) = 3.81, p = .06, (2 = .15, on time. %ere was a signi#cant interaction between intelligence mindset and puzzle type on time, F(1, 21) = 9.87, p = .005, (2 = .32. As seen in Figure 1, individuals spent more time overall on Puzzle 4 than Puzzle 5, and participants with more of an intelligence growth mindset lasted longer on Puzzle 4 before giving up than those with less of an intelligence growth mindset. Participants with more of an intelligence growth mindset solved Puzzle 5 faster than those with more of an intelligence #xed mindset. %e main e"ect and interaction involving talent mindset on time were not signi#cant. !e E"ects of Di#culty and Individual Di"erences in Mindset on Persistence Aisthesis Volume 11, 202021 With number of attempts as the outcome measure, there was a signi#cant main e"ect of intelligence mindset F(1, 21) = 5.30, p = .03, (2 = .20, on attempts, and a signi#cant interaction between intelligence mindset and puzzle type on attempts, F(1, 21) = 6.61, p = .02, (2 = .24. As shown in Figure 2, individuals had more attempts overall on Puzzle 4 than Puzzle 5, and those with more of an intelligence growth mindset attempted Puzzle 4 more times than those with less of an intelligence growth mindset. %e main e"ect and interaction involving talent mindset on attempts were not signi#cant. Discussion Findings suggest that individuals who believe intelligence is adaptable, holding a growth mindset, persist longer on unsolvable tasks before giving up than their #xed mindset counterparts. On the talent subscale, time and attempts did not vary due to the participant’s level of growth mindset. It is suggested that results di"ered between the intelligence and talent subscales of mindset due to the described nature of the task. %e researcher described the task to participants as intellectual and related to cognitive abilities. Individuals with a more growth intelligence mindset lasted longer before giving up on Puzzle 4 and were less negatively in&uenced by a more di!cult puzzle following failure (Puzzle 5) than those with a more #xed intelligence mindset. Findings suggest individuals with more intelligence growth mindset recover better from failure, indicating higher resiliency. Findings were congruent with results from Lurquin (2013) in that di!culty and mindset alone did not predict individual di"erences in persistence. %e interaction between di!culty and mindset was signi#cant, similar to the interaction between working memory capacity and explanatory style (Lurquin, 2013). Growth mindset, speci#cally an individual’s belief that intelligence can change over time, best explained persistence in this study. Limitations of the present research include the lack of account for possible ego-depletion e"ects. %ere may have been decreased self-regulation over time, as research identi#es that energy on a previous task depletes energy for a subsequent task (Baumeister et al., 1998). Other limitations include mood changes, personal interest in the task, and knowledge of the task. Mood was assessed only before the task. Future studies should include a mood assessment during or a$er the task. Some individuals are more susceptible to enjoying puzzle-like tasks, and some participants might have known that Puzzle 4 was mathematically unsolvable. %ough the internal validity of the present study seems strong, participants were of a convenience sample of college- aged students and results may di"er with the general public. %ese #ndings contribute to a better understanding of individual di"erences in persistence, speci#cally how di!culty and mindset interact within individuals. %is study also supports and encourages additional research on the combination of individual characteristics that contribute to persistence. Future research may include additional characteristics, both cognitive and personal. Future examination of the intelligence versus talent subscales of mindset !e E"ects of Di#culty and Individual Di"erences in Mindset on Persistence Aisthesis Volume 11, 202022 and how they in&uence persistence di"erently is also warranted. %ese #ndings have relevance for applied settings, such as education and organization development, including training development, team building, orientation programs, and athletic training. Persistence research could also contribute to studies on mindfulness, self-growth, self-awareness, conversational skills, and even relationships. References Armor, D. A., & Taylor, S. E. (2003). %e e"ects of mindset of behavior: Self-regulation in deliberative and implemental frames of mind. Personality and Social Psychology Bulletin, 29(1), 86-95. Baumeister, R. F., Bratslavsky, E., Muraven, M., & Tice, D. M. (1998). Ego-depletion: Is the active self a limited resource? Journal of Personality and Social Psychology, 74, 1252-1265. Baumeister, R. F., Vohs, K. D., & Tice, D. T. (2007). %e strength model of self-control. Current Directions in Psychological Science, 16, 351-355. doi: 10.1111/j.1467-8721.2007.00534.x Dvorak, R. D. & Simons, J. S. (2009). Moderation of resource depletion in the self-control strength model: Di"ering e"ects of two modes of self- control. Personality & Social Psychology Bulletin, 35, 572-583. doi: 10.1177/0146167208330855 Dweck, C. S. (2000). Self-theories: !eir role in motivation, personality, and development. Psychology Press. Dweck, C. S., Chiu C., & Hong, Y. (1995). Implicit theories and their role in judgments and reactions: A world from two perspectives. Psychology Inquiry, 6(4), 267-285. Feather, N. T. (1961). %e relationship of persistence at a task to expectation of success and achievement related motives. Journal of Abnormal and Social Psychology, 63, 552-561. Feiring, C., Taska, L., Chen, K. (2002). Trying to understand why horrible things happen: Attribution, shame, and symptom development following sexual abuse. Child Maltreatment, 7, 26-41. doi: 10.1177/1077559502007001003 Glass, D. C., Singer, J. E., & Friedman, L. N. (1969). Psychic cost of adaptation to an environmental stressor. Journal of Personality and Social Psychology, 12, 200-210. Lurquin, J. H. (2013). Persisting through di#cult tasks: Individual di"erences in working memory capacity and explanatory style. [Master’s %esis, University of Colorado Boulder]. Psychology and Neuroscience Graduate %eses & Dissertations, 41. Watson, D., Clark, L. A., & Tellegen, A. (1988). Development and validation of brief measures of positive and negative a"ect: %e PANAS scales. Journal of Personality and Social Psychology, 54(6), 1063-1070.