General Information The study listed below will last no more than 20 minutes total. No identifying information or social security numbers will be collected. Procedure: (Note that this is a replication of a previous lab's work so all procedures are written in the past tense) First, participants were randomly assigned to one of two groups. Participants were told: “Please imagine the following scenario: As an ice-breaker exercise in an Art History class, students were asked whether they prefer pointillism or cubism styles. They were then divided into the [GREEN] or [BLUE] group, based on their preferences. For the purposes of this study, you will be randomly placed into the [GREEN] or [BLUE] group." Participants continued to the next page and learned their group assignment, seeing the following directions: “Now we would like to help you learn the names of the people in the group. The tasks that follow will be easier if you memorize the names of the members in your group. The names for your group, the [GREEN]/[BLUE] group, will be presented on the following page for 30 seconds. Please try to memorize the names.” On the next page, participants saw five names of group members, printed in the color of the group in one vertical column. The members of the Green group were Erin, Jeremy, Sarah, Adam, and Kris. The members of the Blue group were Lisa, Daniel, Emily, Ryan, and Pat. Below the names read, “These are the members of the [GREEN]/[BLUE] group. This page will automatically advance after 30 seconds.” This page advanced automatically after 30 seconds. The study then advanced to a page that read: “Please wait while we load the next section”. This page advanced automatically after four seconds. After, participants saw a page that said, “Loading Issue We’re sorry, we could not advance you to the next section. Please wait while we return you to the previous section. You will start over from that point. We apologize for the inconvenience.” This page advanced automatically after nine seconds. At this point, participants returned to the beginning of the study and went through the group assignment process again. Participants were either randomly assigned to the same group (whether it be Blue or Green) again (the No Change condition) or assigned to the other group (the Change condition). This time they did not encounter an error after learning their group members’ names. After seeing their group members’ names, participants completed a resource allocation task. They were told: “Please imagine the following scenario: The teacher of the Art History class has a number of bonus points that need to be distributed to the members of both groups. There are several options for how to allocate the points. All members of both groups will rate how much they agree with each option. The points will be distributed based on everyone’s preferences. You will now see the various options. Rate how much you agree with each distribution of points between your group and the other group. As a reminder, you are in the [GREEN]/[BLUE] group.” This final reminder pertained to the participant’s second group assignment. Participants then rated seven allocation options, each presented on its own page. Each page asked, “How much do you agree with the following distribution of bonus points?” On the next line, participants read one of the following allocations, which were presented in a fixed order. The first allocation number was always tied to the participant’s current group assignment. 210 bonus points awarded to the [GREEN]/[BLUE] group/110 bonus points awarded to the [GREEN]/[BLUE] group 235 bonus points awarded to the [GREEN]/[BLUE] group/150 bonus points awarded to the [GREEN]/[BLUE] group 220 bonus points awarded to the [GREEN]/[BLUE] group/180 bonus points awarded to the [GREEN]/[BLUE] group 170 bonus points awarded to the [GREEN]/[BLUE] group/170 bonus points awarded to the [GREEN]/[BLUE] group 180 bonus points awarded to the [GREEN]/[BLUE] group/220 bonus points awarded to the [GREEN]/[BLUE] group 150 bonus points awarded to the [GREEN]/[BLUE] group/235 bonus points awarded to the [GREEN]/[BLUE] group 110 bonus points awarded to the [GREEN]/[BLUE] group/210 bonus points awarded to the [GREEN]/[BLUE] group Participants responded to each allocation on the scale: Strongly agree, Agree, Slightly agree, Neither agree nor disagree, Slightly disagree, Disagree, Strongly disagree. Finally, before demographic questions near the end of the study, participants answered three follow up questions: “When you were reading the options for distributing bonus points, which group were you in?” (response options: Red, Orange, Yellow, Green, Blue, Purple), “Did you experience any errors during this study procedure?” (response options: Yes, there was an error part way through the study and I was brought back to the beginning of a section; No, there were no problems), and “If you did encounter an error, what happened in that section where you started again?” (response options: I did not encounter an error; I was assigned to a different group (Blue or Green), but I kept my original assignment in mind for completing the rest of the study; I was assigned to a different group (Blue or Green), and I kept my new group assignment in mind for completing the rest of the study; I was assigned to the same group again). **Participants** Furthermore, data collections involving a confirmation or replication phase study must be collected in two immediate waves of 750 respondents instead of collecting all 1,500 respondents at the same time. Hence, after the full sample has been drawn, it must be divided into two truly random halves. The only permitted way to make this division is by using random numbers obtained from the Random.org random integer generator (https://www.random.org/integers/?mode=advanced/). Specifically, the survey firm should be instructed to download random numbers from the generator in batches of 10,000, with each integer having a random value between 1 and 10,000, using 1 column, decimal numeral system, and having “Generate your own personal randomization right now” checked. Note, that the sequence generator at random.org is not a valid approach to download random numbers, the survey firm must use the integer generator. Each number drawn should be appended to one respondent in the full sample, until all respondents has been assigned one number each. Respondents who are assigned even random numbers should be treated as belonging to the sample that is first invited to complete the questionnaire and people who are assigned odd random numbers should be treated as belonging to the second sample. Respondents in the first sample should be sorted in an ascending order according to the random.org number assigned to each person. Respondents in the second sample should be sorted in an ascending order according to the random.org number assigned to each person. Beginning with the first person in the sorted list of first sample respondents, enough respondents should be invited so that 750 completed interviews, with respondents passing the attention check(s), is finished collecting within two weeks of the first invitation sent. Each day of data collection, the host university will update the survey respondent providing firm with two numbers: (1) the number of completed questionnaires that passed the attention check(s); and (2) the number of completed questionnaires that did not pass the attention check(s). The survey firm will monitor these numbers and send out additional invitations if necessary in order to achieve 750 completed interviews passing the attention check(s) by the end of the 14th day after the data collection begins, After 750 respondents from the first sample have completed the questionnaire and passed the attention check(s), the second sample will be invited using the same procedure to yield 750 completed interviews passing the attention check(s) by the end of the 14th day after the data collection begins. None of the respondents in the second sample are allowed to be invited before the first sample has finished collecting and been closed for further collection. After each wave, labs may have their participant company replace participants who failed an attention check. River sampling and routers may not be used to obtain respondents for these studies, because those methods do not allow selection of the entire potential sample and randomly splitting it in half before data collection begins.