// interactive demo · ~5 minutes
This is a condensed, playable replica of the laboratory experiment in Gender and the Demand for Comparative Information (with Nisvan Erkal and Boon Han Koh; revise & resubmit at the Journal of Economic Psychology). You will go through the screens participants saw: the task instructions and practice round, the timed puzzle task, incentivized beliefs about your absolute and relative performance, a deliberately noisy score, an interactive comprehension check — and then the paper's core decision: whether, and how much, to pay to learn how specific groups of other participants performed.
Like the real participants, you will be randomly assigned to one of the two information environments — non-instrumental (the session ends after the feedback) or instrumental (you expect to do the task again). Toggle Researcher notes (top right) to see what each screen measures, or Skip to debrief at any time.
In Part 1, you will be given 10 minutes to solve up to 30 puzzles.
A high level of effort, concentration, and attention to detail are likely to improve your performance in the task.
Numerous studies have shown that performance in solving similar types of puzzles is related to outcomes such as higher salaries, better job positions, and higher life satisfaction (Gottfretson, 2003; Niesser et al., 1996; Strenze, 2007; Bergman, 2015).
On the next page, you will see an example.
Consider the images below (on the left), which are arranged in a pattern.
You will need to identify the image from the solution set (on the right) that best fits the rest of the puzzle.
In the example above, the correct answer would be "a".
In the puzzle, the fill of the square darkens across the columns while every image remains a square of the same size. Hence, options "c", "e", and "f" are wrong because they are not squares, and options "b" and "d" are wrong because their fill does not complete the pattern.
To submit your answer to each puzzle, you will need to type the letter corresponding to the option you have chosen in the box provided on the screen. Then, click the button to submit your answer.
Once you submit your answer, you will not be permitted to change it.
Your earnings in this part of the experiment will depend on your performance in the task. Specifically, you will receive 40 ECU for every puzzle you solve correctly. Wrong answers will not affect your score.
Before we begin the task, you will participate in a practice round where you will be given three puzzles to solve. There is no time limit in the practice round, and your performance in the practice round will not affect your earnings in this experiment.
Please click the "Next" button to start the practice round.
Please select an image that fits the pattern.
Puzzle 1
Puzzle 1
Your answer:
Solution: e
Explanation: Note that all the images in the puzzle are circles with the same fill, therefore the correct answer must be one of "a", "d", or "e".
Since the circles grow larger across the columns, options "a" and "d" are incorrect.
Hence, the correct answer is "e".
This is the end of the instructions and practice round. We are now ready to begin the actual task.
You will receive 40 ECU for every puzzle you solve correctly. Wrong answers will not affect your score.
Unlike the practice round, you will not be given feedback after each question on whether or not you have answered it correctly.
When you are ready, please click the "Start Task" button to proceed with the task. You will have 10 minutes to solve up to 30 puzzles.
Please note that you cannot pause and restart the task. The timer will expire in exactly 10 minutes after you click the "Start Task" button.
Please select an image that fits the pattern.
You attempted demo puzzle(s) and solved correctly.
For the rest of the demo, your Part 1 score is set to out of 30.
Please answer the following questions.
You will earn 100 ECU for each correct answer on this page.
In Part 1, how many puzzles do you think you attempted?
How many puzzles do you think you solved correctly?
You guessed that you solved puzzle(s) correctly.
How confident are you about this guess?
If your guess is wrong, do you think you have over or under-estimated the number of puzzles you solved correctly?
There were 100 participants (university students) who completed the same task you just did. If we randomly pick one participant, how many puzzles do you think this person solved correctly?
You will earn 100 ECU if your answer is correct.
How confident are you about this guess?
The computer will now generate a score based on your Part 1 performance. However, the computer sometimes makes mistakes.
The score that the computer report is as follows:
score = number of puzzles you solved correctly + a random number X
where the random number X is a number randomly drawn by the computer. X can be either -3, -2, -1, 0, 1, 2, or 3. This means that each of these 7 numbers is equally likely to be X.
Example: Suppose the number of puzzles you solved correctly is 18, and the computer randomly draws X to be -2. In this case, the score revealed to you will be 18 − 2 = 16.
As it is mentioned earlier, there were 100 participants (university students) who completed the same task in Part 1 as you just did. Where do you think your performance, as measured by the number of correctly solved puzzles, is ranked against these participants?
Please indicate the likelihood that your performance is in each of the five intervals. A higher number means that your performance is more likely to be in that interval. The total likelihood assigned to all five intervals must add up to 100%.
You may receive 200 ECU based on how accurate your guess is compared to the true outcome. The rule to determine your payment is designed such that your chance of receiving 200 ECU is highest if you state your most accurate prediction.
Total likelihood assigned: 0%
The likelihood must add up to 100%!
We will now proceed to Part 3.
Please read the instructions carefully and complete the comprehension questions.
Part 3 is the last part of the experiment. After Part 3, you will be asked to complete a short survey.
In Part 3, you will have the chance to see the performance of other participants who did the same task.
Specifically, 100 other participants have been ranked based on their performance into five intervals. You can pick one of the following five intervals, and you will be given information about the average performance of participants ranked in that interval.
You will be asked to state how much you are willing to pay for this information.
Specifically, for the interval you have chosen, please tell us the highest amount (between 0 and 100 ECU) you would be willing to pay to receive the information. This amount indicates the value of the information to you. For example, if the information is worth 50 ECU to you, then you should be prepared to pay up to 50 ECU but not more than that.
After you make your decision, the computer will randomly draw a price between 0 ECU and 100 ECU.
If you say that you are willing to pay up to 50 ECU and the computer draws a price of 60 ECU, then you will not receive the information because this price exceeds your value for the information. But if the computer draws a price of 40 ECU, then you will receive the information and pay 40 ECU for it.
To make sure you understand the instructions, please answer the following questions.
Suppose you state that you are willing to pay up to 70 ECU for the information of your choice. Which of the following statements is correct?
Your answer is incorrect.
The computer will randomly draw a price between 0 ECU and 100 ECU. Whether or not you receive the information, and how much you pay for it, will depend on the price that the computer randomly draws.
Please try again.
Your answer is correct. Please click "Next" to proceed.
Of these intervals, which interval would you most prefer to receive information about?
If you are not interested in receiving information about any of these intervals, please select "Not interested".
You have indicated that you would like to learn the average performance of participants in
.
Up to how many ECU would you be willing to pay for this information?
Suppose you are given a chance to learn about the performance of others in one more interval. Which interval would you choose?
If you are not interested in receiving information about any of these intervals, please select "Not interested".
Suppose after Part 3, there is another Part 4 where you can solve another round of puzzles, just like in Part 1.
In Part 2, you guessed that you had solved puzzle(s) correctly in Part 1. How many puzzles do you think you will be able to solve correctly in this hypothetical Part 4?
Your payment in this experiment will not be affected by your responses to these questions.
You have indicated that you would like to learn the average performance of participants in
.
On average, how many puzzles do you think participants in this interval solved?
// demo complete