| Popis: |
A study implementing a canonical risky choice task as a choice blindness study. Participants will complete 40 risky choice trials (choosing between pair-wise presented monetary gambles - each composed of value and probability information). They may for example chose between; p=.8 £10 and p=.5 £5, where each option is to be read as a certain probability of obtaining a certain reward (or for negative outcomes - loosing the equivalent amount). After each decision, participants will indicate how confident they are in that decision. After the initial part, participants will complete a Big 5 personality factor questionnaire. The key function of this is to to take the participants minds of the risky choice task, and a secondary function to explore the role personality may play in choice blindness. After this, participants will complete the 40 risky choices again (in a different random order), under the guise of the study exploring choice consistency. This part will indicate to participants what their previous choices were and ask whether they would like to maintain them or not. Unbeknownst to participants, a subset (5) of option-pairs will involve false feedback. That is, they will be told that they had chosen the left option (for example) when in fact they had chosen the right option (and vice versa). We use two option-pairs that have dissimilar expected value (both in ratio and difference terms) and two option pairs with similar expected value. One in each pair of high/low similarity will be with positive outcomes and one with negative outcomes. The 5th pair is a dominated pair (both probability and money is higher for one option than another). This 5th pair is included to allow us to exclude participants who do not engage seriously with the task in the first choice phase (chose the dominated option), or who do not engage seriously with the second choice phase (perhaps because they thought deeply about options in the first stage and just click on whatever is indicated as their old choice precisely because why give it additional thought). We predict that the similarity in expected value manipulation, as well as participants indicated confidence level for each decision, will modulate choice blindness (the tendency of people to accept false feedback preferences as their own). Unlike much work in the area, we will use a control condition with no false feedback to check whether false feedback induces more changes of mind than baseline changes of mind. This pre-registration is similar to a previous pre-registration, except that: 1) we now also include in the control condition a dominated option which allows the ruling out of participants who do not pay attention in the second choice task in this condition as well as in the treatment (false feedback condition). |
