Title: Aberrant Salience as a Moderator of the Link between Perceived Similarity and Romantic Interest and as a Predictor of Change Detection Ability Collaborators: Madeleine Gross, Benjamin Kaveladze, Jonathan Schooler. For more information, please contact Benjamin Kaveladze: benjamin.kaveladze@psych.ucsb.edu Background & Rationale: The purpose of Study 1 is to test our hypothesis that individuals will feel greater and more remarkable similarity between themselves and an individual whom they are interested in romantically than someone they are not interested in. We also hypothesize that the increased meaning in these similarities will be predicted by trait aberrant salience *and* bring about a state of increased aberrant salience- in which participants see greater meaning in mundane things. Previous work in our lab has found that aberrant salience is linked to creativity, thus we also hypothesize that those in the "attractive" image condition will exhibit increased creativity. Relatedly, we will evaluate creative behaviors, curiosity, and magical thinking to explore how they correlate with aberrant salience. To further examine aberrant salience's role in attraction, we will test whether an aberrant feature (an experimentally-induced "glitch") associated with a face will make that face appear more attractive. We hypothesize that a glitching image of a face will be viewed as more attractive than a non-glitching image of a face – especially by those high in aberrant salience. We are also interested in the perceptual side of aberrant salience. Thus, we will also be including a change blindness task at the end of this study to look for possible connections between performance on this task and states of aberrant salience based on work by Ronald Rensink (2002) on visual sensing. Based on the observation that people high in aberrant salience tend to notice things that others do not, we predict that high trait aberrant salience will cause an increase in performance on the change blindness task. In other words, high aberrant salience folks will notice that there is a difference between the two photos more often. The purpose of Study 2 is to test our hypothesis that a someone writing a message to a highly attractive potential partner will be perceived as less desirable by a rater than some writing a moderately attractive potential partner. We believe that an overly-eager message writer will be seen as having a low mate value in relation to the recipient. Procedure: Study 1 will recruit workers from MTurk and/or CriticalMix. They will see a consent form and then complete a measure of schizotypy to assess proneness to aberrant salience, as well as a creativity measure (the creative behavior inventory), a magical thinking scale, and a curiosity measure (all scales attached). They will then be asked to choose 6 out of 28 interests (such as cooking and soccer) that they are most interested in. Next, they will complete the "Tinder" task. In this task they will first be asked whether they are romantically interested in men, women, or both. Next, they will be shown a page with 35 faces (of the gender[s] to which they stated they would be romantically interested) in random order and asked to pick five of those individuals to whom they are attracted. They will also be asked if they even found five attractive. They will then be asked to rank each individual they previously selected by their level of romantic interest in them. Examples of these images are attached in the materials folder (headshot.jpegs). Next, the participants randomly assigned to the "attractive face" group will be shown only the face they ranked as most attractive in the 'Tinder' task, but this time the face will be presented alongside a "bio" listing 5 of the pictured individual's interests. Three of these interests will always be three of those they selected when they indicated their interests previously, and the other two will not have been from the 28 options they were presented previously. Each participant will be asked to indicate how attractive they find the pictured individual, rate from 1-10 the level of similarity they feel to the pictured individual, indicate how likely is it that they would share this many interests with a random person on a dating app, indicate how remarkable it is that they have this level of similarity, and indicate how much the level of similarity between them would influence their interest in dating that person. Those randomly assigned to the "control face" group will do the same, but instead with the face they ranked as least attractive previously. Finally, participants will be instructed to write a message to the same pictured individual, imagining that they had "matched" on an online dating website and wanted to reach out to them. They will be informed that their response will be reviewed by a rater, reminded that their message will be anonymous, and advised to avoid putting any potentially identifying information in the message. This part of the task will be counter-balanced, with half of the participants doing it immediately after the similarities evaluation, and the other half doing it near the end of the survey. Next, they will complete a "forward flow" creativity task. This task involves giving participants a seed word, namely "bear", and asking them to generate a noun that follows from the word. After they generate that word, they generate another word that follows from the one they generated, and so on until they have generated 20 words. For example, bear, fur, purse, fashion, runway etc. These nouns are then entered into the forwardflow.org site and analyzed automatically by outputting a "flow" and semantic "distance" metric. Participants will then be randomized into either the "glitch" or "no glitch" condition. In each condition, they will see the same five faces of the gender in which they indicated they were interested (if they indicate they were interested in both men and women, they will see only female faces) presented in a random order, and asked to rate the attractiveness of each face. In the glitch condition, one of the faces (the same face for all participants attracted to that gender) will rapidly flash grey a few times every second or so, so that the image appears to be glitching (example in materials tab). Afterwards, they will be asked a series of questions to assess their current relationship or previous relationship if they are not currently in one (these questions are in the materials tab). If they have never been in a relationship they will skip this section. *the following bracketed section may be excluded from the study* {Participants will then complete the change blindness task where they will be shown 40 gifs (example "giphy" in attach tab) of neutral scenes that flick on and back again. Change blindness means that most people do not notice there is a difference between two images that are flicked on and off again (with a blank screen in between). The participants will be instructed to spot the difference between the two flickering images, and are also told that, importantly, some of these images actually do not have a change. So their secondary task is also to decide if they think there is a change, even if they do not find it. They will have 10 seconds to spot each difference. After each image gif, we will ask them if they thought there was a difference, and if so, what it was. We will also ask them if they thought there was a difference but didn't find it, and how confident they are that there was or was not a difference. This task is expected to take 10-12 minutes.} They will then get several demographic questions including several on lucid dreams and psychedelics use. They will also respond whether they have used a dating app in the past, their self-rated attractiveness, and how many romantic partners lasting at least two weeks they have had in the past. Those that were in the "glitch" condition will be asked to indicate whether they noticed anything about the faces that they felt was unusual. If they did notice something, they will be asked if they were suspicious that the researchers had caused the glitch intentionally. Finally, they will get a code that they will enter on the mturk site to get their payment. The study is expected to take about 30 minutes. Study 2 will take place using MTurk workers from Amazon, and may also involve an in-lab component. In this study, participants will be shown a picture of an average looking individual (we may also vary the attractiveness of the pictured individual) alongside one of the message written by participants in the previous study. Participants will be asked to imagine that the message was intended for them. Participants will then rate the quality of the message and their hypothetical level of romantic interest in the pictured individual, taking into account the message they wrote. We may drop the change blindness task from the survey in the interest of saving time. We may also run a shortened version of the study using only the aberrant salience scale and half of the change blindness gifs (15 real and 5 fake change gifs). Data Collection Done to Date: We attempted to run the tinder study on its own, but due to an error in the study programming, much of the data were unusable. We also separately ran the change blindness study, and found links between aberrant salience and change detection. Decision Rules: A change is considered correctly detected if the changed object is identified by name. Spelling errors and item descriptions without explicit naming will be accepted. Messages from the Tinder task will not be included in study 2 only if they are blank, nonsense, or otherwise irrelevant to the task. Planned Analyses: Any exploratory analyses will be updated at the bottom of this entry. Hypotheses: Study 1: 1) Participants will feel greater and more remarkable similarity to the pictured individual when they are more attracted to that individual. 2) Participants higher in aberrant salience will feel greater and more remarkable similarity between themselves and the pictured individual. 3) Participants in the "attractive face" group will feel greater and more remarkable similarity between themselves and the pictured individual. They will also feel that this similarity would influence their interest in dating the pictured individual. 4) Participants in the "attractive face" group will display greater creativity in the "forward flow" creativity task. 5) Participants high in aberrant salience will notice/ have noticed more coincidences that made it seem like they and their partners are/were meant to be together, and will feel that these coincidences are/were more meaningful and suggest/suggested that they and there partner are/were fated to be together to a greater extent. 6) Aberrant salience will correlate positively with the extent to which partners are in love. 7) Participants currently in romantic relationships will feel less and less remarkable similarity between themselves and the pictured individual. 8) Participants high in aberrant salience will feel more strongly that the similarity between themselves and the pictured individual would influence their interest in dating that individual. 9) Those shown the glitch face will rate it more highly than those shown the corresponding non-glitch face. 10) Higher trait aberrant salience will predict greater attraction to the glitch face. 11) Individuals higher in trait aberrant salience will more accurately detect changes in the change blindness paradigm. 12) Individuals higher in trait aberrant salience will indicate that there was a change or that they thought there was a change even though they couldn't find it more often than those low in aberrant salience. 13) Individuals higher in trait aberrant salience will be more confident in their change detection decisions. 14) Trait aberrant salience will be positively related to creativity, magical ideation, and curiosity. Study 2: 1) Messages written by participants who believed they were writing a message to attract a highly attractive potential parter will be perceived as less desirable by a rater than messages from participants who believed they were writing a message to a moderately attractive partner. 2) The perceived quality of a message will influence participant interest in the pictured individual associated with it. 3) A message written by someone of high attractiveness will be seen as being of higher quality than one by someone of average attractiveness. Analysis Plan: Study 1: For the forward flow task, the nouns are entered into an excel .csv file and then uploaded to forwardflow.org. This site automates the calculation of the "forward flow" score. This score and the creative behavior inventory represent creativity (DV) for this task. In total the DVs are: 1. How similar participants feel they are to the pictured individual 2. How attractive they feel the pictured individual is 3. How likely is it that they would share this many interests with a random person on a dating app 4. How remarkable the level of similarity between P's and the individual pictured is 5. how much the level of similarity between P's and the pictured individual would influence P's interest in dating them. 6. The amount of coincidences P's notice(d) that make/made it seem like they and their partner are/were "meant to be together?" 7. How meaningful these coincidences are/were? 8. The extent to which these coincidences make/made them feel they and there partner are/were meant to be together. 9. How attractive participants feel the glitch face versus the non-glitch face is. 10. Flow score on forward flow task 11. The amount of times a specific change is detected in the change blindness task 12. Whether P's thought there was a change (on trials where they do not find a change) 13. How confident P's were that there was or was not a change. 14. The number of relationships lasting at least two weeks that participants have had in the past. For DV's 1-5, separate ANOVAs will be run with the treatment (favorite individual vs. non-favorite individual) submitted as IV and the described scores submitted as DVs. Moderated linear regressions will also be run to see if trait creativity or trait aberrant salience interacts with the effects of the treatment. For DV's 6-8, linear regressions will be run between trait aberrant salience and the described scores. Additional exploratory moderation analyses will be run to see if these links are moderated by 1) the extents to which they are in love with their partners, 2) trait creativity, 3) how long they have been in their relationship, and 4) how serious their relationship is. For DV 9, A moderated linear regression will be conducted with aberrant salience moderating the impact of condition (glitch/no glitch) on attraction to the face. For DV 10, an ANOVA will be run with the treatment (favorite individual vs. non-favorite individual) submitted as the IV and state creativity as DV. A moderation analysis will be run to see if aberrant salience interacts with the effects of the treatment. For DV's 11-13, separate ANOVAs will be run with the treatment (favorite individual vs. non-favorite individual) submitted as the IV and the described scores submitted as DVs. A moderation analysis will be run to see if creativity and aberrant salience interacts with the effects of the treatment. Correlational analyses will be run to determine whether aberrant salience is related to creativity, magical thinking, and curiosity, as our past research suggests. For DV's 11 and 12, signal detection theory will also be applied. Subjects criterion and sensitivity (d') will be calculated, by first calculating the hit and false alarm rate and using standard SDT calculations e.g. criterion is -Z(fa) and d' is criterion + Z(h). Bivariate correlation analyses will determine whether criterion and sensitivity is correlated to aberrant salience. For DV 14 correlations will be run between the DV and aberrant salience and perceptual curiosity. Study 2: The DVs are: 1. The quality of the message from study 1 as rated by the participant from study 2. 2. The perceived attractiveness of the message writer, as rated by the participant from study 2. For DV 1, an ANOVA will be run with the treatment from study 1 (favorite individual vs. non-favorite individual) submitted as the IV and the described scores submitted as DVs. For DV 2, an ANOVA will be run with the treatment from study 1 (favorite individual vs. non-favorite individual) submitted as the IV and the described scores submitted as DVs. We will run a moderation analysis for both ANOVA's to see if self-rated attraction interacts with the effects of the treatment. If we decide to vary the attractiveness of the pictured individual presented to the participant in association with the message, another moderation analysis will be run to see if the pictured individual's attractiveness interacts with the effects of the treatment.