取り入れる情報によって幸福度が下がったり、上がったりする

A Effect of Unfollowing on Main Outcome Variables Standardized Regression Coefficient Feeling toward outgroup Positive Twitter feed Well-being Feeling toward ingroup B Effect of Unfollowing on Content Seen in Feed Standardized Regression Coefficient Politicians seen Political tweets seen News accounts seen False news seen Immediate Follow-Up One Month Follow-Up Six Month Follow-Up 11 Month Follow-Up Figure 2. Unfollowing hyperpartisan accounts on social media durably reduced out-party animosity, with effects persisting up to six months after the treatment was completed. It also reduced the amount of political content people reported seeing in their feed up to eleven months after the treatment was completed, or one year after the experiment began. Panel A shows intent-to-treat effects of the two unfollow treatments on the main pre-registered outcome variables at the immediate follow-up, the one-month follow-up, the six-month follow-up, and the 11-month follow-up. Panel B shows intent-to-treat effects for variables that measured content seen in one's feed at the immediate follow-up and all subsequent follow-ups. Error bars represent 95% confidence intervals. Feelings About Twitter Feed A Unfollow Positive Intriguing Educational Uplifting Inspiring Depressing Informative Angry Unsettling Polarizing Divisive Sad B Follow Informative Educational Polarizing Intriguing Sad Unsettling Inspiring Depressing Positive Divisive Angry Uplifting Well-Being Items C Unfollow Boredom Awe Fulfillment Loneliness Anxiety Satisfaction Isolation Depression Curiosity Joy D Follow Awe Joy Curiosity Fulfillment Loneliness Satisfaction Depression Boredom Isolation Anxiety Standardized Regression Coefficient -0.4 -0.2 0.0 0.2 0.4 Standardized Regression Coefficient -0.4 -0.2 0.0 0.2 0.4 Standardized Regression Coefficient -0.4 -0.2 0.0 0.2 0.4 Standardized Regression Coefficient -0.4 -0.2 0.0 0.2 0.4