WEBVTT 00:00:06.979 --> 00:00:08.722 Imagine you're at a football game 00:00:08.722 --> 00:00:11.206 when this obnoxious guy sits next to you. 00:00:11.206 --> 00:00:11.906 He's loud, 00:00:11.906 --> 00:00:13.240 he spills his drink on you, 00:00:13.240 --> 00:00:14.995 and he makes fun of your team. 00:00:14.995 --> 00:00:17.226 Days later, you're walking in the park 00:00:17.226 --> 00:00:19.823 when suddenly it starts to pour rain. 00:00:19.823 --> 00:00:21.130 Who should show up at your side 00:00:21.130 --> 00:00:22.417 to offer you an umbrella? 00:00:22.417 --> 00:00:25.052 The same guy from the football game. 00:00:25.052 --> 00:00:26.489 Do you change your mind about him 00:00:26.489 --> 00:00:28.213 based on this second encounter, 00:00:28.213 --> 00:00:29.709 or do you go with your first impression 00:00:29.709 --> 00:00:31.907 and write him off? 00:00:31.907 --> 00:00:34.298 Research in social psychology suggests 00:00:34.298 --> 00:00:37.112 that we're quick to form lasting impressions of others 00:00:37.112 --> 00:00:39.020 based on their behaviors. 00:00:39.020 --> 00:00:41.314 We manage to do this with little effort, 00:00:41.314 --> 00:00:43.310 inferring stable character traits 00:00:43.310 --> 00:00:44.647 from a single behavior, 00:00:44.647 --> 00:00:45.481 like a harsh word 00:00:45.481 --> 00:00:46.974 or a clumsy step. 00:00:46.974 --> 00:00:48.733 Using our impressions as guides, 00:00:48.733 --> 00:00:50.040 we can accurately predict 00:00:50.040 --> 00:00:53.187 how people are going to behave in the future. 00:00:53.187 --> 00:00:54.075 Armed with the knowledge 00:00:54.075 --> 00:00:55.462 the guy from the football game 00:00:55.462 --> 00:00:57.293 was a jerk the first time you met him, 00:00:57.293 --> 00:00:59.793 you might expect more of the same down the road. 00:00:59.793 --> 00:01:01.674 If so, you might choose to avoid him 00:01:01.674 --> 00:01:03.009 the next time you see him. 00:01:03.009 --> 00:01:05.293 That said, we can change our impressions 00:01:05.293 --> 00:01:07.337 in light of new information. 00:01:07.337 --> 00:01:09.873 Behavioral researchers have identified 00:01:09.873 --> 00:01:12.210 consistent patterns that seem to guide 00:01:12.210 --> 00:01:14.627 this process of impression updating. 00:01:14.627 --> 00:01:17.014 On one hand, learning very negative, 00:01:17.014 --> 00:01:19.714 highly immoral information about someone 00:01:19.714 --> 00:01:21.435 typically has a stronger impact 00:01:21.435 --> 00:01:25.521 than learning very positive, highly moral information. 00:01:25.521 --> 00:01:27.796 So, unfortunately for our new friend 00:01:27.796 --> 00:01:28.821 from the football game, 00:01:28.821 --> 00:01:30.243 his bad behavior at the game 00:01:30.243 --> 00:01:33.070 might outweigh his good behavior at the park. 00:01:33.070 --> 00:01:35.628 Research suggests that this bias occurs 00:01:35.628 --> 00:01:38.828 because immoral behaviors are more diagnostic, 00:01:38.828 --> 00:01:39.875 or revealing, 00:01:39.875 --> 00:01:42.043 of a person's true character. 00:01:42.043 --> 00:01:43.877 Okay, so by this logic, 00:01:43.877 --> 00:01:46.253 bad is always stronger than good 00:01:46.253 --> 00:01:47.923 when it comes to updating. 00:01:47.923 --> 00:01:49.754 Well, not necessarily. 00:01:49.754 --> 00:01:51.882 Certain types of learning don't seem to lead 00:01:51.882 --> 00:01:54.208 to this sort of negativity bias. 00:01:54.208 --> 00:01:57.431 When learning about another person's abilities and competencies, 00:01:57.431 --> 00:01:58.297 for instance, 00:01:58.297 --> 00:01:59.880 this bias flips. 00:01:59.880 --> 00:02:01.348 It's actually the positive information 00:02:01.348 --> 00:02:03.631 that gets weighted more heavily. 00:02:03.631 --> 00:02:05.128 Let's go back to that football game. 00:02:05.128 --> 00:02:06.624 If a player scores a goal, 00:02:06.624 --> 00:02:08.292 it ultimately has a stronger impact 00:02:08.292 --> 00:02:10.200 on your impression of their skills 00:02:10.200 --> 00:02:11.959 than if they miss the net. 00:02:11.959 --> 00:02:13.902 The two sides of the updating story 00:02:13.902 --> 00:02:16.209 are ultimately quite consistent. 00:02:16.209 --> 00:02:18.310 Overall, behaviors that are perceived 00:02:18.310 --> 00:02:20.542 as being less frequent are also the ones 00:02:20.542 --> 00:02:23.379 that people tend to weigh more heavily 00:02:23.379 --> 00:02:25.627 when forming and updating impressions, 00:02:25.627 --> 00:02:26.914 highly immoral actions 00:02:26.914 --> 00:02:29.293 and highly competent actions. 00:02:29.293 --> 00:02:32.004 So, what's happening at the level of the brain 00:02:32.004 --> 00:02:33.874 when we're updating our impressions? 00:02:33.874 --> 00:02:35.257 Using fMRI, 00:02:35.257 --> 00:02:37.878 or functional Magnetic Resonance Imaging, 00:02:37.878 --> 00:02:39.294 researchers have identified 00:02:39.294 --> 00:02:41.478 an extended network of brain regions 00:02:41.478 --> 00:02:43.213 that respond to new information 00:02:43.213 --> 00:02:46.104 that's inconsistent with initial impressions. 00:02:46.104 --> 00:02:48.296 These include areas typically associated 00:02:48.296 --> 00:02:49.574 with social cognition, 00:02:49.574 --> 00:02:50.652 attention, 00:02:50.652 --> 00:02:52.465 and cognitive control. 00:02:52.465 --> 00:02:54.568 Moreover, when updating impressions 00:02:54.568 --> 00:02:56.570 based on people's behaviors, 00:02:56.570 --> 00:02:59.709 activity in the ventrolateral prefrontal cortex 00:02:59.709 --> 00:03:02.061 and the superior temporal sulcus 00:03:02.061 --> 00:03:03.494 correlates with perceptions 00:03:03.494 --> 00:03:07.879 of how frequently those behaviors occur in daily life. 00:03:07.879 --> 00:03:10.290 In other words, the brain seems to be tracking 00:03:10.290 --> 00:03:13.180 low-level, statistical properties of behavior 00:03:13.180 --> 00:03:15.900 in order to make complex decisions 00:03:15.900 --> 00:03:17.423 regarding other people's character. 00:03:17.423 --> 00:03:18.461 It needs to decide 00:03:18.461 --> 00:03:20.642 is this person's behavior typical 00:03:20.642 --> 00:03:22.483 or is it out of the ordinary? 00:03:22.483 --> 00:03:23.309 In the situation 00:03:23.309 --> 00:03:26.312 with the obnoxious-football-fan-turned-good-samaritan, 00:03:26.312 --> 00:03:27.479 your brain says, 00:03:27.479 --> 00:03:28.649 "Well, in my experience, 00:03:28.649 --> 00:03:31.782 pretty much anyone would lend someone their umbrella, 00:03:31.782 --> 00:03:34.452 but the way this guy acted at the football game, 00:03:34.452 --> 00:03:36.297 that was unusual." 00:03:36.297 --> 00:03:39.247 And so, you decide to go with your first impression. 00:03:39.247 --> 00:03:40.985 There's a good moral in this data: 00:03:40.985 --> 00:03:43.412 your brain, and by extension you, 00:03:43.412 --> 00:03:44.794 might care more about 00:03:44.794 --> 00:03:46.548 the very negative, immoral things 00:03:46.548 --> 00:03:47.792 another person has done 00:03:47.792 --> 00:03:50.665 compared to the very positive, moral things, 00:03:50.665 --> 00:03:52.194 but it's a direct result 00:03:52.194 --> 00:03:55.693 of the comparative rarity of those bad behaviors. 00:03:55.693 --> 00:03:58.329 We're more used to people being basically good, 00:03:58.329 --> 00:04:00.671 like taking time to help a stranger in need. 00:04:00.671 --> 00:04:04.200 In this context, bad might be stronger than good, 00:04:04.200 --> 00:04:06.803 but only because good is more plentiful. 00:04:06.803 --> 00:04:08.828 Think about the last time you judged someone 00:04:08.828 --> 00:04:10.135 based on their behavior, 00:04:10.135 --> 00:04:11.843 especially a time when you really feel 00:04:11.843 --> 00:04:14.270 like you changed your mind about someone. 00:04:14.270 --> 00:04:15.720 Was the behavior that caused you 00:04:15.720 --> 00:04:16.899 to update your impression 00:04:16.899 --> 00:04:19.203 something you'd expect anyone to do, 00:04:19.203 --> 00:04:22.398 or was it something totally out of the ordinary?