Better Profiles
The larger multidimensional universe
I’ve been coming up with practical ways to deal with social reality ever since I left school in the 80s. The first thing that fascinated me was cultural geography, which was my term for how different people behaved in their own particular neighorhoods. The kickoff book for that was The Clustering of America, by Michael J. Weiss. That solidified the concept in my head. He went on to found Claritas and I imagine made a hodload of ducats. He basically came up with 40 demographic profiles and matched them to every zipcode in the nation. I’d pay a lot to get that information. Hmmm.
The other day Wink and I discussed how bored we get by race. I find that most of the time people who are colorstruck have a difficult time distinguishing between society, culture and psychology. I always throw region and economic class into the mix, but a lot of times that’s not necessary. When we’re talking about reasons for behavior I always look to these five things plus the context of my Peasant Theory which breaks us down into three functional classes.
I’ve been happy for the most part with leaving a lot of psychology on the table and wrestling more with personality, and specifically things like Enneagrams, Myers-Briggs (MBTI) and the big five OCEAN. When I’m being simple minded about such matters or talking to young people I talk Hogwarts Houses. (I’m Gryffindor / Slytherin just like Harry). I used to be Ravenclaw. I love NYC too much to be any sliver of Hufflepuff, but my wife and oldest daughter are, so there’s that. I’ve also spoken a lot about the WEIRD world but guess what? Maybe that doesn’t matter much at all.
Along comes Cremiuex and drops the following:
So it has given me the opportunity to have GPT go fetch me some psychological depth. I’ll be using the below studies from now on:
To provide a summary of the psychological studies and their conclusions based on the image you uploaded, I will perform an OCR (Optical Character Recognition) of the text in the image. Here’s a brief description of the studies and their main findings based on the extracted text:
Disgust & Homophobia (Inbar et al., 2009): This study investigates the relationship between feelings of disgust and homophobic attitudes, suggesting a potential link where disgust can amplify negative attitudes toward homosexuality.
Assimilation & Contrast (Schwarz et al., 1991): This research explores how individuals' perceptions are influenced by contextual factors, leading to assimilation (conforming to surrounding opinions) or contrast (differentiating from them).
Correspondence Bias (Miyamoto & Kitayama, 2002): This study examines how people tend to attribute others' behaviors to their dispositions rather than situational factors, known as the fundamental attribution error.
Intentional Side-Effects (Knobe, 2003): This study delves into how people perceive the intentionality of side effects, with findings suggesting that negative side effects are more likely to be perceived as intentional than positive ones.
Trolley Dilemma 1 (Hauser et al., 2007): The study explores moral decision-making in the classic trolley problem, highlighting how people weigh the ethical implications of harming one to save many.
False Consensus 1 (Ross et al., 1977): This research looks into the false consensus effect, where individuals overestimate the extent to which others share their beliefs and behaviors.
Moral Typecasting (Gray & Wegner, 2009): This study suggests that people tend to categorize others as either moral agents (capable of good or bad actions) or moral patients (deserving of moral consideration), but not both.
False Consensus 2 (Ross et al., 1977): Similar to the first study, this examines how people incorrectly assume that others share their attitudes and behaviors.
Intuitive Reasoning (Norenzayan et al., 2002): This study explores how intuitive thinking influences decision-making and how different cultures may rely on intuition versus analytic reasoning.
Less is Better (Hsee, 1998): The research suggests that people sometimes prefer less over more when the smaller option is presented in isolation, highlighting inconsistencies in human preferences.
Framing (Tversky & Kahneman, 1981): This famous study demonstrates how the way choices are framed (e.g., as gains or losses) can significantly impact decision-making.
Direction & SES (Huang et al., 2014): This study investigates the relationship between social class (SES) and directional preferences, suggesting that higher SES individuals might have different spatial and directional biases.
Moral Foundations (Graham et al., 2009): This research examines the moral foundations theory, which posits that different cultures and individuals prioritize different moral values (e.g., harm, fairness).
Tempting Fate (Risen & Gilovich, 2008): The study explores the belief in tempting fate, where people feel that taking unnecessary risks increases the likelihood of negative outcomes.
Trolley Dilemma 2 (Hauser et al., 2007): Another exploration of the moral dilemmas presented in the trolley problem, focusing on different variations of the ethical scenario.
Priming Consumerism (Bauer et al., 2012): This study looks into how consumerism primes can influence people's attitudes and behaviors, possibly leading to increased materialism.
Incidental Anchors (Critcher & Gilovich, 2008): The research investigates how unrelated anchors (numbers or information) can influence people's judgments and decisions in an unrelated domain.
Social Value Orientation (Van Lange et al., 1997): This study examines individual differences in social value orientation, which describes people's preferences for distributing resources between themselves and others.
Moral Cleansing (Zhong & Liljenquist, 2006): The study explores how people engage in moral cleansing behaviors, such as physical cleansing, to alleviate feelings of guilt after unethical actions.
Position & Power (Giessner & Schubert, 2007): This research investigates the relationship between physical position (e.g., height) and perceived power, suggesting that higher positions are often associated with greater power.
Direction & Similarity (Tversky & Gati, 1978): This study examines how the perceived similarity between objects or concepts can influence directional preferences.
SMS & Well-Being (Anderson et al., 2012): The study explores the relationship between social media use (specifically SMS) and individuals' well-being, highlighting potential negative impacts.
Priming Warmth (Zaval et al., 2014): This research investigates how priming warmth (e.g., through physical warmth or warm imagery) can influence people's social judgments and behaviors.
Structure & Goal Pursuit (Kay et al., 2014): The study explores how structural factors (such as societal norms) influence individuals' goal pursuit and achievement.
Incidental Disfluency (Alter et al., 2007): This research examines how disfluency (difficulty in processing information) can lead to more deliberate and careful thinking, potentially improving decision-making.
Choosing or Rejecting (Shafir, 1993): The study investigates how the framing of choices as either accepting or rejecting can influence decision-making, with people often finding rejection easier than acceptance.
Affect & Risk (Rottenstreich & Hsee, 2001): This study examines the role of affect (emotions) in risk perception, suggesting that emotional responses can heavily influence risk-related decisions.
Actions are Choices (Savani et al., 2010): This research explores how the perception of actions as choices influences people's satisfaction and sense of control, with implications for well-being.
Each of these studies represents a significant contribution to the field of psychology, offering insights into human behavior, cognition, and decision-making processes. The image contrasts the effect sizes of these studies across WEIRD (Western, Educated, Industrialized, Rich, Democratic) and less WEIRD populations.
This weekend I attended a data conference in Long Beach, and stopped in for a moment to hear a scientist talk about patient phenotypes. It reminded me about the details with which we are now capable of handling multiple dimensions of details about people but how rarely we do so in our sociological definitions at the level of journalism and politics. I’m going to keep this in mind. I hope you will to. More later.
MD Embedding
Well, let me give you a clue as to what I’m going to do. One of the first aspects of working with vector databases for data on the way to building an LLM is to enter small chunks of data into something call an embedding. At this point, it is relatively simple to do embeddings with words as tokens, but I think they can be done with questions and answers, or tests and results. So what if I have I have a questionnaire about people in the realm of X and there are 250 questions? Do we have anything more than dating apps that even approach this level of individuation? Probably not, considering the cost of DNA analysis from Ancestry.com. So I’m going to putter around with some hybrid database stuff that gets me closer to the WWID Project that I’ll be talking about soon. I think people are more interesting than we’ve allowed them to be in our shortsighted ways of narrowcasting. I’m all about multidimensional embedding.