Dealing with our "groupstruckness" and "boundedness" (with Katja Grace)
Spencer Greenberg and Katya Grace discuss "group struck" behavior, where social pressure overrides individual judgment, and how to overcome it. They also explore how "bounded" humans can make better decisions using heuristics and principles, and the pitfalls of quantitative optimization.
Deep Dive Analysis
19 Topic Outline
Introduction to the 'Groupstruck' Phenomenon
Real-World Examples of Groupstruck Behavior
Defining and Understanding Groupstruck
The Evolutionary Roots of Social Rejection Fear
Strategies to Break Out of Being Groupstruck
Applying Groupstruck Concepts to AI Safety
The Concept of a 'Bounded Person' in Decision-Making
Decision-Making Through Salient Binary Choices
The Power of Heuristics and Principles for Bounded People
Rethinking Long-Term Planning and Goals
Quantitative vs. Qualitative Methods and the Distaste for Efficiency
The Perils of Ruthless Metric Optimization
Managing Goals with Multiple Indicators
Why People Distrust Quantification in Relationships
The 'Mode' of Human Bonding and Friendship
Rapid Fire: The Sleeping Beauty Argument
Rapid Fire: Agentic AI and Global Narratives
Rapid Fire: The Great Filter and Anthropic Reasoning
Rapid Fire: Changing Views on Philosophy
7 Key Concepts
Groupstruck
A phenomenon where individuals in a group act against their own clear interests, often due to unclear social pressure or an inability to break from group norms, even in situations of personal risk. It's characterized by a paralysis to act differently from the group, as seen in the smoke room experiment.
Bounded Person
A realistic model of a human being who is limited in their computational ability, number of options they can consider, and working memory, unlike the 'unbounded' rational agents often assumed in classical economics. This limitation necessitates the use of heuristics and different decision-making approaches.
Salience in Decision Making
The idea that a bounded person's choices are often determined by what options are most 'salient' or visible to them at a given moment, rather than a comprehensive evaluation of all possibilities. This highlights the importance of controlling one's environment to influence future choices, as advertising does.
Principles (Pre-decisions)
Decisions made in advance of a situation arising, serving as a heuristic for a bounded person. They help navigate complex situations, reduce cognitive load, and can contribute to identity formation, leading to more consistent and predictable behavior.
Agentic AI
Artificial intelligence that is 'goal-directed,' meaning it cares about the world being a certain way and actively works towards making that goal happen, rather than merely following patterns of action or responding to stimuli without future-oriented intent. This is seen as a useful characteristic for AI.
Great Filter Argument
The hypothesis that the absence of observable extraterrestrial civilizations (Fermi paradox) implies there are extremely difficult steps (filters) in the evolution of life from a random planet to an advanced, space-faring civilization. The central question is whether these filters are in our past or future.
Anthropic Reasoning
A form of reasoning that takes one's own existence as evidence when evaluating hypotheses about the universe. In the context of the Great Filter, it suggests that if we exist, we might be more likely to be in a universe state with many 'things like us,' which could imply the filter is still ahead of us.
10 Questions Answered
Groupstruck describes strange behaviors in groups where individuals act against their own interests due to social pressure, even in dangerous situations like a fire. This phenomenon can prevent collective action on important challenges because people wait for social permission to react.
While related, groupstruck is particularly focused on situations where an individual's *own* interests or safety are at stake, whereas the bystander effect typically describes inaction when *someone else* is in need.
This intense fear may stem from our ancestral environment, where being ostracized from a tribe could mean death, leading to a deep-seated, almost paranoid, aversion to social rejection even in modern, less dangerous contexts.
Strategies include developing the personal skill to push through discomfort for harmless 'embarrassing' actions, and at a group level, providing objective evidence for concern or offering plausible deniability and alternative incentives for desired actions.
A bounded person is limited by computational ability, memory, and the number of options they can consider, meaning they cannot perform infinite expected utility maximization. Instead, they rely on heuristics and a limited set of salient choices.
Instead of trying to approximate perfect rationality, bounded individuals should focus on influencing the salient choices presented to them, developing useful heuristics, and adopting principles (pre-decisions) to guide behavior without constant re-evaluation.
People tend to associate quantification and efficiency with coldness, a focus on easily measurable but potentially less important aspects, or a 'Machiavellian' strategic mindset that overlooks deeper, qualitative values or genuine care.
Genuine connection relies on a specific 'mode' of interaction where people act from care, not explicit calculation or ulterior motives. While quantification for self-improvement might be quirky, explicit strategic calculation for personal gain can break this mode.
The Sleeping Beauty argument is a thought experiment about probabilities when an observer is uncertain of their position within a sequence of events. Katya Grace supports the 'one third' solution, meaning the probability of the coin being heads, given she's awake, is 1/3.
Anthropic reasoning, by up-weighting hypotheses where one is likely to exist, can tilt the balance towards believing the Great Filter (the hard step preventing widespread advanced civilizations) is in humanity's future, as this scenario allows for more 'observers' like us to exist at our current stage.
19 Actionable Insights
1. Overcome Group Struck Behavior
Act when you know something is wrong, even if others aren’t, as social pressure can lead to collective inaction, especially in emergencies or when facing risks like pandemics.
2. Set Explicit Social Norms
Actively define and communicate desired social rules for group interactions, especially in new or ambiguous settings, to encourage specific behaviors (e.g., leaving conversations freely).
3. Manipulate Environment for Choices
Proactively arrange your physical and digital surroundings to make desired options more salient and accessible, increasing the likelihood you’ll consider and choose them.
4. Form Automatic Habits
Build routines so that desired behaviors become automatic, reducing the need for willpower and conscious decision-making at each choice point.
5. Adopt Principles as Pre-Decisions
Establish clear principles in advance (e.g., “always tell the truth”) to guide your actions in recurring situations, saving cognitive effort and promoting consistent behavior.
6. Practice Comfort Zone Expansion
Intentionally engage in harmless, uncomfortable activities to build resilience against social discomfort and the fear of looking “weird” when acting independently.
7. Seek Objective Evidence
When facing potential problems, look for objective data or external validation to justify your concerns, making it easier to act without fear of social judgment.
8. Find Fun Task Versions
If a beneficial activity feels like a chore, seek out enjoyable alternatives that achieve the same goal, making it more likely you’ll stick with it.
9. Prioritize Good Situations
Instead of rigid long-term plans, develop a strong sense of what constitutes a “good situation” and move in that general direction, allowing for flexibility and adaptation.
10. Build Identity with Principles
Internalize your principles to the point where they become part of your identity, fostering automatic adherence and signaling your values to others.
11. Use Plausible Deniability
Create or utilize excuses (e.g., needing the restroom) to gracefully exit social situations or deviate from norms without perceived social punishment.
12. Gracefully Exit Conversations
When ready to move on from a conversation, simply state, “It’s really nice to meet you,” with a warm tone, rather than feeling obligated to provide an excuse.
13. Avoid Ruthless Metric Optimization
Be wary of intensely optimizing for a single quantitative metric, as it rarely encapsulates true value and can lead to perverse or undesirable outcomes.
14. Utilize Multiple Metrics
Instead of one perfect metric, track multiple imperfect indicators, alternating focus between them, to get a holistic view and avoid over-optimizing any single dimension.
15. Recognize Metrics as Proxies
Always remember that quantitative metrics are mere correlations, not the true things you care about, which are often complex and best described qualitatively.
16. Be Mindful of Sacred Goods
Avoid quantifying deeply meaningful personal interactions (e.g., time with children, friendships) as it can change the nature of the relationship and be perceived as distasteful.
17. Model Your Future Self
When planning behavior change, envision your future self as a separate entity and design interventions (like reminders) to guide their choices effectively.
18. Conduct Monthly Life Ratings
Regularly assess different dimensions of your life to track progress and identify areas for improvement, using a heuristic approach to personal development.
19. Suppress Ulterior Motive Thoughts
In personal relationships, avoid consciously entertaining thoughts about how the relationship benefits you strategically, as this can undermine genuine connection.
7 Key Quotes
It seems just in general that we treat social rejection as way more important than it is by any like reasonable quote, objective standard.
Spencer Greenberg
It's almost like we're paranoid about social rejection, right? It's like, it's like, it's not like we're just worried about the like, you know, the chance that people like tease us a little bit. It's like, we're worried that like we're literally going to die or something.
Spencer Greenberg
I think in some of these cases, like even if we accept that humans really don't like social rejection, it's sort of hard to understand like why they're expecting so much social rejection anyway.
Katya Grace
Everything that can be counted does not necessarily count. Everything that counts can't necessarily be counted.
Spencer Greenberg
I think in that particular scenario, I do imagine feeling more wary of the calculating person, which is sort of interesting in that the calculating person is more predictable in some sense.
Katya Grace
It feels like we'd, we'd much rather hang out with the first person, even if their behavior so far has been the same.
Spencer Greenberg
I think that my life will go better, and the world will go better, if I just have some people I really care about. And then they go ahead and really care about some people, where it somehow is doing fine on the utilitarian calculus, and also involves kind of genuinely caring about the person.
Katya Grace
3 Protocols
Comfort Zone Expansion Exercise (CIFAR)
Spencer Greenberg- Go out and pick something that makes you uncomfortable.
- Ensure it won't make other people uncomfortable (harmless or amusing).
- Try to do it.
Gather Town Social Norm for Leaving Conversations
Spencer Greenberg- If there is more than one other person in the conversation, you can leave without saying anything.
Party Conversation Group Size Management
Katya Grace- At the start of the party, give people stickers indicating if they prefer large or small group conversations.
- If someone prefers small group conversations, they are not allowed to have more than three people in a conversation; if another person joins, someone must leave.