The Future Of Reasoning

"The Future Of Reasoning" by Vsauce [0] is a fascinating pre-AI-era breakdown of how human reasoning works. Thinking about it in terms of LLMS is really interesting. [0]: [VIDEO]

Journey to the End of Time - [LINK] Veritasium on Thermite reaction - [LINK] Problems of a Shared Office - [LINK] History of the entire world, I guess - [LINK] Colin Furze Underground Activities - [LINK] Future of Reasoning - [VIDEO] Sex in Trek - [LINK]

> my brain defaults to looking for flaws in arguments rather than reasons to support them May be due to how reasoning is a social activity humans evolved to save time and cognitive load via division of labor. 1. An individual's reasoning workload is reduced by forming a bias. The gaps left by this bias can be filled with others' reasoning. For example if Gail focuses on why a business idea is good and Bob focuses on why that same idea is bad, they reduced the total amount of reasoning required by 50%. 2. If strong reasons are always prepared for everything, effort is wasted when others are easy to convince. So people start with minimal effort, producing weak reasons (why the business idea is good or bad). Gail and Bob keep responding to each other's reasons with stronger reasons only until necessary. After reaching consensus further reasoning is not required. Source: [VIDEO]

This interesting theory might help explain: reasoning is a social activity humans evolved to save time and cognitive load via division of labor. It requires negative feedback. 1. An individual's reasoning workload is reduced by forming a bias. The gaps left by this bias can be filled with others' reasoning. For example if Gail focuses on why a business idea is good and Bob focuses on why that same idea is bad, they reduced the total amount of reasoning required by 50%. 2. If strong reasons are always prepared for everything, effort is wasted when others are easy to convince. So people start with minimal effort, producing weak reasons (why the business idea is good or bad). Gail and Bob keep responding to each other's reasons with stronger reasons only until necessary. After reaching consensus further reasoning is not required. Source: [VIDEO]