Episode Summary
Executive Summary: Rob Woodland and Spencer Greenberg discuss practical self-improvement, the value of self-experiments, line managers for accountability, the nature of human values, harms from research ethics bureaucracy, and the strengths/limits of prediction markets. The episode argues for systems that work with human behavior as it is, not as we wish it were, and for more permissive, evidence-seeking institutions.
Main Topics: Self-improvement through systems, not personality change: Rob argues people should first solve problems by buying tools, learning concrete skills, or building workaround systems before trying to change traits like conscientiousness or extroversion. Self-experimentation and evidence from personal trials: They debate when self-experiments are useful, how to distinguish placebo from real effects, and why high-variance interventions can be especially informative at the individual level. The value of line managers and weekly accountability: Rob claims students, freelancers, and PhD students would benefit enormously from having regular check-ins that force reflection, prioritization, and goal-setting. How complicated human values really are: The conversation explores whether values are fundamentally about well-being or whether people also value truth, honesty, loyalty, purity, and other non-utilitarian goods. Research ethics, IRBs, and regulatory overreach: Rob criticizes institutional review boards and regulators for slowing beneficial research, especially in COVID-era vaccine trials, and argues that many safeguards are overbroad and misaligned. Prediction markets and forecasting reliability: They discuss why prediction markets can be useful but also distorted—especially in politics—due to partisan money, thin liquidity, and market frictions.
Key Arguments: Trying to improve your personality is often a less effective first move than using external fixes: buy a better chair, set reminders, learn a tool, or build a system that compensates for your weakness. People should forecast their own future behavior realistically, using an outside view; plans that assume a better future self often create frustration and failure. Self-experiments can work when effects are immediate, large, and outside your normal range; many experiments fail, but that does not make the method pointless. High-variance interventions are attractive for self-experimentation because the upside can be huge and the downside limited if the trial is reversible. Weekly line-manager-style check-ins can improve productivity, emotional health, and task prioritization by surfacing problems and forcing goal review. Humans may not be purely utility-maximizers: many people appear to value honesty, truth, loyalty, and purity intrinsically, not just as means to happiness. Thought experiments are useful but imperfect because intuitions can be contaminated by learned associations and by the ordinary usefulness of concepts like knowledge or honesty. IRBs and ethics regulators are often biased toward preventing visible harms and ignoring invisible opportunity costs, which can slow or block life-saving research. Human challenge trials could have sped up vaccine discovery materially during COVID, and in some contexts informed, voluntary risk-taking should be allowed. Prediction markets can be informative, but political markets may be skewed by partisanship, thin participation, fees, and limits on arbitrage capital. A better forecasting strategy is to combine multiple sources—prediction markets, polling, and expert forecasters—weighted by trustworthiness rather than relying on one oracle.
Data Points: Typical podcast length: 2 to 6 hours - Rob says 80,000 Hours episodes are unusually long-form; typical length is about 2.5 hours. Longest episode mentioned: 5 hours - Rob notes the longest 80,000 Hours episode had been five hours. Sample exclusion rate in intrinsic values test: 30% and 50% - During development of Clearer Thinking’s intrinsic values test, they discarded 30% of one sample and 50% of another for failing a comprehension quiz. Back pain experiment count: 7 chairs - Rob says he tried seven different desk chairs before switching to a standing desk. Self-experiment base rate: 5 to 10 experiments per 1 keeper - Rob says most of his self-experiments fail, and he expects many trials before one is worth keeping. Perceived value of an antidepressant trial: Massive positive effect - Spencer says antidepressants were one of the best life-improving experiments he has done. Current sleep self-experiment horizon: 100 days - Spencer is tracking tiredness three times a day and correlating it with sleep data over roughly 100 days. Data collection frequency: 3 times per day - Spencer pings himself with a survey three times daily to measure fatigue. COVID challenge trial proposal size: 100 people per vaccine candidate - Rob suggests a human challenge trial could test vaccine candidates with around 100 healthy volunteers. Low fatality risk estimate for healthy young participants: 1 in 10,000 - Rob says healthy young volunteers might face an extremely low risk of death from COVID. Potential research acceleration: 3 to 4 months - Rob estimates challenge trials could have sped up vaccine identification by months. Trillion-scale value claim: Hundreds of billions, possibly trillions - Rob argues that faster vaccine discovery had enormous social value. Individual compensation thought experiment: $1,000,000 - Rob says he would personally be willing to get COVID for a million dollars, illustrating his view on voluntary risk-taking. Predicted payout example: 6% return - Rob says he could earn about a 6% return betting on Biden in a market after the outcome was already clear. Betting-market cap example: $850 - PredictIt limits each person to $850 per market, which can make markets easier to distort. Episode audience-favorite references: Episodes 11 and 39 - Rob is introduced as a guest from previous 80,000 Hours episodes that were audience favorites. Clearer Thinking episode references: Episode 2, 8, and 10 - The intro recommends specific Clearer Thinking episodes including Aesthetics and Polyamory, Life Experiments and Philosophical Thinking, and Education and Charity.
Pivotal Quotes: "I think often when people have problems in their life, very often they'll jump to trying to shift their personality... and I think focusing on that is often a mistake." — Rob Woodland: Opening argument for prioritizing external fixes and systems over personality change. "If you don't do that, if you make plans that are based around the idea that in the future you'll be better and different than you are now, you're just constantly going to be angry with yourself." — Rob Woodland: On why people should plan around their actual future behavior rather than an idealized self. "We could probably get a pretty good signal about whether these vaccines work very quickly... by testing maybe 100 people with each of these vaccine candidates." — Rob Woodland: Rob’s case for human challenge trials during COVID.
Implications: Listeners get a pragmatic framework: optimize for systems, not self-reinvention; use self-experiments selectively; demand less bureaucratic friction in research; and treat prediction markets as helpful but fallible inputs, especially in politics.