The independent Scott Alexander archive
Essays, arguments, and unexpected connections. An independent archive of Slate Star Codex, Astral Codex Ten, and LessWrong.
Scott presents a framework of 'liking,' 'wanting,' and 'approving' to explain behavior, then shows how people use low-effort approving-influenced actions (like making public commitments) to influence harder behaviors through strategic self-manipulation.
Scott explains how the brain presents socially acceptable desires as genuine values (ego syntonic) and unacceptable ones as external compulsions (ego dystonic) to improve social signaling, with willpower representing the balance between these forces.
Scott reviews psychological research showing that people don't have introspective access to their own mental processes and instead guess at their reasons for behavior in the same way they would guess at others' motivations.
Scott argues against negotiating with unconscious desires, distinguishing between behavior-executors (which have context-dependent urges) and utility-maximizers (which have stable preferences), and suggests we mistakenly model ourselves as the latter when we're actually the former.
Scott explains Robert Trivers' evolutionary theory that consciousness evolved as a 'public relations agency' to help people deceive others by first deceiving themselves about their true motives.
Scott presents Skinner's definition of voluntary behavior as that which is transparent to reinforcement, then extends this framework to explain the perceived unity of consciousness and the feeling of a core 'self'.
Scott explains how thoughts and intentions are reinforced separately from actions, using B.F. Skinner's concept of "mental behavior" to show why people often intend to do things but never actually do them.
Scott explains prospect theory, Kahneman and Tversky's model of how people actually make decisions, showing how its value and probability weighting functions predict numerous cognitive biases including loss aversion, scope insensitivity, and framing effects.
Scott explains why 'wanting' and 'liking' diverge by examining reinforcement learning mechanisms like context-dependence, intermittent reinforcement, and hyperbolic discounting, concluding that this undermines the concept of stable preferences.
Scott explains time and effort discounting through examples like candy placement affecting consumption, argues that hyperbolic discounting creates preference reversals, and suggests these are better understood as reward functions rather than true preferences.
Scott explains how reinforcement learning concepts like secondary reinforcers, generalization, and social learning apply to human behavior, while acknowledging these mechanisms don't fully explain complex decisions like quitting smoking on medical advice.
Scott explains the basics of classical and operant conditioning, focusing on B.F. Skinner's work with pigeons and different reinforcement schedules, ultimately connecting variable ratio reinforcement to slot machines and human behavior.
Scott defends behaviorism as a sophisticated methodological approach that avoided using mental phenomena as fundamental explanations, connecting it to reinforcement learning and the idea that minds are behavior-executors rather than utility-maximizers.
Scott uses a thought experiment about a robot that shoots blue things to argue that we mistakenly interpret simple programmed behaviors as goal-directed utility maximization, when really the robot is just executing code, not minimizing blue.
Scott shares a draft FAQ arguing for consequentialism, targeting intelligent readers without philosophy backgrounds, and requests feedback especially from non-consequentialists.
Scott uses a thought experiment about aliens making British people sensitive to salmon photos to argue that causing offense should be viewed through a consequentialist harm-minimization framework rather than a deontological blame-assignment framework.
Scott proposes that guilt evolved as a costly signal to prove trustworthiness to potential cooperation partners, explaining why guilt involves public self-punishment and can even occur when someone hasn't done anything wrong.
Scott explores whether philosophers genuinely seek answers or prefer endless debate, noting that while philosophy generates many solutions to problems, it lacks mechanisms like scientific experimentation to settle which solutions are correct.
Scott asks whether there's an accepted method for measuring calibration separately from domain knowledge, noting that common scoring rules conflate the two.
Scott describes six techniques for translating vague probability feelings into accurate numerical estimates, explaining the strengths and weaknesses of each method.
Scott argues that charitable giving should be approached with rigorous cost-benefit analysis, comparing it to choosing survival gear for an Arctic expedition, and demonstrates how common charitable choices waste opportunities to save lives.
Scott distinguishes between an argument's internal confidence level and our external confidence in that argument, showing that extremely high probability claims should be moderated by the baseline probability that the argument itself is flawed.
Scott analyzes seven game theory concepts through his experiences playing Diplomacy, demonstrating how trust, enforcement mechanisms, revenge, and coalition-building work in practice, then proposes organizing a game among Less Wrong readers.
Scott announces a London rationality meetup on November 5th (Guy Fawkes Day) at Waterstone's in Piccadilly Circus, humorously noting he's again scheduling it on a holiday despite promising otherwise.
Start with a reader favorite, or follow an idea somewhere unexpected.