AI ALIGNMENT FORUM
AF

554
Wikitags

Psychology

Edited by Collisteru, et al. last updated 8th Feb 2025

Psychology is the study of the mind. Typically, psychology focuses on the human mind, but animal minds are also sometimes studied.

Just like much of the research on animal psychology also applies to humans, research on human psychology may inform how we develop and use AI.

Modern psychology began with the Freudian model of psychoanalysis, which quickly supplanted the prevailing pseudosciences like phrenology. Psychoanalysis is now largely obsolete, and has been replaced by other schools. The most prominent among these include:

Behavioral Psychology, which focuses on observable learning and behavior.

Humanistic Psychology, which focuses on theories of self-actualization and self-help.

Cognitive Psychology, which uses modern research on brain anatomy and other subjects to create theories about how the brain learns, develops, and makes decisions.

From an academic standpoint, much of the content of Lesswrong can be described as trying to implement the results of Decision Theory to people's daily actions using knowledge of human psychology. Sometimes, this involves overriding our native tendencies. Other times, it involves using them to our advantage.

Subscribe
Discussion
1
Subscribe
Discussion
1
Posts tagged Psychology
8Goodbye, Shoggoth: The Stage, its Animatronics, & the Puppeteer – a New Metaphor
RogerDearnaley
2y
0
45Shard Theory: An Overview
David Udell
3y
2
29Studying The Alien Mind
Quentin FEUILLADE--MONTIXI, NicholasKees
2y
0
5How much do personal biases in risk assessment affect assessment of AI risks?
Q
Gordon Seidoh Worley
2y
Q
1
19How to Control an LLM's Behavior (why my P(DOOM) went down)
RogerDearnaley
2y
0
12Why modelling multi-objective homeostasis is essential for AI alignment (and how it helps with AI safety as well). Subtleties and Open Challenges.
Roland Pihlakas
9mo
2
11Acknowledging Human Preference Types to Support Value Learning
Nandi
7y
0
14A brief review of the reasons multi-objective RL could be important in AI Safety Research
Ben Smith, Roland Pihlakas
4y
3
7Policy Entropy, Learning, and Alignment (Or Maybe Your LLM Needs Therapy)
sdeture
4mo
5
Add Posts