[ Question ]

What are the highest impact questions in the behavioral sciences?

by Abby Hoskin1 min read7th Apr 20218 comments


EA PsychologyGlobal priorities researchInstitutional decision-makingCause Prioritization

If you were able to assign a bunch of social scientists (psychologists, sociologists, political scientists, etc.) to study the most important questions in the world, what questions would you ask?

This post is inspired by the lack of EA relevant academic literature in the behavioral sciences. Excellent exceptions to this rule include work by Lucius Caviola and Stefan Schubert (e.g., "The Many Obstacles to Effective Giving", http://journal.sjdm.org/19/190810/jdm190810.pdf ; and "The Psychology of Existential Risk: Moral Judgments about Human Extinction", https://www.nature.com/articles/s41598-019-50145-9 ).  But outside of their work, there seems to be very few people studying human behavior from an EA perspective.

For a few more examples of the kinds of important questions social scientists could be working on, check out this article on why AI Safety research needs more social scientists: https://distill.pub/2019/safety-needs-social-scientists/

Personal background: I am a Princeton Psychology PhD student working on projects ranging from neuroimaging to infohazard reduction and promotion of effective giving. I would love to hear what the EA Forum thinks academics should be studying :)

New Answer
Ask Related Question
New Comment

4 Answers

My brain dump "Potential priority areas within cognitive sciences (psychology, neuroscience, and philosophy of mind)":


This is cool, thanks for sharing! Looks like Lucius Caviola's and Stefan Schubert's research projects are already on your radar ;)

Some quick thoughts (there is certainly already research on these but they seem important, and I don't know about reliability of existing research): 

  • Scope insensitivity: e.g. why do people find it hard to care proportionally more about proportionally bigger things?
  • Probabilistic reasoning: e.g. how can decision makers be ‘taught’ to take seriously low probability, high impact events?
  • Decision-making under uncertainty: e.g. how can this be improved? Can be people be efficiently taught to become more bayesian?
  • Group decision making: e.g. do more diverse groups really make better decisions?
  • Meta: e.g. how can social science become more reliable?

I like these a lot! Thanks for sharing :)

Great question! These are just some initial instincts - I'm not sure any of these questions are overly neglected, and they are too broad to be research questions, but curious to hear what others think:

  • How can we [most effectively - implicit in all questions below] expand individuals' moral circles (including over temporal dimensions), particular during key choice points (e.g. choosing what to eat, whether/where to donate, what/who to vote for)?
  • How can we improve political decision-making (i.e. ensuring political choices are consistent with individuals' values)?
  • How can we improve judgement and decision-making under uncertainty (including deep uncertainty) more generally? And do these differ for individuals vs. small groups vs. large groups/organisations?
  • How can we ensure individuals and groups behave more consistently with their decisions/intentions?
  • How can we increase effective democratic behaviours more generally (i.e. behaviours outside of well-defined political decisions that are generally regarded to contribute towards a well-functioning democracy)?
  • How can we increase cooperation between individuals and groups, particular individuals and groups that are distant (physically, temporally, socially, etc)?

How must we get systematically organized to alleviate suffering in the world?

2 comments, sorted by Highlighting new comments since Today at 4:37 AM

I've done a bit of thinking on this topic, main post here:

I'm most excited about fundamental research in the behavioral sciences, just ideally done much better. I think the work of people like Joseph Henrick/David Graeber/Robin Hanson was useful and revealing. It seems to me like right now our general state of understanding is quite poor, so what I imagine as minor improvements in particular areas feel less impactful than just better overall understanding. 

It has been very frustrating sitting in Psychology seminars led by big prestigious professors, listening to them spout absolute nonsense completely unsupported by quantitative analysis. So I feel your pain for sure! 

Digging up one of my old tweets: Social Psych talk: no error bars, description of stats, or listing size of subject groups. p values displayed as p=0. This is accepted?