We conducted a series of meta-analytic tests on experiments in which participants read perspective-taking instructions—i.e., written instructions to imagine a distressed persons’ point of view (“imagine-self” and “imagine-other” instructions), or to inhibit such actions (“remain-objective” instructions)—and afterwards reported how much empathic concern they experienced after learning about the distressed person. If people spontaneously empathize with others, then participants who receive remain-objective instructions should report less empathic concern than do participants who do not receive instructions; if people can deliberately increase how much empathic concern they experience, then imagine-self and imagine-other instructions should increase empathic concern relative to not receiving any instructions. Random-effects models revealed that medium-sized differences between imagine and remain-objective instructions were driven by remain-objective instructions. The results were robust to most corrections for bias. Our conclusions were not qualified by the study characteristics we examined, but most theoretically relevant moderators have not yet been thoroughly studied.

Recent theorizing suggests that religious people’s moral convictions are quite strategic (albeit unconsciously so), designed to make their worlds more amenable to their favored approaches to solving life’s basic challenges. In a meta-analysis of 5 experiments and a preregistered replication, we find that religious identity places a sex premium on moral judgments, causing people to judge violations of conventional sexual morality as particularly objectionable. The sex premium is especially strong among highly religious people, and applies to both legal and illegal acts. Religion’s influence on moral reasoning emphasizes conventional sexual norms, and may reflect the strategic projects to which religion has been applied throughout history.

Well-validated screening tools have been developed to identify people at high risk for psychosis, but these are rarely used outside of specialty clinics or research settings. The development of extremely brief and simple screening tools could increase dissemination, especially in settings with low buy-in such as those with low base rates of psychosis and/or time constraints. We sought to identify such a brief measure by modeling participant responses to three psychosis screening questionnaires (Prime Screen; Prodromal Questionnaire-Brief; Youth Psychosis At Risk Questionnaire) in a sample of 139 help-seeking individuals and 335 college students (age range: 12-25). Two screening questions with especially strong information characteristics were identified: "Do you see things that others can’t or don’t see?" and "Have you ever felt that someone was playing with your mind?" (Alternative two-item screens with similarly strong properties were also identified and validated using uncertainty quantified through Bayesian modeling.) The resulting measure was validated against clinician ratings of psychosis. The screen performed with a sensitivity of 53% and specificity 98% for clinically significant hallucinations or delusions, and sensitivity of 32% and specificity 99% for identifying people in an early phase of psychosis (clinical high risk or first episode psychosis).

Meta-analysis represents the promise of cumulative science--that each successive study brings us greater understanding of a given phenomenon. As such, meta-analyses are highly influential and gaining in popularity. However, there are well-known threats to the validity of meta-analytic results, such as processes like publication bias and questionable research practices which can cause researchers to massively overestimate the evidence in support of a claim. There are many statistical methods to correct for such bias, but no single method has been found to be robust in all realistic conditions. Here, I describe a method that merges statistical simulation and deep learning to achieve an unprecedented level of robust meta-analytic estimation in the face of numerous forms of bias and other historically problematic conditions. Furthermore, the resulting estimator, called DeepMA, has the unique property that it can easily evolve: As new conditions for which robustness is needed are identified, DeepMA can be re-trained to maintain high performance. Given the weaknesses that have been identified for meta-analysis, the current consensus is that it should serve as simply another data point, rather than residing at the top of the hierarchy of evidence. The novel approach I describe, however, holds the potential to eliminate these weaknesses, possibly solidifying meta-analysis as the platinum standard in scientific debate.

No abstract