Large language models could change the future of behavioral healthcare: a proposal for responsible development and evaluation.

Large language models (LLMs) such as Open AI's GPT-4 (which power ChatGPT) and Google's Gemini, built on artificial intelligence, hold immense potential to support, augment, or even eventually automate psychotherapy. Enthusiasm about such applications is mounting in the field as well as industry. These developments promise to address insufficient mental healthcare system capacity and scale individual access to personalized treatments. However, clinical psychology is an uncommonly high stakes application domain for AI systems, as responsible and evidence-based therapy requires nuanced expertise. This paper provides a roadmap for the ambitious yet responsible application of clinical LLMs in psychotherapy. First, a technical overview of clinical LLMs is presented. Second, the stages of integration of LLMs into psychotherapy are discussed while highlighting parallels to the development of autonomous vehicle technology. Third, potential applications of LLMs in clinical care, training, and research are discussed, highlighting areas of risk given the complex nature of psychotherapy. Fourth, recommendations for the responsible development and evaluation of clinical LLMs are provided, which include centering clinical science, involving robust interdisciplinary collaboration, and attending to issues like assessment, risk detection, transparency, and bias. Lastly, a vision is outlined for how LLMs might enable a new generation of studies of evidence-based interventions at scale, and how these studies may challenge assumptions about psychotherapy.

Correcting misperceptions of the material benefits associated with union membership increases Americans' interest in joining unions.

How accurate are Americans' perceptions of the material benefits associated with union membership, and do these perceptions influence their support for, and interest in joining, unions? We explore these questions in a preregistered, survey experiment conducted on a national sample, representative of the US population on a number of demographic benchmarks (n = 1,430). We find that Americans exhibit large and consistent underestimates of the benefits associated with unionization, as compared to evidence from the Bureau of Labor Statistics and peer-reviewed academic research. For example, 89% of Americans underestimated the life-time income premium associated with union membership, 72% underestimated the percentage of union members who receive health insurance from their employer, and 97% overestimated the average union dues rate. We next randomly assigned half of the participants to receive a brief, informational correction conveying results of academic and government research on the material benefits associated with union membership, or not. Those who received the correction reported 11.6% greater interest in joining a union, 7.8% greater support for unions, and 6.9% greater interest in helping to organize a union in their workplace, as compared to the control group. These results suggest that, overall, Americans tend to underestimate the material benefits associated with unionization, misperceptions of these benefits are causally linked to Americans' support for unionization, and correcting these misperceptions increases a range of pro-union sentiments in the American mass public.

Academics are more specific, and practitioners more sensitive, in forecasting interventions to strengthen democratic attitudes.

Concern over democratic erosion has led to a proliferation of proposed interventions to strengthen democratic attitudes in the United States. Resource constraints, however, prevent implementing all proposed interventions. One approach to identify promising interventions entails leveraging domain experts, who have knowledge regarding a given field, to forecast the effectiveness of candidate interventions. We recruit experts who develop general knowledge about a social problem (academics), experts who directly intervene on the problem (practitioners), and nonexperts from the public to forecast the effectiveness of interventions to reduce partisan animosity, support for undemocratic practices, and support for partisan violence. Comparing 14,076 forecasts submitted by 1,181 forecasters against the results of a megaexperiment (n = 32,059) that tested 75 hypothesized effects of interventions, we find that both types of experts outperformed members of the public, though experts differed in how they were accurate. While academics' predictions were more specific (i.e., they identified a larger proportion of ineffective interventions and had fewer false-positive forecasts), practitioners' predictions were more sensitive (i.e., they identified a larger proportion of effective interventions and had fewer false-negative forecasts). Consistent with this, practitioners were better at predicting best-performing interventions, while academics were superior in predicting which interventions performed worst. Our paper highlights the importance of differentiating types of experts and types of accuracy. We conclude by discussing factors that affect whether sensitive or specific forecasters are preferable, such as the relative cost of false positives and negatives and the expected rate of intervention success.

Disadvantaged group members are prouder of their group when using the language of the dominant group compared to their native language.

In ethnically and linguistically diverse societies, disadvantaged groups often face pressures to acquire and speak the advantaged group's language to achieve social inclusion and economic mobility. This work investigates how using the advantaged group's language affects disadvantaged group members' in-group pride and collective self-esteem, relative to using their native language. Across six experimental studies involving Palestinian citizens of Israel (total N = 1,348), we test two competing hypotheses: Disadvantaged group members may experience greater in-group pride when using a) their native language, due to its emotional significance (the nativity hypothesis), or b) the language of the advantaged group, due to activation of habituated compensatory responses to dominance relations (the identity enhancement hypothesis). We found that respondents reported significantly higher in-group pride when responding to a Hebrew survey when compared to performing the same activity in Arabic (Studies 1a and 1b), regardless of whether the researchers administering the survey were identified as Jewish or Arab (Studies 2a and 2b). Study 3 replicated this effect while employing the "bogus pipeline" technique, suggesting the pride expression was authentic, not merely driven by social desirability. Finally, Study 4 (pre-registered) examined additional measures of positive regard for the in-group, finding that participants described their group more positively in an attribute selection task, and reported greater collective self-esteem, when surveyed in Hebrew, rather than in Arabic. Taken together, these findings suggest that language use influences disadvantaged group members' perceptions and feelings concerning their group when those languages are associated with relative position in an intergroup hierarchy.