Publications | SAGE : Self, Altruism, Groups, & Empathy

AI‑assisted teams outperform AI‑led teams but not human‑only teams in assessing research reproducibility in quantitative social science

Thu, 28 May 2026 00:00:00 +0000

Large Language Models (LLMs) such as ChatGPT are transforming how scientists conduct and validate research, offering promise as tools to improve scientific reproducibility. However, computational reproducibility and error detection remain expensive and labor-intensive. We experimentally test how collaboration between researchers and LLM assistants influences the reproduction of quantitative social science findings across different levels of AI autonomy. We randomly assigned 288 researchers to 103 teams working under three conditions: human-only, AI-assisted (using ChatGPT as a collaborative tool), or AI-led (ChatGPT operating with minimal human oversight). Teams reproduced published results from leading social science journals, detected coding errors, and proposed robustness checks. Human-only and AI-assisted teams achieved comparable reproduction rates (94% vs. 91%) and performed similarly on most outcomes, except human-only teams identified significantly more major coding errors. Both substantially outperformed AI-led teams, which achieved only a 37% reproduction rate, detected fewer errors across all categories, proposed weaker robustness checks, and required more time. This autonomous approach, however, likely represents only a lower bound of AI capabilities. Despite rapid model advances, expert human judgment currently remains indispensable for reliable empirical verification. While AI assistance did not degrade most outcomes, it provided no measurable advantages and was associated with reduced detection of major errors. However, the 37% autonomous reproduction rate indicates that AI could provide value in settings where scale or cost constraints preclude human review of papers, even though general-purpose LLMs offer no immediate advantages for human-supervised verification.

Reproducibility and robustness of economics and political science research

Wed, 01 Apr 2026 00:00:00 +0000

An Identity-Based Approach to Polarization and Public Health

Thu, 29 Jan 2026 00:00:00 +0000

ABSTRACT A growing literature identifies political polarization as a risk factor adversely shaping public health outcomes. We propose that this relationship can be understood through theories of group processes and behavioral decision‐making. To explain the effects of polarization on public health, we review and integrate classic models from these traditions. Guided by this framework, we review evidence suggesting that identity‐based motives can shape people’s attitudes toward health behaviors, social norms about the behavior, and the perceived ability to control the behavior. This integrative review helps explain divergent patterns of health behaviors across political groups but can also generalize to explain group influences on health behaviors more broadly. We argue that to fully understand the role of social contexts in shaping human health, it is critical to investigate how group identification shapes people’s actions.

Measuring the Semantic Priming Effect Across Many Languages

Tue, 16 Sep 2025 00:00:00 +0000

Crowdsourcing Multiverse Analyses to Explore the Impact of Different Data-Processing and Analysis Decisions: A Tutorial

Mon, 15 Sep 2025 00:00:00 +0000

L’importance de la science ouverte en recherche en psychologie [The importance of open science in psychology research]

Mon, 01 Sep 2025 00:00:00 +0000

La crise de réplication scientifique que le domaine de la psychologie traverse en ce moment remet en question la réputation de notre discipline et notre confiance dans une majorité de recherches passées et présentes. Le présent article couvre les points les plus importants en lien à la crise de réplication et la science ouverte, via une brève revue de littérature. Deux causes principales de la crise de réplication en psychologie ressortent : les pratiques de recherche douteuses et le manque de transparence. Heureusement, la science ouverte, qui met au cœur de sa démarche la transparence, la reproductibilité et les bonnes pratiques de recherche, permet d’adresser ces deux problématiques directement. Celle-ci recommande notamment : (a) le préenregistrement de l’étude; (b) le rapport enregistré; (c) la mise en ligne publique des données désidentifiées; (d) la mise en ligne des matériels et de la syntaxe; (e) l’utilisation de logiciels libres tels que R; (f) la prépublication; et (g) la publication en libre accès. Cet article couvre brièvement ces différentes pratiques.

modelbased: An R package to make the most out of your statistical models through marginal means, marginal effects, and model predictions

Fri, 30 May 2025 00:00:00 +0000

Is the Mind More Powerful Than the Heart? A Randomized Controlled Trial of Two Loving-Kindness Interventions

Mon, 17 Feb 2025 00:00:00 +0000

Context. Many spiritual traditions, like Buddhism, emphasize the importance of developing feelings of love and altruism toward others regardless of circumstances, but doing so can be challenging, especially in situations where individuals have little self-control resources, for example due to cognitive fatigue. These traditions offer techniques to train the mind, such as loving-kindness meditation, in order to facilitate the achievement of this altruistic goal. However, the evidence in favour of the prosocial benefits of loving-kindness meditation is scant and at times conflicting. Methods. To investigate the long-term effects of loving-kindness training on prosociality and aggression, we investigated changes in a range of prosocial variables at pre, post, and follow-up in a preregistered randomized controlled trial comparing three groups: loving-kindness meditation (N = 58), loving-kindness reflection (N = 53), and waitlist control (N = 92). Furthermore, to establish the robustness of the loving-kindness training relative to challenging circumstances such as cognitive fatigue, we introduced an ego depletion task at post-testing. Results. Planned contrasts analyses suggested effects on positive affect only for the meditation group, on self-reported compassion for both the meditation and reflection groups, and explicit social attitudes, aggressive behaviour, cognitive accessibility of a recent altruistic behaviour, willingness to help, only for the reflection group. There were no other group differences in attitudes toward aggression, dehumanization, implicit aggression, or charity donations. Moderation analyses revealed three-way interactions between group, implicit attitudes, and ego depletion, as hypothesized, but only for compassion and cognitive accessibility of a recent altruistic behaviour. Discussion. In conclusion, results suggest that the effects of reflecting on loving-kindness concepts provide robust and reliable prosocial effects, perhaps more than meditating on those concepts. Even those benefits are limited to certain behaviours, cognitions, or affects, and they do not generalize to many common behaviours such as charity donations. However, ego depletion seems to only moderate implicit attitudes when loving-kindness is embodied through meditation and not when it is confined to the intellectual mind.

A manifesto for a globally diverse, equitable, and inclusive open science

Wed, 29 Jan 2025 00:00:00 +0000

Moving positive psychology forward: Building on the psychology of passion

Sun, 08 Dec 2024 00:00:00 +0000

Transformer les habitudes du cœur et de l’esprit : des bonnes intentions à une réponse prosociale fiable [Transforming Habits of Heart and Mind: From Good Intentions to Reliable Prosocial Response]

Mon, 28 Oct 2024 00:00:00 +0000

Les pratiques contemplatives, telles que la présence attentive ou la méditation d’amour-bienveillant, sont de plus en plus populaires dans la société occidentale. De nombreux bienfaits, notamment interpersonnels et sociaux, sont attribués à ces pratiques dans des traditions millénaires, notamment bouddhistes. L’objectif de cette thèse est de clarifier les circonstances dans lesquelles des interventions de méditation d’amour-bienveillant et d’activation cognitive de concepts associés à la présence attentive génèrent les effets les plus importants sur l’agression et la prosocialité. La thèse pose donc essentiellement la question de savoir comment l’effet de ces interventions varie en fonction de la durée d’intervention, de l’individu, du contexte et de la méthode de pratique. Nous testons ces idées dans le cadre de quatre études, dont trois études en ligne (article 1) et un essai randomisé contrôlé de six semaines (article 2). Les trois premières études testent une brève exposition contextuelle (amorçage) de la présence attentive via une tâche de phrases brouillées, et, considérés conjointement, les résultats suggèrent que cette tâche n’affecte pas l’agression ou l’hostilité et qu’il n’y a donc pas de modération par trait de personnalité. La quatrième étude compare deux interventions d’entrainement à l’amour-bienveillant sur six semaines à un groupe témoin par liste d’attente. Le premier groupe méditait sur l’amour-bienveillant, tandis que le deuxième groupe réfléchissait à l’amour-bienveillant (via des extraits de livres populaires et des balados). Nos résultats confirment d’abord que la durée de ces interventions compte, puisque l’on trouve des effets via une longue durée (six semaines de pratique quotidienne, étude 4), mais pas lors d’une brève durée contextuelle (amorçage, études 1-3). Les résultats de la quatrième étude suggèrent également que les attitudes implicites prédisent différemment la prosocialité pour certaines variables en fonction du niveau de fatigue cognitive, manipulé expérimentalement, ce qui supporte l’importance du contexte dans l’influence de ces interventions sur la prosocialité. Enfin, les résultats suggèrent également que la méthode de pratique est importante, puisque réfléchir à l’amour-bienveillant semble plus prometteur que méditer sur la chose. Ultimement, cette thèse contribue de manière significative à la littérature sur la présence attentive et l’amour-bienveillant en apportant des réponses à de nombreuses questions et en ouvrant la porte à de nouvelles questions avec cette prometteuse nouvelle intervention basée sur la réflexion sur l’amour-bienveillant.————————————————————————————————————————————————-Contemplative practices, such as mindfulness or loving-kindness meditation, are increasingly popular in Western society. Numerous benefits, particularly interpersonal and social, are attributed to these practices in ancient traditions, particularly Buddhist ones. The aim of this thesis is to clarify the circumstances in which interventions of loving-kindness meditation and of the cognitive activation of concepts associated with mindfulness generate the most significant effects on aggression and prosociality. The thesis therefore essentially asks the question of how the effect of these interventions varies depending on the duration of intervention, the individual, the context, and the method of practice. We are testing these ideas in four studies, including three online studies (Article 1) and a six-week randomized controlled trial. The first three studies tested brief contextual exposure (priming) of mindfulness via a scrambled sentence task, and taken together, the results suggest that this task does not affect aggression or hostility and that there is therefore no moderation by personality trait. The fourth study compared two loving-kindness training interventions over six weeks to a waitlist control group. The first group meditated on loving-kindness, while the second group reflected on loving-kindness (via excerpts from popular books and podcasts). Our results first confirm that the duration of these interventions is important, since we find effects via long exposure (six weeks of daily practice, study 4), but not during brief contextual exposure (priming, studies 1-3). The results of the fourth study also suggest that implicit attitudes predict prosociality differently for certain variables depending on the level of cognitive fatigue, manipulated experimentally, which supports the importance of context in the influence of these interventions on prosociality. Finally, the results also suggest that the method of practice matters, since thinking about loving-kindness seems more promising than meditating on it. Ultimately, this thesis contributes significantly to the literature on mindfulness and loving-kindness by providing answers to numerous questions and opening the door to new ones with this promising new intervention based on reflection on loving-kindness.

Check your outliers! An introduction to identifying statistical outliers in R with easystats

Mon, 25 Mar 2024 00:00:00 +0000

Pratiques cliniques inspirées par la recherche en psychologie positive / Clinical practices inspired by research in positive psychology

Fri, 08 Mar 2024 00:00:00 +0000

Unmet Expectations: Social Inclusion and the Interaction Between Social Anxiety and Ambiguous or Positive Feedback

Tue, 05 Dec 2023 00:00:00 +0000

Objective. The current study examines the effects of preferential inclusion on fulfilling fundamental needs after having received ambiguous or positive social feedback and investigates how social insecurity moderates this effect. Method. 438 participants (58.7% women, mean age 39) either received positive or ambiguous social feedback, then participated either in a social participation (Cyberball/control task) or a preferential social inclusion (Überball/experimental manipulation) task, then finally reported the fulfillment of their fundamental needs. Participants also completed a measure of social insecurity and other personality measures. Results. The two main results emerging from the current study are: (a) Überball—the preferential social inclusion condition—leads to higher fulfillment of fundamental needs than Cyberball; and (b) socially anxious individuals (those high in fear of negative evaluation) significantly benefit from preferential social inclusion (Überball) when receiving positive feedback but not when receiving ambiguous feedback. Conclusion. Overall, this research shows that Überball leads to higher fulfillment of fundamental needs than a social participation task like Cyberball Inclusion. This study is a valid and valuable condition to study the protective effects of social inclusion. It also suggests that socially insecure individuals benefit most from being preferentially socially included after receiving positive social feedback than simply being included.

Implicitly activating mindfulness: Does trait self‑control moderate its effect on aggressive behaviour?

Fri, 17 Nov 2023 00:00:00 +0000

lavaanExtra: Convenience functions for package lavaan

Wed, 11 Oct 2023 00:00:00 +0000

rempsyc: Convenience functions for psychology

Tue, 25 Jul 2023 00:00:00 +0000

Phi, Fei, Fo, Fum: Effect sizes for categorical data that use the chi-squared statistic

Thu, 20 Apr 2023 00:00:00 +0000

In both theoretical and applied research, it is often of interest to assess the strength of an observed association. Existing guidelines also frequently recommend going beyond null-hypothesis significance testing and reporting effect sizes and their confidence intervals. As such, measures of effect sizes are increasingly reported, valued, and understood. Beyond their value in shaping the interpretation of the results from a given study, reporting effect sizes is critical for meta-analyses, which rely on their aggregation. We review the most common effect sizes for analyses of categorical variables that use the χ2 (chi-square) statistic and introduce a new effect size—פ (Fei, pronounced “fay”). We demonstrate the implementation of these measures and their confidence intervals via the effectsize package in the R programming language.

The Rubber Hand Illusion: Top-down attention modulates embodiment

Sat, 22 Jan 2022 00:00:00 +0000

The Rubber Hand Illusion (RHI) creates distortions of body ownership through multimodal integration of somatosensory and visual inputs. This illusion largely rests on bottom-up (automatic multisensory and perceptual integration) mechanisms. However, the relative contribution from top-down factors, such as controlled processes involving attentional regulation, remains unclear. Following previous work that highlights the putative influence of higher-order cognition in the RHI, we aimed to further examine how modulations of working memory load and task instructions—two conditions engaging top-down cognitive processes—influence the experience of the RHI, as indexed by a number of psychometric dimensions. Relying on exploratory factor analysis for assessing this phenomenology within the RHI, our results confirm the influence of higher-order, top-down mental processes. Whereas task instruction strongly modulated embodiment of the rubber hand, cognitive load altered the affective dimension of the RHI. Our findings corroborate that top-down processes shape the phenomenology of the RHI and herald new ways to improve experimental control over the RHI.

Body swapping with a Black person boosts empathy: Using virtual reality to embody another

Fri, 28 May 2021 00:00:00 +0000

Perspective-taking, whether through imagination or virtual-reality interventions, seems to improve intergroup relations; however, which intervention leads to better outcomes remains unclear. This preregistered study collected measures of empathy and race bias from 90 participants, split into one of three perspective-taking groups: embodied perspective-taking, mental perspective-taking, and a control group. We drew on virtual-reality technology alongside a Black confederate across all conditions. Only in the first group, participants got to exchange real-time viewpoints with the confederate and literally “see through the eyes of another.” In the two other conditions, participants either imagined a day in the life of the Black confederate or in their own life, respectively. Our findings show that, compared with the control group, the embodied perspective-taking group scored higher on empathy sub-components. On the contrary, both perspective-taking interventions differentially affected neither explicit nor implicit race bias. Our study suggests that embodiment of an outgroup can enhance empathy.

Self-concept clarity and the bodily self: Malleability across modalities

Sun, 11 Oct 2020 00:00:00 +0000

The self has fascinated scholars for centuries. Although theory suggests that the self-concept (cognitive self-understanding) and bodily self (pre-reflective awareness of one’s body) are related, little work has examined this notion. To this end, in Study 1, participants reported on self-concept clarity (SCC) and completed the rubber hand illusion (RHI), a paradigm in which synchronous (vs. asynchronous) stimulation between a prosthetic hand and one’s own hand leads one to “embody” the prosthetic hand. Whereas participants were equally susceptible to the RHI during synchronous stroking, low-SCC individuals were more vulnerable to the illusion during asynchronous stroking, when the effect is unwarranted. Conceptually replicating and extending this finding, in Study 2, low-SCC individuals were more susceptible to the body-swap illusion—the impression that another person’s body is one’s own. These findings suggest that a clear sense of self implies clarity and stability of both the self-concept and the bodily self.

Book review: Awareness is freedom: The adventure of psychology and spirituality

Fri, 13 Dec 2019 00:00:00 +0000

Patterns of bronchial challenge testing in Canada

Sun, 29 Apr 2018 00:00:00 +0000

On suggestibility and placebo: A follow-up study

Thu, 16 Mar 2017 00:00:00 +0000