Assessment of bias in facial expressions in Social Anxiety: a systematic review Thiago Dantas, Iasmin Viana-Menezes, Bianca Sampaio Silva Ivo, Pablo Jordão Alcântara Cruz, Julian Tejada Psico Usf, 2026 Resumo Este artigo revisa sistematicamente a literatura sobre intervenções para Transtorno de Ansiedade Social (TAS) que objetivam a modificação do viés de atenção (CBM-A e CMB-I) e seu impacto na percepção de expressões faciais emocionais. CBM-A tem por objetivo desviar a atenção de estímulos ameaçadores, enquanto o CBM-I objetiva alterar a interpretação desses estímulos. O estudo foi cadastrado na plataforma PROSPERO sob o código CRD42024591700 e seguiu as recomendações PRISMA de: identificação, triagem e elegibilidade. Treze artigos elegíveis foram analisados baseados nos descritores, as bases de dados e os critérios de inclusão/exclusão definidos. Os resultados indicam que o CBM-A melhora o viés de atenção e os sintomas de TAS, embora sua eficácia varie devido a limitações metodológicas, como amostras e medidas subjetivas. Evidencia-se a necessidade de pesquisas robustas sobre CBM-I e a integração de ferramentas objetivas, marcadores neurofisiológicos, para aprimorar intervenções clínicas baseadas em evidências.
VENturing into machine learning for the morphological analysis of von Economo neurons Ivan Banovac, Oliver J. Bruton, Luis Mercado-Díaz, Julian Tejada, Fernando Marmolejo-Ramos Scientific Reports, 2025 Von Economo neurons (VENs) are a specialized type of large, highly elongated projection neurons located in specific cortical regions. Despite their implication in higher-order cognitive functions and psychiatric disorders in humans, consistent and objective identification criteria for VENs remain lacking. We analyzed 761 digitally reconstructed neurons from the NeuroMorpho.Org database. We applied six supervised machine learning algorithms and a convolutional neural network with Grad-CAM visualization to classify the reconstructions into VENs and pyramidal neurons. Variable importance was evaluated using information-driven and expert-based selection. We compared the classifications made by machine learning algorithms to the reconstructions’ original labels. Reconstructions misclassified by the classifier models were further examined by a neuroanatomy expert. Machine learning models generally achieved high classification accuracy. Morphometric features such as dendritic length and number of stems emerged as some of the key discriminators. Expert ratings only partially aligned with machine findings, and there was low agreement between experts. Most misclassifications made by the classifier models were attributable to reconstruction artifacts or ambiguous morphology rather than model limitations. Our findings demonstrate the utility of combining machine learning with expert insight for distinguishing VENs from pyramidal neurons. While soma shape remains important for the characterization of VENs, classifier models revealed that dendritic architecture may be equally as specific and could help distinguish between borderline cases. This framework offers a replicable, data-driven method for studying VENs and can be utilized for future research on their distribution and function.
Effects of physically active lessons and active breaks on cognitive performance and health indicators in elementary school children: a cluster randomized trial João Carlos N. Melo, Julian Tejada, Ellen Caroline M. Silva, José Ywgne, David N. Oliveira, et al. International Journal of Behavioral Nutrition and Physical Activity, 2025 Background This cluster‐randomized trial examined the effects of active breaks (AB) and physically active lessons (PAL) on cognitive function and health indicators in elementary school children. Methods Six schools were randomly assigned to three groups: AB group (n = 61), PAL group (n = 77), and a control group (CTL, n = 46). First-year elementary school students participated (6.9 ± 0.6 years; 52.7% girls), and the interventions lasted eight weeks. Cognitive function was measured via reaction time and correct responses on computerized tests (Go/NoGo, DigitSpan, Mental Rotation, Visual Search, and Cueing Posner). Secondary outcomes included physical activity, quality of life, daytime sleepiness, and school perception. Results Significant group-by-time interactions were found in four tests: Go/NoGo (reaction time: p = 0.045), DigitSpan (correct responses: p = 0.020), Mental Rotation (reaction time: p = 0.049), and Cueing Posner (reaction time: p = 0.017). Only the PAL group presented a reduction in reaction time in inhibitory control (Go/NoGo) (change from baseline [Δ] = -106.4 ms; p < 0.001; d = 0.50), with a greater reduction than the AB group (difference-in-differences [DiD] = -107.3 ms; p = 0.019; d = 0.47). Short-term memory (Digit Span) improved only in the PAL group (Δ = + 0.6; p < 0.001; d = 0.44), with larger gains than the CTL group (DiD = + 0.7; p = 0.024; d = 0.54) and AB group (DiD = + 0.7; p = 0.010; d = 0.49). Spatial reasoning (Mental Rotation) improved in both the PAL (Δ = -1967.5 ms; p < 0.001; d = 0.72) and AB groups (Δ = -1477.8 ms; p < 0.001; d = 0.54), but only the PAL group showed a greater change than the CTL group (DiD = -1394.0 ms; p = 0.012; d = 0.54). Spatial orientation (Posner Cueing) improved in all groups (PAL group: Δ = -386.6 ms; p < 0.001; d = 0.68; CTL group: Δ = -183.8 ms; p = 0.024; d = 0.29; AB group: Δ = -158.4 ms; p = 0.007; d = 0.36), with the PAL group presenting greater reductions than the CTL (DiD = -202.8 ms; p = 0.045; d = 0.33) and AB groups (DiD = -228.2 ms; p = 0.007; d = 0.45). Conclusions Physically active lessons enhanced various cognitive functions, while active breaks, although less impactful, also represent a beneficial strategy. Trial registration Brazilian Clinical Trials Registry (REBEC trial: RBR-10zxwdrh, retrospectively registered on 2025-01-09, https://ensaiosclinicos.gov.br/rg/RBR-10zxwdrh).
Big team science reveals promises and limitations of machine learning efforts to model physiological markers of affective experience Nicholas A. Coles, Bartosz Perz, Maciej Behnke, Johannes C. Eichstaedt, Soo Hyung Kim, et al. Royal Society Open Science, 2025 Researchers are increasingly using machine learning to study physiological markers of emotion. We evaluated the promises and limitations of this approach via a big team science competition. Twelve teams competed to predict self-reported affective experiences using a multi-modal set of peripheral nervous system measures. Models were trained and tested in multiple ways: with data divided by participants, targeted emotion, inductions, and time. In 100% of tests, teams outperformed baseline models that made random predictions. In 46% of tests, teams also outperformed baseline models that relied on the simple average of ratings from training datasets. More notably, results uncovered a methodological challenge: multiplicative constraints on generalizability. Inferences about the accuracy and theoretical implications of machine learning efforts depended not only on their architecture, but also how they were trained, tested, and evaluated. For example, some teams performed better when tested on observations from the same (vs. different) subjects seen during training. Such results could be interpreted as evidence against claims of universality. However, such conclusions would be premature because other teams exhibited the opposite pattern. Taken together, results illustrate how big team science can be leveraged to understand the promises and limitations of machine learning methods in affective science and beyond.
Where the ‘bad’ and the ‘good’ go: A multi-lab direct replication report of Casasanto (2009, Experiment 1) Yuki Yamada, Jin Xue, Panpan Li, Susana Ruiz-Fernández, Asil Ali Özdoğru, et al. Memory and Cognition, 2025 Casasanto (Journal of Experimental Psychology: General, 138, 351–367, 2009) conceptualised the body-specificity hypothesis by empirically finding that right-handed people tend to associate a positive valence with the right side and a negative valence with the left side, whilst left-handed people tend to associate a positive valence with the left side and negative valence with the right side. Thus, this was the first paper that showed a body-specific space–valence mapping. These highly influential findings led to a substantial body of research and follow-up studies, which could confirm the original findings on a conceptual level. However, direct replications of the original study are scarce. Against this backdrop and given the replication crisis in psychology, we conducted a direct replication of Casasanto’s original study with 2,222 participants from 12 countries to examine the aforementioned effects in general and also in a cross-cultural comparison. Our results support Casasanto’s findings that right-handed people associate the right side with positivity and the left side with negativity and vice versa for left-handers.
Elicited emotion: effects of inoculation of an art style on emotionally strong images Amparo Caceres Gutierrez, Julián Tejada, Enrique García Fernández-Abascal Experimental Brain Research, 2025 The objective of this research is to study how the application of the Convolutional Neural Network (CNN) artistic filter can be an alternative to mitigate the emotional response to photographs with strong emotional content published in Internet news. Van Gogh’s artistic style was extracted through a CNN and inoculated with 64 IAPS images chosen to cover the entire emotional space. 140 university students of both sexes (70 men and 70 women) with an average age of 22 years, evaluated 128 stimuli, 64 original and 64 digitally inoculated, giving the appearance that they were painted with the artistic style of Van Gogh. For the evaluation of the stimuli, four groups were established under the conditions: 1 high arousal—positive valence, 2 negative valence—low arousal, 3 high arousal—negative valence and 4, low arousal, positive valence. The original images (OI) tended to produce less pleasant effects, while the images inoculated with filters made with CNN provoked reactions with a tendency to calm. The most significant changes in the emotional states are observed in the valence, the stimuli with the inoculation of the artistic style produces alterations with a tendency to pleasant effects. The averages of the comparisons of the dimensions valence and arousal of the OI and the CNN allow to show that there are differences in the emotional states, the results can permit the development of a methodology that, based on the inoculation of the artistic style of original paintings through CNN in emotionally strong images, a new image is created that replaces the strong images published in the Internet news.
Severity Classification of Anxiety and Depression Using Generalized Anxiety Disorder Scale and Patient Health Questionnaire: National Cross-Sectional Study Applying Classification and Regression Tree Models Andre Faro, Julian Tejada, Wael Al-Delaimy Jmir Public Health and Surveillance, 2025 Background Scalable and accurate screening tools are critical for public mental health strategies, especially in low- and middle-income countries (LMICs). While the Generalized Anxiety Disorder Scale (GAD-7) and Patient Health Questionnaire (PHQ-9) are widely used, their full application in large-scale programs can pose feasibility challenges. By contrast, shorter versions like GAD-2 and PHQ-2 reduce burdens but fail to capture symptom diversity. Objective This study aimed to optimize screening for anxiety and depression severity using classification and regression tree (CART) models, identifying concise and high-performing decision rules based on the GAD-7 and PHQ-9 items, and to test their reproducibility in 5 independent datasets. Methods A cross-sectional, nonprobabilistic study was conducted with 20,585 Brazilian adults from all 27 states and more than 3,000 cities, collected using digital outreach. Anxiety and depression symptoms were assessed using the GAD-7 and PHQ-9. CART models were trained and tested on bootstrapped samples (70% training, 30% testing), totaling 45,000 trees per scale. Each model used combinations of scale items and sociodemographic predictors. Robustness was evaluated via 10-fold cross-validation and evaluation across 3 hyperparameter configurations (minsplit and minbucket=500, 1000, 2000). Performance metrics included accuracy, sensitivity, specificity, precision, F1-score, and area under the curve (AUC). Results The CART models produced concise, high-performing decision rules—using only 2 items for the GAD-7 and 3 for the PHQ-9. No sociodemographic variable appeared in the final classification paths. For GAD-7, the models achieved an accuracy of 86.1% for minimal or mild severity and 85.1% for severe cases, with both categories showing AUC values above 0.900. By contrast, the moderate severity class had lower performance, with accuracy around 51% and an AUC of 0.728. For PHQ-9, the models achieved 81.7% accuracy for minimal or mild cases and 78.8% for severe cases, with AUCs again exceeding 0.900 for the extreme classes; the moderate or moderately severe class showed 66.9% accuracy and an AUC of 0.776. The most frequently repeated rules included the following: “GAD2<2 and GAD4<2” for identifying minimal or mild anxiety and “GAD2≥2 and GAD4=3” for severe anxiety; for depression, “PHQ2<2and PHQ4<2” for minimal or mild cases and “PHQ2≥2 and PHQ8≥2” for severe cases. These rule-based models demonstrated stable performance across thousands of bootstrapped replications and showed reproducibility in 5 independent datasets through external validation. Conclusions CART models enabled simplified, symptom-specific pathways for stratifying anxiety and depression severity with high precision and minimal item burden. These rule-based shortcuts offer an efficient alternative to fixed short forms (eg, GAD-2, PHQ-2) by preserving symptom diversity and severity discrimination. The findings support and lay the groundwork for adaptive, cost-effective screening and intervention models, especially in resource-limited settings and LMICs.
Avaliação de viés em expressões faciais na Ansiedade Social: uma revisão sistemática T Dantas, I Viana-Menezes, BSS Ivo, PJA Cruz, J Tejada Psico-USF 31, e295943 , 2026 2026
University students’ perceptions and adoptions of AI: a cross-national study F Marmolejo-Ramos, R Abadia, Ö Karakale, C Barrera-Causil, ... Discover Artificial Intelligence , 2026 2026 Citations: 1
Between tool and trouble: Student attitudes toward AI in programming education F Marmolejo-Ramos, S Rojas-Galeano, J Tejada Learning Letters 5, 66-66 , 2026 2026 Citations: 4
Analyzing the Temporal Factors for Anxiety and Depression Symptoms with the Rashomon Perspective M Cavus, PĹ Biecek, J Tejada, F Marmolejo-Ramos, A Faro arXiv preprint arXiv:2601.20874 , 2026 2026
From human artefact to machine output: automating the “art” of psychological measurement F Marmolejo-Ramos, O Bulut, L Anunciaçáo, L Marques, A Barthakur, ... Journal of Psychology and AI 1 (1), 2561692 , 2025 2025 Citations: 4
VENturing into machine learning for the morphological analysis of von Economo neurons I Banovac, OJ Bruton, L Mercado-Díaz, J Tejada, F Marmolejo-Ramos Scientific Reports , 2025 2025
Longitudinal correlation between frequency and duration of sitting, standing, and walking patterns and executive function in children: A two-year follow-up study from the … DN Oliveira, ECM Silva, JCN Melo, L Barboza, J Ywgne, L Gandarela, ... 2025
Severity Classification of Anxiety and Depression Using Generalized Anxiety Disorder Scale and Patient Health Questionnaire: National Cross-Sectional Study Applying … A Faro, J Tejada, W Al-Delaimy JMIR Public Health and Surveillance 11, e72591 , 2025 2025 Citations: 1
Measuring the semantic priming effect across many languages EM Buchanan, K Cuccolo, T Heyman, N Van Berkel, NA Coles, A Iyer, ... Nature Human Behaviour, 1-20 , 2025 2025 Citations: 11
Effects of physically active lessons and active breaks on cognitive performance and health indicators in elementary school children: A cluster randomized trial JCN Melo, J Tejada, ECM Silva, J Ywgne, DN Oliveira, L Gandarela, ... International Journal of Behavioral Nutrition and Physical Activity 22 (1), 96 , 2025 2025 Citations: 12
Implicit Gender Bias Trumps Spatial Metaphor: Evidence from a Multi Laboratory Study on Ambiguous Face Perception F Marmolejo-Ramos, J Tejada, L Shamoa-Nir, M Parzuchowski, ... ScienceOpen Posters , 2025 2025
Beyond p-values: Rethinking Statistical Frameworks for Addressing the Replication Crisis F Marmolejo-Ramos, JD Perezgonzalez, R Ospina, F Hernandez-Barajas, ... 2025
Big team science reveals promises and limitations of machine learning efforts to model physiological markers of affective experience NA Coles, B Perz, M Behnke, JC Eichstaedt, SH Kim, TN Vu, C Raman, ... Royal Society Open Science 12 (6) , 2025 2025 Citations: 9
Where the ‘bad’and the ‘good’go: A multi-lab direct replication report of Casasanto (2009, Experiment 1) Y Yamada, J Xue, P Li, S Ruiz-Fernández, AA Özdoğru, Ş Sarı, SC Torres, ... Memory & Cognition 53 (4), 1140-1146 , 2025 2025 Citations: 20
Elicited emotion: effects of inoculation of an art style on emotionally strong images AC Gutierrez, J Tejada, EG Fernández-Abascal Experimental Brain Research 243 (4), 89 , 2025 2025 Citations: 2
Epilepsy, compulsion and oxytocin: Insights from behavioral sequences, using neuroethology and complexity systems approaches SS Marroni, VR Santos, OW Castro, J Tejada, J Santos, JAC de Oliveira, ... Epilepsy & Behavior 164, 110273 , 2025 2025
Factors influencing trust in algorithmic decision-making: an indirect scenario-based experiment F Marmolejo-Ramos, R Marrone, M Korolkiewicz, F Gabriel, G Siemens, ... Frontiers in Artificial Intelligence 7, 1465605 , 2025 2025 Citations: 23
Network analysis of the symptoms of anxiety and depression, and sequelae due to COVID-19 W Lisboa, J Tejada, DC Seco-Ferreira, A Faro Suma Psicológica 32 (2), 158-169 , 2025 2025
The effect of context switching on predictive learning with funeral and marketing stimuli JA Aristizábal, LY Rodríguez-González, J Tejada, KX Pardo-Fajardo, ... DIGITAL. CSIC , 2025 2025
A worldwide test of the predictive validity of ideal partner preference matching. PW Eastwick, J Sparks, EJ Finkel, EM Meza, M Adamkovič, P Adu, T Ai, ... Journal of Personality and Social Psychology 128 (1), 123 , 2025 2025 Citations: 27
MOST CITED SCHOLAR PUBLICATIONS
A multi-lab test of the facial feedback hypothesis by the Many Smiles Collaboration NA Coles, DS March, F Marmolejo-Ramos, JT Larsen, NC Arinze, ... Nature human behaviour 6 (12), 1731-1742 , 2022 2022 Citations: 153
Scaffolding in immersive virtual reality environments for learning English: an eye tracking study J Bacca-Acosta, J Tejada, R Fabregat, Kinshuk, J Guevara Educational technology research and development 70 (1), 339-362 , 2022 2022 Citations: 84
The Sci-Hub effect on papers’ citations JC Correa, H Laverde-Rojas, J Tejada, F Marmolejo-Ramos Scientometrics 127 (1), 99-126 , 2022 2022 Citations: 66
X-PloRat: a software for scoring animal behavior in enclosed spaces J Tejada, KT Chaim, S Morato Psicologia: Teoria e Pesquisa 33, e3322 , 2017 2017 Citations: 64
The epilepsies: Complex challenges needing complex solutions J Tejada, KM Costa, P Bertti, N Garcia-Cairasco Epilepsy & Behavior , 2013 2013 Citations: 53
Combined Role of Seizure-Induced Dendritic Morphology Alterations and Spine Loss in Newborn Granule Cells with Mossy Fiber Sprouting on the Hyperexcitability of a Computer … J Tejada, N Garcia-Cairasco, AC Roque PLOS Computational Biology 10 (5), e1003601 , 2014 2014 Citations: 41
Characterization of the rat exploratory behavior in the elevated plus-maze with Markov chains J Tejada, GG Bosco, S Morato, AC Roque Journal of neuroscience methods 193 (2), 288-295 , 2010 2010 Citations: 32
Mathematical methods to model rodent behavior in the elevated plus-maze R Arantes, J Tejada, GG Bosco, S Morato, AC Roque Journal of neuroscience methods 220 (2), 141-148 , 2013 2013 Citations: 29
A worldwide test of the predictive validity of ideal partner preference matching. PW Eastwick, J Sparks, EJ Finkel, EM Meza, M Adamkovič, P Adu, T Ai, ... Journal of Personality and Social Psychology 128 (1), 123 , 2025 2025 Citations: 27
Computational models of dentate gyrus with epilepsy-induced morphological alterations in granule cells J Tejada, AC Roque Epilepsy & Behavior 38, 63-70 , 2014 2014 Citations: 27
Factors influencing trust in algorithmic decision-making: an indirect scenario-based experiment F Marmolejo-Ramos, R Marrone, M Korolkiewicz, F Gabriel, G Siemens, ... Frontiers in Artificial Intelligence 7, 1465605 , 2025 2025 Citations: 23
Where the ‘bad’and the ‘good’go: A multi-lab direct replication report of Casasanto (2009, Experiment 1) Y Yamada, J Xue, P Li, S Ruiz-Fernández, AA Özdoğru, Ş Sarı, SC Torres, ... Memory & Cognition 53 (4), 1140-1146 , 2025 2025 Citations: 20
Building and validation of a set of facial expression images to detect emotions: a transcultural study J Tejada, RMK Freitag, BFM Pinheiro, PB Cardoso, VRA Souza, LS Silva Psychological Research 86 (6), 1996-2006 , 2022 2022 Citations: 20
Effects of Physically Active Lessons on Movement Behaviors, Cognitive, and Academic Performance in Elementary Schoolchildren: ERGUER/Aracaju Project LLS Barboza, H Schmitz, J Tejada, ECM Silva, ASS Oliveira, LB Sardinha, ... Journal of Physical Activity and Health 18 (7), 757-766 , 2021 2021 Citations: 20
Looking for complexity in quantitative semiology of frontal and temporal lobe seizures using neuroethology and graph theory P Bertti, J Tejada, APP Martins, MLC Dal-Cól, VC Terra, JAC de Oliveira, ... Epilepsy & Behavior 38, 81-93 , 2014 2014 Citations: 20
Morphological Alterations in Newly Born Dentate Gyrus Granule Cells That Emerge after Status Epilepticus Contribute to Make Them Less Excitable J Tejada, GM Arisi, N García-Cairasco, AC Roque PloS one 7 (7), e40726 , 2012 2012 Citations: 19
Characterization of rat behavior in the elevated plus-maze using a directed graph J Tejada, GG Bosco, S Morato, AC Roque Journal of neuroscience methods 184 (2), 251-255 , 2009 2009 Citations: 15
Effects of physically active lessons and active breaks on cognitive performance and health indicators in elementary school children: A cluster randomized trial JCN Melo, J Tejada, ECM Silva, J Ywgne, DN Oliveira, L Gandarela, ... International Journal of Behavioral Nutrition and Physical Activity 22 (1), 96 , 2025 2025 Citations: 12
Measuring the semantic priming effect across many languages EM Buchanan, K Cuccolo, T Heyman, N Van Berkel, NA Coles, A Iyer, ... Nature Human Behaviour, 1-20 , 2025 2025 Citations: 11
Learning to Follow Directions in English Through a Virtual Reality Environment: An Eye Tracking Study and Evaluation of Usability J Bacca-Acosta, J Tejada, C Ospino-Ibañez Designing, Deploying, and Evaluating Virtual and Augmented Reality in … , 2021 2021 Citations: 11
