arXiv subjects

Alessandro Grecucci

Publications and source records attributed to Alessandro Grecucci.

10 recordsLinked to original sources

arXiv2026

MyMentorLLM: A psychotherapy GenAI environment with multimodal voice/text patients, trainees and experts for deliberate practice

Psychotherapists need repeated training and supervision by experts; however, scalability is problematic. Here we present MyMentorLLM, a multimodal voice- and text-based simulation environment for deliberate practice, used to generate 2,100 complete Cognitive Behavioural Therapy (CBT) training sessions. Each session links a DSM-5-TR-grounded patient (with major depressive, generalised anxiety or borderline personality disorder), a therapist-in-training and an expert supervisor. As an initial implementation, we adopted CBT because its structured procedures and competency-based supervision facilitate standardised simulation and evaluation. Sessions were analysed for emotional dynamics, therapeutic competence and diagnostic accuracy. Simulated patients expressed disorder-congruent emotional profiles, which trainee therapists mirrored as in real human counselling. The quality of supervision differed across LLMs: while most models overestimated trainees' competences, native speech-to-speech was closest to human scores. Supervisors' feedback led to better diagnoses in simulated psychotherapists in 5 out of 7 LLMs, and symptom identification accuracy increased with model size. This work shows that simulation of deliberate practice is possible for CBT training, although patient fidelity, calibration of supervisors, and harmful feedback should be evaluated together.

cs.CL↗

arXiv2026

A Tensor Network Framework for Interpretable Graph Analysis of Brain Networks

Identifying robust neurobiological signatures of brain disorders requires machine learning approaches that combine predictive performance with interpretable representations of feature interactions. Here we introduce a quantum-inspired framework based on tensor network machine learning that learns distributed representations of gray-matter features encoded in a Matrix Product State representation, a variational ansatz originally developed for quantum many-body systems. The trained model is then used not only as a classifier but to extract quantum connected correlations between features, which encode higher-order feature interactions and which we map onto a weighted graph. This construction allows us to track, within a single representation, both: (i) the global spectral properties of the network (capturing collective learning dynamics), and (ii) node-level centrality measures (providing interpretable signatures of individual brain regions). Using repeated train-test sampling schemes, we analyze two classification tasks on structural MRI data as examples of complex brain disorders: healthy controls versus schizophrenia and versus bipolar disorder. Node-level analysis identifies a stable set of gray-matter features, most prominently Heschl gyrus, insular cortex, and frontal regions, that act as hubs across multiple centrality measures and across resamplings. These centralities display lower variability across resamplings than Shapley values, supporting the interpretive value of the network representation. The bipolar feature set emerges as a subset of the schizophrenia one, consistent with the hierarchically organized neuroanatomical alterations reported in neuroimaging studies and offering a network-based characterization of this hierarchy.

q-bio.NC↗

arXiv2026

Cognitive networks reconstruct mindsets about STEM subjects and educational contexts in almost 1000 high-schoolers, University students and LLM-based digital twins

Attitudes toward STEM develop from the interaction of conceptual knowledge, educational experiences, and affect. Here we use cognitive network science to reconstruct group mindsets as behavioural forma mentis networks (BFMNs). In this case, nodes are cue words and free associations, edges are empirical associative links, and each concept is annotated with perceived valence. We analyse BFMNs from N = 994 observations spanning high school students, university students, and early-career STEM experts, alongside LLM (GPT-oss) "digital twins" prompted to emulate comparable profiles. Focusing also on semantic neighbourhoods ("frames") around key target concepts (e.g., STEM subjects or educational actors/places), we quantify frames in terms of valence auras, emotional profiles, network overlap (Jaccard similarity), and concreteness relative to null baselines. Across student groups, science and research are consistently framed positively, while their core quantitative subjects (mathematics and statistics) exhibit more negative and anxiety related auras, amplified in higher math-anxiety subgroups, evidencing a STEM-science cognitive and emotional dissonance. High-anxiety frames are also less concrete than chance, suggesting more abstract and decontextualised representations of threatening quantitative domains. Human networks show greater overlapping between mathematics and anxiety than GPT-oss. The results highlight how BFMNs capture cognitive-affective signatures of mindsets towards the target domains and indicate that LLM-based digital twins approximate cultural attitudes but miss key context-sensitive, experience-based components relevant to replicate human educational anxiety.

cs.CL↗

arXiv2026

Complex networks map test anxiety and wellbeing levels in students and ChatGPT

Academic STEM evaluation can elicit anxiety, yet routine grading rarely captures how students semantically frame exams and wellbeing. We reconstruct these framings using behavioural forma mentis networks (BFMNs), that is, feature-rich networks of concepts linked by memory recalls and enriched with affective ratings and concreteness norms. We build BFMNs from 994 participants spanning STEM experts (N1 = 59), Italian high-schoolers (N2 = 206), physics undergraduates (N3 = 10), psychology undergraduates with math-anxiety levels (N4 = 301), and simulated students (N5 = 497) personified by a large language model (GPT-OSS 20B). Across all human groups, the concepts "exam" and "grade" were (i) perceived negatively, (ii) connected primarily to negatively valenced memory recalls, indicating a clustering of negative emotions around assessment, and (iii) framed through concepts eliciting fear and anticipation in most groups, including physics undergraduates (z-scores in the range [2.04, 2.53]). The semantic neighbourhoods of "anxiety" and "exam" overlapped three times more in human students than in GPT-based simulations, providing structural evidence of test anxiety in student populations. By contrast, experts displayed a neutral and more concrete network neighbourhood for "exam" (z = 1.87), with no clear trace of test anxiety. These negative assessment framings coexisted with positive representations of "wellbeing", which were rich in concrete associations in humans but linked to more abstract concepts in GPT digital twins. Overall, our results show that BFMNs offer a quantitative and interpretable framework to study academic anxiety and to distinguish human affective framing from current AI-based simulations

physics.ed-ph↗

arXiv2025

Textual forma mentis networks bridge language structure, emotional content and psychopathology levels in adolescents

We introduce a network-based AI framework for predicting dimensions of psychopathology in adolescents using natural language. We focused on data capturing psychometric scores of social maladjustment, internalizing behaviors, and neurodevelopmental risk, assessed in 232 adolescents from the Healthy Brain Network. This dataset included structured interviews in which adolescents discussed a common emotion-inducing topic. To model conceptual associations within these interviews, we applied textual forma mentis networks (TFMNs)-a cognitive/AI approach integrating syntactic, semantic, and emotional word-word associations in language. From TFMNs, we extracted network features (semantic/syntactic structure) and emotional profiles to serve as predictors of latent psychopathology factor scores. Using Random Forest and XGBoost regression models, we found significant associations between language-derived features and clinical scores: social maladjustment (r = 0.37, p < .01), specific internalizing behaviors (r = 0.33, p < .05), and neurodevelopmental risk (r = 0.34, p < .05). Explainable AI analysis using SHAP values revealed that higher modularity and a pronounced core-periphery network structure-reflecting clustered conceptual organization in language-predicted increased social maladjustment. Internalizing scores were positively associated with higher betweenness centrality and stronger expressions of disgust, suggesting a linguistic signature of rumination. In contrast, neurodevelopmental risk was inversely related to local efficiency in syntactic/semantic networks, indicating disrupted conceptual integration. These findings demonstrated the potential of cognitive network approaches to capture meaningful links between psychopathology and language use in adolescents.

cs.CY↗

arXiv2025

Increased GM-WM in a prefrontal network and decreased GM in the insula and the precuneus are associated with reappraisal usage: A data fusion approach

Emotion regulation plays a crucial role in mental health, and difficulties in regulating emotions can contribute to psychological disorders. While reappraisal and suppression are well-studied strategies, the combined contributions of gray matter (GM) and white matter (WM) to these strategies remain unclear due to methodological limitations in previous studies. To address this, we applied a data fusion approach using Parallel Independent Component Analysis (Parallel ICA) to GM and WM MRI images from 165 individuals. Parallel ICA identified two networks associated with reappraisal usage. Network 1 included a large lateral and medial prefrontal cortical network, overlapping with the default mode network (DMN) and adjacent WM regions. Higher reappraisal frequency was associated with greater GM-WM density within this network, and this network was negatively correlated with perceived stress. Network 2 included the insula, precuneus, sub-gyral, and lingual gyri in its GM portion, showing a negative association with reappraisal usage. The WM portion, adjacent to regions of the central executive network (CEN), was positively associated with reappraisal usage. Regarding suppression, no significant network was associated with this strategy. This study provides new insights into individual differences in reappraisal use, showing a positive association between reappraisal frequency and increased gray and white matter concentration in a large frontal network, including regions of the frontal DMN and the CEN. Conversely, subcortical areas exhibited reduced gray and white concentration.

q-bio.NC↗

arXiv2024

EEG Based Decoding of the Perception and Regulation of Taboo Words

In daily interactions, emotions are frequently conveyed and triggered through verbal exchanges. Sometimes, we must modulate our emotional reactions to align with societal norms. Among the emotional words, taboo words represent a specific category that has been poorly studied. One intriguing question is whether these word categories can be predicted from EEG responses with the use of machine learning methods. To address this question, Support Vector Machine (SVM) was applied to decode the word categories from Event Related Potential (ERP) in 40 native Italian speakers. 240 neutral, negative and taboo words were used to this aim. Results indicate that the SVM classifier successfully distinguished between the three-word categories, with significant differences in neural activity ascribed to the late positive potential mainly detected in the central-parietal-occipital and anterior right scalp areas in the time windows of 450-649 ms and 650-850 ms. These findings were in line with the established distribution pattern of the late positive potential. Intriguingly, the study also revealed that word categories were still detectable in the regulate condition. This study extends previous results on the domain of the cortical responses of taboo words, and how machine learning methods can be used to predict word categories from EEG responses.

q-bio.NC↗

arXiv2024

Resting-State fingerprints of Acceptance and Reappraisal. The role of Sensorimotor, Executive and Affective networks

Acceptance and Reappraisal are considered adaptive emotion regulation strategies. While previous studies have explored the neural underpinnings of these strategies using task based fMRI and sMRI, a gap exists in the literature concerning resting-state functional brain networks contributions to these abilities, especially for what concerns Acceptance. Another intriguing question is whether these strategies rely on similar or different neural mechanisms. Building on the well-known improved emotion regulation and increased cognitive flexibility of individuals who rely on acceptance, we expected to find decreased activity inside the Affective network and increased activity inside the Executive and Sensorimotor networks to be predicted of acceptance. We also expect that these networks may be associated at least in part with Reappraisal, indicating a common mechanism behind different strategies. To test these hypotheses, we conducted a functional connectivity analysis of resting-state data from 134 individuals (95 females). To assess acceptance and reappraisal abilities, we used the Cognitive Emotion Regulation Questionnaire (CERQ) and a group-ICA unsupervised machine learning approach to identify resting state networks. Subsequently, we conducted backward regression to predict acceptance and reappraisal abilities. As expected, results indicated that acceptance was predicted by decreased Affective, and increased Executive, and Sensorimotor networks, while reappraisal was predicted by an increase in the Sensorimotor network. Notably, these findings suggest both distinct and overlapping brain contributions to acceptance and reappraisal, with the Sensorimotor network potentially serving as a core common mechanism. These results not only align with previous findings but also expand upon them, demonstrating the complex interplay of cognitive, affective, and sensory abilities in emotion regulation.

q-bio.NC↗

arXiv2023

The neural signature of inner peace: morphometric differences between high and low accepters

Acceptance is an adaptive emotion regulation strategy characterized by an open and non-judgmental attitude toward mental and sensory experiences. While a few studies have investigated the neural correlates of acceptance in task-based fMRI studies, a gap remains in the scientific literature in dispositional use of acceptance, and how this is sedimented at a structural level. Therefore, the aim of the present study is to investigate the neural and psychological differences between infrequent acceptance users (i.e., low accepters) and frequent users (i.e., high accepters). Another question is whether high and low accepters differ in personality traits and emotional intelligence. To this aim, we applied, for the first time, a data fusion unsupervised machine learning approach (mCCA-jICA) to the gray matter (GM) and white matter (WM) of high accepters (N = 50), and low accepters (N = 78) to possibly find joint GM-WM differences in both modalities. Our results show that two covarying GM-WM networks separate high from low accepters. The first network showed decreased GM-WM concentration in a fronto-temporal-parietal circuit largely overlapping with the Default Mode Network, while the second network showed increased GM-WM concentration in portions of the orbito-frontal, temporal, and parietal areas, related to a Central Executive Network. At the psychological level, the high accepters display higher openness to experience compared to low accepters. Overall, our findings suggest that high accepters compared to low accepters differ in neural and psychological mechanisms. These findings confirm and extend previous studies on the relevance of acceptance as a strategy associated with well-being.

q-bio.NC↗

arXiv2023

Understanding the neural architecture of emotion regulation by comparing two different strategies: A meta-analytic approach

In the emotion regulation literature, the amount of neuroimaging studies on cognitive reappraisal led the impression that the same top-down, control-related neural mechanisms characterize all emotion regulation strategies. However, top-down processes may coexist with more bottom-up and emotion-focused processes that partially bypass the recruitment of executive functions. A case in point is acceptance-based strategies. To better understand neural commonalities and differences behind different emotion regulation strategies, in the present study we applied a meta-analytic method to fMRI studies of task-related activity of reappraisal and acceptance. Results showed increased activity in left-inferior frontal gyrus and insula for both strategies, and decreased activity in the basal ganglia for reappraisal, and decreased activity in limbic regions for acceptance. These findings are discussed in the context of a model of common and specific neural mechanisms of emotion regulation that support and expand the previous dual-routes models. We suggest that emotion regulation may rely on a core inhibitory circuit, and on strategy-specific top-down and bottom-up processes distinct for different strategies.

q-bio.NC↗