Toxic speech on online civic engagement platforms (CEPs) disproportionately affects marginalized groups and threatens the diversity of citizen voices. However, the deployment of AI-driven toxic speech detection (TSD) tools for CEPs faces complex challenges from legal, psychological, and technical perspectives that remain insufficiently explored. We present a first-of-its-kind interdisciplinary review of these challenges, focusing on the explainability of TSD systems, their compliance with European legal standards and offer a roadmap for ethical deployment. Our review reveals three main findings. First, although transparency in AI decision-making is necessary from both legal and psychological perspectives, assessing the explainability of AI-driven TSD tools, and their compliance with legal regulations within Europe, remains a significant challenge. Second, current explainability approaches, ranging from toxic span identification to advanced explainable AI methods, lack standardized metrics. This makes it difficult to assess their reliability and appropriateness for CEPs. Third, despite the importance of TSD, frameworks and best practices for CEPs are still lacking in existing literature. This paper aims to fill this gap by providing a holistic perspective on the challenges and solutions for TSD deployment. It provides the foundation for collaborative efforts to develop and standardize metrics, evaluation protocols, and best practices that can ensure AI decisions in CEPs are transparent, accountable, and aligned with users’ needs.

Up to 80% of patients with multiple system atrophy (MSA) exhibit tremor or tremor-like phenomena, yet the clinical features of these manifestations remain insufficiently characterized. Clearer phenotypic delineation may support differential diagnosis from other tremor-associated disorders, particularly Parkinson’s disease (PD). We aimed to determine characteristics and prevalence of tremor, distribution across MSA phenotypes and body regions, and its treatment responsiveness. We systematically searched PUBMED for articles reporting tremor in MSA, following PRISMA guidelines. Eligible studies were screened, with data reviewed by two reviewers and risk of bias assessed using JBI critical appraisal tools. Of 127 records, 25 studies met inclusion criteria, comprising 963 MSA patients. Tremor types and prevalence differed by phenotype. In MSA-P, rest tremor occurred in 26.5–38.1% and action tremor in 60.3–81.8%. In MSA-C, rest tremor occurred in 11-22.5%, action tremor in 44.9–97.5%. Among action tremor, a 3-Hz pseudo-orthostatic tremor occurred more often in MSA-C (76.3–97.5%) than MSA-P (24.1–38.3%). Polyminimyoclonus was present in 16.5% of MSA-P and 10.5% of MSA-C. Neurophysiologically, MSA-C typically showed ~ 3-Hz pseudo-orthostatic tremor, whereas MSA-P presented a 7–8-Hz postural tremor and 5-6-Hz rest tremor. Levodopa improved tremor in 11.8–92% of cases, with benefit more often in MSA-P and in action tremor. Preliminary reports suggest potential benefit of clonazepam and botulinum toxin, while a pilot high intensity focused ultrasound therapy showed no improvement. This first systematic review demonstrates that tremor in MSA is common and phenotypically diverse. Rest tremor is generally less frequent than action tremor and a 3-Hz pseudo-orthostatic tremor is characteristic of MSA-C.

Functional movement disorders (FMD) are common and disabling neurological conditions characterized by impaired voluntary motor control in the absence of structural disease. Accumulating evidence suggests that emotion regulation deficits contribute to the pathophysiology of FMD. This randomized, placebo-controlled pilot study investigated the clinical and neural effects of a digital cognitive reappraisal training program in patients with FMD, targeting emotion regulation as a potential therapeutic mechanism. Twenty adult patients diagnosed with clinically definite FMD (mean age = 50.7 years, 11 females) were randomly assigned in a 1:1 ratio to either an intervention group practicing cognitive reappraisal ( n = 11) or an active control group completing reflection exercises ( n = 9). Both groups underwent 14 brief, app-based training sessions over four weeks. Clinical outcomes were assessed before and after the intervention using standardized measures of mood, motor function, and overall illness severity. Simultaneously, brain activity at rest and during a standardized reappraisal task was recorded via electroencephalography, focusing on prefrontal alpha asymmetry as a neural marker of reappraisal engagement. Compared to the control group, the intervention group showed significantly greater improvements in mood (Mdiff = 0.542, p = .010), motor function (Mdiff = 0.542, p = .033), and overall illness severity (Mdiff = 0.528, p = .040). Both groups improved over time in mood ( p = .030), tremor severity ( p = .014), and subjectively rated motor function ( p = .022), suggesting non-specific benefits of structured emotional self-reflection. Crucially, electroencephalography data revealed a significant post-training increase in right-lateralized frontopolar resting-state activity in the intervention group (Mdiff = 13.29, p = .026), aligning with prior evidence linking right prefrontal activity to successful reappraisal. Although task-related EEG activity did not differ significantly between groups, a statistical trend ( p = .115, d = 1.14) suggests increased right frontal activity during reappraisal generation in the intervention group. Reappraisal training performance correlated with EEG changes and with the frequent use of the distancing strategy “relativizing,” implicating right prefrontal involvement in specific cognitive control processes. These findings provide preliminary evidence that cognitive reappraisal training delivered via mobile app is feasible and may enhance both clinical outcomes and prefrontal brain function in FMD. Future studies with larger samples are needed to confirm these effects and better understand the neural mechanisms of digital interventions for emotion regulation in functional neurological disorders.

Intuition often guides our thinking effectively, but it can also lead to consequential reasoning errors, underpinning poor decisions and biased judgments. Little is known about how people globally self-correct such intuitive reasoning errors and what enhances their correction. Defying prevailing models of reasoning, recent research suggests that people spontaneously correct only a few errors during deliberation; however, enhancing error monitoring and motivating further effort should increase error correction. Here, we study whether these mechanisms apply to reasoning across individualistic and collectivistic cultures (expected N = 33,000 participants from 67 regions). Participants will solve problems that elicit incorrect intuitions twice: first intuitively and then reflectively, allowing them to correct initial errors, in a 2 (feedback: absent vs present) × 2 (answer justification: absent vs present) between-participants design. The study will shed more light on the nature, generalisability, and promotion of corrective behaviour, crucial for understanding and improving reasoning worldwide.

Civic online participation platforms offer valuable opportunities to involve citizens in local governance and benefit from collective intelligence. Yet, vulnerable groups, such as older adults, people with disabilities, and the less educated, are often underrepresented in online political engagement. Aiming to empower these citizens to raise their voice online, we conducted two studies using an inclusive-by-design approach for developing an online civic engagement platform. In the first study, 39 individuals from two digitally low-performing European countries were surveyed about functions and features that would motivate and support their online participation. In the second study, focus groups with 13 digital and AI experts identified technical and informational requirements for effective use of the features desired by citizens. Our findings show the wishes for accessible, unbiased and secure AI-driven civic engagement platforms with transparency and user education about AI tools. In particular, chatbots require clear disclaimers and user guidance. Once citizens have been involved in the technical design process, both technical and informational feature preferences must be taken into account to avoid access and usability barriers or misunderstandings during the platform’s use. This is important to facilitate participation, especially for citizens from vulnerable groups.