Le abilità di Teoria della Mente tra docenti e studenti. Strumenti di valutazione e training formativo

Understanding what others think, know, feel and intend is one of the least visible and most pervasive conditions of educational life. This dissertation takes Theory of Mind (ToM; Premack & Woodruff, 1978) out of the laboratory and the early-childhood window where it has mostly been studied (Wellman, 2018). It asks what the construct means for teachers and pupils in primary and lower-secondary school, how it can be assessed from their own point of view, and whether it can be fostered through ordinary classroom activity. The research was funded under Italy’s National Recovery and Resilience Plan (PNRR) and is grounded in a neuroscientific and neuroeducational perspective. It moves from conceptual groundwork to the construction of assessment instruments and, finally, to an intervention in the field. The theoretical part opens from a deliberately wide angle. Evidence on brain-to-brain synchrony, in which neural activity becomes coordinated between people engaged in a shared task (Hasson et al., 2012), is used as the starting point for reading teaching and learning as a fundamentally relational event; studies conducted in real classrooms have linked such coordination to the quality of group interaction (Dikker et al., 2017). The first chapter follows ToM through its developmental trajectory and through the processes the literature has progressively differentiated and related to it: mentalizing, empathy, perspective-taking, cooperation, and the regulatory and reflective processes that allow mental-state inferences to be monitored and revised.The second chapter brings these strands into the school. It argues that explaining, discussing, correcting and evaluating all rest on continuous, fallible readings of the other’s mind, and that these readings shape expectations, participation and the reception of feedback on both sides of the teaching relationship. The perspective that results is symmetrical: teachers and students are both treated as interpreters of one another, and the cognitive and affective sides of that interpretation are kept in view together.Measurement is the subject of the third chapter. ToM has been assessed mostly with instruments of psychological and neuropsychological origin, typically performance tasks that probe mental-state understanding under controlled conditions (Wellman, 2018). These instruments are not designed for the classroom and tell us little about how teachers and pupils recognise the construct in their own practice. The research therefore developed two self-report questionnaires, the QToM Teacher and the QToM Student, through successive rounds of try-out, revision and administration, which the chapter reconstructs in order. The teacher version began with a try-out in December 2023 on 70 students of the corso di laurea in Scienze della Formazione Primaria at the University of Palermo. It covered four areas (responsibility, cooperation, mentalizing, empathy), and no reliability coefficient was computed. The first full administration, in the 2023/24 academic year, involved 166 prospective teachers. By then responsibility had been replaced by self-regulation, and the instrument reached Cronbach’s α = .895 (Cronbach, 1951). In the 2025/26 school year the questionnaire was administered to 110 in-service teachers at the Istituto Comprensivo «Rettore F. Evola» of Balestrate. It now included a fifth area, metacognitive reflection, and its reliability rose to α = .983. The student version, a linguistic and conceptual adaptation of the teacher instrument, followed a parallel course. Its first administration (January-February 2024) involved 100 pupils from schools in the Palermo area and covered four areas (α = .939). The revised version, administered in 2025 to 100 pupils of the Istituto Comprensivo «Rettore F. Evola», also included metacognitive reflection (α = .932). The chapter keeps these figures apart from any claim of equivalence. Samples differed in population, size and setting from one round to the next, so the coefficients document the internal consistency of each version more than a trajectory of improvement. The same caution applies to the structural analyses that guided item revision. The fourth chapter is the empirical centre of the thesis. An exploratory single-group intervention, with a post-test only and no control group (a pre-experimental design in Campbell & Stanley’s terms, 1963), was carried out with 117 pupils in four classes across the two school levels. It was designed from the dimensions identified in the previous chapter. Its method draws on guided discussion of narrative material, in the tradition of the Strange Stories (White et al., 2009) and of earlier school-age training programmes (Lecce et al., 2014) but adapts it freely to the timetable and culture of the institution rather than replicating a protocol. The chapter accounts for the selection of materials and for the structure of the meetings. It then presents a set of purpose-built tools: an initial self-assessment checklist, a monitoring checklist completed by pupils, and an observation checklist for teachers, all administered anonymously online. The QToM Student in its final 47-item form was administered at the end of the programme. Its internal consistency was good (α = .838), and the five areas showed acceptable to good reliability (α between .725 and .813). The substantive picture is more nuanced than a simple story of improvement. Pupils describe themselves as most at ease with perspective-taking and with cooperation when the activity is structured by the teacher. They report most difficulty in sustaining attention and emotional control, in revisiting their own reasoning, and in turning an understanding of others’ emotions into concrete supportive behaviour. Self-regulation had the lowest mean among the five areas (M = 2.77), a finding that the different instruments independently echo. The comparison between instruments is itself informative: it shows that one dimension can occupy different positions depending on whether it is probed through endorsement of a value, through recognition of an emotional experience, or through habitual conduct. Compared with the earlier administration, several self-reported empathic behaviours appear less frequent. This pattern is interpreted with caution: it may reflect a more demanding self-appraisal after a period of work on emotions, rather than a loss of competence. Any reading of the findings must also take account of the study’s limitations. The intervention was implemented without a control group, and the QToM Student was not administered at baseline, so that no pre-test is available and the two groups cannot be considered equivalent. The samples are convenience samples that differ across administrations; the checklists, being anonymous, cannot be linked to individual respondents, and the teachers completed theirs without blinding. For these reasons, the results are best understood as a descriptive profile of the group at the end of the programme, and they do not allow conclusions to be drawn about the efficacy of the intervention.The contribution of the work lies elsewhere: in the coherence of a path that connects a relational and neuroeducational reading of ToM, a set of instruments calibrated for the school context, and a documented, replicable intervention. This path also defines an agenda for future research, which calls for controlled, longitudinal and multi-site designs.