In recent years, the State oral exam has become less and less a mere “recitation” and more and more a skills-based assessment: organizing a speech, arguing, making connections, handling unexpected questions, using subject-specific vocabulary accurately. With theMaturità 2026, this direction is consolidated and calls on schools to make a leap in quality in the design oforal exam simulations: it’s not enough to “do oral questioning”; you need to train authentic performances that can be assessed with clear indicators.
For teachers, the challenge is twofold: on the one hand maintaining disciplinary rigor, on the other making systematic what often remains implicit (argumentative strategies, coherence, time management, quality of connections). This article proposes an operational method to redesign oral assessments in class and shows how tools likeStudierAIcan support the creation of prompts, rubrics, and feedback consistent with the new expectations, without increasing workload.
What changes with the Maturità 2026 in the oral exam: criteria, skills, and new expectations
Although there are differences by track and guidance may be updated over time, the orientation of the oral exam in theMaturità 2026pushes toward a more transparent, competency-based evaluation. Concretely, it strengthens the idea that students must demonstrate not only knowledge, but also the ability to use it appropriately: selecting information, building links, supporting a thesis, responding to follow-up prompts, integrating perspectives.
There are three main teaching implications forexam preparationin class:
- Centrality of transversal skills: clarity of exposition, organization of the discourse, use of examples, management of anxiety and time, listening and responding to questions.
- Argumentation and connections: not “forced connections,” but well-motivated links (cause-effect, analogies, continuities/breaks, applications, ethical and social impacts).
- Indicator-based assessment: the importance of rubrics and observable criteria increases (accuracy, coherence, depth, vocabulary, autonomy, ability to rework).
From a pedagogical standpoint, this orientation is consistent with what we know about durable learning: active recall (retrieval practice) and elaboration (explaining, justifying, connecting) improve understanding and transfer. The oral exam, if well designed, becomes a formative tool: it doesn’t just “measure,” it trains the skills that are then assessed.
For teachers this means making explicit to students what matters: it’s not enough to “know the chapter”; you must be able to turn it into a structured discourse with examples and appropriate lexical choices. And it also means training the management of the unexpected: an authentic oral exam includes follow-up prompts, requests for clarification, shifts in perspective.
From the classic oral questioning to authentic simulation: how to redesign oral assessments in class
The difference between traditional oral questioning and authentic simulation is not only in the content, but in thesetting: instructions, timing, the student’s role, the type of questions, assessment criteria. To align with the expectations of the 2026 oral exam, it is advisable to design assessments that reproduce some key conditions: starting from a prompt, time to organize, structured exposition, follow-up questions for deepening and making connections.
A practical model, replicable in any subject, is the “three-phase simulation” (15–20 minutes total, adaptable):
- Phase 1 – Preparation (2–4 minutes): the student receives a prompt or input (short text, graph described orally, problem, historical document, scenario). They can jot down an outline.
- Phase 2 – Exposition (5–7 minutes): presentation with an explicit structure (introduction, development, examples, conclusion). Assess coherence and subject-specific vocabulary.
- Phase 3 – Interview and follow-ups (5–8 minutes): clarification questions, deepening, connections, applications. This is where the ability to reason—not just remember—emerges.
To increase authenticity without complicating management, two high-yield teaching adjustments can be introduced:
- Prompts with constraints: for example “use at least two key concepts,” “provide an example,” “end with a consequence or a limitation.” Constraints guide structure and make performance assessable.
- Questions of increasing level: start with comprehension (explain/define), move to analysis (compare/distinguish), reach evaluation (argue/justify) and transfer (apply to a case).
This progression connects to evidence on scaffolding: support the student at the beginning and gradually reduce help, increasing autonomy and complexity. In class, this can translate into cycle-based planning: initial simulations more guided (provided outline, predictable questions), then progressively more open (broader prompt, more unpredictable follow-ups).
A crucial issue is assessment: if expectations change, the way of “giving the grade” must change too. Using rubrics (even lean ones, 4 levels) helps to: (1) make criteria transparent, (2) reduce variability between teachers and across different days, (3) turn the grade into feedback oriented toward improvement. In the context oforal exam simulations, the rubric is also a metacognitive tool: the student understands “what to do” to move up a level.
Finally, an often underestimated aspect: inclusion. Authentic simulation must not become a barrier for those with linguistic fragilities, performance anxiety, SEN/SLD. On the contrary, good design makes it possible to offer different access routes to the same competence (for example outlines, maps, timing, clearer instructions), while keeping the assessment objective unchanged.
Personalization and data: building targeted training pathways by level, SEN/SLD, and class objectives
If the 2026 oral exam assesses complex skills, then effective preparation requirespersonalized oral questioningand finer monitoring of “how” the student responds, not only “what” they know. Personalizing does not mean creating 25 different pathways, but designing controlled variants of the same assessment: same objectives, different supports and levels of complexity.
A sustainable approach is to work on three levers, all measurable with observable indicators:
- Lever 1 – Prompt complexity: from guided inputs (questions, keywords) to open inputs (scenario, document).
- Lever 2 – Allowed supports: outline, concept map, subject glossary, list of argumentative connectors, extra time to organize the response (in line with PDP/PEI).
- Lever 3 – Depth of follow-ups: more descriptive questions to consolidate, comparison/evaluation questions to strengthen, transfer questions for excellence.
For SEN/SLD, the key is to distinguish between what is being assessed and what is an ancillary obstacle. If the objective is “argue coherently,” then an outline or map can be allowed without lowering the bar. If the objective is “use of subject-specific vocabulary,” a support glossary can be provided, assessing the choice and appropriate use of terms.
Monitoring progress becomes easier if you work with a few stable indicators for the whole class, for example 5 recurring dimensions:
- Disciplinary accuracy (correctness of concepts, procedures, references).
- Structure and coherence (beginning-development-conclusion, hierarchy of ideas, logical links).
- Argumentation (thesis, evidence/examples, rebuttal or limits).
- Language and communication (vocabulary, precision, register, voice management).
- Autonomy and flexibility (responding to follow-ups, rephrasing, relevant connections).
With these indicators, each simulation produces useful data: not “it went well/badly,” but “needs to improve structure,” “has good knowledge but argues little,” “connections are present but not justified.” It’s a paradigm shift: assessment becomes diagnosis and guidance, and the class can work on weekly micro-goals (for example: one week dedicated to connectors, one to examples, one to an effective conclusion).
How StudierAI supports adapting oral simulations: scenarios, rubrics, and feedback consistent with the 2026 oral exam
When you decide to bring more authentic simulations into the classroom, the main obstacle is often organizational: creating varied prompts, calibrating difficulty, preparing follow-up questions, building rubrics, and writing feedback takes time. Here a tool likeStudierAIcan become a teaching ally, if used intentionally and with clear criteria: the teacher remains the director of the assessment, AI speeds up design and consistency.
Here are four concrete uses, consistent with the new expectations of the oral exam and with evidence-based teaching (clarity of criteria, spaced practice, timely feedback):
- Generating authentic prompts by subject and competence: scenarios, documents, problems, instructions with constraints (e.g., “define, compare, argue a limitation”). This helps make variety systematic and avoid overly repetitive prompts.
- Calibrating difficulty and differentiation: starting from a topic, you can obtain basic/intermediate/advanced versions, or versions with supports (keywords, guided outline) useful for SEN/SLD, while keeping the same assessment objectives.
- Simulating an “exam board” with follow-ups: beyond the initial questions, it is useful to have a bank of follow-ups of increasing level (clarify, give an example, compare, apply, evaluate). This trains the flexibility required in the oral exam and reduces on-the-spot improvisation.
- Structured rubrics and feedback: define indicators (accuracy, coherence, argumentation, language, autonomy) and level descriptors. Feedback can be turned into an operational “next step” (a single improvement priority for the week).
An example of a (sustainable) weekly routine could be: 1 short simulation on a rotating basis (10–12 minutes), a 5-indicator rubric with 4 levels, and two-line feedback:strength+next step. In this way the class does spaced practice (better than a few long simulations concentrated at the end of the year) and the teacher builds documentation that is also useful for assessment transparency.
From a time standpoint, the advantage is clear: prompts, variants, and rubrics can be prepared in advance and reused. But the most important advantage is consistency: when the class works for weeks with the same indicators, oral performance becomes a trainable skill, not an unpredictable event.
If you want to experiment quickly, you canstart for freeorsign up for freeand build a first complete simulation (prompt + follow-ups + rubric) to use already in the next module. To learn more about the team’s approach and educational vision, you can also find the pagewho we are.
In summary: to respond to the changes in the 2026 oral exam, it is worth shifting attention from the quantity of repeated content to the quality of performance: structure, argumentation, well-motivated connections, and the ability to handle follow-ups. Authentic simulations, made sustainable through routines and rubrics, make it possible to train these skills progressively and inclusively.
The guiding teaching question can be: “What is the next oral micro-skill I want to grow in my class?” If the answer is clear, then digital and AI tools become truly useful: they do not replace the educational relationship, but make it easier to design, differentiate, and provide high-quality feedback, in a way consistent with the criteria of theMaturità 2026.
