LGLucrezia Grassi

Lucrezia Grassi

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Lucrezia Grassi has a total of 32 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Modeling Conversation as a Prompt Planning Problem for Adaptive Human–Robot Interaction

    by Giulia Berettieri, Anna Allegra Bixio, Lucrezia Grassi, Carmine Tommaso Recchiuto, Antonio Sgorbissa

    Adaptive behavior is a fundamental requirement for social robots operating in real-world environments, where interactions must dynamically respond to both observable actions and inferred user states. In this work, we propose a novel framework that models human–robot interaction as a planning problem, where adaptive conversational behavior is driven by a symbolic planner based on the Planning Domain Definition Language (PDDL). Unlike other approaches that map actions onto sentences, we model actions as corresponding to high-level verbal interaction strategies, represented by a specific prompt fed to a Large Language Model (LLM) to generate dialogue in real-time. Additionally, the system continuously updates its representation of the interaction state by grounding user mental and nonmental states from dialogue using LLMs. This enables dynamic replanning, allowing the robot to adapt its behavior in real-time according to the evolving interaction context while maintaining a natural flow. We provide a descriptive analysis of the framework through a case study demonstrating behavior adaptivity, involving a robot acting as a tutor and a child asked to perform a task. The results illustrate the potential of combining symbolic planning and LLM-based dialogue generation to achieve flexible, context-aware, and adaptive human–robot interaction.

  • Open Access

    Technology-Assisted Cognitive Support for Older Adults With Mild Cognitive Impairment: Study Protocol for the DHEAL-COM COGNITIVE Quasi-Experimental Pilot Study

    by Roberta Bevilacqua, Elvira Maranesi, Elisa Felici, Rachele Alessandra Marziali, Arianna Margaritini, Arianna Sgolastra, Lucia Paciaroni, Alessandra Raccichini, Giuseppe Pelliccioni, Lucrezia Grassi, et al.

    Background: Mild cognitive impairment (MCI) represents a transitional phase between normal aging and dementia. While no pharmacological treatments have proven effective, nonpharmacological interventions, such as cognitive stimulation and psychosocial support, have shown promise. Integrating digital tools and socially assistive robotics may improve engagement, personalization, and access to care, contributing to the preservation of cognitive function and overall well-being in older adults. Objective: The DHEAL-COM COGNITIVE pilot study aims to evaluate the feasibility, acceptability, and preliminary efficacy of a sociotechnological intervention that combines digital cognitive stimulation, social robot interaction, and group training in digital and health literacy to counteract cognitive decline. Methods: This is a single-blind, quasi-experimental feasibility trial involving 60 older adults with MCI, recruited from the Neurology and Alzheimer Units of IRCCS INRCA (INRCA – IRCCS Istituto Nazionale di Ricovero e Cura per Anziani). Participants were allocated into 2 groups: the experimental group received home-based cognitive training via the Brainer app, a month-long interaction with the NAO social robot, and weekly group sessions on eHealth literacy and cognitive stimulation; the control group received a well-being booklet with optional activities. Assessments were conducted at baseline, postintervention after 12 weeks, and follow-up after 3 months, using validated instruments, such as the Montreal Cognitive Assessment, the EQ-5D-5L, the Italian version of the eHealth Literacy Scale, the System Usability Scale, the unified theory of acceptance and use of technology, and the Psychological Well-Being Scale. Descriptive statistics will be used to summarize feasibility, usability, and acceptability outcomes, while preliminary differences over time will be explored using appropriate inferential analyses, acknowledging that the study is not powered to detect effectiveness. Results: Patient recruitment began in March 2025 and continued through September 2025. As of April 2026, 36 patients have been recruited, with 18 participants per group. The trial started in May 2025 and ran throughout the year. Results will focus on changes in cognitive performance, psychological well-being, quality of life, eHealth literacy, and user acceptability and usability of the technologies. Findings will be analyzed quantitatively and qualitatively and are expected to be published by the end of 2026. Conclusions: The DHEAL-COM COGNITIVE study proposes an innovative, integrated intervention for older adults with MCI, using digital platforms and social robotics to support cognitive health and autonomy. This pilot study represents a promising step toward a more integrated and technology-supported model of care for older adults with MCI.

  • Why Are You Upset?A Framework for Dynamic Behavior Management in Child-Robot Interactions

    by Giulia Berettieri, Anna Allegra Bixio, Lucrezia Grassi, Carmine Tommaso Recchiuto, Antonio Sgorbissa

  • Phase-2 Clinical Trial of a Diversity-Aware Social Robot for Delirium Prevention and Reduction in Hospitalized Older Adults

    by Lucrezia Grassi, Silvia Ottaviani, Carmine Recchiuto, Fiammetta Monacelli, Alessio Nencioni, Antonio Sgorbissa

  • Participatory Design of Socially Assistive Robots for Alzheimer’ Care

    by Francesca Corrao, Maura Casadio, Lucrezia Grassi, Alice Nardelli, Antonio Sgorbissa, Carmine Tommaso Recchiuto

  • Open Access

    A Socially Assistive Robot for Delirium Prevention in Older Inpatients: Study Design and Feasibility Results

    by Lucrezia Grassi, Silvia Ottaviani, Stefania Peruzzo, Fiammetta Monacelli, Alessio Nencioni, Carmine Tommaso Recchiuto, Antonio Sgorbissa

  • Moderating multi-party conversations with social robots

    by Lucrezia Grassi, Carmine Tommaso Recchiuto, Antonio Sgorbissa

    Social robotics is a multidisciplinary field focused on designing and implementing robots capable of interacting with humans in social environments. However, group conversations challenge robots in interpreting social signals for effective participation. This study evaluates control policies for moderating multi-party conversation dynamics using a humanoid robot. The system employs a cloud-based framework to calculate speaker dominance as a weighted combination of speaking time and word count, while the Louvain algorithm identifies subgroups among participants. Control policies aim to minimize dominance disparities and subgroup formation, fostering balanced participation and group cohesion. A study with 300 middle school students compared these policies to a baseline in which the robot did not address individuals directly. The results demonstrated that the proposed policies reduced dominance gaps and subgroup formation, promoting more balanced interactions. These findings highlight the potential applicability of the approach across education, healthcare, and entertainment.

  • Open Access

    Strategies for Controlling the Conversation Dynamics in Multi-Party Human-Robot Interaction

    by Lucrezia Grassi, Carmine Tommaso Recchiuto, Antonio Sgorbissa

    This article tackles the research question of whether it is possible to control conversation dynamics in a multi-party scenario using easily implementable solutions on off-the-shelf robotic platforms. To this end, we expanded upon our previously developed cloud robotic architecture by incorporating policies aimed at managing conversation dynamics through selective addressing of individuals, with the ultimate goal of balancing or unbalancing users’ participation or making subgroups of participants interact. Specifically, we computed the dominance of each speaker as a weighted sum of their speaking time and the number of words spoken within a moving window and used the Louvain algorithm to partition speakers into a set of non-overlapping communities. We then implemented six control policies, which were applied by the robot. Two of them, named BH and BS, aim to reduce dominance error (i.e., the difference in dominance between the most and least dominant speakers—both policies give the floor to the less dominant speaker). Two other policies, UH and US, are designed to increase the dominance error (both give the floor to the most dominant speaker). Finally, CH and CS aim to reduce the community error (i.e., the difference between the actual number of detected subgroups among speakers and the ideal target of a single group to which all speakers belong). Policies BH, UH, and CH (with “H” standing for “hard”) do not allow any exceptions to the policy rules, while BS, US, and CS (with “S” for “soft”) permit exceptions. To test the impact of these policies, we conducted a between-subjects study (N = 300) involving middle school students engaging in dialogue with a humanoid robot acting as a moderator. The study compared five conditions: in four of them, the robot used information gathered during the conversation to decide which speaker to address, applying one of the control policies—BH, BS, CH, or CS. The policies UH and US were excluded, as having a robot consistently give the floor to the most dominant child may raise ethical concerns. In the fifth condition, a baseline neutral policy (N) was applied, in which the robot did not explicitly address any speaker. The results imply that a robot using the proper control policies can influence conversation dynamics to keep both dominance error and community error significantly lower than those of a robot using the baseline policy, leading to more balanced participation and a reduction in the number of subgroups. Indeed, statistically significant differences have been found between the five policies considered in the dominance and community errors. However, no statistically significant differences in user experience—as measured by three scales of the validated SASSI questionnaire—were found when the robot used one of the control policies, as compared to the baseline, suggesting that participants are not negatively impacted by the robot’s attempt to control the conversation.


  • Open Access

    Technology-Assisted Cognitive Support for Older Adults with Mild Cognitive Impairment: Study Protocol for the DHEAL-COM Randomized Pilot Trial (Preprint)

    by Roberta Bevilacqua, Elvira Maranesi, Elisa Felici, Rachele Alessandra Marziali, Arianna Margaritini, Arianna Sgolastra, Lucia Paciaroni, Alessandra Raccichini, Giuseppe Pelliccioni, Lucrezia Grassi, et al.

    BACKGROUND Mild Cognitive Impairment (MCI) represents a transitional phase between normal aging and dementia. While no pharmacological treatments have proven effective, non-pharmacological interventions, such as cognitive stimulation and psychosocial support, have shown promise. Integrating digital tools and social assistive robotics may improve engagement, personalization, and access to care, contributing to the preservation of cognitive function and overall well-being in older adults. OBJECTIVE The DHEAL-COM pilot study aims to evaluate the feasibility, acceptability, and preliminary efficacy of a socio-technological intervention that combines digital cognitive stimulation, social robot interaction, and group training in digital and health literacy to counteract cognitive decline and promote autonomy among older adults with MCI. METHODS This is a single-blind, randomized controlled feasibility trial involving 60 older adults with MCI, recruited from the Neurology and Alzheimer Units of IRCCS INRCA. Participants are randomized into two groups: the experimental group receives home-based cognitive training via the Brainer app, a month-long interaction with the NAO social robot, and weekly group sessions on eHealth and digital literacy; the control group receives a well-being booklet with optional activities. Assessments are conducted at baseline (T0), post-intervention after 12 weeks (T1), and follow-up after 3 months (T2), using validated instruments such as MoCA, EQ-5D-5L, IT-eHEALS, SUS, UTAUT, and PWB. RESULTS Patient recruitment began in March 2025 and will continue through September 2025. The trial started in May 2025 and will run throughout the year. Results will focus on changes in cognitive performance, psychological well-being, quality of life, eHealth literacy, and user acceptability and usability of the technologies. Findings will be analyzed quantitatively and qualitatively and are expected to be published by 2026. CONCLUSIONS The DHEAL-COM study proposes an innovative, integrated intervention for older adults with MCI, using digital platforms and social robotics to support cognitive health and autonomy. This pilot study represents a promising step toward a more integrated and technology-supported model of care for older adults with MCI. CLINICALTRIAL The study protocol was recorded in ClinicalTrials.gov on the number NCT06984367.

  • Open Access

    Moderating Group Conversation Dynamics with Social Robots

    by Lucrezia Grassi, Carmine Tommaso Recchiuto, Antonio Sgorbissa

    This research investigates the impact of social robot participation in group conversations and assesses the effectiveness of various addressing policies. The study involved 300 participants, divided into groups of four, interacting with a humanoid robot serving as the moderator. The robot utilized conversation data to determine the most appropriate speaker to address. The findings indicate that the robot's addressing policy significantly influenced conversation dynamics, resulting in more balanced attention to each participant and a reduction in subgroup formation.