Davide De Tommaso
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Davide De Tommaso has a total of 52 co-authors. Here are the top co-authors by number of shared publications:
Agnieszka Wykowska
42 shared publications- Davide Ghiglino
Social Cognition in Human-Robot Interaction, Istituto Italiano di Tecnologia, Genova, Italy
13 shared publications - Agnieszka Wykowska
Italian Institute of Technology, Genoa, Italy
12 shared publications - Serena Marchesi
Italian Institute of Technology, Genoa, Italy
12 shared publications Francesca Ciardo
10 shared publicationsKyveli Kompatsiari
10 shared publications
Latest Publications
- Open Access
- 7 April 2026
- Published by Springer Nature in International Journal of Social Robotics
- Vol. 18 (4) , 56
The Influence of Robot Autonomy on Sense of Control and Trust Towards a Robot
Autonomous robots can make decisions that lead to outcomes that are better, equal to, or worse than those requested by a human. However, how the value of these outcomes shapes perceptions of robot autonomy remains underexplored. We conducted four online experiments examining how different forms of robot decision-making autonomy influence human–robot interaction. Participants played a game in which they selected a box for a robot avatar to collect, with each box yielding a specific number of points. Some robots were described as autonomous and could disobey instructions. We compared robots that disobeyed to produce a better outcome for the participant (helpful autonomy) with robots that disobeyed but achieved the same outcome (same-outcome autonomy). Experiments 3 and 4 further distinguished helpful autonomy aimed at avoiding losses from autonomy aimed at obtaining additional gains. Trust was assessed using a post-task trust game. Results showed that helpful autonomy generally increased trust compared to same-outcome autonomy, except when loss avoidance was unreliable. Trust was unrelated to perceived control. Perceived control was higher for same-outcome autonomy in Experiments 1–2, where non-autonomous and random robots were present, but higher for helpful autonomy in Experiments 3–4, indicating that perceived control depends on both robot autonomy and the behavioral context in which it is experienced. Finally, helpful autonomy produced bimodal distributions in ratings of robot cooperativeness and experienced sense of joint agency, suggesting that such behavior can elicit qualitatively different attitudes toward the robot.
- 7 April 2026
- Published by Springer Nature in International Journal of Social Robotics
- Vol. 18 (4) , 56
Open AccessThe Influence of Robot Autonomy on Sense of Control and Trust Towards a Robot
Autonomous robots can make decisions that lead to outcomes that are better, equal to, or worse than those requested by a human. However, how the value of these outcomes shapes perceptions of robot autonomy remains underexplored. We conducted four online experiments examining how different forms of robot decision-making autonomy influence human–robot interaction. Participants played a game in which they selected a box for a robot avatar to collect, with each box yielding a specific number of points. Some robots were described as autonomous and could disobey instructions. We compared robots that disobeyed to produce a better outcome for the participant (helpful autonomy) with robots that disobeyed but achieved the same outcome (same-outcome autonomy). Experiments 3 and 4 further distinguished helpful autonomy aimed at avoiding losses from autonomy aimed at obtaining additional gains. Trust was assessed using a post-task trust game. Results showed that helpful autonomy generally increased trust compared to same-outcome autonomy, except when loss avoidance was unreliable. Trust was unrelated to perceived control. Perceived control was higher for same-outcome autonomy in Experiments 1–2, where non-autonomous and random robots were present, but higher for helpful autonomy in Experiments 3–4, indicating that perceived control depends on both robot autonomy and the behavioral context in which it is experienced. Finally, helpful autonomy produced bimodal distributions in ratings of robot cooperativeness and experienced sense of joint agency, suggesting that such behavior can elicit qualitatively different attitudes toward the robot.
- Open Access
- 1 January 2026
- Published by Wiley in Human Behavior and Emerging Technologies
- Vol. 2026 (1)
Theta Frequency Band in the EEG Signal as Neurocognitive Indicator of Cognitive Load for Collaborative Tasks Involving Robot Teleoperation
The aim of the present study was to identify metrics of cognitive effort during robot teleoperation in a social context of joint collaborative task performed with another human user. Robotics teleoperation enables control of artificial systems in remote environments but is often cognitively challenging. Measuring human cognitive effort during teleoperation has largely relied on subjective measures like questionnaires. To evaluate the quality of teleoperated systems, it is critical to investigate objective measures of cognitive effort directly in natural contexts, especially during social interactions with individuals encountering the teleoperated robot. In two separate experiments, we recorded—with dual EEG—brain activity of participants performing a cooperative task. In the first experiment, participants were asked to perform a cooperative task in a direct human–human interaction setup. In the second experiment, the interaction was mediated by the iCub robot, teleoperated by one of the participants using a VR headset (human–robot–human interaction). The EEG data revealed an increase of power in the theta (4–7 Hz) frequency band in the robot‐mediated interaction, relative to human–human interaction, likely reflecting the additional cognitive load required by robot teleoperation. This demonstrates that using EEG in teleoperation might provide a reliable metric of the quality of the system and a way to tap onto cognitive mechanisms underlying the user′s experience in teleoperation.
- 1 January 2026
- Published by Wiley in Human Behavior and Emerging Technologies
- Vol. 2026 (1)
Open AccessTheta Frequency Band in the EEG Signal as Neurocognitive Indicator of Cognitive Load for Collaborative Tasks Involving Robot Teleoperation
The aim of the present study was to identify metrics of cognitive effort during robot teleoperation in a social context of joint collaborative task performed with another human user. Robotics teleoperation enables control of artificial systems in remote environments but is often cognitively challenging. Measuring human cognitive effort during teleoperation has largely relied on subjective measures like questionnaires. To evaluate the quality of teleoperated systems, it is critical to investigate objective measures of cognitive effort directly in natural contexts, especially during social interactions with individuals encountering the teleoperated robot. In two separate experiments, we recorded—with dual EEG—brain activity of participants performing a cooperative task. In the first experiment, participants were asked to perform a cooperative task in a direct human–human interaction setup. In the second experiment, the interaction was mediated by the iCub robot, teleoperated by one of the participants using a VR headset (human–robot–human interaction). The EEG data revealed an increase of power in the theta (4–7 Hz) frequency band in the robot‐mediated interaction, relative to human–human interaction, likely reflecting the additional cognitive load required by robot teleoperation. This demonstrates that using EEG in teleoperation might provide a reliable metric of the quality of the system and a way to tap onto cognitive mechanisms underlying the user′s experience in teleoperation.
- Open Access
- 16 May 2025
- Published by Center for Open Science in OSF Preprints
Evaluating human perception of value-laden decisions made by a humanoid robot
This study investigated human responses to value-laden decisions made by a humanoid robot, focusing on the values of privacy, freedom, and norms. The primary objective was to develop and validate an implicit method for assessing the importance of these values in a domestic context. Participants interacted with a humanoid robot iCub, which verbally presented to them 30 scenarios, each involving actions that varied in the degree value violation. The results revealed that participants were most likely to tolerate robot behaviours that infringed on freedom and least likely to agree to behaviours that violated social norms, with privacy-related violations falling in between. Additionally, response times increased as the severity of value violation increased, suggesting greater cognitive effort when participants faced more severe ethical dilemmas. The analysis of participants’ open text responses explaining rejection of certain robot behaviours highlighted concerns over autonomy, data protection, and adherence to household norms. The study that involved the humanoid robot was then replicated with a human agent to explore whether these patterns were general, or only specific to robot behaviours. Results mirrored the robot condition: participants again showed the greatest tolerance for violations of freedom and the least for social norm violations, with privacy occupying a middle ground. The most important result from our study is that people tend to prioritise social norms over other principles. Such prioritisation of values needs to be taken into account in the design of value-aware AI systems. These findings contribute to a deeper understanding of how people evaluate value-aware behaviour in social robots and introduce an implicit, ecologically valid method for studying human values beyond traditional questionnaire-based approaches.
- 16 May 2025
- Published by Center for Open Science in OSF Preprints
Open AccessEvaluating human perception of value-laden decisions made by a humanoid robot
This study investigated human responses to value-laden decisions made by a humanoid robot, focusing on the values of privacy, freedom, and norms. The primary objective was to develop and validate an implicit method for assessing the importance of these values in a domestic context. Participants interacted with a humanoid robot iCub, which verbally presented to them 30 scenarios, each involving actions that varied in the degree value violation. The results revealed that participants were most likely to tolerate robot behaviours that infringed on freedom and least likely to agree to behaviours that violated social norms, with privacy-related violations falling in between. Additionally, response times increased as the severity of value violation increased, suggesting greater cognitive effort when participants faced more severe ethical dilemmas. The analysis of participants’ open text responses explaining rejection of certain robot behaviours highlighted concerns over autonomy, data protection, and adherence to household norms. The study that involved the humanoid robot was then replicated with a human agent to explore whether these patterns were general, or only specific to robot behaviours. Results mirrored the robot condition: participants again showed the greatest tolerance for violations of freedom and the least for social norm violations, with privacy occupying a middle ground. The most important result from our study is that people tend to prioritise social norms over other principles. Such prioritisation of values needs to be taken into account in the design of value-aware AI systems. These findings contribute to a deeper understanding of how people evaluate value-aware behaviour in social robots and introduce an implicit, ecologically valid method for studying human values beyond traditional questionnaire-based approaches.
- Open Access
- 10 March 2025
- Published by Center for Open Science in PsyArXiv
Adopting the intentional stance interferes with social attention when interacting with a social robot
Joint attention is a pivotal mechanism of human social cognition. Traditionally, joint attention has been experimentally investigated by means of the gaze cueing task in screen-based settings. Recently, the gaze cueing task has been implemented in more naturalistic and interactive settings with the embodied iCub robot, allowing for the investigation of concurrent factors influencing joint attention. Previous studies found that a communicative gaze before the gaze cue (i.e., mutual or avoiding gaze with the participants) influenced the subsequent joint attention by eliciting the gaze-cueing effect only in the mutual gaze condition. In the present work, we first manipulated the attribution of the intentional stance toward the robot by letting participants share a familiar context. Subsequent to this shared experience, we integrated the gaze-cueing paradigm with the iCub robot, such that the robot cued participants’ attention after establishing either mutual or avoiding gaze with them. In contrast to previous studies where the adoption of the intentional stance was not manipulated, our current results show that when the robot was perceived as an intentional agent, a gaze-cueing effect was present for both mutual and avoiding gaze, thereby suggesting that participants might have interpreted both gaze conditions as intentional (and, thus as communicative signals).
- 10 March 2025
- Published by Center for Open Science in PsyArXiv
Open AccessAdopting the intentional stance interferes with social attention when interacting with a social robot
Joint attention is a pivotal mechanism of human social cognition. Traditionally, joint attention has been experimentally investigated by means of the gaze cueing task in screen-based settings. Recently, the gaze cueing task has been implemented in more naturalistic and interactive settings with the embodied iCub robot, allowing for the investigation of concurrent factors influencing joint attention. Previous studies found that a communicative gaze before the gaze cue (i.e., mutual or avoiding gaze with the participants) influenced the subsequent joint attention by eliciting the gaze-cueing effect only in the mutual gaze condition. In the present work, we first manipulated the attribution of the intentional stance toward the robot by letting participants share a familiar context. Subsequent to this shared experience, we integrated the gaze-cueing paradigm with the iCub robot, such that the robot cued participants’ attention after establishing either mutual or avoiding gaze with them. In contrast to previous studies where the adoption of the intentional stance was not manipulated, our current results show that when the robot was perceived as an intentional agent, a gaze-cueing effect was present for both mutual and avoiding gaze, thereby suggesting that participants might have interpreted both gaze conditions as intentional (and, thus as communicative signals).
- Open Access
- 9 March 2025
- Published by Elsevier in Consciousness and Cognition
- Vol. 130, 103835
Does perceived voluntariness of others’ actions induce vicarious sense of agency? Evidence from human-robot interaction
- 9 March 2025
- Published by Elsevier in Consciousness and Cognition
- Vol. 130, 103835
Open AccessDoes perceived voluntariness of others’ actions induce vicarious sense of agency? Evidence from human-robot interaction
- Open Access
- 6 November 2024
- Published by Center for Open Science in OSF Preprints
The influence of autonomy of a teleoperated robot on user's objective and subjective performance
This article describes a study investigating the effects of decision-making autonomy of a robot, which is teleoperated in a simulated unstable or dangerous environment. It specifically focuses on robot’s autonomy to disregard the user’s command if the robot finds an alternative method of achieving the same goal as pursued by the user, but with significantly reduced risk of failure. Such autonomous control module might prove especially useful under circumstances where human operators cannot access or process all available information quickly enough to make the most optimal decision. We conducted an experiment in which subjects participated in a task of teleoperating either a robot that possesses such autonomous cognitive module, or not. We found that such module significantly reduced their sense of agency over the robot as well as the sense of doing the task together with the robot (sense of joint agency). Most interestingly, it reduced their subjective ratings of performance in the task, when in fact their actual performance was better with such an autonomous robot. A further analysis revealed that this counterintuitive finding was due to an effect of bias: in our study loss of control associated with operating an autonomous robot on average lowered the reported subjective performance by 11% points, with other factors staying equal. These results suggest that this type of assistive autonomy can be beneficial for performance, but might lead to unwanted effects that need to be overcome in order for such system to prove useful in practical applications.
- 6 November 2024
- Published by Center for Open Science in OSF Preprints
Open AccessThe influence of autonomy of a teleoperated robot on user's objective and subjective performance
This article describes a study investigating the effects of decision-making autonomy of a robot, which is teleoperated in a simulated unstable or dangerous environment. It specifically focuses on robot’s autonomy to disregard the user’s command if the robot finds an alternative method of achieving the same goal as pursued by the user, but with significantly reduced risk of failure. Such autonomous control module might prove especially useful under circumstances where human operators cannot access or process all available information quickly enough to make the most optimal decision. We conducted an experiment in which subjects participated in a task of teleoperating either a robot that possesses such autonomous cognitive module, or not. We found that such module significantly reduced their sense of agency over the robot as well as the sense of doing the task together with the robot (sense of joint agency). Most interestingly, it reduced their subjective ratings of performance in the task, when in fact their actual performance was better with such an autonomous robot. A further analysis revealed that this counterintuitive finding was due to an effect of bias: in our study loss of control associated with operating an autonomous robot on average lowered the reported subjective performance by 11% points, with other factors staying equal. These results suggest that this type of assistive autonomy can be beneficial for performance, but might lead to unwanted effects that need to be overcome in order for such system to prove useful in practical applications.
- Open Access
- 23 October 2024
- Published by Center for Open Science in OSF Preprints
Atypical eye gazing in autistic children playing with a humanoid: can we use technologies to boost diagnostic precision?
This study explores using humanoid robots to improve the diagnosis and intervention of Autism Spectrum Disorder (ASD) through objective measures, gaze pattern analysis specifically. Traditional ASD diagnostic tools rely on subjective assessments, while our approach uses the iCub robot to provide quantitative metrics. In our proof-of-concept study, the robot engaged children in interactive games to assess mutual gaze. Results showed that neurotypical children maintained more eye contact compared to children with ASD. This highlights the potential of humanoid robots to offer objective, standardized measures that can improve the accuracy of ASD diagnoses and interventions.
- 23 October 2024
- Published by Center for Open Science in OSF Preprints
Open AccessAtypical eye gazing in autistic children playing with a humanoid: can we use technologies to boost diagnostic precision?
This study explores using humanoid robots to improve the diagnosis and intervention of Autism Spectrum Disorder (ASD) through objective measures, gaze pattern analysis specifically. Traditional ASD diagnostic tools rely on subjective assessments, while our approach uses the iCub robot to provide quantitative metrics. In our proof-of-concept study, the robot engaged children in interactive games to assess mutual gaze. Results showed that neurotypical children maintained more eye contact compared to children with ASD. This highlights the potential of humanoid robots to offer objective, standardized measures that can improve the accuracy of ASD diagnoses and interventions.
- Open Access
- 23 October 2024
- Published by Center for Open Science in OSF Preprints
Simulating Social Cognition: A Humanoid-Based Approach to Improve Theory of Mind Skills in Autism
Autism Spectrum Disorders (ASD) often involve challenges in social communication, and various therapies are used to address these. This randomized, two-period crossover trial assessed the effectiveness of integrating Applied Behavior Analysis (ABA) with robot-assisted therapy to improve Theory of Mind (ToM) skills in children with ASD. Thirty-two children (mean age 7.53 ± 1.32 years, 7 females) participated. Group 1 received robot-assisted therapy first, followed by traditional therapy, while Group 2 had the reverse order. ToM skills were evaluated at three points: baseline (T0), after the first period (T1), and after the second (T2). Evaluators were blinded to the treatment conditions. Significant improvements in ToM were observed with robot-assisted therapy compared to traditional methods. A control group confirmed the robot's superiority over human actors in role-playing scenarios. No participants discontinued the study, highlighting the potential of integrating technology into ASD therapy.
- 23 October 2024
- Published by Center for Open Science in OSF Preprints
Open AccessSimulating Social Cognition: A Humanoid-Based Approach to Improve Theory of Mind Skills in Autism
Autism Spectrum Disorders (ASD) often involve challenges in social communication, and various therapies are used to address these. This randomized, two-period crossover trial assessed the effectiveness of integrating Applied Behavior Analysis (ABA) with robot-assisted therapy to improve Theory of Mind (ToM) skills in children with ASD. Thirty-two children (mean age 7.53 ± 1.32 years, 7 females) participated. Group 1 received robot-assisted therapy first, followed by traditional therapy, while Group 2 had the reverse order. ToM skills were evaluated at three points: baseline (T0), after the first period (T1), and after the second (T2). Evaluators were blinded to the treatment conditions. Significant improvements in ToM were observed with robot-assisted therapy compared to traditional methods. A control group confirmed the robot's superiority over human actors in role-playing scenarios. No participants discontinued the study, highlighting the potential of integrating technology into ASD therapy.
- Open Access
- 21 October 2024
- Published by Center for Open Science in PsyArXiv
Value Projection. Humans perception of a humanoid robot in moral contexts.
As robots become collaborators in various domains, understanding human perception of them as moral and intentional agents ispivotal. This study addresses different strategies to promote humans’ perception of social robots as value-aware collaborators,focusing on the alignment of values between humans and the humanoid robot iCub. Employing three different strategiesof describing the iCub robot abilities of moral-decision making, we found that regardless of the strategy used, participantstended to attribute moral agency to iCub, likely influenced by its humanoid appearance. Interestingly, the absence of priorindication of iCub’s ability to make moral decisions led to a lower adoption of intentional stance. These results suggest anuanced relationship between perceived intentionality and morality in human-robot interactions. Moreover, our findings advanceour understanding of how humans perceive different agents in terms of value-awareness, bridging subjective and objectivemeasures through behavioural and self-report indicators.
- 21 October 2024
- Published by Center for Open Science in PsyArXiv
Open AccessValue Projection. Humans perception of a humanoid robot in moral contexts.
As robots become collaborators in various domains, understanding human perception of them as moral and intentional agents ispivotal. This study addresses different strategies to promote humans’ perception of social robots as value-aware collaborators,focusing on the alignment of values between humans and the humanoid robot iCub. Employing three different strategiesof describing the iCub robot abilities of moral-decision making, we found that regardless of the strategy used, participantstended to attribute moral agency to iCub, likely influenced by its humanoid appearance. Interestingly, the absence of priorindication of iCub’s ability to make moral decisions led to a lower adoption of intentional stance. These results suggest anuanced relationship between perceived intentionality and morality in human-robot interactions. Moreover, our findings advanceour understanding of how humans perceive different agents in terms of value-awareness, bridging subjective and objectivemeasures through behavioural and self-report indicators.
- Open Access
- 22 September 2024
- Published by Center for Open Science in PsyArXiv
We-information can facilitate performance in joint teleoperation over a humanoid robot
In this study, we developed a setup allowing two participants (“operators”) to jointly control a single humanoid robot body. Specifically, each operator controlled one robot arm, using their anatomically congruent hand (left hand controlling a left robot arm and vice versa). In our task participants had to move each robot arm into one of two possible positions (arm raised or lowered). We used this setup to investigate (1) whether presenting prior information about the relationship between movements performed by each participant (“We-information”) can facilitate performance in this task, (2) how joint control over a robot body affects sense of control over the robot and sense of joint agency with the other operator, and (3) how it influences the perceived boundaries between oneself and the others (the robot and the other operator), the so-called “self-other” overlap. We found that (1) “We-information” increased the speed of task performance, but only for simpler configurations, (2) participants experienced high level of sense of control over the robot which increased throughout the task, and (3) a short session of joint control over a humanoid led to pronounced increase in self-other overlap (blurring of boundaries) with both the robot and a co-operator. We discuss implications of our results for understanding of human body representation and how they can inform future applications, such as exoskeletons for individuals affected with hemiplegia.
- 22 September 2024
- Published by Center for Open Science in PsyArXiv
Open AccessWe-information can facilitate performance in joint teleoperation over a humanoid robot
In this study, we developed a setup allowing two participants (“operators”) to jointly control a single humanoid robot body. Specifically, each operator controlled one robot arm, using their anatomically congruent hand (left hand controlling a left robot arm and vice versa). In our task participants had to move each robot arm into one of two possible positions (arm raised or lowered). We used this setup to investigate (1) whether presenting prior information about the relationship between movements performed by each participant (“We-information”) can facilitate performance in this task, (2) how joint control over a robot body affects sense of control over the robot and sense of joint agency with the other operator, and (3) how it influences the perceived boundaries between oneself and the others (the robot and the other operator), the so-called “self-other” overlap. We found that (1) “We-information” increased the speed of task performance, but only for simpler configurations, (2) participants experienced high level of sense of control over the robot which increased throughout the task, and (3) a short session of joint control over a humanoid led to pronounced increase in self-other overlap (blurring of boundaries) with both the robot and a co-operator. We discuss implications of our results for understanding of human body representation and how they can inform future applications, such as exoskeletons for individuals affected with hemiplegia.