Rising interest in artificial intelligence in education reinforces the demand for evidence-based implementation. This study investigates how tutor agents’ physical embodiment and anthropomorphism (student-reported sociability, animacy, agency, and disturbance) relate to affective (on-task enjoyment) and cognitive (task performance) learning within an intelligent tutoring system (ITS). Data from 56 students (M = 17.75 years, SD = 2.63 years; 30.4% female), working with an emotionally-adaptive version of the ITS “Betty’s Brain”, were analyzed. The ITS’ agents were either depicted as on-screen robots (condition A) or as both on-screen avatars and physical robots (condition B). Physical presence of the tutor agent was not significantly related to task performance or anthropomorphism, but to higher initial on-task enjoyment. Student-reported disturbance was negatively related to initial on-task enjoyment, and student-reported sociability was negatively related to task performance. While physical robots may increase initial on-task enjoyment, students’ perception of certain characteristics may hinder learning, providing implications for designing social robots for education.
12 August 2013 in Folia Phoniatrica et Logopaedica
The aim of this article is to explicate the uniqueness of the motor activity implied in spoken language production and to emphasize how important it is, from a theoretical and a clinical perspective, to consider the motor events associated with speaking as domain-specific, i.e., as pertaining to the domain of linguistic expression. First, phylogenetic data are reviewed demonstrating the specificity of the human vocal tract motor network regarding (i) the entrenchment of laryngeal motor skills within the organization of vocal tract movements, (ii) the evolution of a neural basis for skill acquisition within this system, and (iii) the integration of this system into an auditory-motor network. Second, ontogenetic evidence and existing knowledge about the experience-dependent plasticity of the brain are reported to explicate that during speech acquisition the vocal tract motor system is constrained by universal properties of speech production and by the specific phonological properties of the speaker's ambient language. Third, clinical data from dysarthria and apraxia of speech provide the background for a discussion about the theoretical underpinnings of domain-general versus domain-specific views of speech motor control. The article ends with a brief sketch of a holistic neurophonetic approach in experimental inquiries, assessment, and treatment of neuromotor speech impairment.
As educational environments become more diverse, adaptive technologies like social robots hold promise for providing individual support to learners. This study investigated the role of adaptive teaching of a robot on students’ learning outcomes, emotions, and self-regulated learning (SRL). A total of 120 participants (aged 18–60 years, Mage = 30.25, SDage = 10.06, 64.3% female) engaged in an interactive vocabulary learning task with varying guidance levels by the robot. Control conditions included fixed guidance, offering either simple guidance with no hints (condition 1) or enhanced guidance, constantly adding hints (condition 2). In the adaptive conditions, the robot adjusted the number of hints based on learners’ recent performance and enjoyment (condition 3) or additionally personalized hints to specific mistakes (condition 4). The results showed no direct or indirect effects of adaptive guidance on task performance or cognitive learning compared to fixed guidance. Instead, adaptive guidance significantly reduced on-task enjoyment compared to enhanced guidance, suggesting that unexpected variability in robot behavior may negatively affect emotional learning experience. However, personalized adaptive guidance increased certain SRL behaviors, indirectly benefiting learning outcomes. These findings highlight the need for further refinement of adaptive social robots to better meet individual learner needs and optimize outcomes.