<![CDATA[This study evaluates user acceptance and motivation in English pronunciation learning across web-based and Augmented Reality (AR) platforms Utilizing a comparative design with a sample of 46 students, a web application (developed with Laravel and Livewire) and an AR application (developed with Unity and Vuforia) were constructed, both integrating the SpeechAce API for real-time pronunciation feedback Usability was measured using the System Usability Scale (SUS), while structural associations between technology acceptance and intrinsic motivation variables were evaluated using Partial Least Squares Structural Equation Modeling (PLS-SEM). Both systems demonstrated excellent usability, yielding average SUS scores of 93.33 for the Web platform and 83.54 for the AR platform. Due to initial instrument instability and construct invalidity in the original measurement model, post-hoc model reconstruction was performed, rendering this structural analysis exploratory in nature. In the reconstructed model, Intrinsic Motivation (IM) exhibited a strong, positive structural association with Perceived Ease of Use (PEOU) across both platforms. However, the association between IM and Perceived Usefulness (PU) was significant only on the AR platform. Conversely, the association between PEOU and PU was significant only on the Web platform. On both platforms, Attitude Toward Using (ATU) was primarily driven by PEOU rather than PU, which had no direct significant association with ATU. These exploratory findings suggest that integrating immersive features in AR can align learners’ subjective enjoyment with their perception of functional utility. This highlights the potential of learner-centered design to enhance engagement in digital pronunciation tools.]]>
