نوع مقاله : مقاله پژوهشی
نویسنده
استادیار، گروه آموزش علوم تربیتی، دانشگاه فرهنگیان، تهران، ایران، ایمیل: m.dadashzadeh@cfu.ac.ir
چکیده
کلیدواژهها
موضوعات
عنوان مقاله [English]
نویسنده [English]
Objective: This study aimed to explore the dimensions and components of the algorithmic halo effect on Iranian university students' experiences during interaction with AI-based teaching assistants, and to clarify its implications for information credibility and quality evaluation, information-seeking behavior, and AI literacy.
Methods: A qualitative methodology employing a phenomenologically informed thematic analysis design was adopted. The target population consisted of students at Farhangian University of Tabriz. Utilizing purposive-judgmental and convenience sampling strategies, participant selection continued until theoretical saturation was reached (N=16). Data were collected through in-depth, semi-structured interviews. Data analysis was conducted following Braun and Clarke’s (2006) six-phase thematic analysis framework using MAXQDA 2024 software. To ensure trustworthiness and rigor, strategies including member checking, triangulation, and peer debriefing were implemented.
Results: Qualitative analysis revealed six primary components: Visual Appeal as Credibility, The Hidden Power of Tone and Diction, Tactical Surrender, The Enigma of Machine Omniscience, The Shift in Judgment from Professor to Algorithm, and Feedback Loops and Error Resilience. Findings indicated that the algorithmic halo effect in students' interactions with AI teaching assistants is not merely confined to technology acceptance or cognitive bias; rather, it significantly influences how the credibility, quality, and trustworthiness of generated information are perceived, evaluated, and validated, as well as the patterns of searching for and using information resources. In the study's conceptual model, the core of the halo is defined by two dimensions- Visual Appeal as Credibility and The Hidden Power of Tone and Diction- which form the initial layer of trust upon encountering the platform. This core develops within a broader socio-cultural context termed The Shift in Judgment from Professor to Algorithm. Alongside this shift, two psychological-situational conditions -Tactical Surrender and The Enigma of Machine Omniscience- reflect the paradoxical experience of feeling recognized by the system. Finally, Feedback Loops and Error Resilience serves as the dynamic engine driving the model.
Conclusion: The algorithmic halo effect can be conceptualized as a phenomenon within human-information interaction that influences users' capacity for critical judgment by altering perceptions of credibility, information quality assessment, and information-seeking behavior.
کلیدواژهها [English]
)
ارسال نظر در مورد این مقاله