The Influence of Human–AI Relational Attachment on Emotional Regulation Among Adults in Pakistan

Abstract

The study aimed to determine the relationship between relational attachment to artificial intelligence (AI) system and emotional regulation in adult population of Pakistan. There is an increasing body of literature showing that users can develop an emotional relationship with chatbots, virtual assistants and conversational agents, but most of this is conducted in western samples and the psychological implications of such relationships in collectivist and fast digitizing societies are not explored. Based on Bowlby's attachment theory and Gross's process model of emotion regulation, the hypothesis of the study was that there is a significant correlation between human–AI relational attachment and emotional regulation among Pakistani adults. A quantitative correlational design was used. The subjects were 120 adults aged 18 years and older, who were obtained by non-probability convenience sampling and completed an online survey consisting of a Demographic Information Form, the AI Attachment Scale (Kasturiratna & Hartanto, 2025) and the Emotion Regulation Questionnaire (Gross & John, 2003). The collected data was analyzed through descriptive Pearson product-moment correlation and simple linear regression using SPSS 22 program. The results show that when an adult feels some level of emotional connection with the AI systems, there is also a higher occurrence of adulthood-performing emotion regulation measures habitually, though nominally. It seems that AI is more of a supporting resource and not a substitute source of emotional support. The study provides culturally placed evidence to the literature on human–AI interaction, along with implications for developers, educators, policymakers, and mental health practitioners in the Pakistani context.