This study examines how information-seeking behaviors and motivations differ between problematic and regular users of Xiaohongshu (RedNote), one of the fastest-growing social media platforms in China. Problematic Social Media Use (PSMU) has been linked to poor mental health outcomes, including anxiety, depression, and stress, but little is known about how patterns of information-seeking contribute to PSMU. By focusing on RedNote, this project aimed to address this gap and explore how information engagement may shape problematic use. An online survey was completed by 168 frequent RedNote users, who were classified as problematic or regular users using the Bergen Social Media Addiction Scale. To capture RedNote-specific patterns, existing scales of information-seeking behavior and motivation were adapted. Statistical analyses compared group differences in five types of information-seeking behaviors (e.g., social searching, social browsing, hedonic proclivity) and three motivational domains (instrumental, hedonic, cognitive). Results indicated that problematic users engaged significantly more in social searching, hedonic proclivity, and general erudition than regular users, but showed no differences in social browsing or consumer trend seeking. Motivationally, problematic users reported higher instrumental and cognitive motives, suggesting they turn to RedNote for practical problem-solving and cognitively salient content. Interestingly, no group difference was found in hedonic motives, implying that pleasure-driven engagement may be universal across user types. These findings suggest that problematic social media use is fueled less by passive browsing and more by targeted, intentional information pursuits that may reinforce addictive cycles, such as seeking validation, reassurance, or knowledge tied to frequent concerns, contrary to prior expectations. The study contributes to understanding the mechanisms underlying PSMU and highlights the importance of platform-specific contexts. Future research should refine measurement tools, recruit larger and more diverse samples, and employ longitudinal designs to examine causality.