Background: The SMHS 2016 revealed that young adults in Singapore had the highest 12-month prevalence of mental disorders, with depression being the most prevalent condition. Additionally, the study found that those with higher education were less likely to seek treatment. The recognition of mental illness and knowledge of where to seek help has been found to influence one's ability to seek timely psychological help. This study thus aims to evaluate the effects of ARTEMIS, an education and contact intervention on university students' recognition of depression and help-seeking preference.Methods: A total of 390 university students were recruited over a period of 6-months (October 2018 to April 2019). Students had to attend a one-off intervention which comprised a lecture on depression and personal contact with a person with lived experience of mental illness. Recognition of depression and help-seeking preference were assessed using a vignette approach, at pre- and post-intervention as well as at 3-month follow-up.Results: The intervention was effective at improving student's recognition of depression and this effect was sustained at 3-months follow-up. The intervention was also effective in shifting student's help-seeking preference, although the effects were not sustained at 3-month follow-up. Having a close friend or family with mental illness was associated with better recognition, and being able to correctly recognize depression was linked to a preference to seek psychiatric over non-psychiatric help.Conclusion: This study elucidated the efficacy of a knowledge-contact-based intervention in improving university students' recognition of depression and help-seeking preference. However, while the benefits on recognition of depression is more enduring, it is more transient for help-seeking beliefs, and booster sessions may be needed to improve the long-term effectiveness of the intervention on help-seeking preference. Lastly, to investigate the generalizability of this study's findings, future studies could replicate the current one across other non-self-selected samples, such as by integrating the intervention as part of student's orientation.