Background: Hearing impairment (HI) in the elderly may be a risk factor for psychosis, but associations between HI and psychotic disorder or psychotic experiences have been reported more consistently in younger than in older populations. The aims of this study were to replicate the positive association between hearing impairment and psychotic experiences and to clarify any differences between groups of young and old individuals in a non-clinical, normal aging general population sample. Methods: HI, assessed at baseline and at 3-year follow-up, and psychotic experiences, assessed at. 3-year follow-up, were analysed in a group of 848 individuals aged 33 to 89 years between 1999 and 2004. HI was determined on the basis of both self-report and audiometric examination. The "psychoticism" and "paranoid ideation" subscales from the SCL-90-R were used to assess level of psychotic experiences. Results: Self-reported hearing problems expressed as conversational HI (beta=0.080, 95% CI: 0.23, 7.90, p=0.038) and subjective HI (beta=0.087, 95%CI: 0.70, 10.30,p=0.025), but not audiometric objective HI, were associated with psychotic experiences. In those with hearing aids, associations with psychotic experiences were only present if accompanied by self-reported hearing problems that persisted in spite of the hearing aid. In addition, HI increased the risk for psychotic experiences specifically in younger rather than older individuals. Conclusions: Self-reported hearing problems rather than audiometric or remediated hearing loss may contribute to the development of psychotic experiences in younger rather than in older individuals.