BackgroundThe etiology of many chronic pain conditions is still poorly understood. According to the perception-filter model, amplification (stage I.) and insufficient filtering (stage II.) of afferent bodily signals results in the increased perception of physical sensations (stage III.).MethodsFocusing on cardiac interoception, we tested if these hypothesized alterations in stages I.-III. apply to a sample of 76 in-patients with chronic pain (Mage = 50.7, 53 females). Heart rate (HR) and heart rate variability (HRV) were assessed as indicators of afferent bodily signal strength (stage I.). The difference between heartbeat-evoked potentials (HEPs) during a resting phase and a heartbeat counting task (HCT), i.e., ΔHEP, was used as an indicator of filter system activity. Interoceptive accuracy (IAcc) in the HCT and a heartbeat discrimination task (HDT) served as indicators of perception of afferent bodily signals.ResultsHR did not predict either indicator of IAcc. LF HRV was a positive predictor of IAccHCT and HF HRV positively predicted both IAccHCT and IAccHDT. HEP amplitudes did not differ between the HCT and the resting period and ΔHEP did not moderate any of the relationships between indicators of bodily signal strength (stage I.) and IAcc (stage III.). Neither indicator of IAcc (stage lll.) was significantly associated with measures of symptom report.ConclusionAlterations in the processing of bodily signals as suggested by the perception-filter model do not apply to chronic pain conditions. Future studies should focus on investigating pain-specific processing of bodily signals.