The aim of the present study was to assess the contribution of somatosensory amplification to somatization. Two questions were addressed: (1) whether somatization is associated with somatosensory amplification independent of level of anxiety or depression, and (2) whether any differences in somatization between patients with panic disorder or depressive disorder and controls can be attributed to differences in somatosensory amplification. The subjects were 673 consecutive psychiatric outpatients. In this sample, 124 patients received a V-code or no psychiatric diagnosis (controls), 82 a single diagnosis of panic disorder, and 68 single diagnosis of major depression. The following measures were used: the Somatosensory Amplification Scale (SSAS), Symptom Checklist-90 (SCL-90), and Hospital Anxiety and Depression Scale (HADS). SCL-90 somatization scores were significantly associated with anxiety scores and to a lesser extent with SSAS scores for somatosensory amplification, independent of gender, presence of physical illness, and depression scores. Moreover, after statistically controlling for differences in somatosensory amplification, panic disorder patients still obtained significantly higher somatization scores than depressive disorder patients, who manifested significantly higher levels of somatization than controls. It is concluded that somatization primarily reflects somatized anxiety and not somatosensory amplification. Somatization associated with panic attacks may result from other mechanisms, such as the psychophysiology of anxiety and cognitive misappraisal of benign bodily symptoms. Further studies into the concurrent validity of the SSAS as a measure of interoceptive acuity are needed.