The noisy input hypothesis of schizophrenia
Gutherie, Audrey Hix
MetadataShow full item record
Introduction: Differences in the level of scotopic intrinsic noise were examined in a group of 15 schizophrenia participants and 15 matched controls in order to test the Noisy Input Hypothesis of Schizophrenia, which states that many of the higher-order disturbances that characterize the disorder may be partly due to increased intrinsic noise at the level of sensory input. In addition to intrinsic noise, a number of other variables were measured including scotopic sensitivity (SS), macular pigment optical density (MPOD), critical flicker fusion frequency (CFF), and lens optical density (OD). With the exception of CFF, none of these variables have been assessed in schizophrenia but, based on past evidence, might be expected to differ for this group. Method: Intrinsic noise was assessed by measuring variability in scotopic thresholds. Thresholds were measured using the method of constant stimuli and a two-alternative forced choice procedure. The stimuli consisted of three 1.85-degree diameter test stimuli (at 410, 510, and 565 nm) presented at 10-degrees in the periphery. Lens OD was derived by subtracting the 410 nm log relative scotopic sensitivity (LRSS) values from the 565 nm values. MPOD was measured with a 1-deg test stimulus using heterochromatic flicker photometry. CFF values were obtained with a 1-deg, 570 nm stimulus using the method of limits. Results: The schizophrenia participants did not differ from the control participants with respect to intrinsic noise, LRSS, or lens OD. In contrast, average CFF thresholds (p < 0.004) were significantly lower in the schizophrenia group. The MP density of the schizophrenia participants was 24% lower than the controls and marginally significant (p < 0.08). There were no significant associations between the dependent variables except for MPOD and LRSS. Discussion: The main result of this study, similar intrinsic noise levels in schizophrenia and control participants, does not support the Noisy Input Hypothesis. In fact, the schizophrenia participants were most similar on those items which might be expected to be most linked to the hypothesis (lens OD, LRSS, and intrinsic noise). The largest difference between the groups was in CFF, which is widely considered to be determined by processing at the level of the visual cortex. One interpretation of these results is that many of the visual disturbances reported in schizophrenia are mediated by post-receptoral mechanisms. More data (e.g., larger samples using non-medicated schizophrenia participants), particularly photopic, however, are needed to confirm this interpretation.