Corrigendum to “The varieties of inner speech: Links between quality of inner speech and psychopathological variables in a sample of young adults” [Consciousness and Cognition 20 (2011) 1586–1593

Abstract

Since the publication of our article, we have noted two corrections necessary to our paper. First, our Method section did not correctly report the Likert scaling used for two of the questionnaires employed, and should have stated that items on both the five-item ‘predisposition to auditory hallucinations’ and four-item ‘predisposition to visual hallucinations and disturbances’ scales were coded on a four-point Likert scale running from ‘‘never’’ (scored 1) to ‘‘almost always’’ (scored 4), meaning scores on these scales could run from 5–20 and 4–16 respectively. Second, due to an error in our spreadsheet formula for summating visual hallucination (VH) scores, analyses involving total VH scores were incorrect. Analyses which utilised individual VH items, rather than their summed total, including the calculation of Cronbach’s alpha for this variable and the exploratory factor analysis involving this variable, were unaffected, as these were based on correct item-level data. Analyses involving total VH scores changed as follows. The revised mean VH score (SD, range) was 6.44 (2.29, 4–14). Correlations between corrected VH scores and the other study variables, originally displayed in Table 4 of our paper, are reported in the revised Table 4 presented here. As partial correlations between auditory hallucination proneness and other study variables included partialling for VH scores, these were also altered, and are included in our revised Table 4. In terms of changes involving previously non-significant correlations now becoming significant, correlations between VH scores and Dialogic Inner Speech, Evaluative/Motivational Inner Speech, auditory hallucination proneness, and depression, became significant. No previously reported significant correlations became non-significant as a result of these changes. In terms of the multiple linear regression predicting auditory hallucination proneness, depression dropped out as a significant predictor of this variable but VH scores became a significant predictor. In terms of the multiple linear regression predicting VH scores, depression now became a significant predictor. The corrected relevant complete text section of our Results (Section 3.5) is reported below.