Intrusive thoughts better predict the strengths of people’s imagined experiences than semantic priming from imagery.

Full description

Data includes all raw information from the online study. Additional spreadsheets sort the data into trials and conditions after exclusions. The excel sheets include: demographics, response times, error-rates, intrusion rates, questionnaire scores. The sheets include averages of trials over conditions. Twenty-three participants failed to follow one or both of the attention-check instructions and were consequently excluded from analyses. We conducted a three-way repeated measures ANOVA to examine the impact of Modality, Imagination, and Priming on word completion response times. Analyses were corrected for multiple comparisons, so significance was evaluated against a Bonferroni adjusted alpha of p < .008. Distributions of CAIS and VVIQ scores violated assumptions of normality. Therefore, all the following correlational analyses were calculated using Spearman’s rho. Results were evaluated against a Bonferroni adjusted alpha of p < .008. Individual datapoints were excluded from analyses based on Mahalanobis distance calculations, using a significance level of .9. We conducted a second three-way repeated measures ANOVA to examine the impact of Modality, Imagination, and Priming on word completion error rates. To quantify how imagination modulates semantic priming, we first calculated individual priming effects: the difference in error rates and response times when words were positively primed, as opposed to negatively primed. This was done separately for Imagine and for Suppression trials. We then subtracted people’s average priming effects on Suppression trials from their average priming effects on Imagine trials. These estimates of how imagination instructions had modulated priming were calculated separately for cued Audio and for cued Visual trials. There was no evidence for a relationship between CAIS scores and audio imagination modulation of semantic priming for word completion error rates. There was also no evidence for an inter-relationship between CAIS scores and audio imagination modulations of semantic priming of response times. By contrast, there was some evidence for a weak inter-relationship between CAIS scores and Audio Intrusion rates. In relation to visualisations, there was no evidence for an inter-relationship between people’s VVIQ scores and visual imagination modulation of the semantic priming of word completion errors-rates. Similarly, there was no evidence for an inter-relationship between VVIQ scores and visual imagination modulation of semantic priming of word completion response times. By contrast, there was very strong evidence for a weak (i.e. rho < 0.3) positive inter-relationship between people’s VVIQ scores and involuntary Visual Intrusions of cued visual scenarios.