Interactions between the neural correlates of dispositional internally directed thought and visual imagery

Abstract

Cognition is not always directed to the events in the here and now and we often self-generate thoughts and images in imagination. Important aspects of these self-generated experiences are associated with various dispositional traits. In this study, we explored whether these psychological associations relate to a common underlying neurocognitive mechanism. We acquired resting state functional magnetic resonance imaging data from a large cohort of participants and asked them to retrospectively report their experience during the scan. Participants also completed questionnaires reflecting a range of dispositional traits. We found thoughts emphasizing visual imagery at rest were associated with dispositional tendency towards internally directed attention (self-consciousness and attentional problems) and linked to a stronger correlation between a posterior parietal network and a lateral fronto-temporal network. Furthermore, decoupling between the brainstem and a lateral visual network was associated with dispositional internally directed attention. Critically, these brain-cognition associations were related: the correlation between parietal-frontal regions and reports of visual imagery was stronger for individuals with increased connectivity between brainstem and visual cortex. Our results highlight neural mechanisms linked to the dispositional basis for patterns of self-generated thought, and suggest that accounting for dispositional traits is important when exploring the neural substrates of self-generated experience (and vice versa). This article is part of the theme issue 'Offline perception: voluntary and spontaneous perceptual experiences without matching external stimulation'.

Keywords: dispositional traits; self-generated experience; visual imagery.

Figure 1.

Behavioural variables and their relationship. (a) Test–retest reliability of responses to items at the end of two resting state fMRI scans for 40 participants. Intraclass correlation coefficient (ICC) visualized using a word cloud. Font size represents the absolute ICC value and font colour its sign (red for positive and blue for negative values). *p < 0.05, **p = 0.001. (b) Component weights from a principal component analysis on the participants’ responses to dispositional measures of physical and mental health. (c) Heat map showing the standardized canonical weights of each item for the significant canonical mode 1. The annotations indicate the item variance explained (%) by the canonical mode.

Figure 2.

Static functional connectivity associated with behaviour. (a) Left. The lateral (LAT) fronto-temporal and precuneus networks. The edge highlights an increase in their partial correlation for people thinking more in images. Right. Scatter plot of thinking in images scores with correlation values between these two networks for each participant. (b) Left. The lateral visual and brainstem networks. The edge highlights a decrease in their partial correlation for higher internal attention scores (PCA component 2). Right. Scatter plot of component 2 scores from the PCA of the physical and mental health questionnaires with correlation values between these two networks for each participant. (Online version in colour.)

Figure 3.

Moderation effect of the lateral (LAT) visual–brainstem coupling on the relationship between thinking in images and lateral fronto-temporal–precuneus functional connectivity. The scatter plot shows the correlation values between the lateral fronto-temporal and precuneus networks and thinking in images scores grouped, by a median split, into higher and lower connectivity groups between the lateral visual and brainstem networks. (Online version in colour.)