Neural substrates of rumination tendency in non-depressed individuals Xiao Chen 陈骁 chenxiao@psych.ac.cn The R-fMRI Lab, Institute of Psychology, Chinese Academy of Science

Outline Introduction Methods Results Discussion

Introduction Rumination “ruminations are repetitive and passive thinking about symptoms of depression and the possible causes and consequences of these symptoms” Susan Nolen-Hoeksema (1959 – 2013)

Introduction

Introduction What am I doing to deserve this? Why can’t I get going? Why do I always react this way?

Introduction Rumination Depression Rumination is an important vulnerability factor in the development of depression Constitute a maladaptive mental habit in depression

Introduction Cognitive mechanisms behind rumination Executive Functions Cognitive inflexibility and difficulties in disengaging attention from irrelevant information

Introduction Rumination? Neural substrates of rumination Emotional Reactivity Rumination? Cognitive control

Introduction Subjects: 19 healthy participants Tasks: recall of autobiographical negative memories and subsequent focus on elicited emotions Results: rumination is associated with activity in subgenual ACC (sgACC) and medial PFC Subjects: MDD Task: rumination induction VS. abstract distraction Results: increased activation in amygdala, rostral anterior cingulate cortex/medial prefrontal cortex (PFC), dorsolateral PFC, and parahippocampal cortex, during rumination relative to an abstract distraction task Subjects: depressed patients Methods: factor structure & A paradigm alternating cognitive and emotional tasks Results: trait rumination correlated not only with amygdala but also hippocampus activity. Other regions in PCC, MPFC, dorsolateral PFC, and anterior insula Subjects: MDD patients Results: Rumination is associated with greater difficulties in deactivating the posterior cingulate cortex (PCC) As well as lower activation in anterior cingulate cortex (ACC) Subjects: 41 healthy participants Tasks: analytical VS. experiential self-focus Results: lower brooding scores measured by the RRS correlated with increased activation of the posterior midline structures Task: Cognitive reappraisal of emotional pictures Subjects: healthy female Results: higher amygdala response when increasing negative affect Lower medial prefrontal activity when decreasing negative affect.

Introduction

Introduction

Introduction Experiment design Active cognitive condition(task switching paradigm) Resting state condition

Introduction Hypothesis Lower attentional shifting abilities Ruminative processes Higher activation of self-related regions during lower cognitive load Higher ruminative tendencies Reduced engagement of the brain areas reflecting attentional control

Methods Participants: 20 healthy subjects ( 10 women & 10 men) Questionnaires: 22 – items of the Ruminative Response Scale (RRS) Beck Depression Inventory – II (BDI) Beck Anxiety Inventory (BAI)

Methods Behavioral task (task switching paradigm)

Methods Reaction? EMOTION

Methods Switch condition (high cognitive load) Emotion Gender Color Switch condition (high cognitive load) Emotion Emotion Emotion Repetition condition (low cognitive load)

Methods Data analysis Pre-processing steps: SPM5 + Matlab R2007b Realign: iterative rigid body transformations Normalize: MNI EPI template Smooth: a Gaussian kernel with FWHM of 8 mm Co-registration: A high-resolution structural image was co-registered with the mean image of the EPI seriesand normalized

Methods Cognitive task data: A two – step analysis: GLM for event-related design in SPM5 1. The onsets of conditions of interest were convolved with the canonical hemodynamic response function (HRF) and used as a regressor in the individual design matrix (first level) 2. The individual statistical images from each condition were used in a second-level ANOVA analysis to create the contrasts of interest, i.e. Repetition (easy) versus Switching (difficult)conditions.

Methods Resting state data Same pre-processing steps 1. independent component analysis (ICA) using the GIFT toolbox 2. The group maps were then inspected to select networks of interest for subsequent analyses

Methods Contrasts from the second level analysis of the cognitive task Maps from the ICA analysis of resting state Using a multiple linear regression model with the scores of each individual from the RRS, BDI, and BAI.

Results

Results Reaction times and switch cost did not correlate with RRS, but accuracy across all trials showed a negative correlation with RRS(r (20) = −.54, p = .011).

Results Resting-state fMRI data an attentional network map a visual network map ICA analysis DMN(medial prefrontal activations) DMN(medial parietal activations)

Results significant positive correlation for the visual network map between resting activity in the entorhinal cortex and the score of rumination

Results A negative correlation for the attention network map between rumination scores and activity in the left middle occipital gyrus

Results The first frontal-dominant DMN map showed a negative correlation of RRS with ACC and PCC, whereas the second map also showed a selective negative correlation in PCC and no effect in ACC.

Results Cognitive tasks Contrast Switch > Repetition significant increases in regions associated with attentional shifting and monitoring, including bilateral superior parietal lobules (SPL) and PCC. Contrast Repetition > Switch significant activations in the left caudate nucleus, the right inferior frontal gyrus, the left superior frontal gyrus, and the right dorsomedial PFC

Results 3. the contrast “easy > difficult” and rumination medial temporal lobe region strikingly similar to that observed in the previous analysis of resting state: i.e. the left entorhinal cortex

Results

Results Higher RRS scores were associated with lower activity in the right anterior insula during the easy vs. difficult condition

Results Brooding is not associated with the insula during the easy condition This weaker effect found for brooding during rest might reflect that ruminations at rest are also linked to more adaptive components of rumination, such as reflection and problem solving in association with mind wandering, whereas intrusive thoughts during low attention demands are more specifically linked to the maladaptive component associated with brooding

Discussion

Discussion Support a hypothesis People with a propensity to ruminate, even when non-depressed, tend to recruit brain systems mediating the retrieval of personal memories and self-related information more strongly or persistently than non-ruminators

Discussion Negative correlations By contrast, the reverse correlation observed in visual areas during rest suggests that individual with higher self-focus and ruminations allocate less resources to the processing of sensory visual inputs from the external world and/or engage less in visual imagery during rest.

Discussion Findings about insula The insula is implicated in self monitoring, saliency detection, and interoceptive awareness, we postulate that rumination tendencies may represent a maladaptive style of response with a relative lack of attention to bodily and affective signals in favor of internal cognitions

Discussion up Cognitive load down Rumination

Discussion Executive Functions Rumination These studies could not conclude if ruminations diminish cognitive resources, or if insufficient cognitive resources predispose to ruminations.

Thank you for your attention!