Dataset Information

Reinforcement Learning Disruptions in Individuals With Depression and Sensitivity to Symptom Change Following Cognitive Behavioral Therapy.

ABSTRACT:

Importance

Major depressive disorder is prevalent and impairing. Parsing neurocomputational substrates of reinforcement learning in individuals with depression may facilitate a mechanistic understanding of the disorder and suggest new cognitive therapeutic targets.

Objective

To determine associations among computational model-derived reinforcement learning parameters, depression symptoms, and symptom changes after treatment.

Design, setting, and participants

In this mixed cross-sectional-cohort study, individuals performed reward and loss variants of a probabilistic learning task during functional magnetic resonance imaging at baseline and follow-up. A volunteer sample with and without a depression diagnosis was recruited from the community. Participants were assessed from July 2011 to February 2017, and data were analyzed from May 2017 to May 2021.

Main outcomes and measures

Computational model-based analyses of participants' choices assessed a priori hypotheses about associations between components of reward-based and loss-based learning with depression symptoms. Changes in both learning parameters and symptoms were then assessed in a subset of participants who received cognitive behavioral therapy (CBT).

Results

Of 101 included adults, 69 (68.3%) were female, and the mean (SD) age was 34.4 (11.2) years. A total of 69 participants with a depression diagnosis and 32 participants without a depression diagnosis were included at baseline; 48 participants (28 with depression who received CBT and 20 without depression) were included at follow-up (mean [SD] of 115.1 [15.6] days). Computational model-based analyses of behavioral choices and neural data identified associations of learning with symptoms during reward learning and loss learning, respectively. During reward learning only, anhedonia (and not negative affect or arousal) was associated with model-derived learning parameters (learning rate: posterior mean regression β = -0.14; 95% credible interval [CrI], -0.12 to -0.03; outcome sensitivity: posterior mean regression β = 0.18; 95% CrI, 0.02 to 0.37) and neural learning signals (moderation of association between striatal prediction error and expected value signals: t97 = -2.10; P = .04). During loss learning only, negative affect (and not anhedonia or arousal) was associated with learning parameters (outcome shift: posterior mean regression β = -0.11; 95% CrI, -0.20 to -0.01) and disrupted neural encoding of learning signals (association with subgenual anterior cingulate prediction error signals: r = -0.28; P = .005). Symptom improvement following CBT was associated with normalization of learning parameters that were disrupted at baseline (reward learning rate: posterior mean regression β = 0.15; 90% CrI, 0.001 to 0.41; loss outcome shift: posterior mean regression β = 0.42; 90% CrI, 0.09 to 0.77).

Conclusions and relevance

In this study, the mapping of reinforcement learning components to symptoms of major depression revealed mechanistic features associated with these symptoms and points to possible learning-based therapeutic processes and targets.

SUBMITTER: Brown VM

PROVIDER: S-EPMC8319827 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:BACKGROUND:Aberrant activity of the subcallosal cingulate (SCC) is a common theme across pharmacologic treatment efficacy prediction studies. The functioning of the SCC in psychotherapeutic interventions is relatively understudied, as are functional differences among SCC subdivisions. We conducted functional connectivity analyses (rsFC) on resting-state functional magnetic resonance imaging (fMRI) data, collected before and after a course of cognitive behavioral therapy (CBT) in patients with major depressive disorder (MDD), using seeds from three SCC subdivisions. METHODS:Resting-state data were collected from unmedicated patients with current MDD (Hamilton Depression Rating Scale-17 > 16) before and after 14-sessions of CBT monotherapy. Treatment outcome was assessed using the Beck Depression Inventory (BDI). Rostral anterior cingulate (rACC), anterior subcallosal cingulate (aSCC), and Brodmann's area 25 (BA25) masks were used as seeds in connectivity analyses that assessed baseline rsFC and symptom severity, changes in connectivity related to symptom improvement after CBT, and prediction of treatment outcomes using whole-brain baseline connectivity. RESULTS:Pretreatment BDI negatively correlated with pretreatment rACC ~ dorsolateral prefrontal cortex and aSCC ~ lateral prefrontal cortex rsFC. In a region-of-interest longitudinal analysis, rsFC between these regions increased post-treatment (p < 0.05FDR). In whole-brain analyses, BA25 ~ paracentral lobule and rACC ~ paracentral lobule connectivities decreased post-treatment. Whole-brain baseline rsFC with SCC did not predict clinical improvement. CONCLUSIONS:rsFC features of rACC and aSCC, but not BA25, correlated inversely with baseline depression severity, and increased following CBT. Subdivisions of SCC involved in top-down emotion regulation may be more involved in cognitive interventions, while BA25 may be more informative for interventions targeting bottom-up processing. Results emphasize the importance of subdividing the SCC in connectivity analyses.

Project description:BackgroundInternet- and mobile-based interventions are most efficacious in the treatment of depression when they involve some form of guidance, but providing guidance requires resources such as trained personnel, who might not always be available (eg, during lockdowns to contain the COVID-19 pandemic).ObjectiveThe current analysis focuses on changes in symptoms of depression in a guided sample of patients with depression who registered for an internet-based intervention, the iFightDepression tool, as well as the extent of intervention use, compared to an unguided sample. The objective is to further understand the effects of guidance and adherence on the intervention's potential to induce symptom change.MethodsLog data from two convenience samples in German routine care were used to assess symptom change after 6-9 weeks of intervention as well as minimal dose (finishing at least two workshops). A linear regression model with changes in Patient Health Questionnaire (PHQ-9) score as a dependent variable and guidance and minimal dose as well as their interaction as independent variables was specified.ResultsData from 1423 people with symptoms of depression (n=940 unguided, 66.1%) were included in the current analysis. In the linear regression model predicting symptom change, a significant interaction of guidance and minimal dose revealed a specifically greater improvement for patients who received guidance and also worked with the intervention content (β=-1.75, t=-2.37, P=.02), while there was little difference in symptom change due to guidance in the group that did not use the intervention. In this model, the main effect of guidance was only marginally significant (β=-.53, t=-1.78, P=.08).ConclusionsGuidance in internet-based interventions for depression is not only an important factor to facilitate adherence, but also seems to further improve results for patients adhering to the intervention compared to those who do the same but without guidance.