Inducing Expansive Thought Dynamics To Support Emotional Recovery

Emotions and spontaneous thoughts are tightly intertwined, with each shaping the other over time. While emotions influence the content of spontaneous thought, the way thoughts unfold over time–thought dynamics–may in turn regulate emotional experience. In particular, expansive thinking, where thoughts move quickly across diverse ideas, may facilitate emotional recovery from negative states. Our recent work provides correlational evidence for this possibility. Across two studies (N = 605), participants completed a negative mood induction followed by a Think-Aloud task in which they narrated their stream of thought. Emotional recovery was indexed as the change in mood from before to after the task. We quantified expansiveness using three complementary metrics (see Methods) capturing the semantic network structure, momentum, and novelty of the thought stream. Across both studies, individuals who exhibited more expansive thought dynamics showed greater emotional recovery. The present study tests whether expansiveness causally promotes emotional recovery by experimentally manipulating thought dynamics in real time. After a negative mood induction via autobiographical memory recall, participants will complete a Think-Aloud task in one of three conditions. During the task, participants’ speech is transcribed and analyzed continuously, and participants see a visual meter that provides ongoing feedback indicating whether their recent speech falls above or below a predefined similarity threshold relative to their prior thoughts. Participants are randomly assigned to one of three conditions: (1) Expansive: participants are instructed to generate conceptually different thoughts and are told that their goal is to keep the meter above the threshold. (2) Constrained: participants are instructed to generate conceptually similar thoughts and are told that their goal is to keep the meter below the threshold. (3) Control: participants are told to think naturally. They see the meter and are informed what it represents, but they are not given any instructions to try to move above or below the threshold. The task will run on a custom Python-based platform that transcribes speech, computes cosine similarity between recent speech and earlier portions of the speech stream, and provides feedback based on condition-specific similarity thresholds. Emotional recovery will be indexed as the change in self-reported emotion from before to after the Think-Aloud task. This design allows us to test whether inducing expansive thought dynamics causally improves emotional recovery following a negative mood induction. Thus, this project aims to test (i) whether thought expansiveness can be experimentally manipulated and (ii) whether doing so facilitates emotional recovery following a negative mood induction. We hypothesize that expansiveness will be highest in the expansive condition, lowest in the constrained condition, and intermediate in the control condition, and that greater expansiveness will lead to greater improvements in mood.