The habenula encodes negative motivational value associated with primary punishment in humans

Abstract

Learning what to approach, and what to avoid, involves assigning value to environmental cues that predict positive and negative events. Studies in animals indicate that the lateral habenula encodes the previously learned negative motivational value of stimuli. However, involvement of the habenula in dynamic trial-by-trial aversive learning has not been assessed, and the functional role of this structure in humans remains poorly characterized, in part, due to its small size. Using high-resolution functional neuroimaging and computational modeling of reinforcement learning, we demonstrate positive habenula responses to the dynamically changing values of cues signaling painful electric shocks, which predict behavioral suppression of responses to those cues across individuals. By contrast, negative habenula responses to monetary reward cue values predict behavioral invigoration. Our findings show that the habenula plays a key role in an online aversive learning system and in generating associated motivated behavior in humans.

Keywords: conditioned behavior; high-resolution fMRI; pallidum.

Fig. 1.

Conditioning task and multiple indices of learning. (A) Exemplar trial (a detailed description is provided in the main text). (B) Explicit preference scores for win, loss, shock, and neutral (NEU) CSs (maximum score of 24). (C) Reaction times to respond to fixation flickers on win, loss, shock, and neutral CSs. (D) Pupil responses to win, loss, shock, and neutral CSs. (E) Relationship between autonomic (pupil responses to shock relative to neutral CSs) and implicit (conditioned suppression) measures of conditioning. Error bars and the shaded region in D represent SEMs. *P < 0.01; **P < 0.005.

Fig. 2.

Habenula results. (A) Location of the habenula on a coronal slice of a representative subject (Upper) and the trial-by-trial evolution of shock CS value during a single task block for a representative subject (Lower). Empty markers (□) indicate high-probability trials, and filled markers (■) indicate low-probability trials. (B) BOLD responses from the right habenula correspond to the dynamically changing values of win, loss, and shock CSs. (C) There is a positive correlation between the right habenula response to shock CS value and conditioned suppression. (D) There is a negative correlation between the right habenula response to win CS value and conditioned invigoration. Error bars represent SEM. *P < 0.01; **P < 0.001.

Fig. 3.

Whole-brain analysis showing activation to shock CS value. Pallidal BOLD responses correspond to shock CS value. Images are thresholded at P < 0.005 (uncorrected) and at k ≥ 10, and they are overlaid on the average normalized anatomical image; the color bar represents t values.