Dopaminergic Response to Drug Words in Cocaine Addiction

Rita Z. Goldstein, Dardo Tomasi, Nelly Alia‐Klein, Jean Honorio, Thomas Maloney, Patricia A. Woicik, Ruiliang Wang, Frank Telang, Nora D. VolkowJournal of Neuroscience 2009 · DOI

Abstract

When exposed to drug conditioned cues (stimuli associated with the drug), addicted individuals experience an intense desire for the drug, which is associated with increased dopamine cell firing. We hypothesized that drug-related words can trigger activation in the mesencephalon, where dopaminergic cells are located. During functional magnetic resonance imaging (fMRI), 15 individuals with cocaine use disorders and 15 demographically matched healthy control subjects pressed buttons for color of drug-related versus neutral words. Results showed that the drug words, but not neutral words, activated the mesencephalon in the cocaine users only. Further, in the cocaine users only, these increased drug-related mesencephalic responses were associated with enhanced verbal fluency specifically for drug words. Our results for the first time demonstrate fMRI response to drug words in cocaine-addicted individuals in mesencephalic regions as possibly associated with dopaminergic mechanisms and with conditioning to language (in this case drug words). The correlation between the brief verbal fluency test, which can be easily administered (crucial for clinical studies), and fMRI cue reactivity could be used as a biomarker of neurobiological changes in addiction.

Introduction

In addiction research, "cue reactivity" describes the phenomenon where stimuli associated with drug use—such as paraphernalia or specific environments—trigger intense cravings and activation of the brain's reward circuitry. This activation is primarily driven by the dopaminergic system, specifically within the midbrain or mesencephalon. While visual cues are well-studied, humans are uniquely sensitive to symbolic representations. This study investigates whether abstract linguistic symbols, such as drug-related words, can serve as conditioned stimuli powerful enough to activate these primitive reward centers in individuals with Cocaine Use Disorders (CUDs). Furthermore, the researchers aimed to identify a behavioral proxy for this neural activity that could be easily implemented in clinical settings.

Methods

The study compared 15 individuals with CUDs (primarily crack cocaine users, not currently seeking treatment) against 15 demographically matched healthy controls. Participants underwent two main assessments:

  1. Drug Word fMRI Task: Inside a 4T MRI scanner, participants performed a color-naming task (a variation of the Stroop task) involving 40 drug-related words and 40 neutral household words. This ensured semantic processing of the words while minimizing cognitive conflict.
  2. Verbal Fluency Task: Outside the scanner, participants were given one minute to generate as many words as possible within specific categories (e.g., "animals" vs. "drug-related words").

Functional MRI data were analyzed using Statistical Parametric Mapping (SPM2), with a specific focus on the mesencephalon—the anatomical home of the Ventral Tegmental Area (VTA) and Substantia Nigra—as a predefined Region of Interest (ROI).

Key Results

The study revealed a stark difference in how the brains of cocaine users process language compared to healthy controls.

  • Selective Midbrain Activation: In the CUD group, drug-related words triggered significant activation in the mesencephalon ($t = 4.1$), whereas neutral words did not. Healthy controls showed no significant mesencephalic response to either word type. A direct comparison confirmed a significant group-by-word interaction in the right mesencephalon ($t = 3.3$).
  • Brain-Behavior Correlation: A key finding was the strong positive correlation ($r = 0.58, p < 0.05$) between the BOLD signal change in the mesencephalon during the fMRI task and performance on the drug-related verbal fluency task. Cocaine users who were more "fluent" in drug-related vocabulary also showed greater midbrain reactivity to those words.
  • Prefrontal Regulation: Interestingly, higher midbrain reactivity was linked to decreased activity in the limbic prefrontal cortex. This suggests that the intense bottom-up signal from the reward system may overwhelm top-down regulatory mechanisms in addicted individuals.
Figure 4: Summary of behavioral and neural findings, including verbal fluency (A), mesencephalon reactivity (B), and the brain-behavior correlation (C) alongside fMRI localization.

Discussion and Limitations

These findings suggest that for individuals with addiction, language itself becomes a potent conditioned stimulus capable of activating the brain's "engine room" for dopamine. The correlation between the simple verbal fluency test and sophisticated fMRI measures is particularly promising, as it suggests the fluency test could serve as a low-cost, accessible biomarker for neurobiological changes in addiction.

However, the study's small sample size (N=30) necessitates caution. Additionally, the use of monetary rewards during the fMRI task may have interacted with drug-cue processing. Future research should validate these findings in larger, treatment-seeking populations to determine if drug fluency can reliably predict clinical outcomes such as relapse risk or treatment success.

Summary updated

Abstract

When exposed to drug conditioned cues (stimuli associated with the drug), addicted individuals experience an intense desire for the drug, which is associated with increased dopamine cell firing. We hypothesized that drug-related words can trigger activation in the mesencephalon, where dopaminergic cells are located. During functional magnetic resonance imaging (fMRI), 15 individuals with cocaine use disorders and 15 demographically matched healthy control subjects pressed buttons for color of drug-related versus neutral words. Results showed that the drug words, but not neutral words, activated the mesencephalon in the cocaine users only. Further, in the cocaine users only, these increased drug-related mesencephalic responses were associated with enhanced verbal fluency specifically for drug words. Our results for the first time demonstrate fMRI response to drug words in cocaine-addicted individuals in mesencephalic regions as possibly associated with dopaminergic mechanisms and with conditioning to language (in this case drug words). The correlation between the brief verbal fluency test, which can be easily administered (crucial for clinical studies), and fMRI cue reactivity could be used as a biomarker of neurobiological changes in addiction.
BibTeX
@article{goldstein2009dopaminergic,
  author = {Rita Z. Goldstein and Dardo Tomasi and Nelly Alia‐Klein and Jean Honorio and Thomas Maloney and Patricia A. Woicik and Ruiliang Wang and Frank Telang and Nora D. Volkow},
  title = {{Dopaminergic Response to Drug Words in Cocaine Addiction}},
  journal = {Journal of Neuroscience},
  year = {2009},
  doi = {10.1523/jneurosci.4247-08.2009},
}
Figure