Pharmacological modulation of subliminal learning in Parkinson's and Tourette's syndromes

Stefano Palminteri, Maël Lebreton, Yulia Worbe, David Grabli, Andréas Hartmann, Mathias PessiglioneProceedings of the National Academy of Sciences 2009 · DOI

TL;DR

This study investigates how dopamine influences learning, specifically whether it affects conscious or subconscious processes. By using a subliminal learning task in patients with Parkinson's disease (PD) and Tourette's syndrome (TS), researchers found that dopamine modulation can bias learning towards seeking rewards or avoiding punishments, even when participants are unaware of the cues. This suggests dopamine plays a crucial role in subconscious reinforcement learning.

What Problem Does This Paper Address?

Instrumental learning theories propose that we learn from our successes and failures by updating our choices based on "reward prediction errors" – the difference between what we expected and what actually happened. Dopamine is thought to be a key neurotransmitter involved in signaling these errors. While drugs that affect dopamine have been shown to influence learning, it's been unclear whether this influence operates on conscious learning (where we're aware of the cues and outcomes) or subconscious learning (where we're not). This paper aims to disentangle these two by testing the effects of dopamine-modulating drugs on learning that occurs without conscious awareness of the stimuli.

Methods

The researchers employed a subliminal instrumental learning paradigm. Participants were presented with abstract visual cues that were rapidly flashed and then masked, preventing conscious perception. These masked cues were associated with different monetary outcomes: a reward (+1€), a punishment (-1€), or a neutral outcome (0€). Participants had to make a choice (either press a button or not) and then received the outcome. The core idea is that if participants learn to associate the masked cues with outcomes, their choices will reflect this learning, even without conscious awareness.

To assess the role of dopamine, they tested:

  • Parkinson's Disease (PD) patients: These patients have a deficiency in dopamine. They were tested both "off" their dopamine-enhancing medication (levodopa) and "on" it.
  • Tourette's Syndrome (TS) patients: These patients often have an overactive dopamine system and are treated with dopamine blockers (neuroleptics). They were also tested both "off" and "on" their medication.

They measured:

  • Monetary payoff: This reflects overall learning performance.
  • Reward bias: This specifically measures the difference in learning between reward and punishment conditions.
  • Visual discrimination sensitivity (d'): This was used as an objective measure to confirm that cues were indeed perceived subliminally. If d' is near zero, it indicates participants couldn't consciously distinguish between cues.

Datasets

Not discussed in the paper. The study uses a custom experimental task with human participants.

Key Insights and Ideas

  • Dopamine drives subconscious learning: The most striking finding is that dopamine modulation significantly impacts learning even when stimuli are presented subliminally, meaning participants are not consciously aware of the cues. This strongly implicates dopamine in non-conscious reinforcement processes.
  • Opposing effects in PD and TS: The study revealed a clear double dissociation. Dopamine enhancers (in PD) improved reward-seeking behavior, while dopamine blockers (in TS) favored punishment avoidance. This highlights the specific roles of dopamine in these opposing learning biases.
  • Learning is possible without conscious awareness: The results extend previous work by demonstrating that complex instrumental learning, involving prediction errors and value updating, can occur outside of conscious perception.
  • Medication effects are more pronounced than disease effects: The differences in learning observed when comparing patients "on" versus "off" medication were generally larger and more reliable than the differences between patient groups and healthy controls. This underscores the direct impact of pharmacological manipulation.

Key Results

  • PD patients on medication: Showed improved reward learning but worsened punishment avoidance.
  • PD patients off medication: Showed impaired reward learning but better punishment avoidance.
  • TS patients on medication: Showed improved punishment avoidance but impaired reward seeking.
  • TS patients off medication: Showed improved reward seeking but impaired punishment avoidance.
  • Subliminal learning confirmed: Participants' learning performance (payoffs) was not correlated with their ability to consciously discriminate the cues (d'), indicating that the observed learning was largely subconscious.
  • Reward bias manipulation: The "reward bias" (difference between learning from rewards and punishments) was significantly increased by dopamine enhancers in PD and significantly decreased by dopamine blockers in TS, demonstrating a direct pharmacological influence on this bias.

Limitations and Future Work

  • Subliminal perception is hard to guarantee: While masking and d' measures were used, it's impossible to definitively prove that all cues were subliminal for all participants in all trials. The authors acknowledge this inherent difficulty in studying subconscious processes.
  • Generalizability of findings: The study focused on specific patient populations (PD and TS) and a particular type of learning. Future work could explore other neurological conditions or different forms of learning.
  • Mechanistic details: The exact dopamine receptors and pathways involved in this subconscious learning remain speculative. Further research using neuroimaging or more targeted pharmacological interventions could shed light on these mechanisms.
  • Conscious vs. subconscious interaction: While the study demonstrates subconscious learning, it doesn't fully explore how conscious feelings or awareness might interact with or be influenced by these subconscious processes. Future experiments could investigate this interplay.
  • Drug effects on conscious perception: The study focused on learning, but it's possible that medications also affect conscious perception or feelings associated with cues, which could indirectly influence learning. This was not directly assessed.

Summary updated

Abstract

Theories of instrumental learning aim to elucidate the mechanisms that integrate success and failure to improve future decisions. One computational solution consists of updating the value of choices in proportion to reward prediction errors, which are potentially encoded in dopamine signals. Accordingly, drugs that modulate dopamine transmission were shown to impact instrumental learning performance. However, whether these drugs act on conscious or subconscious learning processes remains unclear. To address this issue, we examined the effects of dopamine-related medications in a subliminal instrumental learning paradigm. To assess generality of dopamine implication, we tested both dopamine enhancers in Parkinson's disease (PD) and dopamine blockers in Tourette's syndrome (TS). During the task, patients had to learn from monetary outcomes the expected value of a risky choice. The different outcomes (rewards and punishments) were announced by visual cues, which were masked such that patients could not consciously perceive them. Boosting dopamine transmission in PD patients improved reward learning but worsened punishment avoidance. Conversely, blocking dopamine transmission in TS patients favored punishment avoidance but impaired reward seeking. These results thus extend previous findings in PD to subliminal situations and to another pathological condition, TS. More generally, they suggest that pharmacological manipulation of dopamine transmission can subconsciously drive us to either get more rewards or avoid more punishments.
BibTeX
@inproceedings{palminteri2009pharmacological,
  author = {Stefano Palminteri and Maël Lebreton and Yulia Worbe and David Grabli and Andréas Hartmann and Mathias Pessiglione},
  title = {{Pharmacological modulation of subliminal learning in Parkinson's and Tourette's syndromes}},
  booktitle = {Proceedings of the National Academy of Sciences},
  year = {2009},
  doi = {10.1073/pnas.0904035106},
}
Figure