A specific population of brain cells involved in risky decision-making is identified in a study in rats published online this week in Nature. Modulating the activity of these cells can convert risk-seeking rats into risk-avoiding animals, the paper shows.
Risk-preference varies across individuals and within an individual from time to time, but the source of this variability in the brain is not known. Dopamine receptor type-2 (D2R)-expressing cells may have a role in regulating risky decisions, as studies have shown that problem gambling increases in patients with Parkinson’s disease who take medications that target these receptors.
Karl Deisseroth and colleagues developed a model of risk-seeking in rats that parallels human behaviour and showed that D2R-expressing cells within the nucleus accumbens drive risk-preference. When rats were presented with a choice between a ‘safe’ lever that consistently delivers a small amount of sucrose reward or a ‘risky’ lever that inconsistently delivers either a large reward (the more favourable outcome) or nothing (the less favourable outcome), the D2R-expressing cells signalled the less favourable outcome of a previous choice at an appropriate time during the decision period. This signal was higher in risk-averse animals than in risk-seeking animals and predicted whether, on the subsequent trial, a rat would choose the safe or risky option. The authors also showed that optogenetic (light-mediated) stimulation of the D2R neurons during the decision period decreased risky choices in risk-seeking rats, but not in rats that were already risk-averse.
Future research may explore whether distinct subpopulations of D2R-expressing nucleus accumbens cells encode even more specific components of risky decision-making.
Biomedical engineering: A ‘smart toilet’ for health monitoringNature Biomedical Engineering
Health: Maternal microbe associated with lower food allergy risk in infantsNature Communications
Environment: Opening plastic bags and bottles may generate microplasticsScientific Reports
Machine learning: An algorithm designed to smellNature Machine Intelligence