21 December 2017
Decision-making under risk: from drugs to dopamine
How can you tell if a decision was impulsive or risky? Using the Cambridge Gambling Task to unpick how different brain regions, different drugs and different dopamine levels impact on decision-making.
The Cambridge Gambling Task (CGT) was designed to measure decision-making under risk. The term decision-making under risk refers to when the outcome of a decision is unknown, but the probability of such an outcome is known (Camerer & Weber, 1992).
The odds of winning or losing are made explicit in the CGT by the ratio of red to blue boxes. The task requires participants to decide whether a token will be hidden behind a red box or a blue box. Participants are then asked to bet a proportion of their points to support this decision.
The CGT separates itself from more traditional gambling tasks with the crucial ability to dissociate risk-taking from impulsivity. This is due to the different bet conditions of the task. In the ascending condition, the bet value incrementally increases on screen so the participant is required to wait (not be impulsive) in order to place a high-stakes bet (take a risk).
Furthermore, the task places limited demands on learning and working memory, as information from one trial does not inform the next. Consequently, the CGT is a clean measure of decision-making under risk, and therefore a good paradigm to explore the neuroscience behind such decision-making. Here we will discuss some key research into this area that has used CGT.
The role of different brain regions in decision-making
Insights from lesion studies
Lesion studies can be very informative in differentiating the roles of various brain regions in decision-making.
The extent to which the ventromedial prefrontal cortex (vmPFC) is implicated in decision-making under risk, and the role of the insular cortex in this process, was investigated in a lesion study by Clark et al. (2008). The study found that participants with vmPFC lesions bet more, irrespective of the probability for success. In contrast, the participants with lesions to the insular cortex only made riskier bets when the odds of winning were poor, which led to more ‘bankruptcies’.
These findings suggest that the vmPFC and insular cortex play crucial roles in decision-making under risk.
The effect of drugs on decision-making
An example from frontotemporal dementia
The frontal variant of frontotemporal dementia (fvFTD) leads to compromise in functions served by the prefrontal cortex, including decision-making.
A study conducted by Rahman et al. (2006), investigated the extent to which a single dose (40mg) of methylphenidate (‘Ritalin’) could mitigate risky decision-making behaviour in fvFTD. The study found that drug administration resulted in patients making significantly less risky decisions on the CGT task, as indicated by placing lower value bets on the drug than off (Rahman et al., 2006).
The authors suggested that this ‘normalised’ decision-making could be the result of drug-action on the orbitofrontal cortex (akin to vmPFC) (Rahman et al., 2006), which is typically compromised in fvFTD (Salmon et al., 2003).
The link between dopamine and decision-making
As exemplified by Parkinson’s disease
From the age of about 20 onwards, dopamine function is lost at a rate of about 10% per decade (Mukherjee et al., 2002). This loss of dopamine is thought be partly responsible for the reduced sensitivity to reward that results in a selective decrease in appetite for risk as we age. A more extreme loss of dopaminergic function results in Parkinson’s disease.
L-Dopa is prescribed for Parkinson’s disease to normalise dopamine levels in severely affected brain regions (i.e. the dorsal striatum). However, other regions and networks which are relatively intact early in the disease (i.e. the ventral striatum), may be ‘overdosed’ by this medication (Cools, Barker, Sahakian, & Robbins, 2003).
In a study by Cools, Barker, Sahakian, and Robbins (2003), patients with Parkinson’s disease on stable daily L-Dopa medication were assessed on two occasions: one with their usual treatment, and one when they were asked to abstain from their normal medication. The study found that patients ‘on’ medication made rational decisions (i.e. decision-making in line with the relevant probabilities), but placed impulsive bets (Cools et al., 2003). The authors linked this finding with ‘over-dosed’ ventral-striatal orbitofrontal circuitries, leading to motor impulsivity.
In summary the CGT is a useful and well-validated measure of decision-making. Indeed, the CGT assesses decision-making outside of a learning context and task-performance can parse impulsive from risky decisions.
As a robust measure of decision-making, CGT offers a tool for investigating which neural networks are impaired in abnormal decision-making behaviours, and how they are modifiable with drug-action.
Interested in measuring decision-making?
Camerer, C., & Weber, M. (1992). Recent developments in modeling preferences: Uncertainty and ambiguity. Journal of Risk and Uncertainty, 5(4), 325–370. https://doi.org/10.1007/BF00122575
Clark, L., Bechara, A., Damasio, H., Aitken, M. R. F., Sahakian, B. J., & Robbins, T. W. (2008). Differential effects of insular and ventromedial prefrontal cortex lesions on risky decision-making. Brain, 131(5), 1311–1322. https://doi.org/10.1093/brain/awn066
Cools, R., Barker, R. A., Sahakian, B. J., & Robbins, T. W. (2003). L-Dopa medication remediates cognitive inflexibility, but increases impulsivity in patients with Parkinson’s disease. Neuropsychologia, 41(11), 1431–1441. https://doi.org/10.1016/S0028-3932(03)00117-9
Mukherjee, J., Christian, B. T., Dunigan, K. A., Shi, B., Narayanan, T. K., Satter, M., & Mantil, J. (2002). Brain imaging of18F-fallypride in normal volunteers: Blood analysis, distribution, test-retest studies, and preliminary assessment of sensitivity to aging effects on dopamine D-2/D-3 receptors. Synapse, 46(3), 170–188. https://doi.org/10.1002/syn.10128
Rahman, S., Robbins, T. W., Hodges, J. R., Mehta, M. A., Nestor, P. J., Clark, L., & Sahakian, B. J. (2006). Methylphenidate (’Ritalin’) can ameliorate abnormal risk-taking behavior in the frontal variant of frontotemporal dementia. Neuropsychopharmacology : Official Publication of the American College of Neuropsychopharmacology, 31(3), 651–8. https://doi.org/10.1038/sj.npp.1300886
Salmon, E., Garraux, G., Delbeuck, X., Collette, F., Kalbe, E., Zuendorf, G., … Herholz, K. (2003). Predominant ventromedial frontopolar metabolic impairment in frontotemporal dementia. NeuroImage, 20(1), 435–40. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/14527604