“How green is the grass on the other side?” HIGHLIGHTS FROM Dr. RUSHWORTH PLENARY LECTURE (OXFORD, UK)
After two years of intense preparation and the usual last-minute changes, the long-awaited FENS Forum was finally held at the CCIB in Barcelona last week. It was a real success in terms of attendance and engagement in the scientific and social events throughout the Forum. From SENC, we would like to thank the FENS Organization Committee and especially the invited speakers and participants. It would be our deepest pleasure to meet you all again in the next FENS Forum (Milan 2014).
In this post, we would like to give you a flavour of the first plenary lecture presented by Dr. Matthew Rushworth, principal investigator at the Decision and Action laboratory in the Department of Experimental Psychology at Oxford University (rushworth.psy.ox.ac.uk/). His talk was introduced by Sten Grillner (President of FENS) and focused on the reward-guided decision-making and learning mechanisms in the frontal lobe. More specifically, he gave a thoughtful insight into the functional role of different reward-related frontal brain regions including the FPC (Frontopolar Cortex), vmPFC (ventromedial Prefrontal Cortex), the mOFC (medial Orbital Frontal Cortex), the lOFC (lateral Orbital Frontal Cortex), the aPFC (anterior part of the prefrontal cortex) and the ACC (Anterior Cingulate Cortex).
So what are the prefrontal computations that underlie our decision-making behaviour? In an early fMRI study with human subjects (Boorman, Behrens, Woolrich, & Rushworth, 2009), the Rushworth group successfully dissociated the complementary role of FPC and vmPFC during voluntary choice. “Every time we make a decision we also make a prediction about it” said Dr. Rushtworth. Indeed, in this study they used a binary decision-making task in which the subjects were required to choose between two different actions depending on the expected value (EV) associated with them. The EV was a function of the reward probability (which the subjects estimated from recent outcomes) and magnitude (which was displayed on the screen as a number of points inside a green rectangle). What they concluded after the fMRI analyses is that the FPC was continually tracking the evidence favouring a switch in behaviour. By contrast, the vmPFC activated at the time of the decision and was encoding the value of the chosen relative to the unchosen action. This added a new layer of knowledge to the previous view of the FPC being merely active at the point of switching.
If there is a reason why Dr. Rushworth prefers to keep his personal image private, it must be found in his search for human analogs in monkeys, which has caused a series of unfortunate encounters with animal right groups at Oxford. Several studies have suggested that human patients with mOFC/vmPFC lesions make unusual decisions. This closely resembles to what the Rushworth group found in a selective lesion study in macaques (Noonan et al., 2010). When they bilaterally aspirated the mOFC, the animals tended to make irrational choices that were influenced by the presence of irrelevant alternatives. In other words, the value comparison or second stage of the “standard model” of decision making was impaired. Lesions in the lOFC, another functional specialization of the OFC, were correlated with anomalies in value learning. The animals failed to assign correct credit to rewards, i.e. the learning association between an outcome and the particular stimulus that was chosen. Although it might seem counterintuitive at first, the lOFC lesioned animals struggled to learn when each option’s value was further apart from the others because they erroneously estimated the value of each option as close to the mean.
Dr. Rushworth moved on to explain the special role of the aPFC. In this line of research, they aim to understand the counterfactual choices, meaning the choices that we might have pursued but, in the end, did not (Boorman, Behrens, & Rushworth, 2011). Human fMRI images were acquired while the subjects performed a three-alternative decision making task. Data analysis revealed the aPFC as the brain region encoding counterfactual prediction error. In other words, the aPFC monitors how good it would have been to switch to unchosen options. This ability to track the value of an alternative course of action lies at the very heart of our adaptive behaviour in real-world contexts.
To further extend our knowledge on real-world situations, the latest publication from Rushworth lab focused on foraging mechanisms and their human counterparts, which are of great relevance in behavioural economics (Kolling, Behrens, Mars, & Rushworth, 2012). Dr.Rushworth introduced the topic with the amusing but illustrative example of deciding on our relationship with a potential spouse or partner: “We just can’t sample every possibility even when we take important decisions, we rather evaluate whether switching or not to an alternative option”. In this Science paper, the main variation with respect to the aforementioned studies lied in a multiple-choice task design that allowed the distinction of two separate steps in each trial: 1) the stay or foraging step in which the subjects decide whether to engage with the item they have encountered or to search for better alternatives in the environment but paying a cost for it; 2) the decision step where they make a choice. The results pointed to a functional dissociation between the vmPFC and the ACC. The vmPFC codes the value comparison of chosen and unchosen options whereas the ACC signals the value of switching to a course of action alternative to that which is taken. Together, this seems to indicate that we can alternate between decision-making and foraging behaviour relying on specific neural mechanisms in the vmPFC and the ACC.
Boorman, E. D., Behrens, T. E. J., Woolrich, M. W., & Rushworth, M. F. S. (2009). How green is the grass on the other side? Frontopolar cortex and the evidence in favor of alternative courses of action. Neuron, 62(5), 733-43. Elsevier Ltd. doi:10.1016/j.neuron.2009.05.014
Boorman, E. D., Behrens, T. E., & Rushworth, M. F. (2011). Counterfactual choice and learning in a neural network centered on human lateral frontopolar cortex. PLoS biology, 9(6), e1001093. doi:10.1371/journal.pbio.1001093
Kolling, N., Behrens, T. E. J., Mars, R. B., & Rushworth, M. F. S. (2012). Neural mechanisms of foraging. Science (New York, N.Y.), 336(6077), 95-8. doi:10.1126/science.1216930
Noonan, M. P., Walton, M. E., Behrens, T. E. J., Sallet, J., Buckley, M. J., & Rushworth, M. F. S. (2010). Separate value comparison and learning mechanisms in macaque medial and lateral orbitofrontal cortex. Proceedings of the National Academy of Sciences of the United States of America, 107(47), 20547-52. doi:10.1073/pnas.1012246107
Estudiante pre-doctoral de la Universidad de Barcelona