Although a negative finding would not necessarily contradict the role of DLPFC in decision-making, a positive result would lend credence to the neuronal implementation of competitive decision models, similar to previous single-neuron representations of complex decision variables 41. We tested whether DLPFC neurons encode values of specific choice objects termed ‘object values’, in analogy to action values 6 and in line with competitive choice mechanisms 6, 39, 40. We hypothesized that individual DLPFC neurons encode the construction of values from experience, their formatting into object-specific decision variables, and their conversion to object choices. DLPFC is also connected to sensory, motor and reward systems 29, 37, including parietal cortex and striatum, where experience-based value signals are found 1, 3, 5, and anterior cingulate cortex, where lesions impair performance based on reward experience 38. Previous neurophysiological studies showed that DLPFC neurons encode important economic decision variables including reward probability, reward magnitude, effort 19, 26, reward and choice history 20, 35, 36. The DLPFC is implicated in diverse functions including decision-making 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, behavioural control 29, 30, 31, 32, 33, 34 and reinforcement learning 35, 36. Here we recorded the activity of single neurons in the dorsolateral prefrontal cortex (DLPFC) of monkeys performing an object-based foraging task. However, it is unclear whether object value neurons also encode recent reward experiences, as implied by the concept of value construction 2, 3, 4, 7, and whether they directly convert values to choices, as predicted by computational models 8, 9. ![]() Orbitofrontal cortex (OFC) neurons encode economic object valuations when value is explicitly signalled by external cues 16, 17, 18. This distinction is significant, as objects constitute the fundamental choice unit in economic theory. Second, although neuronal values were typically referenced to actions in previous studies 1, 3, 5, decisions are often made between objects. Although object-specific valuations seem computationally advantageous, relative valuations-which can be derived from object-specific values-are frequently observed in human imaging neural population signals 10, 11, 12, 13, 14, 15. Biologically realistic decision models use separate value inputs for different choice objects that compete through winner-take-all mechanisms 6, 7, 8, 9, rather than explicit relative (comparative) valuations. ![]() Although neurophysiological studies uncovered experience-based value signals in different brain structures 1, 3, 5, key questions about the neural value code remain unresolved.įirst, it is unclear how individual neurons encode value estimates as input for decision mechanisms. Such value estimates constitute critical elements in reinforcement learning 6 and computational decision theories 7, 8, 9. In natural environments, reward probabilities are often unknown and decision-making requires internal value estimation from recent experience 1, 2, 3, 4, 5. Rewards are essential goals for economic decisions and behaviour. These findings suggest a dynamic single-neuron and population value code in DLPFC that advances from reward experiences to economic object values and future choices. Decoding from unselected populations enables a read-out of motivational and decision variables not emphasized by individual neurons. Individual neurons dynamically encode both, the updating of object values from recently experienced rewards, and their subsequent conversion to object choices during decision-making. These neuronal object values satisfy principles of competitive choice mechanisms, track performance fluctuations and follow predictions of a classical behavioural model (Herrnstein’s matching law). ![]() As monkeys perform a reward-based foraging task, individual DLPFC neurons signal the value of specific choice objects derived from recent experience. Here we show that the dorsolateral prefrontal cortex (DLPFC) implements a flexible value code based on object-specific valuations by single neurons. Yet, how such valuations are converted to economic decisions remains unclear. Neuronal reward valuations provide the physiological basis for economic behaviour.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |