Abstract: Human orbitofrontal cortex (OFC) has long been implicated in value-based decision making. In recent years, convergent evidence from human and model organisms has further elucidated its role in representing reward-related computations underlying decision making. However, a detailed description of these processes remains elusive due in part to (1) limitations in our ability to observe human OFC neural dynamics at the timescale of decision processes and (2) methodological and interspecies differences that make it challenging to connect human and animal findings or to resolve discrepancies when they arise. Here, we sought to address these challenges by conducting multi-electrode electrocorticography (ECoG) recordings in neurosurgical patients during economic decision making to elucidate the electrophysiological signature, sub-second temporal profile, and anatomical distribution of reward-related computations within human OFC. We found that high-frequency activity (HFA) (70–200 Hz) reflected multiple valuation components grouped in two classes of valuation signals that were dissociable in temporal profile and information content: (1) fast, transient responses reflecting signals associated with choice and outcome processing, including anticipated risk and outcome regret, and (2) sustained responses explicitly encoding what happened in the immediately preceding trial. Anatomically, these responses were widely distributed in partially overlapping networks, including regions in the central OFC (Brodmann areas 11 and 13), which have been consistently implicated in reward processing in animal single-unit studies. Together, these results integrate insights drawn from human and animal studies and provide evidence for a role of human OFC in representing multiple reward computations.
Abstract: Human intracranial electroencephalography (iEEG) recordings provide data with much greater spatiotemporal precision than is possible from data obtained using scalp EEG, magnetoencephalography (MEG), or functional MRI. Until recently, the fusion of anatomical data (MRI and computed tomography (CT) images) with electrophysiological data and their subsequent analysis have required the use of technologically and conceptually challenging combinations of software. Here, we describe a comprehensive protocol that enables complex raw human iEEG data to be converted into more readily comprehensible illustrative representations. The protocol uses an open-source toolbox for electrophysiological data analysis (FieldTrip). This allows iEEG researchers to build on a continuously growing body of scriptable and reproducible analysis methods that, over the past decade, have been developed and used by a large research community. In this protocol, we describe how to analyze complex iEEG datasets by providing an intuitive and rapid approach that can handle both neuroanatomical information and large electrophysiological datasets. We provide a worked example using an example dataset. We also explain how to automate the protocol and adjust the settings to enable analysis of iEEG datasets with other characteristics. The protocol can be implemented by a graduate student or postdoctoral fellow with minimal MATLAB experience and takes approximately an hour to execute, excluding the automated cortical surface extraction.
Abstract: The precise role of serotonin in human brain function remains elusive due at least in part to a lack of rapid measurement technologies available for this neurotransmitter. In order to provide a coherent account of serotonin function, we used a machine learning approach to extract serotonergic signals from fast scan cyclic voltammetry recordings acquired from fourteen human brains. Here we report these first sub- second measurements of serotonin fluctuations in human striatum and show how they correlate with outcomes and decisions in a sequential investment game. We find that serotonergic concentrations transiently increase following negative reward prediction errors. Importantly, these fluctuations reverse when counterfactual losses predominate, suggesting that serotonin tracks the negative outcomes associated with foregone gains. Crucially, these data provide evidence that the serotonergic system acts as an opponent to dopamine signaling, as anticipated by theoretical models. Strikingly, serotonin transients also encoded next actions in a manner that correlated with decreased exposure to both forms of loss. Taken together, these novel findings suggest that serotonin encodes a protective action strategy that mitigates risk, shaping action selection contingent upon negative environmental events.
Abstract: In the mammalian brain, dopamine is a critical neuromodulator whose actions underlie learning, decision-making, and behavioral control. Degeneration of dopamine neurons causes Parkinson’s disease, whereas dysregulation of dopamine signaling is believed to contribute to psychiatric conditions such as schizophrenia, addiction, and depression. Experiments in animal models suggest the hypothesis that dopamine release in human striatum encodes reward prediction errors (RPEs) (the difference between actual and expected outcomes) during ongoing decision-making. Blood oxygen level-dependent (BOLD) imaging experiments in humans support the idea that RPEs are tracked in the striatum; however, BOLD measurements cannot be used to infer the action of any one specific neurotransmitter. We monitored dopamine levels with subsecond temporal resolution in humans (n = 17) with Parkinson’s disease while they executed a sequential decision-making task. Participants placed bets and experienced monetary gains or losses. Dopamine fluctuations in the striatum fail to encode RPEs, as anticipated by a large body of work in model organisms. Instead, subsecond dopamine fluctuations encode an integration of RPEs with counterfactual prediction errors, the latter defined by how much better or worse the experienced outcome could have been. How dopamine fluctuations combine the actual and counterfactual is unknown. One possibility is that this process is the normal behavior of reward processing dopamine neurons, which previously had not been tested by experiments in animal models. Alternatively, this superposition of error terms may result from an additional yet-to-be-identified subclass of dopamine neurons.
Abstract: Egalitarian motives form a powerful force in pro- moting prosocial behavior and enabling large-scale cooperation in the human species. At the neural level, there is substantial, albeit correlational, evidence suggesting a link between dopamine and such behavior [2, 3]. However, important questions remain about the specific role of dopamine in setting or modulating behavioral sensitivity to prosocial concerns. Here, using a combination of pharmacological tools and economic games, we provide critical evidence for a causal involvement of dopamine in human egalitarian tendencies. Specifically, using the brain penetrant catechol-O-methyl transferase (COMT) in- hibitor tolcapone, we investigated the causal relationship between dopaminergic mechanisms and two prosocial concerns at the core of a number of widely used economic games: (1) the extent to which individuals directly value the material payoffs of others, i.e., generosity, and (2) the extent to which they are averse to differences between their own payoffs and those of others, i.e., inequity. We found that dopaminergic augmentation via COMT inhibition increased egalitarian tendencies in participants who played an extended version of the dictator game. Strikingly, computational modeling of choice behavior revealed that tolcapone exerted selective effects on inequity aversion, and not on other computational components such as the extent to which indi- viduals directly value the material payoffs of others. Together, these data shed light on the causal relation- ship between neurochemical systems and human prosocial behavior and have potential implications for our understanding of the complex array of social impairments accompanying neuropsychiatric disorders involving dopaminergic dysregulation.
Abstract: Connecting neural mechanisms of behavior to their underlying molecular and genetic substrates has important scientific and clinical implications. However, despite rapid growth in our knowledge of the functions and computational properties of neural circuitry underlying behavior in a number of important domains, there has been much less progress in extending this understanding to their molecular and genetic substrates, even in an age marked by exploding availability of genomic data. Here we describe recent advances in analytical strategies that aim to overcome two important challenges associated with studying the complex relationship between genes and behavior: (i) reducing distal behavioral phenotypes to a set of molecular, physiological, and neural processes that render them closer to the actions of genetic forces, and (ii) striking a balance between the competing demands of discovery and interpretability when dealing with genomic data containing up to millions of markers. Our proposed approach involves linking, on one hand, models of neural computations and circuits hypothesized to underlie behavior, and on the other hand, the set of the genes carrying out biochemical processes related to the functioning of these neural systems. In particular, we focus on the specific example of value-based decision-making, and discuss how such a combination allows researchers to leverage existing biological knowledge at both neural and genetic levels to advance our understanding of the neurogenetic mechanisms underlying behavior.