Kevin Kahn

Performing behavioral tasks in subjects with intracranial electrodes.

Patients having stereo-electroencephalography (SEEG) electrode, subdural grid or depth electrode implants have a multitude of electrodes implanted in different areas of their brain for the localization of their seizure focus and eloquent areas. After implantation, the patient must remain in the hospital until the pathological area of brain is found and possibly resected. During this time, these patients offer a unique opportunity to the research community because any number of behavioral paradigms can be performed to uncover the neural correlates that guide behavior. Here we present a method for recording brain activity from intracranial implants as subjects perform a behavioral task designed to assess decision-making and reward encoding. All electrophysiological data from the intracranial electrodes are recorded during the behavioral task, allowing for the examination of the many brain areas involved in a single function at time scales relevant to behavior. Moreover, and unlike animal studies, human patients can learn a wide variety of behavioral tasks quickly, allowing for the ability to perform more than one task in the same subject or for performing controls. Despite the many advantages of this technique for understanding human brain function, there are also methodological limitations that we discuss, including environmental factors, analgesic effects, time constraints and recordings from diseased tissue. This method may be easily implemented by any institution that performs intracranial assessments; providing the opportunity to directly examine human brain function during behavior.

Lucky Rhythms in Orbitofrontal Cortex Bias Gambling Decisions in Humans

It is well established that emotions influence our decisions, yet the neural basis of this biasing effect is not well understood. Here we directly recorded local field potentials from the OrbitoFrontal Cortex (OFC) in five human subjects performing a financial decision-making task. We observed a striking increase in gamma-band (36–50 Hz) oscillatory activity that reflected subjects’ decisions to make riskier choices. Additionally, these gamma rhythms were linked back to mismatched expectations or “luck” occurring in past trials. Specifically, when a subject expected to win but lost, the trial was defined as “unlucky” and when the subject expected to lose but won, the trial was defined as “lucky”. Finally, a fading memory model of luck correlated to an objective measure of emotion, heart rate variability. Our findings suggest OFC may play a pivotal role in processing a subject’s internal (emotional) state during financial decision-making, a particularly interesting result in light of the more recent “cognitive map” theory of OFC function.

Neuron selection for decoding dexterous finger movements

Many brain machine interfaces (BMI) seek to use the activity from hundreds of simultaneously recorded neurons to reconstruct an individuals kinematics. However, many of these neurons are not task related since there is no way to surgically target those neurons. This causes model based decoding to suffer easily from over-fitting on noisy unrelated neurons. Previous methods, such as correlation analysis and sensitivity analysis, seek to select neurons based on which reduced order model best matches the ensemble model and thus does not worry about over fitting. To address this issue, this paper presents a new method, cross model validation, that ranks neuron importance on the neuron models ability to generalize well to data from correct movements and poorly to data from incorrect movements. This method attempts to highlight the neurons that are able to distinguish between movements the best and decode accurately. Selecting neurons using cross model validation scores as opposed to randomly selecting them can increase decoding accuracy up to 2.5 times or by 44%. These results showcase the importance of neuron selection in decoding and the ability of cross model validation in discerning each neurons utility in decoding.

The Role of Associative Cortices and Hippocampus during Movement Perturbations

Although motor control has been extensively studied, most research involving neural recordings has focused on primary motor cortex, pre-motor cortex, supplementary motor area, and cerebellum. These regions are involved during normal movements, however, associative cortices and hippocampus are also likely involved during perturbed movements as one must detect the unexpected disturbance, inhibit the previous motor plan, and create a new plan to compensate. Minimal data is available on these brain regions during such “robust” movements. Here, epileptic patients implanted with intracerebral electrodes performed reaching movements while experiencing occasional unexpected force perturbations allowing study of the fronto-parietal, limbic and hippocampal network at unprecedented high spatial, and temporal scales. Areas including orbitofrontal cortex (OFC) and hippocampus showed increased activation during perturbed trials. These results, coupled with a visual novelty control task, suggest the hippocampal MTL-P300 novelty response is modality independent, and that the OFC is involved in modifying motor plans during robust movement.

The precuneus may encode irrationality in human gambling

Humans often make irrational decisions, especially psychiatric patients who have dysfunctional cognitive and emotional circuitry. Understanding the neural basis of decision-making is therefore essential towards patient management, yet current studies suffer from several limitations. Functional magnetic resonance imaging (fMRI) studies in humans have dominated decision-making neuroscience, but have poor temporal resolution and the blood oxygenation level-dependent signal is only a proxy for neural activity. On the other hand, lesion studies in humans used to infer functionality in decision-making lack characterization of neural activity altogether. Using a combination of local field potential recordings in human subjects performing a financial decision-making task, spectral analyses, and non-parametric cluster statistics, we analyzed the activity in the precuneus. In nine subjects, the neural activity modulated significantly between rational and irrational trials in the precuneus (p <; 0.001). In particular, high-frequency activity (70-100 Hz) increased when irrational decisions were made. Although preliminary, these results suggest suppression of gamma rhythms via electrical stimulation in the precuneus as a therapeutic intervention for pathological decision-making.

Winning versus losing during gambling and its neural correlates

Humans often make decisions which maximize an internal utility function. For example, humans often maximize their expected reward when gambling and this is considered as a “rational” decision. However, humans tend to change their betting strategies depending on how they “feel”. If someone has experienced a losing streak, they may “feel” that they are more likely to win on the next hand even though the odds of the game have not changed. That is, their decisions are driven by their emotional state. In this paper, we investigate how the human brain responds to wins and losses during gambling. Using a combination of local field potential recordings in human subjects performing a financial decision-making task, spectral analyses, and non-parametric cluster statistics, we investigated whether neural responses in different cognitive and limbic brain areas differ between wins and losses after decisions are made. In eleven subjects, the neural activity modulated significantly between win and loss trials in one brain region: the anterior insula (p = 0.01). In particular, gamma activity (30-70 Hz) increased in the anterior insula when subjects just realized that they won. Modulation of metabolic activity in the anterior insula has been observed previously in functional magnetic resonance imaging studies during decision making and when emotions are elicited. However, our study is able to characterize temporal dynamics of electrical activity in this brain region at the millisecond resolution while decisions are made and after outcomes are revealed.

High frequency activity correlates of robust movement in humans

The neural circuitry underlying fast robust human motor control is not well understood. In this study we record neural activity from multiple stereotactic encephalograph (SEEG) depth electrodes in a human subject while he/she performs a center-out reaching task holding a robotic manipulandum that occasionally introduces an interfering force field. Collecting neural data from humans during motor tasks is rare, and SEEG provides an unusual opportunity to examine neural correlates of movement at a millisecond time scale in multiple brain regions. Time-frequency analysis shows that high frequency activity (50-150 Hz) increases significantly in the left precuneus and left hippocampus when the subject is compensating for a perturbation to their movement. These increases in activity occur with different durations indicating differing roles in the motor control process.

Erratum: Lucky Rhythms in Orbitofrontal Cortex Bias Gambling Decisions in Humans

Scientific Reports 6: Article number: 36206; published online: 10 November 2016; updated: 16 February 2017 Affiliation 1 was incorrectly listed as ‘Department of Biomedical Engineering, Johns Hopkins University, Baltimore, MD 21211, USA’ in the original version of the Article. The correct affiliation is listed below:

Oscillations in human orbitofrontal cortex during even chance gambling

Evaluating value and risk as well as comparing expected and actual outcomes is the crux of decision making and reinforcement based learning. In this study, we record from stereotactic electroencephalograph depth electrodes in a human subject in numerous areas in the brain. We focus on the lateral and medial orbitofrontal cortex while they perform a gambling task involving betting on a high card. Preliminary time-frequency analysis shows modulations in the 5-15Hz band that is well synced to the different events of the task. These oscillations increase in both high betting scenarios as well as in losing scenarios though their effects cannot be decoupled. However, the activity between lateral and medial orbitofrontal cortex is a lot more homogenous than previously seen. Additionally, the timing of some of these oscillations occurs before even a response in the visual cortex. This evidence hints that these areas encode priors that influence our decision in future statistically ambiguous scenarios.