Bruno B. Averbeck

Neural correlations, population coding and computation

How the brain encodes information in population activity, and how it combines and manipulates that activity as it carries out computations, are questions that lie at the heart of systems neuroscience. During the past decade, with the advent of multi-electrode recording and improved theoretical models, these questions have begun to yield answers. However, a complete understanding of neuronal variability, and, in particular, how it affects population codes, is missing. This is because variability in the brain is typically correlated, and although the exact effects of these correlations are not known, it is known that they can be large. Here, we review studies that address the interaction between neuronal noise and population codes, and discuss their implications for population coding in general.

Parallel processing of serial movements in prefrontal cortex.

A key idea in Lashleys formulation of the problem of serial order in behavior is the postulated neural representation of all serial elements before the action begins. We studied this question by recording the activity of individual neurons simultaneously in small ensembles in prefrontal cortex while monkeys copied geometrical shapes shown on a screen. Monkeys drew the shapes as sequences of movement segments, and these segments were associated with distinct patterns of neuronal ensemble activity. Here we show that these patterns were present during the time preceding the actual drawing. The rank of the strength of representation of a segment in the neuronal population during this time, as assessed by discriminant analysis, predicted the serial position of the segment in the motor sequence. An analysis of errors in copying and their neural correlates supplied additional evidence for this code and provided a neural basis for Lashleys hypothesis that errors in motor sequences would be most likely to occur when executing elements that had prior representations of nearly equal strength.

Coding and transmission of information by neural ensembles

The brain processes information about sensory stimuli and motor intentions using a massive ensemble of neurons arrayed in parallel. Individual neurons receive convergent inputs from thousands of other neurons, leading to the possibility that patterns of spikes across the input neurons might be crucial components of the neural code. Recently, advances in multielectrode recording techniques have allowed several laboratories to investigate the nature of the interactions between neurons, and their potential role in information coding. Several recent studies have found that the amount of information coded by correlated activity about sensory and motor variables is small, casting doubt on the hypothesis that correlations between pairs of neurons are important for information coding. However, other studies have documented the appearance of coherent oscillations, during particular task epochs and conditions that require selective processing of sensory information, supporting the hypothesis that coherent oscillations between neurons might reflect the dynamic flow of information in the brain.

The Primate Cortical Auditory System and Neural Representation of Conspecific Vocalizations

Over the past decade, renewed interest in the auditory system has resulted in a surge of anatomical and physiological research in the primate auditory cortex and its targets. Anatomical studies have delineated multiple areas in and around primary auditory cortex and demonstrated connectivity among these areas, as well as between these areas and the rest of the cortex, including prefrontal cortex. Physiological recordings of auditory neurons have found that species-specific vocalizations are useful in probing the selectivity and potential functions of acoustic neurons. A number of cortical regions contain neurons that are robustly responsive to vocalizations, and some auditory responsive neurons show more selectivity for vocalizations than for other complex sounds. Demonstration of selectivity for vocalizations has prompted the question of which features are encoded by higher-order auditory neurons. Results based on detailed studies of the structure of these vocalizations, as well as the tuning and information-coding properties of neurons sensitive to these vocalizations, have begun to provide answers to this question. In future studies, these and other methods may help to define the way in which cells, ensembles, and brain regions process communication sounds. Moreover, the discovery that several nonprimary auditory cortical regions may be multisensory and responsive to vocalizations with corresponding facial gestures may change the way in which we view the processing of communication information by the auditory system.

CSF and blood oxytocin concentration changes following intranasal delivery in macaque.

Oxytocin (OT) in the central nervous system (CNS) influences social cognition and behavior, making it a candidate for treating clinical disorders such as schizophrenia and autism. Intranasal administration has been proposed as a possible route of delivery to the CNS for molecules like OT. While intranasal administration of OT influences social cognition and behavior, it is not well established whether this is an effective means for delivering OT to CNS targets. We administered OT or its vehicle (saline) to 15 primates (Macaca mulatta), using either intranasal spray or a nebulizer, and measured OT concentration changes in the cerebral spinal fluid (CSF) and in blood. All subjects received both delivery methods and both drug conditions. Baseline samples of blood and CSF were taken immediately before drug administration. Blood was collected every 10 minutes after administration for 40 minutes and CSF was collected once post-delivery, at the 40 minutes time point. We found that intranasal administration of exogenous OT increased concentrations in both CSF and plasma compared to saline. Both delivery methods resulted in similar elevations of OT concentration in CSF, while the changes in plasma OT concentration were greater after nasal spray compared to nebulizer. In conclusion our study provides evidence that both nebulizer and nasal spray OT administration can elevate CSF OT levels.

Resonance in subthalamo-cortical circuits in Parkinson's disease

Neuronal activity within and across the cortex and basal ganglia is pathologically synchronized, particularly at ∼ 20 Hz in patients with Parkinsons disease. Defining how activities in spatially distributed brain regions overtly synchronize in narrow frequency bands is critical for understanding disease processes like Parkinsons disease. To address this, we studied cortical responses to electrical stimulation of the subthalamic nucleus (STN) at various frequencies between 5 and 30 Hz in two cohorts of eight patients with Parkinsons disease from two different surgical centres. We found that evoked activity consisted of a series of diminishing waves with a peak latency of 21 ms for the first wave in the series. The cortical evoked potentials (cEPs) averaged in each group were well fitted by a damped oscillator function (r ≥0.9, P < 0.00001). Fits suggested that the natural frequency of the subthalamo-cortical circuit was around 20 Hz. When the system was forced at this frequency by stimulation of the STN at 20 Hz, the undamped amplitude of the modelled cortical response increased relative to that with 5 Hz stimulation in both groups (P ≤ 0.005), consistent with resonance. Restoration of dopaminergic input by treatment with levodopa increased the damping of oscillatory activity (as measured by the modelled damping factor) in both patient groups (P ≤0.001). The increased damping would tend to limit resonance, as confirmed in simulations. Our results show that the basal ganglia–cortical network involving the STN has a tendency to resonate at ∼ 20 Hz in Parkinsonian patients. This resonance phenomenon may underlie the propagation and amplification of activities synchronized around this frequency. Crucially, dopamine acts to increase damping and thereby limit resonance in this basal ganglia–cortical network.

The statistical neuroanatomy of frontal networks in the macaque

We were interested in gaining insight into the functional properties of frontal networks based upon their anatomical inputs. We took a neuroinformatics approach, carrying out maximum likelihood hierarchical cluster analysis on 25 frontal cortical areas based upon their anatomical connections, with 68 input areas representing exterosensory, chemosensory, motor, limbic, and other frontal inputs. The analysis revealed a set of statistically robust clusters. We used these clusters to divide the frontal areas into 5 groups, including ventral-lateral, ventral-medial, dorsal-medial, dorsal-lateral, and caudal-orbital groups. Each of these groups was defined by a unique set of inputs. This organization provides insight into the differential roles of each group of areas and suggests a gradient by which orbital and ventral-medial areas may be responsible for decision-making processes based on emotion and primary reinforcers, and lateral frontal areas are more involved in integrating affective and rational information into a common framework.

Rapid sequences of population activity patterns dynamically encode task-critical spatial information in parietal cortex

We characterized the temporal dynamics of population activity in parietal cortex of monkeys as they solved a spatial cognitive problem posed by an object construction task. We applied pattern classification techniques to characterize patterns of activity coding object-centered side, a task-defined variable specifying whether an object component was located on the left or right side of a reference object, regardless of its retinocentric position. During a period in which the value of object-centered side, as defined by task events, remained constant, parietal cortex represented this variable using a dynamic neural code by activating neurons with the same spatial preference in rapid succession so that the pattern of active neurons changed dramatically while the spatial information they collectively encoded remained stable. Furthermore, if the neurons shared the same spatial preference, then their pretrial activity (measured before objects were shown) was correlated to a degree that scaled as a positive linear function of how close together in time the neurons would be activated later in the trial. Finally, we found that while parietal cortex represented task-critical spatial information using a dynamic neural code, it simultaneously represented task-irrelevant spatial information using a stationary neural code. These data demonstrate that dynamic spatial representations exist in parietal cortex, provide novel insight into the synaptic mechanisms that generate them, and suggest they may preferentially encode task-critical spatial information.

Oxytocin decreases aversion to angry faces in an associative learning task.

Social and financial considerations are often integrated when real life decisions are made, and recent studies have provided evidence that similar brain networks are engaged when either social or financial information is integrated. Other studies, however, have suggested that the neuropeptide oxytocin can specifically affect social behaviors, which would suggest separable mechanisms at the pharmacological level. Thus, we examined the hypothesis that oxytocin would specifically affect social and not financial information in a decision making task, in which participants learned which of the two faces, one smiling and the other angry or sad, was most often being rewarded. We found that oxytocin specifically decreased aversion to angry faces, without affecting integration of positive or negative financial feedback or choices related to happy vs sad faces.

Risk and learning in impulsive and nonimpulsive patients with Parkinson's disease.

Relatively little is known about the interaction between behavioral changes, medication, and cognitive function in Parkinsons disease (PD). We examined working memory, learning and risk aversion in PD patients with and without impulsive or compulsive behavior (ICB) and compared the results with those in a group of age‐matched control subjects. Parkinson patients with PD+ICB had poorer working memory performance than either controls or PD patients without ICB. PD+ICB patients also showed decreased learning from negative feedback and increased learning from positive feedback in off compared with on dopaminergic medication. This interaction between medication status and learning was the opposite of that found in the PD patients without a diagnosis of ICB. Finally, the PD group showed increased risk preference on medication relative to controls, and the subgroup of PD+ICB patients with pathological gambling were overall more risk prone than the PD group. Thus, medication status and an impulsive behavioral diagnosis differentially affect several behaviors in PD.