Prefrontal high gamma during a magnetoencephalographic working memory task.

In human electrophysiology research, the high gamma part of the power spectrum (~>60 Hz) is a relatively new area of investigation. Despite a low signal-to-noise ratio, evidence exists that it contains significant information about activity in local cortical networks. Here, using magnetoencephalography (MEG), we found high gamma activity when comparing data from an n-back working memory task to resting data in a large sample of normal volunteers. Initial analysis of power spectra from 0-back, 2-back, and rest trials showed three frequency bands exhibiting task-related differences: alpha, beta, and high gamma. Unlike alpha and beta, the high gamma spectrum was broad, without a peak at a single frequency. In addition, power in high gamma was highest for the 2-back and lowest during rest, while the opposite pattern occurred in the other bands. Beamformer source localization of each of the three frequency bands revealed a distinct set of sources for high gamma. These included several regions of prefrontal cortex that exhibited greater power when both n-back conditions were compared to rest. A subset of these regions had more power when the 2-back was compared to 0-back, which indicates a role in working memory performance. Our results show that high gamma will be important for understanding cortical processing during cognitive and other tasks. Furthermore, data from human intracortical recordings suggest that high gamma is the aggregate of spiking in local cortical networks, which implies that MEG could serve to bridge experimental modalities by noninvasively observing task-related modulation of spiking rates.

Presynaptic Dopamine Synthesis Capacity in Schizophrenia and Striatal Blood Flow Change During Antipsychotic Treatment and Medication-Free Conditions.

Standard-of-care biological treatment of schizophrenia remains dependent upon antipsychotic medications, which demonstrate D2 receptor affinity and elicit variable, partial clinical responses via neural mechanisms that are not entirely understood. In the striatum, where D2 receptors are abundant, antipsychotic medications may affect neural function in studies of animals, healthy volunteers, and patients, yet the relevance of this to pharmacotherapeutic actions remains unresolved. In this same brain region, some individuals with schizophrenia may demonstrate phenotypes consistent with exaggerated dopaminergic signaling, including alterations in dopamine synthesis capacity; however, the hypothesis that dopamine system characteristics underlie variance in medication-induced regional blood flow changes has not been directly tested. We therefore studied a cohort of 30 individuals with schizophrenia using longitudinal, multi-session [15O]-water and [18F]-FDOPA positron emission tomography to determine striatal blood flow during active atypical antipsychotic medication treatment and after at least 3 weeks of placebo treatment, along with presynaptic dopamine synthesis capacity (ie, DOPA decarboxylase activity). Regional striatal blood flow was significantly higher during active treatment than during the placebo condition. Furthermore, medication-related increases in ventral striatal blood flow were associated with more robust amelioration of excited factor symptoms during active medication and with higher dopamine synthesis capacity. These data indicate that atypical medications enact measureable physiological alterations in limbic striatal circuitry that vary as a function of dopaminergic tone and may have relevance to aspects of therapeutic responses.

Statistical learning is constrained to less abstract patterns in complex sensory input (but not the least).

The influence of statistical information on behavior (either through learning or adaptation) is quickly becoming foundational to many domains of cognitive psychology and cognitive neuroscience, from language comprehension to visual development. We investigate a central problem impacting these diverse fields: when encountering input with rich statistical information, are there any constraints on learning? This paper examines learning outcomes when adult learners are given statistical information across multiple levels of abstraction simultaneously: from abstract, semantic categories of everyday objects to individual viewpoints on these objects. After revealing statistical learning of abstract, semantic categories with scrambled individual exemplars (Exp. 1), participants viewed pictures where the categories as well as the individual objects predicted picture order (e.g., bird1-dog1, bird2-dog2). Our findings suggest that participants preferentially encode the relationships between the individual objects, even in the presence of statistical regularities linking semantic categories (Exps. 2 and 3). In a final experiment we investigate whether learners are biased towards learning object-level regularities or simply construct the most detailed model given the data (and therefore best able to predict the specifics of the upcoming stimulus) by investigating whether participants preferentially learn from the statistical regularities linking individual snapshots of objects or the relationship between the objects themselves (e.g., bird_picture1-dog_picture1, bird_picture2-dog_picture2). We find that participants fail to learn the relationships between individual snapshots, suggesting a bias towards object-level statistical regularities as opposed to merely constructing the most complete model of the input. This work moves beyond the previous existence proofs that statistical learning is possible at both very high and very low levels of abstraction (categories vs. individual objects) and suggests that, at least with the current categories and type of learner, there are biases to pick up on statistical regularities between individual objects even when robust statistical information is present at other levels of abstraction. These findings speak directly to emerging theories about how systems supporting statistical learning and prediction operate in our structure-rich environments. Moreover, the theoretical implications of the current work across multiple domains of study is already clear: statistical learning cannot be assumed to be unconstrained even if statistical learning has previously been established at a given level of abstraction when that information is presented in isolation.