Per B. Sederberg

Theta and Gamma Oscillations during Encoding Predict Subsequent Recall

Electrophysiological and hemodynamic measures of human brain activity have been shown to distinguish between episodes of encoding items that are later recalled versus those that are not recalled (Paller and Wagner, 2002). Using intracranial recordings from 793 widespread cortical and subcortical sites in 10 epileptic patients undergoing invasive monitoring, we compared oscillatory power at frequencies ranging from 2 to 64 Hz as participants studied lists of common nouns. Significant increases in oscillatory power during encoding predicted subsequent recall, with this effect predominantly in the 4-8 Hz (theta) and 28-64 Hz (gamma) frequency bands. Sites exhibiting increased theta activity during successful encoding were clustered in right temporal and frontal cortex, whereas those exhibiting increased gamma activity appeared bilaterally at widespread cortical locations. These findings implicate theta and gamma oscillatory activity, across a widespread network of cortical regions, in the formation of new episodic memories.

PyMVPA: a Python Toolbox for Multivariate Pattern Analysis of fMRI Data

Decoding patterns of neural activity onto cognitive states is one of the central goals of functional brain imaging. Standard univariate fMRI analysis methods, which correlate cognitive and perceptual function with the blood oxygenation-level dependent (BOLD) signal, have proven successful in identifying anatomical regions based on signal increases during cognitive and perceptual tasks. Recently, researchers have begun to explore new multivariate techniques that have proven to be more flexible, more reliable, and more sensitive than standard univariate analysis. Drawing on the field of statistical learning theory, these new classifier-based analysis techniques possess explanatory power that could provide new insights into the functional properties of the brain. However, unlike the wealth of software packages for univariate analyses, there are few packages that facilitate multivariate pattern classification analyses of fMRI data. Here we introduce a Python-based, cross-platform, and open-source software toolbox, called PyMVPA, for the application of classifier-based analysis techniques to fMRI datasets. PyMVPA makes use of Python’s ability to access libraries written in a large variety of programming languages and computing environments to interface with the wealth of existing machine learning packages. We present the framework in this paper and provide illustrative examples on its usage, features, and programmability.

A context-based theory of recency and contiguity in free recall

The authors present a new model of free recall on the basis of M. W. Howard and M. J. Kahanas temporal context model and M. Usher and J. L. McClellands leaky-accumulator decision model. In this model, contextual drift gives rise to both short-term and long-term recency effects, and contextual retrieval gives rise to short-term and long-term contiguity effects. Recall decisions are controlled by a race between competitive leaky accumulators. The model captures the dynamics of immediate, delayed, and continual distractor free recall, demonstrating that dissociations between short- and long-term recency can naturally arise from a model in which an internal contextual state is used as the sole cue for retrieval across time scales.

Restoring cortical control of functional movement in a human with quadriplegia

Millions of people worldwide suffer from diseases that lead to paralysis through disruption of signal pathways between the brain and the muscles. Neuroprosthetic devices are designed to restore lost function and could be used to form an electronic ‘neural bypass’ to circumvent disconnected pathways in the nervous system. It has previously been shown that intracortically recorded signals can be decoded to extract information related to motion, allowing non-human primates and paralysed humans to control computers and robotic arms through imagined movements. In non-human primates, these types of signal have also been used to drive activation of chemically paralysed arm muscles. Here we show that intracortically recorded signals can be linked in real-time to muscle activation to restore movement in a paralysed human. We used a chronically implanted intracortical microelectrode array to record multiunit activity from the motor cortex in a study participant with quadriplegia from cervical spinal cord injury. We applied machine-learning algorithms to decode the neuronal activity and control activation of the participant’s forearm muscles through a custom-built high-resolution neuromuscular electrical stimulation system. The system provided isolated finger movements and the participant achieved continuous cortical control of six different wrist and hand motions. Furthermore, he was able to use the system to complete functional tasks relevant to daily living. Clinical assessment showed that, when using the system, his motor impairment improved from the fifth to the sixth cervical (C5–C6) to the seventh cervical to first thoracic (C7–T1) level unilaterally, conferring on him the critical abilities to grasp, manipulate, and release objects. This is the first demonstration to our knowledge of successful control of muscle activation using intracortically recorded signals in a paralysed human. These results have significant implications in advancing neuroprosthetic technology for people worldwide living with the effects of paralysis.

Gamma Oscillations Distinguish True From False Memories

To test whether distinct patterns of electrophysiological activity prior to a response can distinguish true from false memories, we analyzed intracranial electroencephalographic recordings while 52 patients undergoing treatment for epilepsy performed a verbal free-recall task. These analyses revealed that the same pattern of gamma-band (28–100 Hz) oscillatory activity that predicts successful memory formation at item encoding—increased gamma power in the hippocampus, prefrontal cortex, and left temporal lobe—reemerges at retrieval to distinguish correct from incorrect responses. The timing of these oscillatory effects suggests that self-cued memory retrieval begins in the hippocampus and then spreads to the cortex. Thus, retrieval of true, as compared with false, memories induces a distinct pattern of gamma oscillations, possibly reflecting recollection of contextual information associated with past experience.

Comparison of spectral analysis methods for characterizing brain oscillations

Spectral analysis methods are now routinely used in electrophysiological studies of human and animal cognition. Although a wide variety of spectral methods has been used, the ways in which these methods differ are not generally understood. Here we use simulation methods to characterize the similarities and differences between three spectral analysis methods: wavelets, multitapers and P(episode). P(episode) is a novel method that quantifies the fraction of time that oscillations exceed amplitude and duration thresholds. We show that wavelets and P(episode) used side-by-side helps to disentangle length and amplitude of a signal. P(episode) is especially sensitive to fluctuations around its thresholds, puts frequencies on a more equal footing, and is sensitive to long but low-amplitude signals. In contrast, multitaper methods are less sensitive to weak signals, but are very frequency-specific. If frequency specificity is not essential, then wavelets and P(episode) are recommended.

Oscillatory correlates of the primacy effect in episodic memory

Both intracranial and scalp EEG studies have demonstrated that oscillatory activity, especially in the gamma band (28 to 100 Hz), can differentiate successful and unsuccessful episodic encoding [Sederberg, P.B., Kahana, M.J., Howard, M.W., Donner, E.J., Madsen, J.R., 2003. Theta and gamma oscillations during encoding predict subsequent recall. Journal of Neuroscience, 23(34), 10809-10814; Fell, J., Klaver, P., Lehnertz, K., Grunwald, T., Schaller, C., Elger, C.E., Fernandez, G., 2001. Human memory formation is accompanied by rhinal-hippocampal coupling and decoupling. Nature Neuroscience, 4 (12), 1259-1264; Gruber, T., Tsivilis, D., Montaldi, D., and Müller, M. (2004). Induced gamma band responses: An early marker of memory encoding and retrieval. Neuroreport, 15, 1837-1841; Summerfield, C., Mangels, J.A., in press. Dissociable neural mechanisms for encoding predictable and unpredictable events. Journal of Cognitive Neuroscience]. Although the probability of recalling an item varies as a function of where it appeared in the list, the relation between the oscillatory dynamics of successful encoding and serial position remains unexplored. We recorded scalp EEG as participants studied lists of common nouns in a delayed free-recall task. Because early list items were recalled better than items from later serial positions (the primacy effect), we analyzed encoding-related changes in 2 to 100 Hz oscillatory power as a function of serial position. Increases in gamma power in posterior regions predicted successful encoding at early serial positions; widespread low-frequency (4-14 Hz) power decreases predicted successful memory formation for later serial positions. These results suggest that items in early serial positions receive an encoding boost due to focused encoding without having to divide resources among numerous list items. Later in the list, as memory load increases, encoding is divided between multiple items.

PyMVPA: A Unifying Approach to the Analysis of Neuroscientific Data

The Python programming language is steadily increasing in popularity as the language of choice for scientific computing. The ability of this scripting environment to access a huge code base in various languages, combined with its syntactical simplicity, make it the ideal tool for implementing and sharing ideas among scientists from numerous fields and with heterogeneous methodological backgrounds. The recent rise of reciprocal interest between the machine learning (ML) and neuroscience communities is an example of the desire for an inter-disciplinary transfer of computational methods that can benefit from a Python-based framework. For many years, a large fraction of both research communities have addressed, almost independently, very high-dimensional problems with almost completely non-overlapping methods. However, a number of recently published studies that applied ML methods to neuroscience research questions attracted a lot of attention from researchers from both fields, as well as the general public, and showed that this approach can provide novel and fruitful insights into the functioning of the brain. In this article we show how PyMVPA, a specialized Python framework for machine learning based data analysis, can help to facilitate this inter-disciplinary technology transfer by providing a single interface to a wide array of machine learning libraries and neural data-processing methods. We demonstrate the general applicability and power of PyMVPA via analyses of a number of neural data modalities, including fMRI, EEG, MEG, and extracellular recordings.

A Bayesian framework for simultaneously modeling neural and behavioral data

Scientists who study cognition infer underlying processes either by observing behavior (e.g., response times, percentage correct) or by observing neural activity (e.g., the BOLD response). These two types of observations have traditionally supported two separate lines of study. The first is led by cognitive modelers, who rely on behavior alone to support their computational theories. The second is led by cognitive neuroimagers, who rely on statistical models to link patterns of neural activity to experimental manipulations, often without any attempt to make a direct connection to an explicit computational theory. Here we present a flexible Bayesian framework for combining neural and cognitive models. Joining neuroimaging and computational modeling in a single hierarchical framework allows the neural data to influence the parameters of the cognitive model and allows behavioral data, even in the absence of neural data, to constrain the neural model. Critically, our Bayesian approach can reveal interactions between behavioral and neural parameters, and hence between neural activity and cognitive mechanisms. We demonstrate the utility of our approach with applications to simulated fMRI data with a recognition model and to diffusion-weighted imaging data with a response time model of perceptual choice.

A method for efficiently sampling from distributions with correlated dimensions.

Bayesian estimation has played a pivotal role in the understanding of individual differences. However, for many models in psychology, Bayesian estimation of model parameters can be difficult. One reason for this difficulty is that conventional sampling algorithms, such as Markov chain Monte Carlo (MCMC), can be inefficient and impractical when little is known about the target distribution--particularly the target distributions covariance structure. In this article, we highlight some reasons for this inefficiency and advocate the use of a population MCMC algorithm, called differential evolution Markov chain Monte Carlo (DE-MCMC), as a means of efficient proposal generation. We demonstrate in a simulation study that the performance of the DE-MCMC algorithm is unaffected by the correlation of the target distribution, whereas conventional MCMC performs substantially worse as the correlation increases. We then show that the DE-MCMC algorithm can be used to efficiently fit a hierarchical version of the linear ballistic accumulator model to response time data, which has proven to be a difficult task when conventional MCMC is used.