Semantic and phonological task-set priming and stimulus processing investigated using magnetoencephalography (MEG).

In this study the neural substrates of semantic and phonological task priming and task performance were investigated using single word task-primes. Magnetoencephalography (MEG) data were analysed using Synthetic Aperture Magnetometry (SAM) to determine the spatiotemporal and spectral characteristics of cortical responses. Comparisons were made between the task-prime conditions for evidence of differential effects as a function of the nature of the task being primed, and between the task-prime and the task performance responses for evidence of parallels in activation associated with preparation for and completion of a specific task. Differential priming effects were found. Left middle temporal and inferior frontal voxels showed a statistically significant power decrease associated with the semantic task-prime, and a power increase associated with the phonological task-prime, within beta and gamma frequency bands respectively. Similarities between the task-related differential effects associated with task-prime presentation and those associated with target stimulus presentation were also found. For example, within the semantic task condition, left superior frontal and middle temporal regions showed a significant power decrease within both task-prime and target epochs; within the phonological task condition there were significant parietal and cerebellar power decreases within both types of epoch. In addition there was evidence within the priming epochs of dissociable patterns of activity which could be interpreted as indices of de-activation of task-irrelevant networks. Following a phonological task-prime, significant power increases were observed in those inferior frontal and middle temporal regions in which significant power decreases were associated with semantic task priming and performance.

Phase-locked gamma band responses to semantic violation stimuli.

This paper addresses the role of gamma-band activity in semantic networks associated with the processing of words and sentences in humans. Magnetoencephalography (MEG) was used to compare the responses of eleven normal volunteers to semantically congruous and incongruous words at the end of syntactically correct sentences. The averaged low frequency responses evoked by the two word classes were clearly different within the latency range associated with N400 components. The oscillatory characteristics of the evoked responses were analysed using Gabor transform techniques in conjunction with statistical re-sampling. This revealed transient gamma-oscillations ( approximately 35 Hz) that were significantly phase-locked to both types of stimuli but preferentially present at intermediate ( approximately 300 ms) and long (>500 ms) latencies for incongruous words. This stimulus dependent phase locked gamma-activity occurred at latencies that were distinct from the short latency and evoked N400 components. The findings suggest that synchronised oscillations provide independent information about brain dynamics and that some semantic processes may dissociate into distinct functional stages.

Task-dependent early latency (30-60 ms) visual processing of human faces and other objects.

Electrophysiological responses to previously seen faces reportedly differ from those to novel faces at shorter latencies than generally associated with complex visual analysis. It is unclear, however, whether such observations are unique to faces, and which stages of visual processing they reflect. MEG was used in 21 normal adults to record neural responses to images of faces, other objects and abstract patterns presented individually as part of a classification task and in sequential pairs as part of an image comparison task. The amplitudes of the short latency responses (30-60 ms) to the first image in pairs of faces were significantly greater than the responses to both the second faces and the individual face images. These early responses were recorded over predominantly right hemisphere parietal and occipito-temporal cortical regions including areas that, at longer latencies, have been associated with face specific activity. The differences in the responses within pairs were less for non-face objects and absent for abstract geometrical patterns. No early neuronal activity was observed in the classification task. The results indicate the existence of early latency neural networks that are sensitive to both stimulus type and task and are strongly activated by faces.

Analysis of magnetoencephalographic data using the homogeneous sphere model: empirical tests.

The external magnetic fields produced by artificial current dipoles placed in conducting-gel filled spheres, partial spheres and human skulls have been measured. Using a homogeneously conducting sphere model, the dipole parameters have been estimated. Although good quality data fitting is obtainable for a range of positions of the centre of the sphere used in the modelling, the predictions for the dipole obtained with those different assumed positions differ considerably. It is found that the most reliable method of obtaining accurate estimates of the dipole parameters is by taking an average of the predictions of several of the best-fit analyses. The information about the accuracy of the predictions that is obtainable from the fitting routine's residual error parameter is discussed.

Macroscopic ionic currents within the human leg.

Recent research on developing and healing tissues suggests that small quasi-DC ionic currents (of magnitude 10-20 microA) may play a controlling role in the initiation and organisation of growing tissues, but the difficulties of measuring such small currents have led to confusing results. Sensitive magnetometry provides a method of demonstrating and, to some extent, locating such currents. A SQUID magnetometer system has been built and used to investigate the magnetic fields around the uninjured human leg. Analysis of the magnetic fields reveals the presence of slowly changing macroscopic current loops (of magnitude up to 12 microA) within the leg. These currents are broadly similar in all subjects, and show day-to-day reproducibility in individuals. They change predictably with time of muscle relaxation (over an hour), and revert to the original form on muscular exertion. These currents are of significance when considering the therapeutic use of injected current for the healing of non-union in bone.

A distributed quasi-static ionic current source in the 3-4 day old chicken embryo.

We report measurements of slowly varying magnetic field patterns close to fertilized eggs of the chicken Gallus domesticus during the first few days of incubation. These fields are generated by ionic currents within the egg that are associated with the development of the embryo. Since they are very weak (no greater than tens of pT) and vary over distances of a few millimetres, it has been necessary to develop specialized instrumentation and analysis techniques. We describe the use of high-spatial-resolution SQUID magnetometers to measure the field patterns and appropriate imaging algorithms to model the current sources responsible for producing the fields. Our results provide strong evidence for a distributed source in the extra-embryonic membranes. There is also indication of a more localized source within the embryo itself.

Identification of discrete regions of activity using correlation coefficient scanning of distributed current maps.

Biomagnetic measurements of the brain are often analysed in terms of a number of discrete primary generators. In this paper we describe an objective method of identifying the number of approximate location of such generators. The method is based on the matching of an instrument-independent representation of the data with a template whose pattern is characteristic of a localized primary source. The method is shown to be insensitive to severe noise and to be capable of resolving closely spaced generators.

Non-invasive monitoring of ionic current flow during development by SQUID magnetometry.

The ionic currents flowing in developing organisms produce weak magnetic fields that can be detected using SQUID magnetometers. The method is non-invasive and dc recording is possible. To date SQUID magnetometers have mainly been used in human studies. The features of the technique are described and the prospects of extending its use to developmental studies are discussed. Feasible instrumental specifications are indicated. A recent SQUID magnetometer investigation of ionic current flow in the developing chick in ovo is summarised as an illustration of the magnetometer method. The paper as a whole argues that magnetometry is a useful alternative or adjunct to electrode-based experiments on the electrophysiology of developing organisms.

Magnetoencephalographic signals identify stages in real-life decision processes.

We used magnetoencephalography (MEG) to study the dynamics of neural responses in eight subjects engaged in shopping for day-to-day items from supermarket shelves. This behavior not only has personal and economic importance but also provides an example of an experience that is both personal and shared between individuals. The shopping experience enables the exploration of neural mechanisms underlying choice based on complex memories. Choosing among different brands of closely related products activated a robust sequence of signals within the first second after the presentation of the choice images. This sequence engaged first the visual cortex (80-100 ms), then as the images were analyzed, predominantly the left temporal regions (310-340 ms). At longer latency, characteristic neural activation was found in motor speech areas (500-520 ms) for images requiring low salience choices with respect to previous (brand) memory, and in right parietal cortex for high salience choices (850-920 ms). We argue that the neural processes associated with the particular brand-choice stimulus can be separated into identifiable stages through observation of MEG responses and knowledge of functional anatomy.

Neural processing of human faces: a magnetoencephalographic study.

This is a whole head magnetoencephalographic (MEG) study of the neural processing of briefly presented images of human faces in 14 normal subjects. The experiments involved three tasks of increasing complexity, involving image categorisation, image comparison and the identification of emotion. The analyses were based on average responses to repeated stimuli in the different image categories. These averages were processed to give numerical measures of the power within defined regions and latency spans. The only statistically significant difference in these data between the response to faces and other images is in the right occipito-temporal channels at a latency of 140 ms. The face-specific response is largely independent of the task. Source modelling suggests an extended source in the ventral occipito-temporal region. The analysis supports the notions of both face-specificity and right hemisphere dominance for all image types at early latencies.