Syntactic and semantic restrictions on morphological recomposition: MEG evidence from Greek.

Complex morphological processing has been extensively studied in the past decades. However, most of this work has either focused on only certain steps involved in this process, or it has been conducted on a few languages, like English. The purpose of the present study is to investigate the spatiotemporal cortical processing profile of the distinct steps previously reported in the literature, from decomposition to re-composition of morphologically complex items, in a relatively understudied language, Greek. Using magnetoencephalography, we confirm the role of the fusiform gyrus in early, form-based morphological decomposition, we relate the syntactic licensing of stem-suffix combinations to the ventral visual processing stream, somewhat independent from lexical access for the stem, and we further elucidate the role of orbitofrontal regions in semantic composition. Thus, the current study offers the most comprehensive test to date of visual morphological processing and additional, crosslinguistic validation of the steps involved in it.

Morphological representations are extrapolated from morpho-syntactic rules.

The field of psycho- and neuro-linguistics has long-debated the decompositional model of visual word processing: Are written words processed via the visual forms of stem and affix morphemes, or as complex wholes? Although many have now settled upon a decompositional view, it is unclear what heuristic the brain uses to generate these visual morpheme-forms in the first place. Here we conduct a magneto-encephalography study to test two hypotheses for how this may be done: i) the brain encodes representations of the morphemes that follow the morpho-syntactic rules governing constituents: A stem morpheme will be represented if the word obeys the grammatical behaviour associated with its suffix; ii) the brain only encodes stem morphemes that occur with multiple suffixes or as words in isolation. Our results indicate that words with morpho-syntactic wellformedness as stem-suffix combinations are decomposed by the system, thus supporting the former hypothesis. This suggests that knowledge of morpho-syntactic rules can be used to form morphological representations of written words, in absence of independent experience with all of their constituent morphemes. Possible mechanisms supporting this computation are discussed.

Functional characterisation of letter-specific responses in time, space and current polarity using magnetoencephalography.

Recent neurophysiological evidence suggests that a hierarchical neural network of low-to-high level processing subserves written language comprehension. While a considerable amount of research has identified distinct regions and stages of processing, the relations between them and to this hierarchical model remain unclear. Magnetoencephalography (MEG) is a technique frequently employed in such investigations; however, no studies have sought to test whether the conventional method of reconstructing currents at the source of the magnetic field is best suited for such across-subject designs. The present study details the results of three MEG experiments addressing these issues. Neuronal populations supporting responses to low-level orthographic properties were housed posteriorly near the primary visual cortex. More anterior regions along the fusiform gyrus encoded higher-level processes and became active ~80ms later. A functional localiser of these early letter-specific responses was developed for the production of functional regions of interest in future studies. Previously established response components were successfully grouped based on proximity to the localiser, which characterised location, latency and functional sensitivity. Unconventional anatomically constrained signed minimum norm estimates of MEG data were most sensitive to the primary experimental manipulation, suggesting that the conventional unsigned unconstrained method is sub-optimal for studying written word processing.

Non-linear processing of a linear speech stream: The influence of morphological structure on the recognition of spoken Arabic words.

Although the significance of morphological structure is established in visual word processing, its role in auditory processing remains unclear. Using magnetoencephalography we probe the significance of the root morpheme for spoken Arabic words with two experimental manipulations. First we compare a model of auditory processing that calculates probable lexical outcomes based on whole-word competitors, versus a model that only considers the root as relevant to lexical identification. Second, we assess violations to the root-specific Obligatory Contour Principle (OCP), which disallows root-initial consonant gemination. Our results show root prediction to significantly correlate with neural activity in superior temporal regions, independent of predictions based on whole-word competitors. Furthermore, words that violated the OCP constraint were significantly easier to dismiss as valid words than probability-matched counterparts. The findings suggest that lexical auditory processing is dependent upon morphological structure, and that the root forms a principal unit through which spoken words are recognised.

Phase II study of gemcitabine, 5-fluorouracil, and leucovorin in patients with pancreatic cancer.

The primary goal of this phase II study was to determine the efficacy of gemcitabine (Gemzar; Eli Lilly and Company, Indianapolis, IN) plus 5-fluorouracil in patients with pancreatic cancer. Eligibility criteria included nonresectable locally advanced or metastatic pancreatic adenocarcinoma and measurable disease. Gemcitabine at 1,000 mg/m(2) and leucovorin at 20 mg/m(2) were administered intravenously 30 minutes before 5-fluorouracil 600 mg/m(2), weekly for 3 of every 4 weeks. Twenty nine patients were enrolled. The overall response rate was 21% (95% confidence interval: 8% to 40%), consisting of one complete response and five partial responses; 16 patients (55%) had stable disease. Median survival was 8.4 months (95% confidence interval: 2.6 to 14.2), and actuarial 1-year survival was 36%. Neutropenia (grade 3 only) was reported in 3.4% of patients, but was generally of short duration. No thrombocytopenia or evidence of cumulative myelosuppression was observed. The only significant nonhematologic events were grade 3 diarrhea and alopecia (both 3.4%). Gemcitabine plus 5-fluorouracil is active and well tolerated compared with results reported for each of these single agents. Thus, this combination justifies future comparative clinical trials. Semin Oncol 28 (suppl 10):44-49.

Reconstructing spatio-temporal activities of neural sources using an MEG vector beamformer technique.

We have developed a method suitable for reconstructing spatio-temporal activities of neural sources by using magnetoencephalogram (MEG) data. The method extends the adaptive beamformer technique originally proposed by Borgiotti and Kaplan to incorporate the vector beamformer formulation in which a set of three weight vectors are used to detect the source activity in three orthogonal directions. The weight vectors of the vector-extended version of the Borgiotti-Kaplan beamformer are then projected onto the signal subspace of the measurement covariance matrix to obtain the final form of the proposed beamformer's weight vectors. Our numerical experiments show that both spatial resolution and output signal-to-noise ratio of the proposed beamformer are significantly higher than those of the minimum-variance-based vector beamformer used in previous investigations. We also applied the proposed beamformer to two sets of auditory-evoked MEG data, and the results clearly demonstrated the method's capability of reconstructing spatio-temporal activities of neural sources.

A magnetoencephalographic component whose latency reflects lexical frequency.

Magnetoencephalography (MEG) measured subjects' brain responses during a lexical decision task. The words employed come from six frequency categories, which were defined in terms of a linear decrease in log-frequency. Although frequency effects in reaction-time are well-documented in studies of lexical access, a neural component whose latency predicts reaction time has not been discovered. This study identifies an MEG component (the M350) whose latency mirrors the frequency-effect.

A syntactic specialization for Broca's area.

Despite numerous aphasia and functional imaging studies, the exact correlation between cortical language areas and subcomponents of the linguistic system has not been established. Here, we used functional MRI to identify cortical areas specifically involved in syntactic processing. An experimental design contrasted sentences containing grammatical errors with sentences containing spelling errors. The ungrammatical sentences produced more activation in cortical language areas than did the sentences with spelling errors, and the difference in activation was significantly greater in Broca's area than in Wernicke's area or in the angular gyrus/supramarginal gyrus. The present findings provide direct evidence of a syntactic specialization for Broca's area and establish the existence of distinct modules for our knowledge of language.

The selectivity of the occipitotemporal M170 for faces.

Evidence from fMRI, ERPs and intracranial recordings suggests the existence of face-specific mechanisms in the primate occipitotemporal cortex. The present study used a 64-channel MEG system to monitor neural activity while normal subjects viewed a sequence of grayscale photographs of a variety of unfamiliar faces and non-face stimuli. In 14 of 15 subjects, face stimuli evoked a larger response than non-face stimuli at a latency of 160 ms after stimulus onset at bilateral occipitotemporal sensors. Inverted face stimuli elicited responses that were no different in amplitude but 13 ms later in latency than upright faces. The profile of this M170 response across stimulus conditions is largely consistent with prior results using scalp and subdural ERPs.

Auditory cortex accesses phonological categories: an MEG mismatch study.

The studies presented here use an adapted oddball paradigm to show evidence that representations of discrete phonological categories are available to the human auditory cortex. Brain activity was recorded using a 37-channel biomagnetometer while eight subjects listened passively to synthetic speech sounds. In the phonological condition, which contrasted stimuli from an acoustic /dae/-/tae/ continuum, a magnetic mismatch field (MMF) was elicited in a sequence of stimuli in which phonological categories occurred in a many-to-one ratio, but no acoustic many-to-one ratio was present. In order to isolate the contribution of phonological categories to the MMF responses, the acoustic parameter of voice onset time, which distinguished standard and deviant stimuli, was also varied within the standard and deviant categories. No MMF was elicited in the acoustic condition, in which the acoustic distribution of stimuli was identical to the first experiment, but the many-to-one distribution of phonological categories was removed. The design of these studies makes it possible to demonstrate the all-or-nothing property of phonological category membership. This approach contrasts with a number of previous studies of phonetic perception using the mismatch paradigm, which have demonstrated the graded property of enhanced acoustic discrimination at or near phonetic category boundaries.

Neoadjuvant chemotherapy with cisplatin, ifosfamide and 5-fluorouracil in the treatment of locally advanced cervical cancer.

The objective of this study was to evaluate clinical and histological response, resectability, and survival in patients with cervical epidermoid carcinoma stage IB2 to IIIB with the use of neoadjuvant chemotherapy followed by radical surgery and/or radiation therapy. Between September 1989 and February 1996, 53 patients were admitted to this study. They were given three cycles of cisplatin 30 mg/m2/day, 5-fluorouracil 500 mg/m2/day, ifosfamide 2000 mg/m2/day i.v., and mesna 400 mg/m2/day i.v. at hour 0 and 400 mg/m2 at hours 4 and 8 during three days every 21-28 days. We evaluated 47 patients. Global clinical response obtained was 85% {95% (CI), 75-97%, CR in 14 patients (30%) and PR in 26 patients (55%)}. Twenty-three patients underwent surgery. Six patients (13%) had a complete histological response. Median follow-up was 42 months (5-96). In resected patients, with a median follow-up of 57 months (5-96), the estimated five-year disease-free survival was 78%. Global survival estimated to 60 months was 83% for stage IB2, 70% for IIB, and 20% for IIIB. This mode of therapy offers a new option to improve survival in locally advanced cervical cancer. Randomized trials are required in order to establish a definitive role for this therapeutic strategy.

MEG spatio-temporal analysis using a covariance matrix calculated from nonaveraged multiple-epoch data.

We propose a magnetoencephalographic (MEG) spatio-temporal analysis in which the measurement-covariance matrix is calculated using nonaveraged multiple epoch data. The proposed analysis has two advantages. First, a very narrow time window can be used for the source estimation. Second, accurate localization is possible even when the source activation has a time jitter. Experiments using auditory evoked MEG data clearly demonstrate these advantages.

MEG covariance difference analysis: a method to extract target source activities by using task and control measurements.

A method is proposed for extracting target dipolesource activities from two sets of evoked magnetoencephalographic (MEG) data, one measured using task stimuli and the other using control stimuli. The difference matrix between the two covariance matrices obtained from these two measurements is calculated, and a procedure similar to the MEG-multiple signal classification (MUSIC) algorithm is applied to this difference matrix to extract the target dipole-source configuration. This configuration corresponds to the source-configuration difference between the two measurements. Computer simulation verified the validity of the proposed method. The method was applied to actual evoked-field data obtained from simulated task-and-control experiments. In these measurements, a combination of auditory and somatosensory stimuli was used as the task stimulus and the somatosensory stimulus alone was used as the control stimulus. The proposed covariance difference analysis successfully extracted the target auditory source and eliminated the disturbance from the somatosensory sources.