Intersubject variability in functional neuroanatomy of silent verb generation: assessment by a new activation detection algorithm based on amplitude and size information

Testing for signals with unknown location and scale in a χ2random field, with an application to fMRI

Siegmund and Worsley (1995) considered the problem of testing for signals with unknown location and scale in a Gaussian random field defined on ℝN. The test statistic was the maximum of a Gaussian random field in anN+1 dimensional ‘scale space’,Ndimensions for location and 1 dimension for the scale of a smoothing filter. Scale space is identical to a continuous wavelet transform with a kernel smoother as the wavelet, though the emphasis here is on signal detection rather than image compression or enhancement. Two methods were used to derive an approximate null distribution forN=2 andN=3: one based on the method of volumes of tubes, the other based on the expected Euler characteristic of the excursion set. The purpose of this paper is two-fold: to show how the latter method can be extended to higher dimensions, and to apply this more general result to χ2fields. The result of Siegmund and Worsley (1995) then follows as a special case. In this paper the results are applied to the problem of searching for activation in brain images obtained by functional magnetic resonance imaging (fMRI).

Testing for signals with unknown location and scale in a χ2 random field, with an application to fMRI

Siegmund and Worsley (1995) considered the problem of testing for signals with unknown location and scale in a Gaussian random field defined on ℝN. The test statistic was the maximum of a Gaussian random field in anN+1 dimensional ‘scale space’,Ndimensions for location and 1 dimension for the scale of a smoothing filter. Scale space is identical to a continuous wavelet transform with a kernel smoother as the wavelet, though the emphasis here is on signal detection rather than image compression or enhancement. Two methods were used to derive an approximate null distribution forN=2 andN=3: one based on the method of volumes of tubes, the other based on the expected Euler characteristic of the excursion set. The purpose of this paper is two-fold: to show how the latter method can be extended to higher dimensions, and to apply this more general result to χ2fields. The result of Siegmund and Worsley (1995) then follows as a special case. In this paper the results are applied to the problem of searching for activation in brain images obtained by functional magnetic resonance imaging (fMRI).

Characteristics of canonical intrinsic connectivity networks across tasks and monozygotic twin pairs

Intrinsic connectivity networks (ICNs) are becoming more prominent in the analyses of in vivo brain activity as the field of neurometrics has revealed their importance for augmenting traditional cognitive neuroscience approaches. Consequently, tools that assess the coherence, or connectivity, and morphology of ICNs are being developed to support inferences and assumptions about the dynamics of the brain. Recently, we reported trait-like profiles of ICNs showing reliability over time and reproducibility across different contexts. This study further examined the trait-like and familial nature of ICNs by utilizing two divergent task paradigms in twins. The study aimed to identify stable network phenotypes that exhibited sensitivity to individual differences and external perturbations in task demands. Analogous ICNs were detected in each task and these ICNs showed consistency in morphology and intranetwork coherence across tasks, whereas the ICN timecourse dynamics showed sensitivity to task demands. Specifically, the timecourse of an arm/hand sensorimotor network showed the strongest correlation with the timeline of a hand imitation task, and the timecourse of a language-processing network showed the strongest temporal association with a verb generation task. The area V1/simple visual stimuli network exhibited the most consistency in morphology, coherence, and timecourse dynamics within and across tasks. Similarly, this network exhibited familiality in all three domains as well. Hence, this experiment is a proof of principle that the morphology and coherence of ICNs can be consistent both within and across tasks, that ICN timecourses can be differentially and meaningfully modulated by a task, and that these domains can exhibit familiality.

Variability of fMRI-response patterns at different spatial observation scales

Functional organization units of the cerebral cortex exist over a wide range of spatial scales, from local circuits to entire cortical areas. In the last decades, scale-space representations of neuroimaging data suited to probe the multi-scale nature of cortical functional organization have been introduced and methodologically elaborated. For this purpose, responses are statistically detected over a range of spatial scales using a family of Gaussian filters, with small filters being related to fine and large filters-to coarse spatial scales. The goal of the present study was to investigate the degree of variability of fMRI-response patterns over a broad range of observation scales. To this aim, the same fMRI data set obtained from 18 subjects during a visuomotor task was analyzed with a range of filters from 4- to 16-mm full width at half-maximum (FWHM). We found substantial observation-scale-related variability. For example, using filter widths of 6- to 8-mm FWHM, in the group-level results, significant responses in the right secondary visual but not in the primary visual cortex were detected. However, when larger filters were used, the responses in the right primary visual cortex reached significance. Often, responses in probabilistically defined areas were significant when both small and large filters, but not intermediate filter widths were applied. This suggests that brain responses can be organized in local clusters of multiple distinct activation foci. Our findings illustrate the potential of multi-scale fMRI analysis to reveal novel features in the spatial organization of human brain responses.

Left hemisphere lateralization for language in right-handers is controlled in part by familial sinistrality, manual preference strength, and head size

We investigated the effects of familial sinistrality (FS+; presence of left-handedness in one's close relatives), manual preference strength (MPS), and head size on the hemispheric lateralization of language in right-handers. Functional magnetic resonance imaging was used to map 49 individuals while listening to a story in their mother tongue. We found that individuals who had both the FS+ trait and weak MPS had no left hemisphere dominance for this lexicosyntactic task, whereas others showed a leftward functional asymmetry. In addition, the smaller the brain size, the smaller the leftward asymmetry for language, independent of FS and MPS. None of these effects were observed when the same subjects performed a spatial attention task that elicited right hemispheric functional asymmetry. These results demonstrate that the left hemisphere dominance for language in right-handers is a variable controlled, in part, by a number of specific factors, including FS, MPS, and head size.

Imaging of language-related brain regions in detoxified alcoholics

BACKGROUND

Neuroimaging studies showed clear evidence of alcoholism-related damage to the frontal lobes and cerebellum. Although these regions have been involved in language processing, language skills are relatively spared in alcoholics. Here, we aimed at identifying neural substrates associated with the preserved mechanisms of language processing in alcoholics. We hypothesized that alcoholics would show a different pattern of neural activity compared with the controls.

METHODS

Alcoholic and nonalcoholic subjects performed an auditory language task while receiving a functional magnetic resonance imaging (fMRI) scan in a 1.5 T magnet. This task has been previously shown to solicit the comprehension processing in healthy controls, with reliable fMRI response in the left frontal and temporal/parietal lobes.

RESULTS

Behavioral results showed comparable performance (error rates, response time) between the alcoholics and the matched controls. However, analysis of the functional data revealed that the alcoholics exhibited greater fMRI response in the left middle frontal gyrus (pars triangularis), the right superior frontal gyrus, and the cerebellar vermis relative to the controls.

CONCLUSIONS

These findings suggest that frontocerebellar neural activity, supporting the comprehension processing of the auditory language task, may require compensatory mechanisms in alcoholics in order to maintain the same level of performance as the controls.

Processing nouns and verbs in the left frontal cortex: a transcranial magnetic stimulation study

Neuropsychological and neurophysiological studies suggest that the production of verbs in speech depends on cortical regions in the left frontal lobe. However, the precise topography of these regions, and their functional roles in verb production, remains matters of debate. In an earlier study with repetitive transcranial magnetic stimulation (rTMS), we showed that stimulation to the left anterior midfrontal gyrus disrupted verb production, but not noun production, in a task that required subjects to perform simple morphological alternations. This result raises a number of questions: for example, is the effect of stimulation focal and specific to that brain region? Is the behavioral effect limited to rule-based, regular transformations, or can it be generalized over the grammatical category? In the present study, we used rTMS to suppress the excitability of distinct parts of the left prefrontal cortex to assess their role in producing regular and irregular verbs compared to nouns. We compared rTMS to sham stimulation and to stimulation of homologous areas in the right hemisphere. Response latencies increased for verbs, but were unaffected for nouns, following stimulation to the left anterior midfrontal gyrus. No significant interference specific for verbs resulted after stimulation to two other areas in the left frontal lobe, the posterior midfrontal gyrus and Broca's area. These results therefore reinforce the idea that the left anterior midfrontal cortex is critical for processing verbs. Moreover, none of the regions stimulated was preferentially engaged in the production of regular or irregular inflection, raising questions about the role of the frontal lobes in processing inflectional morphology.

Hemispheric specialization for language: Brain volume matters

Increasing brain volume may impose constraints, through longer information transfer delays, on the distributed networks supporting language. Here, we assessed the relative effects of brain volume and other putative predictors of the functional variability of perisylvian language areas, as probed with PET, during both a language comprehension and a language production task. In the case of language comprehension (story listening), a linear combination of planum temporale surface, brain volume and handedness could explain almost 60% of the functional asymmetry observed in the perisylvian area. Without brain volume, the goodness of fit was significantly decreased (39%, P < 0.05), and furthermore, the effect of handedness was not detected anymore. This was due to the fact that in our sample, left-handers (n = 12) had a significantly larger brain volume as compared to right-handers (n = 8, P = 0.03). As for language production (verb generation), brain volume and the planum temporale also played a role. However, in this case, the main predictor of functional variability was handedness, where a greater degree of right-handedness was associated with larger activation of left inferior frontal regions. Depending on the language component of interest, these results support different (yet compatible) theories on hemispheric specialization. Left specialization for comprehension could be attributed to the constraints of processing speech stimuli, while a gestural origin of language is mostly supported by the relation we observed between left specialization for production and right-handedness.

Dissociating neural correlates for nouns and verbs

Dissociations in the ability to produce words of different grammatical categories are well established in neuropsychology but have not been corroborated fully with evidence from brain imaging. Here we report on a PET study designed to reveal the anatomical correlates of grammatical processes involving nouns and verbs. German-speaking subjects were asked to produce either plural and singular nouns, or first-person plural and singular verbs. Verbs, relative to nouns, activated a left frontal cortical network, while the opposite contrast (nouns-verbs) showed greater activation in temporal regions bilaterally. Similar patterns emerged when subjects performed the task with pseudowords used as nouns or as verbs. These results converge with findings from lesion studies and suggest that grammatical category is an important dimension of organization for knowledge of language in the brain.

The spatial and temporal signatures of word production components

This paper presents the results of a comprehensive meta-analysis of the relevant imaging literature on word production (82 experiments). In addition to the spatial overlap of activated regions, we also analyzed the available data on the time course of activations. The analysis specified regions and time windows of activation for the core processes of word production: lexical selection, phonological code retrieval, syllabification, and phonetic/articulatory preparation. A comparison of the word production results with studies on auditory word/non-word perception and reading showed that the time course of activations in word production is, on the whole, compatible with the temporal constraints that perception processes impose on the production processes they affect in picture/word interference paradigms.

Interindividual variability in the hemispheric organization for speech

A PET activation study was designed to investigate hemispheric specialization during speech comprehension and production in right- and left-handed subjects. Normalized regional cerebral blood flow (NrCBF) was repeatedly monitored while subjects either listened to factual stories (Story) or covertly generated verbs semantically related to heard nouns (Gener), using silent resting (Rest) as a common control condition. NrCBF variations in each task, as compared to Rest, as well as functional asymmetry indices (FAI = right minus left NrCBF variations), were computed in anatomical regions of interest (AROIs) defined on the single-subject MNI template. FAIs were predominantly leftward in all regions during both tasks, although larger FAIs were observed during Gener. Subjects were declared "typical" for language hemispheric specialization based on the presence of significant leftward asymmetries (FAI < 0) in the pars triangularis and opercularis of the inferior frontal gyrus during Gener, and in the middle and inferior temporal AROIs during Story. Six subjects (including five LH) showed an atypical language representation. Among them, one presented a right hemisphere specialization during both tasks, another a shift in hemispheric specialization from production to comprehension (left during Gener, right during Story). The group of 14 typical subjects showed significant positive correlation between homologous left and right AROIs NrCBF variations in temporal areas during Story, and in temporal and inferior frontal areas during Gener, almost all regions presenting a leftward FAI. Such correlations were also present in deactivated areas with strong leftward asymmetry (supramarginalis gyrus, inferior parietal region). These results suggest that entry into a language task translates into a hemispheric reconfiguration of lateral cortical areas with global NrCBF increase in the dominant hemisphere and decrease in the minor hemisphere. This can be considered as the setting up of a "language mode", under the control of a mechanism that operates at a perisylvian level. On top of this global organization, regional variations carry on the performance of the cognitive operations specific to the language task to be performed. Hemispheric relationships could be different in atypical subjects, with either between task hemispheric regulation differences or differences in regional specialization.

Aspirin or amiloride for cerebral perfusion defects in cocaine dependence

Cocaine dependent (CD) patients have regional cerebral blood flow (rCBF) deficits that may be related to occlusion of blood vessels by vasoconstriction and abnormal platelet aggregation. This study determined whether aspirin, which reverses platelet aggregation, or amiloride, a vasodilator, significantly reversed this rCBF hypoperfusion. This 1-month randomized trial compared clusters of voxels with significant hypoperfusion in recently abstinent CD patients after aspirin (325 mg daily), amiloride (10 mg daily) or placebo treatment. Forty-nine primary CD patients and 18 non-drug abusing controls were compared using single photon emission computed tomography (SPECT) neuroimaging with 99mTc-hexamethyl-propyleneamine-oxime and statistical parametric mapping (SPM). Platelet aggregation to adenosine diphosphate (ADP) was examined after treatment to determine whether rCBF improvement was related to decreased platelet aggregation. Following treatment, areas of hypoperfusion were improved with amiloride, unchanged with aspirin, and worsened with placebo in comparison to baseline levels. Platelet aggregation after ADP showed no significant change during the month, but reduced rCBF significantly improved after 1-month treatment with amiloride compared with placebo and cocaine abstinence alone.

A common language network for comprehension and production: a contribution to the definition of language epicenters with PET

In this paper, we report on a PET activation study designed to assess whether functional neuroimaging would help to uncover essential language areas in normal volunteers and to provide a more accurate definition of their localization. Regional cerebral blood flow was repeatedly monitored in eight right-handed male volunteers, while performing a language comprehension task (listening to factual stories) and a language production task (covert generation of verbs semantically related to heard nouns), using silent resting as a control condition. The conjunction analysis, conducted with SPM, was used to uncover the network of activations common to both task that included three left hemisphere areas, namely (1) the pars opercularis and triangularis of the inferior frontal gyrus, (2) the posterior part of the superior temporal cortex centered around the superior temporal sulcus, extending to the planum temporale posterior part but sparing the supramarginalis and angular gyri, and (3) the most anterior part of the left inferior temporal gyrus at the junction with the anterior fusiform gyrus. The inferior and lateral parts of the right cerebellar cortex were also included in the conjunction network. Each of the three cortical areas, when they are site of lesion or electrical stimulation, elicit impairment in both language comprehension and production and can thus be considered as essential to language. Accordingly, the present results provide conservative anatomofunctional definitions of the Broca, Wernicke, and basal language areas. Interestingly, contralateral homologues of Broca's and Wernicke's areas also lighted up in the conjunction analysis that could be related to the interindividual variability of hemispheric language dominance.

[A comparison between group-average and individual evoked potential analysis]

PURPOSE

Event-related potentials (ERPs) studies in human subjects have shown inter-individual response variations, probably linked to anatomical and functional brain disparities. The present study was conducted to compare the results obtained by a standard grand-average method and a single subject analysis of VEPs to faces.

MATERIAL AND METHOD

Fifty-eight channel ERPs (analysis time: 1,024 ms) were recorded in 13 normal volunteers during gender or familiarity judgements on unknown and known faces, as well as on a control task using meaningless patterns. Data were then submitted to individual and group averages.

RESULTS AND CONCLUSION

Three activities were identified by both procedures: a P1/N1 complex, a vertex positive potential (P2 or VPP) associated with a temporal negativity, and a N2 negativity. These peaks displayed a marked inter-individual topographical variability. Regarding the outcome of statistical analyses, a certain number of differences were found: on P1, in which individual analyses revealed a strong effect of experimental conditions, while the grand-average method did not; on VPP, in which grand-average analyses suggested an interaction between experimental conditions, face familiarity and cerebral lateralization, while individual analyses did not; and on N2, in which grand-average data showed a clear lateralization effect, while individual analyses did not. A P3 component (Pz, 250 ms) was also defined in grand-average data, but could not be clearly described in individual data. Statistical analyses on this P3 component were thus only performed on group data and revealed a right lateralization and an interaction between face familiarity and experimental conditions. These findings confirmed the existence of a marked topographical variability of ERPs to face and, therefore, question the validity of grand-average studies. Moreover, these results suggest a better efficiency of individual analyses for studying short and middle-latency peaks, while grand-averages appear to be better suited for studying late components.

Statistical limitations in functional neuroimaging. II. Signal detection and statistical inference

The field of functional neuroimaging (FNI) methodology has developed into a mature but evolving area of knowledge and its applications have been extensive. A general problem in the analysis of FNI data is finding a signal embedded in noise. This is sometimes called signal detection. Signal detection theory focuses in general on issues relating to the optimization of conditions for separating the signal from noise. When methods from probability theory and mathematical statistics are directly applied in this procedure it is also called statistical inference. In this paper we briefly discuss some aspects of signal detection theory relevant to FNI and, in addition, some common approaches to statistical inference used in FNI. Low-pass filtering in relation to functional-anatomical variability and some effects of filtering on signal detection of interest to FNI are discussed. Also, some general aspects of hypothesis testing and statistical inference are discussed. This includes the need for characterizing the signal in data when the null hypothesis is rejected, the problem of multiple comparisons that is central to FNI data analysis, omnibus tests and some issues related to statistical power in the context of FNI. In turn, random field, scale space, non-parametric and Monte Carlo approaches are reviewed, representing the most common approaches to statistical inference used in FNI. Complementary to these issues an overview and discussion of non-inferential descriptive methods, common statistical models and the problem of model selection is given in a companion paper. In general, model selection is an important prelude to subsequent statistical inference. The emphasis in both papers is on the assumptions and inherent limitations of the methods presented. Most of the methods described here generally serve their purposes well when the inherent assumptions and limitations are taken into account. Significant differences in results between different methods are most apparent in extreme parameter ranges, for example at low effective degrees of freedom or at small spatial autocorrelation. In such situations or in situations when assumptions and approximations are seriously violated it is of central importance to choose the most suitable method in order to obtain valid results.

A probabilistic ribbon model for shape analysis of the cerebral sulci: application to the central sulcus

PURPOSE

An approach for quantifying the shapes of the cerebral sulci is presented, utilizing a probabilistic geometric model, and it is applied to the central sulcus.

METHOD

The geometric structure of the central sulcus is determined from a set of outlines on cross-sectional images and is used by a procedure that automatically labels the major crest lines, i.e., curves of locally maximal curvature, along the sulcus. An automated procedure then determines a parametric representation of the central sulcus that is consistent across individuals, in that it assigns the same parametric coordinates to corresponding regions of the sulcus.

RESULTS

The method is applied to the central sulci from 20 subjects. The use of this shape representation in cortical morphometric analysis applications is demonstrated, in particular in obtaining local depth and curvature measurements of a sulcus as well as in determining average shapes and variability.

CONCLUSION

With this method, we were able to build parametric representations of the sulcal ribbons by preserving anatomical homologies.

Assumptions and validations of statistical tests for functional neuroimaging

We contrast two statistical methods: three-dimensional cluster analysis and statistical parametric mapping. We show that three-dimensional cluster analysis is based on a neurobiological theory of the regulation of blood flow and, unlike statistical parametric mapping, carries a minimum of assumptions that are tested. Statistical parametric mapping is a formal approach, which is based on a multitude of assumptions of which the majority have not been validated. We also demonstrate that in practice three-dimensional cluster analysis has a reasonable balance between sensitivity and the probability of false positives, giving high reproducibility with data on e.g. colour discrimination.

Functional anatomy of a prelearned sequence of horizontal saccades in humans

We have used positron emission tomography (PET) to study the functional anatomy of the repetition of a prelearned sequence of horizontal saccadic eye movements. Five subjects had to memorize a sequence of six successive horizontal saccades. The subjects were scanned in total darkness under three different conditions: at rest, during the execution of self-paced horizontal saccades, and while repeating a prelearned saccades sequence. The repetition of the prelearned saccades sequence led to specific normalized regional cerebral blood flow (NrCBF) increases at the depth of the superior frontal sulcus as well as at the rostral part of the supplementary motor area, whereas at the parietal level an important activation was observed in the intraparietal sulcus extending up to the precuneus. In addition, it was noticed that compared with the resting control condition, both oculomotor tasks activated a common set of cortical and subcortical areas. At the cortical level, this network was composed of the frontal eye fields, the supplementary eye fields, the median part of the cingulate gyrus, and the insula. At the subcortical level, the lenticular nucleus and the thalamus as well as the cerebellar vermis were activated consistently. A direct comparison of our results with those of other PET studies on spatial vision suggest that the dorsal visuospatial pathway could be extended toward the frontal premotor region. In such a scheme, visuospatial information computed in the intraparietal sulcus would be transmitted to the frontal premotor cortex to optimize a spatial-oriented behavior. This is consistent with the early proposal that perceptual and intentional components of spatial information are mediated through superior parietal and frontal areas, respectively.