gatsbyproperty.co.uk/meve-mejor-clutchless-transmisin.php Furthermore, previous research has shown differential patterns of brain activity for accurate word retrieval versus error responses Meinzer et al. In this study, both the timing and number of naming exposures was manipulated across critical conditions, allowing comparison of accurately named items that had not previously been facilitated to items that had received one or more earlier naming attempts. It was hypothesized that if the priming mechanisms at work in unimpaired speakers do indeed underlie effective naming treatment in individuals with aphasia, then repetition suppression effects in similar regions for short-term and long-term conditions may be evident across controls and aphasic participants.
It was further hypothesized that, in line with evidence from previous behavioral research in unimpaired speakers and in aphasia Cave, ; Nickels, a , a single prior exposure could effectively facilitate naming performance in participants with aphasia. Ethical approval was obtained from the University of Queensland and all participants gave informed written consent under an approved Medical Research Ethical Review Committee protocol.
Eighteen control participants seven male meeting inclusionary criteria participated in the study. Their average age was Full details of the experiment conducted with controls are reported elsewhere MacDonald et al. Two female participants with aphasia were also recruited. Table 1 provides full demographic, clinical, and assessment battery details, as well as the possible levels of impairment in spoken word production for both participants. Figure 1 displays lesion location, which for both individuals primarily included temporo-parietal regions.
All participants reported English as their first language, were right handed, and had normal or corrected to normal vision. Every participant was screened for cognitive impairment by administration of the Mini-Mental State Examination Folstein et al. Lesion locations. Axial T 1 weighted MRI slices at 5 mm intervals for both participants with aphasia.
TABLE 1. Demographic, clinical, and assessment battery results for participants with aphasia. A large collection of similarly sized gray-scaled picture stimuli was obtained Hemera Photo-Objects, Hemera, Hull, QC, Canada and royalty free digital stock photographs. For controls of these pictures were then divided into nine sets of 20 items, with assignment of one set to the short-term facilitated condition and one set to the long-term facilitated condition counterbalanced across participants.
For participants with aphasia two pre-test sessions were conducted, in which the entire collection was presented at each session in a random order for naming. Based on the results of both pre-test sessions, 60 pictures were then chosen as experimental stimuli and divided into three sets of 20 items. Stimuli for the short-term facilitated and long-term facilitated conditions were sourced from those items participants found difficult to name across both pre-test sessions and therefore represented suitable targets for facilitation. A third set of stimuli formed an unfacilitated condition, chosen from items that were named consistently by participants during pre-test sessions.
It should be noted that P01 was unable to name most items during both pre-tests. However, P02 was more inconsistent, with the majority of items inaccurately named or not responded to in at least one pre-test. All sets were matched on the basis of International Picture Naming Project naming reaction time Szekely et al.
Additionally, control sets were matched for age of acquisition Morrison et al. Following the pre-test sessions, a facilitation phase and an experimental phase were conducted. The facilitation phase required all participants to complete two sessions within 2—3 days, during which only the long-term facilitated set of stimuli was presented, in random order, three times for attempted overt naming six times in total over the two sessions.
A single facilitation trial consisted of a fixation cross 1. No feedback was provided on performance, nor were any cues provided. During the experimental phase, all sets of stimuli were presented for naming during an fMRI scanning session. A single trial lasted The items presented in the previous facilitation sessions were presented again in the scanner to investigate any long-term effects from previous naming attempts.
The short-term facilitation set was presented twice within the scanner: the first presentation and naming attempt served as a potential prime and the second as a target, with an average lag of 7. For all participants, the initial short-term prime presentations represented an unfacilitated condition. Therefore, the three main conditions of interest presented in the scanner for controls were unprimed short-term prime , short-term facilitated short-term target and long-term facilitated long-term target. Refer to Figure 2 for a breakdown of conditions and procedures for participants with aphasia.
During each session a high resolution 1. Images were acquired in 36 axial planes with an in-plane resolution of 3. A behavioral interleaved gradient design was utilized to avoid the artifacts associated with head movement during overt speech, as well as to enable recording of responses and reaction times. To minimize scanner noise during picture presentation and response periods 4. Image acquisition for each trial therefore occurred during the remaining A point-spread function mapping sequence was acquired prior to GE-EPI acquisitions, allowing the distortion in geometry and intensity to be corrected in the time series data.
Target trials that elicited no response or an incorrect response from participants were excluded from the behavioral and imaging analyses. For the two participants with aphasia, the number of correct trials for each condition within the scanner were: P01 — unfacilitated-unnamed 3, short-term 2, long-term 6, unfacilitated-named 13; P02 — unfacilitated-unnamed 7, short-term 8, long-term 13, unfacilitated-named 15 refer to Figure 3.
Behavioral data. Facilitation effects in percentage accuracy data for both participants with aphasia for all conditions. The initial short-term prime presentations within the scanner In-Scan represent the UNF-unnamed condition. Images were processed using statistical parametric mapping software SPM5 1. Spatial smoothing 8 mm full-width half-maximum Gaussian kernel was applied to normalized volumes 3. A general linear model GLM was constructed for the fMRI time series using finite impulse response functions due to the behavioral interleaved design resulting in partial collection of hemodynamic response, with onsets and durations chosen to reflect the expected peak BOLD response.
Single subject whole brain analyses for the participants with aphasia were conducted. Results for control participants are reported separately MacDonald et al. Briefly, the behavioral results showed significant priming effects for both long-term and short-term facilitated conditions. The whole brain results identified a relative decrease of activation for long-term facilitated items when compared to unfacilitated items comparable to the unfacilitated-named condition in participants with aphasia in the bilateral pars triangularis of the inferior frontal gyri IFG , in the bilateral posterior inferior temporal gyri, and in the right precentral gyrus.
Greater activation, however, was found for facilitated items than for unprimed items in several regions: in the bilateral precuneus for long-term facilitated items, and in the right precuneus and the right middle frontal gyrus for short-term facilitated items. Percentage accuracy data for the pre-test sessions and within the scanner are shown in Figure 3 for both participants. A weighted Wilcoxon One-Sample test was used to determine whether accuracy differed from pre-facilitation Pre-Tests 1 and 2 to post-facilitation In-Scan for each condition.
A Mann—Whitney U -test was also subsequently conducted to determine whether the magnitude of change from pre-to post-facilitation differed between specific conditions. However, there were no significant differences for either participant between short-term and unfacilitated-unnamed, or between short-term and long-term conditions. An analysis of reaction times was not conducted for the participants with aphasia. Full whole brain neuroimaging results for participants with aphasia are set out in Supplementary Table S1 , however, we report and discuss only those results for contrasts involving two conditions, the long-term and unfacilitated-named conditions.
This is due to the fact that we did not find significant behavioral results for the short-term or unfacilitated-unnamed conditions and these two conditions also had very few trials in which an accurate response was produced within the scanner. The only significant difference in neural activity was found for P02, with greater activity for long-term facilitated items facilitated and previously difficult to name when compared to unfacilitated-named unfacilitated and previously accurately named items in the left precuneus see Figure 4.
In controls, the same precuneus region was identified for the equivalent contrast. Further, as we found long-term results in the whole brain analyses within the left inferior frontal region for controls, a subsequent bilateral region of interest ROI analysis in this area was conducted for both individuals with aphasia.
However our results make it clear that posterior areas play an essential role in the word retrieval process. However, regardless of the source of facilitation, this finding indicates that some individuals with aphasia may engage highly networked regions to support word production processes. Meinzer, M. Advance article alerts. Stimuli and Procedure A large collection of similarly sized gray-scaled picture stimuli was obtained Hemera Photo-Objects, Hemera, Hull, QC, Canada and royalty free digital stock photographs. However there was no evidence for effects of other epileptic variables age of seizure onset, duration of epilepsy, anti-epileptic drugs , hence this proposal will have to be re-examined in future studies.
Six ROIs were analyzed for each participant, including bilateral pars orbitalis, pars triangularis, and pars opercularis. No significant results were identified for either participant for conditions of interest. Whole brain result for P02 3dClustSim minimum cluster threshold 55 voxels. LT, long-term facilitated and previously difficult to name; UNF-named, unfacilitated and previously accurately named.
A detailed discussion of control results has been reported elsewhere MacDonald et al. The majority of whole brain results were identified within right hemisphere regions, confirming previous findings that the right hemisphere can be heavily involved in normal word production processes Fridriksson et al. Focusing on results in language-related regions: repetition suppression effects for only long-term facilitated items were identified in two bilateral areas previously linked to semantic processing pars triangularis of the inferior frontal gyrus and the posterior inferior temporal gyrus.
However, no facilitatory effects were identified in the short-term condition, despite this condition showing the greatest behavioral priming. It could be that priming of these representations has indeed occurred, but was not identified due to insufficient sensitivity of the imaging paradigm, or that effects only become apparent when strengthened by multiple repetitions of stimuli as in the long-term condition. While falling outside the scope of discussion in MacDonald et al. These findings outside traditional language-related regions could represent a marker of successful episodic recall precuneus; Cavanna and Trimble, ; Cavanna, and attention middle frontal gyrus; Naghavi and Nyberg, In contrast to the behavioral results for the group of control participants, no improvement was identified in picture naming accuracy for the individual participants with aphasia following short-term facilitation.
However, both participants showed improvement in accuracy following multiple repeated attempts at naming several days earlier in the long-term facilitation condition. For P01 there was not a significant difference between the unfacilitated-unnamed control condition and the long-term facilitated condition, and hence we cannot rule out that for P01 the improvement reflected regression to the mean, given that stimuli were purposefully selected from items that were consistently difficult to name at pretest.
For the unfacilitated stimuli that were previously able to be named, as expected under the stressful conditions of the scanner and also consistent with regression to the mean, performance decreased during the scanning session see Figure 3. Consideration of the behavioral results suggests that a single prior attempt at naming by these two individuals with aphasia did not facilitate subsequent naming, or did not facilitate sufficiently to improve accuracy, but that the cumulative effect of multiple attempts was facilitatory.
The only significant neuroimaging result was for P02, with a repetition enhancement effect identified for long-term facilitated items when compared to unfacilitated-named items in the left precuneus. While P01 also showed behavioral benefits for long-term facilitation, no significant neuroimaging results were identified for relevant contrasts, possibly due to a lack of power for this condition. Interestingly, the result for P02 mirrored one of the findings identified for control subjects in the same neural region left precuneus and for the equivalent contrast long-term versus unprimed.
A relative increase in neural activation, or repetition enhancement, is often present when some form of additional processing is required Henson et al. For our two conditions the same task of overt naming was required and, as such, there should not have been additional processing required specifically for long-term items. However, the conditions differed in two respects: facilitation several prior exposures for long-term items and no prior exposure for unfacilitated-named items and previous ability to name difficult to name for long-term and correctly named for unfacilitated-named.
It should be noted that the precuneus is commonly engaged during semantic tasks Binder et al. We suspect that for P02 an increase in activation for long-term items in this region is more likely related to episodic encoding or object recognition systems being enhanced during subsequent naming of facilitated items with multiple prior exposures, rather than any difference in lexical-semantic processing.
Other picture naming studies with aphasic participants have also found increased activity in the precuneus region, with the authors making similar assumptions to the current study with respect to episodic memory mechanisms and further proposing that modulation of activity in regions not traditionally associated with language processing may represent a form of compensatory cortical adaptation Musso et al. In other words, an intact neural region performs strategic compensatory functions enabling completion of, in this case, a linguistic task Grafman, For the long-term condition in controls, we also saw repetition suppression effects in the left IFG pars orbitalis, pars triangularis, and pars opercularis when compared to the unprimed condition.
However, a subsequent ROI analysis in participants with aphasia in this same region did not reveal any significant results, despite the IFG being spared in both participants. Taken together, these findings demonstrate that while some processing associated with successful word retrieval may be similar across control subjects and individuals with aphasia, facilitation is not necessarily tapping into mechanisms within unimpaired or homologous language-related regions.
In this respect our results challenge the assumption that spared left hemisphere language areas and their right hemisphere homologues necessarily support the reorganization of language following stroke Saur et al. The current study could be improved in a number of ways.
Only two individuals with aphasia participated, therefore, it is not possible to make claims that may be generalized to the wider aphasic population. A study with a larger series of people with aphasia, after controlling for such factors as symptoms or lesion characteristics, could utilize a whole brain regression analysis to identify whether variation in behavioral facilitation effects relate to variation in brain activity.
Additionally, a high percentage of incorrect responses were excluded from the neuroimaging analyses due to the methodological decision to source facilitated conditions for the participants with aphasia from difficult to name items. Therefore, some neuroimaging results were based on only a small number of critical trials and could not be considered due to lack of power see Supplementary Table S1. Even bearing in mind the limitations mentioned above, the findings of the current study have clear implications for the clinical treatment of anomia. Both the control group and the participants with aphasia showed long-term behavioral facilitation effects from repeated picture naming.
Although the current study included data from only two individuals with aphasia, our results add further weight to previous evidence that improvement in word retrieval abilities is possible even in chronic aphasia and without direct treatment. However, while the control group showed significant short-term facilitation effects from repeated naming, the aphasic participants did not. In contrast to our original hypothesis, a single prior exposure over a period of minutes did not effectively facilitate naming performance in participants with aphasia.
Longer lasting benefits appeared to require multiple exposures, suggesting that the number of times a particular task, or particular items, are presented may be a predictor of treatment outcome. It should be noted though, that the number of repeated naming attempts in the current study involved only six exposures of the same pictorial stimuli across two sessions spanning several days.
If a small number of prior naming attempts can result in improvements in behavioral performance, then the impact of any assessment tasks administered prior to beginning a formal treatment regime must be considered. Assessment of anomia, particularly within a cognitive neuropsychological approach, often involves repeated use of the same stimuli across tasks, yet it is generally assumed that assessment has little impact on subsequent naming performance.
Our finding of improvement in performance following a limited number of presentations, and in the absence of feedback or correction, emphasizes the power of assessment in a clinical setting, as well as the importance of baselines and control tasks in experimental settings. Moreover, although it is not yet clear that these results would replicate across all people with aphasia, the fact that this task simply required attempts at naming highlights its potential as a straightforward method by which individuals can self-administer treatment.
This study also sought to determine whether any facilitatory effects associated with repeated picture naming act through normal priming mechanisms. We hypothesized that if the mechanisms at work in unimpaired speakers underlie effective anomia treatment, then similar effects would be evident across controls and participants with aphasia in the same language-related neural regions.
As expected, neuroimaging results for the control group did identify repetition suppression effects for long-term facilitated items in several regions known to be involved in language processing, mirroring the behavioral improvement for this condition, and possibly reflecting more efficient lexical-semantic processing. In addition to repetition suppression, a repetition enhancement effect was evident in the bilateral precuneus for long-term facilitated items when compared to unprimed items for controls. Conversely, we did not find any neuroimaging results surviving correction in language-related regions for the participants with aphasia.
The only significant result identified for one of the participants with aphasia was a repetition enhancement effect outside the language network. Interestingly, this result was identified within the same neural region left precuneus and for the equivalent contrast long-term versus unfacilitated-named items as in the control group. In line with previous research investigating the role of this brain region, we conclude that an increase in activation for long-term items in the left precuneus is related to episodic encoding or enhancement of object recognition systems, rather than a change in lexical-semantic processing.
However, regardless of the source of facilitation, this finding indicates that some individuals with aphasia may engage highly networked regions to support word production processes. Brain areas not traditionally associated with language processing may contribute significantly to recovery in anomia, and it is therefore imperative that, in addition to language, clinicians consider other areas of cognition such as memory and attention in the assessment and treatment of word finding difficulties.
Approaches to word retrieval treatment are not well-understood in terms of underlying neurocognitive mechanisms, however, the findings of the current study add to our understanding of the brain-behavior relationship responsible for positive effects during treatment. This translational knowledge may provide clinicians with some insight into how treatment works in the brain, as well as contribute to the provision of more targeted therapy for individuals with aphasia.
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. Table S1 Full whole brain results. Automated anatomical labeling software Tzourio-Mazoyer et al. Abrahams, S. Functional magnetic resonance imaging of verbal fluency and confrontation naming using compressed image acquisition to permit overt responses. Brain Mapp. Baayen, R. Google Scholar. Badre, D. Left ventrolateral prefrontal cortex and the cognitive control of memory.
Neuropsychologia 45, — Barry, C. Age of acquisition, word frequency, and the locus of repetition priming of picture naming. Binder, J. Where is the semantic system? A critical review and meta-analysis of functional neuroimaging studies. Cortex 19, — Human brain language areas identified by functional magnetic resonance imaging. Bookheimer, S. Functional MRI of language: new approaches to understanding the cortical organization of semantic processing.
Cavanna, A. The precuneus and consciousness. CNS Spectr. The precuneus: a review of its functional anatomy and behavioural correlates. Brain , — Cave, C. Very long-lasting priming in picture naming. Cruice, M. Boston naming test results for healthy older Australians: a longitudinal and cross-sectional study. Aphasiology 14, — Damasio, H. The anatomy of phonological and semantic processing in normal subjects.
Renewal of the neurophysiology of language: functional neuroimaging. Eden, G. Utilizing hemodynamic delay and dispersion to detect fMRI signal change without auditory interference: the behavior interleaved gradients technique. Evans, A.
Folstein, M. Mini-mental state: a practical method for grading the cognitive state of patients for the clinician. Freire, L. What is the best similarity measure for motion correction in fMRI time series? IEEE Trans. Fridriksson, J. Cortical mapping of naming errors in aphasia.
Age-related variability in cortical activity during language processing. Speech Lang. Hear Res. Neural recruitment associated with anomia treatment in aphasia. Neuroimage 32, — Neural correlates of phonological and semantic-based anomia treatment in aphasia. Goodglass, H. Goodglass and A. The main goal of this study was to determine the processing level impaired in naming difficulties in pharmaco-resistant TLE and its possible anatomo-functional substrate.
A standard Oldfield test indicated that 34 patients were right-handed, six left-handed and three ambidextrous Oldfield, A total of 18 patients including all left-handed and ambidextrous patients underwent fMRI to determine hemispheric language dominance. When doubt regarding language dominance persisted three patients , a Wada test was performed. Forty-one patients in the study were left-hemisphere dominant, while two patients with atypical language lateralization were excluded. Neurological examinations were normal. The epilepsy diagnosis was made during pre-surgical assessment for drug-resistant partial epilepsy on the basis of seizure semiology, ictal and interictal EEG recordings as well as structural and functional neuro-imaging MRI and 18FDG PET, respectively.
Video-electroencephalographic EEG recording, performed to analyse usual seizures and interictal EEG, was used to determine the side and extent of the epileptogenic zone. All patients had temporal epilepsy in the broad sense of the term. MRI was normal in 15 patients, and showed hippocampal sclerosis HS in 17 patients, and other temporal lobe lesion s in 8 patients [dysplasia 6 , cavernoma 1 , oligodendroglioma 1 ]. Patients were not subjected to any invasive procedure, were fully informed about the study and gave their written consent.
Two naming tests were administered to all patients: picture naming and auditory definition naming. Each item's name was characterized by its lexical frequency and length in syllables. For auditory definition naming, we adapted the test devised by Hamberger and Seidel in which patients are required to name items in response to definitions e. We constructed 80 such definitions whose responses were different from those of the picture naming test above, but matched for various relevant variables same distribution of man-made and natural objects, paired lexical frequency and syllabic length.
The terms used in the definitions were expected to be understood by the patients. The two naming tests were conducted in two independent blocks. The pictures were printed on paper, one item per sheet.
The definitions were spoken aloud by the experimenter to the patient. Participants were asked to name aloud the object depicted, or referred to, using a single word. Response times were measured with one second accuracy, from the onset of picture presentation in the visual task, and from the end of the definition in the auditory task. A response was considered to be correct when it coincided with the name of the picture given in Lexis, or to the name of the definition in the pilot study conducted to obtain normative data.
Trials in which the patient failed to provide any answer before the time limit were given special attention. Once the time limit was expired, an attempt was made to differentiate recognition failure from word retrieval failure.
To do so, the availability of semantic information regarding the target item was tested. Patients were asked to use gestures or circumlocutions to describe the meaning of the word they were searching. The latter, interpreted as genuine word retrieval failures, were the critical anomic states.
Anomic states were further explored by testing the availability of linguistic information with standard methods previously used in their investigation. In case the patient failed to provide the correct responses, the examiner provided a phonemic cue initial phoneme. A correct response at that point was considered to result from phonemic cueing. Other erroneous trials included phonetic paraphasias when the target word was produced but inappropriately articulated and semantic paraphasias when a word semantically related to the target was produced, irrespective of how correctly it was articulated.
Patients rested during the min uptake period. The 24 rings of the PET system allow 47 images to be obtained, with a 3. Attenuation-corrected images were reconstructed using the ordered subsets expectation maximization algorithm, with 5 iterations and 32 subsets. Ipsilateral 18FDG-PET metabolism of medial and lateral temporal regions was visually described by the same experienced nuclear medicine physician EG for each patient as normal, moderate or severe.
Naming performance at the trial level i. We constructed two models. First, we started by estimating the effects of the major predictors: Laterality of epilepsy left TLE versus right TLE and Task picture versus definition , and their interaction. This led us to restrict further analysis to the sub-population of left TLE patients see details below. In the second restricted analysis, the model comprised three types of predictors. We tested effects of i demographic of the patients; ii linguistic properties of the stimuli; and iii performance in related non-linguistic neuropsychological tasks.
Firstly we conducted a logistic linear model similar to those reported above, with epilepsy and neuroimaging features as predictors of the rate of anomic states. The data were standardized onto the Montreal Neurological Institute atlas MNI by using a parameter affine transformation, followed by non-linear transformations and a trilinear interpolation. Global normalization was performed using proportional scaling. Spearman's correlation was used to assess the relationship between the occurrence of anomic states and brain PET metabolism of each extracted cluster.
In addition, to confirm specific correlation with anomic state, Spearman's correlations between brain PET metabolism of the same extracted cluster and language word fluency test, token test, syllable deletion task, word and sentence repetition task or memory working memory and general memory tasks were performed.
A summary is provided in Fig. In trials where there was an anomic state, patients were able to provide the correct gender This global pattern is typical of anomic states in aphasics and tip-of-the-tongue states in healthy speakers. A Naming performance in the picture and definition tasks. B and C Variability of word finding difficulty in both tasks. This rate was considerably lower for right TLE patients C.
The data set comprised the performance of all patients in both naming tasks, yielding a total of trials a few definitions could not be tested on all patients. Given our focus on anomic states, we excluded phonetic and semantic paraphasias from the analysis, as well as trials in which participants reported not recognizing or understanding the stimulus or the intended target trials, 4.
This left trials from 41 patients. Performance was poorer for left TLE than for right TLE patients, and poorer for the definition than for the picture task in the whole population. We found no evidence for an interaction between the two factors. This suggests that the difference between the definition and the picture tasks was comparable across the two populations of left TLE and right TLE patients.
Left TLE patients produced a high and variable number of tip-of-the-tongue states, both in the picture and the definition tasks Fig. This rate was considerably lower for right TLE patients, and hence no further attempt was made to understand their origin. Model 1 showed strong effects of laterality and task. Performance was worse for left TLE than right TLE patients, and it was worst for the definition than for the picture task. Model 2 left TLE patients showed an effect of the gender males tended to produce fewer anomic states than females , an effect of lexical frequency low lexical frequency word tended to produce fewer anomic states than high lexical frequency and an effect of IQ.
Model 3 left TLE patients showed the effect of hypometabolism in the lateral areas the rate of anomic states was greater when lateral hypometabolism was more severe , a positive effect of type of epilepsy anomic states were more common in medio-lateral epilepsy than in medial epilepsy and an effect of the frequency of the seizures an increase in frequency lead to an increase in rate of anomic states.
Similar effects were observed when the model was computed over 27 patients, including those for which IQ measure was not available. We observed a significant effect of gender males tended to produce fewer anomic states than females. When psycholinguistic variables characterizing the target items were tested, we observed a robust effect of lexical frequency.
As shown in Fig. Finally, we considered the role of IQ and memory performance, in an attempt to capture variability associated with different levels of baseline performance in neuropsychological testing. For this analysis, detailed neuropsychological assessment was only available for 23 of the 27 left TLE patients.
Tip-of-the-tongue states were evenly distributed across these two categories. The analysis was conducted on trials from 25 patients. We also observed a marginally significant effect of seizure frequency computed in seizure per month on a logarithmic scale ; an increase in frequency lead to an increase in rate of anomic states.
Finally, we found a marginally significant positive effect of type of epilepsy; anomic states were more common in medio-lateral epilepsy than in medial epilepsy.
This analysis provides a first indication of an involvement of lateral areas in the occurrence of anomic states. This indication comes from the observation that the rate of anomic states increased in patients with lateral hypometabolism, and in medio-lateral epilepsy patients compared to medial epilepsy. These suggestive effects were observed while the effect of seizure frequency was controlled.
To further ascertain these findings, we conducted the next analysis, in which we sought to identify correlations between brain 18FDG-PET metabolism and the rate of occurrence of anomic states in an unconstrained whole-brain analysis. P -values are not corrected for multi-comparison for whole-brain analysis.
In addition, no lesion was located within this cluster. In addition, the metabolism of the left temporo-parietal cluster was not correlated with scores in other language tasks such as sentence and repetition tasks, word fluency test or syllable deletion task, even when the analysis was restricted to left TLE patients.
The present study provides evidence that many cases of word finding difficulty in TLE can be considered as anomic states. This deficit is related to the lateralization of epilepsy, as anomic states are considerably more frequent in left than in right TLE. Furthermore, this impairment is strongly correlated with resting-state metabolism of the postero-infero-temporo-basal region BA , of the posterior part of GTS BA39 and inferior parietal lobule BA We discuss the behavioural and neural aspects of these findings in turn. However, clearly distinguishing between different types of deficit in particular, word retrieval failure versus impaired concept retrieval is difficult, as a global naming score e.
In light of the difference between picture and definition naming scores, Hamberger and Seidel suggested an impairment at the level of the semantic system. In the current study, we identified trials in which word finding difficulties could be unambiguously established, based on the model of anomic or tip-of-the-tongue states. We selected trials in which participants i provided evidence of having understood the concept that the picture or definition referred to; and ii were not able to produce the corresponding name. In addition, during many of these trials, patients were able to provide reliable lexical information in terms of the grammatical gender of the missing word and benefited from phonological cues.
We showed that rate of anomic states was sensitive to the laterality of epilepsy and the task. Left TLE patients presented more lexical access difficulty than right TLE patients in both modalities, with worse performance in the definition than in the picture task without interaction between these two factors. This result is in contrast with a previous report which only showed difference in the score of the definition task, but not in the picture task between left and right TLE Hamberger and Tamny, ; Hamberger and Seidel, This discrepancy could be explained by our analysis at the single trial level, which evidenced a difference between left and right TLE patients in the picture task that was not observed in previous studies.
This being said, as shown by Hamberger and colleagues and the present data, the definition task is indeed more sensitive than the picture task. Anomic states are considerably more frequent in left than in right TLE. Furthermore, no difference in performance in other language tasks such as comprehension tasks, token test, repetition tasks and syllable deletion task was observed. This suggests that the left TLE deficit is relatively circumscribed to the ability to retrieve words from memory, not a general language deficit. In addition, the anomic state was further characterized by considering possible contributions of psycholinguistic and neuropsychological factors.
This is a standard observation in healthy and impaired language performance Nickels and Howard, In contrast, the analysis provides no evidence for an effect of object properties man-made or not. This provides a further argument in favour of deficit at the lexical rather than semantic level of processing. Among neuropsychological factors, no significant effect was observed for variables measuring immediate memory and working memory, yet a significant effect of IQ was observed.
This result is probably the sum of several facts: i assessment of mental ability depends upon multiple cognitive processes and one explanation could be that poor lexical access promotes lower IQ; and ii severe pharmaco-resistant epilepsy could impair lexical acquisition and other general learning leading to poor IQ and greater rate of anomic states. A recent review of longitudinal studies has suggested the existence of a relationship between seizure frequency and changes in IQ or memory function Dodrill, A marginal effect of lexical access difficulty with increased seizure frequency was observed in our study; it might be hypothesized that higher seizure frequency could induce a larger functional language deficit due to post-ictal disturbance.
However there was no evidence for effects of other epileptic variables age of seizure onset, duration of epilepsy, anti-epileptic drugs , hence this proposal will have to be re-examined in future studies. We found no evidence for effects of hippocampal sclerosis or medial hypometabolism in anomic states. However, this study looked only at medial temporal structures and did not consider the possible role of other temporal regions. Overall, the question of the role of mesial temporal region in naming thus remains open.
In our study, the rate of anomic states increased in patients with lateral hypometabolism as measured using PET, or in those patients defined as having medio-lateral epilepsy as opposed to purely medial epilepsy. This result indicates an involvement of lateral areas in the occurrence of anomic states. To further ascertain these findings, the anatomo-functional substrate of lexical access difficulty was investigated on the basis of unconstrained whole-brain metabolism evaluated with 18FDG-PET. The rate of tip-of-the-tongue states showed a correlation with brain metabolism mainly in inferior and posterior temporal regions BA and to a much lesser degree in the parietal region BA This finding was confirmed both within the subgroup of left TLE and within the subgroup of patients with severe anomia.
To our knowledge this provides the first evidence of a correlation between verbal processing dysfunction in TLE and brain metabolism. Our observations of a negative correlation between temporal lobe metabolism and the rate of occurrence of anomic state are consistent with a growing body of literature demonstrating a relationship between lexical access and the left infero-posterior temporal region. Evidence regarding language processing in TLE has also come from post-surgical data and intracranial stimulation.
Postoperative naming deficits have been observed after anterior temporal lobectomy. Such a result has prompted the conclusion that the anterior part of the temporal lobe plays a critical role in the occurrence of word finding difficulties. Numerous intra-cranial stimulation data have demonstrated a widespread temporal area associated with speech arrest and naming failure.
Recent studies of intra-cranial stimulation have sought to specify these conclusions by investigating sites that compromise word production in the picture task, the definition task, or both. Because the definition task was shown to be more sensitive than the picture task, the role of the anterior temporal lobe was particularly highlighted in these studies. However our results make it clear that posterior areas play an essential role in the word retrieval process.
Interestingly, in the present study, the correlation between rate of tip-of-the-tongue states and PET brain metabolism was also observed in the inferior-basal temporal region. Despite the fact that these studies did not characterize naming difficulties in detail, the conclusions reached provide additional support for the role of posterior temporal structures in lexical retrieval.
We propose, therefore, that lexical retrieval in TLE is associated with a temporal lobe network, within which the left posterior and basal areas play a crucial role. The distinction between semantic and lexical processes is of course somewhat artificial as they are so closely connected in the course of error free speech production. Yet the study of certain pathological states e. This proposal is in contrast to those of previous studies, which insisted on the involvement of anterior sites in TLE in producing word finding difficulty while our results evidence the importance of posterior sites.
This observation might be tentatively explained by a functional deafferentation between anterior and neocortical posterior temporal structures due to epileptic disorganization. Similarly, the naming failures that follow from anterior lobe resection could be due to a disconnection of the network linking semantic and lexical processing.
Finally, the current data provide no indication that the word retrieval system in TLE should basically differ from that observed in studies of normal subjects or patients with ischemic stroke. The present study reports an investigation of word finding difficulties present in left TLE patients irrespective of input modality picture or definition tasks. The results support the idea that this deficit, similar to anomic or tip-of-the tongue states, concerns the stage of lexical access. A significant correlation between the rate of occurrence of anomic states and resting brain metabolism was observed in inferior and posterior temporal regions BA Such correlation highlights the importance of posterior regions in the specific process of word retrieval in TLE, as part of a large antero-posterior-basal temporal network.
Supplementary material is available at Brain online. The authors thank Dr Fabrice Bartolomei, Dr Martine Gavaret and Dr Maxime Guye for clinical and electrophysiological assessment of studied patients, Kristell Blaise and Albane Dejax for the language assessment of studied patients, Dr Aileen McGonigal and anonymous reviewers for discussions and helpful comments. Oxford University Press is a department of the University of Oxford.
It furthers the University's objective of excellence in research, scholarship, and education by publishing worldwide.