The role of segmentation in phonological processing: an fMRI investigation

A cross-linguistic FMRI study of spectral and temporal cues underlying phonological processing

It remains a matter of controversy precisely what kind of neural mechanisms underlie functional asymmetries in speech processing. Whereas some studies support speech-specific circuits, others suggest that lateralization is dictated by relative computational demands of complex auditory signals in the spectral or time domains. To examine how the brain processes linguistically relevant spectral and temporal information, a functional magnetic resonance imaging study was conducted using Thai speech, in which spectral processing associated with lexical tones and temporal processing associated with vowel length can be differentiated. Ten Thai and 10 Chinese subjects were asked to perform discrimination judgments of pitch and timing patterns presented in the same auditory stimuli under two different conditions: speech (Thai) and nonspeech (hums). In the speech condition, tasks required judging Thai tones (T) and vowel length (VL); in the nonspeech condition, homologous pitch contours (P) and duration patterns (D). A remaining task required listening passively to nonspeech hums (L). Only the Thai group showed activation in the left inferior prefrontal cortex in speech minus nonspeech contrasts for spectral (T vs. P) and temporal (VL vs. D) cues. Thai and Chinese groups, however, exhibited similar fronto-parietal activation patterns in nonspeech hums minus passive listening contrasts for spectral (P vs. L) and temporal (D vs. L) cues. It appears that lower level specialization for acoustic cues in the spectral and temporal domains cannot be generalized to abstract higher order levels of phonological processing. Regardless of the neural mechanisms underlying low-level auditory processing, our findings clearly indicate that hemispheric specialization is sensitive to language-specific factors.

Phonology and neuropsychology of the English past tense

The double dissociation between the regular and irregular past tense in English has been explained in terms of dual and single mechanism accounts. In previous research we have argued that problems with the regular past tense in patients with left inferior frontal damage arise from morpho-phonological parsing difficulties [Trends in Cognitive Science 2 (1998) 428]. This claim has recently been challenged by a single mechanism connectionist account which argues that a general phonological processing deficit causes the poor performance on the regular past tense, with morphological factors playing no explicit role [Proceedings of the National Academy of Sciences 96 (1999) 7592]. We used a speeded judgement task with four patients who have documented difficulties with the regular past tense to contrast the claims made by these different approaches. We compared patients' ability to detect the difference between the past tense and stem of regular (hugged/hug) and irregular (taught/teach) past tense verbs, as well as matched "pseudo" pairs (trade/tray and port/peach). These real word conditions were accompanied by matched sets of non-words (e.g. nugged/nug). Patients' latencies to the regular past tense real word-pairs were consistently slower than in any other condition. To test for a general phonological processing deficit, we conducted several tests of phonological processing ability. The results show that the patients had a range of difficulties in phonological processing, from very mild to severe. This did not correlate with their performance on the speeded judgement task. We interpret this pattern of results as support for a specialised morpho-phonological processing mechanism which can be dissociated from other phonological processes and which is used directly in the processing of the regular past tense in a dual-mechanism system.

Anatomical correlates of learning novel speech sounds

We examined the relationship between brain anatomy and the ability to learn nonnative speech sounds, as well as rapidly changing and steady-state nonlinguistic sounds, using voxel-based morphometry in 59 healthy adults. Faster phonetic learners appeared to have more white matter in parietal regions, especially in the left hemisphere. The pattern of results was similar for the rapidly changing but not for the steady-state nonlinguistic stimuli, suggesting that morphological correlates of phonetic learning are related to the ability to process rapid temporal variation. Greater asymmetry in the amount of white matter in faster learners may be related to greater myelination allowing more efficient neural processing, which is critical for the ability to process certain speech sounds.

Cerebellum and speech perception: a functional magnetic resonance imaging study

A variety of data indicate that the cerebellum participates in perceptual tasks requiring the precise representation of temporal information. Access to the word form of a lexical item requires, among other functions, the processing of durational parameters of verbal utterances. Therefore, cerebellar dysfunctions must be expected to impair word recognition. In order to specify the topography of the assumed cerebellar speech perception mechanism, a functional magnetic resonance imaging study was performed using the German lexical items "Boden" ([bodn], Engl. "floor") and "Boten" ([botn], "messengers") as test materials. The contrast in sound structure of these two lexical items can be signaled either by the length of the wordmedial pause (closure time, CLT; an exclusively temporal measure) or by the aspiration noise of wordmedial "d" or "t" (voice onset time, VOT; an intrasegmental cue). A previous study found bilateral cerebellar disorders to compromise word recognition based on CLT whereas the encoding of VOT remained unimpaired. In the present study, two series of "Boden - Boten" utterances were resynthesized, systematically varying either in CLT or VOT. Subjects had to identify both words "Boden" and "Boten" by analysis of either the durational parameter CLT or the VOT aspiration segment. In a subtraction design, CLT categorization as compared to VOT identification (CLT - VOT) yielded a significant hemodynamic response of the right cerebellar hemisphere (neocerebellum Crus I) and the frontal lobe (anterior to Broca's area). The reversed contrast ( VOT - CLT) resulted in a single activation cluster located at the level of the supratemporal plane of the dominant hemisphere. These findings provide first evidence for a distinct contribution of the right cerebellar hemisphere to speech perception in terms of encoding of durational parameters of verbal utterances. Verbal working memory tasks, lexical response selection, and auditory imagery of word strings have been reported to elicit activation clusters of a similar location. Conceivably, representation of the temporal structure of speech sound sequences represents the common denominator of cerebellar participation in cognitive tasks acting on a phonetic code.

Modulation of activity in temporal cortex during generation of inner speech

Monitoring one's thoughts (in the verbal modality) is thought to be critically dependent on the interaction between areas that generate and perceive inner speech in the frontal and temporal cortex, respectively. We used functional magnetic resonance imaging (fMRI) to examine the relationship between activity in these areas while the rate of inner speech generation was varied experimentally. The faster rate was associated with activation in the left inferior frontal gyrus, the right pre- and postcentral gyri and both superior temporal gyri. Thus, temporal cortical activation was associated with increasing the rate of covert articulation, in the absence of external auditory input, suggesting that there is effective fronto-temporal connectivity. Furthermore, this may provide support for the existence of feed forward models, which suggest that activity in regions responsible for verbal perception is modulated by activity in areas that generate inner speech.

Dissociating a common working memory network from different neural substrates of phonological and spatial stimulus processing

Positron emission tomography was used to investigate common versus specific cortical regions for the maintenance of spatial versus phonological information in working memory (WM). Group and single-subject analyses of regional cerebral blood flow during a new 2 x 2 factorial n-back task were performed. Eight subjects had to memorize either phonological features or the location of serially presented syllables. Brain activation during phonological judgment and spatial judgment (0-back) was compared with that during two corresponding WM conditions (2-back). We observed a common network associated with the requirement of maintaining and sequencing items in WM. Seven or more subjects activated (posterior) superior frontal sulcus (pSFS, BA 6/8, global maximum) and/or adjacent gyri, posterior parietal cortex, and precuneus (BA 7). Less consistently, bilateral middle frontal gyrus (BA 9/46) was involved. Bilateral anterior (BA 39/40) and posterior (BA 7) intraparietal sulcus, as well as right pSFS, exhibited dominance for spatial WM. Although underlying stimulus processing pathways for both types of information were different, no region specific for phonological WM was found. Robust activation within the left inferior frontal gyrus (BA 44 and 45) was present, during both phonological WM and phonological judgment. We conclude that the controversial left prefrontal lateralization for verbal WM reflects more general phonological processing strategies, not necessarily required by tasks using letters. We propose a stimulus-independent role for the bilateral pSFS and its vicinity for maintenance and manipulation of different context-dependent information within working memory.

Rate-dependent activation of a prefrontal-insular-cerebellar network during passive listening to trains of click stimuli: an fMRI study

Eight volunteers underwent fMRI during passive listening to click trains. Using a parametric approach, rate-response profiles across the frequency band considered (2-6 Hz) were determined. Several cerebral structures outside the central-auditory pathways and target areas displayed distinct activation patterns each: rate-response profiles resembling high-pass (left side) or low-pass filtered (right side) signal series emerged at the level of the anterior insula, band-pass like characteristics (center frequency: 3-4 Hz) were observed within the left inferior frontal gyrus, and click train rates > 4 Hz yielded enhanced activation of the right cerebellar hemisphere. A variety of clinical and experimental data indicate that the left and right cerebral hemispheres act as high- and low-pass filters, respectively, on auditory input (double filtering by frequency theory). In light of the present fMRI data, the anterior insula contributes to the assumed double filtering by frequency functions. Furthermore, these intrasylvian areas seem to join up with the right cerebellum and the left inferior frontal gyrus to a network subserving parsing/timing functions within the auditory-verbal domain.

Human brain mapping of auditory imagery: event-related functional MRI study

We used event-related fMRI methodology to investigate human brain activity during auditory imagery. A series of susceptibility-weighted MR images covering the whole brain were acquired to obtain blood oxygenation level-dependent (BOLD) signal changes associated with the imagery event of hearing simple monotone. Group analysis across the 12 right-handed subjects revealed activations in the medial and inferior frontal gyri, precuneus, middle frontal gyri, superior temporal gyri, and anterior cingulate gyri. Bilateral primary and secondary auditory areas in the superior temporal gyri also exhibited the event-related MR signal changes. The proposed method allowed for the analysis of brain areas responsive to the event of auditory imagery while our results suggest that auditory imagery and actual audition share common neural substrates.

Baseline conditions and subtractive logic in neuroimaging

Discrepancies in the patterns of cortical activation across studies may be attributable, in part, to differences in baseline tasks, and hence, reflect the limits of the subtractive logic underlying much of neuroimaging. To assess the extent of these effects, three of the most commonly used baseline conditions (rest, tone monitoring, and passive listening) were compared using phoneme discrimination as the experimental task. Eight participants were studied in a fMRI study with a 4.1 T system. The three baseline conditions systematically affected the amount of activation observed in the identical phoneme task with major affects in Broca's area, the left posterior superior temporal gyrus, and the left and right inferior parietal regions. Two central findings were: 1) a differential effect of baseline within each region, with the rest baseline condition producing the greatest amount of activation and the passive listening condition producing the least, and 2) systematic baseline task activation in the inferior parietal regions. These results emphasize the relativity of activation patterns observed in functional neuroimaging, and the necessity to specify the baseline processes in context to the experimental task processes.

Differences in auditory processing of words and pseudowords: an fMRI study

Although there has been great interest in the neuroanatomical basis of reading, little attention has been focused on auditory language processing. The purpose of this study was to examine the differential neuroanatomical response to the auditory processing of real words and pseudowords. Eight healthy right-handed participants performed two phoneme monitoring tasks (one with real word stimuli and one with pseudowords) during a functional magnetic resonance imaging (fMRI) scan with a 4.1 T system. Both tasks activated the inferior frontal gyrus (IFG), the posterior superior temporal gyrus (pSTG) and the inferior parietal lobe (IPL). Pseudoword processing elicited significantly more activation within the posterior cortical regions compared with real word processing. Previous reading studies have suggested that this increase is due to an increased demand on the lexical access system. The left inferior frontal gyrus, on the other hand, did not reveal a significant difference in the amount of activation as a function of stimulus type. The lack of a differential response in IFG for auditory processing supports its hypothesized involvement in grapheme to phoneme conversion processes. These results are consistent with those from previous neuroimaging reading studies and emphasize the utility of examining both input modalities (e.g., visual or auditory) to compose a more complete picture of the language network.

Relations between the neural bases of dynamic auditory processing and phonological processing: evidence from fMRI

Functional magnetic resonance imaging (fMRI) was used to examine how the brain responds to temporal compression of speech and to determine whether the same regions are also involved in phonological processes associated with reading. Recorded speech was temporally compressed to varying degrees and presented in a sentence verification task. Regions involved in phonological processing were identified in a separate scan using a rhyming judgment task with pseudowords compared to a lettercase judgment task. The left inferior frontal and left superior temporal regions (Broca's and Wernicke's areas), along with the right inferior frontal cortex, demonstrated a convex response to speech compression; their activity increased as compression increased, but then decreased when speech became incomprehensible. Other regions exhibited linear increases in activity as compression increased, including the middle frontal gyri bilaterally. The auditory cortices exhibited compression-related decreases bilaterally, primarily reflecting a decrease in activity when speech became incomprehensible. Rhyme judgments engaged two left inferior frontal gyrus regions (pars triangularis and pars opercularis), of which only the pars triangularis region exhibited significant compression-related activity. These results directly demonstrate that a subset of the left inferior frontal regions involved in phonological processing is also sensitive to transient acoustic features within the range of comprehensible speech.

Phonological processing in visual rhyming: a developmental erp study

We employed a visual rhyming priming paradigm to characterize the development of brain systems important for phonological processing in reading. We studied 109 righthanded, native English speakers within eight age groups: 7-8, 9-10, 11-12, 13-14, 15-16, 17-18, 19-20, and 21-23. Participants decided whether two written words (prime-target) rhymed (JUICE-MOOSE) or not (CHAIR-MOOSE). In similar studies of adults, two main event-related potential (ERP) effects have been described: a negative slow wave to primes, larger over anterior regions of the left hemisphere and hypothesized to index rehearsal of the primes, and a negative deflection to targets, peaking at 400-450 msec, maximal over right temporal-parietal regions, larger for nonrhyming than rhyming targets, and hypothesized index phonological matching. In this study, these two ERP effects were observed in all age groups; however, the two effects showed different developmental timecourses. On the one hand, the frontal asymmetry to primes increased with age; moreover, this asymmetry was correlated with reading and spelling scores, even after controlling for age. On the other hand, the distribution and onset of the more posterior rhyming effect (RE) were stable across age groups, suggesting that phonological matching relied on similar neural systems across these ages. Behaviorally, both reaction times and accuracy improved with age. These results suggest that different aspects of phonological processing rely on different neural systems that have different to developmental timecourses.

Cortical degeneration associated with phonologic and semantic language impairments in AD

OBJECTIVE

To compare the pattern of cortical degeneration associated with different language deficits in cases of AD.

METHODS

Cases for detailed neuropathologic analysis (Patients 1 and 2) were selected because of their detailed clinical and neuropsychological assessments of language dysfunction in AD. Patient 1 had severe phonologic impairment with relatively preserved semantic aspects of language. Patient 2 had severe semantic language impairment with relatively preserved phonologic skills. The tissue volume of cortical regions associated with speech and language function was measured using standardized three-dimensional techniques. Neuronal areal fraction was also measured from histologic tissue samples. The degree of volume atrophy and neuronal loss was calculated in comparison to control measures (n = 10 men and 11 women). Measurements more than 2 SD from controls were considered abnormal.

RESULTS

Both AD cases had significant degeneration of the superior temporal gyrus and area 37. Cortical language regions affected only in Patient 1 included the anterior and posterior insula and part of Broca's area. In contrast, Patient 2 had a greater degree of degeneration in the temporal gyri and their white matter connections with the hippocampal/entorhinal complex.

CONCLUSIONS

Variable patterns of neurodegeneration underlie the clinical differences observed in patients with AD. Disconnection within the temporal lobe appears associated with semantic language difficulties, whereas disconnection of the anterior and posterior language areas appears associated with phonologic and grammatical impairment.

A cross-linguistic PET study of tone perception in Mandarin Chinese and English speakers

PET was used in a cross-linguistic study to determine whether neural mechanisms subserving pitch perception differ as a function of linguistic relevance. We compared tone perception in 12 native Mandarin speakers, who use tonal patterns to distinguish lexical meaning, with that of 12 native speakers of a nontone language, English. Subjects were scanned under two conditions: a silent resting baseline and a tonal task involving discrimination of pitch patterns in Mandarin words. Both groups showed common regions of CBF increase, but only Mandarin speakers showed additional activation in frontal, parietal, and parieto-occipital regions of the left hemisphere; this latter finding indicates that language experience may influence brain circuitry in the processing of auditory cues. In contrast, only the English group showed activity in the right inferior frontal cortex, consistent with a right-hemispheric role in pitch perception.