Predictive and perceptual phonemic processing in articulatory motor areas: A prediction potential & mismatch negativity study.

The recent finding of predictive brain signals preceding anticipated perceptual and linguistic stimuli opens new questions for experimental research. Here, we address the possible brain basis of phonological predictions regarding the features of specific speech sounds and their relationship to phonological priming. To this end, we recorded EEG correlates of both pre- and post-stimulus brain responses in a phonological priming study. Redundant spoken sounds induced stimulus expectations, which manifested as a slow-wave anticipatory activity (the Prediction Potential, PP), whereas articulatory-congruent (e.g.,/bƏ/in the context of expected/pƏ/) pairs elicited weaker post-stimulus MMN-like responses as compared to the articulatory-incongruent (e.g.,/bƏ/in the context of expected/dƏ/) pairs, a pattern reminiscent of perceptual priming mediated by articulatory-motor areas. Source analysis reveal clusters of activation in lateral prefrontal, temporal and ventral motor areas, thus providing the proof of the relevance of multimodal representation units subserving predictive and perceptual phonemic processing.

Ultra-rapid access to words in chronic aphasia: the effects of intensive language action therapy (ILAT).

Effects of intensive language action therapy (ILAT) on automatic language processing were assessed using Magnetoencephalography (MEG). Auditory magnetic mismatch negativity (MMNm) responses to words and pseudowords were recorded in twelve patients with chronic aphasia before and immediately after two weeks of ILAT. Following therapy, Patients showed significant clinical improvements of auditory comprehension as measured by the Token Test and in word retrieval and naming as measured by the Boston Naming Test. Neuromagnetic responses dissociated between meaningful words and meaningless word-like stimuli ultra-rapidly, approximately 50 ms after acoustic information first allowed for stimulus identification. Over treatment, there was a significant increase in the left-lateralisation of this early word-elicited activation, observed in perilesional fronto-temporal regions. No comparable change was seen for pseudowords. The results may reflect successful, therapy-induced, language restitution in the left hemisphere.

Unconscious automatic brain activation of acoustic and action-related conceptual features during masked repetition priming.

Classical theories of semantic memory assume that concepts are represented in a unitary amodal memory system. In challenging this classical view, pure or hybrid modality-specific theories propose that conceptual representations are grounded in the sensory-motor brain areas, which typically process sensory and action-related information. Although neuroimaging studies provided evidence for a functional-anatomical link between conceptual processing of sensory or action-related features and the sensory-motor brain systems, it has been argued that aspects of such sensory-motor activation may not directly reflect conceptual processing but rather strategic imagery or postconceptual elaboration. In the present ERP study, we investigated masked effects of acoustic and action-related conceptual features to probe unconscious automatic conceptual processing in isolation. Subliminal feature-specific ERP effects at frontocentral electrodes were observed, which differed with regard to polarity, topography, and underlying brain electrical sources in congruency with earlier findings under conscious viewing conditions. These findings suggest that conceptual acoustic and action representations can also be unconsciously accessed, thereby excluding any postconceptual strategic processes. This study therefore further substantiates a grounding of conceptual and semantic processing in action and perception.

They played with the trade: MEG investigation of the processing of past tense verbs and their phonological twins.

How regular and irregular verbs are processed remains a matter of debate. Some English-speaking patients with nonfluent aphasia are especially impaired on regular past-tense forms like played, whether the task requires production, comprehension or even the judgement that "play" and "played" sound different. Within a dual-mechanism account of inflectional morphology, these deficits reflect disruption to the rule-based process that adds (or strips) the suffix -ed to regular verb stems; but the fact that the patients are also impaired at detecting the difference between word pairs like "tray" and "trade" (the latter being a phonological but not a morphological twin to "played") suggests an important role for phonological characteristics of the regular past tense. The present study examined MEG brain responses in healthy participants evoked by spoken regular past-tense forms and phonological twin words (plus twin pseudowords and a non-speech control) presented in a passive oddball paradigm. Deviant forms (played, trade, kwade/kwayed) relative to their standards (play, tray, kway) elicited a pronounced neuromagnetic response at approximately 130 ms after the onset of the affix; this response was maximal at sensors over temporal areas of both hemispheres but stronger on the left, especially for played and kwayed. Relative to the same standards, a different set of deviants ending in /t/--plate, trait and kwate--produced stronger difference responses especially over the right hemisphere. Results are discussed with regard to dual- and single-mechanism theories of past tense processing and the need to consider neurobiological evidence in attempts to understand inflectional morphology.

Attention to language: novel MEG paradigm for registering involuntary language processing in the brain.

Previous research indicates that, under explicit instructions to listen to spoken stimuli or in speech-oriented behavioural tasks, the brain's responses to senseless pseudowords are larger than those to meaningful words; the reverse is true in non-attended conditions. These differential responses could be used as a tool to trace linguistic processes in the brain and their interaction with attention. However, as previous studies relied on explicit instructions to attend or ignore the stimuli, a technique for automatic attention modulation (i.e., not dependent on explicit instruction) would be more advantageous, especially when cooperation with instructions may not be guaranteed (e.g., neurological patients, children etc). Here we present a novel paradigm in which the stimulus context automatically draws attention to speech. In a non-attend passive auditory oddball sequence, rare words and pseudowords were presented among frequent non-speech tones of variable frequency and length. The low percentage of spoken stimuli guarantees an involuntary attention switch to them. The speech stimuli, in turn, could be disambiguated as words or pseudowords only in their end, at the last phoneme, after the attention switch would have already occurred. Our results confirmed that this paradigm can indeed be used to induce automatic shifts of attention to spoken input. At ~250ms after the stimulus onset, a P3a-like neuromagnetic deflection was registered to spoken (but not tone) stimuli indicating an involuntary attention shift. Later, after the word-pseudoword divergence point, we found a larger oddball response to pseudowords than words, best explained by neural processes of lexical search facilitated through increased attention. Furthermore, we demonstrate a breakdown of this orderly pattern of neurocognitive processes as a result of sleep deprivation. The new paradigm may thus be an efficient way to assess language comprehension processes and their dynamic interaction with those of attention allocation. It does it in an automatic and task-free fashion, indicating its potential benefit for assessing uncooperative clinical populations.

Dissociating the representation of action- and sound-related concepts in middle temporal cortex.

Modality-specific models of conceptual memory propose close links between concepts and the sensory-motor systems. Neuroimaging studies found, in different subject groups, that action-related and sound-related concepts activated different parts of posterior middle temporal gyrus (pMTG), suggesting a modality-specific representation of conceptual features. However, as these different parts of pMTG are close to each other, it is possible that the observed anatomical difference is merely related to interindividual variability. In the present functional magnetic resonance imaging study, we now investigated within the same participant group a possible conceptual feature-specific organization in pMTG. Participants performed lexical decisions on sound-related (e.g., telephone) and action-related (hammer) words. Sound words elicited higher activity in anterior pMTG adjacent to auditory association cortex, but action-related words did so in posterior pMTG close to motion sensitive areas. These results confirm distinct conceptual representations of sound and action in pMTG, just adjacent to the respective modality-specific cortices.

Ultra-rapid access to words in the brain.

Rapid information processing in the human brain is vital to survival in a highly dynamic environment. The key tool humans use to exchange information is spoken language, but the exact speed of the neuronal mechanisms underpinning speech comprehension is still unknown. Here we investigate the time course of neuro-lexical processing by analyzing neuromagnetic brain activity elicited in response to psycholinguistically and acoustically matched groups of words and pseudowords. We show an ultra-early dissociation in cortical activation elicited by these stimulus types, emerging ∼50 ms after acoustic information required for word identification first becomes available. This dissociation is the earliest brain signature of lexical processing of words so far reported, and may help explain the evolutionary advantage of human spoken language.

Reading salt activates gustatory brain regions: fMRI evidence for semantic grounding in a novel sensory modality.

Because many words are typically used in the context of their referent objects and actions, distributed cortical circuits for these words may bind information about their form with perceptual and motor aspects of their meaning. Previous work has demonstrated such semantic grounding for sensorimotor, visual, auditory, and olfactory knowledge linked to words, which is manifest in activation of the corresponding areas of the cortex. Here, we explore the brain basis of gustatory semantic links of words whose meaning is primarily related to taste. In a blocked functional magnetic resonance imaging design, Spanish taste words and control words matched for a range of factors (including valence, arousal, image-ability, frequency of use, number of letters and syllables) were presented to 59 right-handed participants in a passive reading task. Whereas all the words activated the left inferior frontal (BA44/45) and the posterior middle and superior temporal gyri (BA21/22), taste-related words produced a significantly stronger activation in these same areas and also in the anterior insula, frontal operculum, lateral orbitofrontal gyrus, and thalamus among others. As these areas comprise primary and secondary gustatory cortices, we conclude that the meaning of taste words is grounded in distributed cortical circuits reaching into areas that process taste sensations.

Strength of word-specific neural memory traces assessed electrophysiologically.

Memory traces for words are frequently conceptualized neurobiologically as networks of neurons interconnected via reciprocal links developed through associative learning in the process of language acquisition. Neurophysiological reflection of activation of such memory traces has been reported using the mismatch negativity brain potential (MMN), which demonstrates an enhanced response to meaningful words over meaningless items. This enhancement is believed to be generated by the activation of strongly intraconnected long-term memory circuits for words that can be automatically triggered by spoken linguistic input and that are absent for unfamiliar phonological stimuli. This conceptual framework critically predicts different amounts of activation depending on the strength of the word's lexical representation in the brain. The frequent use of words should lead to more strongly connected representations, whereas less frequent items would be associated with more weakly linked circuits. A word with higher frequency of occurrence in the subject's language should therefore lead to a more pronounced lexical MMN response than its low-frequency counterpart. We tested this prediction by comparing the event-related potentials elicited by low- and high-frequency words in a passive oddball paradigm; physical stimulus contrasts were kept identical. We found that, consistent with our prediction, presenting the high-frequency stimulus led to a significantly more pronounced MMN response relative to the low-frequency one, a finding that is highly similar to previously reported MMN enhancement to words over meaningless pseudowords. Furthermore, activation elicited by the higher-frequency word peaked earlier relative to low-frequency one, suggesting more rapid access to frequently used lexical entries. These results lend further support to the above view on word memory traces as strongly connected assemblies of neurons. The speed and magnitude of their activation appears to be linked to the strength of internal connections in a memory circuit, which is in turn determined by the everyday use of language elements.

Event-related potentials reflecting the frequency of unattended spoken words: a neuronal index of connection strength in lexical memory circuits?

How are words represented in the human brain and can these representations be qualitatively assessed with respect to their structure and properties? Recent research demonstrates that neurophysiological signatures of individual words can be measured when subjects do not focus their attention on speech input. These automatic activations, which take the form of negative deflections of event-related potentials, can appear surprisingly early (within ~200 ms) and are based on robust connections within neuronal memory circuits encoding individual words that ignite even when attentional resources are scarce. A new and critical prediction of this framework is that words with high frequency of occurrence have especially strong connections of their underlying memory circuits and should thus yield more negative ERPs compared with rarer words. We tested this prediction by presenting our subjects, in passive non-attend conditions, with acoustically matched high- and low-frequency words along with pseudo-words. Using factorial and correlation analyses, we found that already at ~120 ms after the spoken stimulus information was available, amplitude of brain responses was modulated by the words' lexical frequency. Topographic mapping and source analysis suggested that this early automatic frequency effect originates from the left inferior-frontal cortices. While, at this early latency, lexical differences between words and pseudo-words (more negative-going potential for meaningful words) could be seen only for the most frequent word stimuli, later-on (~270 ms), a more global lexicality effect with bilateral perisylvian sources was found for all stimuli, suggesting faster access to more frequent lexical entries. Our results support the account of word memory traces as interconnected neuronal circuits, and suggest that speed and magnitude of their activation are determined by their internal connection strength, which, in turn, is determined by the everyday language use.

From sounds to words: a neurocomputational model of adaptation, inhibition and memory processes in auditory change detection.

Most animals detect sudden changes in trains of repeated stimuli but only some can learn a wide range of sensory patterns and recognise them later, a skill crucial for the evolutionary success of higher mammals. Here we use a neural model mimicking the cortical anatomy of sensory and motor areas and their connections to explain brain activity indexing auditory change and memory access. Our simulations indicate that while neuronal adaptation and local inhibition of cortical activity can explain aspects of change detection as observed when a repeated unfamiliar sound changes in frequency, the brain dynamics elicited by auditory stimulation with well-known patterns (such as meaningful words) cannot be accounted for on the basis of adaptation and inhibition alone. Specifically, we show that the stronger brain responses observed to familiar stimuli in passive oddball tasks are best explained in terms of activation of memory circuits that emerged in the cortex during the learning of these stimuli. Such memory circuits, and the activation enhancement they entail, are absent for unfamiliar stimuli. The model illustrates how basic neurobiological mechanisms, including neuronal adaptation, lateral inhibition, and Hebbian learning, underlie neuronal assembly formation and dynamics, and differentially contribute to the brain's major change detection response, the mismatch negativity.

Interactions between language and attention systems: early automatic lexical processing?

An ongoing debate is whether and to what extent access to cortical representations is automatic or dependent on attentional processes. To address this, we modulated the level of attention on auditory input and recorded ERPs elicited by syllables completing acoustically matched words and pseudowords. Under nonattend conditions, the word-elicited response (peaking at approximately 120 msec) was larger than that to pseudowords, confirming early activation of lexical memory traces. However, when attention was directed toward the auditory input, such word-pseudoword difference disappeared. Whereas responses to words seemed unchanged by attentional variation, early pseudoword responses were modulated significantly by attention. Later on, attention modulated a positive deflection at approximately 230 msec and a second negativity at approximately 370 msec for all stimuli. The data indicate that the earliest stages of word processing are not affected by attentional demands and may thus possess certain automaticity, with attention effects on lexical processing accumulating after 150-200 msec. We explain this by robustness of preexisting memory networks for words whose strong internal connections guarantee rapid full-scale activation irrespective of the attentional resources available. Conversely, the processing of pseudowords, which do not have such stimulus-specific cortical representations, appears to be strongly modulated by the availability of attentional resources, even at its earliest stages. Topography analysis and source reconstruction indicated that left peri-sylvian cortices mediate attention effects on memory trace activation.

Changes in the perceived duration of a narrowband sound induced by a preceding stimulus.

The authors show that a narrowband noise (NBN) is perceived as longer when presented immediately after a wideband noise (WBN), compared to when the WBN is absent. This effect depended on the WBN's frequency spectrum overlapping that of the NBN, and it increased as the duration of the WBN increased up to 300 ms. It decreased when a silent gap was introduced between the WBN and NBN, but remained significant for an easily detectable gap of 40 ms. A correlate of the effect was observed in the mismatch negativity (MMN) to a deviant stimulus, consisting of a WBN + NBN, presented in a sequence of more common isolated WBNs. The MMN latency was longer for an on-frequency than for an off-frequency WBN; and, more importantly, the size of this difference correlated across participants with the difference in perceived duration. A rhythm-adjustment experiment showed that the presence of an on-frequency WBN immediately preceding a tone caused that tone to be heard as starting earlier than when the WBN was absent. The results are discussed in relation to the continuity illusion and models of duration encoding.

Auditory size-deviant detection in adults and newborn infants.

Auditory size perception refers to the ability to make accurate judgements of the size of a sound source based solely upon the sound emitted from the source. Electro-physiological and behavioural data were collected to test whether sound-source size parameters are detected from task-irrelevant sequences in adults and newborn infants. The mismatch negativity (MMN) obtained from adults indexed automatic detection of changes in size for voices, musical instruments and animal calls, regardless of whether the acoustic change indicated larger or smaller sources. Neonates detected changes in the size of a musical instrument. The data are consistent with the notion that auditory size-deviant detection in humans is an innate automatic process. This conclusion is compatible with the theory that the ability to assess the size of sound sources evolved because it provided selective advantage of being able to detect larger (more competent) suitors and larger (more dangerous) predators.

Spatiotemporal signatures of large-scale synfire chains for speech processing as revealed by MEG.

We report a new brain signature of memory trace activation in the human brain revealed by magnetoencephalography and distributed source localization. Spatiotemporal patterns of cortical activation can be picked up in the time course of source images underlying magnetic brain responses to speech and noise stimuli, especially the generators of the magnetic mismatch negativity. We found that acoustic signals perceived as speech elicited a well-defined spatiotemporal pattern of sequential activation of superior-temporal and inferior-frontal cortex, whereas the same identical stimuli, when perceived as noise, did not elicit temporally structured activation. Strength of local sources constituting large-scale spatiotemporal patterns reflected additional lexical and syntactic features of speech. Morphological processing of the critical sound as verb inflection led to particularly pronounced early left inferior-frontal activation, whereas the same sound functioning as inflectional affix of a noun activated superior-temporal cortex more strongly. We conclude that precisely timed spatiotemporal patterns involving specific cortical areas may represent a brain code of memory circuit activation. These spatiotemporal patterns are best explained in terms of synfire mechanisms linking neuronal populations in different cortical areas. The large-scale synfire chains appear to reflect the processing of stimuli together with the context-dependent perceptual and cognitive information bound to them.

Imagery or meaning? Evidence for a semantic origin of category-specific brain activity in metabolic imaging.

Category-specific brain activation distinguishing between semantic word types has imposed challenges on theories of semantic representations and processes. However, existing metabolic imaging data are still ambiguous about whether these category-specific activations reflect processes involved in accessing the semantic representation of the stimuli, or secondary processes such as deliberate mental imagery. Further information about the response characteristics of category-specific activation is still required. Our study for the first time investigated the differential impact of word frequency on functional magnetic resonance imaging (fMRI) responses to action-related words and visually related words, respectively. First, we corroborated previous results showing that action-relatedness modulates neural responses in action-related areas, while word imageability modulates activation in object processing areas. Second, we provide novel results showing that activation negatively correlated with word frequency in the left fusiform gyrus was specific for visually related words, while in the left middle temporal gyrus word frequency effects emerged only for action-related words. Following the dominant view in the literature that effects of word frequency mainly reflect access to lexico-semantic information, we suggest that category-specific brain activation reflects distributed neuronal ensembles, which ground language and concepts in perception-action systems of the human brain. Our approach can be applied to any event-related data using single-stimulus presentation, and allows a detailed characterization of the functional role of category-specific activation patterns.

Grammar or serial order?: discrete combinatorial brain mechanisms reflected by the syntactic mismatch negativity.

If word strings violate grammatical rules, they elicit neurophysiological brain responses commonly attributed to a specifically human language processor or grammar module. However, an ungrammatical string of words is always also a very rare sequence of events and it is, therefore, not always evident whether specifically linguistic processes are at work when neurophysiological grammar indexes are being reported. We here investigate the magnetic mismatch negativity (MNN) to ungrammatical word strings, to very rare grammatical strings, and to common grammatical phrases. In this design, serial order mechanism mapping the sequential probability of words should neurophysiologically dissociate frequent grammatical phrases from both ungrammatical and rare grammatical strings. However, if syntax as a discrete combinatorial system is reflected, the prediction is that the rare, correctly combined items group with the highly frequent grammatical strings and stand out against ungrammatical strings. Using magnetoencephalography as a measure of human brain activity, we replicated the previously reported syntactic mismatch negativity (sMMN), which distinguishes highly unfamiliar ungrammatical word sequences from common grammatical strings. Crucially, a significant interaction demonstrated that the sMMN specifically distinguished syntactic violations from common grammatical strings, but not uncommon from common grammatical word strings. This significant interaction argues in favor of a genuinely grammatical origin of the sMMN and provides direct neurophysiological evidence for a discrete combinatorial system for word and morpheme sequences in the human brain. The data are more difficult to explain in the context of serial order models that map co-occurrence probabilities of words.

Language outside the focus of attention: the mismatch negativity as a tool for studying higher cognitive processes.

Which aspects of language and cognitive processing take place irrespective of whether subjects focus their attention on incoming stimuli and are, in this sense, automatic? The Mismatch Negativity (MMN), a neurophysiological brain response recorded in the EEG and MEG, is elicited by attended and unattended stimuli alike. Recent studies investigating the cognitive processes underlying spoken language processing found that even under attentional withdrawal, MMN size and topography reflect the activation of memory traces for language elements in the human brain. Familiar sounds of one's native language elicit a larger MMN than unfamiliar sounds, and at the level of meaningful language units, words elicit a larger MMN than meaningless pseudowords. This suggests that the MMN reflects the activation of memory networks for language sounds and spoken words. Unattended word stimuli elicit an activation sequence starting in superior-temporal cortex and rapidly progressing to left-inferior-frontal lobe. The spatio-temporal patterns of cortical activation depend on lexical and semantic properties of word stems and affixes, thus indicating that the MMN can give clues about lexico-semantic information processing stored in long term memory. At the syntactic level, MMN size was found to reflect whether a word string conforms to abstract grammatical rules. This growing body of results suggests that lexical, semantic and syntactic information can be processed by the central nervous system outside the focus of attention in a largely automatic manner. Analysis of spatio-temporal patterns of generator activations underlying the MMN to speech may be an important tool for investigating the brain dynamics of spoken language processing and the activated distributed cortical circuits acting at long-term memory traces.

Reading cinnamon activates olfactory brain regions.

Some words immediately and automatically remind us of odours, smells and scents, whereas other language items do not evoke such associations. This study investigated, for the first time, the abstract linking of linguistic and odour information using modern neuroimaging techniques (functional MRI). Subjects passively read odour-related words ('garlic', 'cinnamon', 'jasmine') and neutral language items. The odour-related terms elicited activation in the primary olfactory cortex, which include the piriform cortex and the amygdala. Our results suggest the activation of widely distributed cortical cell assemblies in the processing of olfactory words. These distributed neuron populations extend into language areas but also reach some parts of the olfactory system. These distributed neural systems may be the basis of the processing of language elements, their related conceptual and semantic information and the associated sensory information.

Brain signatures of meaning access in action word recognition.

The brain basis of action words may be neuron ensembles binding language- and action-related information that are dispersed over both language- and action-related cortical areas. This predicts fast spreading of neuronal activity from language areas to specific sensorimotor areas when action words semantically related to different parts of the body are being perceived. To test this, fast neurophysiological imaging was applied to reveal spatiotemporal activity patterns elicited by words with different action-related meaning. Spoken words referring to actions involving the face or leg were presented while subjects engaged in a distraction task and their brain activity was recorded using high-density magnetoencephalography. Shortly after the words could be recognized as unique lexical items, objective source localization using minimum norm current estimates revealed activation in superior temporal (130 msec) and inferior frontocentral areas (142-146 msec). Face-word stimuli activated inferior frontocentral areas more strongly than leg words, whereas the reverse was found at superior central sites (170 msec), thus reflecting the cortical somatotopy of motor actions signified by the words. Significant correlations were found between local source strengths in the frontocentral cortex calculated for all participants and their semantic ratings of the stimulus words, thus further establishing a close relationship between word meaning access and neurophysiology. These results show that meaning access in action word recognition is an early automatic process ref lected by spatiotemporal signatures of word-evoked activity. Word-related distributed neuronal assemblies with specific cortical topographies can explain the observed spatiotemporal dynamics reflecting word meaning access.