Publications

Abstract

The aim of the present study was to investigate whether reading failure in the context of an orthography of intermediate consistency is linked to inefficient use of the lexical orthographic reading procedure. The performance of typically developing and dyslexic Portuguese-speaking children was examined in a lexical decision task, where the stimulus lexicality, word frequency and length were manipulated. Both lexicality and length effects were larger in the dyslexic group than in controls, although the interaction between group and frequency disappeared when the data were transformed to control for general performance factors. Children with dyslexia were influenced in lexical decision making by the stimulus length of words and pseudowords, whereas age-matched controls were influenced by the length of pseudowords only. These findings suggest that non-impaired readers rely mainly on lexical orthographic information, but children with dyslexia preferentially use the phonological decoding procedure—albeit poorly—most likely because they struggle to process orthographic inputs as a whole such as controls do. Accordingly, dyslexic children showed significantly poorer performance than controls for all types of stimuli, including words that could be considered over-learned, such as high-frequency words. This suggests that their orthographic lexical entries are less established in the orthographic lexicon

Abstract

Even though language allows us to say exactly what we mean, we often use language to say things indirectly, in a way that depends on the specific communicative context. For example, we can use an apparently straightforward sentence like "It is hard to give a good presentation" to convey deeper meanings, like "Your talk was a mess!" One of the big puzzles in language science is how listeners work out what speakers really mean, which is a skill absolutely central to communication. However, most neuroimaging studies of language comprehension have focused on the arguably much simpler, context-independent process of understanding direct utterances. To examine the neural systems involved in getting at contextually constrained indirect meaning, we used functional magnetic resonance imaging as people listened to indirect replies in spoken dialog. Relative to direct control utterances, indirect replies engaged dorsomedial prefrontal cortex, right temporo-parietal junction and insula, as well as bilateral inferior frontal gyrus and right medial temporal gyrus. This suggests that listeners take the speaker's perspective on both cognitive (theory of mind) and affective (empathy-like) levels. In line with classic pragmatic theories, our results also indicate that currently popular "simulationist" accounts of language comprehension fail to explain how listeners understand the speaker's intended message.

Abstract

In this event-related FMRI study we investigated the effect of five days of implicit acquisition on preference classification by means of an artificial grammar learning (AGL) paradigm based on the structural mere-exposure effect and preference classification using a simple right-linear unification grammar. This allowed us to investigate implicit AGL in a proper learning design by including baseline measurements prior to grammar exposure. After 5 days of implicit acquisition, the FMRI results showed activations in a network of brain regions including the inferior frontal (centered on BA 44/45) and the medial prefrontal regions (centered on BA 8/32). Importantly, and central to this study, the inclusion of a naive preference FMRI baseline measurement allowed us to conclude that these FMRI findings were the intrinsic outcomes of the learning process itself and not a reflection of a preexisting functionality recruited during classification, independent of acquisition. Support for the implicit nature of the knowledge utilized during preference classification on day 5 come from the fact that the basal ganglia, associated with implicit procedural learning, were activated during classification, while the medial temporal lobe system, associated with explicit declarative memory, was consistently deactivated. Thus, preference classification in combination with structural mere-exposure can be used to investigate structural sequence processing (syntax) in unsupervised AGL paradigms with proper learning designs.

Abstract

Recent studies have emphasized that developmental dyslexia is a multiple-deficit disorder, in contrast to the traditional single-deficit view. In this context, cognitive profiling of children with dyslexia may be a relevant contribution to this unresolved discussion. The aim of this study was to profile 36 Portuguese children with dyslexia from the 2nd to 5th grade. Hierarchical cluster analysis was used to group participants according to their phonological awareness, rapid automatized naming, verbal short-term memory, vocabulary, and nonverbal intelligence abilities. The results suggested a two-cluster solution: a group with poorer performance on phoneme deletion and rapid automatized naming compared with the remaining variables (Cluster 1) and a group characterized by underperforming on the variables most related to phonological processing (phoneme deletion and digit span), but not on rapid automatized naming (Cluster 2). Overall, the results seem more consistent with a hybrid perspective, such as that proposed by Pennington and colleagues (2012), for understanding the heterogeneity of dyslexia. The importance of characterizing the profiles of individuals with dyslexia becomes clear within the context of constructing remediation programs that are specifically targeted and are more effective in terms of intervention outcome.

Abstract

We investigated global integration (wrap-up) processes at the boundaries of musical phrases by comparing the effects of well and non-well formed phrases on event-related potentials time-locked to two boundary points: the onset and the offset of the boundary pause. The Closure Positive Shift, which is elicited at the boundary offset, was not modulated by the quality of phrase structure (well vs. non-well formed). In contrast, the boundary onset potentials showed different patterns for well and non-well formed phrases. Our results contribute to specify the functional meaning of the Closure Positive Shift in music, shed light on the large-scale structural integration of musical input, and raise new hypotheses concerning shared resources between music and language.

Abstract

Neuroscientific and musicological approaches to music cognition indicate that listeners familiarized in the Western tonal tradition expect a musical phrase boundary at predictable time intervals. However, phrase boundary prediction processes in music remain untested. We analyzed event-related potentials (ERPs) and event-related induced power changes at the onset and offset of a boundary pause. We made comparisons with modified melodies, where the pause was omitted and filled by tones. The offset of the pause elicited a closure positive shift (CPS), indexing phrase boundary detection. The onset of the filling tones elicited significant increases in theta and beta powers. In addition, the P2 component was larger when the filling tones started than when they ended. The responses to boundary omission suggest that listeners expected to hear a boundary pause. Therefore, boundary prediction seems to coexist with boundary detection in music segmentation.

Abstract

In this fMRI study we investigate the neural correlates of information structure integration during sentence comprehension in Dutch. We looked into how prosodic cues (pitch accents) that signal the information status of constituents to the listener (new information) are combined with other types of information during the unification process. The difficulty of unifying the prosodic cues into overall sentence meaning was manipulated by constructing sentences in which the pitch accent did (focus-accent agreement), and sentences in which the pitch accent did not (focus-accent disagreement) match the expectations for focus constituents of the sentence. In case of a mismatch, the load on unification processes increases. Our results show two anatomically distinct effects of focus-accent disagreement, one located in the posterior left inferior frontal gyrus (LIFG, BA6/44), and one in the more anterior-ventral LIFG (BA 47/45). Our results confirm that information structure is taken into account during unification, and imply an important role for the LIFG in unification processes, in line with previous fMRI studies.

Abstract

The current study investigated which time components of rapid automatized naming (RAN) predict group differences between dyslexic and non-dyslexic readers (matched for age and reading level), and how these components relate to different reading measures. Subjects performed two RAN tasks (letters and objects), and data were analyzed through a response time analysis. Our results demonstrated that impaired RAN performance in dyslexic readers mainly stem from enhanced inter-item pause times and not from difficulties at the level of post-access motor production (expressed as articulation rates). Moreover, inter-item pause times account for a significant proportion of variance in reading ability in addition to the effect of phonological awareness in the dyslexic group. This suggests that non-phonological factors may lie at the root of the association between RAN inter-item pauses and reading ability. In normal readers, RAN performance was associated with reading ability only at early ages (i.e. in the reading-matched controls), and again it was the RAN inter-item pause times that explain the association.

Abstract

In the present study, the authors investigate how some visual factors related to early stages of visual-object naming modulate naming performance in dyslexia. The performance of dyslexic children was compared with 2 control groups—normal readers matched for age and normal readers matched for reading level—while performing a discrete naming task in which color and dimensionality of the visually presented objects were manipulated. The results showed that 2-dimensional naming performance improved for color representations in control readers but not in dyslexics. In contrast to control readers, dyslexics were also insensitive to the stimulus's dimensionality. These findings are unlikely to be explained by a phonological processing problem related to phonological access or retrieval but suggest that dyslexics have a lower capacity for coding and decoding visual surface features of 2-dimensional representations or problems with the integration of visual information stored in long-term memory.

Abstract

This article summarizes some of the important findings from research evaluating the relationship between poor rapid naming and impaired reading performance. Substantial evidence shows that dyslexic readers have problems with rapid naming of visual items. Early research assumed that this was a consequence of phonological processing deficits, but recent findings suggest that non-phonological processes may lie at the root of the association between slow naming speed and poor reading. The hypothesis that rapid naming reflects an independent core deficit in dyslexia is supported by the main findings: (1) some dyslexics are characterized by rapid naming difficulties but intact phonological skills; (2) evidence for an independent association between rapid naming and reading competence in the dyslexic readers, when the effect of phonological skills was controlled; (3) rapid naming and phonological processing measures are not reliably correlated. Recent research also reveals greater predictive power of rapid naming, in particular the inter-item pause time, for high-frequency word reading compared to pseudoword reading in developmental dyslexia. Altogether, the results are more consistent with the view that a phonological component alone cannot account for the rapid naming performance in dyslexia. Rather, rapid naming problems may emerge from the inefficiencies in visual-orthographic processing as well as in phonological processing.

Abstract

In this study, we systematically review the scientific literature on the effect of color on object recognition. Thirty-five independent experiments, comprising 1535 participants, were included in a meta-analysis. We found a moderate effect of color on object recognition (d = 0.28). Specific effects of moderator variables were analyzed and we found that color diagnosticity is the factor with the greatest moderator effect on the influence of color in object recognition; studies using color diagnostic objects showed a significant color effect (d = 0.43), whereas a marginal color effect was found in studies that used non-color diagnostic objects (d = 0.18). The present study did not permit the drawing of specific conclusions about the moderator effect of the object recognition task; while the meta-analytic review showed that color information improves object recognition mainly in studies using naming tasks (d = 0.36), the literature review revealed a large body of evidence showing positive effects of color information on object recognition in studies using a large variety of visual recognition tasks. We also found that color is important for the ability to recognize artifacts and natural objects, to recognize objects presented as types (line-drawings) or as tokens (photographs), and to recognize objects that are presented without surface details, such as texture or shadow. Taken together, the results of the meta-analysis strongly support the contention that color plays a role in object recognition. This suggests that the role of color should be taken into account in models of visual object recognition.

Abstract

In the present study, the authors explore in detail the level of visual object recognition at which perceptual color information improves the recognition of color diagnostic and noncolor diagnostic objects. To address this issue, 3 object recognition tasks, with different cognitive demands, were designed: (a) an object verification task; (b) a category verification task; and (c) a name verification task. They found that perceptual color information improved color diagnostic object recognition mainly in tasks for which access to the semantic knowledge about the object was necessary to perform the task; that is, in category and name verification. In contrast, the authors found that perceptual color information facilitates noncolor diagnostic object recognition when access to the object’s structural description from long-term memory was necessary—that is, object verification. In summary, the present study shows that the role of perceptual color information in object recognition is dependent on color diagnosticity

Abstract

The first objective of this study was to compare the brain network engaged by preference classification and the standard grammaticality classification after implicit artificial syntax acquisition by re-analyzing previously reported event-related fMRI data. The results show that preference and grammaticality classification engage virtually identical brain networks, including Broca’s region, consistent with previous behavioral findings. Moreover, the results showed that the effects related to artificial syntax in Broca’s region were essentially the same when masked with variability related to natural syntax processing in the same participants. The second objective was to explore CNTNAP2-related effects in implicit artificial syntax learning by analyzing behavioral and event-related fMRI data from a subsample. The CNTNAP2 gene has been linked to specific language impairment and is controlled by the FOXP2 transcription factor. CNTNAP2 is expressed in language related brain networks in the developing human brain and the FOXP2–CNTNAP2 pathway provides a mechanistic link between clinically distinct syndromes involving disrupted language. Finally, we discuss the implication of taking natural language to be a neurobiological system in terms of bounded recursion and suggest that the left inferior frontal region is a generic on-line sequence processor that unifies information from various sources in an incremental and recursive manner.

Abstract

Neste artigo fazemos uma revisão breve de investigações actuais com técnicas comportamentais e de neuroimagem funcional sobre a aprendizagem de uma linguagem artificial em crianças e adultos. Na secção final, discutimos uma possível associação entre dislexia e aprendizagem implícita. Resultados recentes sugerem que a presença de um défice ao nível da aprendizagem implícita pode contribuir para as dificuldades de leitura e escrita observadas em indivíduos disléxicos.

Abstract

A aquisição de competências de leitura e de escrita pode ser vista como um processo formal de transmissão cultural, onde interagem factores neurobiológicos e culturais. O treino sistemático exigido pela aprendizagem da leitura e da escrita poderá produzir mudanças quantitativas e qualitativas tanto a nível cognitivo como ao nível da organização do cérebro. Estudar sujeitos iletrados e letrados representa, assim, uma oportunidade para investigar efeitos de uma aprendizagem específica no desenvolvimento cognitivo e suas bases cerebrais. Neste trabalho, revemos um conjunto de investigações comportamentais e com métodos de imagem cerebral que indicam que a literacia tem um impacto nas nossas funções cognitivas e na organização cerebral. Mais especificamente, discutiremos diferenças entre letrados e iletrados para domínios cognitivos verbais e não-verbais, sugestivas de que a arquitectura cognitiva é formatada, em parte, pela aprendizagem da leitura e da escrita. Os dados de neuroimagem funcionais e estruturais são também indicadores que a aquisição de uma ortografia alfabética interfere nos processos de organização e lateralização das funções cognitivas.

Abstract

Work on animals indicates that BOLD is preferentially sensitive to local field potentials, and that it correlates most strongly with gamma band neuronal synchronization. Here we investigate how the BOLD signal in humans performing a cognitive task is related to neuronal synchronization across different frequency bands. We simultaneously recorded EEG and BOLD while subjects engaged in a visual attention task known to induce sustained changes in neuronal synchronization across a wide range of frequencies. Trial-by-trial BOLD luctuations correlated positively with trial-by-trial fluctuations in high-EEG gamma power (60–80 Hz) and negatively with alpha and beta power. Gamma power on the one hand, and alpha and beta power on the other hand, independently contributed to explaining BOLD variance. These results indicate that the BOLD-gamma coupling observed in animals can be extrapolated to humans performing a task and that neuronal dynamics underlying high- and low-frequency synchronization contribute independently to the BOLD signal.

Abstract

In individuals with ASD, difficulties with language comprehension are most evident when higher-level semantic-pragmatic language processing is required, for instance when context has to be used to interpret the meaning of an utterance. Until now, it is unclear at what level of processing and for what type of context these difficulties in language comprehension occur. Therefore, in the current fMRI study, we investigated the neural correlates of the integration of contextual information during auditory language comprehension in 24 adults with ASD and 24 matched control participants. Different levels of context processing were manipulated by using spoken sentences that were correct or contained either a semantic or world knowledge anomaly. Our findings demonstrated significant differences between the groups in inferior frontal cortex that were only present for sentences with a world knowledge anomaly. Relative to the ASD group, the control group showed significantly increased activation in left inferior frontal gyrus (LIFG) for sentences with a world knowledge anomaly compared to correct sentences. This effect possibly indicates reduced integrative capacities of the ASD group. Furthermore, world knowledge anomalies elicited significantly stronger activation in right inferior frontal gyrus (RIFG) in the control group compared to the ASD group. This additional RIFG activation probably reflects revision of the situation model after new, conflicting information. The lack of recruitment of RIFG is possibly related to difficulties with exception handling in the ASD group.

Abstract

Language acquisition in both natural and artificial language learning settings crucially depends on extracting information from sequence input. A shared sequence learning mechanism is thus assumed to underlie both natural and artificial language learning. A growing body of empirical evidence is consistent with this hypothesis. By means of artificial language learning experiments, we may therefore gain more insight in this shared mechanism. In this paper, we review empirical evidence from artificial language learning and computational modelling studies, as well as natural language data, and suggest that there are two key factors that help determine processing complexity in sequence learning, and thus in natural language processing. We propose that the specific ordering of non-adjacent dependencies (i.e., nested or crossed), as well as the number of non-adjacent dependencies to be resolved simultaneously (i.e., two or three) are important factors in gaining more insight into the boundaries of human sequence learning; and thus, also in natural language processing. The implications for theories of linguistic competence are discussed.

Abstract

In this paper we explore the feasibility of artificial (formal) grammar recognition (AGR) using spiking neural networks. A biologically inspired minicolumn architecture is designed as the basic computational unit. A network topography is defined based on the minicolumn architecture, here referred to as nodes, connected with excitatory and inhibitory connections. Nodes in the network represent unique internal states of the grammar’s finite state machine (FSM). Future work to improve the performance of the networks is discussed. The modeling framework developed can be used by neurophysiological research to implement network layouts and compare simulated performance characteristics to actual subject performance.

Abstract

In a 2x2 event-related FMRI study we find support for the idea that the inferior frontal cortex, centered on Broca’s region and its homologue, is involved in constructive unification operations during the structure-building process in parsing for comprehension. Tentatively, we provide evidence for a role of the dorsolateral prefrontal cortex centered on BA 9/46 in the control component of the language system. Finally, the left temporo-parietal cortex, in the vicinity of Wernicke’s region, supports the interaction between the syntax of gender agreement and sentence-level semantics.

Abstract

Human implicit learning can be investigated with implicit artificial grammar learning, a simple model for aspects of natural language acquisition. In this paper we investigate the remaining effect of modality transfer in syntactic classification of an acquired grammatical sequence structure after implicit grammar acquisition. Participants practiced either on acoustically presented syllable sequences or visually presented consonant letter sequences. During classification we independently manipulated the statistical frequency-based and rule-based characteristics of the classification stimuli. Participants performed reliably above chance on the within modality classification task although more so for those working on syllable sequence acquisition. These subjects were also the only group that kept a significant performance level in transfer classification. We speculate that this finding is of particular relevance in consideration of an ecological validity in the input signal in the use of artificial grammar learning and in language learning paradigms at large.

Abstract

Both local discourse and world knowledge are known to influence sentence processing. We investigated how these two sources of information conspire in language comprehension. Two types of critical sentences, correct and world knowledge anomalies, were preceded by either a neutral or a local context. The latter made the world knowledge anomalies more acceptable or plausible. We predicted that the effect of world knowledge anomalies would be weaker for the local context. World knowledge effects have previously been observed in the left inferior frontal region (Brodmann's area 45/47). In the current study, an effect of world knowledge was present in this region in the neutral context. We also observed an effect in the right inferior frontal gyrus, which was more sensitive to the discourse manipulation than the left inferior frontal gyrus. In addition, the left angular gyrus reacted strongly to the degree of discourse coherence between the context and critical sentence. Overall, both world knowledge and the discourse context affect the process of meaning unification, but do so by recruiting partly different sets of brain areas.

Abstract

A fundamental and intensively discussed question is whether medial temporal lobe (MTL) processes that lead to non-associative item memories differ in their anatomical substrate from processes underlying associative memory formation. Using event-related functional magnetic resonance imaging, we implemented a novel design to dissociate brain activity related to item and associative memory formation not only by subsequent memory performance and anatomy but also in time, because the two constituents of each pair to be memorized were presented sequentially with an intra-pair delay of several seconds. Furthermore, the design enabled us to reduce potential differences in memory strength between item and associative memory by increasing task difficulty in the item recognition memory test. Confidence ratings for correct item recognition for both constituents did not differ between trials in which only item memory was correct and trials in which item and associative memory were correct. Specific subsequent memory analyses for item and associative memory formation revealed brain activity that appears selectively related to item memory formation in the posterior inferior temporal, posterior parahippocampal, and perirhinal cortices. In contrast, hippocampal and inferior prefrontal activity predicted successful retrieval of newly formed inter-item associations. Our findings therefore suggest that different MTL subregions indeed play distinct roles in the formation of item memory and inter-item associative memory as expected by several dual process models of the MTL memory system.

Abstract

PET and fMRI experiments have previously shown that several brain regions in the frontal and parietal lobe are involved in working memory maintenance. MEG and EEG experiments have shown parametric increases with load for oscillatory activity in posterior alpha and frontal theta power. In the current study we investigated whether the areas found with fMRI can be associated with these alpha and theta effects by measuring simultaneous EEG and fMRI during a modified Sternberg task This allowed us to correlate EEG at the single trial level with the fMRI BOLD signal by forming a regressor based on single trial alpha and theta
power estimates. We observed a right posterior, parametric alpha power increase, which was functionally related to decreases in BOLD in the primary visual cortex and in the posterior part of the right middle temporal gyrus. We relate this finding to the inhibition of neuronal activity that may interfere with WM maintenance. An observed parametric increase in frontal theta power was correlated to a decrease in BOLD in
regions that together form the default mode network. We did not observe correlations between oscillatory EEG phenomena and BOLD in the traditional WM areas. In conclusion, the study shows that simultaneous EEG fMRI recordings can be successfully used to identify the emergence of functional networks in the brain during the execution of a cognitive task.

Abstract

Sentence comprehension requires the retrieval of single word information from long-term memory, and the integration of this information into multiword representations. The current functional magnetic resonance imaging study explored the hypothesis that the left posterior temporal gyrus supports the retrieval of lexical-syntactic information, whereas left inferior frontal gyrus (LIFG) contributes to syntactic unification. Twenty-eight subjects read sentences and word sequences containing word-category (noun–verb) ambiguous words at critical positions. Regions contributing to the syntactic unification process should show enhanced activation for sentences compared to words, and only within sentences display a larger signal for ambiguous than unambiguous conditions. The posterior LIFG showed exactly this predicted pattern, confirming our hypothesis that LIFG contributes to syntactic unification. The left posterior middle temporal gyrus was activated more for ambiguous than unambiguous conditions (main effect over both sentences and word sequences), as predicted for regions subserving the retrieval of lexical-syntactic information from memory. We conclude that understanding language involves the dynamic interplay between left inferior frontal and left posterior temporal regions.

Abstract

Difficulties with pragmatic aspects of communication are universal across individuals with autism spectrum disorders (ASDs). Here we focused on an aspect of pragmatic language comprehension that is relevant to social interaction in daily life: the integration of speaker characteristics inferred from the voice with the content of a message. Using functional magnetic resonance imaging (fMRI), we examined the neural correlates of the integration of voice-based inferences about the speaker’s age, gender or social background, and sentence content in adults with ASD and matched control participants. Relative to the control group, the ASD group showed increased activation in right inferior frontal gyrus (RIFG; Brodmann area 47) for speakerincongruent sentences compared to speaker-congruent sentences. Given that both groups performed behaviourally at a similar level on a debriefing interview outside the scanner, the increased activation in RIFG for the ASD group was interpreted as being compensatory in nature. It presumably reflects spill-over processing from the language dominant left hemisphere due to higher task demands faced by the participants with ASD when integrating speaker characteristics and the content of a spoken sentence. Furthermore, only the control group showed decreased activation for speaker-incongruent relative to speaker-congruent sentences in right ventral medial prefrontal cortex (vMPFC; Brodmann area 10), including right anterior cingulate cortex (ACC; Brodmann area 24/32). Since vMPFC is involved in self-referential processing related to judgments and inferences about self and others, the absence of such a modulation in vMPFC activation in the ASD group possibly points to atypical default self-referential mental activity in ASD. Our results show that in ASD compensatory mechanisms are necessary in implicit, low-level inferential processes in spoken language understanding. This indicates that pragmatic language problems in ASD are not restricted to high-level inferential processes, but encompass the most basic aspects of pragmatic language processing.

Abstract

When interpreting a message, a listener takes into account several sources of linguistic and extralinguistic information. Here we focused on one particular form of extralinguistic information, certain speaker characteristics as conveyed by the voice. Using functional magnetic resonance imaging, we examined the neural structures involved in the unification of sentence meaning and voice-based inferences about the speaker's age, sex, or social background. We found enhanced activation in the inferior frontal gyrus bilaterally (BA 45/47) during listening to sentences whose meaning was incongruent with inferred speaker characteristics. Furthermore, our results showed an overlap in brain regions involved in unification of speaker-related information and those used for the unification of semantic and world knowledge information [inferior frontal gyrus bilaterally (BA 45/47) and left middle temporal gyrus (BA 21)]. These findings provide evidence for a shared neural unification system for linguistic and extralinguistic sources of information and extend the existing knowledge about the role of inferior frontal cortex as a crucial component for unification during language comprehension.

Abstract

A dominant hypothesis in empirical research on the evolution of language is the following: the fundamental difference between animal and human communication systems is captured by the distinction between regular and more complex non-regular grammars. Studies reporting successful artificial grammar learning of nested recursive structures and imaging studies of the same have methodological shortcomings since they typically allow explicit problem solving strategies and this has been shown to account for the learning effect in subsequent behavioral studies. The present study overcomes these shortcomings by using subtle violations of agreement structure in a preference classification task. In contrast to the studies conducted so far, we use an implicit learning paradigm, allowing the time needed for both abstraction processes and consolidation to take place. Our results demonstrate robust implicit learning of recursively embedded structures (context-free grammar) and recursive structures with cross-dependencies (context-sensitive grammar) in an artificial grammar learning task spanning 9 days. Keywords: Implicit artificial grammar learning; centre embedded; cross-dependency; implicit learning; context-sensitive grammar; context-free grammar; regular grammar; non-regular grammar