Alessandro Tavano

Temporal aspects of prediction in audition: Cortical and subcortical neural mechanisms

Tracing the temporal structure of acoustic events is crucial in order to efficiently adapt to dynamic changes in the environment. In turn, regularity in temporal structure may facilitate tracing of the acoustic signal and its likely spatial source. However, temporal processing in audition extends beyond a domain-general facilitatory function. Temporal regularity and temporal order of auditory events correspond to contextually extracted, statistically sampled relations among sounds. These relations are the backbone of prediction in audition, determining both when an event is likely to occur (temporal structure) and also what type of event can be expected at a specific point in time (formal structure, e.g. spectral information). Here, we develop a model of temporal processing in audition and speech that involves a division of labor between the cerebellum and the basal ganglia in tracing acoustic events in time. As for the cerebellum and its associated thalamo-cortical connections, we refer to its role in the automatic encoding of event-based temporal structure with high temporal precision, while the basal ganglia-thalamo-cortical system engages in the attention-dependent evaluation of longer-range intervals. Recent electrophysiological and neurofunctional evidence suggests that neocortical processing of spectral structure relies on concurrent extraction of event-based temporal information. We propose that spectrotemporal predictive processes may be facilitated by subcortical coding of relevant changes in sound energy as temporal event markers.

Evidence for a link among cognition, language and emotion in cerebellar malformations.

We compared the neurobehavioral profiles of children with Joubert syndrome (JS participants), a rare autosomal recessive condition characterized on magnetic resonance imaging (MRI) by hypoplasia of the cerebellar vermis and midbrain-hindbrain malformations, and children with malformations confined to the cerebellar vermis and one or both hemispheres (Cerebellar malformations--CM participants). We aimed at investigating the influence of anatomo-clinical similarities (vermian malformation) and differences (intact cerebellar hemispheres vs sparing of the pons, respectively) with respect to cognitive, linguistic and emotional development, assuming as a reference framework the Cerebellar Cognitive Affective Syndrome (CCAS). Results show that severe to moderate mental retardation is infrequent in JS children, while it is present in more than half the sample of CM children. Affect development was generally preserved in JS, in high-functioning CM individuals and also in some of the CM children with moderate mental retardation, which raised questions as to the role of a cerebellar vermis lesion in determining affect disorders. Further, cognitive and linguistic profiles on both intellectual and neuropsychological evaluations provided evidence for distinct patterns of peaks and valleys in the two groups, with JS children being significantly more impaired in language and verbal working memory and CM individuals showing a significant impairment of executive functions and emotional development. The overall evidence provides support for an important role of cerebellar structures per se in shaping emotional, cognitive and linguistic development, when vermian lesions are associated to cerebellar hemispheric lesions. Cerebellar vermis and brainstem lesions instead appear to have a major impact on motor-related skills, including oro-motor abilities and verbal working memory.

Specific linguistic and pragmatic deficits in Italian patients with schizophrenia

OBJECTIVE Verbal communication impairments are prominent features of schizophrenia. The grammatical and pragmatic components of expressive and receptive verbal abilities were systematically examined, for the first time, in Italian patients with schizophrenia. Indeed, most of the language literature is composed of studies on English speaking people. METHOD Elicited narrative production, and syntactic and pragmatic receptive abilities were analyzed in a cohort of 37 patients with schizophrenia and 37 healthy controls. Furthermore, a conversational speech production task was administered to an age- and gender-matched subset of this population. The level of significance was set at p<or=0.01. RESULTS Participants with schizophrenia produced significantly less words on the narrative task and were less fluent on the conversational task than healthy controls. In both narrative and conversational speech they showed significantly poorer syntactic diversity skills. Errors at word level did not distinguish the two groups. At a receptive level, syntactic abilities were selectively impaired in patients with schizophrenia, who were also slower than controls in providing their answers. Metaphor and idiom explanations revealed consistent deficits in patients with respect to controls. CONCLUSIONS Reduced syntactic diversity characterized expressive language skills in schizophrenia. Syntactic abilities were selectively impaired also at the receptive level, suggesting an underlying processing deficit. On the pragmatic test schizophrenia patients were significantly less able to produce appropriate interpretations, indicating the presence of abnormal pragmatic inferential abilities. These findings confirm that language impairment is a key feature of schizophrenia independent of mother language and suggest a possible deficit involving hemispheric lateralization processes.

Spontaneous Language Production in Bilingual Parkinson's Disease Patients: Evidence of Greater Phonological, Morphological and Syntactic Impairments in Native Language.

Nine early non-demented bilingual (L1 - Friulian, L2 - Italian) patients with Parkinsons disease and nine normal controls matched for age, sex and years of education were studied on a spontaneous language production task. All subjects had acquired L1 from birth in a home environment and L2 at the age of six at school formally. Patients with PD evidenced more phonological, morphological and syntactic errors in L1 than in L2. The opposite pattern was observed in normal controls as far as grammar was concerned. These findings suggest that implicit language processing is more impaired than explicit language processing in Parkinsons disease.

A callosal transfer deficit in children with developmental language disorder.

Twenty-two control children (aged 6-12 years) and forty-three children with developmental language disorder (DLD) (aged 7-12 years) received a test of callosal transfer of tactile information. Among the children with dysphasia, 30 had a diagnosis of receptive dysphasia and 13 of expressive dysphasia. Both control children and children with DLD made a significantly larger number of errors in the crossed localization condition (implying callosal transfer of tactile information) versus the uncrossed localization condition. In the crossed localization condition, children with DLD made a significantly larger number of errors than controls, while no differences were found in the two groups of children with DLD. These data suggest that the corpus callosum may be involved in the pathogenesis of DLD.

The role of emotion in dynamic audiovisual integration of faces and voices

We used human electroencephalogram to study early audiovisual integration of dynamic angry and neutral expressions. An auditory-only condition served as a baseline for the interpretation of integration effects. In the audiovisual conditions, the validity of visual information was manipulated using facial expressions that were either emotionally congruent or incongruent with the vocal expressions. First, we report an N1 suppression effect for angry compared with neutral vocalizations in the auditory-only condition. Second, we confirm early integration of congruent visual and auditory information as indexed by a suppression of the auditory N1 and P2 components in the audiovisual compared with the auditory-only condition. Third, audiovisual N1 suppression was modulated by audiovisual congruency in interaction with emotion: for neutral vocalizations, there was N1 suppression in both the congruent and the incongruent audiovisual conditions. For angry vocalizations, there was N1 suppression only in the congruent but not in the incongruent condition. Extending previous findings of dynamic audiovisual integration, the current results suggest that audiovisual N1 suppression is congruency- and emotion-specific and indicate that dynamic emotional expressions compared with non-emotional expressions are preferentially processed in early audiovisual integration.

Generative modeling and classification of dialogs by a low-level turn-taking feature

In the last few years, a growing attention has been paid to the problem of human-human communication, trying to devise artificial systems able to mediate a conversational setting between two or more people. In this paper, we propose an automatic system based on a generative structure able to classify dialog scenarios. The generative model is composed by integrating a Gaussian mixture model and a (observed) Markovian influence model, and it is fed with a novel low-level acoustic feature termed steady conversational period (SCP). SCPs are built on duration of continuous slots of silence or speech, taking also into account conversational turn-taking. The interactional dynamics built upon the transitions among SCPs provides a behavioral blueprint of conversational settings without relying on segmental or continuous phonetic features, and may be important for predicting the evolution of typical conversational situations in different dialog scenarios. The model has been tested on an extensive set of real, dyadic and multi-person conversational settings, including a recent dyadic dataset and the AMI meeting corpus. Comparative tests are made using conventional acoustic features and classification methods, showing that the proposed scheme provides superior classification performances for all conversational settings in our datasets. Moreover, we prove that our approach is able to characterize the nature of multi-person conversation (namely, the role of the participants) in a very accurate way, thus demonstrating great versatility.

Language and Social Communication in Children with Cerebellar Dysgenesis

Objective: Acquired cerebellar lesions in children and adults may determine deficits of executive functions, visuoperceptual skills, expressive language and modulation of affect; a complex pattern termed ‘cerebellar cognitive affective syndrome’. However, the long-term sequelae of malformative cerebellar lesions have yet to be systematically investigated, particularly in children. The purpose of this study was to present preliminary longitudinal data on the development of language and social communication skills in children with congenital malformations confined to the cerebellum. Patients and Methods: Five children (3 males, 2 females) with cerebellar malformations confined to the cerebellum were selected. Three patients presented with cerebellar hypoplasia involving the vermis and the hemispheres, while the remaining 2 had a malformation affecting only the cerebellar hemispheres. Neurobehavioral and language development were traced through access to available clinical data. Results: In the patients with cerebellar vermis malformation, language and social communicative skills were affected to a variable extent: 1 patient did not present with social disturbances during development. Those with hemispheric cerebellar lesions presented with selective linguistic impairments. Conclusions: The neurobehavioral profile of children with cerebellar malformations supports a key role of the cerebellum in language acquisition and affect regulation as distinguished functional domains.

Temporal regularity facilitates higher-order sensory predictions in fast auditory sequences

Does temporal regularity facilitate prediction in audition? To test this, we recorded human event‐related potentials to frequent standard tones and infrequent pitch deviant tones, pre‐attentively delivered within isochronous and anisochronous (20% onset jitter) rapid sequences. Deviant tones were repeated, either with high or low probability. Standard tone repetition sets a first‐order prediction, which is violated by deviant tone onset, leading to a first‐order prediction error response (Mismatch Negativity). The response to highly probable deviant repetitions is, however, attenuated relative to less probable repetitions, reflecting the formation of higher‐order sensory predictions. Results show that temporal regularity is required for higher‐order predictions, but does not modulate first‐order prediction error responses. Inverse solution analyses (Variable Resolution Electrical Tomography; VARETA) localized the error response attenuation to posterior regions of the left superior temporal gyrus. In a control experiment with a slower stimulus rate, we found no evidence for higher‐order predictions, and again no effect of temporal information on first‐order prediction error. We conclude that: (i) temporal regularity facilitates the establishing of higher‐order sensory predictions, i.e. ‘knowing what next’, in fast auditory sequences; (ii) first‐order prediction error relies predominantly on stimulus feature mismatch, reflecting the adaptive fit of fast deviance detection processes.

Sequence memory skills in Spastic Bilateral Cerebral Palsy are age independent as in normally developing children

Purpose: To study the development of sequence memory skills in a group of participants with Spastic Bilateral Cerebral Palsy (CP) and their matched controls (TD). Sequence memory skills are defined as a blend of implicit and explicit competences that are crucial for the acquisition and consolidation of most adaptive skills along the lifespan. Method: A computerized sequence learning task was administered to 51 participants with CP (age range: 4.1–14.7) and their controls. General performance, accuracy and learning strategy were analyzed, as well as cognitive competencies (IQ and explicit visual spatial memory). Results: Explicit learning developed along with age in all participants. Sequence learning skills were age independent and unevenly distributed among CP participants: most TD (96.1%) and only about half (58.8%) of CP participants succeeded in sequence learning, in dynamic relation with cognitive and manipulation abilities. Conclusion: Sequence memory skills should be verified to plan therapeutic strategies. Therapeutic plans based on implicit learning (more resistant to disruption and stress) could be effective and highly advantageous for most but not for all CP children. Independently from age, many CP children could fix sequences more efficiently by explicit strategies, a more effortful but probably more effective way. Implications for Rehabilitation “Sequence memory skills in Spastic Bilateral Cerebral Palsy (CP) are age-independent as in normally developing children” Sequence memory skills (a blend of explicit and implicit components) represent a basic competence whose impairment could in a dynamic perspective affect multiple motor and non-motor developmental features. The prevalence and importance of implicit learning as a point of strength in therapeutic choice has been formerly emphasized: implicit learning is far more resistant to disruption and stress during rehabilitation and therefore potentially far reaching. Sequence memory skills are unevenly distributed in our clinical group. A majority of CP participants (58.8%) can rely on efficient sequence learning and therefore may specifically benefit from therapeutic programs privileging implicit learning (that is, relying on progressive and not necessarily explicit consolidation of sequences). In those children failing to fix sequences, sequence learning skills need to be specifically supported and explicit strategies (such as for example verbal and visual guides along the task) could be helpful to support consolidation.