During speech perception listeners receive both linguistic information about the speech content as well as information regarding the identity of the talker. While these two aspects have been traditionally studied in isolation, with a dominant interest for linguistic information over talker identity, it is now a widely accepted notion that these two kinds of information are processed in an integrated way. The inclusion of talker-related information in the domain of speech perception highlighted both benefits and challenges for listeners. On the one hand, linguistic and talker-identity information appear to be mutually beneficial for the extraction of both kinds of information from the speech signal. On the other hand, listeners must take care of the great acoustic variability that characterizes the physical dimensions linked to the two kinds of information. The aim of the present dissertation is to study three specific cognitive mechanisms that listeners can use to access the benefits of the integrated processing of linguistic and talker-related information as well as to deal with their intrinsic variability. Three empirical studies employing both behavioural and neurophysiological techniques highlight peculiar aspects of abstraction, memory retrieval and perceptual learning mechanisms in relation to the consequences of including the talker in the study of speech perception.

When we observe other people's actions, a network of temporal, parietal and frontal regions is recruited, known as action observation network (AON). This network includes areas that have been reported to be involved when we perform actions ourselves. Such findings support the view that action understanding occurs by simulating actions in our own motor system (motor theories of action understanding). Alternatively, it has been argued that actions are understood based on a perceptual analysis, with access to action knowledge stored in the conceptual system (cognitive theories of action understanding). It has been argued earlier that areas that play a crucial role for action understanding should be able to (a) distinguish between different actions, and (b) generalize across the ways in which the action is performed (e.g. Dinstein, Thomas, Behrmann, & Heeger, 2008; Oosterhof, Tipper, & Downing, 2013; Caramazza, Anzelotti, Strnad, & Lingnau, 2014). Here we argue that one additional criterion needs to be met: an area that plays a crucial role for action understanding should have access to such abstract action information early, around the time when the action is recognized. An area that has access to abstract action information after the action has been recognized is unlikely to contribute to the process of action understanding. In this thesis, I report three neuroimaging studies in which we used magnetoencephalography (MEG) and functional magnetic resonance imaging (fMRI) to characterize the temporal dynamics of abstract representations of observed actions (Study 1 and 2), meaning that generalize across lower level dimensions, and to characterize the type of information encoded in the regions of the AON (Study 3). Specifically, in Study 1 we examined where in the brain and at which point in time it is possible to distinguish between pointing and grasping actions irrespective of the way in which they are performed (reach direction, effector) using MEG in combination with multivariate pattern analysis (MVPA) and source analysis. We show that regions in the left lateral occipitotemporal cortex (LOTC) have the earliest access to abstract action representations. By contrast, precentral regions, though recruited relatively early, have access to abstract action representations substantially later than left LOTC. In Study 2, we tested the temporal dynamics of the neural decoding related to the oscillatory activity induced by observation of actions performed with different effectors (hand, foot). We observed that temporal regions are able to discriminate all the presented actions before effector-related decoding within effector-specific motor regions. Finally, in Study 3 we investigated what aspect of an action is encoded within the regions of the AON. Object-directed actions induce a change of states, e.g. opening a bottle means changing its state from closed to open. It is still unclear how and in which brain regions these neural representations are encoded. Using fMRI-based multivoxel pattern decoding, we aimed at dissociating the neural representations of states and action functions. Participants observed stills of objects (e.g., window blinds) that were in either open or closed states, and videos of actions involving the same objects, i.e., open or close window. Action videos could show the object manipulation only (invisible change), or the complete action scene (visible change). This design allowed us to detect neural representations of action scenes, states and action functions independently of each other. We found different sub-regions within LOTC containing information related to object states, action functions, or both. These findings provide important information regarding the organization of action semantics in the brain and the role of LOTC in action understanding.

My PhD project mainly focused on understanding how a child or adolescent copes with pain associated with a disease, intended in a broader sense (i.e. procedures, treatments and disease-related). I tried to prove an innovative perspective that can help understand the wide variation in children’s pain experience, by considering intra-interpersonal influences, contextual factors, and intrapsychic factors that focus on needs, defenses, and self-structure. Overall, the whole project involved three pediatric units in Italy: the pediatric wards of Trento and Rovereto hospitals and the pediatric clinic of San Gerardo hospital, Monza (Milan). This doctoral thesis has achieved five goals: 1. Providing a selective overview on current relevant topics in the pediatric pain research and state of the art regarding the existing models of pediatric pain. 2. Developing a multi-dimensional protocol with an intra-method design for the assessment of pediatric pain in several chronic illnesses (cystic fibrosis, rheumatic diseases, cancer), by using also a battery of projective tests (drawings) to screen the emotional adjustment. 3. Validating the protocol by extending the methodology of projective drawings’ scoring with a control group and adding other assessment variables on a single cohort of patients (with malignant hematologic cancer) to test the new model that I developed. Quantitative analysis phase preceded qualitative analysis phase within the same framework to yield a parallel mixed analysis. 4. Planning specific training modules about pain management, starting from a bottom-up process concerning the local health professionals’ needs. I investigated these training needs through a series of open-ended questions, analyzed by a thematic analysis method. 5. Evaluate treatment’s feasibility, acceptability, and satisfaction of a problem-solving skills training for parents of children who have received an intensive pain rehabilitation from one pediatric pain rehabilitation program (Seattle Children's Hospital). I provided a methodological contribute within the mixed-method approach (statistical analysis and grounded theory). The results presented and their implications, are discussed in a clinical perspective since the rationale of this dissertation is that effective pain assessment must be multidimensional, multidisciplinary and at the same time feasible and practical to meet each pediatric patient’s needs.

La ricerca indaga due aree delle cognizione sociale: la Teoria della Mente (ToM) e la valutazione sociale. Nella prima parte è stato indagato quando e come emerge il ragionamento psicologico, inteso come l’abilità di inferire gli stati mentali altrui. Nel primo studio gli esperimenti valutano la capacità negli infanti di attribuire le false credenze attraverso lo sguardo anticipatorio (misure implicite) e anche negli adulti, attraverso compiti verbali (misure esplicite). I risultati hanno rivelato nei bambini di 17 mesi il possesso di un implicita TOM e danno quindi sostegno alle teorie modulari che prevedono una precoce attivazione di capacità metarappresentazionali. Nel secondo studio, sono stati condotti due esperimenti e confrontate due popolazioni di infanti, udenti e sordi. I risultati hanno mostrato il ruolo della precoce esperienza comunicativa, verbale o dei segni, sullo sviluppo della ToM. Un terzo studio condotto sui bambini prescolari bilingui e monolingui, ha rivelato migliori abilità di ragionamento inferenziale e pragmatico nei bambini bilingui. Il quarto studio ha indagato il substrato neurale, sottostante la ToM, per verificare l’ipotesi modulare, analizzando le prestazioni a compiti di ToM di pazienti neurologici, con lesione prevalentemente focale nella corteccia prefrontale, e controllando la co-azione di altri processi cognitivi, come le funzioni esecutive e l’intelligenza generale. I risultati hanno confermato l’ipotesi modulare e del dominio specifico della Teoria della Mente, esaminando una delle regioni neurali, ritenuta specializzata nelle rappresentazione degli stati mentali altrui: la corteccia prefrontale ventromediale. Nella seconda parte del lavoro è stata studiata la valutazione sociale delle azioni distributive nella prima infanzia. Sono stati condotti due studi che hanno indagato le inferenze delle disposizioni altrui, anche in seguito a giudizi su comportamenti distributivi operati da diversi agenti. I risultati hanno rivelato nei bambini di 16 mesi la capacità di inferire le disposizioni degli agenti, quando queste sono successive ad un processo di valutazione sociale dei comportamenti altrui. Le conclusioni hanno implicazioni teoriche che sostengono le nuove proposte sull’origine evolutiva del senso morale.

The ability of moving freely in the environment gives us the great advantage to directly interact with it, improving our discriminative abilities. For example, if we were to inspect an object without the chance to actively moving around it, then we could only rely on the information that we can extract from a single point of view with respect of the object. We would have restricted access to the object properties and we would then establish our decisions within those limits. Moving actively allow us to overcome these limitations and gain access to a more complete set of informations regarding the object. This would help us decide what to do next, whether or not to interact with an external object and, in case, providing hints on how to interact. To this extent moving and exploring the environment augment our discrimination abilities. Moreover, active movements help us to form a complete sense of space.

One of the core human abilities is that of interpreting symbols. Prompted with a perceptual stimulus devoid of any intrinsic meaning, such as a written word, our brain can access a complex multidimensional representation, called semantic representation, which corresponds to its meaning. Notwithstanding decades of neuropsychological and neuroimaging work on the cognitive and neural substrate of semantic representations, many questions are left unanswered. The research in this dissertation attempts to unravel one of them: are the neural substrates of different components of concrete word meaning dissociated? In the first part, I review the different theoretical positions and empirical findings on the cognitive and neural correlates of semantic representations. I highlight how recent methodological advances, namely the introduction of multivariate methods for the analysis of distributed patterns of brain activity, broaden the set of hypotheses that can be empirically tested. In particular, they allow the exploration of the representational geometries of different brain areas, which is instrumental to the understanding of where and when the various dimensions of the semantic space are activated in the brain. Crucially, I propose an operational distinction between motor-perceptual dimensions (i.e., those attributes of the objects referred to by the words that are perceived through the senses) and conceptual ones (i.e., the information that is built via a complex integration of multiple perceptual features). In the second part, I present the results of the studies I conducted in order to investigate the automaticity of retrieval, topographical organization, and temporal dynamics of motor-perceptual and conceptual dimensions of word meaning. First, I show how the representational spaces retrieved with different behavioral and corpora-based methods (i.e., Semantic Distance Judgment, Semantic Feature Listing, WordNet) appear to be highly correlated and overall consistent within and across subjects. Second, I present the results of four priming experiments suggesting that perceptual dimensions of word meaning (such as implied real world size and sound) are recovered in an automatic but task-dependent way during reading. Third, thanks to a functional magnetic resonance imaging experiment, I show a representational shift along the ventral visual path: from perceptual features, preferentially encoded in primary visual areas, to conceptual ones, preferentially encoded in mid and anterior temporal areas. This result indicates that complementary dimensions of the semantic space are encoded in a distributed yet partially dissociated way across the cortex. Fourth, by means of a study conducted with magnetoencephalography, I present evidence of an early (around 200 ms after stimulus onset) simultaneous access to both motor-perceptual and conceptual dimensions of the semantic space thanks to different aspects of the signal: inter-trial phase coherence appears to be key for the encoding of perceptual while spectral power changes appear to support encoding of conceptual dimensions. These observations suggest that the neural substrates of different components of symbol meaning can be dissociated in terms of localization and of the feature of the signal encoding them, while sharing a similar temporal evolution.

From the methodological perspective, we validated the use of a recently introduced multivariate searchlight pattern classification method for the analysis of training effects in language studies. We found that, compared to the standard GLM method, the searchlight analysis has comparable and, in some cases, greater sensitivity in localizing BOLD signal changes, and thus it represents a promising complementary tool in studies of training. "

Growing up as bilingual seem to exert some positive and negative effects on general cognitive functioning. Positive gains concern a earlier maturation of the attentional system in childhood and its later decline in adulthood. Bilinguals have been shown to be advantaged , compared to their monolingual peers in tasks requiring control of attention: they are more accurate in judging the grammaticality of sentences (Bialystok et al.1986); faster in the Attentional Network Test and in the Simon task (Costa et al 2007; Bialystok et al.2005).On the other hand bilinguals are disadvantaged in t asks requiring speech production : they are slower in acceding to the lexical representation of words (Ivanova, Costa,2007); they exhibit more tip of the tonguge states (Gollan & Silvenberg, 2001) and they show lower rates in retrieving verbal stimuli (Gollan,2002).If bilingualism can exert these influence sin boosting and hampering these so general cognitive functions, bilinguals children might also show these effects across domains.We surveyed the presence of these advantages and disadvantages in a domain so relevant as it is speaking two languages that is the number field. Bilinguals and monolinguals children were asked to perform three different number tasks: number Stroop, verification and dot counting.Unlike in the study of Bialystok et al. (2004; 2005) who employed a classical Stroop task, no difference was found between mono- and bilinguals as for the number Stroop effect. In the verification task, an associative confusion effect was found in the bilingual but not in the monolingual group. Finally, when children were asked to count, bilinguals performed equally well as their monolingual peers in counting items in the subitizing range only, whereas they were slower than their peers when they counted from 4 to 9 dots. This latter result, in particular, is consistent with many psycholinguistic studies claiming that bilinguals are disadvantaged in tasks requiring lexical access (Costa & Caramazza, 1999), and extend them to the number processing domain.

Parental sensitivity towards infants’ needs influences both the way caregiver-infant interactions unfold and individuals’ own development throughout lifetime. The pivotal role of this early interaction with caregivers is especially highlighted by the fact that when the interaction is non-adaptive, infants’ development may be severely hindered in various domains, such as cognitive, social, and emotional. Moreover, the quality of the early interaction with caregivers has long-lasting effects since it constitutes a lens through which individuals interpret the social world throughout lifetime. Caregivers’ influence on individuals’ subsequent behavior is also moderated by their own genetic predispositions. However the way behavioral, physiological and genetic mechanisms dynamically interact over time in shaping the development of caregiver-infant bonding and the long term effects on individuals remains largely unknown. The present project aimed to investigate behavioral and physiological mechanisms underling caregiver-infant interactions and their long-term effects applying a multilevel approach including behavioral, physiological and genetic measurements as well as a comparative approach between human and non-human primates. Specifically, in a first study focused on human adults we investigated the effect of the interaction between early parental care and individuals' own genetic predispositions in moderating adults' subsequent peripheral physiological responses to distressing social stimuli. Next, a second study on a primate model, the marmosets (\textit{Callithrix Jacchus}), focused directly on caregiver-infant dyads applying micro-behavioral analysis during infants' first month of life. Overall findings highlighted a differential importance of environmental and genetic factors in moderating caregiver-infant dyads vs individuals' long-term development. The leading role of environmental factors, namely parental behaviors, in determining infants' responses to specific caregivers within caregiver-infant interactions seems to be, on the contrary, smoothed out by individuals' own genetic predispositions when focusing on the long-term effects. Indeed, genetic characteristics determine individuals' sensitivity to environment, either weakening or strengthening the effect of environmental contribution in shaping individuals' physiological mechanisms. Also, interesting similarities between marmosets’ and humans’ caregiver-infant interactions’ structure have been found, making way for future studies investigating the brain mechanisms underlying the development of parent-infant bonding.

Object manipulation is central to our daily interactions with the environment. Failing to select, prepare or perform correct prehension movements results in dramatic limitations for the affected individual. Whereas we begin to have a better understanding of the neural mechanisms underlying the execution of object-directed movements, less is known about how exactly our brain makes the plan for action. Previous studies examining movement planning suggested that neuronal populations in parieto-frontal areas contain information about upcoming movements moments before they actually take place. However, such studies typically used experiments in which the participant was instructed about the movement to plan with visual or auditory cues, making it difficult to disentangle movement planning from the processing of cues and stimulus- response (S-R) mapping. In our first functional magnetic resonance imaging (fMRI) study (Study I), we compared an instructed condition with a free-choice condition that allowed participants to select which prehension movement to perform: a condition in which the task was not tied to specific external cues (i.e., no direct S-R mapping). Using multi-variate pattern analysis (MVPA), we found contralateral parietal and frontal regions containing abstract representations of planned movements that generalize across the way these movements were generated (internally vs externally). The majority of previous studies were based on delayed-movement tasks, which introduce brain responses unrelated to movement preparation. Consequently, whether these findings would generalize to immediate movements remained unclear. In our second fMRI study (Study II), we directly compared delayed and immediate reaching and grasping movements. Using time-resolved MVPA allowed us to reveal shared representations for delayed and non-delayed movement planning in human primary motor cortex and examine how movement representations unfolded throughout the different stages of planning and execution. Overall, our findings expand previous understanding of the regions implicated in movement planning and offer new insights into the dynamics of the human prehension system.