The experiments reported in the thesis investigate the nature of word stress and its lexical representation. Focusing on Italian, I considered two research questions: How is lexical stress represented? How does this representation intervene in perceiving or producing a word? Italian is a polysyllabic language with free-stress position: Stress may appear on one of the last three-syllables (e.g., TAvolo, ‘table’, paROla, ‘word’, coliBRI, ‘hummingbird’, capitals indicate stress) and its position is not predictable by rules. Moreover, there is a large asymmetry in the distribution of the stress patterns, with about 80% of words bearing stress on the penultimate syllable (e.g., paROla, ‘word’). On the assumption that stress is a lexical feature and that the stress pattern of a word is part of the knowledge stored in the lexicon, three studies were designed in which a priming paradigm and a visual word paradigm were used. Specifically, we investigated lexical stress in two domains, i.e., spoken-word recognition (Chapter 2) and reading aloud (Chapters 4 and 5). The results shed new light on the nature of the stored prosodic knowledge about lexical stress and on what extent processing of lexical stress is similar in spoken-word recognition and reading aloud. In synthesis, the empirical evidence indicates that lexical stress is part of the abstract prosodic knowledge stored in the lexicon: It pertains to the suprasegmental level of word representation and it is dissociable from the information pertaining to the segmental level.

The project has explored the developmental trajectories of several cognitive functions related to different brain regions: parietal cortex (quantity manipulation, finger gnosis, visuo-spatial memory and grasping abilities) and occipito-temporal cortex (face and object processing), in order to investigate their contributions to the acquisition of formal arithmetic in the first year of schooling. We tested preschooler, first grader and adult subjects, using correlational cross-sectional and longitudinal approaches. Results show that anatomical proximity is a strong predictor of behavioural correlations and of segregation between dorsal and ventral streams’ functions. This observation is particularly prominent in children: within parietal functions, there is a progressive separation across functions during development. During preschool age, presymbolic and symbolic number systems follow distinct developmental trajectories that converge during the first year of primary school. Indeed a possible cause of this phenomenon could be due to the refinement of the numerosity acuity during the acquisition of symbolic knowledge for numbers. Among the tested parietal functions, we observe a strong association between the numerical and the finger domain, especially in children. In preschoolers, finger gnosis is strongly associated with non-symbolic quantity processing, while in first graders it links up to symbolic mental arithmetic. This finding may reflect a pre-existing anatomical connection between the cortical regions supporting the quantity and finger-related functions in early childhood. In contrast, first graders exhibit a finger-arithmetic association more influenced by functional factors and cultural-based strategies (e.g. finger counting). Longitudinal data has allowed us to individuate which cognitive functions measured in kindergarteners predicts better the success in mental arithmetic in the first year of school. Results show that finger gnosis, as well as quantity and space–related abilities all concur at shaping the success in mental calculation in first graders. These results are important because, primarily, they are the first to observe a strong relation between visuo-spatial, finger and quantity related abilities in young children, and, secondly, because the longitudinal design provides strong evidence for a causal link between these functions and the success in formal arithmetic. These results suggest that educational programs should include training in each of these cognitive domains in mathematic classes. Finally, specific applications of these findings can be found within the domain of educational neuroscience and for the rehabilitation of children with numerical deficits (dyscalculia).

The present thesis investigates the graphemic stage of the spelling process. Aim of thesis is to study the sub-processes occurring at the orthographic working memory level and the interaction between graphemic representations and working memory that holds these representations active during spelling. Chapter 1, after a brief description of the two-routes spelling model adopted in this research, deals with, presenting neuropsychological evidences, some of the most important issues about the relations between the different levels of elaboration that are engaged in the spelling process. Final part of this chapter is dedicated to the review of the neuropsychological researches regarding the structural organization and the processing of the orthographic representation. Chapter 2 reports the cases of GSI and CRI, two dysgraphic subjects with a selective deficit for consonants and a graphemic buffer disorder (GBD), whose spelling patterns are consistent with the hypothesis that their deficits affect different properties of orthographic working memory: temporal stability for GSI and representational distinctiveness for CRI. Their performance on spelling task demonstrate two things: first, GBD is not a homogeneous deficit because different sub-processes, involved in graphemic buffering, can be selectively affected by cerebral damage; second, different patterns of GSI and CRI arise from interaction of consonant representation and WM properties, both impaired in these subjects. Chapter 3 reports the case of a third dysgraphic subject, PPO, with a selective disorder for consonants but whose spelling picture was not identifiable as a clear GBD. Spelling pattern of this subject, quite different from those of both subjects of Chapter 2, demonstrates that the internal structure of orthographic representation, holding at working memory level, can be selectively impaired in the absence of working memory deficit. Moreover, PPO’s results on spelling task confirm the role of temporal stability and representational distinctiveness in the spelling and the interaction between representations and WM. Finally, Chapter 4 summarizes all the presented results and discusses the implications.

The study of the interactions between olfaction and the decision making processes has mainly focused on the investigation of what is considered the most useful odor to disperse in the air to drive the consumers' choices to prefer a product rather than another one. Despite the fact that some studies showed the existence of associations between odors and prosocial behavior, much less data are available on the links between olfaction and donation in favor of public goods. Thus, the main purpose of the series of experiments described in this thesis is precisely that to shed some light on the investigation of the nature of pleasant odor-decision and on the chemosignal-decision associations. In order to achieve this, the presence of an odor and the congruence between odor and decision task has been manipulated, and the decision to donate has been tested in different domains. Therefore, in the first series of experiments, we manipulated the presence of an ambient pleasant odor, expecting that the congruent stimuli sharing the more stable association odor-concept, would have resulted in an overall increase in the WTC (willingness to contribute) and WTP (willingness to pay) in the decision to donate in favor of a public good. The results confirmed the existence of an association between the olfactory stimulus congruence and the amount donated in favor of a public good. Interestingly, this effect is confirmed in both hypothetical and real decision settings. Moreover, the stimulus modality was manipulated presenting congruent olfactory or visual stimuli. Our findings confirmed the preferential link between olfaction and cognitive processes and showed higher donations in the olfactory setting (compared to the visual one). The existence of associations between odors and words are examined to know if the odors could semantically drive cognitive processes different from decision making. The results confirmed the existence of an association between the olfactory stimulus pleasantness and the performance in a lexical decision task (LDT). Thus, this effect is mediated by the presence of a pleasant odor and not by the semantic congruence between odor and the presented words. Moreover, the visual modality tend to worsen the LDT performance even if the visual stimulus was semantically congruent with the word presented. Finally, even though the semantic link appears to be crucial for cognitive processes such as decision, it seems not so important for memory and linguistic processes involved in the LDT. The second series of studies involved the presence of chemosignal, expecting that the congruity between stimulus and decision (the chemosignal used is AND, known as the best candidate to be considered a human chemosignal) would have resulted in an overall increase in amount of money donate to unknown persons during Dictator and Trust Game. The results confirmed the existence of an association between the chemosignal stimulus presence (and its olfactory experience) and the amount donated, and a mediation effect due to the positive mood (in presence of AND) especially in females and with high AND concentration.

Standard economic models explain decision making under risk as a utility maximization process. Developments in cognitive psychology and neuroeconomics showed the volatility of such conceptualization highlighting human bounded rationality and discussing the role of decision maker’s affective state (basic reactions to any emotionally charged event) in cognitive evaluation of risk (risk as feeling). In particular, an affective-based evaluation of choice options may determine whether decision maker’s behavior will be risk averse or risk seeking. Evidence indicates that affective reactions carry over significant information about the goodness of certain choice options directly influencing risk taking behavior. Affective influences on decisions may be directly associated with the evaluation of the choice options and the anticipation of future outcomes (integral affect) or may be associated with stimuli or event unrelated to the decision at hand, for example contextual factors or environmental cues (incidental affect). A classic advertisement strategy, such as a smiling face presented in association with a good, is an example of contextual affective manipulation. Research shows that experiencing a positive affective state may lead to risk aversion behavior while negative affect may lead to risk seeking. However, previous studies mostly adopted a valence-based approach to the study of affect ignoring its multidimensional nature. In particular, the role of arousal has been largely neglected. Recent studies showed that emotional states with the same valence may have opposite consequences on risk taking. Therefore, the main purpose of the series of studies described in this dissertation was to investigate the effect of inducing incidental affective states at high and low levels of negative arousal or positive arousal on preferences for monetary options varying in risk. Research shows that elevated arousal is associated with cognitive depletion, increased sensitivity to rewards, immediate gratification, less resistance to temptation. Therefore, we hypothesized that affective states characterized by high levels of arousal might increase preferences for the riskier option. We further predicted that including arousing stimuli as contextual factor of a decision scenario would capture individual attention interfering with information processing of risk. In order to achieve this goal, in a first series of experiments we asked participants to make choices between couples of two-outcomes lotteries with the same expected value but different risk. Arousal was manipulated by presenting participants with visual stimuli (IAPS pictures) varying in the level of arousal (high or low) keeping the valence unvaried (negative or positive). By adopting the technique of contextual priming, participants were simultaneously exposed to stimuli (the lotteries) and the contextual factor (the arousing/unarousing image). An effect of arousal on predicting risky choice was found. Probability of selecting the riskier lottery was higher when an arousing stimuli (unpleasant or pleasant) was included as part of the decision context. In some cases, positive arousal was found to interact with gender: risk taking was higher in males than females when a pleasant arousing cue was presented. In a second series of studies, participants performed the same task and, by using an eye tracker, eye fixations and looking times were recorded. The predicted effect of arousal on attention was found. Participants spent more time looking at the arousing image (unpleasant or pleasant). This result is in line with arousal theories which correlate the level of arousal to attention allocated to the arousing stimuli. Furthermore, participants seemed to process less risky information (as indicated by decreased looking times toward the riskier option) when the arousing stimuli was contextually presented, as opposed to when the unarousing stimuli was presented.

This thesis is composed of two individual but interconnected studies. The first study investigated the gestural ability of aphasic patients in comparison with healthy speakers, by analysing both qualitatively and quantitatively co-speech gestures during a story-retelling task. The objective was to understand the relation between language and gesture ability in the aphasic patient: whether impairments in language production influence gesture production, as suggested by the long existing notion “asymbolia†. If this is not the case, gesture then may play a very potential role in aphasic daily communication and rehabilitation, as suggested by researchers and clinicians (e.g., Marshall, 2006; Rodriguez et al., 2006) who proposed the use of gesture as a compensatory and facilitative means to assist aphasic individuals to communicate. In our first study, four aphasic patients and four age-matched healthy speakers were recruited. They were requested to retell stories after watching eight short films from the cartoon “Tweety and Silvester†. Both verbal and non-verbal production from the participant were video-taped for analyses. Group and individual analyses were performed to examine representational and non-representational gestures in per-100-word and per-minute measures. We found that in aphasic subjects, as a group, gestures were quantitatively indistinguishable from those produced by normal controls. Also, qualitative analyses demonstrated that the aphasic subjects tended to use representational gestures to cue or substitute for difficult-to-name words. This supports the notion that gesture may cue naming and may be a potential treatment approach in aphasia rehabilitation. The second study explored treatment efficacy of three approaches in aphasia rehabilitation – the Gesture-based, the Language-based, and the Combined approach, aiming to understand the effects elicited by these techniques in improving single word naming ability in aphasic patients. Previous research suggested that gesture training can facilitate word naming (see Rose, 2006 for review). Language-based treatment aiming to reconstruct concepts and restore phonological information on difficult-to-name words has been widely studied, but the therapeutic role played by gesture in language recovery has been rarely considered. Our second study recruited four chronic aphasic patients with word-finding difficulty to explore the effects of three types of treatment – Gesture-based, Language-based, and Combined, on the retrieval of nouns and verbs. It was hypothesized that gesture and language-based treatments alone would yield positive effects and that combined treatment would result in the largest improvement of single-word naming. In gesture-based treatment, patients were trained to produce a gesture that can be mapped onto a corresponding word. In language-based treatment, Semantic Feature Analysis and Phonological Component Analysis were used. The combined treatment includes the same materials used in the gesture-based and language-based treatments, but materials were alternated across sessions. Training materials included verbs of hand-related actions and nouns of manipulable objects. We found that all types of treatment, as hypothesized, led to significant item-specific improvement in both verb and noun naming. Three of four subjects showed the largest recovery following combined treatment, especially on verbs. This suggests that gesture, when combined with logopedic treatment, can boost naming skills.

Technology has a great impact on our everyday lives; computers, smart devices, sensors and digital technology in general, try to communicate with us to accomplish some task. Each step of the communication however, requires understanding of the future behavioral utterance, deciding on what is the circumstance and the social context, and finally predicting the individual’s needs. Even if computers are so deeply involved in our daily lives, they lack basic social skills that would allow for natural communication. We believe automatic personality recognition will provide computers with an essential social notion, improving the quality of services, such as in intelligent tutoring systems or information retrieval systems among many other uses. Over the past few years, researcher in social computing have shown that personality trait recognition from nonverbal behavior is feasible, yet, the accuracy rate never exceeds a certain level, due to a phenomenon called within-person variability. This means that individuals may vary their behavioral manifestation according to the situational context in which they are in. In this thesis, we propose a shift from the traditional personality trait theory, to an approach which incorporates the personality fluctuations. This new perspective defines personality as dynamic episodes, the so called personality states, which relate to situational factors. Based on this property, we define the notion of social situations and propose a fully data-driven approach based on the Topic Modeling theory. The active situational characteristics that emerge from the model are interpreted according to their interrelation to the personality states fluctuations. We also present an automatic framework based on topic modeling, which handles dynamic spatio-temporal patterns of behavior and aims to predict the semantic meaning of the situational patterns, in meaningful situations, without the need of expert annotators.

How humans make choices in uncertain and competitive situations is a key determinant of viability and successful living. Improving those choices requires sometimes encountering undesirable outcomes and avoiding them, eventually even anticipating them in novel situations. Learning depends on making choices, encountering errors and updating evaluations of options. Various models extended from the reinforcement learning framework compared to human behavior describe in part how individuals heterogeneously make choices. To peer into the components of these mechanisms, strategic games that emulate real-world situations provide measurable and manageable environments in which to examine slight differences in choice behavior among different people. Such differences may be endogenous to participants (e.g. age or learning disposition) while others derive from external events (e.g. emotional induction or brain stimulation). We contrasted such behavior in three situations involving learning or competition, leveraging differences in age, emotional induction and brain stimulation. We aimed to describe the variations in choice behavior across these differences and investigated, when possible, how prior conditions generated a transfer of learning from one domain to another. The work here builds on recent investigations of neural mechanisms underlying choice behavior during strategic or competitive interaction.

The study of cross-modal and cross-dimensional associations in non-synesthetes has become an increasingly hot topic in recent times (see Spence, 2011, for a review). Despite the many examples of associations between shape/color-stimuli on one hand and stimuli of different sensory nature on the other hand, little is known about the way shape and color are interrelated with each other. The purpose of our research was to test whether non-synesthetes also exhibit systematic associations between these two different dimensions of the visual domain, that is, shape and color. Furthermore, our study also relates to art, and in particular to Kandinsky’s theory on shape-color correspondences (1912, 1926), which could be tested indirectly. The project consisted of six experiments, some of them (Experiments 1-3 and 6) carried out in the Experimental Psychology Labs at the University of Trento under the supervision of Prof. Liliana Albertazzi, and the others (Experiments 4-5) carried out in the Visual Perception and Aesthetics Lab at the University of California Berkeley, under the supervision of Prof. Stephen E. Palmer. Experiment 1 provided the first evidence that people tend to match certain hues to certain simple geometric shapes. The strongest relations were found between the triangle and yellows, and the circle and square with reds, confirming only in part Kandinsky’s artistic findings. Experiment 2 replicated the results of Experiment 1 with a different group of participants and showed that the pattern of shape-color associations that previously emerged was independent of a shape’s size, area/perimeter, and stability. Experiment 3 examined the relation between parts of shapes (angles) and hues, also in order to assess if the choice of a hue to be matched with a given shape could be partly driven by its angles. In the remaining experiments (4-5), we extended the previous studies to color dimensions other than the hue, and to an even wider variety of shapes and shape-features. The shape-features studied included pointedness, intersections, symmetry-axes, concavities, and the number of generating-points. We found that specific shape-features of line-shapes (Experiment 4) or closed geometric shapes (Experiment 5) influence specific color attributes (saturation, lightness, redness/greenness, and yellowness/blueness) of the associated colors. Our results also suggested that shape-color associations may be mediated, in part, by emotions; indeed people tend to match colors and shapes that have similar emotional associations (e.g., angry colors are matched to angry shapes). Finally, Experiment 6 assessed in an exploratory way the association between words related to abstract concepts and hues. Altogether, additional examples of associations in the non-synesthetic population were reported. We suggest that this new trend of research, at large, could improve the study of human perception and cognition, guide the search for neural correlates, as well as find possible applications in a variety of disciplines, including ergonomics, art, and design.

The research presented in this thesis addresses the neural mechanisms of auditory motion processing and the impact of early visual deprivation on motion-responsive brain regions, by using functional magnetic resonance imaging. Visual motion, and in particular direction selectivity, is one of the most investigated aspects of mammalian brain function. In comparison, little is known about how the brain processes moving sounds. More precisely, we have a poor understanding of how the human brain codes for the direction of auditory motion and how this process differs from auditory sound-source localization. In the first study, we characterized the neural representations of auditory motion within the Planum Temporale (PT), and how motion direction and sound source location are represented within this auditory motion responsive region. We further explore if the distribution of orientation responsive neurons (topographic representations) within the PT shares similar organizational features to what is observed within the visual motion area MT/V5. The spatial representations would, therefore, be more systematic for axis of motion/space, rather than for within-axis direction/location. Despite the shared representations between auditory spatial conditions, we show that motion directions and sound source locations generate highly distinct patterns of activity. The second study focused on the impact of early visual deprivation on auditory motion processing. Studying visual deprivation-induced plasticity sheds light on how sensory experience alters the functional organization of motion processing areas, and exploits intrinsic computational bias implemented in cortical regions. In addition to enhanced auditory motion responses within the hMT+/V5, we demonstrate that this region maintains direction selectivity tuning, but enhances its modality preference to auditory input in case of early blindness. Crucially, the enhanced computational role of hMT+/V5 is followed by a reduced role of PT for processing both motion direction and sound source location. These results suggest that early blindness triggers interplay between visual and auditory motion areas, and their computational roles could be re-distributed for effective processing of auditory spatial tasks. Overall, our findings suggest (1) auditory motion-specific processing in the typically developed auditory cortex, and (2) interplay between cross- and intra-modal plasticity to compute auditory motion and space in early blind individuals.