409 research outputs found
Sort by
Sharing is caring: Investigating accountability practices in Italian autonomous state museums
No full textSince 2014, the Italian state museum system has been modified by
— among other innovations — the introduction of the “autonomous
museum” (d.p.c.m.August 29th 2014, n. 171 and d.m. December 23rd
2014, with subsequent amendments). Thanks to the reform, autonomous
museums are accountable for their own management and for the
fulfilment of their mission, are provided with specific management
bodies and internal organization structures, are able to directly manage
revenues and are required to prepare yearly accounting documents.
Public museums operate as hybrid institutions at the intersection
of public administration, cultural management, and heritage
conservation, and they must address multiple accountability demands.
However, empirical evidence on how museums engage with
accountability remains limited, particularly concerning non-financial
reporting and stakeholder engagement.
Drawing from accountability theory, this thesis investigates the
accountability practices of Italian autonomous state museums,
examining how these institutions navigate the tensions between
financial, managerial, and cultural responsibilities. The research
employs a qualitative approach, using case studies of two autonomous
museums to assess their accountability practices through semi-
structured interviews and document analysis.
Findings reveal that while autonomous museums comply with
mandatory financial reporting, voluntary disclosure remains
inconsistent and underdeveloped. The mainly hierarchical approach to
accountability applied by the Ministry of Culture primarily emphasizes
financial oversight and bureaucratic disclosure. The findings are also
analyzed considering the broader constraints and limitations imposed
on state museums by the current framework of autonomy.
By addressing these gaps, this research contributes to the
literature on museum accountability and cultural governance. It
underscores the need for an integrated and multidimensional approach
to accountability that can foster a more meaningful dialogue between
museums, policymakers, and the public
Policy evaluation and machine learning in international economics
Full text linkThis thesis explores innovative empirical models in interna-
tional economics, leveraging machine learning techniques and
a dose-response method to address issues of multidimension-
ality, heterogeneity, and nonlinearity, while exploiting detailed
firm- and product-level microdata.
Firstly, we investigate the capacity of machine learning tech-
niques to forecast the firm’s exporting status. Analyzing com-
prehensive financial accounts and firm-and industry-specific
data from French manufacturing firms (2010-2018), we demon-
strate that machine-learning methodologies can accurately fore-
cast a firm’s exporting status with up to 90% accuracy. Unlike
traditional econometrics, our method handles multidimen-
sional data and exploits it to model non-linear relationships
among endogenous predictors, thus proving a valuable tool
for targeted trade promotion programs.
Next, we assess the heterogeneous impacts of the EU-Canada
Comprehensive Economic and Trade Agreement (CETA) on
French trade using a causal machine learning approach. Em-
ploying a non-parametric matrix completion algorithm rooted
in potential outcome models, we predict multidimensional
counterfactuals at the firm, product, and destination levels,
capturing complex interactions without assuming functional
forms. Using predicted potential outcomes allows us to un-
cover significant heterogeneity in the trade agreement’s ef-
fects, which conventional average effects models might over-
look. Furthermore, our methodology is suitable to evaluate
spillover effects. Within our framework, these manifest as
classical Vinerian diversion effects, wherein trade to Canada partially substitutes for trade outside Canada, especially for
products with a higher elasticity of substitution.
Lastly, we examine the learning-by-exporting phenomenon
by isolating the effect of export intensity on firm productivity
from the endogenous selection into exporting status. Using a
dose-response model that treats export intensity as a contin-
uous treatment affecting firm productivity, we move beyond
traditional binary treatment models to provide insights into
how this relationship evolves across the full spectrum of ex-
port intensity values. Our findings indicate that productiv-
ity gains from exporting are non-linear, with firms needing
to achieve a 60% export intensity threshold to fully capitalize
on knowledge spillovers and effectively compete in interna-
tional markets.
Overall, this research expands the frontier of empirical re-
search in international economics, revealing insights into the
complex dynamics of trade through innovative methodolo-
gies
Aesthetic Pleasure, Computational Models and Brain Activation. The mechanisms behind aesthetic appeal
No full textAesthetic appeal is a peculiarity of human behaviour that involves the
integration of sensory information, individual experiences, cultural
factors, emotional responses, and context, leading to unique and varied
evaluation. Till today there is no consensus of what the main influence
of aesthetic appeal is and if aesthetic value is truly a unique form of
sensory valuation that is encoded separately in the brain from other
forms of value, for example monetary value. In this thesis, we design
several experiments to investigate these two issues. In chapter one, we
explore the various psychological and neuroscience theories of
aesthetic appeal to reveal the current status of the field. In chapter two,
we designed a behavioural experiment including 1,190 artworks and
408 participants to determine if emotional instances, subjective factors
or formal perceptual features have the most influence on aesthetic
appeal. In chapter three, we conducted a behavioural and a fMRI
experiment to investigate if there is a behavioural or neural dissociation
between aesthetic value compared to incentive salience. Incentive
salience was manipulated with a monetary reinforcement paradigm.
The results, indicate that aesthetic value is primarily determined by
participant-specific influences, but formal perceptual features have a
significant but small effect. Also, aesthetic value can be dissociated
from other forms of value (incentive salience due to monetary
reinforcement) both at a neural and behavioural level. Even though
incentive salience had effects on aesthetic value these effects were not
robustly observed across experiments. Lastly, the brain region of the
anterior medial prefrontal cortex is a potential candidate for context
specific value encoding while the ventro medial prefrontal cortex is
candidate for context general value encoding. Altogether, the evidence
suggests that aesthetic value is not simply a conditioned response,
instead aesthetic value is dependent on the insight we gain towards a
stimuli based on highly individual experiences
Sensory Disconnection and Dreaming: The Functional and Phenomenological Impact of Sensory Stimulation During Sleep
Full text linkSleep is often perceived as a state of disconnection from the
environment. Yet, accumulating evidence suggests that the brain can
monitor and process external stimuli even while asleep. The
accompanying subjective experiences, commonly referred to as dreams,
are also thought to be influenced by sensory perceptions. However, the
precise mechanisms through which sensory stimulation affects
dreaming activity remain largely unknown.
This work seeks to address this gap through a comprehensive,
multi-faceted approach. It begins with a systematic review of the
existing literature on the influence of sensory stimulation on dreams,
uncovering key findings and identifying current limitations in the field.
Following this, an experimental study investigates the use of
multimodal sensory stimulation to enhance dream lucidity during
REM sleep, highlighting the potential of sensory-based protocols for
facilitating real-time communication with dreamers and objectively
exploring perceptual awareness during sleep. Finally, the relationship
between multimodal stimulation during NREM sleep and EEG
aperiodic activity is empirically explored, indicating that aperiodic
spectral slopes may serve as informative markers of subjective sleep
experiences.
By integrating theoretical, experimental, and analytical
perspectives, this work aims to deepen the understanding of how
external stimuli influence consciousness during sleep. The findings
contribute to the growing body of knowledge on the dynamic interplay
between the sleeping brain and sensory stimulation, offering valuable
insights into how these interactions shape our dreams
Essays in International Trade: competition, uncertainty and digitalization in a globalized economy
Full text linkGlobalization has significantly impacted global economic growth by enhancing
the flows and integration of goods, human capital, and knowledge.
This process has facilitated the exchange of know-how, promoted
specialization, and led to more efficient resource allocation, thus accelerating
global growth and helping to reduce inequalities between developing
and developed countries. Some of the main drivers through which
globalization pushed economic growth include the reduction of tariff
barriers through extensive free trade agreements, a new organization of
production that has been increasingly fragmented across global supply
chains and the rise of digital technologies. The reorganization of production
along supply chains that span across countries promoted innovation
and technological advancements by enabling technological spillovers.
Many firms managed to become multinationals thanks to cross-border
investments that played a crucial role in boosting productivity and sustaining
competition across markets. Moreover, digital technologies further
supported the integration of the global economy by creating new job
opportunities and innovative tools for conducting business. The digital
revolution has also empowered developing countries, enabling them to
better integrate into the global economy
Spectral signature of Breaking of ensemble equivalence
Full text linkIn this thesis we explore the concept of Breaking of Ensem-
ble Equivalence (BEE) within the context of random graph
models, focusing on spectral properties of adjacency matri-
ces. Our research aims to identify spectral quantities that
can distinguish between different random graph ensembles,
thereby providing new insights into the structure and behav-
ior of complex networks. We cover both theoretical aspects
and practical implications, including simulations and sam-
pling methods for random graph models.
In Chapter 1 we introduce some basic notions of random graph
theory, and discuss how maximum entropy graph models are
fundamental in modeling real-world networks. We explain
what BEE is, what is its characterization in the context of sta-
tistical mechanics, and how it is intimately connected to dif-
ferences that arise naturally between the canonical versus the
microcanonical description of random graph ensembles. In
order to do so, we delve into the spectral theory of random
graphs and use it to investigate BEE.
In Chapter 2 we formulate a conjecture on the equivalence of
measure-BEE and the presence of a gap between the largest
non-centered and non-scaled largest eigenvalues of the adja-
cency matrix in the canonical and the microcanonical ensem-
ble. We prove this conjecture in the setting of homogeneous
graphs.
In Chapter 3 we study the same question for Chung-Lu ran-
dom graphs. In particular, we prove central limit theorems
for the largest eigenvalue and its associated eigenvector.
In Chapter 4 we compute the expectation of the largest eigen-
value for the configuration model, which verifies our conjec-
ture in the setting of inhomogeneous graphs as well.
In Chapter 5 we provide numerical evidence for our findings
through simulation, after a brief introduction to graph sam-
pling. We formulate the main conclusions of our work and
indicate possible further directions of research
Stratification of first episode psychosis based on clinical and neurobiological features: from single-center studies to big data
No full textPsychosis is a common and functionally disruptive clinical syndrome that might be present in many psychiatric, neurodevelopmental, neurologic, and medical conditions. Rather than a nosological entity, psychosis is a syndrome characterized by different symptoms and domains.
Therefore, an increasing amount of pointed out the importance of recognising and treating a first episode as soon as possible. For these reasons, first episode psychosis (FEP) rapidly became a very important population of study and assessment. More than just the first symptomatic presentation of a disease, FEP often shows already some of the features of the advanced psychiatric illnesses, although to a minor extent. On the other hand, great efforts are being made in order to establish an effective intervention, given the fact that early treatment has been proved to ameliorate the course of the disease, ranging from symptoms, relapse, and number of hospitalisations to quality-of-life measures such as involvement in school or work and global functioning.
Given the multifactorial nature of FEP and the different trajectories it can follow (e.g., affective vs. non-affective psychosis), the possibility of predicting future trajectories and
to obtain clear and valid biomarkers is becoming of paramount importance.
Prediction modelling has the potential to revolutionize medicine by predicting individual patient outcome. Early identification of those with good and poor outcomes would allow for a more personalised approach to care, matching interventions and resources to those most at need.
Through a series of studies, we explored: 1) the possibility to stratify FEP patients based on neuroimaging and biological measures; 2) the possibility of use cutting edge machine learning techniques to improve classification and cluster subtypes of FEP patients; 3) the presence of autoimmune features in FEP in a multi-site study I had the opportunity to coordinate as Co-PI (namely the PHLAMES study).
Specifically, in single-site studies we showed that neuroimaging and biological variables can be predictive of the course of the disease. Moreover, in large multi-site bid data analyses we presented how machine learning can improve the prediction of the disorder and help in stratify the risk, using both clinical and neuroimaging data. Finally, in the first results of PHLAMES study emerged that a subsample of FEP with
autoimmune characteristics might be defined. This subsample shows some unique features in terms of neurological symptoms, cognitive deficits, and brain imaging alterations.
The studies presented in this dissertation point out that it is possible to dissect the clinical and biological heterogeneity of psychosis at the beginning of its disease course, by defining meaningful groups of patients and therefore tailor personalized management. In conclusion, these data foster the research for subtypes of FEP and the definition of disease trajectories. These advances might have a great impact on patients’ lives, by defining specific subgroups or progression that benefit of tailored interventions
Exploring the Behavioral and Neural Underpinnings of Nonverbal Affective Communication: Insights from Facial Expressions, and Human Vocalizations in Distinct Vigilance States
No full textThis Thesis offers a thorough exploration concerning the behavioral and brain
functional bases of nonverbal affective communication, with a particular focus
on facial expressions and human vocalizations. The processing of nonverbal
human vocalizations, in particular, was explored during both wakefulness and
sleep, thus giving us the opportunity to get novel insight into the physiological
mechanism responsible for sensory disconnection during sleep. The Thesis
encompasses three distinct studies. In the first study, we developed a facial
motion tracking procedure for the examination of spontaneous affective
expressions and validated it in a group of healthy adult individuals. This tool
allowed us to capture the diversity of subjective emotional states beyond
predefined categories. Utilizing naturalistic video stimuli and data-driven
analytic techniques, this study identified low-dimensional descriptors of facial
configurations that map subjective experience across individuals, suggesting a
potentially universal aspect of emotional expression. The second study
investigated how taking different perspectives (core affect, CA, and perceived
affective qualities, PAQ, ratings) influenced the evaluation of emotional content
conveyed through nonverbal human vocalizations. Results showed a V-shaped
relationship between assessed arousal and valence, consistent with previous
findings in visual, olfactory, and auditory stimuli. While no significant average
differences emerged between arousal and valence between conditions, there
might be variations in the magnitude and variability of ratings between CA
and PAQ evaluation. These findings suggest the importance of careful
consideration when designing experimental instructions and imply that
nonverbal emotional communication may share a common semantic
representation across sensory modalities. The third study deepened into brain
functional responses to affective human vocalizations during both wakefulness
and (NREM) sleep. Brain activity patterns were analyzed using high-density
EEG. We showed that vocal bursts entailing positive or negative valence were
processed differently from neutral stimuli across vigilance states. This finding
supports the notion of a 'sentinel' mechanism that is active during sleep,
whereby the brain remains sensitive to salient cues in the surroundings of the
sleeper. Collectively, the described studies provided new insights into the
complex mechanisms underlying non-verbal human communication
encompassing emotion expression and recognition. All the performed studies
relied on naturalistic stimuli, which better reflect the complexity of real-world
emotional experiences with respect to artificial, simplified stimuli often used in
cognitive studies exploring brain physiology. Our approaches and findings
underscore the importance of studying affective communication in ecologically
valid contexts, laying a foundation for further exploration in this captivating
field
Numerical Modeling and Optimization of Fractured Structures via Machine Learning and Topology Optimization
Full text linkDuring the continuous development of science and technol-
ogy, optimization plays a tremendous role in improving our
resources without compromising the quality of performance.
This thesis work investigates the application of the phase-
field method for fracture (PFF) in brittle materials, focusing
on the understanding of the influence of the model parame-
ters, both for the isotropic and the anisotropic cases, in cap-
turing the mechanical response of experimental results. For
the PFF isotropic case, an experimental investigation was car-
ried out on an ABS co-polymers. A MATLAB-based algo-
rithm combining particle swarm optimization (PSO) with PFF
has been utilized to determine optimal values of Young’s mod-
ulus (E), fracture toughness (Gc), and the PFF internal length
scale (lc) through uni-axial tensile and three-point bending
tests. To understand the potential of bio-polymers in vari-
ous industrial applications, 3D printed PLA materials were
fabricated via fusion deposition modeling, and due to their
anisotropic behavior, an anisotropic PFF approach was ex-
ploited. A metaheuristic machine learning algorithm coupled
with PFF demonstrates robustness in estimating fracture pa-
rameters (Gc, lc, β) and a strong influence of β the penalty
parameter on the predicted force-displacement curves.
The thesis examine also the critical issue of delamination at
internal interfaces/adhesive joints and internal cracks in com-
posite and multi-material components, which can lead to catas-
trophic failures. Existing structural topology optimization
(TO) methods typically assumes perfect bonding, which urges
the development of approaches that explicitly optimize struc-tures against delamination. The proposed data-driven heuris-
tic optimization strategy has been applied to identify optimal
cohesive interface properties with linear grading, enhancing
the composite structure’s resistance to peeling. Additionally,
it explored the application of the Solid Isotropic Material with
Penalty (SIMP) topology optimization approach to optimize
substrate internal structures affected by interface delamina-
tion.
The integration of a phase-field for fracture (PFF) approach
with TO has been highlighted as a robust mathematical frame-
work to mitigate crack progression in structures compromised
by initial damage under operational loads. Employing the
SIMP technique and optimality criteria (OC) method, the re-
search validated its effectiveness through numerical exam-
ples, demonstrating potential improvements in fracture re-
sistance for damaged structures crucial in aerospace, marine,
automotive, and civil engineering industries
Essay on Geography of Production
Full text linkEconomic activities are unevenly distributed in space. The
economic literature has extensively investigated the potential
reasons behind the emergence of cities and local specializa-
tion, on one hand, and foreign trade and investment patterns,
on the other. These essays aim to analyze some of the compo-
nents explaining the spatial variation of economic activities,
leveraging an extensive amount of firm-level data. First, I in-
vestigate the interplay between regional productivity dispari-
ties and local agglomeration advantages. Although I find evi-
dence for agglomeration advantages stemming within Italian
firms’ clusters, I find them to be much smaller than the pro-
ductivity premium for firms located in northern Italy. Sec-
ond, I investigate the rationale for the ownership chains de-
veloped by multinational enterprises across different national
borders. Based on the insight that the placement of subsidiaries
along ownership chains is driven by the existence of commu-
nication costs to transmit management decisions, I develop
a theoretical model for corporate control where parent com-
panies can delegate monitoring activities to middlemen sub-
sidiaries that are located in intermediate jurisdictions. I de-
rive a two-step empirical strategy enabling the structural es-
timation of a gravity equation for foreign investments. Third,
I investigate the ability of European Union regions to retain
foreign investments, evaluating how crucial regional charac-
teristics, such as an R&D-friendly economic environment and
good local institutions, affect the life duration of companies
targeted by foreign investments