CIMeC - Center for Mind/Brain Sciences

CIMeC Colloquium Series

The CIMeC Colloquium Series is an annual set of invited talks given by leading researchers in the mind/brain sciences, both from Italy and abroad, aimed principally at our PhD Students. Given the multi-disciplinary backgrounds of the CIMeC students and researchers, the colloquia are aimed at a general scientific level rather than at a more specialized audience. For school credit, all first, second and third year students attend the Colloquia. First and second year students prepare an essay based on one of the Colloquia of their choice summarizing the Colloquium, critically assessing the claims made and discussing the Colloquium in a broader context.

Find out who’s speaking at the next CIMeC Colloquium

Seminars

Talk

​When: Tuesday, September 26th, 3 PM

Where: CIMeC 3rd floor seminar room, Corso Bettini 31- Rovereto

Speaker: Charley Wu, predoctoral fellow at Max Planck Institute for Human Development in Berlin

​Title: Generalization and Exploration in Vast Spaces

​Summary: Many real world decisions occur in environments with a vast number of possible actions, where under time or resource constraints, optimal solutions are generally unobtainable. How do humans generalize and learn which actions to take when not all outcomes can be explored? From two behavioural experiments and a large-scale comparison 27 models for predicting individual search decisions, we find that a Bayesian function learning model, combined with an optimistic sampling strategy, robustly captures how humans use generalization to guide search behaviour. Taken together, these two form a model of exploration and generalization that leads to reproducible and psychologically meaningful parameter estimates, providing novel insights into the nature of human search in vast spaces. Importantly, our modelling results and parameter estimates are recoverable, and can be used to simulate human-like performance, bridging a critical gap between human and machine learning.

Host: Folco Panizza & Giorgio Coricelli's lab

Transfer-Learning Talk

​When: Thursday, 27th July, 4 - 5 PM

Where: Mattarello, Via delle Regole 101 , Conference Room

Speaker: Carlo Reverberi, Department of Psychology, Università Milano - Bicocca, Milano, Italy

​Title: Long range connectivity patterns reflect progressive learning and global strategy shifts

​Summary: People learn and improve their performance in a task by both incremental optimisation of the implemented strategy or by discovering a new more effective strategy. We re-analyzed a recent fMRI experiment (Schuck et al., 2015) where subjects had to press one of two buttons based on spatial features of the visual stimulus - the instructed strategy - or based on the stimulus color - a more effective but uninstructed strategy. Here we used a new fMRI analysis technique, CPDC (Allegra et al, 2016), to investigate brain network dynamics and its association with learning. Using CDPC with sliding-windows allows to cluster voxels with similar time-series in activation level and to integrate the resulting information in a connectivity network whose nodes are optimised on the basis of the clustering results. CPDC can detect even short-lived (e.g. 20 seconds) coherence clusters. We identify a network of correlated activity involving areas in the visual cortex, the parietal cortex, the precuneus, and the prefrontal cortex. For subjects following the instructed strategy, incremental learning reflected into a progressive strengthening of several network links parallel to a reduction in response times. In the sudden passage from the spatial to the color strategy, we observed a weakening of some network links, including those that previously showed the strongest strengthening. Our results contribute to the current debate on network neuroscience and learning. CDPC proved to be highly effective for tracking network dynamics and could be fruitfully applied to other tasks.

Host: Giorgio Coricelli

CIMeC Seminar

​When: Wednesday, 26th July, 2 - 3 PM

Where: CIMeC 3rd floor seminar room, Corso Bettini 31- Rovereto

Speaker: Franco Pestilli

​Title: Multidimensional encoding of brain connectomes: building quantitative biological networks with preserved edge properties to study the visual white matter and brain aging

​Summary: The ability to map brain networks in living individuals is fundamental in efforts to chart the relation between brain and behavior in health and disease. We present a framework to encode brain connectomes and diffusion-weighted magnetic resonance data into multidimensional arrays. The framework goes beyond current methods by integrating the relation between connectome nodes, edges, white matter fascicles and diffusion data. We demonstrate the utility of the framework for in vivo white matter mapping and anatomical computing by evaluating more than 3,000 connectomes across thirteen tractography methods and four data sets in normal and clinical populations. We show that this framework allows mapping connectivity matrices, edge anatomy, and microstructural properties of the white matter tissue in each connectome edge. The framework is based on statistical evaluation principles introduced with the Linear Fascicle Evaluation and virtual lesions methods (LiFE; Pestilli et al., 2014). In short, instead of building networks by relying uniquely on the terminations of fascicles into the cortex, we exploit the full measured signal available for each connectome edge by extracting a forward-prediction of the biological tissue properties of the edge. We validated the framework by comparing results with standard connectome measures (fiber count and density). To do so, we generated ten repeated-measures connectomes in each individual brain in various datasets, using different tracking methods. For each connectome estimated in an individual, we computed the mean network clustering coefficient across repeated measures. We demonstrate high reliability of the clustering coefficients. We also demonstrate profound differences in connectomes across brains, beyond what can be captured using standard measures (fiber density). We also show that the proposed method is highly sensitive to differences between individuals by improving subject classification into various diagnostic groups. Finally, we show that the framework is useful in clarifying fundamental properties of the human visual white matter as well as identifying useful network science biomarkers for predicting degenerative changes in the Alzheimer's brain. We publish the method with software compatible with data from the Human Connectome Project, the Alzheimer Disease Neuroimaging Initiative, and Indiana Alzheimer Disease Center Data. The software integrates the Brain Connectivity Toolbox and is available open source GitHub.com/brain-life.

Host: Paolo Avesani

Co-Host: David Melcher

Transfer-Learning Talk

​When: Thursday, 20th July, 4 - 5 PM

Where: Mattarello, Via delle Regole 101 , Conference Room

Speaker: Grégoire Borst

​Title: Cognitive control, neurocognitive development and education from childhood to adulthood

​Summary: I will present researches conducted in my lab on the role of cognitive control in neurocognitive development and school learnings from childhood to adulthood. In particular, I will present data showing (a) that the progressive ability to solve logicomathematical problems, to adopt a third person perspective and to generate creative ideas is rooted on the growing ability to inhibit misleading strategies, (b) that cognitive control can generalize over different tasks and can operate outside of consciousness and (c) that the sulcal morphology of the brain, a stable feature of the brain determined in utero and not affected by brain maturation, constraints the development of cognitive control and reading in children, adolescents and adults.

Host: Giorgio Coricelli

Transfer Learning Seminar

​When: Thursday, 13th July, 3.30 - 4.30 PM

Where: Mattarello, Via delle Regole 101 , Conference Room

Speaker: Mehdi Keramati, UCL (UK)

​Title: Planning, under cognitive limitations

​Summary: The computational theory of reinforcement learning offers several ways of evaluating possible actions and making choices. Habitual decision making, forward planning, backward planning, and pruning are some of those strategies that are proven to be particularly relevant to how humans and other animals make decisions. However, each strategy has its pros and cons in terms of cognitive resources and time it consumes. This suggests a competition and cooperation between these strategies in order to make the best use of cognitive resources, given the features of the task in hand. In this talk, I will review several works where we and others studied the involvement of these strategies in decision-making and their competition and cooperation in the face of different cognitive limitations.

Host: Giorgio Coricelli

Talk

​When: Thursday, 13th July, 10.30 - 11.30 AM

Where: CIMeC Corso Bettini 31, Rovereto 1st floor seminar room

Speaker: Matthias Niemeier, Associate Professor at the University of Toronto Scarborough

​Title: Right-brain lateralization for action and perception

​Summary: The corpus callosum is a computational bottleneck that is believed to constitute a driving force of brain organization. Specifically it should cause lateralization of functions underlying tasks that require information integration (a) from and/or for both hemispheres, (b) that is irreducible, and (c) that is computationally expensive. All three criteria apply to the two sets of tasks that I will focus on in my talk. The first is bimanual grasping where two hands are used to prehend one object and where the sight of the object needs to be used to holistically compute the points at which to grasp the object, arguably within one hemisphere. I will show that bimanual grasp control is right-brain dominant: It features a left visual field preference; transcranial stimulation suggests that it relies on grasp- and bimanual coordination-related areas in parietal cortex; and it yields motor priming benefits when people switch from unimanual grasping with their left (but not their right) hand to bimanual grasping. Furthermore, I will argue that similar mechanisms will be employed to compute grasp-like actions jointly performed with another person, consistent with the long-held assumption that sensorimotor control serves as a seed mechanism of social cognitions. The second focus of my talk is perceptual judgments of spatial magnitudes in the left and right visual field. These tasks have garnered interest in the context of functions of spatial attention and visual awareness and reveal deficits in right-brain damaged patients with spatial neglect. However, beyond the observations in patients the precise nature of the intact functions of attention and awareness, often called “pseudoneglect,” remain unknown. I will present new paradigms and effects together with electrophysiological and functional imaging data to suggest that pseudoneglect is associated with a parietal and ventro-frontal network that attains right dominance through interhemispheric competition. Further I will show that pixel noise superimposed onto perceptual judgment stimuli, much like a TV image with poor reception, creates substantial behavioural biases to the left side opposite to spatial neglect. Our research suggests that this noise effect results from visual activation and that it is found for perceptual judgments but not during visual search, another measure of pseudoneglect. Crucially, visual search and perceptual judgments differ in that only the latter requires data-intensive integration of information across hemispheres, perhaps relying on iconic memory buffers in right frontal cortex. Our results suggest that lateralization together with computational affordances serve to attain a more fundamental understanding of the human brain’s functions for action and perception.

Host: Simona Monaco

Transfer-Learning Seminar

​When: Wednesday, 12th July, 2 - 3 PM

Where: CIMeC Mattarello via delle regole 101, Conference room

Speaker: Andrea Brovelli

​Title: Neural and computational bases of instrumental learning in individual and social context 

​Summary: I will present results from fMRI studies on the role of fronto-striatal circuits in the acquisition and early consolidation of arbitrary visuomotor associations (Brovelli et al., 2008; 2011). Then, I will describe a dual-system computational model that can predict both performance (i.e., participants' choices) and modulations in reaction times during learning (Viejo et al., 2015). Finally, I will present results from two fMRI studies investigating the neural circuits that govern associative learning by observation (Monfardini t t al., 2013) and neural modulations due to social comparison effects (Brovelli et al., in preparation).

Host: Giorgio Coricelli

Premesor Talk

​When: Wednesday, 12th July, 11.30 AM

Where: ACN lab - seminar room, II floor, Piazza della Manifattura 1 - Building n. 14

Speaker: Mark E. Hauber, City University of New York and University of Illinois, Urbana-Champaign

​Title: Self- and species recognition in brood parasitic birds

​Summary: Obligate brood parasitic birds, including cuckoos and cowbirds, face critical constraints in their social recognition systems: in the absence of suitable referents (parents and siblings), they must use cues other than social partners to learn about their own species. Here I review 20 years of our research into the behavioral and neural mechanisms of species recognition in brood parasitic birds and contrast these to the species recognition mechanisms of their hosts, focusing on the context of egg and nestling discrimination in parasitized broods.

Host: Giorgio Vallortigara

Transfer-Learning Seminar

​When: Monday, 10th July, 11 AM - 12 AM

Where: CIMeC, via delle Regole 101, Mattarello

Speaker: Kai Görgen, Research Associate, Charité, University Hospital Berlin

​Title: The Same Analysis Approach: Detect, Avoid & Eliminate Confounds in Neuroimaging and other Data Analysis

​Summary: Classical design principles (e.g. randomization) and control analyses (e.g. on behavioural errors, reaction time, age) are routinely applied in many studies. It is typically not tested however whether these work together with new analysis methods, that involving e.g. cross-validation, classifiers, or permutation testing. I will show that – counterintuitively – such standard practices can lead to the exact opposite of what they should achieve: Classical design principles can induce confounds instead of controlling them, and standard control analyses can give false certainty that confounds have been controlled, even if they have not. This can cause systematic positive or negative biases (such as significant below-chance accuracies), potentially yielding false positive results or suppressing real effects. As a remedy, I present "the same analysis approach (SAA)" — a framework to detect, avoid, and eliminate a large class of potential confounds and other potential errors. The main idea is to perform the to-be-employed analysis on (i) design variables, (ii) control data, and (iii) artificial simulations. Although our examples come from neuroimaging, similar arguments apply to other fields such as psychology or machine learning.

Host: Giorgio Coricelli

CIMeC Talk

​When: Thursday, 13th July, 10:30 AM - 11:30 AM

Where: CIMeC Corso Bettini 31, Rovereto 1st floor seminar room 

Speakers: Matthias Niemeier, Associate Professor at the University of Toronto Scarborough

​Title: Right-brain lateralization for action and perception

​Summary: The corpus callosum is a computational bottleneck that is believed to constitute a driving force of brain organization. Specifically it should cause lateralization of functions underlying tasks that require information integration (a) from and/or for both hemispheres, (b) that is irreducible, and (c) that is computationally expensive. All three criteria apply to the two sets of tasks that I will focus on in my talk. The first is bimanual grasping where two hands are used to prehend one object and where the sight of the object needs to be used to holistically compute the points at which to grasp the object, arguably within one hemisphere. I will show that bimanual grasp control is right-brain dominant: It features a left visual field preference; transcranial stimulation suggests that it relies on grasp- and bimanual coordination-related areas in parietal cortex; and it yields motor priming benefits when people switch from unimanual grasping with their left (but not their right) hand to bimanual grasping. Furthermore, I will argue that similar mechanisms will be employed to compute grasp-like actions jointly performed with another person, consistent with the long-held assumption that sensorimotor control serves as a seed mechanism of social cognitions. The second focus of my talk is perceptual judgments of spatial magnitudes in the left and right visual field. These tasks have garnered interest in the context of functions of spatial attention and visual awareness and reveal deficits in right-brain damaged patients with spatial neglect. However, beyond the observations in patients the precise nature of the intact functions of attention and awareness, often called “pseudoneglect,” remain unknown. I will present new paradigms and effects together with electrophysiological and functional imaging data to suggest that pseudoneglect is associated with a parietal and ventro-frontal network that attains right dominance through interhemispheric competition. Further I will show that pixel noise superimposed onto perceptual judgment stimuli, much like a TV image with poor reception, creates substantial behavioural biases to the left side opposite to spatial neglect. Our research suggests that this noise effect results from visual activation and that it is found for perceptual judgments but not during visual search, another measure of pseudoneglect. Crucially, visual search and perceptual judgments differ in that only the latter requires data-intensive integration of information across hemispheres, perhaps relying on iconic memory buffers in right frontal cortex. Our results suggest that lateralization together with computational affordances serve to attain a more fundamental understanding of the human brain’s functions for action and perception.

Host: Simona Monaco

CIMeC PhD Colloquim

​When: Friday, 7th July, 10AM - 11AM

Where: CIMeC, Corso Bettini 31 Rovereto, 1st floor Conference Room

Speakers: Valeria della Maggiore, Director of the Physiology of Action Lab Universidad de Buenos Aires

​Title: Impact of conflicting material on motor learning and consolidation

​Summary: Anterograde interference (AI) usually refers to the impact of prior learning on the ability to successfully learn a subsequent, conflicting task. AI has been unambiguously reported in visuomotor and force-field adaptation tasks. It has even been suggested that AI may be stronger than retrograde effects, thereby masking the effect of interest in retrograde protocols aimed at studying consolidation. Yet, none of the work aimed at examining the magnitude of AI have actually quantified its effect on the speed of learning. In this seminar, I will show recent data from our lab in which we dissociated the impact of prior learning on the initial level of performance from its impact on the speed of learning and long-term memory. Contrary to current views, our results indicate that anterograde effects in visuomotor adaptation do not significantly affect the ability to learn conflicting material. Yet, it hampers memory retention, suggesting that it may however interfere with memory consolidation. Our findings may explain the failure of retrograde protocols to unveil memory consolidation in this type of motor learning.

Host: Luca Turella

CLIC Seminar

​When: Monday, 19th June, 3.30PM - 5.00PM

Where: Seminar room, third floor, CIMeC (Palazzo Fedrigotti), Corso Bettini 31, Rovereto

Speakers: Sudeep Bhatia (University of Pennsylvania)

​Title: Knowledge Representation in Decision Making

​Summary: I discuss how insights from computational linguistics can be used to build models of judgment and decision making with human-like knowledge representations. In addition to specifying the psychological mechanisms people use to form beliefs and preferences, these models also represent the information on which these psychological mechanisms operate. Subsequently, they are able to deliberate over and respond to a large variety of naturalistic decision problems, and moreover, mimic human responses to these problems. These models shed light on the processes at play in everyday decision making, and illustrate a novel approach to predicting real-world behavior.

Workshop

​When: Friday, 16th June, 11.00AM - 6.00PM

Where: Palazzo Istruzione, Corso Bettini 84, Rovereto Aula Magna

Speakers: Leonardo Chelazzi (University of Verona, IT) Giorgio Coricelli (University of Trento, IT; University of Southern California, USA) Jackie Gottlieb (Columbia University, USA) Clayton Hickey (University of Trento, IT) Jane Raymond (Birmingham University, UK)

​Title: ​Motivation, Selection and Information-seeking Workshop

​Summary: Talks will address the role of motivation and learning in cognition with a focus on attention, selection, and information-seeking. Each speaker will give a 40-minute talk to be followed by a 20-minute question period.

PREMESOR Seminar

When: Wednesday, 7th of June, 11.00AM - 12.00PM

Where: ACN lab - seminar room, II floor, Piazza della Manifattura 1 - Building n. 14 Borgo Sacco (Rovereto)

Speaker: Massimo Piattelli-Palmarini - Professor of Cognitive Science, Cognitive Science, Linguistics, & Psychology at the University of Arizona

Title: ​"Normal languages in abnormal brains"

Abstract: There is little doubt that, in the adult, specific brain lesions cause specific language deficits. Yet, brain localizations of linguistic functions are made problematic by several reported cases of normal language in spite of major brain anomalies, mostly, but not exclusively, occurring early in life. The signal cases are hydrocephaly, spina bifida and hemispherectomy. These cases are discussed and possible solutions are suggested: namely a vast redundancy of neurons and/or the role of microtubules as neuron-internal processors and key factors in signaling and guiding the growth and reconfiguration of the brain. ​  

CLIC Seminar

When: Thursday, 18th of May, 3.30PM - 5.00PM

Where:  Meeting Room, second floor, CIMeC (Palazzo Fedrigotti), Corso Bettini 31, Rovereto

Speaker:  Ivan Donadello, FBK (Trento)

Title: Logic Tensor Networks for Semantic Image Interpretation

Abstract: Semantic Image Interpretation (SII) is the task of extracting structured semantic descriptions from images. It is widely agreed that the combined use of visual data and background knowledge is of great importance for SII. Recently, statistical relational learning (SRL) approaches have been developed to deal with reasoning and learning in the presence of rich data and knowledge under uncertainty. Logic Tensor Networks (LTNs) is an SRL framework which integrates neural networks with first-order (fuzzy) logic in an attempt to allow (i) efficient learning from noisy data in the presence of logical constraints and (ii) reasoning with logical formulas describing properties of the data. We develop and apply LTNs to two of the main tasks of SII, namely, classification of bounding boxes and the detection of part-of relations between bounding boxes. To the best of our knowledge, this is the first application of SRL to such SII tasks. The proposed approach is evaluated on a standard image processing dataset. Experiments show that the use of background knowledge in the form of logical axioms improves the performance of the state-of-the-art data-driven approaches in both tasks. Moreover, we show that the use of background logical knowledge adds robustness to the learning system in the presence of  erroneous training data.

CLIC Seminar

When: Wednesday, 10th of May, 3.30PM - 5.00PM

Where:  Seminar Room, third floor, CIMeC (Palazzo Fedrigotti), Corso Bettini 31, Rovereto

Speaker:  Davide Crepaldi

Title: The Reading Brain as a Statistical Learning Machine

Abstract: Morphology's very heart lies in the bridging between form and meaning in the classical definition provided by theoretical linguistics. However, recent results have shown that, at least early during visual word identification, morphological phenomena may arise on the basis of form alone -- genuinely complex words (e.g., DEALER, which is someone who deals) and pseudo-morphological words (e.g., CORNER, which isn't someone who corns) are treated alike [1,2]. The reason behind these surprising set of data is still unclear. Bringing together data from animal research [3,4], positional constraints in morpheme identification in humans [5], and morphological effects in simple word recognition [6], I will illustrate one possible interpretation of these data, namely, that reading builds heavily on general learning mechanisms implemented into the ventral stream (and perhaps more widely in the brain), which capture regularities in letter co-occurrence whose primary origin is morphology itself.

Transfer-Learning Seminar

When: Thursday, 20th of April, 4.00PM - 5.00PM

Where:  Conference Room, CIMeC Via delle Regole 101, Mattarello

Speaker: Adam Zylbersztejn, Assistant Professor, University of Lyon 2 and GATE-LSE

Title: Commitment with and without money: an experiment

Abstract: We experimentally study commitment devices in the trust game with pre-play communication. We first replicate the well-known finding that bare communication does not improve cooperation. Then, we enrich this communication environment by introducing additional mechanisms: (i) monetary commitment for truth-telling achieved through a cost of lying (either "mild" or "deterrent"), and (ii) non-monetary commitment for truth-telling achieved via a truth-telling oath. The deterrent cost of lying induces the highest levels of cooperation, outperforming both the mild cost of lying and the truth-telling oath. The truth-telling oath has a positive effect on cooperation and performs as good as a mild cost of lying. However, combining monetary and non-monetary devices does not bring any improvement as compared to either of these mechanisms. Furthermore, maintaining cooperation over time through communication requires strong monetary incentives for honesty. Commitment devices also change the patterns of communication: there are fewer messages announcing cooperation and higher levels of cooperation conditional on such messages. The intensity of these effects varies across treatments and reaches its peak for the deterrent cost of lying. These data support the hypothesis that external commitment devices dissuade untrustworthy agents from making deceptive announcements.

Host: Giorgio Coricelli

IIT Seminar

When: Wednesday, 19th of April, 2.00PM - 3.00PM

Where:  Seminar Room, third floor, CIMeC (Palazzo Fedrigotti), Corso Bettini 31, Rovereto

Speaker: Jennifer Whitesell, Scientist at the Allen Institute for Brain Science, Seattle, Washington, United States

Title: An anatomical correlate of the mouse default mode network

Abstract: The default mode network (DMN) has been described in humans and nonhuman primates using resting state functional MRI (fMRI), and recently a rodent correlate was observed using fMRI in anesthetized rats and mice. Altered DMN functional connectivity has been observed in a variety of pathological conditions including Alzheimer’s and Parkinson’s diseases. As a first step in understanding the mechanism of this pathological network degeneration, we are examining the structural connectivity underlying the functionally-defined rodent DMN using several viral-based mesoscale connectivity mapping approaches in mice. Using a DMN consensus map based on fMRI data registered to the Allen Institute Common Coordinate Framework, we identified brain regions belonging to the DMN, then performed paired stereotaxic injections of retrograde CAV2-Cre virus and anterograde rAAV expressing Cre-dependent fluorescent protein (eGFP) in pairs of regions inside and outside the DMN. One DMN region, the retrosplenial cortex, possesses at least two classes of target-defined cell types, one that projects preferentially to DMN regions and another that projects to regions outside the DMN, while other DMN regions seem to have only one projection pattern. I will present the results from our ongoing analysis of anatomical connectivity tracing in the context of a functionally-defined network.

Host: Alessandro Gozzi

CIMeC PhD Seminar

When: Wednesday, 12th of April, 10.30PM - 12.00PM

Where:  Seminar Room, third floor, CIMeC (Palazzo Fedrigotti), Corso Bettini 31, Rovereto

Speaker:  Femke T.A. Buisman-Pijlman

Title: Addiction: impacting on pleasure, attention and decision making

Abstract: Addiction is a common problem, but we are still increasing our knowledge of the underlying biological processes that maintain it. Although specific drugs target specific targets in the brain, addiction to a range of drugs results in similar adaptations. Chronic use changes the pleasure from use, attention to drugs cues over other information, decision making and other brain processes. In this lecture we will explore these changes and their underlying biological footprint as well as the research methods that can be used to investigate the different aspects of addiction.

Host: Massimiliano Zampini

CIMeC Seminar

When: Tuesday, 11th of April, 3.00PM - 4.00PM

Where:  Seminar Room, third floor, CIMeC (Palazzo Fedrigotti), Corso Bettini 31, Rovereto

Speaker:  Konstantinos Tsetsos

Title: Selective integration: The mechanistic basis and normative justification of decision irrationality

Abstract: Humans violate rational choice theory but why the do so remains unclear. I will present a new model, dubbed selective integration, according to which decision irrationality stems from an early processing bottleneck that dampens the gain of processing of weaker inputs. This bottleneck, asides from leading to violations of axiomatic decision theory (such as violations of the axioms of transitivity and regularity), can have a positive role by being able to nullify the influence of late noise arising beyond the sensory stage. I will present a series of experiments, in which the ameliorative role of selective integration is confirmed, suggesting that apparently irrational decisions are a side effect of a rational evidence accumulation process. I will close by alluding to the neurochemical basis of selective integration, showing that the pharmacological enhancement of cortical inhibition results in stronger decision irrationality patterns in healthy individuals.

Host: Katya Tentori

MADVIS Seminar

When: Friday, 7th of April, 2.00 PM - 3.30PM

Where:  Seminar Room, third floor, CIMeC (Palazzo Fedrigotti), Corso Bettini 31, Rovereto

Speaker:  Daniel Casasanto, Associate Professor of Psychology and Neuroscience, University of Chicago

Title: Spatializing time and number: How culture shapes cognitive universals

Abstract: People use space to conceptualize abstract domains like time and number. This tendency may be a cognitive universal, but the specifics of people’s implicit space-time and space-number associations vary across cultures. How does culture shape our abstract concepts? In Western cultures, both time and numbers are arranged in people’s minds along an imaginary horizontal line, from left to right, but in other cultures the directions of the mental timeline (MTL) and mental number line (MNL) are reversed. The directions of both the MTL and the MNL have long been assumed to depend on the direction in which people read and write text. In this talk I’ll argue that this assumption is false, and show how different aspects of cultural experience determine the direction of the MTL and the MNL. The distinct experiential bases of the MTL and MNL were predicted on the basis of a general principle, which we call the CORelations in Experience (CORE) principle: People spatialize abstract domains in their minds according to the ways these domains are spatialized in the world.

Host: Olivier Collignon

ACN Seminar

When: Thursday, 6th of April, 3.00PM - 4.00PM

Where:  ACN lab II°Floor, P. Manifattura 1, Rovereto (TN)

Speaker: Gillian Forrester, PhD, Birkbeck, University of London

Title: Hand dominance and the Evolution of Human Cognition: A Cross-Species Perspective

Abstract: The investigation of human cognition benefits from a dual perspective approach, as the evolution and development of modern human abilities are inextricably linked. It is widely acknowledged that humans demonstrate population-level right-handedness, linked with dominant left hemisphere control of language processes. Additionally, it is recognized that a significant majority of mothers cradle their infants with their heads resting on the left arm, associated with a right hemisphere dominance for social-emotional processing. Yet, in both cases, species-unique, causal relationships are debated. The following presentation focuses on behavioural evidence from human and non-human animals supporting a theoretical supposition that evolutionarily old left and right hemisphere dominances for primitive survival behaviours provided a platform for the evolution and development of sophisticated communication and social-emotional abilities in modern humans. Moreover, it is hypothesised that primitive cerebral dominances still play a critical in the typical development of cognition in children.

Host: Giorgio Vallortigara

ACN Seminar

When: Wednesday, 5th of April, 10.00PM - 11.00PM

Where:  ACN lab II°Floor, P. Manifattura 1, Rovereto (TN)

Speaker: Ivo D. Popivanov, PhD, Department of Cognitive Science and Psychology at the New Bulgarian University

Title: Neural representation of bodies in the primate inferotemporal cortex

Abstract: Perception of bodies is an important process which animals (including non-human primates and humans) perform effortlessly every day. The information which is decoded from visual inspection of bodies can be instrumental for survival – e.g. distinguishing bodies of dangerous animals from conspecifics. Furthermore, analysis of body posture can provide various non-verbal signals useful for decoding the emotional state, intentions, attitude, etc. Despite the importance of body perception, the neural mechanisms underlying it are not fully understood yet. During this talk several studies performed in the Laboratory of Neuro- and Psychophysiology in Leuven, Belgium, will be presented aiming at shedding more light on how body perception is performed in the brain. Two distinct brain regions involved in representing bodies were localised by functional Magnetic Resonance Imaging (fMRI) in the inferotemporal cortex of four rhesus monkeys. Then, one of these regions was targeted with electrodes for recording single cell responses to 100 stimuli of different classes (bodies, faces, and artificial objects). The tolerance of the neurons to different image transformations preserving the stimulus identity was also studied. We found that the neurons in the more posterior body patch (midSTS) respond on average higher to bodies than to other stimuli (including artificial objects and faces), however individual neurons had heterogeneous responses, responding only to few body exemplars but never to all bodies (high within-class selectivity). Most of the neurons were tolerant to identity preserving transformations (such as scaling and change of the location within the receptive field). On the other hand, they were not tolerant to planar rotation of the stimuli.

Host: Giorgio Vallortigara

IIT Talk

When: Tuesday, 4th of April, 2.30PM - 3.30PM

Where:  Seminar Room, third floor, CIMeC (Palazzo Fedrigotti), Corso Bettini 31, Rovereto

Speaker: Alexander Thiele, Professor of Visual Neuroscience at the University of Newcastle

Title: Attentional modulation of information exchange between cortical layers and cortical areas

Abstract: Attention improves perceptual abilities and is required for decision making. Many aspects of the effects of attention at the cellular level have been investigated, but how it affects information exchange between different cortical layers and different cortical areas remains poorly understood. We recorded from all cortical layers in macaque area V1 and V4 simultaneously while macaque monkeys performed a cued top-down spatial attention task. In the talk I will delineate how different aspects of feed-forward and feedback signals are routed between layers in V1 and in V4 as well as between different layers of V1 to different layers of V4 (and vice versa).

Host: Stefano Panzeri

CIMeC Talk

When: Monday, 3rd of April, 3.30PM

Where:  Sala Convegni 1°Floor, (Palazzo Fedrigotti), Corso Bettini 31, Rovereto

Speaker:  M.Sc Alexandre Gauvin, SCIL - Université de Sherbrooke, Québec, Canada | GIN - Université de Bordeaux, France

Title: Every (neural) fiber of your being?

Abstract: Tractography is the only method enabling neuroscientists to virtually dissect white matter neural pathways in vivo. Unfortunately this method is highly dependent on the chosen models and parameters. Also, among algorithms available today, none can claim to accurately represent the underlying biology. In fact, there are many challenges to face along the processing pipeline, from data acquisition to tracts analysis. How do we assess the quality and reproducibility of tractograms?

Host: Jorge Jovicich

CLIC Seminar

When: Thursday, 30th of March, 3.30PM - 5.00PM

Where:  Seminar Room, third floor, CIMeC (Palazzo Fedrigotti), Corso Bettini 31, Rovereto

Speaker: Francesca Franzon and Chiara Zanini, University of Padova

Title: The conceiving of (un)countability and its encoding into language.

Abstract: In our experience of the world, concrete entities, objects and substances, are visually perceived as having boundaries. Literature has highlighted the crucial role of boundaries in conceiving an entity and accordingly name it (Prasada et al., 2002; Chesney & Gelman, 2015). However, boundaries are not pertinent when denoting substances, thus references expressed by means of mass nouns and mass morpho-syntax (much butter) are more abstract than references expressed by means of count nouns and count morpho-syntax (a ring). The capability to recognize entities independently from their incidental shape is linked to non–linguistic logical operations such as abstraction, deduction, conservation (Vianello & Marin, 1997). Such abilities are not mature until the age of eight, whereas the acquisition of language is completed at five (Tomasello, 2003). Data from language acquisition (Zanini et al., 2016) will be compared with data collected on adult Italian native speakers. It will be argued that the nature of the differences between mass and count nouns reported in the literature is not totally lexical, and that it could be more economically explained in terms of frequency of occurrence and effects related to extra-linguistic abstraction abilities.

Contact: Sandro Pezzelle

Premesor Talk

When: Wednesday, 22nd of March, 11PM - 12PM

Where:  ACN lab II°Floor, P. Manifattura 1, Rovereto (TN)

Speaker: Dr. Antone Martinho, Fellow by Examination | Magdalen College, Retained Lecturer | Pembroke College, Department of Zoology - University of Oxford

Title: Concept Learning in Imprinting

Abstract: Filial imprinting is the iconic ability of precocious birds to spontaneously learn to identify and follow their mother in the first days of life. While filial imprinting effects a limited portion of a birds’ life history and is limited in the behaviours to which it pertains, recent evidence suggests a richer understanding of this form of learning, including the ability of juvenile birds to learn and represent abstract concepts through the imprinting mechanism. This talk will present some of the research that has evidenced this capability of abstraction and discuss why capacity for abstraction is necessary, even in this narrow learning modality.

Hosts: NGiorgio Vallortigara

CIMeC Talk

When: Thursday, 16th of March, 12PM

Where:  Seminar Room, third floor, CIMeC (Palazzo Fedrigotti), Corso Bettini 31, Rovereto

Speaker: Professor Bharat Biswal, Professor and Chair of Biomedical Engineering, New Jersey Institute of Technology, USA

Title: Dynamic functional connectivity

Abstract: Prof. Biswal is an internationally known researcher recognized for starting the field of mapping functional connectivity of the brain "at rest". In this talk he will present his most recent work characterizing dynamic aspects of functional connectivity.

Hosts: Nivedita Agarwal, Jorge Jovicich

IIT Seminar

When: Tuesday, 7th of March 2017, 3:00PM – 4.00PM

Where:  Seminar Room, third floor, CIMeC (Palazzo Fedrigotti), Corso Bettini 31, Rovereto

Speaker: prof. Jason Lerch, Associate Professor, Department of Medical Biophysics, University of Toronto

Title: Autism: the surprising role of the cerebellum

Abstract: The cerebellum, as traditionally taught, serves to integrate motor outputs with sensorimotor inputs, and thus coordinates smooth and effective movements. Yet more and more evidence is emerging that the cerebellum plays a potentially critical role in autism, a disorder characterized by social deficits and repetitive behaviours. Here I will talk about a series of both mouse and human brain imaging studies that implicate cerebellar alterations in autism and related neurodevelopmental disorders, including its role in brain networks and sensitivity to environmental manipulation.

Host: Alessandro Gozzi

CLIC Seminar

When: Thursday, 2nd of March 2017, 3:30PM – 4.30PM

Where:  Seminar Room, third floor, CIMeC (Palazzo Fedrigotti), Corso Bettini 31, Rovereto

Speaker: Laura Rimell, University of Cambridge

Title: Learning to Negate Adjectives with Bilinear Models

Abstract: We learn a mapping that negates adjectives by predicting an adjective’s antonym in an arbitrary word embedding model. We show that both linear models and neural networks improve on this task when they have access to a vector representing the semantic domain of the input word, e.g. a centroid of temperature words when predicting the antonym of 'cold'. We introduce a continuous class-conditional bilinear neural network which is able to negate adjectives with high precision.

Contact: Aurelie Herbelot

IIT Seminar

When: Thursday 23rd of February 2017, 11PM – 12PM

Where:  Seminar Room, third floor CIMeC, (Palazzo Fedrigotti), Corso Bettini 31, Rovereto

Speaker: prof. Michela Fagiolini, Associate Professor at Children's Hospital of Harvard University

Title: Circuit analysis and novel therapies in neurodevelopmental disorders

Abstract: Our research has identified for the first time a visual phenotype in mouse models of Rett Syndrome and demonstrated its rescue by environmental and genetic manipulation (Durand et al, Neuron 2012). These results have paved the way to the identification of a specific role for Mecp2 in the experience-dependent refinement of cortical circuits by regulating the excitation of pivotal inhibitory neurons (Mierau et al., Biological Psychiatry 2015). The identification of a particular receptor pathway within a specific cortical circuit offers an accessible membrane target for drug intervention strategies that do not rely on the re-expression of Mecp2 itself. Dr. Fagiolini’s laboratory has recently completed a pre-clinical trial of a low dose of ketamine, a FDA-approved NMDA antagonist, in a murine model of RTT (Patrizi et al., Biological Psychiatry 2016). Daily exposure to ketamine reverses deficits in cortical neuronal activity and connectivity in conjunction with significant improvements in general health and survival. Since human studies have also identified abnormalities in NMDA receptor subunits early on in RTT, it is possible that drugs modulating NMDA receptors may be able to prevent or delay the developmental regression that occurs in girls with RTT. The results found in the animal models strongly suggests that visual processing in RTT patients may also be altered and can be used as a robust biomarker of both cortical status and its response to therapy. To this end, Dr. Fagiolini has begun a fruitful collaboration with the Rett Clinic at BCH and the laboratory of Cognitive Neuroscience to assess the cortical function of the visual system in young girls with RTT using visual evoked potentials (VEP) as previously done in Mecp2 mutant mice. Remarkably, they found significant differences between typically developing children and RTT patients (Le Blanc et al., Annals of Neurology 2015) supporting the introduction of standardized VEP analysis in clinical and research settings to probe the neurobiological mechanism underlying functional impairment and to longitudinally monitor progression of the disorder and response to treatment.

Host: Alessandro Gozzi

CIMeC PhD Talk

When: Monday 20th of February 2017, 10:30PM – 12:00PM

Where:  Palazzo Istruzione, Aula 6, Aula Magna, Corso Bettini 84, Rovereto, Province of Trento, Italy

Speaker: Chris Chambers, Professor of Cognitive Neuroscience, School of Psychology, Cardiff University, UK

Title: The Registered Reports project: A vaccine against research bias?

Abstract: In 2013 the journal Cortex became the first outlet to offer Registered Reports, a format of pre-registered empirical publication in which peer review happens prior to data collection and analysis (see https://cos.io/rr/). The philosophy of Registered Reports is that in order to counteract publication bias and various forms of researcher bias (such as p-hacking and hindsight bias), the publishability of a scientific study should be decided by the importance of the research question and rigour of the methodology, and never based on the results of hypothesis testing. In this talk I will provide an update on the progress of Registered Reports at Cortex and beyond, including uptake by more than 40 journals. I will focus in particular on some of the emerging challenges of the format as it has expanded, together with insights it has offered into forms of bias within both research and the peer review process. Together with allied initiatives, Registered Reports are helping to reshape the incentive structure of the life sciences to place transparency and reproducibility on par with conventional indicators of scientific quality.

Host: Francesco Pavani

CIMeC Talk

When: Wednesday 15th of February 2017, 1:00PM – 2:30PM

Where: CIMeC Corso Bettini, 31, Rovereto, Province of Trento, Italy 1st floor meeting room 

Speaker: Aurelie Herbelot (University of Trento)

Title: High-risk learning: acquiring concepts and things from tiny data

Abstract: Humans are able to grasp the meaning of a new word extremely rapidly: often, a single sentence suffices for an educated guess. In a similar fashion, they can build a complex picture of a particular person or object from very reduced information. This extraordinary ability is still out of reach for state-of-the-art computational systems. Whilst the field of distributional semantics has made much progress in modelling the meaning of words and their composition, current systems still require exposure to huge corpora to simulate basic human semantic judgments. In this talk, I'll present a neural model of nonce word acquisition which, given some previously learnt semantic knowledge, can derive a reasonable representation of a new lexical item from tiny data. The strategy used is 'high-risk' in that the system has to trust the informativeness of the provided data, and accordingly update its parameters, in a way that would normally be seen as detrimental over a large corpus. Contrarily to previously proposed methods, this approach is an extension of a standard distributional semantics architecture, which is desirable from the point of view of creating a generic and incremental model of semantic knowledge acquisition. To illustrate this, I will sketch how, in principle, the model can be applied to learning representations of single instances, relating the acquisition of 'concepts' and 'things' in a single framework.

CIMeC Talk

When: Tuesday, 24th of January 2017, 10:30AM – 11:30AM

Where: Corso Bettini, 31, Rovereto, Province of Trento, Italy 1st floor Conference Room

Speaker: Dr. Gesa Hartwigsen, Department of Neuropsychology Max Planck Institute for Human Cognitive and Brain Sciences

Title: Modulation of language networks: Insights from TMS and neuroimaging

Abstract: Language is an elementary mental capability that humans use to communicate. In this talk, I will discuss how transcranial magnetic stimulation (TMS) and functional neuroimaging can be used to characterize interactions, adaptive plasticity and effective connectivity in the healthy language network. The first part of my talk focuses on the use of TMS during a language task to elucidate the contribution of the stimulated area to a specific language function. Here, I will address the role of different key regions for language comprehension on the word and sentence level. The second part of my talk is related to the combination of plasticity-inducing TMS before a task and subsequent functional neuroimaging to elucidate adaptive plasticity during semantic processing. I will show that a TMS-induced perturbation might suppress task-related activity not only in the stimulated area, but in a large task-specific network for semantic processing. Moreover, TMS can also modulate the effective connectivity within this network. Finally, I will discuss how the upregulation of neighboring regions after a TMS-induced perturbation of a semantic key region can contribute to a better understanding of short-term reorganization and plasticity in the healthy language network. As an outlook, I will present novel TMS-fMRI data on the investigation of adaptive plasticity in the reorganized language network in patients with chronic post-stroke aphasia.

Host: Giorgio Vallortigara