一般口演
知覚 / 発達と加齢 / その他
Perception / Development and Aging / Others
座長：天野 薫（東京大学） |
|---|
| 2022年7月3日 11:00～11:15 沖縄コンベンションセンター 会議場A1 第2会場
4O02a1-01
An Information-theoretic Measure To Investigate The Content of Communication Between Simultaneously Recorded Neural Populations
*Marco Celotto(1,2,3), Jan Bím(1,4), Alejandro Tlaie(1), Vito De Feo(1,5), Stefan Lemke(1), Daniel Chicharro(1,6), Malte Bieler(7,8), Ileana Livia Hanganu-Opatz(7), Tobias Donner(9), Andrea Brovelli(10), Stefano Panzeri(1,3)
1. Neural Computation Laboratory, Istituto Italiano di Tecnologia, 38068 Rovereto, Italy, 2. Department of Pharmacy and Biotechnology, University of Bologna, 40126 Bologna, Italy, 3. Department of Neural Information Processing, Center For Molecular Neurobiology (ZMNH), University Medical Center, Hamburg-Eppendorf, Falkenried 94, D-20251 Hamburg, Germany , 4. Datamole, s. r. o, Vítězné náměstí 577-2 Dejvice, 160 00 Praha 6, The Czech Republic, 5. Artificial Intelligence Group, Future Health Technology Lab, School of Computer Science and Electronic Engineering, University of Essex, Colchester, United Kingdom, 6. Department of Computer Science, City, University of London, London, UK, 7. Developmental Neurophysiology, Institute of Neuroanatomy, University Medical Center, Hamburg-Eppendorf, 20251 Hamburg, Germany, 8. Mobile Technology Lab, School of Economics, Innovation and Technology, University College Kristiania, 0152 Oslo, Norway, 9. Section Computational Cognitive Neuroscience, Department of Neurophysiology and Pathophysiology, University Medical Center, Hamburg-Eppendorf, Martinistrasse 52, 20251 Hamburg, Germany , 10. Institut de Neurosciences de la Timone, UMR 7289, Aix Marseille Université, CNRS, 13385 Marseille, France
Keyword: Information transmission, Sensory processing, Partial information decomposition, Causality
A key element for the understanding of how different areas or neural populations in the brain exchange information to produce brain function is the ability to measure both how much information neural populations exchange, as well as what information they exchange. Current methodologies rely on the Wiener-Granger causality principle as a criterion to measure directed information flow between simultaneously recorded neural signals capturing the activity of different neural populations. These methods include both parametric measures, such as Granger Causality, and nonparametric ones, such as information-theoretic Directed Information (DI). However, while these commonly-used measures allow for quantifying the magnitude and directionality of information flow, they provide no insight into communication content.
Here, we define a new measure, which we termed Feature-specific Information Transfer (FIT). FIT allows quantifying the information transmitted between neural signals about specific external target variables, such as a feature of a sensory stimulus. To define FIT, we build on a recent multivariate generalization of Shannon’s information theory called Partial Information Decomposition (PID). PID allows breaking down the joint mutual information that several source variables carry about a target variable into pieces of redundant, unique and synergistic information. Within this framework, we define FIT as the amount of information that is redundant between the past of an emitter neural signal X and the present of a receiver signal Y about a specific stimulus feature S, which is also unique with respect to the past of the receiver Y. Such a definition does, therefore, merge the Wiener-Granger causality principle of information transfer with the content specificity about a target variable of interest via PID.
We validate FIT on simulated data and we compare the results to those obtained using DI, of which FIT is a subpart, to highlight the properties of both measures. Then, we tested FIT on real data to reveal interesting inter-area stimulus-related communication in electroencephalographic (EEG), magnetoencephalographic (MEG) and multi-unit activity (MUA) data. All in all, we found that FIT allows unveiling novel properties of sensory information processing in the brain that cannot be detected using standard measures of information flow. | | 2022年7月3日 11:15～11:30 沖縄コンベンションセンター 会議場A1 第2会場
4O02a1-02
Nicotinic autoreceptors expressed by cholinergic interneurons in the striatum control striatal neuronal activity and explorative behavior
*Helena Janickova(1), Alice Abbondanza(1,6), Martin Modrak(2), Martin Capek(3,4), Sylvie Dumas(5), Veronique Bernard(6)
1. Laboratory of Neurochemistry, Institute of Physiology of the Czech Academy of Sciences, Prague, Czech Republic, 2. Bioinformatics Core Facility, Institute of Microbiology of the Czech Academy of Sciences, Prague, Czech Republic, 3. Laboratory of Biomathematics, Institute of Physiology of the Czech Academy of Sciences, Prague, Czech Republic, 4. Light Microscopy Core Facility, Institute of Molecular Genetics of the Czech Academy of Sciences, Prague, Czech Republic, 5. Oramacell Paris, France, 6. Neuroscience Paris Seine - Institut de Biologie Paris Seine (NPS - IBPS) INSERM, CNRS, Sorbonne Université, Paris, France
Keyword: striatum, cholinergic, nicotinic, interneurons
In the striatum, cholinergic interneurons (CINs) are crucial for the control of signaling and associated behavior. Striatal projection neurons, also called medium spiny neurons (MSNs), are modulated by CINs through muscarinic acetylcholine receptors they express. CINs can also modulate various types of interneurons (INs) that overall represent less than 5 % of striatal neurons. Unlike MSNs, some striatal INs express nicotinic acetylcholine receptors (nAChRs) that increase their activity when stimulated by CINs. However, it is not clear which striatal INs express nAChRs and if these receptors have any significance for behavior. The aim of our study was to determine the expression of beta2 nicotinic subunit by individual types of striatal INs. In addition, we investigated if these receptors control activity of striatal neurons and show any effect on behavior. To this aim, we used double-probe fluorescent in situ hybridization (FISH) to evaluate beta2 expression. By using Cre/loxP approach, we also deleted beta2 nicotinic subunit in striatal INs and tested the effect of the deletion in a battery of behavioral tasks and by measuring c-Fos expression in individual neuronal types. Our FISH analysis revealed that the vast majority of beta2-containing nicotinic receptors in adult mouse striatum is expressed by CINs themselves as autoreceptors. To less extent, we found beta2 expression in parvalbumin- and neuropeptide Y-positive INs. The deletion of beta2 subunit induced anxiety-like behavior in some behavioral tasks and altered behavior in the social preference task where the mutated mice spent more time exploring non-social inanimate object. We also found a slight impairment of discrimination learning in the response-based T-maze task. The deletion increased the overall c-Fos expression in the dorsal striatum and increased sensitivity to a stimulant drug amphetamine. When we investigated the c-Fos expression in individual neuronal types, we found that both MSNs and INs showed c-Fos expression but in INs it was not affected by the beta2 deletion or the stimulant. As an exception, CINs showed a decrease of the c-Fos expression after the beta2 deletion which is in line with the presence of beta2 on these neurons. We conclude that beta2-containing nAChRs are primarily expressed by CINs in the striatum and that the deletion of these receptors affects both striatal signaling and striatal-based explorative behavior in mice. | | 2022年7月3日 11:30～11:45 沖縄コンベンションセンター 会議場A1 第2会場
4O02a1-03
The metacognitive link between detection and discrimination tasks
*Hugo Six(1), Mahiko Konishi(2), Mitsuo Kawato(1), Hakwan Lau(3), Aurelio Cortese(1)
1. Advanced Telecommunications Research Institute International, 2. École Normale Supérieure, Paris, France, 3. RIKEN Center for Brain Science, Wako, Japan
Keyword: Perception, Decision making, Metacognition, Uncertainty
Metacognition is defined as the ability to reflect upon and estimate the quality of our decisions. Studies on its mechanisms have used discrimination tasks (e.g., 2-alternative forced choice) and detection tasks (e.g., yes/no choice to report the presence/absence of a target) interchangeably and sometimes with confusion. Yet, intuitively, the two tasks probably tap into different cognitive mechanisms. This could explain the differences observed between the two tasks at the behavioral and neural levels. Only in a detection task the observer has to determine the probability of a present but missed target, which could rely on counter-factual higher order processes dependent on prefrontal cortex computations. Here, we sought to understand to what extent the mechanisms allowing the execution of a detection task (type 1 task) are similar to those allowing the evaluation of one’s performance (type 2 task). We hypothesized that these decisions share a common metacognitive mechanism. To test this hypothesis, we developed a within-subject, performance-matched design with identical stimuli and trial structure across detection and discrimination tasks. During the detection task, participants had to assess if the motion of a Random Dot Kinematograms (RDK) was coherent (Yes) or just random (No). Instead, in the course of the discrimination task, participants had to perform a forced choice on the direction of the RDK motion, i.e., report Left or Right. In both tasks, following their perceptual choice, participants judged the degree of confidence in their decision. Participants’ brain activity was measured throughout the experiment sessions with functional magnetic resonance imaging (fMRI). Behavioral measurements were analyzed through the lenses of Signal Detection Theory (SDT). We first replicated the strong relationship between task accuracy and confidence, such as significantly higher d’ in high vs low confidence trials. Supporting our original hypothesis, we observed a significant positive correlation between type 1 (d’) and type 2 (meta-d’) sensitivity estimators in the detection task, but not in the discrimination task. Multivoxel pattern analysis of fMRI data allowed us to predict task type and behavior from activity patterns in the prefrontal cortex. These findings reveal that, in humans, the simple act of detecting a target relies on metacognitive processes and therefore differs from a discrimination task as cognitive mechanisms involved are not identical. | | 2022年7月3日 11:45～12:00 沖縄コンベンションセンター 会議場A1 第2会場
4O02a1-04
Characteristics of the aging brain of healthy older adults: A longitudinal study
*Epifanio Bagarinao(1,2), Hirohisa Watanabe(2,3,4), Satoshi Maesawa(2,5), Kazuya Kawabata(2,4), Kazuhiro Hara(4), Reiko Ohdake(3), Aya Ogura(2,4), Daisuke Mori(2), Shuji Koyama(1,2), Masahisa Katsuno(4), Minoru Hoshiyama(1,2), Haruo Isoda(1,2), Shinji Naganawa(6), Norio Ozaki(2,7), Gen Sobue(2,8)
1. Dept Int Health Sci, Nagoya Univ Grad Sch Med, Nagoya, Japan, 2. Brain & Mind Res Ctr, Nagoya Univ, Nagoya, Japan, 3. Dept Neurol, Fujita Health Univ Sch Med, Toyoake, Japan, 4. Dept Neurol, Nagoya Univ Grad Sch Med, Nagoya, Japan, 5. Dept Neurosurg, Nagoya Univ Grad Sch Med, Nagoya, Japan, 6. Dep Radiol, Nagoya Univ Grad Sch Med, Nagoya, Japan, 7. Dept Psych, Nagoya Univ Grad Sch Med, Nagoya, Japan, 8. Aichi Med Univ, Nagakute, Japan
Keyword: aging, longitudinal, reserve, maintenance
The aging brain undergoes atrophic changes even in very healthy individuals. In spite of this, some individuals tend to remain cognitively healthy even in advanced age, whereas others are prone to cognitive decline. Concepts such as cognitive/brain reserve [1] or brain maintenance [2] have been advanced to explain this inconsistency. According to these concepts, individuals with greater reserve or well maintain brain integrity in old age are less vulnerable to cognitive decline. But direct anatomical evidence supporting these hypotheses still remains very limited. Using longitudinal magnetic resonance imaging data from carefully selected healthy older adults, aging 50 to 80 years old at baseline scan, we examined age-related volumetric changes in the brain of this cohort to uncover potential mechanisms differentiating the brains of healthy agers. We used linear mixed effects models to account for both the cross sectional and longitudinal components of these changes.
After accounting for differences in total intracranial volume and sex, our results showed that longitudinally, the rates of tissue loss in the total gray matter and white matter volumes were 2,497.5 mm3 and 2,579.8 mm3 per year, respectively. In addition, the rates of gray matter decline varied across the whole brain with regions in the frontal and parietal lobes exhibiting faster rates of decline, whereas regions in the occipital and temporal lobes appeared relatively preserved. The rates of decline were also not symmetric with many regions in the right hemisphere showing faster declines than those in the left hemisphere. By contrast, cross sectional changes were mainly observed in the temporal-occipital regions. Overall, regions maturing late in development (frontal/parietal) showed more vulnerability to longitudinal decline, whereas those that fully mature in the early stages (temporal/occipital) were mainly affected by cross sectional changes in this healthy older cohort. This may indicate that successful cognitive aging is associated with the development of reserve capabilities in frontal-parietal regions at an earlier age in order to compensate for the unavoidable decline later in life and the maintenance of temporal-occipital regions in old age, consistent with both concepts of reserve and brain maintenance.
REFERENCES
[1] Stern, Y. J Int Neuropsychol Soc 8 (2002) 448 – 460
[2] Nyberg, et al., Trends Cogn Sci 16 (2012) 292 – 305 | |
|
|
