DECNEF

Learning is essential to biological organisms, and efficient learning a much-coveted goal in artificial intelligence. In this mini-symposium, we are excited to welcome 3 distinguished speakers who will discuss their recent discoveries on the mechanisms of learning from the perspective of visual perceptual learning. The talks will cover contrasting yet complementary findings, from the role of visual information, sleep (and its various stages), to neural circuits’ excitatory/inhibitory balance in post-encoding processing.

This workshop is supported by AMED under Grant Number JP22dm0307008 and partly by KDDI collaborative research contract.

<Date and Time>

December 16, 2022 (Fri) 13:15-18:30

13:15 – 14:30             Kazuhisa Shibata, RIKEN Center for Brain Science

14:30 – 14:40            break

14:40 – 15:55            Takeo Watanabe, Brown University

15:55 – 16:05            break

16:05 – 17:20            Yuka Sasaki, Brown University

17:30 –  18:30           Open discussion and refreshments

<Venue>

1FWS-A

Advanced Telecommunications Research Institute International (ATR)

13:15 – 14:30      Kazuhisa Shibata, Team Leader, Human Cognition and Learning, RIKEN

Title : Visual perceptual learning as a result of exposure to artificial and natural stimuli

Abstract :
Visual perceptual learning (VPL) refers to performance improvement in a visual task as a result of visual experience. Although almost all VPL studies have used artificial stimuli including Gabor patch and random dot motion, it has been suggested that visual processing of natural scenes/stimuli is different from that of artificial stimuli. Thus, it remains unclear whether VPL of natural stimuli follows the same mechanisms as VPL of artificial stimuli. In this talk, we first review VPL studies that used artificial stimuli. We particularly focus on the roles of attention and reward in VPL as a result of exposure to visual stimuli that are irrelevant to the main task (task-irrelevant VPL). Second, we report our recent findings on task-irrelevant VPL of natural stimuli and discuss possible differences between VPL of artificial and natural stimuli.

Brief Bio:
1999 – 2003  Tokyo University of Agriculture and Technology (Bachelor’s degree, Engineering)
2003 – 2008  Nara Institutes of Science and Technology (Master’s and Doctor’s degree, Science)
2008 – 2009  Advanced Telecommunications Research institutes international (Postdoctoral researcher)
2009 – 2012  Boston University (Research scholar)
2012 – 2014  Brown University (Research scholar, JSPS postdoctoral researcher)
2014 – 2016  Brown University (Assistant professor of research)
2016 – 2018  Nagoya University (Associate professor)
2018 – 2019  National Institutes for Quantum and Radiological Science and Technology (Senior chief researcher)
2019 – Current            RIKEN Center for Brain Science (Team Leader)

14:40 – 15:55      Takeo Watanabe, Distinguished Professor, Linguistic and Psychological Sciences, Brown University

Title : Roles of plasticity and stability in visual learning, revealed by the concentrations of glutamate and GABA

Abstract :
In this talk, I will discuss mechanisms of post-encoding stages of learning. Although mechanisms of encoding of learning have been investigated to a great degree, those in post-encoding stages (e.g., stabilization, consolidation, reactivation) have been less clarified. We have investigated how plasticity and stability occur in post-encoding stages, using paradigms developed in studies of visual perceptual learning (VPL; Watanabe, Nanez &Sasaki, Nature, 2001; Watanabe et al., Nature. Neuroscience, 2002; Seitz &Watanabe, 2003, Nature; Tsushima, Sasaki &Watanabe, 2006, Science; Shibata et al, 2011, Science). By means of magnetic resonance spectroscopy, we obtained an E/I ratio, which refers to the concentration of glutamate and glutamine for excitation, divided by the concentration of GABA for inhibition in human visual areas at different post-encoding stages of VPL. E/I ratios and. performance indicate that plasticity and stability occur in significantly different manners in different post-training stages (Shibata et al, Nature Neuroscience, 2017; Bang et al, Nature Human Behavior, 2018; Tamaki et al, Nature Neuroscience, 2020；Frank et al, Current Biology, 2022).

Brief Bio:
1985                    Dept of Psychology, U Tokyo
1989                    PhD, Dept of Psychology, U Tokyo (as youngest in the whole history of Faculty of Letters in U Tokyo at that time)
1989-1992           Research Associate Dept of Psychology, Harvard
1992-1995           Assistant Prof, Dept of System Engineering, ASU
1995-2012           Assistant to Full Professor, Dept of Psychology, BU
2012-2014           Professor, Dept of Cognitive. Linguistic and Psychological Sciences, Brown University
2014 – Current    The Fred M Seed Chair, Distinguished Professor, Linguistic and Psychological Sciences, Brown University, ATR Invited Director of Neurofeedback Lab, University Paris Visiting Professor, CiNet Invited Researcher, NICT Foreign Advisory Board member

16:05 – 17:20      Yuka Sasaki, Professor, the Department of Cognitive, Linguistic, and Psychological Sciences, Brown University

Title : Sleep and visual perceptual learning

Abstract :

While sleep is a familiar phenomenon, neural mechanisms and functions of sleep are poorly understood. Our lab studies sleep functions in human adults using non-invasive neuroimaging techniques. We are especially interested in how sleep facilitates visual perceptual learning.
Because many factors impact human sleep and vice versa, a strict screening process for participants and controlling contaminating factors are necessary. For instance, shift work, traveling abroad, aging, and mental and neurological disorders affect sleep structures.
Additionally, in human sleep studies, it is often the case that young and healthy participants, who do not have any sleep problems, do not sleep well in the very first experiment. This phenomenon is termed the first-night effect (FNE). Since the FNE is known to the field, researchers have treated the FNE as noise on sleep data. Thus, before investigating the roles of sleep in learning, one must remove the FNE.
In the symposium, I will talk about our recent two studies from our lab. One is about the FNE. We found that the FNE is a manifestation of one hemisphere being more vigilant than the other as a surveillance system. The other is the role of sleep stages in visual perceptual learning. It was controversial whether Non-REM and REM sleep stages have the same function on learning. We have found that each sleep stage has a different role and works complementary to visual perceptual learning.

Brief Bio:
Yuka Sasaki is a Professor at the Department of Cognitive, Linguistic, and Psychological Sciences at Brown University. Yuka was awarded a Ph. D on sleep studies from Waseda University. After working as an instructor or Joshu (at the time in Japanese) at Waseda University, and then Research Fellow at Japan Science and Technology Corporation (Kagicho Tokubetu Kenkyuin), Yuka went to Boston, Massachusetts, USA in 1998 to work at Massachusetts General Hospital with Roger Tootell. It was meant to be just one year of training abroad, but Boston became her home eventually. She was promoted to Assistant Professor at Harvard Med School and then moved to Brown University with Takeo. She studies vision and sleep.