Penn Scientists Reflect on One Year of ChatGPT

by Erica Moser

René Vidal, at the podium, introduces the event “ChatGPT turns one: How is generative AI reshaping science?” Bhuvnesh Jain, left at the table, moderated the discussion with Sudeep Bhatia, Konrad Kording, Andrew Zahrt, and Nick Pangakis.

As a neuroscientist surveying the landscape of generative AI—artificial intelligence capable of generating text, images, or other media—Konrad Kording cites two potential directions forward: One is the “weird future” of political use and manipulation, and the other is the “power tool direction,” where people use ChatGPT to get information as they would use a drill to build furniture.

“I’m not sure which of those two directions we’re going but I think a lot of the AI people are working to move us into the power tool direction,” says Kording, a Penn Integrates Knowledge (PIK) University professor with appointments in the Perelman School of Medicine and School of Engineering and Applied Science. Reflecting on how generative AI is shifting the paradigm of science as a discipline, Kording said he thinks “it will push science as a whole into a much more collaborative direction,” though he has concerns about ChatGPT’s blind spots.

Kording joined three University of Pennsylvania researchers from the chemistry, political science, and psychology departments sharing their perspectives in the recent panel “ChatGPT turns one: How is generative AI reshaping science?” PIK Professor René Vidal opened the event, which was hosted by the School of Arts & Sciences’ Data Driven Discovery Initiative (DDDI), and Bhuvnesh Jain, physics and astronomy professor and co-faculty director of DDDI, moderated the discussion.

“Generative AI is moving so rapidly that even if it’s a snapshot, it will be very interesting for all of us to get that snapshot from these wonderful experts,” Jain said. OpenAI launched ChatGPT, a large language model (LLM)-based chatbot, on Nov. 30, 2022, and it rapidly ascended to ubiquity in news reports, faculty discussions, and research papers. Colin Twomey, interim executive director of DDDI, told Penn Today that it’s an open question as to how it will change the landscape of scientific research, and the` idea of the event was to solicit colleagues’ opinions on interesting directions in their fields.

Read the full story in Penn Today.

Konrad Paul Kording is Nathan Francis Mossell University Professor in Bioengineering and Computer and Information Science in Penn Engineering and in Neuroscience in the Perelman School of Medicine.

How the Hippocampus Distinguishes True and False Memories

by Erica Moser

Image: iStock/metamorworks

Let’s say you typically eat eggs for breakfast but were running late and ate cereal. As you crunched on a spoonful of Raisin Bran, other contextual similarities remained: You ate at the same table, at the same time, preparing to go to the same job. When someone asks later what you had for breakfast, you incorrectly remember eating eggs.

This would be a real-world example of a false memory. But what happens in your brain before recalling eggs, compared to what would happen if you correctly recalled cereal?

In a paper published in Proceedings of the National Academy of Sciences, University of Pennsylvania neuroscientists show for the first time that electrical signals in the human hippocampus differ immediately before recollection of true and false memories. They also found that low-frequency activity in the hippocampus decreases as a function of contextual similarity between a falsely recalled word and the target word.

“Whereas prior studies established the role of the hippocampus in event memory, we did not know that electrical signals generated in this region would distinguish the imminent recall of true from false memories,” says psychology professor Michael Jacob Kahana, director of the Computational Memory Lab and the study’s senior author. He says this shows that the hippocampus stores information about an item with the context in which it was presented.

Researchers also found that, relative to correct recalls, the brain exhibited lower theta and high-frequency oscillations and higher alpha/beta oscillations ahead of false memories. The findings came from recording neural activity in epilepsy patients who were already undergoing invasive monitoring to pinpoint the source of their seizures.

Noa Herz, lead author and a postdoctoral fellow in Kahana’s lab at the time of the research, explains that the monitoring was done through intracranial electrodes, the methodology researchers wanted to use for this study. She says that, compared to scalp electrodes, this method “allowed us to more precisely, and directly, measure the neural signals that were generated in deep brain structures, so the activity we are getting is much more localized.”

Read the full story in Penn Today.

Michael Kahana is the Edmund J. and Louise W. Kahn Term Professor of Psychology in the School of Arts & Sciences and director of the Computational Memory Lab at the University of Pennsylvania. He is a member of the Penn Bioengineering Graduate Group.

AI-guided Brain Stimulation Aids Memory in Traumatic Brain Injury

by Erica Moser

Illustration of a human brain
Image: iStock/Ogzu Arslan

Traumatic brain injury (TBI) has disabled 1 to 2% of the population, and one of their most common disabilities is problems with short-term memory. Electrical stimulation has emerged as a viable tool to improve brain function in people with other neurological disorders.

Now, a new study in the journal Brain Stimulation shows that targeted electrical stimulation in patients with traumatic brain injury led to an average 19% boost in recalling words.

Led by University of Pennsylvania psychology professor Michael Jacob Kahana, a team of neuroscientists studied TBI patients with implanted electrodes, analyzed neural data as patients studied words, and used a machine learning algorithm to predict momentary memory lapses. Other lead authors included Wesleyan University psychology professor Youssef Ezzyat and Penn research scientist Paul Wanda.

“The last decade has seen tremendous advances in the use of brain stimulation as a therapy for several neurological and psychiatric disorders including epilepsy, Parkinson’s disease, and depression,” Kahana says. “Memory loss, however, represents a huge burden on society. We lack effective therapies for the 27 million Americans suffering.”

Read the full story in Penn Today.

Michael Kahana is the Edmund J. and Louise W. Kahn Term Professor of Psychology at the University of Pennsylvania. He is a member of the Penn Bioengineering Graduate Group.

Two from Penn Bioengineering Graduate Group Elected to the National Academy of Sciences

Four faculty from the University of Pennsylvania have been elected to the United States National Academy of Sciences (NAS). They are David Brainard of the School of Arts & Sciences; Duncan Watts of the Annenberg School of Communication, School of Engineering and Applied Science, and Wharton School; and Susan R. Weiss and Kenneth S. Zaret of the Perelman School of Medicine.

They join 120 members and 23 international members elected by their peers this year to NAS. Recognized for “distinguished and continuing achievements in original research,” this new class brings the total number of active members to 2,565 and of international members to 526.

Brainard and Zaret are members of the Penn Bioengineering Graduate Group.

David Brainard is the RRL Professor of Psychology, director of the Vision Research Center, and associate dean for the natural sciences in the School of Arts & Sciences. His research focuses on human vision, using both experiments and computer modeling of visual processing, to understand how the visual system deciphers information about objects from light entering the eye. Specifically, he and his lab are interested in color vision, conducting psychophysical experiments to investigate how the appearance of color is affected by an object’s surface properties and ambient light, and how color perception aids in identifying objects. Brainard is the recipient of many honors, including the Macbeth Award from the Inter-Society Color Council, Stein Innovation Award from Research to Prevent Blindness, and Edgard D. Tillyer Award from Optica. He is an elected member of the Society of Experimental Psychologists, a Silver Fellow of the Association for Research in Vision and Ophthalmology, and a Fellow of the Association for Psychological Science.

Kenneth Zaret

Kenneth S. Zaret is the Joseph Leidy Professor in the Department of Cell and Developmental Biology at the Perelman School of Medicine, director of the Institute for Regenerative Medicine, and a member of the Cell and Molecular Biology Graduate Program. His research focuses on gene regulation, cell differentiation, and chromatin structure, with a goal of elucidating these phenomena in the context of embryonic development and tissue regeneration. Pinpointing these aspects of development at the cellular level can serve as the basis for developing future therapeutics and experimental models that further scientists’ ability to understand and cure disease. Zaret has been the recipient of many honors, including a MERIT Award from the National Institutes of Health, the Stanley N. Cohen Biomedical Research Award, and election as a fellow of the American Association for the Advancement of Science.

Read the full announcement in Penn Today.

Grace Hopper Distinguished Lecture: “How Memory Guides Value-Based Decisions” (Daphna Shohamy, Columbia University)

We hope you will join us for the 2022 Grace Hopper Distinguished Lecture by Dr. Jennifer Lewis, presented by the Department of Bioengineering and hosted by Dani S. Bassett, J. Peter Skirkanich Professor in Bioengineering, Electrical and Systems Engineering, Physics & Astronomy, Neurology and Psychiatry.

Date: Thursday, December 8, 2022
Start Time: 3:30 PM EST
Location: Glandt Forum, Singh Center for Nanotechnology, 3205 Walnut Street, Philadelphia, PA 19104

Join us after the live lecture for a light reception!

Daphna Shohamy, Ph.D.

 

Speaker: Daphna Shohamy, Ph.D.
Kavli Professor of Brain Science, Co-Director of the Kavli Institute for Brain Science, Professor in the Department of Psychology & Zuckerman Mind Brain Behavior Institute
Columbia University

Title: “How Memory Guides Value-Based Decisions”

Zoom link
Passcode: 704696

 

Lecture Abstract:

From robots to humans, the ability to learn from experience turns a rigid response system into a flexible, adaptive one. In the past several decades, major advances have been made in understanding how humans and other animals learn from experience to make decisions. However, most of this progress has focused on rather simple forms of stimulus-response learning, such as automatic responses or habits. In this talk, I will turn to consider how past experience guides more complex decisions, such as those requiring flexible reasoning, inference, and deliberation. Across a range of behavioral contexts, I will demonstrate a critical role for memory in such decisions and will discuss how multiple brain regions interact to support learning, what this means for how memories are used, and the consequences for how decisions are made. Uncovering the pervasive role of memory in decision-making challenges the way we think about what memory is for, suggesting that memory’s primary purpose may be to guide future behavior and that storing a record of the past is just one way to do so.

Dr. Shohamy Bio:

Daphna Shohamy, PhD is a professor at Columbia University where she co-directs the Kavli Center for Neural Sciences and is Associate Director of the Zuckerman Mind, Brain Behavior Institute. Dr. Shohamy’s work focuses on the link between memory, and decision-making. Combining brain imaging in healthy humans with studies of patients with neurological and psychiatric disorders, Dr. Shohamy seeks to understand how the brain transforms experiences into memories; how memories shape decisions and actions; and how motivation and exploration affect human behavior.

Information on the Grace Hopper Lecture:
In support of its educational mission of promoting the role of all engineers in society, the School of Engineering and Applied Science presents the Grace Hopper Lecture Series. This series is intended to serve the dual purpose of recognizing successful women in engineering and of inspiring students to achieve at the highest level.

Rear Admiral Grace Hopper was a mathematician, computer scientist, systems designer and the inventor of the compiler. Her outstanding contributions to computer science benefited academia, industry and the military. In 1928 she graduated from Vassar College with a B.A. in mathematics and physics and joined the Vassar faculty. While an instructor, she continued her studies in mathematics at Yale University where she earned an M.A. in 1930 and a Ph.D. in 1934. Grace Hopper is known worldwide for her work with the first large-scale digital computer, the Navy’s Mark I. In 1949 she joined Philadelphia’s Eckert-Mauchly, founded by the builders of ENIAC, which was building UNIVAC I. Her work on compilers and on making machines understand ordinary language instructions lead ultimately to the development of the business language, COBOL. Grace Hopper served on the faculty of the Moore School for 15 years, and in 1974 received an honorary degree from the University. In support of the accomplishments of women in engineering, each department within the School invites a prominent speaker for a one or two-day visit that incorporates a public lecture, various mini-talks and opportunities to interact with undergraduate and graduate students and faculty.

Defining Neural “Representation”

by Marilyn Perkins

Neuroscientists frequently say that neural activity ‘represents’ certain phenomena, PIK Professor Konrad Kording and postdoc Ben Baker led a study that took a philosophical approach to tease out what the term means.

Monitors Show EEG Reading and Graphical Brain Model. In the Background Laboratory Man Wearing Brainwave Scanning Headset Sits in a Chair with Closed Eyes. In the Modern Brain Study Research Laboratory
Neuroscientists use the word “represent” to encompass multifaceted relationships between brain activity, behavior, and the environment.

One of neuroscience’s greatest challenges is to bridge the gaps between the external environment, the brain’s internal electrical activity, and the abstract workings of behavior and cognition. Many neuroscientists rely on the word “representation” to connect these phenomena: A burst of neural activity in the visual cortex may represent the face of a friend or neurons in the brain’s memory centers may represent a childhood memory.

But with the many complex relationships between mind, brain, and environment, it’s not always clear what neuroscientists mean when they say neural activity “represents” something. Lack of clarity around this concept can lead to miscommunication, flawed conclusions, and unnecessary disagreements.

To tackle this issue, an interdisciplinary paper takes a philosophical approach to delineating the many aspects of the word “representation” in neuroscience. The work, published in Trends in Cognitive Sciences, comes from the lab of Konrad Kording, a Penn Integrates Knowledge University Professor and senior author on the study whose research lies at the intersection of neuroscience and machine learning.

“The term ‘representation’ is probably one of the most common words in all of neuroscience,” says Kording, who has appointments in the Perelman School of Medicine and School of Engineering and Applied Science. “But it might mean something very different from one professor to another.”

Read the full story in Penn Today.

Konrad Kording is a Penn Integrates Knowledge University Professor with joint appointments in the Department of Neuroscience the Perelman School of Medicine and in the Department of Bioengineering in the School of Engineering and Applied Science.

Ben Baker is a postdoctoral researcher in the Kording lab and a Provost Postdoctoral Fellow. Baker received his Ph.D. in philosophy from Penn.

Also coauthor on the paper is Benjamin Lansdell, a data scientist in the Department of Developmental Neurobiology at St. Jude Children’s Hospital and former postdoctoral researcher in the Kording lab.

Funding for this study came from the National Institutes of Health (awards 1-R01-EB028162-01 and R01EY021579) and the University of Pennsylvania Office of the Vice Provost for Research.

Using Big Data to Measure Emotional Well-being in the Wake of George Floyd’s Murder

by Melissa Pappas

George Floyd’s murder had an undeniable emotional impact on people around the world, as evidenced by this memorial mural in Berlin, but quantifying that impact is challenging. Researchers from Penn Engineering and Stanford have used a computational approach on U.S. survey data to break down this emotional toll along racial and geographic lines. Their results show a significantly larger amount of self-reported anger and sadness among Black Americans than their White counterparts. (Photo: Leonhard Lenz)

The murder of George Floyd, an unarmed Black man who was killed by a White police officer, affected the mental well-being of many Americans. The effects were multifaceted as it was an act of police brutality and example of systemic racism that occurred during the uncertainty of a global pandemic, creating an even more complex dynamic and emotional response.

Because poor mental health can lead to a myriad of additional ailments, including poor physical health, inability to hold a job and an overall decrease in quality of life, it is important to understand how certain events affect it. This is especially critical when the emotional burden of these events  falls most on demographics affected by systemic racism. However, unlike physical health, mental health is challenging to characterize and measure, and thus, population-level data on mental health has been limited.

To better understand patterns of mental health on a population scale, Penn Engineers Lyle H. Ungar, Professor of Computer and Information Science (CIS), and Sharath Chandra Guntuku, Research Assistant Professor in CIS, take a computational approach to this challenge. Drawing on large-scale surveys as well as language analysis in social media through their work with the World Well-Being Project, they have developed visualizations of these patterns across the U.S.

Their latest study involves tracking changes in emotional and mental health following George Floyd’s murder. Combining polling data from the U.S. Census and Gallup, Guntuku, Ungar and colleagues have shown that Floyd’s murder spiked a wave of unprecedented sadness and anger across the U.S. population, the largest since relevant data began being recorded in 2009.

Read the full story in Penn Engineering Today.

N.B. Lyle Ungar is also a member of the Penn Bioengineering Graduate Group.

Looking Towards the Future Through an Interdisciplinary Lens

by Erica K. Brockmeier

Yasmina Al Ghadban, a senior in the School of Engineering and Applied Science from Beirut, was able to connect her undergraduate education in bioengineering and psychology with her passion for public health through teaching, research, and extracurricular activities. Now, she is poised to leverage her “interdisciplinary lens” towards a future career in public health.

While reflecting on her undergraduate journey at Penn, senior Yasmina Al Ghadban says that she has a “ton of memories” she will take with her: lifelong friends made and skills developed through coursework, research, and teaching experiences, the chance to engage with public health communities on campus, and traveling for courses and internships. “That’s the beauty of Penn,” she says. “There’s just so many opportunities everywhere.”

As a double major in bioengineering and psychology, Al Ghadban, who is from Beirut, has certainly taken advantage of many such opportunities. Now, she is poised to leverage her “interdisciplinary lens” towards a future career in public health.

Problem-solving perspectives

Looking for a place to grow and become more independent, Al Ghadban decided to come to Penn after graduating from the International College in Lebanon. After taking an introduction to bioengineering course during her freshman year, she became enthralled by the hands-on nature of the program and enrolled in the School of Engineering and Applied Science. “I really enjoyed working with circuits and Arduino, being able to synthesize things, and I felt like being in engineering was the place where I was going to gain the most skills,” she says.

Al Ghadban is applying those skills as she completes her senior design project. She and a team of four seniors are building an autonomous robot equipped with Lidar sensors that it uses to create a map of a physical space. The team also programmed their robot to recognize high-touch surfaces that it then disinfects with UV light. “It’s a technology that is completely autonomous, cheaper than what’s on the market, and doesn’t put people at risk when they go in to disinfect,” she says. The team recently put the finishing touches on the project and presented their robot as part of a demonstration on April 14.

In addition to her degree in engineering, Al Ghadban’s interests in public and mental health spurred her to take courses and eventually pursue a double major in psychology, a field that she sees as complementary to engineering. “In psychology, we focus a lot on research and study design, research bias, and these things are similar in engineering and psychology,” she says. “Overall, I think they gave me different perspectives in terms of problem solving, and it’s nice to have that interdisciplinary lens.”

One place where Al Ghadban was able to use this interdisciplinary lens was while working as an research assistant in the Rehabilitation Robotics Lab with Michelle Johnson during her sophomore year. “The focus of the lab is to create robots for post-stroke rehabilitation, and the robotics part is very engineering-focused, but there is another part where people struggle doing the exercises,” she says. “Being able to engage with people and increasing their likelihood of doing that intervention, you rely on a lot from psychology, like interventions from positive psychology or research on how people stay engaged.”

Continue reading at Penn Today.

An ‘Electronic Nose’ to Sniff Out COVID-19

by Erica K. Brockmeier

Postdoc Scott Zhang at work in the Johnson lab. (Photo: Eric Sucar, University Communications)

Even as COVID-19 vaccines are being rolled out across the country, the numerous challenges posed by the pandemic won’t all be solved immediately. Because herd immunity will take some time to reach and the vaccine has not yet been approved for some groups, such as children under 16 years of age, the coming months will see a continued need for tools to rapidly track the disease using real-time community monitoring.

A team of Penn researchers is working on a new “electronic nose” that could help track the spread of COVID-19. Led by physicist Charlie Johnson, the project, which was recently awarded a $2 million grant from the NIH, aims to develop rapid and scalable handheld devices that could spot people with COVID-19 based on the disease’s unique odor profile.

Dogs and devices that can detect diseases

Long before “coronavirus” entered into the vernacular, Johnson was collaborating with Cynthia Otto, director of the Penn Vet Working Dog Center, and Monell Chemical Senses Center’s George Preti to diagnose diseases using odor. Diseases are known to alter a number of physical processes, including body odors, and the goal of the collaboration was to develop new ways to detect the volatile organic compounds (VOCs) that were unique to ovarian cancer.

The next step is to scale down the current device, and the researchers are aiming to develop a prototype for testing on patients within the next year.

Since 2012, the researchers have been developing new ways to diagnose early-stage ovarian cancer. Otto trained dogs to recognize blood plasma samples from patients with ovarian cancer using their acute sense of smell. Preti, who passed away last March, was looking for the specific VOCs that gave ovarian cancer a unique odor. Johnson developed a sensor array, an electronic version of the dog’s nose, made of carbon nanotubes interwoven with single-stranded DNA. This device binds to VOCs and can determine samples that came from patients with ovarian cancer.

Last spring, as the pandemic’s threat became increasingly apparent, Johnson and Otto shifted their efforts to see if they could train their disease-detecting devices and dogs to spot patients with COVID-19.

Continue reading at Penn Today.

N.B.: A.T. Charlie Johnson, Rebecca W. Bushnell Professor of Physics and Astronomy at the Penn School of Arts & Sciences, and Lyle Ungar, Professor in Computer and Information Science at Penn Engineering and Psychology at the School of Arts & Sciences, are both members of the Penn Bioengineering Graduate Group.

Penn Postdoctoral Researcher David Lydon-Staley Appointed Assistant Professor in Annenberg School for Communication

by Sophie Burkholder

A Penn Bioengineer will soon join the Annenberg School for Communication as an Assistant Professor of Communication. David Lydon-Staley, Ph.D., recently completed two years as a Postdoctoral Researcher in Penn’s Complex Systems Lab, led by Danielle Bassett, Ph.D., the J. Peter Skirkanich Professor of Bioengineering and Electrical and Systems Engineering.

David Lydon-Staley, Ph.D.

Lydon-Staley started out studying English and Psychology in his undergraduate education, going on to pursue a Ph.D. from Penn State University in Human Development and Family Studies. What brought him to Bassett’s lab was his interest in using cognitive neuroscience to understand the brain patterns and behaviors behind substance abuse and addiction. There, Lydon-Staley examined networks of nicotine withdrawal behaviors, how those behaviors impact each other, and what information they might hold about how to help smokers in their quit attempts. “David’s breadth of interest is only rivalled by his expansive expertise and bottomless enthusiasm,” says Bassett. “I feel incredibly lucky to have had the chance to work with him.”

In his new role at Annenberg, Lydon-Staley will launch the Addiction, Health, and Adolescence Lab, or “AHA!” for short. “My recent work examines engagement with new media during the course of daily life, and how the information sought and encountered relates to both curiosity and substance use,” he says. Lydon-Staley’s new lab will use methods like experience-sampling and functional Magnetic Resonance Imaging to understand brain and behavior, while drawing on theories and tools from  communication, psychology, cognitive neuroscience, network science, and more.

Even though Lydon-Staley will be working out of a new school at Penn, he still has plans to continue collaborating with the Bassett Lab. One ongoing project he has with the lab involves studying how curiosity works in everyday life, and another looks at moment-to-moment patterns of cigarette withdrawal in daily smokers. “Working in the Bassett Lab gave me the confidence and ability to stretch my wings, chase ideas across traditional disciplinary lines, learn new skills, and collaborate with creative and capable scientists every day,” says Lydon-Staley. Those are opportunities he hopes to keep chasing and fostering in his new position.

Beyond continuing his prior research from a communication-based angle, Lydon-Staley is also excited to develop new classes in the Annenberg School. “Annenberg is a very special place. It is an active school, with frequent seminars and many vibrant research centers,” he says. Informed and inspired by the breadth of research from Annenberg scholars, Lydon-Staley hopes that he can create classes that focus on the psychology of time and timing in everyday life—topics that he spends a lot of time thinking about himself.

Above all, Lydon-Staley is excited by the opportunity to stay at Penn and continue the kind of versatile and multi-faceted studies that have been the bedrock of his research so far. He hopes to continue expanding his previous work with not only the Engineering School, but the School of Medicine and the Graduate School of Education as well. “The opportunities for interdisciplinary collaboration at Penn are unrivaled, and I am constantly in awe of the quality of students here.”