Memory Lane

Your memory holds a vast trove of information. Surprises can come to us unbidden, like the lyrics to a song not heard in decades or the click of a knee that recalls a long-ago accident. Other memories are more frequently called upon—the route home, a coworker’s name—or exist at the periphery, an infrequently acknowledged part of our foundation, like our family history or facts and figures learned in high school.

But what makes a memory real? How can we tell, and does it matter? How are our minds and bodies marked by our experiences, and how do our social, cultural, and political moments shape what we remember? 

“Perception and memory are built upon what you already know,” says Mark Liberman, Christopher H. Browne Distinguished Professor of Linguistics. “When you’re learning something new, you are retrieving information from your past. Every instance of stimulus and response is interdependent.” 

With language acquisition, for instance, that can mean mapping the structure of a known language onto the structure of one that’s being learned, or a toddler thinking that all words have two repeated syllables (like “mama” or “dada”). Sometimes, people form and recall memories in ways that support what they believe about the world, framing or reframing history. Still other times, this process indicates more, a harbinger of something malfunctioning in the brain. 

Then there is collective memory, which is not about cognitive processes or individual experiences, but rather a set of repeated narratives or tropes. “The types of stories people tell about the past get mobilized for different kinds of social and political projects,” says Anna Lehr Mueser, a PhD candidate in history and sociology of science studying collective memory about a New York watershed. “Memory helps us understand how people feel about the past, how they feel about the present, and what they might imagine about the future.” 

Scholars across Penn Arts & Sciences are exploring memory through the lens of their own disciplines. They’re using tools from archival records and ethnographic studies to AI-guided brain stimulation. What they are learning has the potential to protect our memories in the face of injury or aging and change how we understand our own minds and bodies in the context of our communities and shared histories. 

Finding the Words 

Our memory is always at work, storing and retrieving information essential to our everyday functions. Communication falls into this category. “The most extraordinary thing, and the most central thing, about human language is the large number of words that people know,” says Liberman, who runs Penn’s Linguistic Data Consortium (LDC). “Normal speakers of pretty much every language know between 10,000 and 100,000 different words.” 

What’s more, “people learn almost 100 percent of their words by hearing them used in context by other people,” he says. “The rate of learning is surprisingly high and the amount of required exposure is surprisingly low. Do the math: If you’ve learned 40,000 words by the time you’re 17 or 18 years old, then on average, you’ve learned about 10 words a day.” 

Watching a child go from a babbling baby to a chatty toddler involves a certain amount of awe. But once conversant, we tend to take our linguistic recall for granted—until, that is, words start to disappear. Forgetting a name or place can cue commonplace worries about aging, sometimes indicating real problems like developmental or degenerative disorders. According to Liberman, the majority of such neuropsychological diagnostic tests are linguistically mediated and involve memory, like asking a participant to listen to a story and then repeat details.

Memory helps us understand how people feel about the past, how they feel about the present, and what they might imagine about the future. 

Such testing has its advantages: It doesn’t require more complex measures like an MRI or EEG to observe or measure symptoms. But there are limits, too, including the immense variation in what’s considered “normal” language and recall. Give any group of college undergraduates one of these neurophysiological tests, Liberman says, and you’ll get a range of results. Everyone’s baseline is different, he adds, meaning for one of these tests to give a definitive diagnosis of, say, neurological decline, the person would have to score incredibly low, and that wouldn’t transpire unless the decline had been happening for a long time.

Liberman, with colleagues at LDC and Penn Medicine, has been working to improve these tests by building a database that tracks neural health across time, enabling doctors to make earlier diagnoses and researchers to evaluate medications and other treatments. This project, called the Speech Biomarkers Study, has already gathered data from a range of participants. Researchers involved have published on testing and diagnostics for disorders including Alzheimer’s disease. 

One key finding is that the standard battery of tests is too short to account for individual variation. Higher volumes of data collected over a longer period could yield more accurate, timely results. The researchers are considering ways to motivate people to participate in tests that measure language and memory, like through an app. 

“You have to make it fun,” Liberman says. “People don’t do it just because it’s going to tell whether they’re having cognitive problems, because that’s no fun. But they might be willing to do it if it involved playing a game, right? Gamifying these tests is a direction worth pursuing.”

Neural Insight

While Liberman is studying memory formation and loss from the outside, Michael Jacob Kahana is coming at it from the inside—literally. The professor of psychology has a habit of looking into people’s brains and finding surprising things. He studies the neural basis of human memory using behavioral, computational, and neurophysiological methods. In addition to revealing the neural mechanics of brain activity, he researches and develops therapies to restore memory in patients with brain injury or neurological disease.

Image

“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,” says Kahana, who is director of the Computational Memory Lab. “Memory loss, however, represents a huge burden on society. We lack effective therapies for the 27 million Americans suffering.” 

In 2023, Kahana and a team of neuroscientists published a paper in the journal Brain Stimulation showing that targeted electrical stimulation in patients with traumatic brain injury (TBI) led to an average 19 percent boost in recalling words. They determined this by studying TBI patients with implanted electrodes, analyzing neural data as patients studied words, and using a machine learning algorithm to predict momentary memory lapses.

Another study Kahana and colleagues published that same year, in Proceedings of the National Academy of Sciences, showed that electric signals in the human hippocampus—the brain’s memory and learning center—differed immediately before recollection of true and false memories, the latter being exactly as they sound, where the brain incorrectly fills in the blanks on a situation similar to, but not the same as, one that happened in the past. 

In Kahana’s study of true and false memories, epilepsy patients already undergoing invasive monitoring to pinpoint the source of their seizures agreed to have their neural activity recorded. They then studied a list of unrelated words and were distracted before being asked to recite the words they’d seen. 

When you’re learning something new, you are retrieving information from your past. Every instance of stimulus and response is interdependent.

Kahana and colleagues analyzed patterns of electricity generated in the hippocampus, capturing brain activity leading up to the recall. Beyond noting a difference in electrical activity in the hippocampus before correct or false memories, the researchers also predicted that the neural activity would reflect how similar to each other the memories were—something further work did confirm. Because conditions such as post-traumatic stress disorder involve false or incongruous recalls that cause distress, these novel insights about memory retrieval in the brain could lead to new interventions or clinical treatments.

Embodied Histories

The work from Kahana and Liberman involves measuring recall or reactions in the moment. But sometimes, events or experiences leave a mark on our minds and bodies that continue to affect us even if the memory has faded. How do these experiences reverberate across time? Historian Kathleen Brown is trying to answer that question in the context of American slavery and its legacy.

“We all have very personal histories of our bodies,” says Brown, David Boies Professor of History. “We carry scars from surgeries or, in some cases, traumatic memories that make us go sweaty, make our hearts beat faster, make us approach certain life events with great foreboding. Those personal histories, these memories, are always in a dynamic interplay with the bigger historical context.”

While a faded scar from a skinned knee might not be one for the history books, what about the physical relics of growing up drinking lead-contaminated water, living in a war zone, or surviving a wildfire in an area affected by climate change? Situations like that, inseparable from the big picture of policy and legislation, impact the body long after the immediate experience ends, Brown notes. It’s one of the reasons her recent work centers on the body, casting it as a dynamic and consequential record of what happened in the past.

Image

Brown sees connections, for example, between current discussions of trauma and memory and how 19th-century abolitionists focused on slavery’s effect on the human body and psyche. “Today, we think deeply about how experiences of trauma can have a lasting impact,” says Brown, lead faculty historian for the student-run Penn & Slavery Project. People fighting against slavery, she adds, saw the body as formed by its circumstances, so enslavement, then, was an intolerable set of physical constraints, injuries, and harms done to people who might otherwise thrive.

The experience could leave its mark on an individual’s body and mind, from injury, brutal punishment, malnutrition, or the trauma of family separation. Even after a body healed or a person escaped enslavement, the memory of it would remain. And more broadly, Brown says, slavery’s legacy left and continues to influence American history. She explains that racist scientific beliefs propped up legal structures that harmed and limited the movement of Black bodies—from slavery being inheritable to Jim Crow–era laws that enforced segregation—allowing the memory of enslavement and its afterlives to reenact and reinflict on people’s bodies throughout the 20th century.

We are imprinted by our history.

To that end, it is perhaps not surprising, she adds, that if you map out regions of health disadvantage in the United States, places where people are more likely to suffer from diabetes, drug use, and mental health crises, and are more susceptible to lack of dental care and general healthcare access, this closely follows historical maps of slavery and Native American forced migration to reservations.

“It bears our attention as historians to understand how the world of health outcomes has been made historically,” Brown argues. “They didn’t just happen. They are the product of histories that have created patterns of wealth and poverty and diet and access to healthcare. We are imprinted by our history.”

Memory As Contested Object 

Anna Lehr Mueser’s personal backstory includes working at an environmental nonprofit, where she frequently heard arguments against horizontal hydrofracking in New York State. The popular refrain was that New York’s agricultural legacy needed protecting and therefore couldn’t evolve to include fracking. 

“But of course, you could also look at New York’s history and say it’s an incredibly industrial state, right?” says Mueser, a doctoral candidate in history and sociology of science. “There are many different things in the state’s history, and I realized that its agricultural past was a really helpful motivator for the activists. It was a memory that had power in that time and place.” 

Now, Mueser’s PhD research focuses on collective memory as it relates to people living in the New York City watershed, located about 120 miles away from the city in the Catskills. Villages were destroyed to construct the reservoirs and today, activities including farming are highly regulated to protect the water supply. 

Alongside archival work at state and local institutions, Mueser interviews farmers in the region, asking broad questions about their daily practice and how they work within the complex watershed rules.

“Emotions and familial connections—often going back generations—come up frequently in interviews. There is also a sense of ever-present political engagement,” says Mueser. “In the 1990s, farmer protests and citizen activism led to a nonprofit, funded largely by New York City, that helps farms stay financially viable in the face of watershed regulations that limit agricultural activity.” 

In her research, Mueser contends that the reservoir’s construction was just one phase in ongoing negotiations between residents of the Catskills and residents of New York City, between their needs, rights, pasts, and presents. Once the reservoirs were in place, the difficult act of living with them began. Knowing how people understand the past can guide decisions in the present, she says.

“Collective memory is important when we look at ongoing politics and policy decisions, say, about where a dam might be sited or a landfill located,” Mueser says. “The kinds of stories that activists, lawmakers, and different stakeholders tell are shaped by memories about what kind of place that potential site is. They’re activating these stories in moments in which the past becomes a special kind of evidence.”

Lauren Rebecca Thacker is a Providence-based higher ed writer who covers topics from poetry to politics. Additional contributions by Erica Moser.