Mind Matters Natural and Artificial Intelligence News and Analysis
model biological micro organism paramecium caudatum 3d illustration

Scientists Try To Understand How One-Celled Life Forms Learn

Artificial intelligence may offer a model for learning without a brain

According to a recent article in The Scientist, in the mid-twentieth century, several labs produced results that suggested that one-celled organisms could learn, in the sense that they could alter future behavior based on past experience. At the time, such findings were dismissed as flukes or mistakes because it was unclear how a unicellular life form like paramecium, with no brain or nervous system, could store memories.

Today, a team from Harvard, Rutgers, and MIT is taking a second look at the findings of learning in paramecium:

We exhume the experiments of Beatrice Gelber on Pavlovian conditioning in the ciliate Paramecium aurelia, and suggest that criticisms of her findings can now be reinterpreted. Gelber was a remarkable scientist whose absence from the historical record testifies to the prevailing orthodoxy that single cells cannot learn. Her work, and more recent studies, suggest that such learning may be evolutionarily more widespread and fundamental to life than previously thought and we discuss the implications for different aspects of biology.

Samuel J Gershman, Petra EM Balbi, C Randy Gallistel, Jeremy Gunawardena, “Reconsidering the evidence for learning in single cells” at eLife (open access) (January 4, 2021)

Surveying the literature from the turn of the twentieth century to the present, the researchers conclude,

Single cells continue to surprise us… Among their many capabilities, it is now appreciated that cells have memory, possibly in the form of a ‘histone code’ (Jenuwein and Allis, 2001; Turner, 2002), though a precise computational understanding of this code has remained elusive. Whatever the memory code may be, its implications for neuroscience are far-reaching: we may finally be poised to link cellular memory codes with cognitive information processing. In this context, the studies by Gelber and others of learning in Paramecia become freighted with significance. They suggest that single cells have the ability to carry out a form of information processing that neuroscientists have traditionally attributed to networks of cells. We still do not understand how Paramecia accomplish this feat. If the hypothesis is correct, then single cells hold more surprises in store for us.

Samuel J Gershman, Petra EM Balbi, C Randy Gallistel, Jeremy Gunawardena, “Reconsidering the evidence for learning in single cells” at eLife (open access) (January 4, 2021)

In short, the mainstream researchers agree that one-celled life forms like paramecium have memory but we don’t know how they manage it.

Others have been looking at the question too. In 2016, researchers reported that a single-celled organism called Physarum polycephalum (slime mold) could learn via a method called habituation:

During a nine-day experiment, the scientists thus challenged different groups of this mold with bitter but harmless substances that they needed to pass through in order to reach a food source. Two groups were confronted either by a “bridge” impregnated with quinine, or with caffeine, while the control group only needed to cross a non-impregnated bridge. Initially reluctant to travel through the bitter substances, the molds gradually realized that they were harmless, and crossed them increasingly rapidly — behaving after six days in the same way as the control group. The cell thus learned not to fear a harmless substance after being confronted with it on several occasions, a phenomenon that the scientists refer to as habituation. After two days without contact with the bitter substance, the mold returned to its initial behavior of distrust. Furthermore, a protist habituated to caffeine displayed distrustful behavior towards quinine, and vice versa. Habituation was therefore clearly specific to a given substance…

This form of learning exists in all animals, but had never previously been observed in a non-neural organism. This discovery in a slime mold, a distant cousin of plants, fungi and animals that appeared on Earth some 500 million years before humans, improves existing understanding of the origins of learning, which markedly preceded those of nervous systems. It also offers an opportunity to study learning types in other very simple organisms, such as viruses or bacteria.

CNRS/Université Toulouse III, “A single-celled organism capable of learning” at ScienceDaily (April 27, 2016) The paper is open access.

Slime molds have been demonstrated to solve mazes in recent years:

Could machine learning provide a model for how one-celled organisms learn?

Key to the conundrum of how creatures with no brain or nervous system can learn is the assumption that memories are stored physically. Perhaps not. Michael Levin of Tufts University suggests an alternative approach that adapts concepts from artificial intelligence:

His group at Tufts University has been studying gene regulatory networks, which control gene expression, in individual cells. In a computational study published earlier this year, Levin and colleagues explored how these networks could shift their responses to certain stimuli or inputs without requiring underlying physical changes—much like how a computer doesn’t need to physically change its hardware when it records a piece of information typed into word processor.

In the simplest version of such a network, genes are assumed to be activated or inactivated by interactions with other genes or by stimuli from the external environment. Memory arises because the current state of genes in the network is dependent on all the interactions and inputs that occurred until now. In some situations the team has studied, this means that the network can be trained to learn certain associations and adapt its future behavior “not because we’ve changed the connections between genes A and B. . . . It’s simply that certain experiences change the overall stable state of the system in a way that changes how it reacts to those stimuli in the future.” Levin says.

Catherine Offord, “Can Single Cells Learn?” at The Scientist (May 1, 2021) The paper is open access.

In short, changes in the state of the life form, resulting in changed behavior — which amounts to learning — need not “be someplace” or weigh something, for the same reasons as a full USB stick doesn’t weigh any more than an empty one. This potential new way of understanding how simple life forms can store and use information is a fascinating direction for research.

Note: Here’s an infographic on the investigation as to whether single cells learn.

You may also wish to read: Why do many scientists see cells as intelligent? Bacteria appear to show intelligent behavior. But what about individual cells in our bodies?


What neuroscientists now know about how memories are born and die. Where, exactly are our memories? Are modern media destroying them? Could we erase them if we wanted to?

Mind Matters News

Breaking and noteworthy news from the exciting world of natural and artificial intelligence at MindMatters.ai.

Scientists Try To Understand How One-Celled Life Forms Learn