MIT scientists built over 400 tiny ecosystems with 12 to 20 bacterial species; after adding an outsider, the more changeable communities were easier to invade

A new explanation for why some probiotic supplements may not work could lie in a petri dish of soil. A team of MIT physicists created over 400 tiny ecosystems, each containing 12 to 20 different species of soil bacteria, then introduced one new species into each community to see which ecosystems would let the newcomer establish itself. A 2025 study titled “Collective dynamical regimes predict invasion success and impacts in microbial communities,” published in the journal Nature Ecology and Evolution, led by physicist Jiliang Hu and MIT professor Jeff Gore, found that the ecosystems most easily invaded were not the most stable ones, but the most chaotic: communities with constantly fluctuating bacterial populations. In contrast, the more ordered ecosystems resisted the new arrival. This may seem like a narrow finding, but it challenges decades of ecological thinking about biodiversity and invasion, and it has implications for the wellness industry built around boosting gut bacteria.

Biodiversity was meant to protect you

Conventionally, ecologists would expect a highly biodiverse ecosystem to be the hardest to invade. With so many species present, there are fewer opportunities to exploit because the number of available niches is occupied. There is less available space for new species to exploit. However, the MIT team found that this is not always true. Real ecosystems such as coral reefs or grasslands do not always follow this logic.

Some biodiverse ecosystems are very resistant to invasion, and others are overrun. What they found was that it was not the biodiversity that stopped the ecosystems from being invaded, but the stability. If the nutrient levels within their ecosystems were high, the bacteria were competing more for resources. Some ecosystems reached a stable equilibrium, whereas others remained chaotic. Chaotic ecosystems had higher biodiversity but were easily invaded. Stable ecosystems were more resistant.

Why your probiotic gummies may be a waste of money

The MIT researchers have said they now plan to test the invasion theory in the human gut microbiome, which could explain why some probiotics work and others do not. Earlier findings appear to support this idea. A 2018 Cell study led by Eran Elinav and Eran Segal at the Weizmann Institute, titled “Personalized Gut Mucosal Colonization Resistance to Empiric Probiotics Is Associated with Unique Host and Microbiome Features,” found that the effect of an 11-strain probiotic supplement varied widely from person to person, even though the strains were detected reliably in stool samples. Direct testing of the gut lining itself, however, told a different story: some individuals’ gut linings readily took up the new bacteria, while others’ showed little to no colonisation.

Taken together, the MIT and Cell findings suggest that a stable gut may simply be harder to invade. A probiotic supplement may be struggling to establish itself in a stable gut ecosystem, and there is currently no easy way to tell which category any individual falls into.

When it's actually useful

There is, however, a scenario where an unstable gut is exactly what’s needed: fecal microbiota transplants (FMT), used to displace a harmful resident species. FMT is now an FDA-approved treatment for recurrent C. difficile infections. A 2017 randomised clinical trial in JAMA, “Effect of Oral Capsule– vs Colonoscopy-Delivered Fecal Microbiota Transplantation on Recurrent Clostridium difficile Infection,” found that both capsule- and colonoscopy-delivered FMT resulted in approximately 90% of patients remaining free of relapse. In these cases, doctors are intentionally destabilising the gut ecosystem to let healthy donor bacteria take hold.

The main takeaway

Whilst none of this has been proven in real human guts, the formula the MIT team developed is based entirely on bacteria and soil. The researchers have said their next step is testing it in real-world ecosystems. That said, it is an uncomfortable finding for an entire industry built around increasing the number of bacteria in the gut. It seems that gut resilience may depend more on how stable the ecosystem is than on how much bacterial activity it contains.