Social partner cooperativeness influences brain oxytocin transcription in Trinidadian guppies (Poecilia reticulata)

Sylvia Dimitriadou, Eduarda M. Santos, Darren P. Croft, Ronny van Aerle, Indar W. Ramnarine, Amy L. Filby, Safi K. Darden

Preprint posted on March 02, 2021

Oxytocin paradox: New research shows that fish change their brain chemistry when they experience cooperation - but not in the way you might expect

Selected by Sophia Friesen

Background and context:

What was the last thing you achieved entirely on your own, without any advice, support, or outside resources? Cooperation is fundamental to most human endeavors, but from an evolutionary perspective, cooperation is a tricky task. For non-kin cooperation to be advantageous, it must be reciprocated, which means that when organisms are the recipient of altruism, they must change their behavior. Recent research by Dimitriadou et al. focused on the biological bases for these changes in a cooperative fish, the Trinidadian guppy. When a shoal of guppies detects a potential threat, one or more guppies may separate from the group to approach the threat more closely and return with information on its danger level, a behavior known as “predator inspection”. This is a dangerous task, but if multiple guppies go at the same time, they can share the risk – taking on danger to benefit each other.

The researchers’ key candidate for adapting brain chemistry to cooperative behavior was the peptide hormone oxytocin. Oxytocin has been implicated in social behaviors and emotions in humans as well as in fish. A dose of oxytocin makes humans trust each other more when social cues are present, although intriguingly, when such cues are absent, oxytocin actually decreases cooperation (1). Oxytocin is also elevated when cleaner wrasse detect “client” fish with which they can form a mutualistic relationship (2).

The researchers investigated how the experience of cooperation in threatening situations altered the guppies’ behavior and brain oxytocin transcription. Based on previous research in humans and fish, they expected that cooperation would increase oxytocin transcription – but the results were more complicated than that.


Key findings:

Fish from a low-predation population approach threats more closely when they have a partner

To control for the experience of social behavior, independent of cooperation under threat, the researchers performed all assays in fish from two populations. The “high-predation” population was taken from a region with many guppy predators; high-predation guppies perform predator inspection under conditions of high risk, and having a partner share that risk is an instance of cooperation. In the “low-predation” population, from a region with few guppy predators, leaving the shoal is not risky and sharing that task is therefore not cooperative.

Fish from the low-predation population approached a visual stimulus – either a replica of a predatory fish, or an innocuous plastic plant – more closely when they perceived that they had a partner. The researchers simulated partner presence by placing a mirror along one side of the experimentation tank, such that the guppy’s reflection accompanied it on its journey. Interestingly, how closely high-predation fish approached the stimulus was unaffected by whether or not they had a partner.


Fish from a high-predation population change oxytocin transcription based on cooperation and threat level

After guppies performed predator inspection (or plastic plant inspection) with or without a simulated partner, the researchers harvested their brains and measured the level of oxytocin transcription in the midbrain. In the low-predation population, oxytocin transcription was unaffected by partner presence, suggesting that other neurobiological changes drive the observed behavioral changes. However, in fish from the high-predation population, oxytocin transcription increases with partner presence – when the fish are inspecting a low-threat plastic plant. Unexpectedly, when the fish inspects a threatening replica of a predator, the presence of a partner actually decreases oxytocin transcription, the opposite of what the researchers predicted.


Why I liked this paper:

Behavior is a complicated thing to assess, so I loved the use of simplified stimuli that provoked an appropriate response while minimizing unwanted variation. Where a live predator could behave unpredictably, a realistic replica is both predictable and convenient. Similarly, I thought that using a mirror to simulate cooperation with a partner was ingenious! Instead of relying on the unreliable behavior of a freely moving guppy shoal, the mirror (combined with a compartment to restrain the majority of the shoal) allowed the researchers to simulate cooperation in a consistent, standardized way.

I also appreciated that the results of this paper were unexpected! Oxytocin has been branded as the “love hormone”, but social behavior is such a complicated phenomenon that it makes sense that oxytocin would play a more nuanced role. On a topic that’s often sensationalized, I liked that the researchers avoided unnecessary speculation and emphasized that there’s still a lot we don’t understand.


Image source: Per Harald Olsen.


Questions for the authors:

  1. It’s interesting that partner presence leads to a transcriptional change only in high-predation fish, but leads to a behavioral change (distance from the stimulus) only in low-predation fish. Do you expect that there are partner-dependent behavioral changes in the HP fish that you might not have measured in your assay?
  2. When using a mirror to simulate the presence of a partner, is it ever possible for the fish to see a mirror image of the stimulus as well? In other words, could the fish in the partner situation perceive two predators?
  3. While the transcriptional changes you observed occurred quite rapidly, they are (presumably) not fast enough to alter behavior during a single instance of predator inspection. Has any work been done on how predator inspection (with or without a partner) impacts fish behaviors at later times, or is that work that you would be interested in doing?



  1. Declerck CH, Boone C, Kiyonari T. 2010. Oxytocin and cooperation under conditions of uncertainty: The modulating role of incentives and social information. Horm. Behav. 57:368–374.
  2. de Abreu MS, Kulczykowska E, Cardoso SC, André GI, Morais M, Gozdowska M, Soares M. 2018. Nonapeptide levels in male cleaner fish’ brains during interactions with unfamiliar intra and interspecific partners. Ecol. Sociobiol. 72:122.

Tags: cooperation, guppies, oxytocin

Posted on: 15th March 2021


Read preprint (No Ratings Yet)

Have your say

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Sign up to customise the site to your preferences and to receive alerts

Register here