Berg, van den Berg et al. (2007) — Behavior Genetics
Evaluation of the Serotonergic Genes htr1A, htr1B, htr2A, and slc6A4 in Aggressive Behavior of Golden Retriever Dogs
Published: August 04, 2026
Aggression in Golden Retrievers is heritable — the genetics behind it are real and measurable. The serotonin system has long been implicated in canine aggression through physiological and pharmacological evidence. Putting those two facts together, the four most obvious candidate genes in the serotonergic pathway — coding for serotonin receptors 1A, 1B, and 2A, and the serotonin transporter — seem like the natural place to look for the genetic basis of aggressive behaviour in this breed. A rigorous multi-method genetic study looked. They found nothing. And that null result is more important than it might first appear. 🐾
Researchers L. Berg, L.V.D. Berg, Peter Leegwater, and their team conducted mutation screens, linkage analysis, association studies, and quantitative genetic analysis of the four candidate serotonergic genes in a study group of Golden Retrievers with documented aggression history and non-aggressive controls. The analyses were thorough enough — covering 41 SNPs across the 1 Mb regions flanking each candidate gene in 49 aggressive and 49 non-aggressive dogs — to have detected a major effect at these loci if one existed.
What the Genetic Analysis Found and Did Not Find
The coding DNA sequences of the four candidate genes showed no systematic difference between aggressive and non-aggressive Golden Retrievers. Linkage analysis found no strong major locus effect on human-directed aggression in the regions surrounding the candidate genes. The SNP association analysis found no association of alleles, genotypes, or haplotypes with aggression at any of the four candidate loci. And the quantitative genetic analysis — which estimated the effect of candidate gene haplotypes on a range of aggression-related phenotypic measures using a linear animal model — found no effect across multiple types of aggressive behaviour.
The researchers are precise in their conclusion: it is unlikely that these four genes play a major role in the variation in aggression in the Golden Retrievers studied. They acknowledge that smaller phenotypic effects cannot be ruled out given sample size constraints — the study was powered to detect major gene effects rather than minor contributions. But the absence of any signal across multiple independent analytical approaches in a well-characterised breed group is strong evidence against a major role for these specific genes.
This finding sits alongside the earlier confirmation in the same study group that aggression in Golden Retrievers is highly heritable. These two facts together are not contradictory. They are a precise scientific statement: the genetic architecture of aggression in this breed is real, but it is not driven by the four serotonergic candidate genes that represented the most biologically plausible starting point for the search. The genetics are there. They are just elsewhere in the genome.
Why Null Results Matter in Behavioural Genetics ⚠️
The serotonin hypothesis of aggression has substantial supporting evidence from pharmacological and physiological research — altered serotonin metabolism has been documented in aggressive dogs, serotonin-modulating drugs are used as adjunct treatments for aggression in veterinary behavioural medicine, and the dietary protein-serotonin-aggression pathway covered in earlier research in this series provides a mechanistic bridge between nutrition and behavioural expression. None of that evidence is invalidated by this genetic null result.
What the null result establishes is that the relationship between serotonin and aggression does not operate through major coding or regulatory variants in these four specific receptor and transporter genes — at least not in this breed population. Serotonin is involved in aggression modulation. The specific genetic path through which it is involved is more complex than the candidate gene approach here could capture.
This is a significant message for the broader field. Behavioural genetics in dogs has sometimes been characterised by the assumption that if a neurotransmitter system is implicated in a behaviour, the genes coding for that system’s receptors and transporters must be the genetic source of heritable variation in that behaviour. This study demonstrates that assumption is not reliable. The genetic architecture of complex behavioural traits rarely reduces to obvious single-pathway candidate genes — and the search for the genetic basis of canine aggression must extend beyond the most intuitive starting points into more comprehensive genome-wide approaches.
What This Means for How We Think About Aggressive Golden Retrievers 🐕
For owners and trainers working with aggressive Golden Retrievers, this research reinforces a message that should run through every piece of aggression science covered in this series: aggressive behaviour in dogs is multifactorial, and the factors driving it in any individual dog are unlikely to be fully explained by any single system — neurochemical, genetic, developmental, or relational.
A Golden Retriever showing human-directed aggression carries heritable genetic risk that is real but not yet mapped to specific variants. Their serotonin system may be functioning differently than in non-aggressive dogs, but the genetic source of that difference is not in the receptor genes studied here. Their early developmental history, their physiological stress coping capacity, the quality of their owner relationship, and the dietary protein environment they are living in all contribute to whether and how that heritable risk is expressed.
At Zoeta Dogsoul, this complexity is exactly why NeuroBond does not reduce the relationship between dog and owner to any single variable. Aggression is not a serotonin problem. It is not a breed problem. It is not a training problem. It is a whole-system expression of a dog operating within their particular genetic architecture, developmental history, physiological state, and relational environment — and addressing it requires looking at all of those dimensions rather than seeking a single explanatory gene or a single corrective intervention. 🐾
Source: Berg, L., van den Berg, L. V. D., et al. (2007). Evaluation of the Serotonergic Genes htr1A, htr1B, htr2A, and slc6A4 in Aggressive Behavior of Golden Retriever Dogs. Behavior Genetics. Published December 7, 2007.
