Behavioral Science|15 min read|Last reviewed 2026-11-01|Mixed EvidencePartially Verified

Understanding Canine Aggression: Development, Context, and Prevention

Compound evidence detail4 SCRs / 8 parts

SCR-084

Documentedthe specified associations
Heuristica hormonal-mechanism interpretation

SCR-406

Documenteddevelopmental-window science and behavioral outcomes data from structured socialization environments in puppies
HeuristicJB position that conventional puppy-class environments often conflict with the soft-landing approach during the first month

SCR-484

Documentedthe association
Heuristicthe mentorship interpretation

SCR-531

DocumentedThe bite-context patterns, the fatality framework, and the surveillance position.
HeuristicThe JB-specific synthesis that the prevention pillar at the raising level addresses several of the preventable factors identified at the fatality level.

Aggression research can help families ask better questions. Which behavior occurred, toward whom, and in what situation? What does the dog's history tell us, and what does it leave uncertain? These questions are more useful than treating aggression as one fixed quality that a dog either has or does not have. Studies use different measures, from family questionnaires to working-dog tests and injury records; their results are not interchangeable. Documented

Just Behaving approaches prevention through the environment, relationships and continuity of raising. This is a practical philosophy with a scientific rationale, but the complete programme has not been shown superior to competent alternatives in a controlled comparison of adult aggression outcomes. The evidence can inform the philosophy without turning it into a guarantee. Heuristic The comparison gap supports neither JB superiority nor the superiority of formal training over competent relational raising. Evidence Gap (SCR-167)

What It Means

Start with the outcome

Aggression toward a visitor, a household member and another dog can be measured separately. Fear, avoidance, excitability and aggression are also different outcomes, even when they occur together. A study reporting less fear has not necessarily demonstrated fewer bites. C-BARQ, a structured family-report questionnaire, helps describe behavior in specified situations; it does not replace an individual assessment. Documented

Think of these measures as different instruments on a dashboard. A fuel gauge and a speedometer both tell you something about a car, but one reading cannot stand in for the other. In the same way, a puppy-test score, a family's report and a hospital injury record each answer a different question. The analogy is about measurement, not a model of how aggression develops.

For an ordinary observation, "the dog growled when a visitor reached toward its bed" is more informative than "the dog is aggressive." It identifies an action, target and situation without pretending to diagnose the cause.

Development contributes without fixing the future

Retrospective surveys associate broader early social exposure with lower later fear. Controlled research adds a useful qualification: Stolzlechner and colleagues' study of 83 puppies found selected early test advantages after graded challenges, while handled controls showed greater interest in a friendly stranger. The 67 questionnaires returned at six months showed no significant group differences in fear or aggression. This does not demonstrate lasting aggression prevention, equivalence or the irrelevance of early experience. Documented (SCR-246; SCR-538)

Specific exercises have equally specific limits. Boone and colleagues studied early neurological stimulation in 76 small-breed puppies from one commercial kennel. Assessment ended around eight weeks, and the stranger-approach measure did not show a significant treatment effect. The study cannot establish prevention of adult aggression because it did not follow that outcome into adulthood. Documented

Acquisition and history involve several influences at once. McMillan and colleagues found higher reported family-member-, stranger- and dog-directed aggression among dogs acquired from pet stores than among dogs acquired from noncommercial breeders; the study's owner-directed association differed by neuter status. The comparison did not isolate genetics, early care, transport or another single cause. Espinosa and colleagues found modest associations between recalled early adversity and later fear and aggression. Their exploratory Golden Retriever tests showed a fear difference but no significant aggression difference, without correction for multiple comparisons. These associations do not determine an individual dog's future. Documented (SCR-532; SCR-534)

Care, other dogs and timing need context

Maternal-care findings vary by population and outcome. Foyer and colleagues associated greater care with higher aggression-test scores in German Shepherd working dogs. Bray and colleagues associated greater investment with behavioral release from a guide-dog programme. Neither result means that withholding care prevents household aggression. Test reactions and occupational qualification are different from household bites, and shared genetics complicates causal interpretation. Documented

Serpell and Duffy found that other dogs in a guide puppy's raising household were associated with lower aggression toward household members. That result survived correction for multiple comparisons; the dog-directed-aggression result did not. Table 5 also reported nominal training-success associations for households with other dogs and for households with other pets but no dogs, each compared with no other pets. Those training findings were not marked as surviving the multiple-comparison correction. Resident-animal age and mentorship behavior were not measured. Documented (SCR-484) JB interprets this literature within its supervised mentorship framework; it does not demonstrate that adding a dog prevents aggression. Heuristic

The date a puppy moves home is only part of its experience. Cocco's survey of 107 dogs found higher fear/anxiety in its earliest adoption groups than in the oldest group, with attachment/attention-seeking differences but no significant aggression difference. Small, uneven groups and inconsistent printed age boundaries limit interpretation. Documented (SCR-533) The evidence does not establish that eight-week placement or twelve-week retention is superior in otherwise equivalent environments. JB's approximately twelve-week programme remains its developmental choice; no homing-policy change is proposed here. Heuristic

Classes do not have one uniform result. Kutsumi reported favorable stranger responses in selected comparisons but no significant C-BARQ aggression differences. Dinwoodie associated training by six months with lower odds of reported aggression. In Lord's referral sample, more than five classes versus none was associated with higher odds of a human-aggression case rather than another behavioral case. This was not a direct comparison with one to five classes or population aggression risk. Selection, different outcomes and emerging problems limit causal conclusions. Documented (SCR-540) JB's case-by-case first-month guidance remains a practical interpretation, not a conclusion that classes uniformly help or harm. Heuristic (SCR-406)

Why It Matters for Your Dog

Prevention is a responsibility, not a diagnosis

For JB, prevention means adults organize manageable experiences, consistent boundaries and calm relationships while a puppy develops. It is a philosophy of raising. It does not establish that every later aggressive event proves a family failed to apply the philosophy. A dog's presentation cannot be classified as caused by poor implementation simply by working backward from the outcome. Heuristic

Prevention, present-day management and treatment address different questions. The practical task for a family is to describe what is happening and seek the appropriate level of help, rather than intensify the raising approach when it is no longer sufficient. This is a boundary of the JB programme, not a clinical treatment instruction.

Golden Retrievers still vary

Golden Retrievers have breed-typical tendencies alongside meaningful individual variation. Breed ancestry was a modest predictor of behavior in Morrill and colleagues' analysis; an average cannot certify the safety of one dog. Documented (SCR-070; SCR-338)

Alex and colleagues' Golden Retriever study reported a dog-directed-aggression association near PTPN1 and SNP heritability of 50.51 percent, with a standard error of 14.88 percentage points. SNPs are the measured genetic variants used in this analysis. Heritability describes variation in that population, not the percentage of one puppy's behavior that is fixed. The result establishes neither a causal gene nor an individual aggression prediction test. A separate disputed trait estimate remains withheld in the source record. Documented (SCR-539)

Knowledge and clinical outcomes have their own limits

Two preschool studies illustrate the limits of dog-safety education. Meints reported improved recognition of overt threatening signals without an independent control group; subtle conflict-defusing signals remained poorly recognized and avoiding-signal gains were not durable. Schwebel's randomized website trial found mostly nonsignificant outcomes in the assigned groups, with selected gains only among children who used the programme as intended. That subgroup comparison no longer preserved randomization. Neither study establishes fewer real-world bites. Documented (SCR-536) Adults remain responsible for arranging and supervising interactions. Heuristic

Professional care deserves accurate expectations. In Powell's cohort of 131 dog-family pairs, modeled stranger-directed aggression, owner-directed aggression, dog-directed aggression and dog rivalry did not change significantly across six months. Attrition, missing questionnaire responses and the absence of an untreated comparison group limit interpretation. This establishes neither a general cure rate nor that care is ineffective. Documented (SCR-537) Neutering studies also do not establish surgery as a reliable general aggression-prevention intervention. Observational timing associations do not settle a particular dog's health and reproductive decisions. Documented (SCR-083; SCR-084)

When to Seek Professional Help

JB is a raising programme, not an aggression treatment service. A bite, escalating threats, sudden behavior change or a situation the family cannot safely manage belongs with veterinary assessment and appropriate behavioral support. This is a referral boundary, not a diagnosis or a triage instrument. Professional assessment should not depend on deciding whether a family implemented JB correctly.

Key Takeaways

Describe the behavior, its target and the situation before explaining its cause.
Early care and experience matter, but a fear association or puppy-test result is not proof of adult aggression prevention.
Maternal-care findings are mixed; other household dogs were not experimentally tested as JB mentors.
Golden Retriever averages and genetic findings do not guarantee an individual dog's safety.
Prevention is an adult responsibility. Existing aggression may require professional assessment rather than more intensive application of a raising philosophy.

The Evidence

Hsu and Serpell (2003); Foyer et al. (2016); Powell et al. (2021); Patronek et al. (2013)Dogs in family-report, military working-test and referral settings; humans in fatality records
SCR-247 and SCR-531: target-specific questionnaire scales, working-test reactions and injury records describe different endpoints; none substitutes for individual assessment.

Puurunen et al. (2020); Hakanen et al. (2020)Domestic dogs in retrospective surveys
SCR-246: early social exposure was associated with later fear. Overlapping populations, recall and confounding limit independent-replication and causal claims.
Stolzlechner et al. (2022)83 puppies; 67 six-month questionnaire returns
SCR-538: selected short-term challenge-group advantages, greater stranger interest in handled controls, and no significant six-month fear/aggression group differences.
Boone et al. (2022)76 puppies from one commercial kennel
SCR-535: no significant treatment difference in the approach test around eight weeks; no adult aggression follow-up.

McMillan et al. (2013)413 pet-store and 5,657 noncommercial-breeder dogs
SCR-532: reported aggression differed by acquisition source; the owner-directed association differed by neuter status. Source bundled multiple influences.
Espinosa et al. (2025)Pet-dog survey; adversity recalled before six months
SCR-534: modest associations with later aggression/fear. Exploratory Golden fear and aggression results differed, without multiplicity correction.

Foyer et al. (2016)German Shepherd military working dogs
SCR-483: greater maternal care associated with higher aggression-test scores, not measured household biting.
Bray et al. (2017)Guide-dog candidates
SCR-037: greater investment associated with behavioral release; not a household-aggression endpoint or separated genetic effect.
Serpell and Duffy (2016)Guide-dog raising households
SCR-484: lower household-member aggression survived correction; dog-directed aggression did not. Table 5 nominal training-success associations used no-other-pets as comparator, not measured adult mentorship.

Cocco et al. (2025)107 dogs in an observational survey
SCR-533: early-group fear/attachment differences and aggression null; inconsistent age labels, small uneven groups and environmental confounding preclude an optimal-age claim.
Kutsumi et al. (2013)Dogs in training/test comparisons
SCR-540: favorable stranger responses in selected comparisons but no significant C-BARQ aggression differences.
Dinwoodie et al. (2021)Dogs in a retrospective family survey
SCR-540: training by six months associated with lower odds of reported aggression; selection and confounding limit interpretation.
Lord et al. (2017)Referral-clinic case-control sample
SCR-540: more than five classes versus none associated with human-aggression rather than other behavioral cases. Household-dog direction is withheld.

Morrill et al. (2022)Multi-breed dog analysis
SCR-070: breed ancestry explains a modest share of individual behavioral variation; it is not an individual safety prediction.
Alex et al. (2025)Golden Retriever cohort
SCR-539: PTPN1-associated locus and dog-directed-aggression SNP heritability 50.51% (SE 14.88 percentage points). No causal-gene or individual-test conclusion; disputed secondary trait estimate withheld.

Powell et al. (2021); Farhoody et al. (2018); McGreevy et al. (2018); Moxon et al. (2022); Sherman and Mills (2008)Domestic dogs; referral cohort, observational surveys and female retriever-cross guide-dog cohort
SCR-537, SCR-164, SCR-083, SCR-084 and SCR-528: Powell group aggression nulls do not establish ineffectiveness. Gonadectomy effects are not uniform; model dependence, confounding and selected populations remain. The veterinary-led care framing is not a drug or surgery prescription.

Meints et al. (2018)Preschool children
SCR-536: overt-threat recognition improved; subtle conflict-defusing recognition remained poor and avoiding-signal gains were not durable. No independent control group or demonstrated bite reduction.
Schwebel et al. (2016)Preschool children in a randomized website trial
SCR-536: mostly nonsignificant intention-to-treat dog-safety outcomes; selected gains in an adherent subgroup do not preserve randomization or demonstrate bite reduction.

Just Behaving Foundations 2.1 and Pillars 2.0Application to family dogs
Supervised relationships, manageable experiences and adult responsibility are philosophical applications. SCR-406 and SCR-484 keep class-policy and mentorship interpretations separate from empirical findings. They are not outcome evidence for the complete programme.

Limitations of the accepted aggression claimsDog studies and human child-learning studies kept distinct
SCR-533, SCR-538, SCR-539 and SCR-536 do not establish an optimal homing age, lasting general aggression prevention, an individual genetic safety test or dependable real-world bite reduction.

SCR-167 comparison-gap recordJB raising and competent alternative approaches
The cited literature does not establish comparative adult-aggression outcomes for the complete JB programme. This gap supports neither automatic superiority nor automatic rejection of JB.

SCR References

Scientific Claims Register

SCR-037In the studied multi-breed guide-dog programme, greater observed maternal investment and ventral nursing were associated with higher odds of behavioral release. These results do not establish a general reduction in stress resilience, an adult household-aggression effect, or a causal effect of maternal care independent of shared genetics.Documented

SCR-083The reviewed canine evidence does not establish gonadectomy as a reliable general intervention to reduce aggression. Farhoody et al. found no statistically significant association with aggression toward familiar people or other dogs. Higher stranger-directed-aggression associations, concentrated in dogs gonadectomized at seven to twelve months, appeared in the restricted model but not the full model; the authors raise possible type I error. A prospective female retriever-cross guide-dog cohort had no significant between-group aggression differences at ages one and three, alongside a qualified change-score result. Results vary by target, sex, timing, population and design.Documented

SCR-084In McGreevy et al.'s observational male-dog survey, lower lifetime gonadal-hormone exposure or earlier castration was associated with higher reported scores on selected fear and aggression measures. Results varied by item and hormone-exposure measure. The design does not isolate a causal effect of surgery or establish that early gonadectomy universally increases fear.Mixed Evidence

SCR-164In 131 veterinary behavioral-medicine cases, baseline dog behavior, selected owner personality traits, and owner-dog attachment were associated with treatment outcomes after behavioral consultation.Documented

SCR-246Early handling and developmental environment are associated with later behavior in the cited canine populations and are relevant to prevention. The evidence does not establish a universal ranking of developmental variables or a specific programme that prevents adult aggression.Documented

SCR-247Standardized owner-report tools such as C-BARQ are valuable for structured description and epidemiology but should not be treated as substitutes for clinical assessment or lived observation.Documented

SCR-338Golden Retrievers have a documented breed-typical behavioral profile shaped by working history, but meaningful variation remains within the breed and breeder selection can pull toward different parts of that temperament distribution.Documented

SCR-406In the Just Behaving transition framework, standard puppy class is often a poor fit in the first month because many classes combine stimulation, peer arousal, and performance pressure in ways that destabilize the calm floor, while calmer or delayed class decisions usually fit the transition better.Mixed Evidence

SCR-483Canine maternal-care associations differ by studied population and outcome. In Foyer et al.'s German Shepherd military working-dog population, greater maternal care was associated with higher aggression-test scores, not a measured increase in household biting. In Bray et al.'s guide-dog programme, greater investment was associated with higher behavioral-release odds. Neither study establishes a universal maternal-care optimum or separates care from shared genetic effects.Documented

SCR-484In Serpell and Duffy's guide-dog cohort, other dogs in the raising household were associated with lower aggression toward household members at approximately twelve months, surviving multiple-comparison correction. The dog-directed-aggression association did not survive that correction, and dog-directed fear was nonsignificant. Separately, Table 5 associated other dogs, and other pets without dogs, with higher odds of training success versus no other pets at nominal significance. Resident-animal age and mentorship behavior were not measured; a causal JB mentorship mechanism was not tested.Mixed Evidence

SCR-532One survey found higher reported aggression among pet-store dogs than dogs from noncommercial breeders. It did not identify the cause or predict the behavior of an individual dog.Documented

SCR-533One small study linked earlier adoption with higher fear ratings but found no statistically significant aggression difference. It does not establish an ideal homing age.Documented

SCR-534Dogs with reported early adversity had higher average fear and aggression ratings in one large survey. That association does not determine an individual dog's behavior or establish what caused it.Documented

SCR-535One ENS study found no detectable difference in its puppy approach test. It did not test whether ENS prevents adult aggression.Documented

SCR-536Dog-safety studies report some learning gains alongside important gaps and null findings. Better knowledge scores do not establish that bites have been prevented.Documented

SCR-070Breed ancestry explains only a modest fraction (~9%) of individual behavioral variation, even though within-breed temperament heritability remains meaningful for selection.Documented

SCR-167No published randomized controlled trial compares dogs raised in structured relational environments without formal training vs. dogs receiving formal obedience training vs. no-intervention controls. The "definitive comparative study" does not exist.Evidence Gap

SCR-528Behavior-modifying pharmacotherapy in canine clinical practice is positioned as adjunctive to environmental management and behavioral modification, applied under veterinary oversight rather than as a standalone treatment modality. The field consensus across the veterinary behavior medicine literature and across documented diplomate practice positions environment and modification as primary intervention modalities, with medication used to enable, support, or accelerate the behavioral work in cases where baseline arousal or anxiety pathology interferes with the dog's capacity to engage with modification.Documented

SCR-531Population-level US dog bite data and the Patronek et al. (2013) fatality analysis document several stable patterns. In clinical case series of medically attended child bites, familiar in-home and known-dog contexts are predominant; children are disproportionately represented at the medically attended and severe end of the bite spectrum across surveillance and clinical-case data; fatal bites are rare relative to non-fatal bites; and fatal cases are characterized by the co-occurrence of multiple preventable husbandry and management factors rather than by breed alone. Patronek et al. (2013) identified seven such co-occurrent factors across 256 US fatalities from 2000 through 2009. The CDC and successor surveillance reviews have not maintained breed-specific bite statistics as a valid surveillance metric, citing methodological limitations of visual breed identification at the population level, an issue traced to the original Sacks et al. (2000) review and to subsequent breed-ID reliability research.Mixed Evidence

SCR-537Aggression scores did not change significantly at group level in one clinical cohort. This does not establish either a cure rate or that professional care is ineffective.Documented

SCR-538An early-challenge study found some short-term puppy-test benefits, but no detectable six-month differences in the measured fear and aggression questionnaire scores.Documented

SCR-539A Golden Retriever study reported a genetic association with dog-directed aggression. It does not provide an individual aggression prediction test.Documented

SCR-540Class-attendance studies report favorable, null and contrary associations in different populations. They do not establish that classes uniformly prevent or cause aggression.Documented

Sources

Kutsumi A, Nagasawa M, Ohta M, Ohtani N (2013). Importance of puppy training for future behavior of the dog. The Journal of veterinary medical science. DOI 10.1292/jvms.12-0008. PubMed 23018794.

McMillan FD, Serpell JA, Duffy DL, Masaoud E, Dohoo IR (2013). Differences in behavioral characteristics between dogs obtained as puppies from pet stores and those obtained from noncommercial breeders. DOI 10.2460/javma.242.10.1359. PubMed 23634679.

Foyer P, Wilsson E, Jensen P (2016). Levels of maternal care in dogs affect adult offspring temperament. Scientific reports. DOI 10.1038/srep19253. PMCID: PMC4725833. PubMed 26758076.

Serpell JA, Duffy DL (2016). Aspects of Juvenile and Adolescent Environment Predict Aggression and Fear in 12-Month-Old Guide Dogs. Frontiers in veterinary science. DOI 10.3389/fvets.2016.00049. PMCID: PMC4916180. PubMed 27446937.

Schwebel DC, Li P, McClure LA, Severson J (2016). Evaluating a Website to Teach Children Safety with Dogs: A Randomized Controlled Trial. International journal of environmental research and public health. DOI 10.3390/ijerph13121198. PMCID: PMC5201339. PubMed 27918466.

Lord M, Casey RA, Loftus BA, Blackwell EJ (2017). Risk factors for human-directed canine aggression in a referral level clinical population. Vet Rec. DOI 10.1136/vr.103638. PubMed 28576767.

Bray EE, Sammel MD, Cheney DL, Serpell JA, Seyfarth RM (2017). Effects of maternal investment, temperament, and cognition on guide dog success. Proceedings of the National Academy of Sciences of the United States of America. DOI 10.1073/pnas.1704303114. PMCID: PMC5576795. PubMed 28784785.

Farhoody P, Mallawaarachchi I, Tarwater PM, Serpell JA, Duffy DL, Zink C (2018). Aggression toward Familiar People, Strangers, and Conspecifics in Gonadectomized and Intact Dogs. Frontiers in veterinary science. DOI 10.3389/fvets.2018.00018. PMCID: PMC5834763. PubMed 29536014.

McGreevy PD, Wilson B, Starling MJ, Serpell JA (2018). Behavioural risks in male dogs with minimal lifetime exposure to gonadal hormones may complicate population-control benefits of desexing. PloS one. DOI 10.1371/journal.pone.0196284. PMCID: PMC5931473. PubMed 29718954.

Meints K, Brelsford V, De Keuster T (2018). Teaching Children and Parents to Understand Dog Signaling. Frontiers in veterinary science. DOI 10.3389/fvets.2018.00257. PMCID: PMC6256863. PubMed 30525045.

Puurunen J, Hakanen E, Salonen MK, Mikkola S, Sulkama S, Araujo C, Lohi H (2020). Inadequate socialisation, inactivity, and urban living environment are associated with social fearfulness in pet dogs. Scientific reports. DOI 10.1038/s41598-020-60546-w. PMCID: PMC7044223. PubMed 32103117.

Hakanen E, Mikkola S, Salonen M, Puurunen J, Sulkama S, Araujo C, Lohi H (2020). Active and social life is associated with lower non-social fearfulness in pet dogs. Scientific reports. DOI 10.1038/s41598-020-70722-7. PMCID: PMC7426946. PubMed 32792641.

Powell L, Stefanovski D, Siracusa C, Serpell JA (2021). Owner Personality, Owner-Dog Attachment, and Canine Demographics Influence Treatment Outcomes in Canine Behavioral Medicine Cases. Frontiers in Veterinary Science 7:630931. DOI 10.3389/fvets.2020.630931. PubMed 33553291; PMCID:PMC7862121.

Dinwoodie IR, Zottola V, Dodman NH (2021). An Investigation into the Impact of Pre-Adolescent Training on Canine Behavior. Animals (Basel). DOI 10.3390/ani11051298. PMCID: PMC8147152. PubMed 33946497.

Moxon R, Freeman S, Payne R, Corr S, England GCW (2022). A Prospective Cohort Study Investigating the Behavioural Development of Bitches in a Guide Dog Training Programme Neutered Prepubertally or Post-Pubertally. Frontiers in veterinary science. DOI 10.3389/fvets.2022.902775. PMCID: PMC9301489. PubMed 35873697.

Stolzlechner L, Bonorand A, Riemer S (2022). Optimising Puppy Socialisation-Short- and Long-Term Effects of a Training Programme during the Early Socialisation Period. Animals : an open access journal from MDPI. DOI 10.3390/ani12223067. PMCID: PMC9687081. PubMed 36428295.

Boone G, Romaniuk AC, Barnard S, Shreyer T, Croney C (2022). The Effect of Early Neurological Stimulation on Puppy Welfare in Commercial Breeding Kennels. Animals : an open access journal from MDPI. DOI 10.3390/ani13010071. PMCID: PMC9818019. PubMed 36611681.

Cocco R, Arfuso F, Sechi S, Piccione G, Giannetto C, Arrigo F, Rizzo M (2025). The Puppies' Age at Adoption Time Influences the Behavioral Responses of Adult Dog. Veterinary sciences. DOI 10.3390/vetsci12020176. PMCID: PMC11860672. PubMed 40005936.

Espinosa J, Zapata I, Alvarez CE, Serpell JA, Kukekova AV, Hecht EE (2025). Influence of early life adversity and breed on aggression and fear in dogs. Scientific reports. DOI 10.1038/s41598-025-18226-0. PMCID: PMC12491534. PubMed 41039027.

Alex E, Gennotte P, Morros Nuevo A, Yu Y, Keep B, Sullivan M, Mills D, Warrier V, Raffan E (2025). GWAS for behavioral traits in golden retrievers identifies genes implicated in human temperament, mental health, and cognition. Proceedings of the National Academy of Sciences of the United States of America. DOI 10.1073/pnas.2421757122. PMCID: PMC12684936. PubMed 41284867.

Hsu, Y., and Serpell, J. A. (2003). Development and validation of a questionnaire for measuring behavior and temperament traits in pet dogs. Journal of the American Veterinary Medical Association, 223(9), 1293-1300.

Morrill, K., et al. (2022). Ancestry-inclusive dog genomics challenges popular breed stereotypes. Science, 376:eabk0639. DOI.

Sherman, B. L., and Mills, D. S. (2008). Canine anxieties and phobias: an update on separation anxiety and noise aversions. Veterinary Clinics of North America: Small Animal Practice, 38(5), 1081-1106. DOI. Used only for the accepted SCR-528 veterinary-led framing, not as a general aggression-efficacy trial.

Patronek, G. J., Sacks, J. J., Delise, K. M., Cleary, D. V., and Marder, A. R. (2013). Co-occurrence of potentially preventable factors in 256 dog bite-related fatalities in the United States. Journal of the American Veterinary Medical Association, 243(12), 1726-1736. DOI.

The scope of JB's interpretation is described in the Prevention pillar, Scientific Claims Register explanation, and verification guide. Internal source documents listed in the metadata are provenance records, not promises of publicly downloadable OneDrive files.