The whole record: an evidence review of dolphin behavior

This is the long version of the record that the main dolphin page summarizes. It exists because a mascot invites people to imagine an animal, and imagination is generous. Research is not. We chose the dolphin as a symbol of cooperation, and that choice obliges us to publish what dolphins also do.

This review is serious and expandable, not exhaustive. It covers the major documented categories of behavior that a friendly mascot leaves out. It does not try to catalog everything every dolphin has ever done, and it does not fill gaps with dramatic stories. Where the evidence runs out, we say so, and the open questions go to the research queue at the end rather than into the text as facts.

Setting and species

There are many dolphin species. This review concerns bottlenose dolphins, and even that word covers more than one animal:

Common bottlenose dolphin (Tursiops truncatus): the species in the captive cooperation studies, the stranding studies from Scotland, Virginia, the German Baltic and the Canary Islands, and the Sarasota Bay study of human interaction pmc.ncbi.nlm.nih.gov pmc.ncbi.nlm.nih.gov pmc.ncbi.nlm.nih.gov digitalcommons.odu.edu link.springer.com pmc.ncbi.nlm.nih.gov pmc.ncbi.nlm.nih.gov.
Indo-Pacific bottlenose dolphin (Tursiops aduncus): the species in the Shark Bay, Western Australia, studies of male alliances and herding doi.org pmc.ncbi.nlm.nih.gov. An earlier Shark Bay paper identifies the animals only to genus, as Tursiops sp. pmc.ncbi.nlm.nih.gov.

When a finding comes from one population, we say which. A behavior documented in Shark Bay is a fact about Shark Bay dolphins until someone documents it elsewhere.

How to read this review

Each finding is labeled with the kind of evidence behind it, because these are not equally strong:

Direct observation: researchers watched the behavior happen.
Forensic inference: researchers examined a dead animal and inferred the cause from injuries, for example the spacing of tooth marks or a pattern of fractures. The event itself was usually not seen.
Experimental result: researchers set up a task and measured what animals did.
Proposed explanation: a hypothesis about why a behavior happens. Explanations can be reasonable and still unproven.

Each finding also names the species, the population or setting, the sample size where the source gives one, and the main uncertainty. Sources marked "abstract only" support only what their abstracts state.

On language. We use behavioral terms and the researchers' own terms. When a source uses a term like "sexual coercion" or "coercive mate guarding," we quote it and explain what the researchers measured. We do not assign human criminal motives to animals, and we do not describe injuries in more detail than the science requires.

Aggression within dolphin populations

NOAA Fisheries lists the social behavior of common bottlenose dolphins as including "breeding, playing, aggression, and gentle body contact" fisheries.noaa.gov. Aggression is part of ordinary dolphin social life, not an aberration.

The clearest documentation of aggression among dolphins of the same population comes from Shark Bay, where male alliances compete over females. In a 25 month study from 1987 to 1989, Richard Connor, Rachel Smolker and Andrew Richards documented herding in 10 alliances and reported that "occasionally, two alliances combined and took females from another alliance or defended females against such efforts" pmc.ncbi.nlm.nih.gov. A 2022 analysis by Connor and colleagues explains that many consortships last a day or less, "particularly if males are unable to defend the female from attacks from rivals" doi.org.

Allied males in Shark Bay also use a sound researchers call a "pop," which Stephanie King and colleagues describe as a threat vocalization, produced in trains of 3 to 30 pops at rates of 6 to 12 per second pmc.ncbi.nlm.nih.gov. In their recordings from 2016 and 2017, which included 1,221 pop trains, pop rates rose when the male nearest the female changed, which the authors interpret as males inducing the female to stay close during a guard switch pmc.ncbi.nlm.nih.gov.

Kind of evidence: direct observation; the interpretation of pops as threats is the researchers' inference from context and female responses.
Species and population: Indo-Pacific bottlenose dolphins, Shark Bay.
Uncertainty: these studies document aggression tied to competition over females in one population.

Male alliances and coercive mating

This is the finding that most directly tests the dolphin as a symbol of cooperation, because the most elaborate cooperation known in dolphins is organized around controlling access to females.

What was observed. In Shark Bay, males form alliances at three levels doi.org. Pairs and trios ("first-order alliances") herd individual females. Almost every adult male also belongs to a larger "second-order alliance" of 4 to 14 males that competes with other alliances over females. And second-order alliances cooperate with one another in "third-order alliances" against other groups. Connor and colleagues analyzed association and consortship data collected from 2001 to 2006 on 121 adult males and describe "the largest nonhuman alliance network known" doi.org.

What herding is. The 2022 paper describes first-order alliances as forming "aggressively maintained consortships with individual females that last from hours to weeks" doi.org. King and colleagues define herding as "an aggressively maintained association, where two to three males use vocal and physical threats to coerce a female to accompany them" pmc.ncbi.nlm.nih.gov. They classify it as coercive mate guarding, "a form of sexual coercion which functions to constrain female mate choice" pmc.ncbi.nlm.nih.gov. The earlier Shark Bay study found that males "preferentially herded nonpregnant females likely to be in estrus" pmc.ncbi.nlm.nih.gov.

What cooperation buys. Males who were better connected to allies within their second-order alliance spent more time consorting females, and males with stronger ties to allies outside it kept females for longer doi.org. In plain terms: the more cooperative a male's network, the longer he and his partners could keep a female with them.

What the researchers are careful not to claim. "Coercion" here refers to controlling a female's movements and associations. Whether matings themselves are forced is a separate question. King and colleagues note that the three dimensional ocean habitat is thought to impede coerced matings pmc.ncbi.nlm.nih.gov, and Connor and colleagues write that "it is currently impossible to determine how matings are shared during consortships" doi.org. We report the coercion of association that was observed and leave the rest where the researchers leave it.

Kind of evidence: long term direct observation and association analysis; the description of herding as sexual coercion is the researchers' classification of what they observed.
Species and population: Indo-Pacific bottlenose dolphins, Shark Bay, Western Australia.
Sample: 121 adult males in the network analysis doi.org; 10 herding alliances in the 1987 to 1989 study pmc.ncbi.nlm.nih.gov; focal follows in 2016 and 2017 pmc.ncbi.nlm.nih.gov.
Uncertainty: the alliance system is described by the researchers as unusual; it should not be assumed to exist in other bottlenose populations without evidence.

The point for a symbol of cooperation is direct: cooperation is a means, not a moral outcome. Cooperation among some individuals can be what makes coercion of others effective.

Infanticide

Northeast Scotland. Ian Patterson and colleagues reported in 1998 that post mortem examinations of stranded bottlenose dolphins showed that "five out of eight young calves from this same area were also killed by bottlenose dolphins," and they reported "direct observations of an aggressive interaction between an adult bottlenose dolphin and a dead bottlenose dolphin calf" pmc.ncbi.nlm.nih.gov. The authors wrote that their findings "appear to provide the first evidence of infanticide in cetaceans" pmc.ncbi.nlm.nih.gov.

Virginia. David Dunn and colleagues examined nine common bottlenose dolphin calves that stranded in Virginia in 1996 and 1997 digitalcommons.odu.edu. All had died of severe blunt force trauma concentrated on the head and chest, with multiple rib fractures and lung damage, and the injuries occurred on both sides of the body in every calf. One calf had a bite wound with punctures "consistent with the tooth placement in an adult bottlenose dolphin" digitalcommons.odu.edu. The pathologists found the injuries "not compatible with predation, boat strike, fisheries interactions, rough-surf injury, or blast injury," and concluded that "the evidence suggests" infanticide occurs in bottlenose dolphins of the western North Atlantic digitalcommons.odu.edu.

Kind of evidence: forensic inference in both studies, plus one direct observation of an adult with an already dead calf in Scotland. The killings themselves were not observed.
Species and population: common bottlenose dolphins, northeast coast of Scotland and coastal Virginia.
Sample: 8 calves examined in Scotland, 5 attributed to dolphins pmc.ncbi.nlm.nih.gov; 9 calves in Virginia digitalcommons.odu.edu.
Proposed explanations: the Scottish authors propose that infanticide "must be considered as a factor shaping sociality" in cetaceans and connect it to attacks on porpoises (next section) pmc.ncbi.nlm.nih.gov. Our access to both papers is limited to their abstracts, so we do not summarize further explanations here.
Uncertainty: these are findings for two populations. They do not show that infanticide is common, or that it occurs in every population. A later veterinary review from the Canary Islands describes the species as "well-known for infanticide" pmc.ncbi.nlm.nih.gov, but that is a summary of the same earlier literature, not new evidence of frequency.

Lethal attacks on harbor porpoises and other small cetaceans

Scotland. The 1998 Scottish study reports that "most harbour porpoises found dead on the north-east coast of Scotland show signs of attack by sympatric bottlenose dolphins," and that the size range of attacked porpoises and attacked dolphin calves is similar pmc.ncbi.nlm.nih.gov. That size match is the basis of their proposal that the two behaviors may be linked. An earlier 1996 report by Harry Ross and Ben Wilson on porpoises in the Moray Firth is cited throughout this literature; we have not yet opened it and do not quote its figures.

German Baltic Sea. In autumn 2016, a single photo identified male bottlenose dolphin stayed in the German Baltic waters of Schleswig-Holstein for three months link.springer.com. Harbor porpoises are the only native cetacean there. Necropsies during that period found trauma in six stranded porpoises, including internal bleeding, multiple mostly bilateral rib fractures and one lung laceration, which the authors say "correspond with the previously reported dolphin-caused injuries in other regions" link.springer.com. Public sighting reports matched the dolphin's presence in time and place to the strandings. The authors are explicit that "no attack has been witnessed in German waters to date," and describe their findings as an "indication" of lethal interactions link.springer.com.

Canary Islands. Raquel Puig-Lozano and colleagues reviewed 540 cetaceans that stranded and were necropsied in the Canary Islands from 2000 to 2017 pmc.ncbi.nlm.nih.gov. Twenty-four animals of eight species had traumatic injuries caused by other cetaceans. In ten with tooth rake marks, the spacing between marks allowed the researchers to suggest the attacking species: four were compatible with killer whales and four with common bottlenose dolphins, affecting three short-beaked common dolphin calves and one adult Atlantic spotted dolphin pmc.ncbi.nlm.nih.gov.

Kind of evidence: forensic inference from injuries and tooth mark spacing, with sighting correlation in the Baltic case. No attack was observed in the Baltic or Canary studies.
Species and population: common bottlenose dolphins as the inferred attackers; harbor porpoises in Scotland and the Baltic; other small dolphins in the Canaries.
Proposed explanations: a link to infanticidal behavior, based on size similarity pmc.ncbi.nlm.nih.gov; practice for fighting, which the Canary authors cite from earlier work pmc.ncbi.nlm.nih.gov. Neither is established. The Baltic case adds that a socially isolated dolphin can be involved link.springer.com.
Uncertainty: forensic attribution depends on matching injuries to known patterns. These studies show that bottlenose dolphins can and do kill porpoises in the documented places; they do not establish how often or why.

Risks in human interaction

Bites and aggression. NOAA Fisheries states that "dozens of bites have been reported, and people have been pulled underwater," and that dolphins that beg can become "pushy, aggressive, and threatening" when food does not come fisheries.noaa.gov. Its guidance on feeding says marine mammals "are known to bite or injure people when teased or expecting food" fisheries.noaa.gov. NOAA also calls the friendly wild dolphin a myth, noting that the television dolphin Flipper was a trained captive animal fisheries.noaa.gov.

Feeding harms dolphins. In Sarasota Bay, Florida, on the Gulf coast, Fredrik Christiansen and colleagues used records spanning more than 45 years, more than 32,000 group sightings and more than 1,100 individual dolphins pmc.ncbi.nlm.nih.gov. By 2014, 190 dolphins, 16.6 percent of all identified individuals, had become conditioned to human interactions such as begging, scavenging and taking bait or catch from anglers. Dolphins became conditioned partly by associating with dolphins that already were, and "conditioned dolphins were more likely to be injured by human interactions" than unconditioned ones pmc.ncbi.nlm.nih.gov. The authors also report, from a team member's field observation rather than their statistical analysis, that roughly 35 percent of dolphins seen with human related injuries had injuries serious enough to have likely contributed to death or to require rescue pmc.ncbi.nlm.nih.gov. NOAA adds that serious and sometimes fatal injuries from rod and reel fishing gear are increasing fisheries.noaa.gov, and that young dolphins may not survive when their mothers compete with them for handouts fisheries.noaa.gov.

Solitary sociable dolphins. A review by Laetitia Nunny and Mark Simmonds counted 32 solitary dolphins recorded since 2008, 27 of them bottlenose pmc.ncbi.nlm.nih.gov. Some of these animals, isolated from other dolphins, come to seek out people. The review describes a progression in which, as an animal becomes a tourist attraction, it "may start to exhibit dominant, aggressive and sexual behaviors," and gives the example of a dolphin in Brittany pushing divers to the bottom and preventing kayaks from maneuvering pmc.ncbi.nlm.nih.gov. The most serious case it reports is from Brazil in 1994: a bottlenose dolphin known as Tião "injured 29 people and caused the death of a 30-year-old man after butting him" pmc.ncbi.nlm.nih.gov. The review stresses context: people had grabbed his fins, hit him and jumped on him, and a management plan introduced afterward was followed by no further recorded incidents pmc.ncbi.nlm.nih.gov. The same review documents that several solitary dolphins have been deliberately injured or killed by people pmc.ncbi.nlm.nih.gov. Its account of the Brazil case relies on a 1997 report by Marcos César de Oliveira Santos, which we have not yet opened.

Kind of evidence: agency guidance summarizing reports (NOAA); long term observation with statistical analysis (Sarasota); a review of published and unpublished case reports (solitary dolphins).
Species and setting: common bottlenose dolphins in Sarasota Bay; bottlenose and a few other species worldwide in the solitary dolphin review.
Uncertainty: NOAA does not give a total count of bites with a denominator, so we do not estimate a rate. Case reports show what can happen, not how often.

The practical conclusion is the same as NOAA's: do not feed, touch, swim with or harass wild dolphins, and stay at least 50 yards away fisheries.noaa.gov fisheries.noaa.gov fisheries.noaa.gov. The full guidance is on the main dolphin page.

Predation and diet

Bottlenose dolphins are predators. NOAA Fisheries lists their prey as "fish, squid, and crustaceans (e.g., crabs and shrimp)" fisheries.noaa.gov. They locate prey by passive listening and echolocation, and they swallow fish whole, head first fisheries.noaa.gov. Some of their most striking cooperation happens here: groups "work together to herd fish into groups and then take turns charging through the school," and dolphins may trap schools against sand bars and seawalls fisheries.noaa.gov.

Eating other animals is not a moral failing and we do not treat it as one. It belongs in this review for a different reason: the same coordination people admire in a dolphin pod is, in this case, coordination against prey. Cooperation describes how animals act together, not whether the result is kind.

Kind of evidence: agency summary of established natural history.
Uncertainty: diet varies by place; the NOAA summary is general, not population specific.

Popular myths, audited

Each claim below gets one of three verdicts: supported, not supported, or unresolved. Unresolved means we could not find evidence strong enough to settle it, not that the claim is false.

"Dolphins are always smiling." Unresolved as an anatomical statement. We have not yet located an authoritative anatomical source to cite, so we publish no claim about how the mouth is built. What the record does establish is that a dolphin's appearance is not evidence of its state: the same species that many people see as smiling is documented biting people, herding females and killing calves pmc.ncbi.nlm.nih.gov digitalcommons.odu.edu fisheries.noaa.gov. A face that people read as cheerful is a human interpretation.

"Wild dolphins are friendly to people." Not supported as a general statement. NOAA calls the friendly wild dolphin a myth and documents bites and people pulled underwater fisheries.noaa.gov. Solitary sociable dolphins seek out human company in some cases, and some of those cases have ended with injuries to people pmc.ncbi.nlm.nih.gov.

"Dolphins rescue drowning swimmers." Unresolved. Stories of this kind circulate widely, but we found no systematic study that tested them, and anecdotes cannot separate rescue from play, curiosity or chance.

"Dolphins chew pufferfish to get high." Unresolved. We found no peer-reviewed study that measured intoxication in dolphins handling pufferfish. Footage of dolphins handling fish does not establish what the dolphins experienced or why they did it.

"Dolphins are altruistic." Unresolved. Some behavior looks like care. A systematic review of 106 cases of aquatic mammals reacting to a death, 81 of them cetaceans, found that cetaceans display carrying behavior with the dead, such as hauling, mouthing and diving with the carcass, and lifting or sinking it, generally with other members of their species present pmc.ncbi.nlm.nih.gov. Those are observations. Whether they reflect altruism, grief, or something else is a proposed explanation that the observations alone cannot settle. The cooperation experiments show coordination for a shared reward, which is not the same as helping at a cost to oneself pmc.ncbi.nlm.nih.gov pmc.ncbi.nlm.nih.gov.

"Dolphins never sleep" or "sleep with half the brain." The second is supported in a precise form. A 2008 review reports that cetaceans examined to date show unihemispheric slow wave sleep, in which one half of the brain shows sleep patterns while the other does not, along with a negligible amount or complete absence of rapid eye movement sleep and an asymmetrical eye state pmc.ncbi.nlm.nih.gov. So dolphins do sleep, in an unusual way. The review's authors propose that the main pressure behind this form of sleep was the need to offset heat loss in water; that is a proposed explanation pmc.ncbi.nlm.nih.gov.

"Dolphins are the second most intelligent animal." Unresolved, and not a claim we will publish. NOAA describes dolphins as highly intelligent fisheries.noaa.gov, and experiments show specific abilities such as learning a partner's role in a task pmc.ncbi.nlm.nih.gov. We found no single accepted measure that ranks intelligence across species, and a ranking would add nothing to the evidence above.

Captivity and stress

The two cooperation experiments on this site were conducted with dolphins in human care pmc.ncbi.nlm.nih.gov pmc.ncbi.nlm.nih.gov. That tells us what those animals could learn in those conditions. It does not tell us how wild dolphins behave, and we do not use captive observations as descriptions of wild animals.

The reverse is also true: conditions of captivity can shape behavior and stress, and the effects of conditions are not the same thing as inherent traits of the species. We have not yet reviewed specific studies of stress in captive dolphins with the rigor this page requires, so we make no claim about it here.

Research queue

These are open questions. None of them is stated as fact on this site until it is verified at the source.

What does the full text of Patterson and colleagues (1998) report about the injuries, the attribution methods, and the explanations the authors considered?
What does the full text of Dunn and colleagues (2002) report about calf sizes, bite measurements and limitations?
What did Ross and Wilson (1996) report about porpoise strandings and observed dolphin attacks in the Moray Firth?
Have peer-reviewed studies since 2002 documented bottlenose infanticide in other populations, and with what evidence?
What do tooth rake scar studies, such as Scott and colleagues (2005), show about aggression by age and sex?
What does Connor and Vollmer's chapter on sexual coercion in dolphin consortships establish beyond the papers reviewed here?
What does the 1997 primary report on the Brazil solitary dolphin case state, and how were the injury counts recorded?
Is there an authoritative anatomical source on the fixed shape of the bottlenose mouth?
Has any peer-reviewed study tested accounts of dolphins assisting swimmers?
Has any peer-reviewed study measured effects on dolphins of handling pufferfish?
What do specific studies show about stress in captive dolphins, and which effects are attributable to conditions rather than the species?

Evidence

Topic	Species and population	Kind of evidence	Sample	Main uncertainty
Aggression between alliances	Indo-Pacific bottlenose, Shark Bay	Direct observation	10 alliances; 121 males pmc.ncbi.nlm.nih.gov doi.org	One population
Coercive mate guarding	Indo-Pacific bottlenose, Shark Bay	Direct observation, researchers' classification	121 males; focal follows doi.org pmc.ncbi.nlm.nih.gov	Whether matings are forced is not established
Infanticide	Common bottlenose, Scotland; Virginia	Forensic inference, one observation	5 of 8 calves; 9 calves pmc.ncbi.nlm.nih.gov digitalcommons.odu.edu	Frequency and cause unknown
Porpoise and small dolphin deaths	Common bottlenose, Scotland, Baltic, Canaries	Forensic inference	6 porpoises; 4 of 24 Canary cases pmc.ncbi.nlm.nih.gov link.springer.com pmc.ncbi.nlm.nih.gov	Motive unknown
Injuries to fed dolphins	Common bottlenose, Sarasota Bay	Long term observation, statistics	More than 1,100 dolphins pmc.ncbi.nlm.nih.gov	One population
Bites and aggression toward people	Bottlenose, various	Agency summary, case reports	No denominator fisheries.noaa.gov pmc.ncbi.nlm.nih.gov	No rate can be estimated
Unihemispheric sleep	Cetaceans examined to date	Review of studies	Not stated in abstract pmc.ncbi.nlm.nih.gov	Function proposed, not proven

What the evidence does not show

It does not show that dolphins are bad animals, any more than the cooperation studies show they are good ones. Moral categories are human, and this review does not apply them.
It does not show that every bottlenose population coerces, commits infanticide, or attacks porpoises. Each finding is tied to a named place.
It does not show motive. Infanticide and porpoise attacks are mostly inferred from bodies, and the explanations offered are proposals.
It does not give rates for bites or injuries to people, because the sources do not provide denominators.
It does not settle the popular myths marked unresolved. We will update this page when evidence does.
It does not say anything about people, parties or voters. A mascot's behavior is not inherited by anyone who uses it.

Sources

Jaakkola K, Guarino E, Donegan K, King SL, "Bottlenose dolphins can understand their partner's role in a cooperative task," Proceedings of the Royal Society B 285(1887): 20180948, 2018. pmc.ncbi.nlm.nih.gov (locator: Methods; Results; Table 3; Discussion). Accessed 2026-10-28. (full text reviewed)
Yamamoto C, Kashiwagi N, Otsuka M, Sakai M, Tomonaga M, "Cooperation in bottlenose dolphins: bidirectional coordination in a rope-pulling task," PeerJ 7: e7826, 2019. pmc.ncbi.nlm.nih.gov (locator: Methods; Results; Conclusions). Accessed 2026-10-28. (full text reviewed)
Connor RC, Smolker RA, Richards AF, "Two levels of alliance formation among male bottlenose dolphins (Tursiops sp.)," Proceedings of the National Academy of Sciences 89(3): 987 to 990, 1992. pmc.ncbi.nlm.nih.gov (locator: Abstract). Accessed 2026-10-28. (abstract only)
Connor RC, Krützen M, Allen SJ, Sherwin WB, King SL, "Strategic intergroup alliances increase access to a contested resource in male bottlenose dolphins," Proceedings of the National Academy of Sciences 119(36): e2121723119, 2022. doi.org (full text read at pmc.ncbi.nlm.nih.gov; locator: Abstract; Introduction; Results; Materials and Methods). Accessed 2026-10-28. (full text reviewed)
King SL, Allen SJ, Krützen M, Connor RC, "Vocal behaviour of allied male dolphins during cooperative mate guarding," Animal Cognition 22: 991 to 1000, 2019. pmc.ncbi.nlm.nih.gov (locator: Abstract; Introduction; Methods, Behavioural data; Results; Discussion). Accessed 2026-10-28. (full text reviewed)
NOAA Fisheries, "Common Bottlenose Dolphin," species page, last updated September 1, 2026. fisheries.noaa.gov (locator: diet, behavior and threats sections). Accessed 2026-10-28. (full text reviewed)
Patterson IAP, Reid RJ, Wilson B, Grellier K, Ross HM, Thompson PM, "Evidence for infanticide in bottlenose dolphins: an explanation for violent interactions with harbour porpoises?" Proceedings of the Royal Society B 265(1402): 1167 to 1170, 1998. pmc.ncbi.nlm.nih.gov (locator: Abstract). Accessed 2026-10-28. (abstract only)
Dunn DG, Barco SG, Pabst DA, McLellan WA, "Evidence for infanticide in bottlenose dolphins of the western North Atlantic," Journal of Wildlife Diseases 38(3): 505 to 510, 2002. digitalcommons.odu.edu (locator: Abstract). Accessed 2026-10-28. (abstract only)
Gross S, Claus P, Wohlsein P, Kesselring T, Lakemeyer J, Reckendorf A, Roller M, Tiedemann R, Siebert U, "Indication of lethal interactions between a solitary bottlenose dolphin (Tursiops truncatus) and harbor porpoises (Phocoena phocoena) in the German Baltic Sea," BMC Zoology 5: 12, 2020. link.springer.com (locator: Abstract, Background, Results, Conclusion). Accessed 2026-10-28. (abstract only)
Puig-Lozano R, Fernández A, Saavedra P, Tejedor M, Sierra E, De la Fuente J, Xuriach A, Díaz-Delgado J, Rivero MA, Andrada M, Bernaldo de Quirós Y, Arbelo M, "Retrospective Study of Traumatic Intra-Interspecific Interactions in Stranded Cetaceans, Canary Islands," Frontiers in Veterinary Science 7: 107, 2020. pmc.ncbi.nlm.nih.gov (locator: Abstract; Materials and Methods; Results, Table 3; Discussion). Accessed 2026-10-28. (full text reviewed)
NOAA Fisheries, "Protect Wild Dolphins: Admire Them from a Distance," last updated October 3, 2023. fisheries.noaa.gov (locator: full page). Accessed 2026-10-28. (full text reviewed)
NOAA Fisheries, "Frequent Questions: Feeding or Harassing Marine Mammals in the Wild," last updated January 20, 2026. fisheries.noaa.gov (locator: full page). Accessed 2026-10-28. (full text reviewed)
NOAA Fisheries, "Marine Life Viewing Guidelines: Guidelines and Distances." fisheries.noaa.gov (locator: General Viewing Guidelines; dolphins and porpoises). Accessed 2026-10-28. (full text reviewed)
Christiansen F, McHugh KA, Bejder L, Siegal EM, Lusseau D, McCabe EB, Lovewell G, Wells RS, "Food provisioning increases the risk of injury in a long-lived marine top predator," Royal Society Open Science 3(12): 160560, 2016. pmc.ncbi.nlm.nih.gov (locator: Abstract; 2.1 Study area; Results). Accessed 2026-10-28. (full text reviewed)
Nunny L, Simmonds MP, "A Global Reassessment of Solitary-Sociable Dolphins," Frontiers in Veterinary Science 5: 331, 2018. pmc.ncbi.nlm.nih.gov (locator: Abstract; Stages of Sociability; interactions with humans; reference 38). Accessed 2026-10-28. (full text reviewed)
Reggente MALV, Papale E, McGinty N, Eddy L, de Lucia GA, Bertulli CG, "Social relationships and death-related behaviour in aquatic mammals: a systematic review," Philosophical Transactions of the Royal Society B 373(1754): 20170260, 2018. pmc.ncbi.nlm.nih.gov (locator: Abstract). Accessed 2026-10-28. (abstract only)
Lyamin OI, Manger PR, Ridgway SH, Mukhametov LM, Siegel JM, "Cetacean sleep: an unusual form of mammalian sleep," Neuroscience and Biobehavioral Reviews 32(8): 1451 to 1484, 2008. pmc.ncbi.nlm.nih.gov (locator: Abstract). Accessed 2026-10-28. (abstract only)

Last checked against its sources on October 28, 2026. Report a correction.