AboutZoos, Since 2008


Wolf species can be iden­ti­fied by their ‘howl­ing dialects’

pub­lished 13 Feb­ru­ary 2016 | mod­i­fied 13 Feb­ru­ary 2016

Howling wolfLargest quan­ti­ta­tive study of howl­ing, and first to use machine learn­ing, defines dif­fer­ent howl types and finds that wolves use these types more or less depend­ing on their species, resem­bling a howl­ing dialect. Researchers say find­ings could help con­ser­va­tion efforts and shed light on the ear­li­est evo­lu­tion of our own use of language.

The largest ever study of howl­ing in the ‘canid’ fam­ily of species — which includes wolves, jack­als and domes­tic dogs — has shown that the var­i­ous species and sub­species have dis­tin­guish­ing reper­toires of howl­ing, or “vocal fin­ger­prints”: dif­fer­ent types of howls are used with vary­ing reg­u­lar­ity depend­ing on the canid species.

Researchers used com­puter algo­rithms for the first time to analyse howl­ing, dis­till­ing over 2,000 dif­fer­ent howls into 21 howl types based on pitch and fluc­tu­a­tion, and then match­ing up pat­terns of howling.

They found that the fre­quency with which types of howls are used — from flat to highly mod­u­lated — cor­re­sponded to the species of canid, whether dog or coy­ote, as well as to the sub­species of wolf.

For exam­ple, the howl­ing reper­toire of the tim­ber wolf is heavy with low, flat howls but doesn’t fea­ture the high, loop­ing vocal that is the most fre­quently used in the range of howls deployed by critically-​endangered red wolves.

The sur­vival of red wolves in the wild is threat­ened by inter­breed­ing with coy­otes, and we found that the howl­ing behav­iour of the two species is very sim­i­lar. This may be one rea­son why they are so likely to mate with each other, and per­haps we can take advan­tage of the sub­tle dif­fer­ences in howl­ing behav­iour we have now dis­cov­ered to keep the pop­u­la­tions apart
Dr. Arik Ker­shen­baum, lead author, Depart­ment of Zool­ogy, Uni­ver­sity of Cam­bridge, UK »

Ker­shen­baum describes these dis­tinc­tive howl reper­toires as resem­bling vocal dialects, with each species hav­ing its own iden­ti­fi­able use of the var­i­ous howl types. He says the find­ings could be used to track and man­age wild wolf pop­u­la­tions bet­ter, and help mit­i­gate con­flict with farmers.

The ori­gins of lan­guage devel­op­ment in humans are mys­te­ri­ous, as the vocal­i­sa­tions of our clos­est exist­ing bio­log­i­cal rel­a­tives such as chim­panzees are rel­a­tively sim­ple. Ker­shen­baum and col­leagues believe that study­ing the sounds of other intel­li­gent species that use vocal com­mu­ni­ca­tion for coop­er­a­tive behav­iour — such as wolves and dol­phins — may pro­vide clues to the ear­li­est evo­lu­tion of our own use of language.

Wolves may not be close to us tax­o­nom­i­cally, but eco­log­i­cally their behav­iour in a social struc­ture is remark­ably close to that of humans. That’s why we domes­ti­cated dogs — they are very sim­i­lar to us,” said Ker­shen­baum, from Cambridge’s Depart­ment of Zoology.

Under­stand­ing the com­mu­ni­ca­tion of exist­ing social species is essen­tial to uncov­er­ing the evo­lu­tion­ary tra­jec­to­ries that led to more com­plex com­mu­ni­ca­tion in the past, even­tu­ally lead­ing to our own lin­guis­tic abil­ity” he said.

The research was con­ducted by a team of sci­en­tists from the UK, US, Spain and India, and the find­ings are pub­lished in the March issue of the jour­nal Behav­ioural Processes.

The researchers made use of howls recorded from both cap­tive and wild ani­mals, from Aus­tralia and India, to Europe and the United States, cre­at­ing a data­base of 6,000 howls that was whit­tled down to 2,000 for the study. This included comb­ing YouTube for domes­tic dog howls.

These were then fed into machine learn­ing algo­rithms to clas­sify the howls into dis­crete types. Stud­ies on howl­ing in the past have had to rely on sub­jec­tive human com­par­isons by look­ing at sound­wave pat­terns, but the new algo­rithms allowed the howl types to be com­pared objec­tively, reveal­ing that the var­i­ous species have char­ac­ter­is­ti­cally dif­fer­ent reper­toires of howl type usage.

While the howl­ing reper­toires of most of the 13 species analysed were very dis­tinct, some bore close sim­i­lar­i­ties to each other that may influ­ence inter­breed­ing and, in at least one case, threaten the sur­vival of a species.

Red wolves, hunted to the brink of extinc­tion in the mid-​20th cen­tury, were the focus of a rein­tro­duc­tion pro­gramme insti­gated by the US gov­ern­ment, which has recently been halted due to a lack of success.

Part of the prob­lem was red wolves breed­ing with coy­otes, and the resul­tant hybridi­s­a­tion diluted attempts to main­tain this rare wolf species. The researchers found sig­nif­i­cant over­lap between the howl­ing vocab­u­lary of the red wolf and the coy­ote — with both favour­ing highly mod­u­lated, whin­ing howls such as the one classed by researchers as ‘type three’.

The sur­vival of red wolves in the wild is threat­ened by inter­breed­ing with coy­otes, and we found that the howl­ing behav­iour of the two species is very sim­i­lar. This may be one rea­son why they are so likely to mate with each other, and per­haps we can take advan­tage of the sub­tle dif­fer­ences in howl­ing behav­iour we have now dis­cov­ered to keep the pop­u­la­tions apart,” said Kershenbaum.

Other con­ser­va­tion uses for the new find­ings may involve refin­ing the use of play­backs to recre­ate more accu­rate howl­ing behav­iours that imi­tate ter­ri­to­r­ial mark­ings, thereby encour­ag­ing wolf packs to steer clear of farms and livestock.

How­ever, we know very lit­tle about the mean­ing of dif­fer­ent howl types and what they are actu­ally com­mu­ni­cat­ing, says Ker­shen­baum, because — as with dol­phins, that other highly vocal, smart and social species which he stud­ies — wolves are extremely dif­fi­cult to study in the wild.

You don’t observe nat­ural wolf behav­iour in zoos, only in the wild, and you need to know where the ani­mals are when howl­ing before you can really begin to try and dis­cern mean­ings. But, as with dol­phin pods, phys­i­cally fol­low­ing a wild wolf pack is vir­tu­ally impos­si­ble,” explained Kershenbaum.

We are cur­rently work­ing on research in Yel­low­stone National Park in the US using mul­ti­ple record­ing devices and tri­an­gu­la­tion tech­nol­ogy to try and pick up howl sounds and loca­tion. In this way we might be able to tell whether cer­tain calls relate to dis­tance com­mu­ni­ca­tion or pack warn­ings, for exam­ple,” he said.

For Ker­shen­baum, wolves and dol­phins show remark­able par­al­lels with each other in social behav­iour, intel­li­gence and vocal com­mu­ni­ca­tion — all com­par­isons that extend to humans.

As well as being intel­li­gent and coop­er­a­tive species, wolves and dol­phins have remark­ably sim­i­lar vocal char­ac­ter­is­tics. If you slow a dol­phin whis­tle down about 30 times it sounds just like a wolf howl, some­thing I often do in my lec­tures,” he said.

The pres­ence of com­plex ref­er­en­tial com­mu­ni­ca­tion in species that must com­mu­ni­cate to sur­vive was prob­a­bly a cru­cial step in the evo­lu­tion of lan­guage. I think we can shed a lot of light on early evo­lu­tion of our own use of lan­guage by study­ing the vocal­i­sa­tion of ani­mals that are socially and behav­iourally sim­i­lar to us, if not nec­es­sar­ily tax­o­nom­i­cally closely related.”

(Source: Uni­ver­sity of Cam­bridge news release, 08.02.2016)

UN Biodiversity decade
WWF Stop Wildlife Crime
Fight for Flight campaign
End Ivory-funded Terrorism
Support Rewilding Europe
NASA State of Flux

Goal: 7000 tigers in the wild

Tiger range countries map

Tiger map” (CC BY 2.5) by Sander­son et al., 2006.

about zoos and their mis­sion regard­ing breed­ing endan­gered species, nature con­ser­va­tion, bio­di­ver­sity and edu­ca­tion, which of course relates to the evo­lu­tion of species.
Fol­low me on: