Nyheter

  1. Du är här:
  2. Start
  3. Institutionen för arkeologi och antikens kultur
  4. Om oss
  5. Nyheter
  6. A new computational theory of social learning in animals

A new computational theory of social learning in animals

Associative mechanisms can result in transfer of information and behaviour from experienced to naive individuals. A new model and results from the Centre for Cultural Evolution show that associative processes supported by genetic predisposition can account for most types of social learning processes found in nonhuman animals.

Social transmission of information is a key phenomenon in the evolution of behavior and in the establishment of traditions and culture. To this day, a plethora of social learning processes have been described. Here, we used a computational formulation of associative learning, taking instinctual aspects of behavior into account, to analyze social learning in nonhuman animals. Our results show that associative processes supported by genetic predisposition can account for most types of social learning processes found in nonhuman animals. We conclude that combining associative learning with instinctual aspects of behavior provides a unitary framework for the study of social learning in nonhuman animals.

Raccoon
Photo: Johan Lind/N

 

The full paper is publicly available:

Johan Lind, Stefano Ghirlanda & Magnus Enquist (2019): Social learning through associative processes: a computational theory. Royal Society Open Science (6), 181777.

Bokmärk och dela Tipsa

KONTAKT

Postadress/leveransadress
Institutionen för arkeologi och antikens kultur
Stockholms universitet
Wallenberglaboratoriet
106 91 Stockholm

Fakturaadress
Stockholms universitet
Ref. 108
Box 50741
20270 Malmö
(för utländska leverantörer gäller dock postadressen)

Besöksadress
Wallenberglaboratoriet
Lilla Frescativägen 7
114 18 Stockholm
(lastkaj finns)

Faxnummer
Antikens kultur och samhällsliv:
08-16 13 62
Arkeologi, Arkeologiska- och osteoarkeologiska forskningslaboratorierna:
08-674 73 66

Organisationsnummer
202100-3062

Information om larm