Affiliations:
1Department of Zoology, University of Oxford; South
Parks Road, Oxford, OX1 3PS, UK.2Departament of Genetics, Ecology and Evolution,
Universidade Federal de Minas Gerais; CP 486, Belo Horizonte, Brazil
(gwilson@icb.ufmg.br).3School of Biological Sciences, Life Sciences Building
85, University of Southampton; Highfield Campus, Southampton, SO17 1BJ,
UK (r.j.morris@soton.ac.uk).
Running title: The hidden network of higher-order interactions
Key words: ants, aphids, Baccharis
dracunculifolia, Cerrado, experimental manipulation, galling insects,
indirect interactions, interaction modification, parasitoid wasps,
parasitism
Number of words in the abstract: 124
Number of words in the main text: 4,519
Number of words in each text box: 0
Number of references: 40
Number of figures: 2
Number of tables: 2
Number of text boxes: 0
Correspondence author: Milton Barbosa, email:
miltonbsjunior@ufmg.br, Departament of Genetics, Ecology and Evolution,
Federal University of Minas Gerais, CP 486/ 30161-970, Belo Horizonte,
Brazil, phone: +55 31 3499 2580
Statement of authorship: MB and RJM designed the experiment, MB
collected the data with assistance from GWF, MB analysed the data with
support from RJM; MB wrote the first draft of the manuscript and all
authors edited the manuscript and contributed substantially to the ideas
presented.
Data accessibility statement: should the manuscript be
accepted, the data supporting the results will be archived in the Dryad
Digital Repository and the data DOI will be included at the end of the
article.
Abstract: Transcending pairwise interactions in ecological networks
remains a challenge. Higher-order interactions, the modulation of a
pairwise interaction by a third species, have so far only been
demonstrated in models or small isolated systems. Their ubiquity at a
community level remains unknown. Using field experiments, we tested how
multiple interactions within a network changed with species composition
by reducing the densities of distinct species in a diverse arthropod
community. We revealed an extensive hidden network of higher-order
interactions modifying each other and the “visible” direct
interactions. Most pairwise interactions were affected by the
manipulation of a non-interacting taxonomic group. The pervasiveness of
these interaction modifications challenges pairwise approaches to
understanding interaction outcomes and could shift our thinking about
the structure and resilience of ecological communities.