Toggle Main Menu Toggle Search

Open Access padlockePrints

A Computational Model of Stereoscopic Prey Capture inPraying Mantises

Lookup NU author(s): Dr James O'Keeffe, Dr Vivek Nityananda, Professor Jenny ReadORCiD



This work is licensed under a Creative Commons Attribution 4.0 International License (CC BY 4.0).


We present a simple model which can account for the stereoscopic sensitivity ofpraying mantis predatory strikes. The model consists of a single “disparity sensor”:a binocular neuron sensitive to stereoscopic disparity and thus to distance fromthe animal. The model is based closely on the known behavioural andneurophysiological properties of mantis stereopsis. The monocular inputs to theneuron reflect temporal change and are insensitive to contrast sign, making thesensor insensitive to interocular correlation. The monocular receptive fields have aexcitatory centre and inhibitory surround, making them tuned to size. Thedisparity sensor combines inputs from the two eyes linearly, applies a thresholdand then an exponent output nonlinearity. The activity of the sensor representsthe model mantis’s instantaneous probability of striking. We integrate this overthe stimulus duration to obtain the expected number of strikes in response tomoving targets with different stereoscopic distance, size and vertical disparity. Weoptimised the parameters of the model so as to bring its predictions intoagreement with our empirical data on mean strike rate as a function of stimulussize and distance. The model proves capable of reproducing the relatively broadtuning to size and narrow tuning to stereoscopic distance seen in mantis strikingbehaviour. Although the model has only a single centre-surround receptive field ineach eye, it displays qualitatively the same interaction between size and distanceas we observed in real mantids: the preferred size increases as prey distanceincreases beyond the preferred distance. We show that this occurs because of astereoscopic “false match” between the leading edge of the stimulus in one eye andits trailing edge in the other; further work will be required to find whether suchfalse matches occur in real mantises. Importantly, the model also displays realisticresponses to stimuli with vertical disparity and to pairs of identical stimuli offeringa ”ghost match”, despite not being fitted to these data. This is the firstimage-computable model of insect stereopsis, and reproduces key features of bothneurophysiology and striking behaviour.

Publication metadata

Author(s): O'Keeffe J, Yap SH, Cornejo IL, Nityananda V, Read JCA

Publication type: Article

Publication status: Published

Journal: PLOS Computational Biology

Year: 2022

Volume: 18

Issue: 5

Online publication date: 19/05/2022

Acceptance date: 10/04/2022

Date deposited: 12/04/2022

ISSN (electronic): 1553-7358

Publisher: Public Library of Science


DOI: 10.1371/journal.pcbi.1009666


Altmetrics provided by Altmetric


Funder referenceFunder name
BB/S009760/1Biotechnology and Biological Sciences Research Council (BBSRC)