talk

Views:
 
Category: Entertainment
     
 

Presentation Description

No description available.

Comments

Presentation Transcript

Flexible house hunting strategies in social insects: 

Flexible house hunting strategies in social insects Bob Planqué Lorentz Center January 29, 2007

How to design for robustness?: 

How to design for robustness?

How to design for robustness?: 

How to design for robustness?

Behavioural diversity in animal groups: 

Behavioural diversity in animal groups Individual complexity vs group integration

Ant emigrations: collective crisis management: 

Ant emigrations: collective crisis management

Much to organize…: 

Much to organize… Information-gathering Evaluation Deliberation Consensus-building Choice and implementation

Much to organize…: 

Much to organize… Information-gathering Evaluation Deliberation Consensus-building Choice and implementation

Much to find out…: 

Much to find out… Size Darkness Entrance width Cavity height

Slide12: 

Mallon et al., 2001, Pratt et al., 2002

Slide13: 

Mallon et al., 2001, Pratt et al., 2002

Slide14: 

searching Mallon et al., 2001, Pratt et al., 2002

Slide15: 

searching Mallon et al., 2001, Pratt et al., 2002

Much to organize…: 

Much to organize… Information-gathering Evaluation Deliberation Consensus-building Choice and implementation

Magic carpet experiment: 

Magic carpet experiment

Buffon’s Needle: 

Buffon’s Needle Estimated area is inversely proportional to number of intersections between two random lines drawn within the area

Much to organize…: 

Much to organize… Information-gathering Evaluation Deliberation Consensus-building Choice and implementation

Slide22: 

Forward tandem runs tandem runs Mallon et al., 2001, Pratt et al., 2002 Franks & Richardson, 2006

Much to organize…: 

Much to organize… Information-gathering Evaluation Deliberation Consensus-building Choice and implementation

Slide26: 

!! Quorum is met tandem runs Mallon et al., 2001, Pratt et al., 2002

Much to organize…: 

Much to organize… Information-gathering Evaluation Deliberation Consensus-building Choice and implementation

Slide28: 

Stephen Pratt Social carrying transport Mallon et al., 2001, Pratt et al., 2002

Slide29: 

transport complete Mallon et al., 2001, Pratt et al., 2002

Slide30: 

Ants in old nest Scouts Recruiters Carried ants in new nest

Slide31: 

Ants in old nest Scouts Recruiters FTR Carried ants in new nest

Slide32: 

Ants in old nest Scouts Recruiters Carried ants in new nest transport

Ant emigrations: An exercise in logistics: 

Ant emigrations: An exercise in logistics

Best of n: 

Best of n

Slide35: 

When pressure is low: use high quorum, make precise choice When pressure is high: use low quorum, make any choice Franks et al., 2003 Trading speed for accuracy

Slide37: 

Ants do not switch equally throughout an emigration Switching without costs

Slide39: 

Gradually stronger commitment improves emigration time and accuracy

An enigma: reverse tandem runs: 

An enigma: reverse tandem runs

Function in behavioural ecology: 

Function in behavioural ecology “What function does a behaviour have?” How does the behaviour add to animal’s fitness? Use speed of emigration as proxy for fitness

Why do ants perform tandem runs after quorum?: 

Why do ants perform tandem runs after quorum? To speed up an emigration: more hands make light work? As a control mechanism: re-uniting worker ants at old nest?

Why do ants perform tandem runs from the new to the old nest?: 

‘Home’ has changed? Ants in new nest have extra knowledge? To avoid initial overcrowding? To make sure there are enough ants to look after brood? To use time more efficiently? Why do ants perform tandem runs from the new to the old nest?

Available evidence: 

Available evidence RTRs occur after quorum has been met RTRs followed by scouts, recruiters Sometimes ants are first carried to new nest, before RTR Ants that followed an RTR later carried other ants or brood Mallon et al., 2001, Pratt et al., 2002

When can RTR be a back-up strategy?: 

When can RTR be a back-up strategy? If FTRs had failed. I.e., when there were no ants at the old nest willing to follow an FTR. This occurs if All these potential recruiters have gone scouting There were few potential recruiters to start with The new nest is hard to find

Slide47: 

Passive ants in old nest Scouts Active ants in old nest Old nest New nest

Slide48: 

Passive ants in old nest Scouts Active ants in old nest Old nest New nest Recruiters FTR

Slide49: 

Passive ants in old nest Scouts Old nest New nest Recruiters

Slide50: 

Passive ants in old nest Scouts Old nest New nest Recruiters Carried ants in old nest

Slide51: 

Passive ants in old nest Scouts Old nest New nest Recruiters Carried ants in old nest RTR? Carried ants for RTRs

Slide52: 

Idea: let ants choose which set of mechanisms gives fastest emigration.

Approach: 

Approach Vary scouting rate fraction of active ants probability to find the new nest Find optimal emigration strategy, w.r.t. fraction of time spent on RTRs quorum threshold

Comparing scouting rate and fraction of active ants: 

Comparing scouting rate and fraction of active ants

Few active ants, high scouting rate: No RTR + High Quorum: 

Few active ants, high scouting rate: No RTR + High Quorum

Slide59: 

Few active ants, high scouting rate: Full RTR + Low Quorum

Slide60: 

Many active ants, low scouting rate: Full RTR + Low Quorum

Many active ants, low scouting rate: No RTR + High Quorum: 

Many active ants, low scouting rate: No RTR + High Quorum

Model prediction: 

Model prediction If panic is great new nest is hard to find there are few active ants then do NOT rely on FTRs, but use a low quorum and switch to a mix of carrying and RTRs

Slide65: 

new nest old nest

Slide66: 

new nest old nest

Slide67: 

new nest old nest scouts

Slide68: 

new nest old nest scouts Control after 20 minutes

Slide69: 

new nest old nest scouts Test : lost scouts after 20 minutes

Slide70: 

new nest old nest

Slide71: 

p = 0.028 p = 0.007

Slide72: 

Ants use decentralized control Can flexibly change alternatives Use quorum sensing to come to collective decision Change strategies if situation demands Conclusions

Future work: 

Future work Together with Dick James, Nick Britton (Bath) Nigel Franks, Ana Sendova-Franks (Bristol) Food transport network (trophallaxis) How to optimize the spread of good substances (food), whilst minimizing the spread of bad ones (diseases)?

Slide74: 

Replicated experiments, with nearly complete interaction network.

Acknowledgements: 

Acknowledgements FX Dechaume-Moncharmont Nigel Franks James Marshall Tim Kovacs Ana Sendova-Franks Eamonn Mallon Stephen Pratt Anna Dornhaus Gemma Harfield Vicky Pook

authorStream Live Help