PhD opportunity exploring fish cognition.

Behaviour and the Origins of Life

Behaviour before genes?

The behaviour of organisms is often assumed to involve sensors and motors that are too complicated to have spontaneously emerged in an abiotic world. The common view is that evolution must have preceded these structures, and for this reason, it may seem counterintuitive to consider how behaviour could have played a role in the origins of life.

However, many examples can be found of non-biological systems that demonstrate life-like behaviour. One of the most compelling examples is that of motile oil-droplets. These are simple systems, involving in some cases less than five common and easily synthesized chemicals, such as olive-oil and soap, but their behaviour is impressive. The droplets don't just move, they propel themselves in a directed manner, moving for instance towards areas that have high concentrations of certain reactants (see Video 1).

Viability-based behaviours

Interestingly, a wide variety of these non-biological systems tend to behave in what might be considered forms of self-preservation. For instance, the motility of the oil-droplets relies upon high alkalinity, and the motion itself tends to cause the droplets to move towards highly alkaline environments. Reaction-diffusion spots, hurricanes, and Belousov-Zhabotinsky gel-worms all perform similar motion towards the conditions necessary for the persistence of either the system, its motility, or both.

These behaviours can be described as viability-based behaviours. In each case, there is an inherently unstable dissipative structure that persists only when there is sufficient accessible free-energy to maintain its ordered state. The behaviours of these systems tend to increase the likelihood of there being sufficient energy available, and this is not mere coincidence. In each case, the behaviour is not an arbitrary response to the environment, but is instead is a response to how the environment affects the self-maintenance of the dissipative structure. This "viability-based" behaviour is reminiscent of the "metabolism-based" behaviour that is observed in a variety of natural organisms, such as the metabolism-based behaviour of bacteria such as Escherichia coli and Azospirillum brasilense, where certain behaviours are not responses to environmnental phenomena, but rather to how well the organism's metabolism is operating (see Figure 1).

Theoretical work has shown that viability-based behaviours can provide a variety of adaptive and evolutionary advantages. For instance, when the behaviour of an organism is in response to internal state-variables (such as blood-sugar, or metabolic rate-indicators), the behaviour:

  1. can adapt to changes in the way that the organism operates, such as those brought about by mutations or other organizational transformations [Nature Scientific Reports];
  2. can evaluate the effects of the environment upon the viability of the organism in an ongoing in-the-moment way that is history-dependent and integrates numerous environmental influences into a single coherent response [PLoS CB];
  3. can facilitate evolution through (i) increasing exposure to environmental variation, (ii) making more likely the fixation of some beneficial metabolic pathways, (iii) providing a mechanism for in-the-moment adaptation to changes in the environment and to changes in the organization of the organism itself, and (iv) generating conditions that are conducive to speciation [Artificial Life].

Next steps

The next step in this line of research is to further investigate the extent to which viability-based behaviour can bootstrap open-ended, complexity-increasing evolution. This involves several parallel branches of research:

Investigate the extent to which viability-based behaviour improves evolvability in a computational model. In this project, we will use a computational model (perhaps similar to AVIDA) to compare the evolution of passively-stable systems (cf RNA-polymers) with the evolution of inherently unstable, self-maintaining systems that perform a form of viability-based behaviour.

Investigate the limitations of viability-based behaviour in a model that fully captures emergent viability-limits. Most models of adaptive behaviour simply assume certain viability limits. In this project, we will simulate a self-maintaining system with emergent viability limits and investigate how various forms of viability-based behaviour are (or are not) capable of responding to viability-affecting environmental changes.

References
Moreno A, Ruiz-Mirazo K. Basic autonomy as a fundamental step in the synthesis of life. Artificial Life 2004 10:3, 235-259
Egbert MD, Pérez-Mercader J. Adapting to Adaptations: Behavioural Strategies that are Robust to Mutations and Other Organisational-Transformations. Scientific Reports. 2016;6 (Article number: 18963).
Egbert MD, Barandiaran XE, Di Paolo EA. A Minimal Model of Metabolism-Based Chemotaxis. PLoS Computational Biology. 2010;6(12):e1001004.
Egbert MD, Barandiaran XE, Di Paolo EA. Behavioral Metabolution: The Adaptive and Evolutionary Potential of Metabolism-Based Chemotaxis. Artificial Life. 2012;18(1):1-25.




Video 1. Chemotactic oil-droplets. These droplets seek low pH conditions. See Maze Solving by Chemotactic Droplets István Lagzi, Siowling Soh, Paul J. Wesson, Kevin P. Browne, and Bartosz A. Grzybowski Journal of the American Chemical Society 2010 132 (4), 1198-1199 DOI: 10.1021/ja9076793
1D Robot
Figure 1. Schematic diagram indicating different relationships between metabolism and behaviour. The top-two figures show metabolism-independent behaviour where the actions of the organism are not influenced by the metabolic-state of the organism, but are purely responses to the organism's environment. In contrast, the two latter relationships show behaviours that are influenced in an ongoing manner by the metabolic state of the organism, processes that are tightly related to the system's viability.