American Society of Naturalists

A membership society whose goal is to advance and to diffuse knowledge of organic evolution and other broad biological principles so as to enhance the conceptual unification of the biological sciences.

“Positive feedback between behavioral and hormonal dynamics leads to differentiation of life-history tactics”

Posted on

Junnosuke Horita, Yoh Iwasa, and Yuuya Tachiki (Dec 2020)

Positive feedback loop between competition and hormone synthesis induces the alternative life-history tactics

Read the Article (Just Accepted)

Stay or migrate? Fish in a school look similar but are different in intrinsic states

Life is not complex. We are complex. Life is simple, and the simple thing is the right thing.
—Oscar Wilde
<i>Masu</i> salmon (<i>Oncorhynchus masou</i>) has multiple phenotypes, and this variation is caused by different hormonal status. Top: migrant male; mid: migrant female; bottom: resident male.<br/>(Credit: J. Horita)
Masu salmon (Oncorhynchus masou) has multiple phenotypes, and this variation is caused by different hormonal status. Top: migrant male; mid: migrant female; bottom: resident male.
(Credit: J. Horita)

In life, we sometimes find ourselves at a crossroads and are perplexed at the choice of paths before us. Wild animals must also make decisions in their lives. For example, some species of salmon have to choose whether to stay in their natal river (called the resident tactic) or undergo a feeding migration to the ocean (called the migrant tactic). Because this life decision largely influences the chances of future success, how individuals choose their way is an important issue. Jun-nosuke Horita (Kyushu Univ.) et al. tackle this problem using a mathematical model. The key factor is individuals’ hormonal state, which triggers the development of the phenotypes required. Even in humankind, which likes to fancy that it makes decisions in a logical way, physiological state affects decision-making.

Horita et al. demonstrate the emergence of discrete patterns of hormonal states arising from a continuous distribution, which would work as an initiation of dimorphisms. They considered the life-history choice of masu salmon, in which maturation in the stream (the resident tactic) is induced by 11-ketoteststerone (11-KT), a kind of androgen. It has been known that 11-KT also affects the individual’s behavior. An individual with a high hormone level moves actively and tends to win competition for resources in a territory. Interestingly, the outcome of such competition, in turn, affects the hormone synthesis: 11-KT is synthesized in the winner of competition. Horita et al. address the possibility that fighting and hormone synthesis form a positive feedback loop, the so-called winner-loser effect. They constructed a mathematical model that describes the events in a fish school in which individuals grow and fight each other for food resources. Their simple model of a dynamical system for size and hormonal state reveals that individuals are finally separated into the two groups (whether they migrate or stay) according to hormonal status, even if individuals are indistinguishable from the viewpoint of size. This work emphasizes the importance of considering behavior and physiology simultaneously thinking about life-history choices.


Competitive interaction among individuals of a single population may result in the differentiation to two or more distinct life history tactics. For example, although they exhibit unimodal size distribution, male juveniles of salmonids differentiate into those going down to the ocean for growth and returning after several years to the natal stream for reproduction (migratory tactic) and those staying in the stream and reproducing for multiple years (resident tactic). In this study, we developed a simple mathematical model for the positive feedback between hormonal and behavioral dynamics, with the expectation of establishing multiple discrete clusters of hormone levels leading to differentiation of life-history tactics. The assumptions were that probability of winning in fighting depends both on the body size and hormone level of the two contestants. An individual with a higher hormone level would be more likely to win the competition, which further enhanced hormone production, forming a positive feedback loop between hormone level and fighting ability. If the positive feedback was strong but not excessive, discrete clusters of hormone level emerged from a continuous distribution. In contrast, no clear clustering structure appeared in the distribution of hormone levels if the winning probability in fighting was controlled by the body size.