25.4 Actimetry

349

Tracking movement, which poses many difficulties of quantification for human

subjects (but see Sect. 25.4), might not be necessary. For elderly people living at

home, minute changes in skin colour can be detected by low-resolution cameras

placed around the house, or even the standard cameras that are now ubiquitous in

mobile devices such as smartphones or tablets. ¹¹

Even bacterial behaviour can be represented at a higher level than mere respira-

tion: rate of growth and tumbling motions are characteristic. As with the fruit flies,

behaviours can be classified into discrete behavioural states. Once that is achieved

one can examine transitions between behaviours. Any system for monitoring bacte-

ria, however, needs to be able to contend with the short generation time, during which

individuals may lose their identity and become two new entities. One approach to

dealing with this problem is to use a system of interconnected compartments and

ensure that each compartment contains at most one bacterium, which can then be

observed unencumbered by congeners. ¹²

Once the behavioural data have been classified into discrete behavioural states,

for any individual the protocol of actual behaviour will be a long one-dimensional

string of symbols identifying the different states. From this stochastic information a

matrix of transition probabilities between the different states can be determined and

we have a Markov chain, ¹³ which can then be further analysed.

von Foerster (1970) has proposed “molecular ethology” as a term to describe

the concept of a bridge linking the macroscopic phenomena of behaviour with the

structure and function of microscopic elementary units, much as molecular genetics

is the bridge that links macroscopic phenomenology such as a taxonomy of species

with microscopic elementary units such as the metabolome.

25.4

Actimetry

Actimetry, also called actigraphy, comprises the measurement of (typically human)

activity through accelerometers worn on the body. ¹⁴ To avoid the problem of the

subject’s behaviour being influenced by the measurement, the field had to wait for

the development of miniature (microsystems) devices, but these are now readily and

cheaply available. Considering the vast potential, rather little work has been done in

the field; it mostly seems to be used in sleep medicine, and even there success has

been limited. ¹⁵

11 Pearce (2017).

12 Wakamoto et al. (2005).

13 Sect. 11.2; see also Ashby (1956), Chap. 9.

14 Dancsházy et al. (2004).

15 Kellner et al. (1997), Shoch et al. (2019).