348
N K Humphrey, G R Keeble
peep behind a tree, a hand thrust into a bush, the turning of a stone. But whatever
form the sampling takes we may expect that monkeys are biologically adapted to
conduct it both methodically and economically. No monkey can afford either to
spend all its time collecting information or, for that matter, to be slack about it;
the sensible strategy, as every human ethological observer knows, is to space out
successive observations in a way which yields maximum information for least effort.
This means (i) generating an appropriate statistical distribution of intervals between
observations, (ii) choosing an appropriate mean sampling frequency.
We suggest that a Poisson distribution, containing a semirandom assortment of
intersample intervals, may represent an efficient strategy for detecting some of the
short-term changes in addition to most of the longer-term changes in an unpredictable
environment. And, further, we suggest that, given this distribution, the mean
sampling frequency (which is inversely proportional to the half-life) ought to be
tuned to the a priori probability of change in the environment.
Let us explain this last suggestion and relate it to the results of the experiment.
Although all environments may be unstable, some are more unstable than others.
In the kind of world that monkeys naturally live in, the probability of significant
change must vary both with general factors such as the time of day, the weather, the
season and also with more specific ones such as the social situation, the threat of
predators, the nature of the immediate habitat, etc. Some of these factors will be
associated with particular kinds of background sensory stimulation—sounds, colours,
temperature, atmospheric pressure, etc. We propose, in the light of our experimental
results, that red light presages more rapid change than blue, and loud noise more
rapid change than quietness.
Although these proposals about noise and colour are post hoc, they make good
sense.
Colour. Red is associated with dusk and dawn (when predators are active), with
fire, with blood, with oestrous swellings, with ripe fruits, with stinging insects, and a
host of other natural ‘signs’ (see Humphrey 1976). Blue is associated with the
midday sky—and very little else.
Noise. Loud noise is associated with thunderstorms, falling trees, approaching
elephants, screams of anger or fear, etc. Quietness is associated with quietness.
No wonder, therefore, that monkeys might be more alert to change in a red and/or
a noisy environment than in a blue and/or a quiet one. It is not that their subjective
clocks go faster under the former conditions, but that the world itself does.
Acknowledgements. This research was supported by the Science Research Council. We are grateful
to P P G Bateson and MJA Simpson for their critical reading of the manuscript.
References
De Marco A, Penengo P, Trabucco A, Borsellino A, Carlini F, Riani M, Tuccio M T, 1977
“Stochastic models and fluctuations in reversal time of ambiguous figures” Perception 6 645-656
Humphrey N K, 1976 “The colour currency of Nature” in Colour for Architecture Eds T Porter,
B Mikellides (London: Studio Vista) pp 95-98
Humphrey N K, Keeble G R, 1977 “Do monkeys’ subjective clocks run faster in red light than in
blue?” Perception 6 7-14
© 1978 a Pion publication printed in Great Britain