Some Aspects of the Embryology of Behavior

Abstract

In the past, the development of behavior in vertebrate embryos has been interpreted in terms of two opposing viewpoints. Coghill, on the basis of his observations on the salamander Amblystoma, has maintained that behavior represents an integrated pattern from beginning to end; partial patterns segregate from total patterns by a developmental process which he calls "individuation." The opposing view, advocated by Windle and other students of the behavior of mammalian embryos and fetuses, contends that local reflexes are the elementary units of behavior that combine secondarily to form integrated patterns. The conceptual and factual foundations of both theories are reappraised, and their limitations and shortcomings are indicated. It becomes evident that Coghill''s concepts, derived from lower forms, cannot be generalized and extended to higher forms. On the other hand, the reflexogenic theory is unacceptable on theoretical and factual grounds. A rein-vestigation of the embryology of behavior in the chick embryo has provided a new frame of reference that places the problem in a different perspective. An old observation of Preyer (1885) to the effect that overt motility is in operation several days before reflex circuits are formed has been followed up, and, in a series of observations and experiments, spontaneous, nonreflexogenic motility was established as a basic component of embryonic behavior. It can be shown that spontaneous motility patterns are not influenced by sensory input. Arguments are presented in favor of the idea that the spontaneous motility in the chick embryo is due to automatic, self-generated activity of neurons rather than to stimulation by agents in the circulation. One of the most significant characteristics of spontaneous motility is its rhythmicity: short phases of activity are followed by longer phases of inactivity. The analysis of recordings of the periodicity in normal embryos and in embryos with transected spinal cords has enabled us to dissociate a trigger mechanism for activation of motility from a device that determines the duration of the activity phases. Spontaneous and reflexogenic motility are regarded as two independent basic constituents of embryonic behavior. The former is considered to be the primary component, not only because it precedes the reflex responses in lower forms but mainly because reflex motility remains latent in the embryo owing to the absence of adequate stimuli; therefore, it cannot contribute to the molding of behavior patterns in prehatching and prenatal stages.