Arousal and the Causation of Behaviour

Abstract

AbstractA. A number of very different meanings have been given to the term "arousal". These have here been distinguished as follows: (1a) Responsiveness. The likelihood at a particular moment that any one of a number of different stimuli will evoke the response appropriate to it if presented. (ib) Behavioural intensity. A variable which affects the vigour and completeness of whichever response is elicited. (2) Activation. A variable, systematic changes in which determine the kind of response which is possible (e.g. grooming or copulation). (3) Arousal as drive. A variable which explains the association together in time of a group of responses, which may be held to include alert responses, cardiac acceleration, increase in postural tonus and a variety of "emotional" responses. (4) Level of sensory input. A variable which regulates the level of sensory input. Mechanisms of selective attention are often included. (5) Property of a functional system in the CNS. The syndrome resulting from increased activity in the ascending activating system. Some or all of these meanings are commonly confused together, which greatly reduces their value in description and analysis. Further it is usually assumed, but with little little firm evidence, that changes in all these variables are continuous and similar in nature from deep sleep to states of extreme wakefulness. These latter states have rarely been clearly characterised. B. For a number of studies the use made of arousal as an explanatory concept is discussed. (1) Changes in responsiveness are important in behaviour, and deserve wider study and theoretical discussion. Data on behavioural intensity are scanty. (2a) Activation models can be used to explain changes in the likelihood of different acts through the course of normal behavioural cycles from one period of sleep to the next. However, except for grooming which tends to occur before and after drowsiness, most acts appear to be affected by a general progressive increase in responsiveness following awakening. The existence of a number of groups of acts, each associated with a particular range of activation (arousal) remains to be proved. (2b) Experimental manipulation of drowsiness in the chick by the administration of hexoses either into the crop or directly into the hypothalamus and other areas of the brain, has at least three relatively independent effects on behaviour. Drowsy periods interfere directly with other responses. At the same time approach to drowsiness is accompanied by a loss of inhibition in responses such as pecking. This is probably due to reduced forebrain activity since it can be reproduced by spreading depression due to KCl. Finally, familiar surroundings are treated as novel immediately after waking. The second and third effects cannot be handled by activation models. (2c) Attempts to vary activation by startling and painful stimuli have shown that some responses may be markedly and unexpectedly facilitated (e.g. copulation by electric shock). However, other startling stimuli may have no effect at all, and recovery from the stimulus does not show the sequential facilitation of a number of different acts which an activation model would predict. (3) Various cardiovascular, respiratory, thermoregulatory and other reflexes, which represent feedforward regulation of the physiological consequences of a predicted increase in exertion, have been held to be indices of heightened arousal. Their use as such is complicated by the fact that the sustained activation of such reflexes in the absence of exertion is inevitably disturbed by feedback regulation of the disturbances which result. Further, a novel stimulus can also produce preparations for immobility (e.g. cardiac deceleration), if the animal examines it carefully rather than at once responding. Attempts to explain changes in these two complexes of reflexes in terms of arousal changes have become so confusing that they are best abandoned. The pattern of alert responses shown at waking is commonly taken as the behavioural counterpart of cortical arousal. However, once waking has occurred it is not possible to identify further patterns of alert responses which can be used to characterise a series of states of progressively increasing arousal. Instead at least two phases of attention, scanning and focussed, can be distinguished over a wide range of states of responsiveness or autonomic arousal. (4) Estimates of the rate of intake or of passage of information are extremely difficult. They are complicated by the fact that studies of selective attention have shown that preliminary recognition of stimuli proceeds in channels other than that in primary use, so that any definition of arousal in terms of information passage must also consider the extent to which attention is confined to the primary channel. Evidence for the homeostatic regulation of arousal, so defined, is indirect and scanty. (5) At least two brain mechanisms are primarily responsible for different phenomena which have been ascribed to arousal. The ascending activating system (AAS) is involved in the appearance of alert responses in a previously drowsy animal, whilst the central mesencephalic grey (CMG) and its diencephalic connections appear to mediate both preparations for exertion and such "emotional" responses as piloerection. The relationship between the two systems and, in particular, the effects of sustained activation of the AAS in the absence of effects on the CMG require further study uncomplicated by prior assumptions about the nature of arousal. It is probable that other of the phenomena discussed here are also primarily mediated by relatively independent brain structures (e.g. focussed attention and the hippocampus).