It is important to note that just as, on the psychological side, memory always involves the awakening of a previous conscious experience by an associated idea, one that was an element in the previous system of associated sensations, perceptions, thoughts, etc., making up the experience, so, on the physiological side, we must suppose that it involves stimulation of the whole system of neurons belonging to this experience by the physiological stimulus corresponding to the conscious element or stimulus. For instance, if I see my friend A, the image is not a memory, though it is one I have had many times before and has left residua of itself capable of being reproduced as memory. But if I see his hat, and immediately previously linked pictorial images of him arise in my mind; or, if, when I see him, there arise images of his library in which I have previously seen him, these images are memory. A conscious memory is always the reproduction of an experience by an associated idea or other element of experience (conscious or subconscious). Similarly we must infer that the neurons correlated with any past mental experience are stimulated by associated neuron processes. This is the foundation-stone of mental physiology; for upon the general principle of the correlation of mental processes with neural processes rests the whole of cerebral localization and brain physiology. Although we assume newly arranged dynamic associations of neurons corresponding to associations of ideas, we do not know how this rearrangement is brought about, though we may conceive of it as following the physiological laws of lowering of thresholds of excitability. Nor do we know whether the modifications left as residua (by which the thresholds are lowered) are physical or chemical in their nature, though there is some reason for believing they may be chemical. =Chemical and physical theories of residua.=—It is possible that, through chemical changes of some kind left in the system of neurons corresponding to an experience, the neurons may become sensitized so as to react again as a whole to a second stimulus applied to one element. In other words a hyper-susceptibility may become established. There is a physiological phenomenon, known as anaphylaxis, which may possibly prove more than analogous, in that it depends upon the production, through chemical changes, of hyper-susceptibility to a stimulus which before was inert. The phenomenon is one of sensitizing the body to certain previously innocuous substances. If, for instance, a serum from a horse be injected into a guinea pig no observable reaction follows. But, if a second dose be injected, a very pronounced reaction follows and the animal dies with striking manifestations called anaphylactic shock. This consists of spasm of the bronchioles of the lungs induced by contraction of their unstriated muscles and results in an attack of asphyxia.[57]