Psychology when James wrote of it was supposed to deal with 'states of consciousness,' which seems to connect it definitely with metaphysics. In its later developments, as experimental psychology and psycho-physics it has become as inseparably connected with neurology, and however we may define it as a science per se, we may, for the purpose of the psycho-neural problem, regard it as a nexus between metaphysics on the one side and neurology on the other and as partaking in some degree of each. But in such a region of our knowledge-a region which is so largely inchoate and nebulous-words are as Huxley once called them only "noise and smoke" and may be left to the contentions of logomachists. It is only by an integration or synthesis of these diverse but complementary categories of knowledge and thought that a solution of the problem can be arrived at.

What we call man's knowledge may be said to consist of multitudinous mental symbols in ever-changing relations with one another-symbols which may be weighted with different kinds of what we usually call imagery-it may be visual, auditory, audito-motor, tactile or kinaesthetic, and which in the absence of any such characteristics are usually called 'abstract.' To these multitudinous symbols and their everchanging inter-relations man has in the long course of his phylogenetic history affixed more or less empirically an equally multitudinous number of verbal symbols which we call 'names1,' and it has through the ages been one of the persistent functions of logic to induce cultivated man to abandon the primitive and youthful practice of attaching indiscriminately either a number of different names to the same mental symbol or a number of different mental symbols to the same name. Even the philosophers have not always complied with this behest of the logicians and we find that these mental symbols, when regarded collectively, have in the past been variously denoted by the terms 'ideas,' 'concepts,' 'presentations,' 'representations,' 'intuitions,' 'notions,' 'images,' etc. After Hume's death, and for the most part as the result of his writings, the English tradition more or less crystallized into the doctrine that the human mind was concerned with forming discrete 'ideas' and making them cohere in accordance with certain laws of 'association.' 'The association of ideas' was held to be the essential mental function. James' Principles of Psychology dealt this doctrine a heavy blow, and Bradley is held by some to have demolished it, but we imagine that Bradley himself might have been more inclined to regard the doctrine as a half-truth, which, like every other half-truth, is ever 1 These 'names' are subject to the same processes of error as the symbols which they denote, whether in their growth, structure or interpretation.

falling deeper and deeper into the pit of error the more it is regarded as a whole truth. James' Principles pointed to the primary continuity of mental life and showed that what first required explanation was not the method of effecting connections or 'associations' but the growth of distinctions, both in "the big, blooming, buzzing confusion" of the immediate perceptual flux and also in the continuity of the trains of thought which are unceasingly passing through what we colloquially call our 'minds.'

We propose to follow the tradition of Wm. James in using for these mental symbols generally, whatever type of imagery be connected with them or whether they be entirely abstract, the term 'Concepts.' This term concept, which goes back to Plato, has long denoted in philosophical discourse a mental entity having in philosophic parlance the characteristics of concreteness, universality and invariability. Many years before his death James saw that the first step towards a synthesis of the complementary but diverse aspects of the psycho-neural problem lay in finding something in the nature of a single constituent element of our cognitive dispositions. "We seem," he wrote (1890), “if we are to have an elementary psycho-physic law at all thrust right back on something like the mental-atom theory since the molecular fact being an element of the brain would seem naturally to correspond not to total thoughts but to elements of thought1." At the end of his life James attempted to find such a mental atom or constituent element of all our cognitions in the concept but he was hampered in this quest by the traditional view of the concept which regarded it as a static rigid entity which never varied. "A concept never varies 2," he said in one of the chapters in his posthumously published volume on Problems of Philosophy. In this brief statement he was voicing the traditional philosophic view.

Bergson saw that any attempt to explain thought by means of rigid and unchanging elements was hopeless, and in his Introduction to Metaphysics postulated a view of the concept or mental symbol as a supple, mobile, almost fluid representation always ready to mould itself on the fleeting forms of intuition. This view of Bergson has recently been

3

1 Text Book of Psychology, p. 464.

2 Some Problems of Philosophy, p. 53.

"Certainly, concepts are necessary to metaphysics for all the other sciences work as a rule with concepts, and metaphysics cannot dispense with the other sciences. But it is only truly itself when it goes beyond the concept, or at least when it frees itself from rigid and ready-made concepts in order to create a kind very different from those which we habitually use; I mean supple, mobile and almost fluid representations, always ready to mould themselves on the fleeting forms of intuition." An Introduction to Metaphysics, by Henri Bergson; translation by T. E. Hulme, p. 18.

elaborated by one of us (G. G. C.) with the aid of Mr Santayana and he shows by a number of concrete examples1 that a concept when once formed grows in the course of the ontogenetic life process in a somewhat similar fashion to the different organs of the body by a growth in mass or bulk accompanied by progressive differentiation of structure. This view of the concept satisfies the requirements of James and Bergson and provides us with a mental element subject to continual modification by means of the new experience which is being progressively acquired through sense-perception-an element of which all our cognitions may be held to be gradually built up in the course of the ontogenetic life process. We adopt this term 'concept' in the sense and with the connotations here noted as the term by which we denote the symbol or mental atom of which James thirty-five years ago indicated the need and we proceed to enquire what this mental element of our various and multitudinous cognitions has as its correlate in the neural elements and processes of the brain.

During the twenty years before his death the late Richard Semon of Vienna in laborious biological studies of the abiding effects of transient stimuli on irritable living tissues-studies which he pursued through a wide range of biological types culminating in Man-gave to the abiding effects of such stimuli the term 'engram.' In the last of his works, Bewusstseinsvorgang und Gehirnprozess, published in 1920, he considers more intimately the engram in relation to the psycho-neural problem, and in this connection he means by it a physiological pattern established by successive stimuli and resulting in a condition in which there have gradually been produced, by recurrent stimuli and faciliation, paths for a ready connection of neural impulse between many and diversely scattered groups of neurones in various cerebral areas. These engrams when once formed may remain in a latent condition or may at any time become active by an impulse discharged through them. In the arrangement of terms which Semon adopted, the 'engram' remained in a latent condition until 'ecphorized' or made active by the discharge of a neural impulse through it. These engrams were held by him to exist in all degrees of complexity, to dichotomize again and again, and at each period of their active or 'ecphorized' condition to form new engrams. within themselves 2. It is further to be noted that Semon regarded the mental correlates of the engrams (or the symbols which we call 'concepts') to be below the 'threshold of consciousness' during the latent

1 Elements in Thought and Emotion, chap. ii, on Percept and Concept.

2 Mnemic Psychology, by Richard Semon, p. 258.

condition of the engram and as above the threshold during its active or 'ecphorized' condition. We are inclined to regard this as an inadequate explanation of this phenomenon and shall refer to the point again later.

We find in studying the masterly paper by Drs Head and Holmes on "Sensory Disturbances from Cerebral Lesions1" that some at any rate of Semon's conclusions had been also reached by these investigators, although the terms in which their conclusions are expressed differ in the way which is usual when workers in the same field of research are engaged on the same problem but working in isolation from one another. This classic piece of research was immediately concerned only with those groups of sensory impulses which reach the brain through the spinal cord and medulla but we consider that the conclusions reached in regard to the slowly-acquired systems of paths in the cortex for these sensory impulses have a much wider and more universal application, and that it will be only by their application to all the afferent nerves of the cerebrospinal system and perhaps also to the sympathetic and para-sympathetic systems that their full significance will be realized and the contribution which they, with Semon's engrams, make to a solution of the psychoneural problem will be fully understood.

Their researches led them to the conclusion that sensory impulses from the cord and medulla, after exciting the essential thalamic centres, leave the optic thalami in five main functional groups for distribution in the cortex2.

1. Those which underlie postural recognition and the appreciation of passive movement.

2. Those which underlie the recognition of tactile differences or the power of appreciating those qualities of touch other than contact and roughness (e.g. weights of objects on hand).

3. Those upon which depend spacial discrimination (compass points) and its allied faculty, the recognition of size and shape.

4. Those impulses which enable the patient to recognize the spot stimulated (localization).

5. Thermal impulses.

They show that the appreciation and recognition of the import of these groups of sensory impulses depend on separate 'schemata' or systems of neural paths slowly formed during the ontogenetic life process and which may be severally destroyed by cortical lesions. These

1 Brain, XXXIV, p. 183.

2 Ibid.

'schemata' along which such impulses pass enable us to recognize, locate and analyse, with an accuracy so subtle as to be utterly beyond expression by means of the clumsy machinery of language, the results of sensory impulses which cannot be resolved into types of imagery and yet which contribute in most important ways to our integrated experience. They have their correlates in our conceptual knowledge but with a refinement and subtlety which often eludes description. In the arrangement of terms adopted by Drs Head and Holmes the 'focus of attention' sweeps over these neural schemata which "modify the impressions produced by incoming sensory impulses in such a way that the final sensations of position or locality rise into consciousness charged with a relation to something that has happened before. Destruction of such schemata by a lesion of the cortex renders impossible all recognition of position or of the locality of a stimulated spot on the affected part of the body1."

Making the necessary allowances for various conditions under which investigators working separately in the same research field, at the same time, will obviously precipitate their conclusions by means of different terms we seem driven inevitably to the conclusion that the neural 'schemata' of Drs Head and Holmes are identical with Semon's neural 'engrams,' and the psychological abstraction which we call the 'focus of attention' has as its neural correlate that part of the entire engrammic' or 'schematic' system which is endowed at any particular moment with the highest neural potential, and the shifting in the engrammic or schematic system of the point of highest neural potential is the physiological correlate of the movement of the 'focus of attention.'

From our present point of view it seems necessary for us to invert this statement. From the biological standpoint it seems to us that in the course of the ontogenetic life process there has gradually been formed in response to the lifelong aggregation of the effects of sensory impulses a huge and complicated network of physiological paths in which are enmeshed innumerable neurones in various parts of the cerebral organs. These physiological paths were termed by Semon 'engrammic systems' and by Drs Head and Holmes neural 'schemata.' We hold that along this huge system of paths, and partly as the result of fresh sensory stimuli, neural impulses are being continually propagated now in one part of the system now in another; that a neural impulse activating one part of the system tends to activate both adjacent and subjacent parts; that those parts which are in a sufficient state of activity tend to throw above the threshold of consciousness their psychological correlates; and 1 "Sensory Disturbances from Cerebral Lesions," Brain, XXXIV, p. 189.