Association of Ideas

ASSOCIATION OF IDEAS. The chief problem that is in volved in a discussion of the Association of Ideas and the Laws of Association is the problem of recollection and the conditions under which it takes place. Although the expression "associa tion of ideas" was introduced by Locke, the problem under consideration was already dealt with by Plato incidentally, and by Aristotle fairly fully. According to Plato, reminiscence takes place in one of the three ways illustrated by the following ex amples : A lyre or garment belonging to the beloved one puts the lover in mind of him. (2) From beholding a picture of Simmias you may remember him. (3) The sight of a weedy youth may put one in mind of a robust athlete. These examples illustrate what are usually known as association by contiguity, similarity and contrast respectively. Aristotle was much more explicit than Plato in formulating these types or laws of associa tion as principles governing the reinstatement of ideas previously experienced. These three laws of association were accepted and taught by various philosophers (Epicureans, Stoics and Scholas tics) during the centuries that intervened between Aristotle and Hobbes. But no great importance was attached to these prin ciples. Even Locke, who, as already remarked, coined the phrase "association of ideas," laid no particular stress on the laws of association. Hobbes attached considerable importance to as sociation of ideas in mental life, but did not advance the subject of association to any appreciable extent. It was Hume who marked the next considerable advance beyond Aristotle in this matter. He recognized association by contiguity, and by sim ilarity; but instead of association of contrast Hume put forward association by cause and effect—the observation of clouds, for instance, puts one in mind of rain. Strictly speaking this was really tantamount to reducing the laws of association to two, namely, those of contiguity and resemblance, since Hume re garded the idea of causal connection as merely a case of habitual association by contiguity (or contiguous sequence) between an antecedent and a consequent. Hume, however, was interested in the epistemological bearing of the subject of association of ideas rather than in its psychological importance. The so-called As sociationist Psychologists (Hartley, James Mill, etc.) attached exaggerated importance to the association of ideas, which they regarded as occupying in the realm of psychology a place anal ogous to the Law of Gravitation in the realm of Physics (see ASSOCIATIONIST PSYCHOLOGY), as Hume had already vaguely sug gested.

Criticism and Restatement.

(a) Thomas Brown, although he may be described as a member of the school of Associationists or Associationist Psychologists, objected to the term Association, for which he substituted the term suggestion. He pointed out that the term association appears to imply some previous associ ation between the idea suggested and the idea which suggested it. Yet in many cases, and especially in cases of so-called association by similarity, there is no previous association—any object seen for the first time may suggest the idea of something resembling it, but which has never been associated with it in the past. Even if the term association be suitable in cases of association by con tiguity (even this Brown questioned), it was clearly inappropriate for cases of suggestion by similarity.

(b) F. H. Bradley went far beyond Brown in his criticism of the laws of association, although he did not repudiate association in principle, when properly formulated and kept within its proper bounds. The laws of association most widely accepted in the 19th century were those of contiguity and of similarity—the only ones really acknowledged by Hume. The main points in Bradley's criticism come out most clearly if attention is paid to the way in which the laws were formulated by the best known British psychologist of the 19th century, A. Bain. (I) The Law of Con tiguity: "Actions, sensations and states of feeling, occurring to gether or in close connection, tend to grow together, or cohere, in such a way that, when any one of them is afterwards presented to the mind, the others are apt to be brought up in idea" (Senses and Intellect, p. 327). To this Bradley said that sensations, etc., are particular experiences or events that do not recur at all. Con sequently the sensations that are contiguous are not really as sociated in the manner stated by the law of contiguity, whereas the associated ideas were never contiguous before, so that their association could not be due to their contiguity. (2) The Law of Similarity: "Present actions, sensations, thoughts or emotions tend to revive their like among previous impressions or states" (ibid. p. 457) . Bradley's criticism based on the non-recurrence of the same sensation, etc., applies here also. And he adds two further criticisms. Two ideas, etc., can only be recognized as similar if they are both present in the mind ; but what the law seeks to explain is how an idea present in the mind calls up an other that is absent. Moreover, ideas alleged to be associated by similarity are usually more unlike than like one another; so that the recall cannot be due to mere similarity. Bradley then formulated his own law of association to which he gave the name (borrowed from Hamilton) of the Law of Redintegration: "Any part of a single state of mind tends, if reproduced, to reinstate the remainder," or "any element tends to reproduce those ele ments with which it has formed one state of mind." At first sight this law appears to be exposed to Bradley's own criticism about the non-recurrence of states of mind. But the most characteristic feature in Bradley's account (though it can really be traced to Aristotle) is his view that association only marries universals, that is, it associates not particular experiences as such, but the elements that are identical in the individual of the same type. In this way Bradley's Law of Redintegration is really a law of identity and contiguity.

(c) Physiological Basis of Association.

Of the older psy chologists, Aristotle and Hartley were almost the only ones to attempt some sort of explanation of association, namely, by reference to its physiological or neural basis. But until quite re cently the subject received very little further attention—most psychologists being content apparently to subscribe to Hume's remark that the causes of association "are mostly unknown, and must be resolved into original qualities of human nature, which I pretend not to explain" (Inquiry Concerning Human Under standing, Part I. § v.). But with the increasing vogue of Physi ological Psychology, the attempt to find a physiological explana tion of association or suggestion has naturally been renewed. So far as association by contiguity is concerned, a physiological ac count seems to suggest itself readily. W. James pointed to the "laws of habit in the nervous system" as its physiological cause or basis. The idea more or less generally accepted is that if two nerve-centres in the brain are excited simultaneously or in im mediate succession, then some connection is established between them, to the extent at least that the resistance at their synapse (the point of contact between two neurones or nerve-centres) is reduced; and the more frequently this simultaneous or successive activity of two neurones occurs, the more does the path between them tend to become a path of least resistance, so that if one of them is afterwards excited, the excitation spreads to the other. If Bradley's Law of Redintegration be accepted as a true account of association, then the foregoing physiological account of associa tion by contiguity may be made to account for all association— similarity being explained by identity and contiguity. And this is so much in favour of Bradley's views on the matter. But the physiological account involves, of course, the larger problem of the relation between body and mind.

(d) J. Ward and G. F. Stout.

Excessive stress on Associa tion, as may be seen most clearly in the case of so-called associa tionist psychology, has had a tendency to betray psychologists in two ways. It has often misled them to overlook the initial unity of consciousness, to treat experiences as if they were ultimately mere atoms put together and held together by the quasi-gravi tation of association. And in the second place, but largely perhaps in consequence of this soulless treatment of psychology, the whole character of association was conceived in a too impersonal and non-conational manner—as if human experiences were the chance result of the play of impersonal tendencies and in no way in fluenced by the desires and strivings of the individual subject of the experiences. These defects curiously enough are not to be found in Aristotle, who indeed laid emphasis on the activity and purposiveness involved in recollection and association—"we hunt," he says, "for the next in the series, starting our train of thought from what is now present or from something else, and from some thing similar or contrary or contiguous to it" (Memory and Rec ollection, 452a, p. ed. G. R. T. Ross). The faults in question were mainly the consequence of a tendency to fall into line with the mechanistic ideas of modern science. Ward and Stout have done much to stem this current of thought in psychology, and to remedy the faults under consideration. Association, according to Ward, is largely a matter of selective attention, and attention largely depends on interest. Hence to a considerable extent as sociation is a form of integration or a way of constructing a con tinuum expressing and subserving some interest of the experienc ing mind. Stout likewise emphasizes the importance of interest in the formation of associations. Association and recall are deter mined by continuity of interest.

BIBLIOGRAPHY.-G. Croom Robertson, Philosophical Remains (conBibliography.-G. Croom Robertson, Philosophical Remains (con- tains a brief history) ; G. R. T. Ross's edition and translation of Aristotle's De Sensu and De Memoria; works on Psychology by Bain, Sully, Ward, Stout, James, etc.; the works of the other writers men tioned in the text. (A. Wo.)