by
Nigel J.T. Thomas
University of Leeds,
Leeds, U.K.
This penultimate draft © Nigel J.T. Thomas, 1989.
The definitive version was published in The American Journal of Psychology
[1989, vol.102, pp. 395-412].
Galton and subsequent investigators find wide divergences in people's subjective reports of mental imagery. Such individual differences might be taken to explain the peculiarly irreconcilable disputes over the nature and cognitive significance of imagery which have periodically broken out among psychologists and philosophers. However, to so explain these disputes is itself to take a substantive and questionable position on the cognitive role of imagery. This article distinguishes three separable issues over which people can be "for" or "against" mental images. Conflation of these issues can lead to theoretical differences being mistaken for experiential differences, even by theorists themselves. This is applied to the case of John B. Watson, who inaugurated a half-century of neglect of image psychology. Watson originally claimed to have vivid imagery; by 1913 he was denying the existence of images. This strange reversal, which made his behaviorism possible, is explicable as a "creative misconstrual" of Dunlap's "motor" theory of imagination.
For most of the history of human thought in the West, at least since Aristotle
wrote his pioneering psychological treatises De Anima and the
Parva Naturalia (Ross, 1931), the mental image - the quasi-perceptual
experience that can occur in the absence of the relevant perceptual object
- has been almost universally considered to be the primary form of human
mental representation. If this strong tradition is correct, then a proper
understanding of the nature and functional role of the mental image would
clearly be of absolutely fundamental significance for psychology. In fact,
the tradition has been much questioned in the twentieth century. But, even
so, we may still say that debates about the reality, significance, and underlying
mechanisms of imagery are debates about the nature, or the very reality,
of mental processes, and thus about the nature, possibility, and proper direction
of psychological science. Furthermore, I believe that the mechanisms of mental
imagery may have an important bearing on basic issues in epistemology and
philosophy of science (Thomas, 1987), and the psychological study of imagery
probably holds out our best hope of getting a scientific purchase on the
obscure, but culturally very salient concept of "imagination" (cf. Nadaner,
1988).
There have been at least two major debates about imagery among psychologists during this century. The first, at the beginning of the century, was the so-called "imageless thought" controversy in which both Wundt at Leipzig and Titchener at Cornell hotly disputed (on rather different grounds, it should be said) the claims of Külpe and his students at Würzburg to have introspectively discovered cognitive mental contents which were not mental images (of any modality) (Humphrey, 1951; Thomas, 1987). The second was the so-called "analog-propositional" dispute of the 1970s, in which Paivio (e.g., 1971, 1977, 1986), Shepard (e.g., 1975, 1978b, 1981; Shepard & Podgorny, 1978), Kosslyn (e.g., 1980, 1981, 1983; Kosslyn, Pinker, Smith, & Shwartz, 1979; Kosslyn & Pomerantz, 1977), and others argued for imagery as a distinct form of mental representation in its own right, whereas Pylyshyn (e.g., 1973, 1978, 1981) and others (e.g., Anderson & Bower, 1973; Hinton, 1979; Moran, 1974, 1979; Simon, 1972) maintained that all mental representation could best be accounted for within a uniform quasi-linguistic or "propositional" format. It is important to realize that neither of these disputes has ever been properly resolved. The imageless thought controversy came to an end not through the victory of one side or the other, nor with a satisfactory synthesis of both points of view, but, in America at least, with a thoroughgoing denial of the problem. Behaviorism rejected the whole notion of psychology as the science of mental life, and effectively brought discussion of mental imagery to a halt.
The "analog-propositional" dispute may perhaps be said to be still gently simmering, despite the influential efforts of Anderson (1978, 1979) as well as Palmer (1978) to declare it a nonissue. This attempt to call time on the debate is not in fact neutral, as it purports to be, but rather offers to assimilate imagery as a distinct form of representation only within and very much subordinate to a basically computational and "propositional" representational format. Such an assimilation is lent credence by Kosslyn, perhaps the most energetic polemicist on the "analog" side of the debate, who has propounded a theory of imagery (Kosslyn, 1980, 1981, 1983; Kosslyn & Shwartz, 1977) that does indeed set imagery within just such a computational context. Kosslyn's "quasi-pictorial" theory is probably the most fully worked out theory of the mental image available, but this does not mean that it is not without serious and even fatal flaws, quite regardless of one's position on computationalism in general (see Thomas, 1987). Certainly other leading proponents of "analog" representation, such as Paivio (1977, 1986) and Neisser (1976, 1979), would reject both Kosslyn's "pictorial" understanding of imagery and the broader computational theory of mind within which it is set. (Shepard's position [1975, 1978a, 1978b, 1981, 1984a, 1984b) can perhaps be interpreted as pictorialist, but it is certainly not computational.) Anderson and Palmer were probably right to imply that the differences between Kosslyn and Pylyshyn are really rather trivial - if images are really what Kosslyn says they are, they probably do not matter very much - but this emphatically does not apply to the differences between Pylyshyn (and, indeed, Anderson, Palmer, and other computationalists) and the other "analog" theorists. If Kosslyn's theory is set aside, and indeed there are recent signs that Kosslyn himself is moving away from his earlier conventionally computationalist stance (Kosslyn & Hatfield, 1984), we may agree with Block (1983) in seeing the psychology of imagery as a crucial test case for the adequacy of computational approaches to the mind in general.
The first expulsion of imagery from psychology led to the narrow sterilities of behaviorism. Psychology ceased its attempt to tell us about the human mind and human experience, and told us instead only of rats and reflexes. If we allow imagery to be denied or shunted to the sidelines once again, we may be led to a psychology that tells of computing machines rather than of living people (Yuille, 1983). The point is not that either behaviorism or computationalism is necessarily a worthless approach, but that both of them are narrow - they leave out many of the most interesting issues. The main purpose of my article is to attempt to clear up some of the cross - purposes which have repeatedly led debates about imagery into the confusions that have so often frustrated empirically minded psychologists. Even contemporary views about imagery cannot be neatly divided into "analog" or "propositional" positions - other, and probably more fundamental, differences cut across this division. The distinctions I shall make will be applied to, and illustrated by, the circumstances of the first banishment of the mental image from psychological science, at the hands of J. B. Watson - a strange episode which seems to have been crucial to the origins of his behaviorism. I have no wish to defend the introspective methodologies that Watson rejected rather the contrary - but if we find that his rejection of images was ill founded, then how much more would that be the case today, when far more objective means for studying imagery have been developed.
Differences about imagery, then, go much deeper than debates over "analog" versus "propositional" representation. They also go beyond the disciplinary boundaries of psychology. As Dennett (1978, chap. 10) has noted, psychologists and other people interested in the workings of the mind can by and large be divided into two, seemingly passionately opposed, camps. These are the "iconophobes" and the "iconophiles," those who deny and those who assert the reality, or at least the psychological significance, of mental imagery. In the early days of experimental psychology, the iconophiles were well in command. Even the Würzburg "imageless thought" school did not deny that images are real and important, just that they are all-important (Humphrey, 1951; Thomas, 1987, sec. I.B.1). There was also a good deal of interest among clinical psychologists in hallucinations and other "pathological" forms of image (Holt, 1964, p. 255). From about 1920 to 1960, however, as all accounts (e.g., Bugelski, 1984; Haber, 1970; Holt, 1964; Kessel, 1972; Kosslyn, 1980; Morris & Hampson, 1983; Paivio, 1971; A. Richardson, 1969; J.T.E. Richardson, 1980; Sheehan, 1972)1 seem to agree, iconophobia, especially in America, reigned supreme. Even a "fringe" area like parapsychology can be seen to turn from an interest in "apparitions"to the study of card guessing and the like (Beloff, 1977; Holt, 1964). Paivio (1971) regards the 1920s and 1930s as "the most arid period" for the mental image, and Holt (1964) dates the first reawakenings of interest to the mid1950s, but even through the 1940s and 1950s, Psychological Abstracts records no more than five references to imagery (Kessel, 1972). Philosophers (e.g., Ryle, 1949; Sartre, 1940/1966; Schlick, 1925/1974; Wittgenstein, 19602) seem also to have turned away from imagery during the same period. Most philosophers are probably still basically iconophobic (although see Hannay, 1971; Johnson, 1987), but in psychology, since about the mid-1960s, the situation has changed again and imagery has been very much back on the agenda (Thomas, 1987, chap. I.C.). (Parapsychology has also again followed the trend with its interest in so-called "remote viewing" [Targ & Puthoff, 1974] and the like.)
The frustrating stalemate of the "analog-propositional" debate, however, threatens to remove imagery from the agenda once again. Even if imagery is real, one suspects that it would be a relief for many psychologists if they could regard it as unimportant, and there is a readily available rationale for such a position. When Galton issued his pioneering questionnaire on the vividness of mental imagery (Galton, 1880; 1883, pp. 83-144), he was "amazed" to find that although most people reported experiencing images of greater or lesser vividness, a few, concentrated among scientists and other intellectuals, denied experiencing any imagery at all.3 Such non-, or at least very poor,4 imagers seem nowadays to form about 10% to 12% of the population (Abelson, 1979), and it seems natural and plausible to conclude that the argument between iconophobes and iconophiles may rest simply on a personal, idiosyncratic difference in visualizing power (e.g., Abelson, 1979; Marks, 1981). Accepting this explanation of the theoretical differences leads to a sort of compromise view: Mental images must exist (because the iconophiles, we admit, have them), but because many iconophobes manage very well without them, images cannot have any significant cognitive function. At best, imagery must be some sort of mental luxury or cognitive crutch; at worst, it is a distraction from the serious business of proper "scientific" thinking.
Tempting as this conclusion is as a way out of a seemingly irreconcilable dispute, I think we can and must resist it. The real stumbling block to resolution lies not in the personal characteristics of the disputants, but in their conceptual confusions. After all, if people's theoretical attitudes toward imagery are determined by idiosyncratic personal differences, then how does one account for the historical record? Was the American psychological community from sometime before 1920 entirely recruited from the 10% to 12% of nonimagers? If so, why did the vetting procedure break down sometime before 1960? Surely we are rather looking at a process of persuasion and change of theoretical view. Differences in subjective experience may account for some of the passion of the debate, but not, I think, for the substantive disagreement - and even the passion may arise more from an apprehension of the far-reaching implications of the issues involved. I would like to suggest that the reported vividness and quantity of people's imagery may be at least as much determined by the theoretical views they (and their peers) hold on the topic as vice versa. There is, in fact, some empirical warrant for thinking that such self-ratings of imagery are very susceptible to social pressures, to the desire to please whoever is asking (DiVesta, Ingersoll, & Sunshine, 1971). The familiar problems of experimental demand characteristics (Orne, 1962) are thus especially acute here, and imagery research in general does seem to be particularly susceptible to experimenter effects (Intons-Peterson, 1983; Neisser, 1970, 1972; but see Marks, 1983a; J.T.E. Richardson, 1980, p. 12 1). Without direct access to other minds, it is hard to know how to properly conceptualize our subjective experience and against what standards to judge it. Unless one is strongly committed to some theory about mental contents, it is not unreasonable to take some implicit guidance from questioners who seem to know better and care more. "Folk psychology," the beliefs of the general public about the workings of the mind, seems to be iconophilic, on the whole (Denis & Carfantan, 1985; cf. Price, 1953, chap. 8), so most ordinary people normally think they experience plenty of imagery, or perhaps we should say that they are inclined to notice the relevant experiences and to conceptualize them in imagery terms. When iconophobic theories are in the air, however, those who are likely to hear about them and are inclined to be impressed by theoretical argument (i.e., scientists and other intellectuals) may tend to conceptualize any such experiences differently, and perhaps generally to downplay them.5 It may even be that they are simply reluctant to call the relevant experiences "imagery." This sort of interpretation of iconophobia seems to me to be supported by the comments made by one of the "nonimaging" scientists replying to Galton's inquiries:
These questions presuppose assent to some sort of a proposition regarding the "mind's eye", and the "images" which it sees . . . . This points to some initial fallacy . . . . It is only by a figure of speech that I can describe my recollection of a scene as a "mental image" which I can "see" with my "mind's eye." . . . I do not see it . . . any more than a man sees the thousand lines of Sophocles which under due pressure he is ready to repeat. The memory possesses it, and the mind can at will roam over the whole, or study minutely any part. (Quoted by Galton, 1883, pp. 85, 91; Galton's ellipses).
I believe it will help if we distinguish certain issues over which people can be iconophobic or iconophilic. We can hope that this may prove a first step toward resolving their antagonisms. The first possible disagreement occurs over whether people ever really do have the experiences which are commonly designated as "having/seeing a mental image," "visualizing," "picturing," "seeing in the mind's eye," etc.: that is, experiences that resemble (faint) perceptual experiences, but that occur in the absence of the things that seem to be "perceived." Let us call these two possible positions "experiential iconophobia" and "experiential iconophilia." In fact, out-and-out experiential iconophobes seem to be very rare; it takes some gall to deny other people's experiences - even so strict a behaviorist as B.F. Skinner acknowledges "seeing in the absence of the thing seen" (Skinner, 1953, pp. 266, 271; 1976, pp. 91ff.). A more common position is that although imagery experiences may sometimes occur, they have no real cognitive function, they may even hinder clear thinking, or positively deceive us - at the extreme, imagery may be entirely assimilated to hallucination (see Holt, 1964, on this tendency in an iconophobic psychological culture; Sarbin & Juhasz, 1967, set it in a much longer historical context). We may call this "functional iconophobia," and its counterpart - the view that imagery plays an important, regular, perhaps vital role in cognition - will be "functional iconophilia." Two important functional iconophobes of the past were Plato (Republic, 598b; Sophist, 236b-c in Hamilton & Cairns, 1961) and Pascal (Pensées, sec. 81 in Lafuma & Barnwell, 1973). Functional iconophilia begins with Aristotle (De Anima, 427b, 431a; De Memoria et Remeniscentia, 449b in Ross, 1931) and forms the mainstream of the Western tradition (McMahon, 1973; Thomas, 1987).
A quite distinct point of difference arises not over the existence or functional importance of mental imagery but over its underlying mechanism. The problem is that the everyday terminology of "images," "the mind's eye," and "picturing" strongly suggests a particular sort of explanation of the phenomenon; they suggest that something significantly like an ordinary material picture, yet somehow mental, is centrally involved. This is probably the form generally taken by "folk" explanations of imagery, and sometimes by scientific and philosophical accounts too (see Thomas, 1987, chaps. II.A, II.B). Let us call the people who accept this sort of commonsense "pictorial" view "pictorial iconophiles." Those who reject it, who regard the everyday terminology of "images" as implying a radically false theory and who thus deny that there are "real" mental pictures, will be "pictorial iconophobes."
Now it should be apparent that functional and pictorial iconophobia/philia are logically independent of one another. You can quite well regard mental images as being picture-like yet unimportant or bad (this seems to have been Plato's position), or as not involving anything pictorial yet very important - a view which I myself share with several contemporary psychologists (e.g., Neisser, 1976; Paivio, 1977, see secs. 3, 3.2). Nevertheless, there has been a strong tendency to associate the various forms of iconophilia and iconophobia, and this, I think, has caused confusion not only in people's understanding of their opponents' views but sometimes in their self-understanding as well. Galton's correspondent quoted above looks to me very much like someone who has allowed his quite reasonable pictorial iconophobia to push him all the way to an entirely unwarranted experiential iconophobia. This, I think, is one way at least in which theoretical belief can affect introspective reports, and perhaps has affected them in our contemporaries. Pictorial iconophiles, such as Kosslyn (see, e.g., Kosslyn, 1980, 1983; Kosslyn & Shwartz, 1977), are likely to readily recognize their images as such, but Hebb, who seems to be in principle a functional iconophile, has found that his pictorial iconophobia has severely impaired his ability to introspect his own imagery. Hebb gives an amusing account of how he largely lost conscious access to his image life through imagining his own "mind's eye" and recognizing the absurdity of it (Hebb, 1968, p. 476; 1969, p. 57).
Pictorial iconophobes, it must be admitted, face a particular problem. If like Ryle (1949, chap. 8), they reject the everyday "image" terminology as "misleading" and try to avoid it, they run a grave risk of being mistaken for out-and-out experiential iconophobes - nonimagers misled by simple ignorance of other people's interior life. Ryle has certainly suffered this fate (see, e.g., Danto, 1958; Lawrie, 1970), and it has sometimes led, I think, to some of his arguments not being given the weight they deserve. In fact, Ryle seems to have been a moderate functional iconophile (Ryle, 1979, chap. 3). Paivio, by contrast, retains the "image" terminology in expounding his own strong functional iconophilia (e.g., Paivio, 1971, 1975, 1977, 1983, 1986), and gets taken, even by his departmental colleague Pylyshyn (1973), to be a pictorial iconophile also - which he seems not to be (Paivio, 1975, p. 277; 1977, secs. 3, 3.2). By the same token, careful reading shows that it would be a mistake to regard Pylyshyn (1973) as an experiential iconophobe, although he is clearly an iconophobe in our other two senses. Such misunderstandings bedevil the imagery literature.
The iconophobia, which as we have noted above reigned among psychologists for over 40 years of this century, can I believe be shown to be in part the historical result of confusions of the type we have just been discussing. The psychological "paradigm" that particularly sustained American iconophobia was, of course, behaviorism in its various forms, and the standard-bearer of behaviorism, J. B. Watson, regarded the notion of "the 'centrally aroused sensation' or 'image' " as "the most serious obstacle" to the establishment of a thoroughgoing behaviorism, of a "truly scientific" psychology (Watson, 1914, pp. 16ff.; 1913, pp. 24lff., my emphasis). The persistent belief that images exist, and that thinking is carried out in the brain rather than in the muscles is castigated by Watson as an unscientific, "mediaeval" (Watson, 1930, pp. 5-6) hangover from religious belief (Watson, 1913, p. 424; 1914, p. 20). (As noted by Cohen, 1979, chap. 1, Watson had reacted strongly against a strict religious upbringing.) Of course, Watson did not impose his own personal iconophobic feelings, nor even the behaviorist methodology, on the rest of the psychological community. Larger forces were at work (see O'Donnell, 1985). However, Watson is a pivotal figure, and the development of his own opinions about imagery probably provides the ideal case study of the influences upon attitudes to imagery.
Watson's iconophobia was not a thing of half measures. In a lengthy footnote appended to his famous "behaviorist manifesto," Watson asserts:
There is need of questioning more and more the existence of what psychology calls imagery . . . . I should throw out imagery altogether and attempt to show that practically all natural thought goes on in terms of sensori-motor processes in the larynx. (Watson, 1913/1961, p. 816n.)
He was looking forward to establishing a psychology which would "never use the terms consciousness, mental states, mind, content, introspectively verifiable, imagery, and the like" (Watson, 1913/1961, p. 808), and he attempted, with considerable consistency, to carry through this program. People's reports of imagery, he asserted, were "sheer bunk" (Watson, 1928, p. 76). By 1919 he was even attempting to explain common hallucinations, such as the "snakes" frequently "seen" by sufferers from delirium tremens in terms of inappropriate responses to "sinuous shadows on the wall" or other such external or peripheral stimuli (Watson, 1919, pp. 111-112).
Watson appears from all this to have been about as clear an example of the experiential iconophobe as you are likely to find committing himself to print. It would be very tempting to conclude that he was simply one of that 10%-12% of non- or very poor imagers - one who erected this personal quirk into an entire psychological system. However, it seems very strange that such a natural iconophobe should have been moved to enter what was, when he entered it, a systematically iconophilic profession. And consider the following enigmatic remarks, also from the long note to "Psychology as the Behaviorist Views It" (Watson 1913/1961, p. 816n.):
Until a few years ago I had thought that centrally aroused visual sensations [i.e., images] were as clear as those peripherally aroused. I had never accredited myself with any other kind. However, closer examination leads me to deny in my own case the presence of imagery in the Galtonian sense.6 The whole doctrine of the centrally aroused image is, I believe, at present on a very insecure foundation.
Note also that for his first 2 years as professor at Johns Hopkins University (from autumn 1908), Watson taught experimental psychology from the manuals of the arch-iconophile Titchener (Watson, 1936, pp. 276-277), and he acknowledged a great intellectual debt to Titchener at that time (see his letter of 1908 to Titchener, quoted by Larson & Sullivan, 1976, p. 339). Moreover, Knight Dunlap, his junior colleague at Johns Hopkins, testifies that Watson still accepted "the old doctrine of 'images' " in his early days there. When Dunlap expressed skepticism about images, Watson insisted on their reality, saying that he himself made very effectual use of visual imagery when designing his apparatus. It looks as if Watson originally thought that he had good mental images - even "as clear as those peripherally aroused" (Watson, 1913/196 1, p. 816n.). It was only as his theoretical views developed that he decided that he (and everybody else) had none at all.7
Of course, Watson did not bring about the behaviorist revolution singlehandedly in 1913. Furthermore, he was by no means the first of his contemporaries to question the prevailing iconophilia. Külpe and his students at Würzburg had raised doubts about the functional significance of imagery in thinking (see Humphrey, 1951), and the ensuing "imageless thought controversy" seems to have played a significant part in throwing the whole introspective psychological method into question. (Watson, in his 1913 "manifesto;' was able to play very effectively on how such disagreements could degenerate into quite irresolvable wrangles about the introspective competence of different "observers" [Watson, 1913/1961, p. 804 & n.]). But there also seem to have been rising doubts about the whole conception of mind which sustained the notion of mental "pictures-in-the-head." In 1914, Lovejoy8 wrote that over the previous ten years9 it had "become the fashion with not a few philosophers" to question the conception of "consciousness" that figured so largely in the introspective psychology. Furthermore:
The fate of "consciousness" has been shared by several other ancient notions which once made up its retinue. The existence of sensations, of images, of ideas, of mental states, of "subjective appearances", and the possibility of "introspection", have all been denied by recent philosophical and psychological iconoclasts. (Lovejoy, 1914, p. 42)
Lovejoy was concerned to resist this "iconoclasm"; Watson embraced it, and by cutting it free from the philosophical subtleties with which he had long since grown impatient,10 he probably did more to ensure its (temporary) triumph than anybody else.
Watson tells us (1924, p. viii; 1936, p. 276) that he had been toying since 1903 or 1904 with the idea of applying to humans the sorts of methods, the observation of behavior, that he had already applied so successfully to the study of learning in rats (Watson, 1903), and which he soon went on to apply to other animals (see Cohen, 1979, chaps. 2-3; Watson, 1910). However, there seems to be no good reason to think that he was part of the "iconoclastic" movement when he arrived at Hopkins in 1908. Rather, as we have seen, the reverse seems to have been the case. Dunlap, however, who was already established at Hopkins when Watson arrived, claims to have been an open "iconoclast," criticizing "the conventional doctrine of 'mental images'" in private, since about 1907 (Dunlap, 1914, p. 25). In 1912, the year at whose end Watson "came out" as a behaviorist, Dunlap had published a paper criticizing the then current conceptions of introspection (Dunlap, 1912a) and a textbook (Dunlap, 1912b) in which he expressed some of his "iconoclastic" views on imagery, albeit in an appropriately toned-down form (he did not express himself freely on the subject in print until 1914). Although he never embraced behaviorism (Dunlap, 1932, p. 46), Dunlap was nevertheless sympathetic to Watson's early tentative ideas about applying the methods of the "behavior men," the animal psychologists, to humans, and he seems to have been the first person to give Watson any real encouragement on this (Cohen, 1979, p. 64). His views on imagery appear to have been crucial. By 1910, and perhaps before, the only real factor preventing Watson from conceiving of the study of behavior as embracing the whole of psychology seems to have been "the problem of the higher thought processes" (Burnham, 1968, p. 150). Thought was supposed to be carried on primarily in imagery, and imagery was not behavior (see Watson, 1913, p. 421; 1914, pp. 16ff.). Watson acknowledges that it was Dunlap's "iconoclastic" arguments which eventually led him to the view that images could simply be dropped altogether and replaced by "implicit" muscular responses (Watson, 1924, p. ix). Dunlap's own account of what happened after Watson arrived at Hopkins, which Watson unreservedly endorses as "true" (Watson, 1936, p. 277), is as follows:
I had already discarded the old doctrine of "images." Watson, however, still accepted it. He, he said, used visual imagery very effectually in designing his apparatus. Watson had not at that time developed his behaviorism and his thinking was, to a large extent, along conventional lines. He was violently interested in animal behavior, and was looking for some simplifications of attitude which would align that work with human psychology. Hence, he was interested in the iconoclastic activity I was developing, and was influenced by my views, but carried them out to extremes. I rejected images as psychic objects, and denounced introspection as held by the orthodox psychologists. Watson carried this further, to the excluding from his psychology of everything to which the word "introspection" could be applied, and excluded imagination along with images. (Dunlap, 1932, p. 45)
What does Dunlap mean here by implying that he would not have had Watson drop "imagination" but only "images"? I believe that this indicates that Dunlap's "iconoclasm" amounted only to what we have called pictorial iconophobia, whereas Watson, failing to make the distinction, carried this right through into functional, and even experiential, iconophobia. Dunlap certainly stoutly denied the existence of mental images in the sense of mental pictures, faint reinstatements of former visual impressions (or, indeed, impressions of other sense modes): "I contend that the image, as a copy or reproduction of sensation of variable mode does not exist" (Dunlap, 1914, p. 28). He offered both theoretical argument and introspective evidence in support of this (Dunlap, 1912b, pp. 156-160; 1914, pp. 38-39). However, he still held that something, something mental, was needed to fill the functional role that images played in the conventional psychology of thinking - something to form the "carriages" in the "train of association" - and he provided not only an account of what these mental contents were, but also of why people so widely took them to be reproductions of former sensation, images in the "pictorial" sense. He suggested:
There is indeed a present content essentially connected with imagination or thought; but this present content is in each case a muscle sensation, or a complex of muscle sensations. We are therefore, in investigating images, dealing not with copies, or pale ghosts, of former sensations but with actual present sensations. (Dunlap, 1914, p. 28)
These muscle sensations were not at all to be confused with the impalpable "imageless thoughts" of Würzburg (Dunlap, 1914, p. 37); rather: "This sensation is the true image" (Dunlap, 1914, p. 34, emphasis in original). The muscle sensations were supposed to be caused by small, outwardly imperceptible, muscular reflex responses to particular stimuli, and Dunlap provides an account of how regular successions of stimuli can cause the relevant muscular responses to become entrained so that perception of one stimulus can make us "imagine," by producing the relevant muscular responses, those which go with it: "In concrete illustrative terms: the visual presentation of an apple no longer arouses visual perception merely, but arouses also the perception of the gustatory, olfactory, tactual, and possibly thermal qualities of the apple" (Dunlap, 1914, p. 33). Dunlap claims that his own introspections reveal only muscle sensations in such cases, and not the "absent" perceptual qualities (1914, pp. 35, 38). However, these muscular response pattern "images" serve us as "ideas," as representations of the stimulus objects which in the first place produce them (just as pictorial images do in more traditional forms of associationism). In the thinking process, they can therefore represent some absent object of thought. Dunlap is thus able to explain away the fact that most people believe that they experience "mental images" as reinstatements of former perceptions (rather than of former responses):
[T]his form of present content (muscular activity) is that which is actually observed by those who report "mental images". These observers correctly notice that there is a present content in addition to the "absent" or ultimate object of thought, but they mistakenly confuse the quality of the ultimate object with the quality of the present content. (Dunlap, 1914, p. 39)
Furthermore:
The muscles of the organs of the special senses are in many cases concerned in "imagery", and there is a strong tendency in these cases to refer the image to the mode of the special sense concerned. If the muscles of the eye are involved in the production of an image, there is a tendency to classify the "image" as visual, and so on. (Dunlap, 1914, p.36)11
The real trouble, however, is that other introspectors hold the wrong theory!
It is of course extremely difficult to separate completely in introspection the direct content from the idea. The difficulty is especially great if we do not understand what the direct content really is. Hence we need not be puzzled by the fact that the direct content has been described in conventional psychology as possessing the modality, and possibly other characteristics, of the idea. (Dunlap, 1914, p. 36).
It seems that our introspections can only be relied upon to reveal to us the true nature of the image, or other thought content, if we already have a "correct" understanding of this. Here, as in the "imageless thought" debate, we see again that futility of introspectionist argument which led Watson to rail so successfully against the whole approach.
But, although he lays much less stress on laryngeal, language responses, Dunlap's theory of "image" association bears a considerable resemblance to Watson's theory of thought (Watson, 1919, chap. IX; 1930, chaps. X-XI). Under both theories, thinking could be described as a matter of a succession of implicit muscular responses serially conditioned to one another. However, Watson's behaviorism disavowed all interest in conscious contents, utterly rejected introspection, and seemingly denied the existence of imagery altogether. Dunlap, it seems to me, was rather trying to explain these things in a new (if perhaps ultimately unconvincing) way. However, lacking a firm grasp of the distinctions between the various forms of iconophobia, Dunlap occasionally talks as if he is denying the reality of imagery altogether. Watson, it seems to me, probably took him to be doing just that.
We may perhaps take Watson as not only the leader but also the exemplar of the switch to behaviorism in American psychology. "Iconoclastic" objections, such as Dunlap's, to "pictorial" accounts of imagery (and to other notions from the "retinue" of "consciousness") were, I suspect, widely seen to be essentially cogent. However, the available alternative accounts of these phenomena, even if clearly distinguished from outright rejections of them, probably seemed far less persuasive. After all, like all theories, they would inevitably have their defects and limitations. It should be no surprise that the hope of cutting through these difficulties once and for all, by entirely excluding the "inner" from psychology, came to hold great attractions. In the case of imagery, the difficulties would have been compounded (as they are today) by the general failure to recognize the distinction between the various forms of iconophobia. Arguments against "pictorial" mechanisms could easily be misunderstood as powerful, if somewhat elusive, arguments against the reality of the experience itself. Anyway, the received wisdom among psychologists became that "mental imagery," especially, was a bad concept and best avoided.
However, denial of the experiential reality of imagery has always looked implausible, whatever the pressure of theory, and in recent times its functional significance has been demonstrated in numerous ways (see, e.g., Kosslyn, 1980, 1983; Paivio, 1971, 1983, 1986; Shepard, 1978a, 1978b; Shepard & Cooper, 1982; for critical appraisal of this and other related work see Thomas, 1987, chap. I.C). None of this, however, demands that imagery should be regarded as pictorial (Thomas, 1987, pt. II), a view which is, indeed, probably untenable, despite the efforts of Kosslyn (e.g., 1980, 1981, 1983; Kosslyn et al., 1979; Kosslyn & Shwartz, 1977) to promote it. But perhaps Kosslyn would never have embarked on his pictorialist project if he had realized that functional iconophilia does not entail pictorial iconophilia. It may even be inconsistent with it. Realizing this, we may hope that psychology will not recapitulate the confusions of its past.
Notes
1. In fact the rest of these accounts probably rely heavily, though not exclusively, on Holt's.
2. Written in 1933-1935, and much circulated before publication.
3. A study by Roe (1951) does not really bear out Galton's view that scientists are largely non- or very poor, imagers, but David Marks has recently found a negative correlation between self-ratings of imagery vividness and examination grades among chemistry students (personal communication, February 10, 1984), suggesting that perhaps less vivid imagers do, indeed, make better scientists.
4. The specific examples quoted by Galton (1880, pp. 305-306; 1883, pp. 91-92) mostly, in fact, admit to occasional vague and dim images when awake or to hypnagogic or dream images. Marks, a leading contemporary researcher on individual differences of self-reported imagery vividness, seems inclined to deny the existence of absolute "nonimagers" (Marks, 1972, p. 107). Sommer's (1978, chap. 7) case study of an extreme "nonimager" seems to bear out this conclusion; after some prodding, the subject admitted to imagery in his dreams. There do seem to be rare cases of people totally losing mental imagery, but without serious impairment of perceptual or cognitive functioning, after brain damage. However, rather than reflecting a loss of the underlying imagery function, this may well be best explained in terms of a broken connection between the brain areas where images are formed and the language centers that would have to be involved in reporting it-even to oneself (Basso, Bisiach, & Luzzatti, 1980). Marks (1986, p. 237) speculates that some of the healthy people who report very little or very poor imagery may, possibly, be suffering from a subclinical and partial "disconnection" of this type.
5. I do not say that such "theoretical" and "social" influences are the only reason for the individual differences in people's self-ratings of imagery vividness, etc. Although results in the field seem to be somewhat conflicting and confusing (see Sheehan, Ashton, & White, 1983, for review) and Kaufmann (1981) has raised general questions about the validity of imagery vividness questionnaires, Marks (1983a) has vigorously defended their use. Marks (1983b) reviews a number of studies that find interesting correlations between scores on the Vividness of Visual Imagery Questionnaire (VVIQ), which he devised (Marks, 1973), and a wide range of other traits [e.g., ability to remember pictures, characteristics of eye movement patterns in vision, susceptibility to the "imagery McCollough effect" (Finke & Schmidt, 1978), hypnotizability]. More recently a surprising negative correlation between VVIQ scores and accuracy of color memory has been found (Huer, Fischman, & Reisberg, 1986; Reisberg, Culver, Huer, & Fischman, 1986). In the light of such evidence, it would seem hard to deny that differences in reported imagery vividness do have some real basis (although this may still be overlain by the sorts of factors I have mentioned). I do not even assert that the influence of theories fully accounts for the seeming difference between intellectuals and others. Galton himself suggested that there may be a certain antagonism between being a habitual vivid imager and "habits of highly-generalized and abstract thought:' although he still thought that the best minds should be able to use imagery or abstract" thinking at will (Galton, 1883, p. 88).
6. Precisely what "the Galtonian sense" of imagery is, and what "other kind" Watson might have accredited himself with, I am not at all sure. Galton does not provide any definition or theory of imagery. Except for a brief and very indirect hint at a "pictorial" theory of the image mechanism (Galton, 1883, p. 111), he seems to be making no claims about the mechanisms of imagery, and seems to think he is using the term in an atheoretical, "ordinary language" way. If you do not have imagery "in the Galtonian sense,' I do not see that you have it at all.
7. Watson's lifelong fear of the dark (Cohen, 1979) also suggests that he always had a vivid imagination, and it hints at further (psychodynamic) reasons why he might nevertheless have been inclined to deny the faculty (I owe this point to Graham D. Richards). The fact remains, however, that he did not deny it before he had produced a theoretical rationale for doing so.
8. Professor of education at Hopkins at the time, and friendly with Watson (Cohen, 1979, p. 54).
9. He seems to be dating events from the publication of William James's article, "Does 'Consciousness' Exist?" (1904).
10. Watson began as a graduate student at the University of Chicago "more interested in philosophy than psychology," but "it wouldn't take hold" (Watson, 1936, p. 273). Only from Hume (a proto-positivist, but also a good iconophile!) and to a lesser extent Locke and Hartley, does he admit to getting anything at all (Watson, 1936, p. 274). In 1913 he urged that the education of behaviorist psychologists would be able to (and should) altogether avoid consideration of such "time-honored relics of philosophical speculation" as the mind-body problem (Watson, 1913/1961, pp. 807-808).
11. Watson (1914, p. 18n.) borrows this argument, with acknowledgment to Dunlap, to explain why people might come to believe in "the fiction of visual imagery."
References
Abelson, R.P. (1979). Imagining the purpose of imagery. Behavioral and Brain Sciences, 2, 548-549.
Anderson, J.R. (1978). Arguments concerning representations for mental imagery. Psychological Review, 85, 249-277.
Anderson, J.R. (1979). Further arguments concerning representations for mental imagery: A response to Hayes-Roth and Pylyshyn. Psychological Review, 86, 395-406.
Anderson, J.R., & Bower, G.H. (1973). Human associative memory. New York: Wiley.
Basso, A., Bisiach, E., & Luzzatti, C. (1980). Loss of mental imagery: A case study. Neuropsychologia, 18, 435-442.
Beloff, J.. (1977). Historical overview. In B. B. Wolman (Ed.), Handbook of Parapsychology, (pp. 3-24). New York: Van Nostrand & Reinhold.
Block, N. (1983). Mental pictures and cognitive science. Philosophical Review, 92, 499-539.
Bugelski, B.R. (1984). Imagery. In R.J. Corsini (Ed.), Encyclopædia of psychology (Vol. 2, pp. 185-187). New York: Wiley.
Burnham, J.C. (1968). On the origins of behaviorism. Journal of the History of the Behavioral Sciences, 4, 143-151.
Cohen, D. (1979).J. B. Watson - the founder of behaviorism: A biography. London: Routledge & Kegan Paul.
Danto, A.C. (1958). Concerning mental pictures. Journal of Philosophy, 55, 12-20.
Denis, M., & Carfantan, M. (1985). People's knowledge about images. Cognition, 20, 49-60.
Dennett, D.C. (1978). Brainstorms. Hassocks, Sussex, U.K.: Harvester.
DiVesta, E.J., Ingersoll, G., & Sunshine, P. (1971). A factor analysis of imagery tests. Journal of Verbal Learning and Verbal Behavior, 10, 471-479.
Dunlap, K. (1912a). The case against introspection. Psychological Review, 19, 404-413.
Dunlap, K. (1912b). A system of psychology. New York: Scribners.
Dunlap, K. (1914, March). Images and ideas. Johns Hopkins University Circular, 3, 25-41.
Dunlap, K. (1932). Knight Dunlap. In C. Murchison (Ed.), A history of psychology in autobiography (Vol. 2, pp. 35-61). Worcester, MA: Clark University Press.
Finke, R.A., & Schmidt, M.J. (1978). The quantitative measure of pattern representation in images using orientation-specific color aftereffects. Perception & Psychophysics, 23, 515-520.
Galton, F. (1880). Statistics of mental imagery. Mind, 5, 301-318.
Galton, F. (1883). Inquiries into human faculty and its development. London: Macmillan.
Haber, R.N. (1970). Imagine! They are finally talking about images again. Contemporary Psychology, 15, 556-559.
Hamilton, E., & Cairns, H. (Eds.). (1961). The collected dialogues of Plato (including the letters). New York: Pantheon.
Hannay, A. (1971). Mental images - a defence. London: Allen & Unwin.
Hebb, D.O. (1968). Concerning imagery. Psychological Review, 75, 466-477.
Hebb, D.O. (1969). The mind's eye. Psychology Today, 2, 54-57, 67-68.
Hinton, G. (1979). Some demonstrations of the effects of structural descriptions in mental imagery. Cognitive Science, 3, 231-250.
Holt, R.R. (1964). Imagery: The return of the ostracized. American Psychologist, 19, 254-266.
Huer, E, Fischman, D., & Reisberg, D. (1986). Why does vivid imagery hurt color memory? Canadian Journal of Psychology, 40, 161-175.
Humphrey, G. (1951). Thinking. London: Methuen.
Intons-Peterson, M.J. (1983). Imagery paradigms: How vulnerable are they to experimenter's expectations? Journal of experimental Psychology: Human Perception and Performance, 9, 394-412.
James, W. (1904). Does "consciousness" exist? Journal of Philosophy, Psychology and Scientific Methods, 1. [Reprinted in James, W. (1912). Essays in radical empiricism (pp. 1-38). New York: Longmans]
Johnson, M. (1987). The body in the mind: The bodily basis of meaning, imagination, and reason. Chicago: University of Chicago Press.
Kaufmann, G. (1981). What is wrong with imagery questionnaires? Scandinavian journal of psychology, 22, 59-64.
Kessel, F.S. (1972). Imagery: A dimension of mind rediscovered. British Journal of Psychology, 63, 149-162.
Kosslyn, S.M. (1980). Image and mind. Cambridge, MA: Harvard University Press.
Kosslyn, S.M. (1981). The medium and the message in mental imagery: A theory. Psychological Review, 88, 46-66.
Kosslyn, S.M. (1983). Ghosts in the mind's machine: Creating and using images in the brain. New York: Norton.
Kosslyn, S.M., & Hatfield, G. (1984). Representation without symbol systems. Social Research, 51, 1019-1045.
Kosslyn, S.M., Pinker, S., Smith, G.E., & Shwartz, S.P. (1979). On the demystification of mental imagery. Behavioral and Brain Sciences, 2, 535-548.
Kosslyn, S.M., & Pomerantz, J.R. (1977). Imagery, propositions and the form of internal representations. Cognitive Psychology, 9, 52-76.
Kosslyn, S.M., & Shwartz, S.P. (1977). A simulation of visual imagery. Cognitive Science, 1, 265-295.
Lafuma, L., & Barnwell, H.T. (Eds.), (1973). Blaise Pascal: Pensées (J. Warrington, Trans.). London: Dent. (Original work published 1670)
Larson, C.A., & Sullivan, J.J. (1965). Watson's relation to Titchener. Journal of the History of the Behavioral Sciences, 1, 338-354.
Lawrie, R. (1970). The existence of mental images. Philosophical Quarterly, 20, 253-257.
Lovejoy, A.O. (1914, March). On the existence of ideas. Johns Hopkins University Circular, 3, 42-99.
Marks, D.F. (1972). Individual differences in the vividness of visual imagery and their effects. In P. W. Sheehan (Ed.), The function and nature of imagery (pp. 83-108). New York: Academic Press.
Marks, D.F. (1973). Visual imagery differences in the recall of pictures. British Journal of Psychology, 64, 17-24.
Marks, D.F. (1981). Imagery, knowledge and the individual. Journal of Mental Imagery, 5, 47-50.
Marks, D.F. (1983a). In defense of imagery questionnaires. Scandinavian Journal of Psychology, 24, 243-246.
Marks, D.F. (1983b). Mental imagery and consciousness: A theoretical review. In A.A. Sheikh (Ed.), Imagery: Current theory, research, and application (pp. 96-130). New York: Wiley.
Marks, D.F. (1986). The neuropsychology of imagery. In D.F. Marks (Ed.), Theories of image formation (pp. 225-241). New York: Brandon House.
McMahon, C.E.(1973). Images as motives and motivators: A historical perspective. American Journal of Psychology, 86, 465-490.
Moran, T.P. (1974). The symbolic imagery hypothesis: An empirical investigation via a production system simulation of human behavior in a visualization task. Dissertation Abstracts International, 35, 55 1 B-552B. (University Microfilms No. 74-14, 657)
Moran, T.P. (1979). The imprecision of mental imagery. Behavioral and Brain Sciences, 2, 560.
Morris, P.E., & Hampson, P.H. (1983). Imagery and consciousness. London: Academic Press.
Nadaner, D. (1988). Visual imagery, imagination, and education. In K. Egan & D. Nadaner (Eds.), Imagination and education (pp. 198-207). Milton Keynes, U.K.: Open University Press.
Neisser, U. (1970). Visual imagery as process and as experience. In J.S. Antrobus (Ed.), Cognition and affect (pp. 159-178). Boston, MA: Little, Brown.
Neisser, U. (1972). Changing conceptions of imagery. In P.W. Sheehan (Ed.), The function and nature of imagery (pp. 234-251). New York: Academic Press.
Neisser, U. (1976). Cognition and reality. San Francisco: Freeman.
Neisser, U. (1979). Images, models, and human nature. Behavioral and Brain Sciences, 2, 561.
O'Donnell, J.M. (1985). The origins of behaviorism: American psychology, 1870-1920. New York: New York University Press.
Orne, M.T. (1962). On the social psychology of the psychology experiment: With particular reference to demand characteristics and their implications. American Psychologist, 17, 776-783.
Paivio, A. (1971). Imagery and verbal processes. New York: Holt, Rinehart & Winston.
Paivio, A. (1975). Imagery and synchronic thinking. Canadian Psychological Review, 16, 147-163.
Paivio, A. (1977). Images, propositions and knowledge. In J.M. Nicholas (Ed.), Images, perception and knowledge [Western Ontario Studies in the Philosophy of Science, 8] (pp. 47-71). Dordrecht, Netherlands: Reidel.
Paivio, A. (1983). The empirical case for dual coding. In J.C. Yuille (Ed.), Imagery, memory and cognition: Essays in honor of Allan Paivio (pp. 307-332). Hillsdale, NJ: Erlbaum.
Paivio, A. (1986). Mental representations: A dual coding approach. New York: Oxford University Press.
Palmer, S.E. (1978). Fundamental aspects of cognitive representation. In E. Rosch & B.B. Lloyd (Eds.), Cognition and categorization (pp. 259-303). Hillsdale, NJ: Erlbaum.
Price, H.H. (1953). Thinking and experience. London: Hutchinson.
Pylyshyn, Z.W. (1973). What the mind's eye tells the mind's brain: A critique of mental imagery. Psychological Bulletin, 80, 1-25.
Pylyshyn, Z.W. (1978). Imagery and artificial intelligence. Minnesota Studies in the Philosophy of Science, 9, 19-55.
Pylyshyn, Z.W. (1981). The imagery debate: Analogue media versus tacit knowledge. Psychological Review, 88, 16-45.
Reisberg, D., Culver, L.C., Huer, F., & Fischman, D. (1986). Visual memory: When imagery vividness makes a difference. Journal of Mental Imagery, 10, 51-74.
Richardson, A. (1969). Mental imagery. London: Routledge & Kegan Paul.
Richardson, J.T.E. (1980). Mental imagery and human memory. London: Macmillan.
Roe, A. (1951). A study of imagery in research scientists. Journal of Personality, 19, 459-470.
Ross, W D. (Ed.). (1931). The works of Aristotle translated into English (Vol. 3). Oxford: Oxford University Press.
Ryle, G. (1949). The concept of mind. London: Hutchinson.
Ryle, G. (1979). On thinking. Oxford: Blackwell.
Sarbin, T.R., & Juhasz, J.B. (1967). The historical background of the concept of hallucination. Journal of the History of the Behavioral Sciences, 3, 339-358.
Sartre, J.-P. (1966). The psychology of imagination. New York: Philosophical Library. (Original French edition, 1940)
Schlick, M. (1974). General theory of knowledge. Vienna: Springer-Verlag. (Translated from 2nd German edition, 1925; 1st German edition, 1918)
Sheehan, P.W. (Ed.). (1972). The function and nature of imagery. New York: Academic Press.
Sheehan, P.W., Ashton, R., & White, K. (1983). Assessment of mental imagery. In A.A. Sheikh (Ed.), Imagery: Current theory, research, and application (pp. 189-221). New York: Wiley.
Shepard, R.N. (1975). Form, formation and transformation of internal representations. In R.L. Solso (Ed.), Information processing and cognition: The Loyola symposium (pp. 87-122). Hillsdale, NJ: Erlbaum.
Shepard, R.N. (1978a). Externalization of mental images and the act of creation. In B.S. Randhawa & B.F. Coffman (Eds.), Visual learning, thinking and communication (pp. 133-189). London: Academic Press.
Shepard, R.N. (1978b). The mental image. American Psychologist, 33, 125-137.
Shepard, R.N. (1981). Psychophysical complementarity. In M. Kubovy & J.R. Pomerantz (Eds.), Perceptual organization (pp. 279-341). Hillsdale, NJ: Erlbaum.
Shepard, R.N. (1984a). Ecological restraints on internal representation. Psychological Review, 91, 417-447.
Shepard, R.N. (1984b, April). How the mind reflects the world. Paper presented at the University of California Cognitive Science Seminar, Berkeley, CA.
Shepard, R.N., & Cooper, L A. (1982). Mental images and their transformations. Cambridge, MA: MIT Press.
Shepard, R.N., & Podgorny, P. (1978). Cognitive processes that resemble perceptual processes. In W. K. Estes (Ed.), Handbook of learning and cognitive processes (pp. 189-237). Hillsdale, NJ: Erlbaum.
Simon, H.A. (1972). What is visual imagery: An information processing interpretation. In L. W. Gregg (Ed.), Cognition in learning and memory (pp. 183-204). New York: Wiley.
Skinner, B.F. (1953). Science and human behavior New York: Free Press.
Skinner, B.F. (1976). About Behaviorism. New York: Vintage. (Original edition, Knopf, 1974)
Sommer, R. (1978). The mind's eye. New York: Delacorte Press.
Targ, R., & Puthoff, H. (1974). Information transfer under conditions of sensory shielding. Nature, 251, 602-607.
Thomas, N.J.T. (1987). The psychology of perception, imagination and mental representation,and twentieth century philosophies of science. Unpublished doctoral dissertation, University of Leeds, Leeds, U.K. (ASLIB Index to Theses, 37-iii, 4561)
Watson, J.B. (1903). Animal education. Chicago: University of Chicago Press.
Watson, J.B. (1910). The new science of animal behavior. Harper's Monthly Magazine, 12O,346-353.
Watson, J.B. (1913). Image and affection in behavior Journal of Philosophy, Psychology and Scientific Methods, 10, 421-428.
Watson, J.B. (1914). Behavior: An introduction to comparative psychology. New York: Holt.
Watson, J.B. (1919). Psychology from the standpoint of a behaviorist (1st ed.). Philadelphia, PA: Lippincott.
Watson, J.B. (1924). Psychology from the standpoint of a behaviorist (2nd ed.). Philadelphia, PA: Lippincott.
Watson, J.B. (1928). The ways of behaviorism. New York: Harper.
Watson, J.B. (1930). Behaviorism (rev. 2nd ed.). Chicago: University of Chicago Press.
Watson, J.B. (1936). John Broadus Watson. In C. Murchison (Ed.), A history of psychology in autobiography (Vol. 3, pp. 271-281). Worcester, MA: Clark University Press.
Watson, J.B. (1961). Psychology as the behaviorist views it. In T. Shipley (Ed.), Classics in psychology (pp. 798-821). New York: Philosophical Library. (Reprinted from Psychological Review, 1913, 20, 158-177)
Wittgenstein, L. (1960). The blue and brown books (2nd ed.). New York: Harper & Row.
Yuille, J.C. (1983). The crisis in theories of mental imagery. In J.C. Yuille (Ed.), Imagery, memory and cognition: Essays in honor of Allan Paivio (pp. 263-284). Hillsdale, NJ: Erlbaum.