Goals and Methods in Classification

Classification of Psychopathology: Goals and Methods in an Empirical Approach
G. Scott Acton University of California, San Francisco

Author Note

Note: This paper is under revision. This version will be significantly different from the forthcoming published version.

This article aims to elucidate the philosophical and methodological assumptions underlying the fourth edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV, American Psychiatric Association, 1994). Operationism, which appears to underwrite the DSM's categorical approach, is criticized as a form of infallibilism. A fallibilist alternative in the conceptualization of diagnosis is provided in the form of fuzzy sets. The present criticisms of the DSM clear the way for a new understanding of the classification enterprise, one that embraces explanatory realism as a goal, fallibilism as a method, and dimensions as a (perhaps yet-to-be-realized) empirical finding.
Keywords: Abnormal psychology, operationism, prototypes, dimensions, traits

The fourth edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV; American Psychiatric Association, 1994) virtually holds a monopoly on the classification of psychopathology. Nearly every textbook on abnormal psychology uses DSM classifications so that professors will choose the book as a way to prepare their students for successful careers in the helping professions; researchers who study psychopathology know that they probably will not receive grant funding unless they define their terms according to the DSM nomenclature. Because of the extraordinary prominence of the DSM in clinical psychology, psychiatry, and related professions, careful attention to the DSM's philosophical and methodological underpinnings seems warranted. It is hoped that this article will peak readers' skepticism concerning some of the DSM's untested assumptions and will point the way to some more generative alternatives.
For persons interested in the classification of psychopathology, this article will draw some crucial and sometimes counterintuitive distinctions. An attempt is made to explore how issues in the philosophy of science and in the theory of measurement bear on psychodiagnostic classification. The present criticisms of the DSM apply to any approach that takes operationism as the basis for a categorical approach. The alternative to operationism is a realist philosophy of science in which it is possible to talk about truth--or truth-likeness--and progress. This article aims to illuminate the philosophical and methodological assumptions underlying the DSM, clearing the way for a new understanding of the classification enterprise, one that embraces explanatory realism as a goal, fallibilism as a method, and dimensions as a (perhaps yet-to-be-realized) empirical finding.

Issues in the Philosophy of Science

Certainty
Most modern philosophers, whether empiricists or rationalists, agree with most radical skeptics that the problem of knowledge is the problem of certain knowledge or justified belief. Dogmatists insist on what skeptics deny, that certain knowledge is possible; both dogmatists and skeptics agree with the following principle of rationality: A belief is reasonable if and only if the belief is certain or justified (Musgrave, 1993). This principle of rationality is presupposed in the DSM approach to classification, as can be seen from the following canonical argument:

(1) A clinician's belief that person X has disorder Y is reasonable if and only if the belief is certain or justified.
(2) The belief is certain or justified if and only if X meets the necessary DSM diagnostic criteria for Y.
_________________________________________________________________________
(3) Therefore, a clinician's belief that person X has disorder Y is certain or justified if and only if X meets the necessary DSM diagnostic criteria for Y.

Logic is the wellspring of certainty--the only beliefs that are certain are true by virtue of logical relations and definitions. For example, tautologies (trivially true statements of the form, If P, then P) can rightly be considered certainties. Skepticism concerning non-trivial certainties, however, is well warranted and is incorporated directly into the methodological approach known variously as critical rationalism, critical realism, or fallibilist realism.

Realisms
There are many versions of realism. Most versions of realism embrace what has been called the copy theory of truth, that there comes to be an agreement between concept and referent that reflects the actual state of the world; truth is defined as accuracy in identifying the items to which a term correctly refers. A similar definition of realism is given by Putnam (1982):

the world consists of some fixed totality of mind-independent objects. There is exactly one true and complete description of 'the way the world is'. Truth involves some sort of correspondence between words or thought-signs and external things and sets of things. (p. 49)

Critical realists disagree with most of the assertions of the above form of realism and yet are still realists. The remainder of this section details how critical realists disagree with the above assertions of typical realists.
Three worlds. That the material world consists of some fixed totality of mind-independent objects seems unproblematic. Some philosophers have gone further, postulating the existence of three "worlds" (Eccles, 1974; Lakatos, 1970; Musgrave, 1974; Popper, 1968a, 1968b; Toulmin, 1967). Lakatos (1970) described the three worlds this way: "The first world is the material world, the second is the world of consciousness, the third is the world of propositions, truth, standards: the world of objective knowledge" (p. 180). [What work does this paragraph do? The issue is that idealizations are possible. Facts can represent ideas, even though the facts are in World 1 and the ideas are in either World 2 or World 3. If ideas are objective knowledge (World 3), then it needs to be possible for the ideas to be about the real world (World 1)--otherwise, we have pure idealism. The way that ideas are about the real world is that the concepts within propositions, or perhaps the propositions themselves, correspond with the real world. But the correspondence is not perfect, because the ideas are incomplete--at least within science, we do not claim to have complete knowledge, and any such claim would warrant complete skepticism. There is a correspondence between our concepts and the material or psychological world, but this correspondence is incomplete, so we have only probabilistic knowledge. Here is where item response theory (IRT) comes in. In IRT, the item is the material world, the person is the world of consciousness, and the probability that the person will endorse the item is the (imperfect) correspondence between these two worlds, which exists separately in the world of objective knowledge. The possibility that objective knowledge can be incomplete might be interesting; this incompleteness, again, is represented by a probability--we know something, but we do not know all. We are modeling this probability, attempting to explain it, but the probability is unobserved--inferred from raw data but not the raw data themselves. Thus, the probability is an idealization. We need a form of realism that permits idealizations. Or perhaps not, if we suppose that the inference to the probability comprises the correspondence relation between the concept and the world. ]
Partial description. The statement that there is exactly one true and complete description of "the way the world is" is problematic because, as Strawson (1959) said, "The idea of an 'exhaustive description' is quite meaningless in general" (p. 120). There cannot be a complete description of anything within science, and to the extent that a description is complete, it is not scientific. Chemistry did not end with the advent of the periodic table of elements, and biology did not end with the completion of the Human Genome Project--there is still more to be said in these respective disciplines beyond what has been said so far, and this will always be the case because, in science, description is never complete.
A complete description is only possible within a closed system, which is more akin to logic or metaphysics than to science. The DSM taxonomy aims to be a complete description, and the method used to attain this aim, operationism, makes the DSM a closed system and precludes scientific discovery for those operating within the DSM framework.
A commonplace observation is that all theories are born false. A more important observation would be that all theories are born incomplete. The point of theories is not to tell us with certainty what is true but rather to structure our ignorance. Theories are not good because they tell us what is true--otherwise, tautological theories would be the best of all. Theories are not good because they contain certain knowledge but rather because they lead to growth of knowledge. Unfortunately, the DSM, by virtue of defining disorders by means of infallible diagnostic criteria, has a methodological predilection for inhibiting the growth of knowledge.
Verisimilitude. Truth indeed involves a correspondence between words or thought-signs and external things and sets of things (Acton, 1998). There is more to be said about truth in a scientific domain, however, than is contained in a correspondence theory, or copy theory, of truth. There is a kind of truth that lies in process, the process that results in the growth of knowledge. This kind of truth might be called verisimilitude or truth-likeness.
In the form of realism known as critical realism, the notion of truth is complemented by the notion of verisimilitude or truth-likeness. This substitution in the aim of theory construction allows a substitution in the correspondence relation between theoretical concepts, known in the philosophy of language as intensions, and the things in the world to which they refer, known as extensions. Rather than a rigid one-to-one correspondence between category and referent, a graded membership in a category is envisioned.

Fallibilism
Fallibilism is the recognition that our basic assumptions could be wrong. Fallibilism may seem to run counter to realism, which is the aim to generate theories that are right. Fallibilism, however, is an integral part of critical realism. Fallibilism as a scientific method comprises an open invitation to criticism. Criticism is deemed necessary for structuring our ignorance, thereby leading to the growth of knowledge. For the critical realist, science does not aim for knowledge but for growth. One who reads the literature in a scientific domain can see that knowledge has grown--more is known now than was known before.

Fuzzy Sets
Growth of knowledge requires a method for defining items regarding which certain knowledge is lacking. Fuzzy sets--that is, sets that do not have clear boundaries, such that a given item can be a member of the set to a certain degree and only to a certain degree--could prove excellent devices for representing such uncertain knowledge.
Despite the advantages of a dimensional approach based on fuzzy sets, fuzzy sets are not universally accepted. Logicians generally do not like fuzzy sets, because fuzzy sets do not have crisp boundaries. Philosophers generally do not like fuzzy sets, because fuzzy sets do not strictly correspond to natural kinds--one does not have Type O blood to a certain degree and only to a certain degree. Engineers, however, generally love fuzzy sets, because fuzzy sets work. Fuzzy logic has been applied to make superior washing machines, cameras, and trains (e.g., see Zimmermann, 1996). Perhaps fuzzy logic could also be applied to make superior theories of psychopathology.
Adlassnig (1980) has tried out this conjecture by using fuzzy sets to develop a system of computer-assisted medical diagnosis. It is possible to look at psychodiagnosis as an expert system in which theories define the interrelations of symptoms, disorders, and people. Adlassnig has made this proposal explicit and programmatic: "a) Medical knowledge should be stored as logical relationships between symptoms and diagnoses. b) The logical relationships might be fuzzy. They are not obliged to correspond to Boolean logic" (p. 143).
One advantage of using fuzzy sets in a diagnostic system is that fuzzy sets provide a way of structuring our ignorance. Therefore, we do not need a perfect and complete theory before creating the diagnostic system--we can proceed on the basis of a few heuristic principles such as the conjecture that psychological disorders are arranged as dimensions.

Applications in the Classification of Psychopathology
Numerous examples in which a dimensional approach based on fuzzy logic could be applied arise in the domain of psychopathology. In many cases, in fact, there are multiple competing dimensional approaches. This healthy competition may prevent a ready consensus from arising with regard to certain disorders, but as the arguments thus far have indicated, the aim of a science of classification should not be artificial unanimity but rather empirical discovery and the growth of knowledge.
An attempt is made below to indicate a variety of dimensional models that might have heuristic value. No attempt is made to advocate a particular dimensional model over the others or to provide an exhaustive, or even detailed, description of the models that exist.

Bipolar Disorder
Consider bipolar disorder (formerly known as manic-depression). Three related disorders are phenomenologically similar, Bipolar I, Bipolar II, and cyclothymia. In fact, there seems to exist a spectrum ranging from normal mood variation, to cyclothymic personality, to cyclothymic disorder, to Bipolar II, to unipolar mania, to Bipolar I (Akiskal, 1991; Goodwin & Jamison, 1990). These disorders share enough characteristics so as to suggest a number of possible dimensions of relation, including severity of depressive states, severity of manic states, severity of mixed states, polarity, cyclicity, duration of episodes, instability or rapidity of state changes (switches), and so on (Goodwin & Jamison, 1990).

Unipolar Major Depressive Disorder
The International College of Neuropsychopharmacology (CINP) arrived by consensus at the following conclusion: "Unipolar major depressive disorder (MDD) is a pleomorphic mood disorder, consisting of a cluster of depressive subtypes existing in a clinically homogeneous symptomatic continuum from subsyndromal depressive symptomology (SSD) through minor depression, dysthymic disorder, major depressive disorder and double depression..." (Judd, 1997). There seems to be some question about whether these authors advocate a dimensional or categorical system of diagnosis; indeed, they seem to advocate a number of categories arranged along a single continuum. This raises the possibility of simply abolishing the categories and describing depression as a single dimension, the specified dimension being severity. Frances, Widiger, and Fyer (1990) note this possibility:
Given that the threshold is to a large extent arbitrary, it may be preferable to measure anxiety and affective disorders on a continuum to allow the full range to be included in analyses. Much information is lost in a nominal distinction of simply the presence versus the absense of dysthymia. It is apparent that patients vary in the extent to which they display dysthymia, panic disorder, and other anxiety and affective disorders. Why not use this information? (p. 49)

Anxiety Disorders
Anxiety disorders share many overlapping features. This has led investigators to apply a hierarchical dimensional model to the anxiety disorders (Zinbarg & Barlow, 1996). In this model, there is a higher-order general factor (Negative Affectivity) that all anxiety disorders share; there are also lower-order factors that allow differentiation of specific anxiety disorders. Particularly noteworthy are the conclusions drawn by these authors: "First, the emergence of the general, higher order factor suggests the existence of significant overlap at the symptom level even in the absence of a comorbid diagnosis.... These intermediate elevations suggest particularly fuzzy diagnostic boundaries and difficult differential diagnoses" (Zinbarg & Barlow, 1996, p. 190). If indeed the anxiety disorders meld imperceptibly into each other, then perhaps the artificial categories imposed on this continuous reality should be replaced by a dimensional model.

Mixed Anxiety and Depression
Depression and anxiety also share many overlapping features; both correspond to a high degree of negative affect, whereas depression alone corresponds to a low degree of positive affect (Clark & Watson, 1991). Negative affect and positive affect are held by these authors not to be opposites but to be unrelated, or orthogonal, and thus are depicted graphically as lying at right angles to one another.
An alternative dimensional model is the affective circumplex (Russell, 1980). A circumplex is a circular array of variables in two-dimensional space. In this case, the dimensions are pleasure-displeasure and degree of arousal. Depression is a state of high displeasure and low arousal, panic disorder is a state of high displeasure and high arousal, mania is a state of high pleasure and high arousal, the absence of anxiety or affective disorder is a state of high pleasure and low arousal, and cyclothymia is alternation between states. One could carve categories at, say, every 45 degrees of the circumplex and give these categories names--but this would sacrifice the simplicity of the underlying model, and the "categories" would be completely arbitrary (Frances, Widiger, & Fyer, 1990).
Nevertheless, the empirical overlap of depression and anxiety has led proponents of a particular dimensional model to propose a new diagnostic category, mixed anxiety and depression (e.g., Clark & Watson, 1991). Here again is a case in which the categorical assumption might profitably be dispensed with altogether by dealing with these disorders dimensionally. The dimensions proposed above seem to encompass the phenomena reasonably well. The only reason to impose an arbitrary category upon the existing dimensions seems to be deference to the sheer inertia exerted by the DSM's categorical system of diagnosis.

Schizoaffective Disorder
Schizoaffective disorder is characterized both by psychotic features and by mood disturbance. Indeed, this disorder seems to exist midway between schizophrenia and depression. Thus, it serves as another example in which a dimensional approach might be applied. Strange (1994) noted the possibility of a dimensional approach: "it has been suggested that schizophrenia and affective disorders may be separate ends of a continuum of psychotic disorders" (p. 228).

Schizophrenia
Schizophrenia is a psychotic disorder characterized by loss of contact with reality. Before the early 1900s, when Kraepelin distinguished the category of schizophrenia from manic-depression, psychopathologists assumed that the functional (as opposed to organic) psychoses were unitary in nature and dubbed this construct Einheitspsychose (Eysenck, 1992). More recently, Eysenck (1992) and Kendell (1991) have strongly advocated abandoning the neo-Kraepelinian categories represented in the DSM in favor of the broader dimensional construct of Psychoticism.
In one study attempting to differentiate the affective psychoses from schizophrenia, researchers found results consistent with a hierarchical dimensional model comprising a large general factor (Psychoticism) and a small bipolar factor (affective disorder versus schizophrenia; Kendell & Gourlay, 1970). Interestingly, even though participants in this study were selected for having a diagnosis of either schizophrenia or affective disorder, schizoaffective cases (blends of schizophrenia and affective disorder) were more common than "pure" cases of either disorder.
The factor most responsible for differential diagnosis of persons high on Psychoticism may be Extraversion (Armstrong, Hottusson, Ries, & Holmses, 1967; Venables & Wing, 1962; Verma & Eysenck, 1973). As noted by Armstrong et al. (1967), "these results raise the possibility that a significant degree of what is included within the process-reactive frame of reference [in schizophrenia diagnosis] may be considered a function of Extraversion-Introversion" (p. 69). Generalizing on the importance of Extraversion in differential diagnosis, Eysenck (1992) claimed that persons with depression are more extraverted, persons with paranoia and schizophrenia more introverted.
The few existing studies on the dimensionality of Psychoticism (reviewed in detail by Eysenck, 1992) indicate that the present distinction between affective disorder and schizophrenia may be misguided and show the possible utility of a dimensional approach. Although further research is needed into the dimensionality question with respect to the functional psychoses, the shortcomings of the neo-Kraepelinian DSM approach to schizophrenia are already a matter of record (Carson, 1991; Sarbin, 1990).

Schizotypal Personality Disorder
Schizotypal personality disorder is characterized by interpersonal deficits, cognitive and perceptual distortions, and eccentric behavior. This disorder has much in common with schizophrenia. For example, consider the following DSM description of symptoms of schizotypal personality disorder:

Individuals with Schizotypal Personality Disorder often have ideas of reference (i.e., incorrect interpretations of causal incidents and external events as having a particular and unusual meaning specifically for the person).... These should be distinguished from delusions of reference, in which the beliefs are held with delusional conviction. (APA, 1994, p. 641)

This distinction seems tenuous at best: ideas of reference are delusional in the case of schizophrenia but not in the case of schizotypal personality disorder if they are delusional! Other than delusions, these two disorders also share the following features: paranoid ideation, hallucinations, disorganized speech, and negative symptoms (e.g., flattened affect). If these disorders differ at all, perhaps the best dimension on which to characterize their differences would be degree of Psychoticism.

Other Personality Disorders
Experts have confidently predicted that "dimensional approaches will gradually supplant the categorical in the classification of personality disorders" (Frances & Widiger, 1986, p. 396; see also Costa & Widiger, 1994; Plutchik, 1997). In fact, personality disorders are to some extent the paradigmatic case for proponents of dimensional classification, because it is so clear that disordered personality blends imperceptibly into normal personality. As Widiger (1994) observed, "Each personality will have its own Achilles' heel and limitations. In this respect, all people can be said to have some degree of personality disorder" (p. 312).

Back to the Future
The recent history of the DSM represents a shift toward a medical model of discrete illness (Kirk & Kutchins, 1992). This shift, however, may have moved the study of psychopathology away from a previous model that had some merit. Consider the alternative approach formulated by the National Advisory Mental Health Council (1955):
The concept of etiology as embraced by modern psychiatry differs from the simple cause and effect system of traditional medicine. It subscribes to a 'field theory' hypothesis in which the interactions and transactions of multiple factors eventuate in degrees of health and sickness. (quoted in Klerman, 1991, p. 78)

Such an approach represents a refreshing alternative to the currently dominant medical model with its attendant categorical system of diagnosis. One obvious application of this alternative is the recognition of the relatively neglected role in psychopathology of life events (e.g., Dohrenwend & Egri, 1981; Dohrenwend & Shrout, 1985; Dohrenwend, Shrout, Egri, & Mendelsohn, 1980), and of protective factors such as social support and cognitive, temperamental, and occupational strengths (e.g., Pierce, Lakey, Sarason, & Sarason, 1997; Seligman, 1990; Weise, Blehar, Maser, & Akiskal, 1996).

Categories Versus Dimensions: An Empirical Approach

Two Kinds of Categorical Approach
Two kinds of categorical approach should be distinguished: the methodological approach and the empirical approach (see Figure 1 [not yet available on the internet]). The methodological approach that underwrites the DSM is operationism. Operationism is the methodological dictum that all scientific concepts must be completely defined in terms of the operations or measurements used to recognize them.
Operationists with respect to categorization come in two varieties: lumpers and splitters (Frances, Widiger, & Fyer, 1990; Goodwin & Jamison, 1990). Lumpers are comfortable with large categories that display considerable within-group heterogeneity. Splitters want to create a new, homogeneous category for every small variation. Both lumpers and splitters prefer a categorical approach based on operationism.
Operationism is a form of infallibilism and is to be contrasted with the alternative methodological approach, fallibilism. If the arguments in this article are correct, then a dimensional approach based simply on methodological fiat would be no better (and no worse) than the present categorical approach, which itself is based on methodological fiat. The alternative to methodological fiat is fallibilism.
Fallibilism allows for the discovery of surprising findings, which form the springboard for the growth of knowledge. It would be surprising to find that such highly similar disorders as schizophrenia and schizotypal personality disorder (discussed above) are in fact unrelated in the way that different blood types are unrelated. The nature of discovery, however, is that such findings could crop up--if and only if empirical methods can be employed to test the categorical versus dimensional alternatives. Such methods should be encouraged if a realistic assessment is the goal of our diagnostic scheme, and such methods are irrelevant if realism is not our goal.
Maybe we will turn out to be right in our conjectures every time--but if we are right by fiat, then we should not deceive ourselves that realism is our goal. Rather, we should be content with social constructionism (e.g., Rothbart & Taylor, 1992; Sarbin, 1990), because that is the only goal we will have succeeded in achieving.

Categories Versus Dimensions
Dimensions, factors, and traits are to be contrasted with categories, taxa, and types. Dimensions can be thought of as differences in degree, whereas categories can be thought of as differences in kind (Engel, 1989; Meehl, 1992). Differences in degree can be large or small--conceptually, infinitely small, as is the case with real numbers on a number line. Differences in kind do not yield to linear conceptualization--for example, blood type is difficult to conceptualize as lying along any continuum. Once the distinction between traits and types is understood, one can still ask whether it is important to draw the distinction conceptually and whether it is possible to detect the distinction empirically.
The conceptual importance of the dimensions versus categories question hinges on a particular approach to the classification enterprise. Simply put, the question matters only if one is a realist. As noted above, there are multiple versions of realism. Most realists share a belief in the ability of theories to represent the structure of reality accurately. Critical realists demand that theories not only represent what is known about reality but also what is unknown and yet-to-be-discovered (Lakatos, 1970).
Once one accepts that the dimensions versus categories distinction matters conceptually, there remain two empirical questions: first is the question of whether the distinction matters empirically; second is the question of whether the distinction can be empirically detected. In psychology, some of the best work on the detection of latent categories has been done by Meehl and colleagues (e.g., Waller & Meehl, 1998). Indeed, Meehl has coined his own term for the numerical aspects of category detection: taxometrics. Whereas the nature of taxometric methods such as those developed by Meehl and others has been detailed elsewhere (e.g., Bailey, 1989; Eysenck, 1950, 1952, 1992; Grayson, 1987; Waller & Meehl, 1998), it is important to call attention here to their potential for answering the empirical questions that our conceptual analysis has shown to be so important for those of a realist philosophical persuasion.

Epilogue: On Scientific Discovery
The arguments in this article have been motivated by a particular perspective that is an alternative to the DSM approach. Over against the DSM's operationism lies a fallibilist brand of scientific realism that holds as literally true the way of speaking in which it is said that scientific research or clinical experience has discovered a particular disorder. Such discovery arises from theoretically guided observations of the interrelations of symptoms, disorders, and people.
The DSM stifles such observations by dictating that each indicator, in the form of a diagnostic criterion, is infallible, leaving no room for discovery. This, the DSM's great methodological shortcoming, prevents thoughtful people from taking the DSM's categorical approach altogether seriously when an empirically based dimensional approach presents such a fruitful alternative.

References
Acton, G. S. (1998). Classification of psychopathology: The nature of language. The Journal of Mind and Behavior, 19, 243-256.
Adlassnig, K.-P. (1980). A fuzzy logical model of computer-assisted medical diagnosis. Methods of Information in Medicine, 19, 141-148.
Akiskal, H. S. (1991). Cyclothymic, hyperthymic, and depressive temperaments as subaffective variants of mood disorders. In A. Tasman & M. B. Riba (Eds.), Review of psychiatry (Vol. 11, pp. 43-62). Washington, DC: American Psychiatric Press.
Armstrong, H. E., Hotusson, M. H., Ries, H. A., & Holmes, D. S. (1967). Extraversion-introversion and process-reactive schizophrenia. British Journal of Social and Clinical Psychology, 6, 69-74.
American Psychiatric Association (1994). Diagnostic and statistical manual of mental disorders (4th ed.). Washington, DC: Author.
Bailey, J. M. (1989). A critique and reinterpretation of Gordon's IQ-commensurability property. International Journal of Sociology and Social Policy, 9, 64-73.
Carson, R. C. (1991). Dilemmas in the pathway of the DSM-IV. Journal of Abnormal Psychology, 100, 302-307.
Clark, L. A., & Watson, D. (1991). Tripartite model of anxiety and depression: Psychometric evidence and taxonomic implications. Journal of Abnormal Psychology, 100, 316-336.
Costa, P. T., Jr., & Widiger, T. A. (Eds., 1994). Personality disorders and the five-factor model of personality. Washington, DC: American Psychological Association.
Dohrenwend, B. P., & Egri, G. (1981). Recent stressful life events and episodes of schizophrenia. Schizophrenia Bulletin, 7, 12-23.
Dohrenwend, B. P., & Shrout, P. E. (1985). "Hassles" in the conceptualization and measurement of life stress variables. American Psychologist, 40, 780-785.
Dohrenwend, B. P., Shrout, P. E., Egri, G., & Mendelsohn, P. S. (1980). Measures of nonspecific psychological distress and other dimensions of psychopathology in the general population. Archives of General Psychiatry, 37, 1229-1236.
Eccles, J. C. (1974). The world of objective knowledge. In P. A. Schilpp (Ed.), The philosophy of Karl Popper (pp. 349-370). La Salle, IL: Open Court.
Engel, R. E. (1989). On degrees. The Journal of Philosophy, 86, 23-37.
Eysenck, H. J. (1950). Criterion analysis--an application of the hypothetico-deductive method to factor analysis. Psychological Review, 57, 38-53.
Eysenck, H. J. (1952). Schizophrenia-cyclothymia as a dimension of personality: II--Experimental. Journal of Personality, 20, 345-384.
Eysenck, H. J. (1992). The definition and measurement of Psychoticism. Personality and Individual Differences, 13, 757-785.
Frances, A., Widiger, T., & Fyer, M. R. (1990). The influence of classification methods on comorbidity. In J. D. Maser & C. R. Cloninger (Eds.), Comorbidity of mood and anxiety disorders (pp. 41-59). Washington, DC: American Psychiatric Press.
Frances, A. J., & Widiger, T. A. (1986). Methodological issues in personality disorder diagnosis. In T. Millon & G. L. Klerman (Eds.), Contemporary directions in psychopathology: Toward the DSM-IV (pp. 381-400). New York: Guilford.
Goodwin, F. K., & Jamison, K. R. (1990). Manic-depressive illness. New York: Oxford University Press.
Grayson, D. A. (1987). Can categorical and dimensional views of psychiatric illness be distinguished? British Journal of Psychiatry, 151, 355-361.
Judd, L. L. (1997). Pleomorphic expressions of unipolar depressive disease: Summary of the 1996 CINP President's Workshop. Journal of Affective Disorders, 45, 109-116.
Kendell, R. E. (1991). The major functional psychoses: Are they independent entities or part of a continuum? Philosophical and conceptual issues underlying the debate. In A. Kerr & H. McClelland (Eds.), Concepts of mental disorder: A continuing debate (pp. 1-16). London: Gaskell.
Kendell, R. E., & Gourlay, J. (1970). The clinical distribution between the affective psychoses and schizophrenia. British Journal of Psychiatry, 117, 261-266.
Kirk, S. A., & Kutchins, H. (1992). The selling of DSM: The rhetoric of science in psychiatry. Hawthorne, NY: Aldine de Gruyter.
Klerman, G. L. (1991). An American perspective on the conceptual approaches to psychopathology. In A. Kerr & H. McClelland (Eds.), Concepts of mental disorder: A continuing debate (pp. 74-83). London: Gaskell.
Lakatos, I. (1970). Falsification and the methodology of scientific research programmes. In I. Lakatos & A. Musgrave (Eds.), Criticism and the growth of knowledge (pp. 91-196). New York: Cambridge University Press.
Meehl, P. E. (1992). Factors and taxa, traits and types, differences of degree and differences in kind. Journal of Personality, 60, 117-174.
Musgrave, A. E. (1974). The objectivism of Popper's epistemology. In P. A. Schlipp (Ed.), The philosophy of Karl Popper (pp. 560-596). La Salle, IL: Open Court.
Musgrave, A. (1993). Common sense, science and scepticism: A historical introduction to the theory of knowledge. New York: Cambridge University Press.
National Advisory Mental Health Council (1955). Minutes and Proceedings. Public Health Service. Washington, DC: U.S. Government Printing Office.
Pierce, G. R., Lakey, B., Sarason, I. G., & Sarason, B. R. (Eds., 1997). Sourcebook of social support and personality. New York: Plenum.
Plutchik, R. (1997). The circumplex as a general model of the structure of emotions and personality. In R. Plutchik & H. R. Conte (Eds.), Circumplex models of personality and emotions (pp. 17-45). Washington, DC: American Psychological Association.
Popper, K. R. (1968a). Epistemology without a knowing subject. In B. van Rootselaar & J. F. Staal (Eds.), Proceedings of the third international congress for logic, methodology and philosophy of science (pp. 333-373). Amsterdam: North-Holland.
Popper, K. R. (1968b). On the theory of the objective mind. Proceedings of the XIV International Congress of Philosophy, 1, 25-53.
Putnam, H. (1982). Reason, truth and history. New York: Cambridge University Press.
Ross, C. A. (1995). Errors of logic in biological psychiatry. In C. A. Ross & A. Pam (Eds.), Pseudoscience in biological psychiatry: Blaming the body (pp. 85-128). New York: Wiley.
Rothbart, M., & Taylor, M. (1992). Category labels and social reality: Do we view social categories as natural kinds? In G. R. Semin & K. Fiedler (Eds.), Language, interaction and social cognition (pp. 11-36). London: Sage.
Russell, J. A. (1980). A circumplex model of affect. Journal of Personality and Social Psychology, 39, 1161-1178.
Sarbin, T. R. (1990). Toward the obsolescence of the schizophrenia hypothesis. The Journal of Mind and Behavior, 11, 259-284.
Seligman, M. E. P. (1990). Learned optimism: How to change your mind and your life. New York: Simon & Schuster.
Strange, P. G. (1994). Brain biochemistry and brain disorders. New York: Oxford University Press.
Strawson, P. F. (1959). Individuals, an essay in descriptive metaphysics. London: Methuen.
Venables, P. H., & Wing, J. K. (1962). Level of arousal and the subclassification of schizophrenia. Archives of General Psychiatry, 7, 114-119.
Verma, R. M., & Eysenck, H. J. (1973). Severity of type of psychotic illness as a function of personality. British Journal of Psychiatry, 122, 573-585.
Toulmin, S. (1967). The evolutionary development of natural science. American Scientist, 55, 456-471.
Waller, N. G., & Meehl, P. E. (1998). Multivariate taxometric procedures: Distinguishing types from continua. Thousand Oaks, California: Sage.
Weise, R. E., Blehar, M. C., Maser, J. D., & Akiskal, H. S. (1996). Competence models in adult psychopathology: A report on a National Institute of Mental Health workshop. Psychotherapy, 33, 61-67.
Widiger, T. A. (1994). Conceptualizing a disorder of personality from the five-factor model. In P. T. Costa, Jr. & T. A. Widiger (Eds.), Personality disorders and the five-factor model of personality (pp. 311-317). Washington, DC: American Psychological Association.
Zimmermann, H.-J. (1996). Fuzzy set theory--and its applications (3rd ed.). Boston: Kluwer.
Zinbarg, R. E., & Barlow, D. H. (1996). Structure of anxiety and the anxiety disorders: A hierarchical model. Journal of Abnormal Psychology, 105, 181-193.

Last modified June 2001
Visited times since July 2001
Comments?

Home to Great Ideas in Personality