Semiotica 162 (2006), 309-321.
On the basis of clinical experience, we find that Roentgen signs may induce feelings, provoke judgments or impart knowledge in the mind of the interpreter. In this study, we develop a triadic typology of mental effects of Roentgen signs that is based on the Peircean categories of thought. In the process, we provide a categorical basis for the valuations used in diagnostic appraisal. Hence, this study is a contribution to a fully categorical phenomenology of Roentgen diagnosis.
In Roentgen diagnosis, signs that present in Roentgen images are used for the detection, localization and identification of disease (Cantor 2002). In this paper, we shall study the effects of Roentgen signs on the mind of the interpreter. From clinical experience, we know that Roentgen signs may induce feelings, provoke judgments or impart knowledge. We shall show that the Peircean Categories constitute an organizing principle for these mental effects. Furthermore, we shall introduce the concept of categorical markedness as a basis for the concept of value in Roentgen diagnosis.
In this section, we briefly review the concepts and terminology to be used in the construction of a categorical typology of mental effects of Roentgen signs.
For our purposes, a Peircean sign will be thought of as a triadic relation consisting of a binary relation between a representamen and an object that is mediated by an interpretant. Each correlate of a sign acquires its meaning in relation to a ground or context chosen by the interpreter. In general, a correlate ground may be the physical world, a physical object or the mind of the interpreter.
A Roentgen sign is defined to be a Peircean sign with a representamen that presents in a Roentgen image. In this study, the interpretant ground of a Roentgen sign is assumed to be the mind of the interpreter. When the representamen of a Roentgen sign is taken to be a distribution of light emanating from a physical image or the image itself, its representamen ground is the physical world. A Roentgen sign for which the physical world is a common ground for its representamen and its object may be termed a physical sign. If the representamen of a sign is taken to be a perception in the mind of the interpreter, its representamen ground is the mind of the interpreter. If the object of a sign is taken to be a retrieved memory or idea of an object or event, its object ground is the mind of the interpreter. A sign for which the mind of the interpreter serves as a common ground for all of its correlates will be termed a mental sign. In studying the effects of Roentgen signs on the mind of the interpreter, we shall deal only with mental signs.
The Peircean categories
The Peircean categories were the result of an attempt by Charles S. Peirce to comprehend the most fundamental categories of being that are accessible by the mind. Peirce discerned a hierarchy of three categories of being which provide a basis for all thought. The First Category may be described as Quality or Attribute, the Second Category as Reaction or Opposition and the Third Category as Representation or Mediation (Peirce 1867). The Peircean categories form a unified system under an inclusion rule or entailment principle by which Thirdness entails Secondness and Secondness entails Firstness (Liszka : 46).
In a general sense, the concept of markedness characterizes the notion of irreducible 'difference'. It was originally employed in theoretical linguistics (cf. Holenstein : 129-131). Recently, the concept has been adapted by Cantor (2000, 2006) for use in a semiotic theory of diagnosis. Markedness is a property of binary relations that distinguishes between correlates, where one correlate is marked and the other unmarked. We consider a binary relation to be the presence of two objects of thought in the same moment of awareness. A binary opposition is defined to be a marked binary relation. As originally conceived, there were two types of markedness: the general and the specific (using the terminology of Roman Jakobson). In a clinical context, markedness is conferred by the expectations of the interpreter. Hence, in markedness of the general type, an attribute that is unexpectedly present (or absent) in the marked correlate is expected to be absent (or present) in the unmarked correlate. In markedness of the specific type, there is unexpectedly more (or less) of an attribute in the marked correlate than is expected in the unmarked correlate. In what follows, we shall extend the concept of markedness to include a third irreducible type of 'difference' and base the entire scheme on the Peircean categories.
In diagnosis, the marked correlate is unexpected and the unmarked correlate is expected. The categorical types of markedness used in diagnosis will be termed cardinal, polar and ordinal markedness.
In First Category or cardinal markedness, the marked correlate is an unexpected presence or absence of an attribute. Hence, cardinal markedness is based on the category of Attribute. Examples of cardinal markedness in diagnosis are the presence of an unexpected finding or the absence of an expected finding in a Roentgen image. Cardinal markedness is represented by the dichotomy present/absent. This corresponds to general markedness as previously defined.
In Second Category or polar markedness, the marked correlate is an unexpected attribute in a pair of opposites. Hence, polar markedness is based on the category of Reaction. Examples of polar markedness are 'true' as opposed to 'false' or 'false' as opposed to 'true. Polar markedness is represented by the dichotomy positive/negative. This corresponds to specific markedness as defined by Jacobson (1957).
In Third Category or ordinal markedness, the marked correlate has a magnitude that is more or less than expected i.e. ordinal markedness entails a rule of succession and is based on the category of Mediation. Examples of ordinal markedness in diagnosis are an unexpected increase or decrease in an expected finding. Ordinal markedness is represented by the dichotomy increased/decreased. This corresponds to specific markedness as previously defined by Cantor (2000).
This triadic concept of diagnostic markedness is more inclusive than the previous concept and I believe that its categorical basis justifies the use of a new and more conceptual terminology.
We know from experience that perceptions may induce feelings. Furthermore, we observe that the mind easily organizes fundamental concepts in dichotomies. Examples of affective dichotomies are feelings induced by perception of the beautiful and the ugly (a polar dichotomy) and feelings of interest and disinterest (a cardinal dichotomy). Formally, an affective dichotomy may be defined as a marked binary relation between affective states. The correlates of an affective dichotomy are termed its values. A polar valuation is a correspondence between a perception and an induced feeling which is the marked correlate of a polar dichotomy in the mind of the interpreter. A cardinal valuation is a correspondence between a perception and an induced feeling which is the marked correlate of a cardinal dichotomy in the mind of the interpreter. An ordinal valuation is defined in a similar manner. These valuations will be used to establish a fully categorical basis for the effects of Roentgen signs on the mind of the interpreter.
In clinical practice, Roentgen signs may induce feelings in the mind of the interpreter. By tacit convention, this fact is not acknowledged in diagnostic reports or professional publications but is freely expressed in collegial conversations. Roentgen signs that induce feelings in the mind of the interpreter may be termed affective signs. Since feeling is a state or quality of mind, these signs have a First Category effect (see Peirce  for the categorization of feelings). In this study, we shall not make use of an important distinction between feelings and emotions (cf. Damasio :42) but will refer to an affective state as simply 'feelings'. There are three categorical subtypes of affective signs that correspond to three types of feelings. These subtypes will be referred to as aesthetic, attentional and motive signs.
Aesthetic signs may induce feelings of positive or negative value. Modes of positivity are usually expressed by predicates such as 'beautiful' (informally, 'great') or 'good' as in 'This is a beautiful case' or 'This is a good case'. Less frequently, negativity may be expressed as 'This is not a good case'. When a negative feeling is accompanied by empathy, it may be expressed by predicates such as 'terrible' or even 'disgusting'. Such strongly negative feelings are induced by signs that represent severe distortions of expected (normal) anatomy. The question arises as to the characteristics of aesthetic signs that induce positive feelings. We recall that a Roentgen image may be thought of as a map of radiation attenuation within a region of the body that is perceived by the interpreter as a two-dimensional distribution of brightness intensities (cf. Cantor 2000). By semiosis, such perceptions lead to the identification of physical forms and substances within the patient (Cantor 2000, 2002). Elements of such perceptions and cognitions are selected for storage in long-term memory and subsequent retrieval. Such retrievable elements of memory constitute ideas. For aesthetic Roentgen signs, the predicate 'beautiful' refers to the feelings induced by perceptions of unexpected clarity and completeness. The experience of the unexpected induces a feeling of surprise. The feeling of clarity is induced by the effortless perception of an image that has been formed in strict conformity to standard technique without distracting artifacts. The feeling of completeness is induced by perceptions that cannot be distinguished from ideas acquired from prior experience. The feeling of clarity may be thought of as a sense of technical perfection. The feeling of completeness may be thought of as a sense of diagnostic perfection. Hence, the feeling of the beautiful that is induced by aesthetic Roentgen signs entails a feeling of surprise and a sense of perfection. The apparent relativity of aesthetic values may be accounted for by individual differences in
expectations based on prior experience
capacity for memorial selection, storage and retrieval
ability to compare perceptions and ideas.
Clearly, what is felt to be beautiful by one interpreter may not be appreciated by another. Hence, aesthetic signs induce feelings of the beautiful or the ugly that express a polar valuation. Furthermore, aesthetic signs constitute a First Category subtype of affective signs since they induce a particular quality of feeling.
The dominant function of attention is understood to be an orienting reaction of the mind to sensory stimuli (Posner 1997). In this sense, orienting refers to the activation of a state of selective awareness of a location or object (Cavanagh 2004). The major function of this activation or arousal is the detection of target events. Another function of attention is the maintenance of selective awareness (Posner 1995).
In the experience of this author, there is a dual orienting reaction of the mind that is directed away from certain sensory stimuli. This is probably due to selective disengagement or non maintenance of attention. In the reading of clinical images, this reaction is a common cause of "misses" (cf. Cantor 2005). It is well known that attention is attracted to high contrast edges and is averted from low contrast edges. Furthermore, unrealized expectations such as the presence of the unexpected or the absence of the expected attract attention while attention is averted from images where there are severe distortions of shape due to nonstandard projections or severe visual noise due to superposition of structures. On this basis, we propose a generalized concept of attention that orients the mind toward or away from sensory stimuli and that induces feelings of attraction or aversion. Hence, attentional signs induce feelings of attraction or aversion that express a polar valuation. Furthermore, attentional signs constitute a Second Category subtype of affective signs since they provoke an orienting reaction in the mind of the interpreter.
It is likely that there is a 'motivational' or 'seeking' system in the brain that sustains the basic drive "to search, investigate and make sense of the environment" and that the arousal of this system has a characteristic quality of feeling (Panksepp : 145). It is possible that 'phasic engagement' of this system produces feelings of interest and 'tonic engagement' feelings of curiosity (Panksepp : 149). In the reading of diagnostic images, Roentgen signs may induce feelings of interest i.e. feelings that motivate the interpreter to search for meaning in an image. Such feelings are expressed in collegial conversation as 'This is an interesting case'. In clinical practice, feelings of interest are induced by events that the interpreter regards as unexpected, unfamiliar or uncommon. Such appraisals are independent of the extent of prior knowledge. Hence, even the most informed interpreter will experience the unexpected, the unfamiliar and the uncommon. Recognition of the relations between perception and recognition that involve the unexpected, the unfamiliar and the uncommon provides a cognitive basis for feelings of interest (see Clore and Ortony , on cognition in emotion). Such motivational feelings may be induced by
an unexpected presentation of a common condition
a conventional presentation of an uncommon condition
a presentation that is unexpected because the condition is unfamiliar
a familiar presentation that has an unexpected interpretation (eg. a simulator).
It is thought that the mind uses the emotions 'to continually appraise situations for personal relevance' (Clore and Ortony , 29). For the radiologist, the personal relevance of the above situations is their relevance to the task of Roentgen diagnosis. Hence, motive signs induce feelings of interest or disinterest that express a cardinal valuation. Furthermore, motive signs constitute a Third Category subtype of affective signs since they act by mediation between perceiving and knowing.
An intentional sign provokes an intentional state in the mind of the interpreter. According to Franz Brentano (1973: 88,198), an intentional state is a judgment that directs the mind upon an object with acceptance (as true) or rejection (as false). Hence, an intentional state that is due to perceptual experience entails orientation, ideation, valuation and an affective mode (acceptance or rejection). According to John Searle (2004:166-169), an intentional state that is due to perceptual experience may be thought of as a representation of an object or event in the world. This involves a direction of awareness (orientation), a propositional content (ideation), appraisal of the representation (valuation) and a psychological mode (belief or desire). Furthermore, Searle postulates that an intentional representation has an 'aspectual shape' that is determined by a point of view or formal constraint and that an intentional state represents the conditions that must be satisfied for its valuation. Hence, an intentional state in Roentgen diagnosis is a perceptual judgment that entails an appraisal of the clinical revelance of an object of perception that is expressed by a statement that the interpreter believes to be true. Specifically, such perceptual judgments entail an awareness of the presence or absence of active disease that is expressed as a statement of fact. In summary, intentional signs provoke perceptual judgments relating to the presence or absence of active disease in a patient. Since a perceptual judgment involves an orienting reaction, intentional signs have a Second Category effect on the mind of the interpreter. Three categorical subtypes of intentional signs are based on the concepts of existence and time in Roentgen diagnosis, to be defined in the next section.
Roentgen ontology and temporality
Roentgen existence refers to states or events in the region of the body under examination that are represented in the Roentgen image. Clearly, many events of diagnostic significance are not detectable by Roentgen imaging and the absence of disease in an image does not necessarily mean the absence of disease in reality. In general, Roentgen existence and nonexistence are inferred rather than perceived states.
Roentgen time (cf. Cantor 2002) is derived from a conventional Roentgen present. A Roentgen present is determined by events in the patient at the time of image formation. From events in the Roentgen present of an image, the interpreter may infer the probable existence of events in a Roentgen past or the possible existence of events in a Roentgen future based on a knowledge of cause-effect relations. Hence, any Roentgen image determines a past, a present and a future in which events are ordered as in physical time. The three categorical subtypes of intentional signs will be referred to as retrodictive, contemporaneous and predictive signs, corresponding to their location in Roentgen time.
Retrodictive signs refer to events in the Roentgen past of an image. Hence, retrodictive intentional signs provoke perceptual judgments of the clinical significance of past events. Clearly, past events are not amenable to treatment. Examples of such judgments are: 'There are signs of healed granulomatous disease', 'There has been cholecystectomy', 'There is a healed fracture' etc. Hence, retrodictive signs provoke perceptual judgments of events in the Roentgen past that express a polar valuation (true/false). Therefore, retrodictive signs constitute a First Category subtype of intentional signs since they provoke a simple awareness without the possibility of action and this is an irreducible attribute of the mind of the interpreter.
Contemporaneous signs refer to events in the Roentgen present of an image. Hence, contemporaneous intentional signs provoke perceptual judgments of the clinical significance of events in the present. A perceptual judgment of present events is an awareness of events that are in principle amenable to treatment. Examples of such judgments are statements of fact: 'There is active disease 'or 'There is no active disease', 'There is improvement of disease' or 'There is worsening of disease'. Hence, contemporaneous signs provoke perceptual judgments of events in the Roentgen present that express a polar valuation (true/false). Therefore, contemporaneous signs constitute a Second Category subtype of intentional signs since they provoke an awareness of events that may be opposed by treatment, which is a reaction of the mind of the interpreter
Predictive signs refer to events in the Roentgen future of an image. Hence, predictive intentional signs provoke perceptual judgments that make it possible in principle to control the future by the mitigation or avoidance of undesirable outcomes. Examples of such judgments are: 'The patient is at risk for ...' and ' A likely sequela is ...'. Hence, predictive signs provoke anticipations of events in the Roentgen future that express a cardinal valuation (expectation/no expectation). Therefore, predictive signs constitute a Third Category subtype of intentional signs since they provoke an awareness of the possibility of intervention in the course of events i.e. a mediation between the present and the future in the mind of the interpreter.
In a previous study, this author developed a categorical typology of Roentgen signs, based on their cognitive effects (Cantor, 2002). Signs that impart knowledge were termed gnostic signs. Signs that impart knowledge of events in the Roentgen past, present or future were termed anagonstic, diagnostic or prognostic signs, respectively. The reader is referred to the original article for a detailed exposition of that typology along with clinical examples.
This completes our presentation of a categorical typology of Roentgen signs based on their effects on the mind of the interpreter. This categorical typology is represented diagrammatically in Figure 1.
Most of the ideas presented here are simply new combinations of old ideas. In what follows, we provide a brief historical perspective.
Feelings and valuations
The idea that perception of the beautiful is a mode of feeling is due to Hume (2000:303). The idea that feelings present values may be found in Meinong (Schubert Kalsi : 6-7, 51, 64). There have been philosophical studies of specific values since antiquity. However, attempts to find a general characterization of value are relatively recent. The concept of valuation as a quantitative representation first appeared in 19th century mathematics. Meinong's general theory of value was influenced by geometric ideas such as polarity and seriality which are properties abstracted from the real number line (cf. Meinong 1996: 129-131, 143-147). Frondizi (1971: 10-11) referred to the basic characteristics of value as polarity and hierarchy, which are relational values. These properties correspond to our concepts of Second Category (polar) valuation and Third Category (ordinal) valuation. Since unexpected 'presence' and unexpected 'absence' both induce feelings of surprise and these feelings present values, we consider 'presence' and absence' as existential values that determine a First Category (cardinal) valuation. Hence, the full concept of categorical valuation follows from the categorization of existential and relational values.
Intentional existence in Roentgen semiotics is a form of immanent objectivity or reference to events that may or may not exist in actuality (cf. Brentano 1973: 88). Roentgen existence may involve the presence of events that do not exist in actuality or the absence of events that do exist in actuality. Hence, simulators may be confused with the real and the perceptual judgment of 'normality' actually means 'no radiographic abnormality', since some real abnormalities are not registered radiographically. This is one source of error in Roentgen diagnosis (cf. Cantor 2005).
Our triadic concept of Roentgen time is based on the intuitive concept of mental time. In this scheme, the present is experienced in a moment of conscious awareness while the past and future are inferred (cf. Brentano 1973: 327). Brentano has also a provided a historical survey of threefold classifications of mental activities (Brentano 1973:177-193). He adopted the classification of Descartes (1984: 25-26) in which the three fundamental classes of mental activities may be described as knowing, judging and feeling, where knowing is an awareness of ideas (Brentano 1902: 13-14). These activities correspond, in reversed order, to the effects on the mind produced by affective, intentional and gnostic signs. Our recognition of a categorical structure in this triadic scheme was anticipated by Peirce (1992[1887-88]:257-258). Brentano (1902:15) also observed that there are positive and negative modes for both feeling and judgment but not for knowing. This parallels our recognition of polar valuations for First and Second category mental effects and a cardinal valuation for Third Category mental effects of Roentgen signs.
We have presented a typology of Roentgen signs based on their effects on the mind of the interpreter. This typology corresponds to the well-known threefold classification of mental activities consisting of feeling, judging and knowing. We have shown that this typology is based on the Peircean categories of thought. By the categorical inclusion rule, knowledge entails judgment and judgment entails feelings. Furthermore, we have found a categorical basis for the valuations used in diagnostic appraisal. It is suggested that the uses of signs in Roentgen diagnosis are determined by their categorical effects on the mind of the interpreter.