UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Prototype categorization of emotion Fehr, Beverley Anne 1982

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-UBC_1982_A8 F45.pdf [ 5.44MB ]
Metadata
JSON: 831-1.0095323.json
JSON-LD: 831-1.0095323-ld.json
RDF/XML (Pretty): 831-1.0095323-rdf.xml
RDF/JSON: 831-1.0095323-rdf.json
Turtle: 831-1.0095323-turtle.txt
N-Triples: 831-1.0095323-rdf-ntriples.txt
Original Record: 831-1.0095323-source.json
Full Text
831-1.0095323-fulltext.txt
Citation
831-1.0095323.ris

Full Text

PROTOTYPE CATEGORIZATION OF EMOTION BEVERLEY ANNE FEHR B.A.(Hons), U n i v e r s i t y of Winnipeg, 1980 . THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF THE FACULTY OF GRADUATE STUDIES (Department of Psychology) We accept t h i s t h e s i s as conforming to the r e q u i r e d standard THE UNIVERSITY OF BRITISH COLUMBIA August 1982 by MASTER OF ARTS i n Beverley Anne Fehr, 1982 In presenting t h i s thesis i n p a r t i a l f u l f i l m e n t of the requirements fo r an advanced degree at the University of B r i t i s h Columbia, I agree that the Library s h a l l make i t f r e e l y available for reference and study. I further agree that permission for extensive copying of t h i s thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. I t i s understood that copying or publication of t h i s thesis f o r f i n a n c i a l gain s h a l l not be allowed without my written permission. Department of The University of B r i t i s h Columbia 1956 Main Mall Vancouver, Canada V6T 1Y3 )E -6 (3/81) i i Abstract Psychologists have yet to agree on a d e f i n i t i o n of emotion. Attempts at a c l a s s i c a l d e f i n i t i o n , whereby a concept is defined by a necessary and s u f f i c i e n t set of c r i t e r i a l attributes have not met with success. The purpose of t h i s research, therefore, was to test the f e a s i b i l i t y of an alt e r n a t i v e to the c l a s s i c a l view, namely prototype theory. According to the prototype view, concepts are organized in terms of prototypes, which are the clearest cases or best examples of the category. Category members are nonequivalent and can be ordered in terms of their degree of resemblance to the prototypical cases. Boundaries between categories are therefore i l l - d e f i n e d . In t h i s research, the f e a s i b i l i t y of conceptualizing the everyday concept of emotion as structured in terms of prototypes was tested using Rosch's approach. Rosch and her associates have recently demonstrated that many natural language categories such as f r u i t , furniture, and vehicle can be conceptualized as pro t o t y p i c a l l y organized. Rosch has also demonstrated that many natural language categories are organized h i e r a r c h i c a l l y . For example, the set f r u i t , apple, Granny Smith apple i l l u s t r a t e s a hierarchy with a superordinate, middle, and subordinate l e v e l . The f i r s t two studies examined the hi e r a r c h i c a l structure of emotion categories. In Study One, "emotion" was taken to be the highest, or superordinate l e v e l . Subjects were asked to l i s t members of the category emotion. As predicted, prototypical category members l i k e "happiness", "anger", and "sadness", were l i s t e d f i r s t and with greater frequency than less t y p i c a l members l i k e "respect", "awe", and "boredom". The purpose of Study Two was to explore the p o s s i b i l i t y of a subordinate l e v e l of the hierarchy. Subjects were asked to l i s t types of emotion categories generated in Study One. It was discovered that unlike the Study One results where a l l responses were single , words, emotion categories at t h i s l e v e l of the hierarchy are not coded monolexemically. Subjects had to "invent" subordinate categories. Consequently there was l i t t l e agreement in their responses. Internal structure refers to that general class of conceptions of categories in which categories are composed of a core meaning and in which items within the category may be considered d i f f e r e n t i a l l y representative of the meaning of the category term. In t h i s research, representativeness was operationally defined by means of subjects' ratings of how good each item i s as an example of i t s category. In Study Three, p r o t o t y p i c a l i t y ratings were obtained for 20 emotion terms (generated in Study One) as a measure of category representativeness. As predicted, subjects found i t meaningful to rate the extent to which each instance was a good example of the category emotion. Moreover, subjects agreed with one another in their responses. Representativeness of items within a category was then shown i v to affect c e r t a i n dependent variables important in psychological research. Study Four concerned speed of processing. Subjects were asked to v e r i f y statements of the form "An {exemplar} is a {category name}". As predicted, response times were shorter for v e r i f i c a t i o n of the category membership of highly prototypical than less t y p i c a l exemplars. In Study Five, subjects were given the 20 target emotions and were asked to give the general category to which each belonged. As predicted, "emotion" was given as the superordinate category name more often for prototypical than nonprototypical exemplars. In Study Six, subjects generated attributes of the 20 target emotions. A family resemblance score was computed for each emotion based on the attributes each had in common with the other category members. As predicted, prototypical category members resembled the entire family to a greater degree ( i . e . had a higher family resemblance score) than nonprototypical members. Overall, the results suggested how people may organize their concept of emotion. People need not be able to define "emotion" in order to use the concept in an orderly and comprehensive way. V TABLE OF CONTENTS A b s t r a c t i i L i s t of Tables v i L i s t of F i g u r e s v i i i Acknowledgments ix I n t r o d u c t i o n 1 The Search f o r a D e f i n i t i o n of Emotion 2 Is a C l a s s i c a l D e f i n i t i o n P o s s i b l e ? 10 The Prototype View 14 Can Emotion be C o n c e p t u a l i z e d as a Prototype? 21 Overview of the Present Study 26 Part I The H i e r a r c h i c a l S t r u c t u r e of Emotion C a t e g o r i e s ..28 Study One: Free L i s t i n g of Exemplars 28 Study Two: Generation of Subordinates 42 Part II I n t e r n a l S t r u c t u r e of C a t e g o r i e s 68 Study Three: P r o t o t y p i c a l i t y Ratings 72 Study Four: Reaction Time as a Measure of I n t e r n a l S t r u c t u r e 77 Study F i v e : Generating the Superordinate Name 83 Study S i x : Family Resemblances 91 General D i s c u s s i o n of I n t e r n a l S t r u c t u r e 111 C o n c l u s i o n 116 References 124 LIST OF TABLES Table 1. Free l i s t i n g of exemplars of emotion 31 Table 2. Frequency of l i s t i n g in a free production of exemplars task for 20 categories of emotion 38 Table 3. Frequency of l i s t i n g in a free production of exemplars task for 30 categories of emotion 40 Table 4. Subordinates for 20 emotion categories 48 Table 5. Pr o t o t y p i c a l i t y ratings for 20 emotion categories . . 75 Table 6. Pr o t o t y p i c a l i t y ratings for 30 emotion categories 76 Table 7. Mean reaction time in a category v e r i f i c a t i o n task 82 Table 8. Generation of the superordinate category for 20 target emotion terms 87 Table 9. Per cent subjects giving "Emotion" as superordinate category ..88 Table 10. Responses given for generation of the superordinate category task 89 Table 11. Attributes l i s t e d for 20 target emotion categories 97 Table 12. Attributes rated for 20 target emotion categories 105 Table 13. Family resemblance scores for 20 target emotion categories 104 Table 14. Correlations between family resemblance scores v i i f o r "Emotion" and each of 20 t a r g e t emotion c a t e g o r i e s ..109 Table 15. Convergence of o p e r a t i o n s 112 Table 16. C o r r e l a t i o n s f o r 20 t a r g e t emotions among measures of i n t e r n a l s t r u c t u r e ' 113 Table 17. Frequency and p r o t o t y p i c a l i t y c o r r e l a t i o n s f o r 30 c a t e g o r i e s of emotion 115 Table 18. Frequency and p r o t o t y p i c a l i t y c o r r e l a t i o n s f o r 50 c a t e g o r i e s of emotion 117 v i i i LIST OF FIGURES Figure 1. Per cent endorsement per emotion category 37 Figure 2. Number of attributes generated for each number of emotion categories 102 ix Acknowledgments I would l i k e to express my appreciation to the following people for their contributions which enabled the completion of t h i s study. F i r s t l y , I would l i k e to thank my supervisor, Dr. Jim Russell, for the guidance, advice, and support that he u n s e l f i s h l y provided throughout t h i s project. I am especially grateful for the e f f o r t he extended to remain in close contact during his absence. I would also l i k e to thank my committee members: Dr. Bob Knox for his i n s i g h t f u l comments and helpful suggestions, and Dr. Merry Bullock for her expert knowledge and willingness to share i t . The graduate students in the Psychology Annex deserve mention for the encouragement and moral support they provided. A special thank you i s extended to Ross Broughton for his invaluable assistance with the data analysis, and to Stephen Holliday for his generous donation of time and expertise in the attribute c l a s s i f i c a t i o n task. I would also l i k e to express gratitude to the professors who so charitably granted me class time for data c o l l e c t i o n . F i n a l l y , I wish to thank Patty Verkaar for providing her competent typing s k i l l s and for the cheery optimism that f a c i l i t a t e d the completion of this project. 1 Introduction The search for a d e f i n i t i o n of "emotion" c a r r i e d out in psychology and other related d i s c i p l i n e s has been marked with c o n f l i c t , confusion, and disagreement. Such a singular lack of success raises the question whether a d e f i n i t i o n of the word emotion i s even p o s s i b l e — o r rather a d e f i n i t i o n in the c l a s s i c a l sense whereby concepts are defined by a necessary and s u f f i c i e n t set of c r i t e r i a l a t t r i b u t e s . Recently, psychologists have begun to explore an alternative form of d e f i n i t i o n known as the prototype view. Prototypes are defined as the clearest cases or best examples of a category. Within the prototype view, category members can be ordered in terms of their degree of resemblance to the prototypical cases. Membership in the category i s a matter of degree, rather than all-or-none as in the c l a s s i c a l view, and there are no sharp boundaries separating members from non-members. The prototype view may shed l i g h t on how people use and understand the concept of emotion without being able to define i t . The purpose of the research to be reported here was to test the p o s s i b i l i t y that the everyday concept of emotion may be better understood from the prototype view. Thus the research to be reported attempts to describe a form of folk knowledge or everyday way of thinking. It does not attempt to t e l l psychologists how they should conceptualize the 2 phenomena commonly referred to by the word "emotion". The Search for a D e f i n i t i o n of Emotion In the famous Wittenberg symposium on emotion held in 1928, one of the participants described the psychology of emotion as "the most confused chapter in a l l psychology"(Claparede, 1928, p.124). Twenty years l a t e r , Hebb(l949) wrote, "The discussion of emotion has been about as confused as that of any topic in psychology" (p.235). Thirty years after that, Plutchik asked "Why i s the study of emotion in such an unsatisfactory state compared to other parts of psychology?" (p. x v i i ) . As a p a r t i a l answer to his question, he suggested: For one thing, i t appears to be exceedingly d i f f i c u l t to create a d e f i n i t i o n of the word emotion that i s acceptable to most investigators. Some writers have defined emotions as disruptive states; others have defined them as organized. A few have claimed that emotions are so subjective, vague, and idiosyncratic that a general d e f i n i t i o n i s a l l but impossible. And some writers have said that since we cannot c l e a r l y d i s tinguish between emotions and other psychological states, 3 we should drop the term e n t i r e l y " (Plutchik, 1980, p. x v i i ) . The l a t t e r position was adopted by Duffy (1941). In a paper e n t i t l e d "An explanation of emotional phenomena without the use of the concept 'Emotion'", her introductory statement was: "For many years the writer has been of the opinion that 'emotion' as a s c i e n t i f i c concept i s less than useless"(Duffy, 1941, p.283). In a later paper she recommended the "abandonment of 'emotion' and other vague and unmeasurable categories"(Duffy, 1943, p.197). H i s t o r i c a l l y , emotion has been defined in various ways. William James (1884) defined emotion thus: "My t h e s i s . . . i s that the bodily changes follow d i r e c t l y the PERCEPTION of the exciting fact, and that our feeling of the same changes as they occur IS the emotion" (James, 1884, p. 189). For James, emotion was a fe e l i n g , a mental event. James' d e f i n i t i o n has sometimes been misinterpreted as a physiological d e f i n i t i o n of emotion. Thus when Lange (cited in Plutchik, 1980, p.9), a physiologist, suggested that emotion i s a physiological event, the conceptualization became known as the James-Lange theory. However, i t should be clear from his statement quoted, that for James, emotion i s the perception of physiological change; whereas for Lange, physiological change i s emotion. Cannon attempted to remove the emphasis in emotion research that had been placed on peripheral physiology, and replace i t with neurology. According to Cannon: 4 For theory that emotional experiences arise from changes in effector organs i s substituted the idea that they are produced by unusual and powerful influences emerging from the region of the thalamus and aff e c t i n g various systems of c o r t i c a l neurones (Cannon, 1928, p.257) Cannon's research involved removing various portions of animals' brains and observing any resultant emotional states. Based on his research, Cannon concluded that the experience of emotion depends on the occurrence of neural discharges from the optic thalamus. He proposed that the thalamic discharge simultaneously produces both an emotional experience and a series of bodily changes. With the advent of behaviorism occurred a s i g n i f i c a n t change in how emotion was defined. Consistent with his behavioral view, John B. Watson rejected a physiological or neurological analysis of emotion. In his opinion, i t i s perfectly possible for a student of behavior e n t i r e l y ignorant of the sympathetic nervous system and of the glands and smooth muscles or even of the central nervous system as a whole, to write a thoroughly comprehensive and accurate study of the emotions--their types, their i n t e r r e l a t i o n s with habits, 5 their role, etc. (Watson, 1919, p.195). Watson formulated emotion as a hereditary pattern reaction, meaning that "the separate d e t a i l s of the response appear with some constancy, with some regularity and in approximately the same sequential order each time the exciting stimulus is presented" (Watson, 1919, p. 195). What this view implied i s that emotion i s a disturbance of organized a c t i v i t y and that the basic patterns of an emotional reaction are unlearned. The function of learning, in t h i s context, is to disassemble and p a r t i a l l y i n h i b i t the hereditary pattern of emotion. B.F. Skinner offered several d e f i n i t i o n s of emotion. In his early work, Skinner defined emotion in the following way: "Emotion i s not primarily a kind of response at a l l but rather a state of strength comparable in many respects with a drive" (Skinner, 1938, p.407). In 1956, the Langian d e f i n i t i o n was resurrected by Wenger, who defined emotion as " a c t i v i t y and r e a c t i v i t y of the tissues and organs innervated by the autonomic nervous system. It may involve, but does not necessarily involve skeletal muscular response or mental a c t i v i t y " (Wenger, 1956, p.343) . More recently, there has been a trend in psychology to move away from d e f i n i t i o n s of emotion as a single entity toward d e f i n i t i o n s of emotion as consisting of several components. The most popular candidates have been cognitive/mental, behavioral, and physiological responses, 6 with theorists varying in terms of the importance assigned to each of these. Izard (1979) suggested that: Since i t i s generally agreed that emotions have a neuro-physiological, subjective, and a f f e c t i v e component, i t would avoid confusion i f students of emotions designated which of these components they are investigating or discussing (Izard, 1979, p. 448). Lazarus (1975) conceptualized emotion as: "a complex disturbance that includes three main components: subjective a f f e c t , physiological changes...and action impulses having both instrumental and expressive q u a l i t i e s " (Lazarus, 1975, p.554). Based on his research with drug-induced states, Schachter(1967) concluded that bodily reactions comprise • a part of emotional experience, but that, depending on cognitive factors, the same reactions can be part of very d i f f e r e n t emotions. The cognitive component was also emphasized by E l l i s ( l 9 6 2 ) , who s i m i l a r l y conceptualized emotion as consisting of cognitive (referred to as the beli e f system within his framework), physiological and behavioral components. What view of emotion emerges from these attempts to define emotion? Having completed an extensive review of the l i t e r a t u r e on d e f i n i t i o n s of emotion, Plutchik (1980) concluded: "For one thing, i t i s evident that there is 7 r e l a t i v e l y l i t t l e consistency or unanimity in the proposed d e f i n i t i o n s " (p.80). He further commented that: A second interesting point about the d e f i n i t i o n s i s that many are not r e a l l y e x p l i c i t d e f i n i t i o n s at a l l . They talk i n d i r e c t l y about some phenomenon (which we might label X) without giving us any clear idea of what X i s in familiar terms. For example, to say that X is a complex process that has physiological,* expressive, and subjective aspects does not r e a l l y t e l l us what X i s , since many states can be described in exactly these same terms. S i m i l a r l y , to say that an emotion i s a state of strength or weakness of an operant response or that emotional feeling i s added to sensation when the thalamus i s aroused also f a i l s to ide n t i f y emotions for someone who does not already know what they are (Plutchik, 1980, p. 83). Plutchik also observed that many writers who have presented theories of emotion never provide an e x p l i c i t d e f i n i t i o n of the word. This i s the position adopted by Strongman(1973), author of the book Psychology of Emotion: ...at present emotion defies d e f i n i t i o n . Some theorists stress psychological 8 factors, some behavioral, some subjective. Some deal only with extremes, some say emotion colours a l l behavior. For the moment I w i l l not add to the confusion by producing my own d e f i n i t i o n of emotion. Most of thi s book w i l l be concerned with describing emotion rather than defining i t . The aim i s to give the reader more connotations to the word than he has at present (Strongman, 1973, p.1). Perhaps emotion truly defies d e f i n i t i o n . Watson(l9l9) observed that "hard and fast d e f i n i t i o n s are not possible in the Psychology of emotion"(Watson, 1919, p.145). Wenger(l956) offers the following comment on thi s ma 11 e r : Emotion is a peculiar word. Almost everyone thinks he understands i t u n t i l he attempts to define i t . Then p r a c t i c a l l y no one claims to understand i t . S c i e n t i s t s who investigate i t disagree. Philosophers, novelists, and others who write about i t disagree. But in the meantime we a l l go about our individual ways, sometimes enjoying our "emotions" and sometimes bemoaning them (Wenger, 1956, p. 339). S i m i l a r l y , Young(l961) observed that "while everybody 9 talks about emotion no one seems to know exactly what emotion is nor what to do about it"(Young, 1961, p.351). Duffy(l94l) lamented that her recommendation has not been followed: But, alas, the concept of emotion has not been abandoned. Psychologists remain convinced that the term refers to a distinguishable category of responses, and they perservere in the attempts to give this category more adequate d e f i n i t i o n . . . The readings of these d e f i n i t i o n s has l e f t the writer with a sentiment similar to that expressed by William James in regard to c l a s s i f a c t o r y descriptions of separate emotions—that he 'should as l i e f read verbal descriptions of the shape of the rocks on a New Hampshire farm as t o i l through them again'(Duffy, 1941, p. 283) What emerges i s a potpourri of views on the matter of .defining emotion. The writers c i t e d agree that attempts at defining emotion have not been successful. Plutchik(1980) maintains that due to the necessity of an e x p l i c i t d e f i n i t i o n these e f f o r t s should continue. Others, l i k e Strongman, said that a d e f i n i t i o n i s not possible, but emotion should s t i l l be studied. Duffy (1941) suggested that emotion is not 10 definable, and that t h i s i s grounds for not continuing to use the concept. Is a C l a s s i c a l D e f i n i t i o n Possible? The prototype analysis of emotion to be offered here w i l l not necessarily t e l l psychologists how best to. conceptualize emotional phenomena. But the fact that experts have been unable to agree on a d e f i n i t i o n does suggest that non-experts may not base their use of the word "emotion" on a c l a s s i c a l d e f i n i t i o n . The experts' arguments over d e f i n i t i o n also suggests that they have assumed that such a c l a s s i c a l d e f i n i t i o n can be found. This may be because concepts have t r a d i t i o n a l l y been thought of as defined by a set of necessary and s u f f i c i e n t c r i t e r i a l a t t r i b u t e s . Category membership i s therefore an all-or-none phenomenon—any instance which meets the c r i t e r i o n i s a member, other things are non-members. Boundaries between categories are thus c l e a r l y defined. Since each member must possess the pa r t i c u l a r set of attributes that i s the c r i t e r i o n for category inclusion, a l l members have a f u l l and equal degree of membership and are therefore equally representative of the category. The t r a d i t i o n a l , c l a s s i c a l view of concepts fosters the assumption that a precise d e f i n i t i o n of emotion i s both necessary and possible. This assumption about the nature of concepts seems to have been i m p l i c i t in the search for a d e f i n i t i o n of emotion. This assumption has not been accepted by everyone, however. William James opened his book The V a r i e t i e s of Religious  Experience with the comment that "Most books on the philosophy of r e l i g i o n try to begin with a precise d e f i n i t i o n of what i t s essence consists of" (James, 1929, p.26). Adhering to t h i s l i t e r a r y t r a d i t i o n , James attempted to define " r e l i g i o u s experiences". He began by making a d i s t i n c t i o n between i n s t i t u t i o n a l and personal r e l i g i o n before offering any d e f i n i t i o n s : Were we to l i m i t our view to i n s t i t u t i o n a l , we should have to define r e l i g i o n as an external art, the art of winning the favor of the gods. In the more personal branch of r e l i g i o n i t i s in the contrary the inner dispositions of man himself which form the center of his interest, his conscience, his deserts, his helplessness, his incompleteness. (James, 1929, p.29) James decided to focus on personal r e l i g i o u s experiences. He discovered that for each component or attribute he proposed, he could readily generate an example of a r e l i g i o u s experience in which that pa r t i c u l a r component was absent. Moreover, the issue of how personal r e l i g i o n could be distinguished from man's conscience or morality was problematic. After much debate, James tentatively suggested that r e l i g i o u s experiences might be characterized by 12 solemnity, seriousness, and tenderness. However, even at t h i s stage, q u a l i f i e r s were necessary. For example, "If glad, we must not grin or snicker...If sad we must not scream or curse..." (James, 1929, p. 38). F i n a l l y , James was compelled to conclude that: ...do what we w i l l with our defining, the truth must at l a s t be confronted that we are dealing with a f i e l d of experience that cannot be sharply drawn. The pretension, under such conditions, to be rigorously ' s c i e n t i f i c ' or 'exact' in our terms would only stamp us as lacking in understanding of our task. Things are more or less divine, states of mind are more or less r e l i g i o u s , reactions are more or less t o t a l , but the boundaries are always misty, and i t is everywhere a question of amount and degree (James, 1929, p. 38). In l i g h t of his o r i g i n a l purpose, namely defining r e l i g i o n , James' f i n a l statement on the matter was,"the word r e l i g i o n cannot stand for any single p r i n c i p l e or essense, but is rather a c o l l e c t i v e name. The theorizing mind tends to the over-simplification of i t s materials" (James, 1929, p. 26) . To i l l u s t r a t e a similar point, Wittgenstein (1953) 1 3 attempted to define "games". He speculated that a game could be defined as an event in which there i s competition between players. However, consider children playing ring-around-the-rosy. This i s a game in which there i s no competition between players. Wittgenstein found that for each supporting example an equal number of counterexamples could be found. He then speculated that games might rather be defined as requiring s k i l l . However, the s k i l l s required to play chess, hide-and-seek, and tennis seem very d i f f e r e n t in nature, and there are games of chance that require no s k i l l at a l l . Thus, that d e f i n i t i o n seemed inadequate. After numerous such attempts, Wittgenstein was forced to declare himself a loser in the defining game. Like James, he eventually concluded that a concept l i k e "game" cannot be e x p l i c i t l y defined. He i l l u s t r a t e d this point with a rather descriptive analogy: in spinning a thread we twist f i b r e on fib r e and the strength of the thread does not reside in the fact that some one fibre runs through the whole length, but in the overlapping of many fi b r e s (Wittgenstein, 1953, p.32). The absence of a defining feature or set of features precludes the establishment of a game/nongame boundary. If a concept l i k e "game" could be defined by a necessary and s u f f i c i e n t set of features, then any instance that possessed the c r i t e r i a l features would be a game, and any instance that did not possess the c r i t e r i a l features would not be a game. 14 Boundaries between games and nongames would therefore be well defined. However, as Wittgenstein demonstrated, the concept "game" does not possess a set of necessary and s u f f i c i e n t c r i t e r i a l a t t r i b u t e s . Different games share d i f f e r e n t attributes that vary in kind and number. Some attributes are also common to neighboring concepts resulting in i n d i s t i n c t between-category boundaries. Wittgenstein argued that a concept with blurred edges i s , however, no less a concept than one in which the boundaries are sharply defined. This last point of Wittgenstein's is an important one in l i g h t of Duffy's widely c i t e d argument that because "emotion" cannot be c l a s s i c a l l y defined, i t i s not a useful concept. The Prototype View One alternative to the c l a s s i c a l view of concepts is subsumed under the rubric "prototype theory". James alluded to much of what is now c a l l e d "prototype theory" in The  V a r i e t i e s of Religious Experience, although Wittgenstein is generally credited as the founding father of t h i s view. In the 18th century, Bishop Berkeley raised questions concerning the role of prototypes in psychological processes, stimulating modern day psychologists to empirically investigate t h i s issue. Recently, Eleanor Rosch and her colleagues have been mainly responsible for rekindling an interest in t h i s idea. Very generally, the prototype point of view i s that categories have an internal structure. Categories are 15 organized around prototypes, which are the clearest cases or best examples of the category. To be a category member is to resemble the prototype. Category members thus d i f f e r in terms of degree of resemblance to the prototypical cases. This means that a l l category members are not equivalent, and can be ordered in terms of how representative they are of the category. Exemplars that are the most representative of the category share the greatest number of attr i b u t e s with a l l other members of the category. Less t y p i c a l exemplars have fewer attributes in common with the prototypical cases and also have a greater proportion of attributes in common with other categories. Boundaries between categories are therefore i l l - d e f i n e d . Consider, for example, some of the q u a l i t i e s o r d i n a r i l y treated as attributes in c l a s s i f y i n g animals: "coat"(fur, feathers), "oral opening" (mouth, beak), and "primary mode of locomotion" ( f l y i n g , on foot) (Mervis and Rosch, 1981). Robins and sparrows, prototypical members of the category "bird", share a l l of these q u a l i t i e s : feathers, beak, and f l y i n g . Turkeys are less prototypical exemplars of the category. Turkeys do not share the "primary mode of locomotion" attribute with robins and sparrows. Penguins are even less representative of the category "bird" and do not share the feathers and f l y i n g attributes with sparrows and robins. The c l a s s i c a l view of concepts, in contrast, advocates d i s t i n c t category boundaries, and a spe c i f i a b l e set of necessary and s u f f i c i e n t c r i t e r i a l a t t r i b u t e s . Each exemplar 1 6 i s therefore equally representative of the category. However, i t i s not necessary to choose between the c l a s s i c a l view and the prototype view—both may be correct in representing something of the way in which we think about some concepts. Or, perhaps the c l a s s i c a l view i s a better view of how we define some concepts, the prototype view a better view for other concepts. Undoubtedly some categories and some kinds of processing of categories are an all-or-none phenomena. Something either is or i s not a one d o l l a r b i l l ; someone either is or i s not pregnant. Conversely, one would probably not say that a particular person i s "sort of" the Prime Minister of Canada. The c l a s s i c a l view of concepts has t y p i c a l l y been i m p l i c i t in the concept formation research paradigm, in which subjects learn c r i t e r i a l a ttributes for a concept, e.g., blue and c i r c l e . If the target subset consists of the conjunction "blue c i r c l e " , with size as an irrelevant a t t r i b u t e , i t does not make sense to ask i f the large or small c i r c l e i s a better example of the concept "blue c i r c l e " . On the other hand, categories l i k e d o l l a r b i l l s and Prime Ministers may not be representative of the majority of concepts. The notion that prototypes play a role in psychological processing of some concepts has been supported amply through research. The empirical studies seeking evidence concerning prototypes have largely addressed the question of how prototypes develop and by what process new exemplars are 1 7 c l a s s i f i e d . The question i s not new; Bishop Berkeley considered i t a long time ago: In his mind's eye a l l images of triangles seemed to have rather s p e c i f i c properties. They were eq u i l a t e r a l or isoceles or right t r i a n g l e s , and he searched in vain for a mental image of the "universal t r i a n g l e " . Although i t is easy to define verbally what we mean by a t r i a n g l e , i t is not clear what the "perfect" triangle looks l i k e . We see lo t s of d i f f e r e n t kinds of triangles; from t h i s variety what do we create in our mind as the basis of recognizing a triangle? (Cited in Calfee, 1975, p. 222) . The speculation invited by Berkeley's search for the "perfect t r i a n g l e " culminated several centuries later in what has i t s e l f become a prototype experiment by Posner, Goldsmith and Welton (1967). These investigators created the prototype of a tr i a n g l e and other forms, and then presented subjects with d i s t o r t i o n s of the prototypes. It was found that subjects could c l a s s i f y the d i s t o r t i o n s of a particular prototype into a common category. Patterns derived from another prototype were grouped together. In a subsequent c l a s s i f i c a t i o n task, the o r i g i n a l prototypes were included in the set of stimuli to be grouped. Subjects c l a s s i f i e d the 18 prototypes (which had not been previously presented) as accurately as the d i s t o r t i o n s that they had grouped in the f i r s t task. Franks and Bransford(1971) constructed a series of figures, one card consisting of the prototype, and the remainder "transformations", which varied in the number of deviations (distance) from the prototype. Subjects were required to reproduce the transformations during the "tr a i n i n g phase" of the experiment. A subsequent recognition task included the prototype (not previously seen) and the transformations. The investigators discovered that subjects "recognized" the unseen prototype with greater probability than i t s previously seen transformations. Moreover, they did so with a greater degree of confidence. It was also found that the recognition ratings were related to transformational distance, with the prototype most frequently recognized, transformations consisting of one permutation next, and so on. Using "real l i f e " figures, Reed (1972) conducted several studies involving faces in which the features (e.g. eye placement, length of nose, height of forehead) were varied. In a t y p i c a l problem subjects were asked to c l a s s i f y these schematic faces into one or the other of two rows of faces. According to Reed, "the dominant strategy was to form an abstract image or prototype to represent each category and to c l a s s i f y test patterns on the basis of s i m i l a r i t y to the two prototypes" (Reed, 1972, p. 401). Eleanor Rosch and her associates have been instrumental 19 in a r t i c u l a t i n g and re-kindling an interest in the approach to categorization suggested by James, Wittgenstein, and the modern day psychologists mentioned. I n i t i a l l y , Rosch's research focused on color categories(Heider 1971,1972). In a series of studies, she demonstrated that there are salient areas of the color space (focal colors) which are given the shortest names and are named most quickly across languages. Focal colors are also most accurately recognized across cultures and are paired with their corresponding names with fewest errors. A developmental study by Mervis, C a t l i n , & Rosch (1975) revealed that foci for color categories become established and s t a b i l i z e d e a r l i e r than boundaries, and focal judgments are more stable than boundary judgments. Rosch then extended her work to semantic categories for everyday objects (Rosch 1973,1974,1975a,1975b; Rosch and Mervis 1976, Rosch, Mervis, Gray, Johnson, & Boyes-Braem 1976; Rosch, Simpson, & M i l l e r 1976; Rosch 1978a,1978b; Rosch 1981; Rosch and Mervis 1981). In t h i s research, representativeness of category members was measured through subjects' ratings of goodness-of-example for natural language categories l i k e f r u i t , sport, vehicle, b i r d , and so on. Reaction times in a category v e r i f i c a t i o n task were shorter to prototypical exemplars (Rosch, 1973). Priming (prior presentation of the category name) f a c i l i t a t e d recognition for highly t y p i c a l but not for less t y p i c a l exemplars (Rosch, 1975b). It was also demonstrated that t y p i c a l exemplars share the most attributes, while less t y p i c a l exemplars have fewer 20 attributes in common with the prototypical cases, and also have a greater number of attributes in common with adjacent categories (Rosch and Mervis, 1975). Some of the major concepts in psychology have recently been re-conceptualized in l i g h t of the prototype view of concepts. Cantor, Smith, French, & Mezzich (1980) addressed a problematic aspect of psychiatric diagnosis—namely that many patients do not f i t into one and only one category. Some patients appear to be prototypical examples of schizophrenia, depression, or other diagnostic categories, but other patients are rather poor examples. From a prototype view, the blurry boundaries of the diagnostic categories can be viewed as orderly and predictable, rather than problematic. Empirical evidence on diagnostic judgments supported the hypotheses derived from the prototype view of these categories. Cantor and Mischel (1979) have also applied this view to an analysis of personality types. Extraversion, for example, can be formulated in terms of a "prototypical extravert". These researchers also presented subjects with statements about various personality types (Cantor and Mischel, 1977). In a subsequent recognition task subjects "recognized" highly prototypical statements that had not been previously presented with greater certainty than statements of intermediate degrees of t y p i c a l i t y that they had seen. Neisser(1979) argued that no single a b i l i t y or single mental process can serve as an adequate d e f i n i t i o n of 21 i n t e l l i g e n c e . He suggested that the concept of i n t e l l i g e n c e is a category which is organized in terms of a prototype. Actual persons resemble the " i n t e l l i g e n t " prototype to varying degrees and along varying dimensions. Thus two equally i n t e l l i g e n t persons could be quite d i f f e r e n t in their actual mental s k i l l s . One i n t e l l i g e n t person might be extremely good at solving the Rubik's cube and cross-word puzzles but mediocre in t e l l i n g jokes and f i l l i n g out income tax forms. Another equally i n t e l l i g e n t person might have these attributes in reverse. Can Emotion be Conceptualized as a Prototype? How might the concept of emotion be conceptualized? In approaching such an issue, the lesson that can be learned from James warrants lengthy quotation: Let us not f a l l immediately into a one-sided view of our subject, but l e t us rather admit freely at the outset that we may very l i k e l y find no one essence, but many characters which may alternately be equally important in r e l i g i o n . If we should inquire for the essence of 'government', for example, one man might t e l l us i t was authority, another submission, another polic e , another an army, another an assembly, another a system of laws; yet a l l the while it-would be true that no concrete government 22 can exist without a l l these things, one of which is more important at one moment and others at another. The man who knows government most completely troubles himself least about a d e f i n i t i o n which s h a l l give their essence. Enjoying an intimate acquaintance with a l l their p a r t i c u l a r i t i e s in turn, he would naturally regard an abstract conception in which these things were unifi e d as a thing more misleading than enlightening and why may not r e l i g i o n be a conception equally complex? (James, 1929, p. 27). Substituting "emotion" for " r e l i g i o n " in t h i s quotation is evocative. As was argued e a r l i e r , attempts at defining "emotion" have not been p a r t i c u l a r l y successful or illuminating, but have met the same fate as the search for d e f i n i t i o n s of games and r e l i g i o u s experiences. The thesis of the present study is that "emotion" lacks a c l a s s i c a l d e f i n i t i o n and i s organized as a prototype, with no clear boundary separating emotion from non-emotion. There i s no empirical evidence currently available in the l i t e r a t u r e to support t h i s thesis, although several theorists have suggested the p o s s i b i l i t y . Based on an examination of various d e f i n i t i o n s of emotion that have been offered, Duffy(l94l) noted: Changes in energy l e v e l , in degree of 23 organization of responses, and in conscious state occur in a continuum. There i s no point on t h i s continuum where a "non-emotional" energy l e v e l changes suddenly to an "emotional" evergy l e v e l ; there is no point at which a "non-emotional" degree of disorganization of response changes suddenly to an "emotional" degree of disorganization; and there is no point at which a "non-emotional" conscious state changes suddenly to an "emotional" one. These c h a r a c t e r i s t i c s of experience and behavior show continuous v a r i a t i o n rather than separation into hard and fast categories. Extremes of the continuum are readily i d e n t i f i e d as "emotion"; intermediate points offer d i f f i c u l t y in i d e n t i f i c a t i o n (Duffy, 1941, p.291). A v e r i l l (1975) set out to c o l l e c t an exhaustive l i s t of emotion terms in the English language, and eventually concluded that: There are some terms which everyone would agree refer to emotional states, e.g. "angry", " f e a r f u l " , "grieving", "loving", but such terms are r e l a t i v e l y few, and they represent only one extreme of a 24 continuum. Between the two ends of t h i s continuum of a f f e c t i v e meaning, any dividing l i n e between emotional and unemotional concepts is necessarily vague and somewhat arb i t r a r y ( A v e r i l l , 1975, p. 6). Leeper (1948) observed that some emotions are commonly regarded as prototypical. His focus, however, was not so much on an empirical examination of how the concept i s organized but was, rather, e s s e n t i a l l y a plea for the inclusion of posit i v e emotions within the set of "prototypical emotions": Unless we take an extremely pessimistic view of human l i f e , we might just as well say that such "pleasurable emotions" or "positive emotions" are, in general, just as numerous and important in human l i f e as are the "unpleasurable" or "negative" emotions. It i s hard to see, therefore, why they secure merely passive mention and why such emotions as fear and anger are discussed as though they are the only v a l i d prototypes of emotion (Leeper, 1948) . While Leeper does not make any systematic investigations of t h i s notion, he deserves mention for suggesting that some emotions can be (are) regarded as more prototypical than 25 others. What are the implications of re-conceptualizing emotion in the way that concepts l i k e i n t e l l i g e n c e , psychiatric diagnosis, personality types and so on have been organized? If emotion is pr o t o t y p i c a l l y organized, one would expect that a l l members would not be equally representative of the category, but rather a given instance would be a better or poorer example of the category in re l a t i o n to the other category members. Further, one would predict that rather than possessing a common set of c r i t e r i a l a t t r ibutes, category exemplars would share various attributes in overlapping and cr i s s - c r o s s i n g ways. On the other hand, i f emotion i s c l a s s i c a l l y defined, one would expect no internal structure. A l l members would be equally representative of the category and would share the same set of defining features. One of the major issues that would be influenced by such a view i s the idea of a definable set of emotions. One would expect that certain emotions would be more exemplary of the category, while others would be less representative. Rather than being considered anomalous, borderline cases in this context would be orderly and expected. Such a view of emotion would imply that a d i s t i n c t emotion/nonemotion boundary cannot be established. A prototype view of emotion would not mean, however, that the concept must be abandoned, as Duffy suggested. It seems that looking at emotion as a concept with blurred edges could be p o t e n t i a l l y useful. Adopting the following argument 26 from Wittgenstein: One might say that the concept 'game' i s a concept with blurred edges. But i s a blurred concept a concept at a l l ? Is i t even always an advantage to replace an i n d i s t i n c t picture by a sharp one? Isn't the i n d i s t i n c t one often exactly what we need? (Wittgenstein, 1953. p. 34). Overview of the Present Study To recapitulate, various writers have alluded to aspects of the structure of emotion which invites speculation and empirical investigation of the notion that the concept of emotion i s p r o t o t y p i c a l l y organized. Such an approach seems warranted given the state of the art of defining emotion. The present research consists of a set of preliminary studies to test the f e a s i b i l i t y of viewing emotion as a pr o t o t y p i c a l l y organized concept. One of the merits of the Roschian approach is that through her research with natural language categories, she has provided a framework that can be applied to new areas. Rosch does not s p e l l out s p e c i f i c hypotheses, but rather offers a general approach that must be specified through empirical means in any p a r t i c u l a r domain. The exploratory nature of the present research must therefore be emphasized. Two kinds of studies were conducted in the present investigation. The f i r s t set of studies examined the notion that emotion categories are structured h i e r a r c h i c a l l y , where 27 the idea of a hierarchy very simply refers to a p r i n c i p l e of inclu s i o n : a higher l e v e l includes the level s below i t . Common sense suggests that wrath, indignation, and fury may be types of anger and that anger, fear, and happiness may be types of emotion. Because no investigations have been car r i e d out on thi s topic, however, the h i e r a r c h i c a l nature of the concept of emotion was the f i r s t topic of study here. The second type of study referred to the internal  structure of categories. In the domain of emotion, one would expect that certain instances would be better examples of the category than others, and that category members could be ordered along a dimension of category representativeness. If successful, then further questions would be raised: Is the category membership of protoytpical cases v e r i f i e d faster than that of less t y p i c a l cases? Is there a greater l i k e l i h o o d that prototypical cases w i l l e l i c i t "emotion" as their category name? Do the prototypical cases have more attributes in common with each other as compared to the less representative exemplars? It i s from the convergence of several measures that internal structure i s demonstrated. Following Rosch, in thi s research, evidence from several studies is r e l i e d upon in order to demonstrate internal structure. If these measures converge, the case that emotion may be a prototype concept can be made with greater confidence. 28 PART I THE HIERARCHICAL STRUCTURE OF. EMOTION Some categories are subordinate to other categories; some are superordinate. For example, the set f r u i t , apple, Granny Smith apple i l l u s t r a t e s a hierarchy with a superordinate, middle, and subordinate l e v e l . Rosch has demonstrated that many natural language categories l i k e vehicle, f r u i t , and furniture are organized h i e r a r c h i c a l l y (Rosch et a l . 1976b). In studies investigating the hi e r a r c h i c a l structure of categories, subjects are t y p i c a l l y given the category name and are asked to generate examples that f a l l within the category. Or, conversely, subjects are given a lower l e v e l instance and are required to give a name higher in the hierarchy. Whether or not emotion forms this kind of hierarchy i s open to investigation. For my purposes, emotion was taken to be the highest l e v e l , and was therefore termed "superordinate". In one study, subjects were asked to l i s t instances (members) of the category emotion. In order to explore the p o s s i b i l i t y of a subordinate l e v e l , subjects in another study were asked to l i s t instances of some of the categories that had been generated in the f i r s t task. Study One Free L i s t i n g of Exemplars In th i s free l i s t i n g task, subjects were simply required to l i s t what they thought was included in the category emotion. 29 Method Subjects. Subjects were 200 students enrolled in various psychology courses at the University of B r i t i s h Columbia. Pa r t i c i p a t i o n was voluntary. Procedure. Subjects read instructions and were given one minute to provide their responses. The following instructions were provided: This study i s part of a larger project on the sorts of things people have in mind when they hear and use words. On thi s questionnaire we are interested in the kinds of things which might belong to general categories. We w i l l give you the category and you w i l l give us the items. For example, i f given the category "SEAFOOD", you might respond with such items as clams, oysters, lobster t a i l s , shrimp salad, pickled herring, and so on. Now, please l i s t as many items of the category "EMOTION" as come readily to mind. Stop after about a minute or 20 items. Results and Discussion Subjects' responses were collapsed across syn t a c t i c a l form (e.g."happy" and "happiness" were combined). The rationale was that in no instance did a subject mention both 30 terms as though they were separate emotions. This l e f t 383 d i f f e r e n t responses, which are l i s t e d in Table 1. Instances l i s t e d for the category emotion seemed to vary tremendously in how readily they came to mind. Of the 383 responses, 188 were idi o s y n c r a t i c , i . e . l i s t e d by only one i n d i v i d u a l . Only four categories, happiness, anger, sadness, and love, were l i s t e d by more than half the subjects. Figure 1 i l l u s t r a t e s t h i s point by showing the d i s t r i b u t i o n of endorsements for the 383 terms—endorsement defined as the per cent subjects who l i s t e d the term. As can be seen, there was considerable v a r i a b i l i t y in the frequency with which each term was generated, with no clear break in the d i s t r i b u t i o n . There was no obvious boundary but rather a gradual change from instances that came readily to mind to those that do not. Twenty emotion terms, l i s t e d in Table 2, were selected as the target terms for further analysis in t h i s and subsequent studies. The items with the ten highest endorsements were selected. The remaining items were chosen at random from among those 108 remaining items, that were generated by at least 4 subjects (2% of the sample). In subsequent studies, the assumption was made that these terms exist at the same hi e r a r c h i c a l l e v e l . At the time of data c o l l e c t i o n i t was unknown that similar frequency data had been col l e c t e d for various categories, including emotion, by Hunt and Hodge (1971). These researchers asked participants to l i s t only four Table 1 Free L i s t i n g of Exemplars of Emotion happiness(152) confusion(17) glad(9) anger(149) surprise(17) a f f e c t i o n (8) sadness(136) despair(16) boredom(8) love(124) hurt(16) delight(8) fear(96) liking(16) greed(8) hate(89) lonely(16) hope(8) joy(82) sympathy(16) lust(8) excitement(53) compassion(14) tenderness(8) anxiety(50) ecstasy(14) annoyed(7) depression(42) envy(14) arousal(7) frustration(39) grief(14) cheerful(7) crying(36) mad(14) di sappo.intment ( 7) feelings(35) sorrow(14) di stress(7) jealousy(29) warmth(14) frightened(7) disgust(27) nervous(13) hoplessness(7) laughter(27) pa i n (1 3) i r r i t a t i o n ( 7 ) elat ion(26) tense(13) kindness(7) caring(24) moody(12) longing(7) guilt(22) pride(12) melancholy(7) embarrassment(20) smiling(12) rage(7) contentment(19) trust(12) r e l i e f ( 7 ) peace(19) passion(11) sensitive(7) upset(19) tears(11) pleased(7) worry(19) pleasure(lO) respect(7) empathy(18) calm(9) scared(7) Table 1 (continued) sex ( 6 ) h o s t i l i t y ( 4 ) e x p e c t a t i o n (3) shyness(6 ) humor(4) e x p r e s s i v e ( 3 ) s i n c e r i t y ( 6 ) l o y a l t y ( 4 ) g i v i n g ( 3 ) strong ( 6 ) miserable ( 4 ) h e l p i n g ( 3 ) a f r a i d ( 5 ) mournful ( 4 ) h e l p l e s s n e s s ( 3 ) a n t i c i p a t i o n ( 5 ) needs (4 ) high ( 3 ) b i t t e r n e s s ( 5 ) pensive ( 4 ) h u m i l i t y ( 3 ) concern ( 5 ) r e j e c t i o n ( 4 ) j u b i l a t i o n ( 3 ) c o n t r o l ( 5 ) remorse (4 ) negative ( 3 ) d i s l i k e ( 5 ) s e r e n i t y ( 4 ) p a s s i v i t y ( 3 ) exuberance(5 ) shame(4) p o s i t i v e ( 3 ) p a n i c ( 5 ) sharing ( 4 ) q u i e t ( 3 ) s a t i s f a c t i o n ^ ) s t r e s s ( 4 ) r e a c t i o n s ( 3 ) touching( 5 ) t h r i l l e d U ) resentment ( 3 ) aggression ( 4 ) t r a n q u i l i t y ( 4 ) t e r r o r ( 3 ) amused(4) unhappy (4) t h i n k i n g ( 3 ) apprehension ( 4 ) v i o l e n c e ( 4 ) wonder (3) awe(4) v u l n e r a b i l i t y ( 4 ) admi r a t ion ( 2 ) deep(4) ambivalence ( 3 ) a l e r t ( 2 ) d e s i r e ( 4 ) a t t r a c t i o n ( 3 ) amazement(2) dismay ( 4 ) b l i s s ( 3 ) a p p r e c i a t i o n ^ ) enjoyment (4 ) confidence ( 3 ) anguish(2 ) enthusiasm(4 ) c o n f 1 i c t ( 3 ) belonging ( 2 ) e x h i l a r a t i o n ( 4 ) d e f e a t ( 3 ) b o i s t e r o u s ( 2 ) gay (4 ) d e j e c t i o n ( 3 ) c l o s e n e s s ( 2 ) Table 1 (continued) commun icat ion(2) complacent(2) contempt(2) c r i t ic i sm(2) cynical(2) devotion(2) d i s t r u s t ( 2 ) disturbed(2) dread(2) edgy(2) expression(2) euphoria(2) frown(2) gentleness(2) hardness(2) heart(2) hyperactive(2) impulse(2) insecurity(2) malicious(2) meditating(2) mixed(2) outgoingness(2) protective(2) rapture(2) relaxed(2) repulsion(2) responsibi1ity(2) responsiveness(2) self-concept(2) self-esteem(2) sent imental(2) softness(2) state(2) stubborness(2) successful(2) tiredness(2) turbulent(2) uncertainty(2) uncontrollable(2) understanding(2) unstable(2) uptight(2) wanting(2) weak(2) withdrawn(2) accepting(1) accompli shment(1) act ion(1) aggravation(1 ) apathy(1) ashamed(1) assertive(1) assessing(1) assurance(1) astonished(1) awareness(1) bedazzled(1) bewildered(1) bigotry(1) blank(1) blush(1) bold(1) bothering(1) b r i l l i a n t (1 ) bubbling(1 ) ' carefree(1) chagr in(1) changeable(1) choleric (1 ) circumstantial(1) comfort(1) competent (1 ) complete (1 ) contemplative(1) T a b l e 1 ( c o n t i n u e d ) craziness(1) evaluating(1 ) c r e a t i v i t y ( 1 ) evasiveness(1) curious(1) exercising(1) daydreaming(1 ) explosive(1) demanding(1 ) f a i t h O ) dependent(1 ) faithfulness(1) despise(1) flabbergasted(1) detached(1 ) fret(1) d i f f i d e n t ( 1 ) f r i v o l o u s ( 1 ) discouragement(1) f u l f i l l e d ( 1 ) disgrace(1 ) fun(1) disgruntled(1) f u r y ( 1 ) disparagement(1) generous(1) dizziness(1 ) gingery(1) doubt(1) glee(1) dreaming(1) good(1) dreamy(1) grabbing(1 ) eagerness(1 ) gratitude(1) ease(1) gregariousness(1 ) enamoured(1) grimace(1) encouraging(1) grin(1 ) endurance(1 ) grumpy(1) energetic(1 ) grunt(1) engrossed(1) harassed(1) essential(1 ) harm(1) harried(1 ) haughtiness(1) hit( 1 ) honesty(1) honor(1) horny(1) horor(1) hot(1) hugging(1) humi1iat ion(1) hunger(1) hypoactive(1) hys t e r i c a l ( 1 ) impat ience(1) impressions(1) inadequacy(1) independent(1) indi f ference(1) infatuat ion(1) injury(1 ) injury(1) i n s t i n c t i v e ( 1 ) interested(1) int imidat ion(1) Table 1 (continued) joking(1) r e f l e c t i v e ( 1 ) solemnity(1 ) j umpy(1) reluctance(1) s p i r i t u a l ( 1 ) k i l l ( 1 ) repr i s a l ( 1 ) s t a b i l i t y d ) l i s t e n i n g ( 1 ) reserve(1 ) startled(1) l i v e l y ( 1 ) resignation(1) s t i l l n e s s ( 1 ) lousy(1) r i d i c u l e ( 1 ) stoic(1 ) low(1 ) romant ic(1) stroking(1) manic (1 ) s a c r i f i c i n g ( 1 ) subdued(1) mellowness(1) sanguinity(1) suf fer ing(1) misbehavior(1) s e c u r i t y ( 1 ) sullen(1) mistrust(1) s e l f - a c t u a l i z a t i o n ( 1 ) 1 s u p e r f i c i a l ( 1 ) moved(1) sef-assured(1) suspic i o n ( 1 ) hasty(1) sef-control(1) tease(1 ) nice(1) s e l f - s a t i s f a c t i o n ( 1 ) temper(1) nostalgia(1) s e l f i s h ( 1 ) temperament(1) nurturance(1) senses(1) tin g l i n g ( 1 ) o p t i m i s t i c ( 1 ) sensual(1) togetherness(1) outraged(1 ) ser iousness(1) tolerance(1) overwhelmed(1) shock(1) transcendence(1 plac idness(1) s i l l y ( 1 ) trepidat ion(1) playing(1 ) sing(1) t r i umph(1) power(1 ) smug(1) troubled(1) pressured(1) sneak(1 ) turmoil(1) punishment(1) snob(1) tyranny(1) 36 Table One (continued) puzzled(1) uneasy(1) unresponsive(1) vengeance(1) v i t a l i t y ( 1 ) well-being(1) snort(1) unity(1) unstable(1) vibrant(1) wail(1 ) ye l l i n g ( 1 ) unbalanced(1) u n r e a l i s t i c ( 1 ) uplifted(1) vivac ious(1) watchful(1) Note: The number in parentheses i s the number of subjects who l i s t e d each item or some syntactic variant of i t . Figure 1. Per cent endorsements per emotion category. 337 + / / / / / / •3 30 + S-l o Cn 2 25 + m u § 20 15 + 10 4 5 4 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 Per Cent Endorsements Per Category 38 Table 2 Frequency of L i s t i n g in a Free Production of Exemplars Task for 20 Categories of Emotion STUDY ONE STUDY ONE HUNT & HODGE ( A l l responses ( F i r s t four (Four responses) in one minute) responses) Category Rank % Subjects Rank % Subjects Rank % Subjects Happiness 1 76.0 1 51 .0 4 29.75 Anger 2 74.5 2 49.0 5 27.75 Sadness 3 68.0 3 42.5 6 22.75 Love 4 62.0 4 38.5 2 52.25 Fear 5 48.0 5 24.5 1 53.25 Hate 6 44.5 6 19.5 3 44.50 Joy 7 41.0 7 13.0 7 1 5.75 Exc itement 8 26.5 8 6.5 10 5.00 Anxiety 9 25.0 10 3.5 9 7.00 Depression 10 21.0 9 4.0 8 7.50 Disgust 1 1 13.5 1 3 2.5 14 1 .00 Guil t 12 11.0 11.5 3.0 13 1 .25 Embarrassment 13 10.0 11.5 3.0 17 .25 Worry 14 9.5 14 1.5 15.5 .75 Envy 15 7.0 15.5 .5 12 1 .5 Pride 16 6.0 18.5 0 15.5 .75 Calmness 17 4.5 18.5 0 1 1 2.0 Boredom 18 4.0 18.5 0 19 0 Respect 19 3.5 15.5 .5 19 0 Awe 20 2.0 18.5 0 19 0 *Note: Study One N=200;Hunt & Hodge H = 400. 39 responses for each of the categories. To enable comparison with Hunt and Hodge, a second frequency count was performed on the present data, t a l l y i n g only the f i r s t four responses. In Table 2 appears frequency of l i s t i n g , raw scores and ranks, for the 20 target terms both as o r i g i n a l l y scored and based only on the f i r s t four responses. The frequency-of-l i s t i n g scores based on a l l responses correlated .97 with those based on the f i r s t four. These two scores correlated .80 and .76 with the Hunt and Hodge frequency scores. The same three correlations computed with the ranks were .98, .89, and .86, respectively. Thus, the frequency scores showed considerable r e l i a b i l i t y across scoring methods and samples of subjects. Of course, t h i s r e l i a b i l i t y occurred for a sample of 20 terms extending over a considerable range of frequency-of-listing scores(2% to 76%). A further analysis was ca r r i e d out to examine the r e l i a b i l i t y of a more r e s t r i c t e d range of emotion categories. Thus 30 items were selected that were among the top 118 ( i . e . endorsed by 2% of the sample) but not in the top 10. Thus the range of frequency scores was r e s t r i c t e d to 2% to 20%. Frequency of l i s t i n g data for the 30 emotion categories appears i n Table 3. For the 30 terms, frequency-of-listing scores based on a l l responses correlated .75 with those based on the f i r s t four. These two scores correlated .47 and .40 with the Hunt and Hodge frequency scores. The same three correlations with the ranks are .68, .51, and .51, respectively. For a l l 50 terms (these 30 plus the 20 of Table "Table 3 Frequency of L i s t i n g in a Free Production of Exemplars Task for' 30 Categories of Emotion STUDY ONE STUDY ONE HUNT & HODGE (A l l responses ( F i r s t four (Four responses) in one minute) responses) Category Rank % Subjects Rank % Subjects Rank % Subjects Frustrat ion 1 19.5 3 6.0 4.5 2.25 Crying 2 18.0 2 8.0 2 . 4.75 Feelings 3 17.5 r 1 13.0 1 2 .5 Jealousy 4 14.5 6 3.5 3 2.75 Laughter 5 13.5 4 5.0 1 6.0 Elat ion 6 13.0 5 4.0 6 6.0 Caring 7 12.0 7 2.5 16 .25 Liking 8 8.0 8 1 .5 24 0 Nervousness 9 6.5 12 .5 7.5 .75 Trust 10 6.0 12 .5 1 2 .5 Hope 1 1 4.0 12 .5 12. .5 Cheerfulness 12. 5 3.5 -12 .5 12 .5 Sensitive 12. 5 3.5 12 ' .5 12 .5 Anticipation 14. 5 2.5 23 0 24 0 Sat i s f a c t i o n 14. 5 2.5 23 0 16 .25 41 Table 3 (continued) Frequency of L i s t i n g in a Free Production of Exemplars Task for 30 Categories of Emotion STUDY ONE STUDY ONE HUNT & HODGE ( A l l responses ( F i r s t four (Four response in one minute) responses) Category Rank % Subjects Rank % Subjects Rank % Subjects Desire 18.5 2.0 23 0 24 2.25 Enthusiasm 18.5 2.0 23 0 24 0 Humor 18.5 2.0 1 2 .5 1 2 .5 Loathing 18.5 2.0 23 0 24 0 Mournful 18.5 2.0 23 0 24 0 Attract ion 23 1 .5 23 0 7.5 .75 Conf idence 23 1 .5 23 0 24 0 Dejection 23 1 .5 23 0 24 0 Humility 23 1 .5 23 0 24 0 Wonder 23 1 .5 23 0 24 0 Alert 28 1.0 23 0 24 0 Closeness 28 1 .0 23 0 24 0 Devotion 28 1.0 23 0 24 0 Gentleness 28 1.0 1 2 .5 16 .25 Tiredness 28 1.0 23 0 24 0 42 2), the frequency-of-listing scores based on a l l responses correlated .97 with those based on the f i r s t four. These two scores correlated .84 and .81 with the Hunt & Hodge frequency scores. The same three correlations computed with the ranks are . 9 1 , .87, and .83, respectively. As a f i n a l comment, within a c l a s s i c a l view of concepts, one would expect that i f a l l members are equally representative of the category, there would not be a great deal of v a r i a b i l i t y in how much instances came to mind. However, as shown by Figure 1, thi s notion was not supported in the presented study. As shown by Table 1, some instances were very borderline l o o k i n g — a finding that i s e a s i l y accounted for by the prototype view. Study Two Generation of Subordinates At the subordinate l e v e l of the hierarchy, a middle l e v e l category is further subdivided. Imagine walking into a specialty coffee store and asking for a pound of coffee. This information would be inadequate—the shopkeeper would want to know i f you were ref e r r i n g to Mocha, B r a z i l i a n Santos, Java, Columbian, and so on. The existence of categories at the subordinate l e v e l has been demonstrated for some natural language, concepts. For example, Rosch et a l . (1976b) have demonstrated that given the category table, subjects l i s t : kitchen table, dining room table; for car: sports car, 4 door sedan car, and so on. 43 Whether or not such a le v e l i s present in the domain of emotion has not been investigated. It was not clear whether or not subjects would be able to l i s t types of happiness, anger, and so on. For example, for love, one might respond with: f i l i a l love, passionate love, love for God, s e l f - l o v e , a f f e c t i o n and so on. On the other hand, what types of awe can we l i s t ? or the kinds of boredom? It seems d i f f i c u l t to generate subordinates in these cases. Perhaps some emotion categories can be further subdivided, while others cannot. Method Subjects. F i f t y - f i v e subjects participated in the study. Respondents were enrolled in psychology classes at the University of B r i t i s h Columbia. P a r t i c i p a t i o n was voluntary. Procedure. Participants were given the following instructions: This study i s part of a larger project on the sorts of things people have in mind when they hear and use words. On each page of the booklet you w i l l find the name of a general category. Your task i s to give us types of that category. For example, i f given the category CHAIR, you might respond with rocking chair, r e c l i n e r , armchair, sto o l , bean bag chair, lawn chair, or other types of chair. Or, i f given the category 44 EXERCISE, you might write jogging, body building, squash, tennis, dancing, and so on. The categories we are interested in involve emotional states, things that you can experience. For each of the psychological events l i s t e d in the booklet, please write down as many types of that experience as you can think of. For some of the categories you w i l l be able to generate many types. For others you might not be able to come up with any. Don't worry about whether your response is right or wrong. Just give us your opinion. You may now proceed to the f i r s t page. Stop after about a minute or two, or about 10 items. Then proceed to the next page, u n t i l you have completed the booklet. Subjects were given as much time as they wished to complete the exercise. Each subject generated subordinates for a l l 20 target emotions presented in random order. These were the 20 items from Study One that were l i s t e d in Table 2, which had endorsement scores ranging from 2% to 76%. 45 Results and Discussion The responses generated by subjects in thi s study were more complex and d i f f i c u l t to interpret than the responses given by subjects when asked to l i s t instances of "emotion" (Study One). There was much less consensus in the present than in the e a r l i e r task. Whereas in Study One the most popular response to "emotion" (happiness) was given by 76% of the subjects, the most popular reponse in these data was "jealousy" given by only 34.5% of the subjects for "envy". Of a l l the responses generated in Study One, 49% were idio s y n c r a t i c , compared to 96% in thi s study. In Study One 100% of a l l responses were single words, whereas in thi s study, 62% of a l l responses were single words. Phrases constituted 38% of the responses. Even where single words were l i s t e d , their interpretation was much less straightforward than in Study One. For example, "death" was l i s t e d as a type of depression. The subject did not mean to say a dead person i s a depressed person. Rather,one type of depression i s depression-caused-by-the-death-of-another. Also, some single words were probably not meant to stand alone, but to modify the o r i g i n a l category, as when " t o t a l " was given as a type of joy. One result to emerge from this study therefore, was that fewer subordinate categories of emotion are coded monolexemically in English--a result that contrasts dramatically with the Study One finding where a l l middle l e v e l categories were monolexemic. In thi s study, i f single 46 words that cannot stand alone as types of emotion or words already l i s t e d as middle l e v e l categories are excluded, monolexemic types were generated for only 8 of the 20 emotion categories. A t o t a l of 50 monolexemic types were given across the 8 emotion categories, (compared with 383 at the middle l e v e l ) , ranging from one to twelve monolexemic types per category. When considering only non-idiosyncratic responses,i.e. responses endorsed by two or more subjects, monolexemic types were produced only for 6 of the 20 emotion categories. A t o t a l of 13 non-idiosyncratic monoleximic types were generated, ranging from one to four types per emotion category. When asked to generate subordinate categories, subjects create their responses in a variety of ways, which i s apparent from the large number of idiosyncratic responses. To i l l u s t r a t e t h i s aspect of the r e s u l t s , the responses were grouped into types according to how the response was formed. The eight p r i n c i p a l ways of forming the responses were: 1. By specifying the cause or object. For example, kinds of anger can be d i f f e r e n t i a t e d according to the cause: anger caused by f r u s t r a t i o n , anger caused by jealousy. In many cases, the cause and object could not be distinguished as when "nature" was given as a response for "awe". Thus cause and object categories were combined. 2. By specifying presence of s p e c i f i c components such as cognitive or behavioral e.g. anger that includes aggression, i r r a t i o n a l anger. 47 3. By modifying the name of the o r i g i n a l category e.g. "anger" can be subdivided by giving an adjective with anger: violent anger, misdirected anger. 4. By l i s t i n g associated emotions e.g. happiness accompanies "joy". 5. By l i s t i n g monolexemic types e.g. infatuation as a type of "love" . 6. By giving synonyms of the o r i g i n a l category e.g. jealousy for "envy". 7. Miscellaneous responses—those responses that did not c l e a r l y f i t in any of categories 1-6. The responses grouped according to this c l a s s i f i c a t i o n system appear in Table 4. Only responses with a frequency of two or more were included. The number of idiosyncratic responses for each category is given. For the present purposes the attempt has been simply to present the data, without imposing any p a r t i c u l a r interpretation thereof. No responses were combined. The following should be noted concerning the grouping procedure: F i r s t , the various categories do not contrast with one another. I r r a t i o n a l anger does not contrast with violent anger—any real instance of anger could be one, the other, or both. Secondly, the syntactical and semantic form of the responses was preserved. Anger that includes aggression may be the same as violent anger. In summary, i f we take emotion as the superordinate T a b l e 4 Subordinates f o r 20 Emotion C a t e g o r i e s dOY CAUSE/OBJECT C h r i s t m a s (3). f r i e n d s h i p ( 2 ) s u c c e s s ! 2 ) MODIFIER COMPONENT SYNONYM (58 r e s p o n s e s ) * (55 f r e q = 1 ) * * happy(2) sexual j o y ( 2 ) c a r e f r e e ( 2 ) ASSOCIATED EMOTION happtness(18) l o v e ( 4 ) , e l a t 1 o n ( 2 ) e x d t e m e n t ( 2 ) g l a d ( 2 ) peace(2) (20 r e s p o n s e s ) (28 responses) (14 responses) (1 response) MONOLEXEMIC TYPES MISCELLANEOUS a l 1 v e ( 2 ) f r e e ( 2 ) warmth(2) ( 18 freq=1) (22 freq=1) (0 r e s p o n s e s ) (4 re s p o n s e s ) (0 freq=1) (1 freq=1) ( 13 freq=1) (1 freq=1) * The number 1n p a r e n t h e s e s means a t o t a l of 58 s u b o r d i n a t e c a t e g o r i e s which were c l a s s i f i e d as a c a u s e / o b j e c t of "Joy" were g e n e r a t e d . N=55. ** F i f t y - f i v e ' of the 58 s u b o r d i n a t e c a t e g o r i e s c l a s s i f i e d as a c a u s e / o b j e c t of "Ooy" were I d i o s y n c r a t i c , I.e., g e n e r a t e d by o n l y one s u b j e c t . The re m a i n i n g 3 c a t e g o r i e s a re l i s t e d i n T a b l e 4, f o l l o w e d by the number of s u b j e c t s who g e n e r a t e d t h a t r e s p o n s e . D E P R E S S I O N C A U S E / O B J E C T l o n e l i n e s s ( 5 ) a1one(2) 1so1at1on(2) ? o s s ( 4 ) weather(2) (69 r e s p o n s e s ) (64 freq=1) M O D I F I E R A S S O C I A T E D C O M P O N E N T E M O T I O N sad(6) sadness(IO) worry(4) s u 1 c i d a l ( 3 ) l o n e J y ( 8 ) c ry1ng(2) worr1ed(2) h e l p i e s s n e s s ( 4 ) t e a r s ( 2 ) S Y N O N Y M M O N O L E X E M I C T Y P E S M I S C E L L A N E O U S Incompetent (2) (60 responses) (36 responses) (14 responses) (0 r e s p o n s e s ) (0 r e s p o n s e s ) (12 r e s p o n s e s ) (57 freq=1) (33 freq=1) (11 freq=1) (0 freq=1) (0 freq=M ) (11 freq=1) CAUSE/OBJECT unknown!4) d e a t h ( 3 ) f a 1 l u r e ( 3 ) of f a 1 l u r e ( 2 ) s p 1 d e r s ( 3 ) s t r e s s ( 3 ) f a l 1 1 n g ( 2 ) of d y i n g ( 2 ) t h u n d e r ( 2 ) (89 r e s p o n s e s ) (80 freq=1) MODIFIER a n x i o u s ( 4 ) nervous(4) h e 1 p l e s s ( 2 ) ASSOCIATED EMOTION anx1ety(6) pan1c(2) (48 responses) (9 responses) (45 freq=1) (7 freq=1) FEAR COMPONENT SYNONYM worry(4) tense(2) MONOLEXEMIC MISCELLANEOUS TYPES f r 1 g h t e n e d ( 2 ) t e r r o r ( 2 ) (23 responses) (21 freq=1) (3 responses) (3 freq=1) (12 r e s p o n s e s ) (5 re s p o n s e s ) (10 freq=1) (5 freq=1) EXCITEMENT CAUSE/OBJECT MODIFIER ASSOCIATED COMPONENT SYNONYM MONOLEXEMIC MISCELLANEOUS EMOTION TYPES f u n ( 3 ) happy(7) happ1ness(7) b u t t e r f 1 l e s ( 2 ) s p o r t s ( 3 ) nervous(3) j o y ( 5 ) s u c c e s s ! 3 ) s e x u a l ( 4 ) l o v e ( 3 ) a d v e n t u r e ! 2 ) e m o t i o n a l ( 2 ) f e a r ( 2 ) a n t 1 d p a t 1 o n ( 2 ) f e v e r 1sh(2) t h r l l 1 ( 2 ) e x p e c t a t i o n s ! 2 ) mus1c(2) p a r t y ( 8 ) (80 r e s p o n s e s ) (45 r e s p o n s e s ) ( 16 responses) (39 responses) (0 responses) (0 r e s p o n s e s ) (10 r e s p o n s e s ) (72 freq=1) <\ (40 freq=1) (11 freq=1) (38 freq=1) (0 freq=1) (0 freq=1) (10 freq=1) ANGER CAUSE/OBdECT f r u s t r a t 1 o n ( 6 ) j e a l o u s y ( 3 ) temper(3) argument(2) h u r t ( 2 ) (49 r e s p o n s e s ) (44 freq=1) MODIFIER ' ASSOCIATED COMPONENT SYNONYM EMOTION I r r a t i o n a l ( 4 ) u p s e t ( 5 ) a g g r e s s i o n ( 3 ) mad(10) r a t i o n a l anger(2) h a t r e d ( 4 ) madness(4) v i o l e n t anger(2) f e a r ( 3 ) d 1 s g u s t ( 2 ) h a t e ( 2 ) unhappy(2) (58 responses) (14 responses) (12 responses) (3 respo n s e s ) MONOLEXEMIC TYPES f u r y ( 3 ) rage(3) MISCELLANEOUS (55 freq=1) (8 freq=1) (11 freq=1) (1 freq=1) (14 r e s p o n s e s ) (4 r e s p o n s e s ) ( 12 freq=1) (4 freq=1 ) ANXIETY CAUSE/OBJECT s t r e s s ( 3 ) t e ns1on(2) (60 r e s p o n s e s ) (58 freq=1) MODIFIER ASSOCIATED COMPONENT EMOTION worr1ed(5) f e a r ( 7 ) n e r v ous(4) u p s e t ( 7 ) anx1ous(2) f r u s t r a t i o n o e x c 1 t e d ( 2 ) apprehens1on(2) frown1ng(2) f e a r f u l ( 2 ) h e l p 1 e s s n e s s ( 2 ) sweat1ng(2) (42 responses) (21 responses) (39 responses) (37 freq=1) (16 freq=1) (34 freq=1) SYNONYM worry(8) nervousness!7 ) tense(3) (0 r e s p o n s e s ) (0 freq=1) MONOLEXEMIC TYPES MISCELLANEOUS (0 r e s p o n s e s ) (5 r e s p o n s e s ) (0 freq=1) (5 freq=1) 0 G U I L T C A U S E / O B J E C T c h e a t i n g ! 3) f a 11ure ~s 1 n (2 ) wrong!2) (74 r e s p o n s e s ) (70 freq=1) M O D I F I E R emot l o n a l ( 2 ) u n l a w f u l ( 2 ) (32 -responses (30 freq=1) A S S O C I A T E D C O M P O N E N T S Y N O N Y M E M O T I O N remorse(3) c o n s c l e n c e ( 2 ) embarrass(2) r e g r e t ( 2 ) sorrow(2) ) (22 responses) (21 resp o n s e s ) <0 r e s p o n s e s ) (18 freq=1) (20 freq=1) (0 freq=1) M O N O L E X E M I C M I S C E L L A N E O U S T Y P E S pun1sh(2) u n t r u s t w o r t h y ( 2 ) (0 r e s p o n s e s ) (14 r e s p o n s e s ) (0 freq=1) (12 freq=1) ENVY CAUSE/OBJECT b e a u t y ( 2 ) 1ndependence( 2) (41 r e s p o n s e s ) (39 freq=1) MODIFIER ASSOCIATED COMPONENT EMOTION hate(6) anger(3) d l s l 1 k e ( 3 ) adm1 r a t 1on( 2) angry(2) (26 responses) (17 responses) (27 responses) (26'freq=1) (12 freq=1) (23 freq=1) des1re(4 ) 1ong1ng(3 ) want(3) w1shfulness(2] SYNONYM MONOLEXEMIC TYPES j ea1ousy(19 ) j e a l o u s ( 4 ) MISCELLANEOUS gre e n ( 3 ) (4 r e s p o n s e s ) (O respo n s e s ) (6 r e s p o n s e s ) (2 freq=1) (0 freq=1) (5 freq=1) E M B A R R A S S M E N T C A U S E / O B J E C T b e i n g wrong(2) (63 r e s p o n s e s ) (62 freq=1) M O D I F I E R A S S O C I A T E D C O M P O N E N T S Y N O N Y M E M O T I O N f o r o t h e r s ( 2 ) shame(6) b l u s h ( 4 ) p e r s o n a l ( 2 ) g u l l t O ) b1ush1ng(2) sma11(2) r e d - f a c e d ( 2 ) s e l f - c o n s c 1 o u s n e s s ( 2 ) (40 responses) (10 responses) (40 responses) (0 r e s p o n s e s ) (37 freq=1) (8 freq=1) (36 freq=1) (0 freq=1) M O N O L E X E M I C T Y P E S M I S C E L L A N E O U S (0 r e s p o n s e s ) (6 r e s p o n s e s ) (0 freq=1) (6 freq=1 ) DISGUST CAUSE/OBJECT p r o s t 1 t u t 1 o n ( 2 ) t o t a l ( 2 ) (48 r e s p o n s e s ) (47 freq=1) nausea(3) d 1 s r e s p e c t ( 2 ) d 1 s t a s t e ( 2 ) s1ck(2) MODIFIER ASSOCIATED COMPONENT SYNONYM EMOTION " hate(6) d1sda1n(4) h a t r e d ( 4 ) d 1 s l 1 k e ( 3 ) l o a t h l n g ( S ) r e p u l s 1 o n ( 3 ) anger(2) d e s p l s e ( 2 ) d1sappo1nted( 2) repe11ed(2) r e v o l t ( 2 ) turned o f f ( 2 ) (18 responses) (31 responses) (25 responses) (0 responses) MONOLEXEMIC TYPES MISCELLANEOUS (17 f req = 1 ) (19 freq=1) (21 freq=1) (0 freq=1) (O r e s p o n s e s ) (11 r e s p o n s e s ) (0 freq=1) (11 freq=1) CALM CAUSE/OBJECT l a k e s ( 2 ) n a t u r e ( 2 ) (37 r e s p o n s e s ) (35 freq=1) MODIFIER re1axed(9) c o n t e n t ( 3 ) happy(2) ASSOCIATED EMOTION p e a c e f u l ( 5 ) peace(4) contentment(2) not e x d t e d ( 2 ) (33 responses) (9 responses) (30 freq=1) (5 freq=1) COMPONENT re1axat1on(6) q u l e t ( 5 ) s1eep(3) c o m f o r t a b l e ( 2 ) c o n t r o l l e d ( 2 ) q u l e t n e s s ( 2 ) s e c u r 1 t y ( 2 ) " t o g e t h e r " ( 2 ) (51 responses) (43 freq=1) SYNONYM MONOLEXEMIC MISCELLANEOUS TYPES s e r e n 1 t y ( 7 ) serene!5) tranqu11(3) tranqu111ty(2 ) s o l 1 t u d e ( 2 ) (6 r e s p o n s e s ) (2 freq=1) (0 r e s p o n s e s ) (10 r e s p o n s e s ) (O freq=1) (9 freq=1) BOREDOM c a u s e / o b j e c t M O D I F I E R A S S O C I A T E D E M O T I O N n o t h i n g t o do(5) I 1 s t l e s s ( 2 ) no e x d tement (2 ) rep 1 t 1 o u s ( 2 ) no one t o t a l k t o ( 2 ) t o t a l ( 2 ) r o u t 1ne(2) s c h o o l ( 2 ) (62 r e s p o n s e s ) (57 freq=1) u n 1 n t e r e s t e d ( 2 ) COMPONENT t 1 r e d ( 5 ) 1azy(3) SYNONYM d1 s 1 n t e r e s t e d ( 2 ) 1ethargy(2) 1 1 f e l e s s ( 2 ) ^ s1eepy(2) • _ un1nterest1ng(2) unst1mulated(2) yawn1ng(2) (32 responses) (15 responses) (41 responses) (1 response) (28 freq=1) (15 freq=1) (32 freq=1) (1 freq=1) M O N O L E X E M I C T Y P E S M I S C E L L A N E O U S d u l 1 ( 4 ) d r a b ( 2 ) (0 responses) (12 r e s p o n s e s ) (O freq=1) (10 freq=1) AWE CAUSE/OBJECT MODIFIER ASSOCIATED COMPONENT SYNONYM MONOLEXEMIC EMOTION TYPES n a t u r e ! 2 ) r e s p e c t f u l ( 3 ) amazement(8) r e s p e c t ( 7 ) r e l 1 g 1 o u s ( 2 ) s u r p r 1 s e ( 6 ) f 1 a b b e r g a s t e d ( 3 ) envy(3) shock(3) wonder(4) s p e e c h l e s s ( 3 ) bew11derment(2) stunned(3) f ! u s t e r e d ( 2 ) dumb-struck(2) 1mpressed(2) shocked(2) -j e a l o u s y ( 2 ) overwhelmed!3) wonderment(2) worsh1p(2) (45 r e s p o n s e s ) (15 responses) (27 responses) (33 responses) (0 r e s p o n s e s ) (0 r e s p o n s e s ) (44 freq=1) (13 freq=1 ) (18 freq=1) (24 freq=1) (0 freq=1) (0 freq=1) MISCELLANEOUS (5 freq=1) HAPPINESS CAUSE/OBJECT f r i e n d s ( 6 ) f a m i l y ( 4 ) b e i n g 1n l o v e ( 2 ) f u n ( 2 ) sunsh1ne(2) t r a v e l 11ng(2) (74 r e s p o n s e s ) (68 freq=1) laugh1ng(4) 1aughter(3) MODIFIER ASSOCIATED COMPONENT SYNONYM EMOTION j o y f u l ( 2 ) l o v e(14) Joy(10) a f f e c t 1 o n a t e ( 2 ) sm111ng(3) n a t u r a l h1gh(2) d M n M n g ( 2 ) peace(2) f r e e ( 2 ) r e l a x a t i o n ( 2 ) secure!2) s e c u r ! t y ( 2 ) sex(2) s1ng!ng(2) (39 responses) (25 responses) (43 responses) (0 respo n s e s ) (38 freq=1) (20 freq=1) (33 freq=1) (0 freq=1) MONOLEXEMIC MISCELLANEOUS TYPES contentment(5) warmth(3) c o n t e n t ( 3 ) c h e e r f u l ( 2 ) p i e a s u r e ( 2 ) (15 responses) (3 r e s p o n s e s ) (11 freq=1) (2 freq=1) SADNESS CAUSE/OBJECT V d e a t h ( 7 ) 1 o n e l i n e s s ( 6 ) 1oss(5) f a l l u r e ( 4 ) 1 o n e l y ( 3 ) (50 r e s p o n s e s ) (45 freq=1) MODIFIER long-term(2) s h o r t - t e r m ( 2 ) ASSOCIATED COMPONENT EMOTION d e p r e s s i o n ) 1 1 ) t e a r s ( 6 ) d e pressed(6) sorrow(5) unhappy(5) melancholy(4) g r i e f ( 2 ) SYNONYM cry1ng(6) frown(3) c r y ( 2 ) h e 1 p l e s s n e s s ( 2 ) mourn1ng(2) unhapp1ness(2) t 1 r e d n e s s ( 2 ) u p s e t ( 2 ) (27 responses) (36 responses) (18 responses) (0 respo n s e s ) MONOLEXEMIC TYPES MISCELLANEOUS (25 freq=1) (28 freq=1) (11 freq=1) (0 freq=1) (0 r e s p o n s e s ) (5 r e s p o n s e s ) (0 freq=1) (5 freq=1) RESPECT CAUSE/OBJECT p a r e n t s ( 4 ) I n t e l 11gence(3) a u t h o r 1 t y ( 2 ) e l d e r s ( 2 ) f o r o t h e r s ( 2 ) f r i e n d l 1 n e s s ( 2 ) f r l e n d s ( 2 ) f r i e n d s h i p ( 2 ) k i n d n e s s ( 2 ) m y s e l f ( 2 ) o l d e r p e o p l e ( 2 ) p e e r s ( 2 ) s e l f ( 2 ) (73 r e s p o n s e s ) (60 f r e q =1) MODIFIER p a r e n t a l ( 2 ) (28 responses) (27 freq=1) ASSOCIATED EMOTION lo v e ( 6 ) honour(4) admlrat1on( 2) awe(2) COMPONENT SYNONYM calm(2) cons i d e r a t l o n ( 2) t r u s t ( 2 ) u nderstanding( 2) (23 responses) (30 responses) (19 freq=1 ) (2G freq=1 ) (0 r e s p o n s e s ) (0 freq=1) MONOLEXEMIC TYPES MISCELLANEOUS (0 responses) (4 r e s p o n s e s ) (O freq=1) (4 freq=1) LOVE CAUSE/OBJECT f r 1 e n d s ( 4 ) n a t u r e ( 2 ) p e t s ( 2 ) (43 r e s p o n s e s ) (40 freq=1) MODIFIER ASSOCIATED COMPONENT EMOTION f a m i l y ( 6 ) h a p p l n e s s ( 5 ) sex(6) fr1endsh1p(6) car1ng(4) l u s t ( 4 ) b r o t h e r 1y(5) happy(3) r e s p e c t ( 4 ) p h y s i c a l ( 5 ) euphor1a(2) s e c u r 1 t y ( 4 ) p a r e n t a l l o v e ( 3 ) t o g e t h e r n e s s ( 4 ) puppy(3) understand 1ng( 3) romant1c(3) c l o s e n e s s ! 2 ) s e l f ( 3 ) c o n f 1 d e n t ( 2 ) u n r e q u l t e d ( 3 ) consIderat1on( 2) c o n s t a n t ( 2 ) h o l d i n g hands(2) j o y f u l ( 2 ) k1ss1ng(2) pass1onate(2) marr1age(2) s1bl1 n g l o v e ( 2 ) pass1on(2) r e l a x e d ( 2 ) shar1ng(2 ) to g e t h e r ( 2 ) t r u s t ( 2 ) (S5 responses) (32 responses) (83 responses) (52 freq=1) (28 freq=1) (G6 freq=1) SYNONYM MONOLEXEMIC MISCELLANEOUS TYPES a f f e c t 1 o n ( 4 ) warmth(4) 1nfatuat1on(4) a11ve(2) romance(2) h e a r t ( 2 ) warm!2) (4 res p o n s e s ) (14 r e s p o n s e s ) (1 freq=1 ) (10 freq=1 ) PRIDE CAUSE/OBJECT MODIFIER accompl1shment(5) p e r s o n a l ( 2 ) a c h i e v e m e n t ( 2 ) s e l f - p r 1 d e ( 2 ) t a l e n t ( 2 ) (68 r e s p o n s e s ) (65 freq=1) ASSOCIATED EMOTION happy(3) p i e a s e d ( 2 ) proud(2) COMPONENT SYNONYM s e l f - e s t e e m ( 3 ) a r r o g a n c e ) 2 ) conf1dencet2) ego(2) (46 responses) (12 responses) (26 responses) (O responses) (44. freq=1) (9 freq=1) (22 freq=1) (O freq=1) MONOLEXEMIC TYPES c o n c e 1 t ( 3 ) MISCELLANEOUS (4 r e s p o n s e s ) (8 r e s p o n s e s ) (3 freq=1) (8 freq=1) p WORRY CAUSE/OBJECT s t r e s s ( 6 ) exams(5) 1 l l n e s s ( 3 ) f r 1 e n d s ( 2 ) f u t u r e ( 2 ) (65 r e s p o n s e s ) (60 freq=1) nervous(3) na1 1. b1t1ng(2) nervous(3) MODIFIER ASSOCIATED COMPONENT SYNONYM EMOTION f e a r f u 1 ( 2 ) anx1ety(11) nagg1ng(2) f e a r ( 8 ) u p s e t ( 3 ) anx1ous(2) c o n c e r n ( 2 ) depress1on(2) f r u s t r a t 1 o n ( 2 ) r e g r e t ( 2 ) s c a r e d ( 2 ) uneas1ness(2) (19 responses) (19 responses) (46 responses) (1 response) MONOLEXEMIC TYPES MISCELLANEOUS (17 freq= 1) (9 freq=1) (43 freq=1) (1 freq=1) (0 responses) (2 r e s p o n s e s ) (0 freq=1) (2 freq=1) HATRED CAUSE/OBdECT (48 r e s p o n s e s ) (48 freq=1) pass1on(2) d e s t r u c t 1 o n ( 2 ) v 1 o l e n c e ( 2 ) MODIFIER ASSOCIATED COMPONENT SYNONYM EMOTION s e 1 f ( 2 ) a n g e r O ) desp1se(5) mad(4) angry(3) envy(3) resentment(3) b l t t e r n e s s ( 2 ) d e t e s t ( 2 ) J e a l o u s y ( 2 ) l o a t h e ( 2 ) rage(2) r e p u l s 1 o n ( 2 ) (41 responses) (30 responses) (29 responses) (0 respo n s e s ) (40 freq=1) (18 freq=1) (2G freq=1) (0 freq=1) MONOLEXEMIC TYPES d l s l 1 k e ( 8 ) MISCELLANEOUS (2 res p o n s e s ) (2 r e s p o n s e s ) ( 1 freq=1) (2 freq=1) 68 l e v e l , subjects can generate a middle level with great ease. The English language provides several hundred single terms that name types of emotion. Subjects may be able to describe a subordinate l e v e l , although with greater d i f f i c u l t y . English provides few single terms that c l e a r l y f a l l at th i s l e v e l . So, subjects must be more inventive in describing types of love, hate, anger, happiness, and so on. As a consequence, their responses tend to be highly i d i o s y n c r a t i c . PART II INTERNAL STRUCTURE OF CATEGORIES One of the major d i s t i n c t i o n s between the c l a s s i c a l view of concepts and the prototype view concerns internal structure. Internal structure refers to that general class of conceptions of categories in which categories are composed of a core meaning and in which items within the category may be considered d i f f e r e n t i a l l y representative of the meaning of the category term. Representativeness may be operationally defined by means of subject's ratings of how good an example an item i s of i t s category. Rosch has obtained inter-subject consistency in such ratings. Individual subjects agree that some exemplars of a category are more representative than others, and diff e r e n t subjects choose the same examples as most representative of the category. Representativeness of items within a category has been shown to af f e c t many of the dependent variables used in psychological research. Most of the studies have focused on common semantic categories(e.g. "dog", "furniture", e t c . ) . 69 Rosch has used such measures as the f o l l o w i n g : speed of p r o c e s s i n g , f r e e p r o d u c t i o n of exemplars, n a t u r a l language use of category terms, asymmetries i n s i m i l a r i t y r e l a t i o n s h i p s between category exemplars, and l e a r n i n g and development. Speed of p r o c e s s i n g ( r e a c t i o n time) has been i n v e s t i g a t e d e x t e n s i v e l y i n category v e r i f i c a t i o n t a s k s . T y p i c a l l y s u b j e c t s are asked to v e r i f y statements of the form "An {exemplar} i s a {category name}" as q u i c k l y as p o s s i b l e . Response times are s h o r t e r f o r v e r i f i c a t i o n of the category membership of r e p r e s e n t a t i v e exemplars than n o n r e p r e s e n t a t i v e exemplars. Rosch et a l . (1976) have a l s o demonstrated t h i s e f f e c t f o r three types of a r t i f i c i a l c a t e g o r i e s , where r e p r e s e n t a t i v e n e s s was d e f i n e d by f a m i l y resemblance, by mean values of a t t r i b u t e s , or by degree of d i s t o r t i o n from the pr o t o t y p e . These d i f f e r e n c e s i n response times were a m p l i f i e d when a prime ( p r i o r mention of the category name) wasprovided. Priming reduced response times to v e r i f y the category membership of r e p r e s e n t a t i v e exemplars but inc r e a s e d response times to v e r i f y the membership of n o n r e p r e s e n t a t i v e exemplars. Order and p r o b a b i l i t y of exemplar p r o d u c t i o n have been i n v e s t i g a t e d p r i m a r i l y f o r su p e r o r d i n a t e semantic c a t e g o r i e s . Frequency of mention of an exemplar i s s i g n i f i c a n t l y c o r r e l a t e d with degree of r e p r e s e n t a t i v e n e s s (Mervis et a l . 1976). N a t u r a l languages possess mechanisms f o r coding 70 g r a d i e n t s of r e p r e s e n t a t i v e n e s s . For example, languages g e n e r a l l y i n c l u d e such q u a l i f y i n g terms ("hedges") as "roughly speaking" and " s o r t o f " ( L a k o f f , 1973). Rosch (1975) has shown that when s u b j e c t s are given sentence frames such as "{X} i s v i r t u a l l y {Y}", they r e l i a b l y p l a c e the more r e p r e s e n t a t i v e member of the p a i r i n the r e f e r e n t {Y} s l o t . S u b s t i t u t a b i l i t y of the sup e r o r d i n a t e i s another measure of exemplar r e p r e s e n t a t i v e n e s s . Rosch (1978) found that r e p r e s e n t a t i v e n e s s r a t i n g s f o r members of supero r d i n a t e c a t e g o r i e s p r e d i c t the extent to which the member i s s u b s t i t u t a b l e f o r the supero r d i n a t e word i n many commonly encountered sentences. For example, i n the sentence "A bowl of f r u i t makes a n i c e c e n t e r p i e c e " , "apples", but not "watermelon" produces a sentence which, s u b j e c t s agree, r e t a i n s i t s n a t u r a l n e s s and t r u t h v a l u e . Asymmetry in s i m i l a r i t y r a t i n g s between members that vary i n r e p r e s e n t a t i v e n e s s i s another way in which members of a category f a i l to be e q u i v a l e n t . Rosch(l975) has shown that l e s s r e p r e s e n t a t i v e exemplars are more s i m i l a r to more r e p r e s e n t a t i v e exemplars than v i c e - v e r s a . For example, s u b j e c t s f e e l that penguins are more s i m i l a r to robins than r o b i n s are to penguins. In the l e a r n i n g and development of c a t e g o r i e s , r e p r e s e n t a t i v e n e s s appears to be a major v a r i a b l e . Representativeness g r a d i e n t s have two b a s i c i m p l i c a t i o n s f o r category a c q u i s i t i o n . The f i r s t i s that category membership i s e s t a b l i s h e d f i r s t f o r the most r e p r e s e n t a t i v e exemplars 71 and l a s t for the least t y p i c a l instances. Rosch(1973,a,b) found that focal (representative) colors and forms were learned more quickly than nonfocal colors and forms by persons whose language did not contain e x p l i c i t labels for these categories. In the next set of studies to be reported, the internal structure of the category emotion was investigated. When we ask i f the concept of emotion is i n t e r n a l l y structured, we are asking whether a l l instances have equal status or some instances of the concept are thought of as "better representatives" than others. From the Study One results we can speculate that emotions l i k e happiness, anger, sadness, and love would be prototypical exemplars, while examples l i k e boredom, awe, and respect would be less representative of the category. If t h i s i s the case, we might assume that the concept of emotion i s organized i n t e r n a l l y ; furthermore, we would expect the internal organization to a f f e c t performance on a variety of tasks. In t h i s project, five measures of internal structure were used. Study Three examined ratings of goodness of example (prot o t y p i c a l i t y ratings) for 50 of the emotion terms generated in Study One. Study Four examined reaction time in a category membership v e r f i c a t i o n task. It was hypothesized that response latency would be increased as a function of deviation from the prototype. Study Five examined whether or not "emotion" was e l i c i t e d as the superordinate more often for prototypical than nonprototypical instances. For many 72 natural language categories there is a greater l i k e l i h o o d that the name of a prototypical category member w i l l e l i c i t the superordinate category name than w i l l a less t y p i c a l exemplar (Rosch, 1973). Study Six examined the "family resemblances" of the 20 target emotions. It was expected that the prototypical instances would have a greater degree of resemblance to the "emotion" family than the less t y p i c a l . Should these measures converge, predictions can be generated with some confidence concerning performance on the other measures of internal structure used by Rosch. For example, prototypical emotions could be expected to f i t more ea s i l y than nonprototypical emotions in a sentence such as "Pat became overly emotional". Study Three P r o t o t y p i c a l i t y Ratings Method Subjects. Subjects were 55 students enrolled in an Introductory Psychology class at the University of B r i t i s h Columbia. P a r t i c i p a t i o n was voluntary. Procedure. The following instructions, borrowed from Rosch(l973), were provided: This study has to do with what we have in mind when we hear and use words. Let's consider the word "red". Close your eyes and imagine a true red. Now imagine an orangish red. Imagine a purplish red. Although you might s t i l l name the orange-73 red or the purple-red with the term "red", they are not as good examples of "red" (not as clear cases of what red refers to ) as the clear, true red. Orange and purple are even poorer examples of "red", perhaps not even red at a l l . Notice that to judge how good an example something i s has nothing to do with how much you l i k e the thing. You might prefer a purple red or purple to a true red, but s t i l l recognize which is the better example of "red". The word we are interested in i s "emotion". We are interested in which experiences or feelings are good or poor examples of "emotion". On the following page is a l i s t of things that you can fee l or experience--things l i k e hunger, happiness, anger, and dizziness. We would l i k e you to rate the extent to which each feeling on the l i s t i s a good or poor example of "emotion". Don't worry about why you think something i s or isn't a good example—just give us your opinion. Subjects rated each of the 20 target emotions (taken 74 from Table 2) on a scale of 1-6 where a score of 1 indicated extremely poor example and 6 meant extremely good example of an emotion. Subjects were given as much time as they wished to complete the task. The order in which the terms were presented was reversed for half the subjects. The same questionnaire was re-administered fiv e months later (in the second term) in order to assess r e l i a b i l i t y over time. At this time p r o t o t y p i c a l i t y ratings were also obtained for an additional 30 emotions terms (taken from Table .3). In order to ensure anonymity while allowing i d e n t i f i c a t i o n of the questionnaires, respondents were asked to print their mother's maiden name on the response sheet. Results and Discussion Mean p r o t o t y p i c a l i t y ratings for the target 20 emotions terms at times 1 and 2 appear in Table 5. The Time 1 t y p i c a l i t y ratings correlated .97 with the Time 2 ratings. P r o t o t y p i c a l i t y ratings for the 30 additional terms appear in Table 6. An intraclass c o r r e l a t i o n c o e f f i c i e n t was computed to assess inter-rater r e l i a b i l i t y across a l l 50 terms, ICC=.96, indicating that there i s high agreement among subjects on ratings of p r o t o t y p i c a l i t y , a result that would not have been expected had the items been equally exemplary of the category. The mean inte r - r a t e r correlation c o e f f i c i e n t was .38. I n i t i a l l y i t seemed quite possible that rating the extent to which an emotion is a good example of the general category would prove a meaningless task to subjects. It 75 Table 5 Pro t o t y p i c a l i t y Ratings for 20 Emotion Categories PROTOTYPICALITY TIME 1 TIME 2 Category Mean Rank Mean Rank Love 5.46 1 5.27 2 Anger 5.15 2 5.36 1 Hate 5.26 3 5.04 3 Sadness 5.04 4 4.49 9 Happiness 5.00 5 4.51 8 Fear 4.78 6 5.00 4 Depression 4.73 7 4.58 6 Joy 4.89 8 4.93 5 Exc i tement 4.58 9 4.47 10 Gui l t 4.55 10 4.55 7 Embarrassment 4.36 1 1 4.31 1 2 Envy 4.13 1 2 4.26 13 Anxiety 4.29 13 4.44 1 1 Worry 3.84 14 3.96 1 4 Di sgust 3.71 1 5 3.89 15 Awe 3.46 16 3.24 1 7 Pride 3.33 1 7 3.51 1 6 Calm 2.75 18 2.82 18 Boredom 2.71 19 2.76 19 Respect 2.49 20 2.51 20 Note: Ratings were made on a scale from 1=extremely poor exai to 6=extremely good example. Time 1 N=55; Time 2 N=53 corr e l a t i o n between mean ratings at Time 1 and Time 2 was .97. Table 6 Pro t o t y p i c a l i t y Ratings for 30 Emotion Categories Category Jealousy Elat ion Frustrat ion Mournful Cheerfulness Loathing Enthusiasm Feelings Desire Dejection Crying Nervousness Anticipation Laughter Caring Mean Rating Rank 4.82 1 4.31 2 4.07 3 4.00 4 3.95 5 3.84 6 3.75 7 3.73 8 3.73 9 3.55 1 0 3.51 1 1 3.47 1 2 3.35 1 3 3.22 1 4 3.16 15 Category Hope Sat i sfact ion Liking Humility Attraction Closeness Sensitive Conf idence Humor Trust Gentleness Devotion Wonder Alert Tiredness Mean Rat ing Rank 3. 1 3 1 6 3. 1 1 17 3. 09 1 7 3. 06 18 3. 02 20 2. 91 21 2. 87 22 2. 86 23 2. 80 25 2. 69 25 2. 58 26 2. 53 27 2. 1 6 28 2. 1 6 29 2. 15 30 77 seemed plausible that they would question the c r e d i b i l i t y of the task, refuse to complete the ratings, or resort to some other judgment such as their preference for each emotion. Further, i t seemed plausible that even i f subjects performed the task, their responses would be highly idiosyncratic because of the subjective nature of emotional experiences. However, subjects found i t meaningful to rate the extent to which an instance was a good example of a category. This i s in and of i t s e l f a noteworthy finding. Moreover, as shown by the inter-rater r e l i a b i l i t y , subjects agree with each other about their responses. This study provides evidence that the category emotion i s thought to be i n t e r n a l l y structured. There are prototypes (clearest cases, best examples) of the category with instances varying from better to poorer examples. A l l emotions are not equal. Study Four Reaction Time as a Measure of Internal Structure If one accepts the notion that in some categories members are not equally representative of the category, one would predict that the extent to which a category member represents the category w i l l a f f e c t the time required to v e r i f y i t s membership. The hypothesis under consideration in this study was that subjects would respond "true" more 78 quickly to a statement of the form "A {member} i s a {category}" when the member i s a central example (a prototypical case) than when the member i s a peripheral or nonprototypical exemplar. Method Subjects. Subjects were 30 students from various classes at the University of B r i t i s h Columbia. P a r t i c i p a t i o n was voluntary. Materials. Subjects were required to respond "True" or "False" to 80 statements of the form "{X} is a/n {Y}". The pool of statements consisted of: 1) Ten true central Emotion statements such as "Anger is a/n emotion". 2) Ten true peripheral Emotion statements such as "Awe is a/n emotion". 3) Ten false central Emotion statements such as "Joy is a/n clothing". 4) Ten false peripheral Emotion statements such as "Respect is a/n f r u i t " . 5) Ten true Central Vehicle statements such as "Truck is a/n vehicle". 6) Ten true peripheral Vehicle statements such as "Wagon i s a/n vehicle". 7) Ten false central Vehicle statements such as "Car is a/n t o o l " . 8) Ten false peripheral Vehicle statements such as "Carriage 79 is a/n b i r d " . Emotion terms were the 20 l i s t e d in Table 2. The ten central terms were the ten terms with the highest p r o t o t y p i c a l i t y ratings in Study Three. Conversely, the ten peripheral emotions were the ten terms with the lowest p r o t o t y p i c a l i t y ratings. The vehicle statements were included in order to replicate Rosch's (1973) findings and to provide " f i l l e r " items for the task. A computer program was designed to present the 80 statements in a d i f f e r e n t random order for each subject. The following instructions were presented on the screen of a TRS-80 Radio Shack Computer: This i s a study of the 'belongingness' of items in categories. You w i l l be presented with a series of statements of the form 'X i s a/n Y'. Your task is to respond TRUE or FALSE to each statement as fast as you can. So, for example, i f the sentence 'Apple is a/n f r u i t ' appeared on the screen, you would press the "1" key for TRUE. If the statement 'Apple i s a/n clothing' appeared, you would press the "2" key for FALSE. Try to respond as quickly and as accurately as you can. 80 Procedure. Subjects were ushered into the laboratory where the instructions for the task were displayed on the computer terminal. Subjects were told to read the instructions and press the <ENTER> key i f they understood the instructions and were ready to begin. The word READY appeared before each sentence to a l e r t the subject that a sentence would be displayed. Subjects took approximately 15-20 minutes to complete the task. Results and Discussion A response was considered correct i f the subject responded "True" to a True central or peripheral statement, and "False" to a False central or peripheral statement. Conversely, a response was considered incorrect i f the subject answered "False" to a True sentence or "True" to a False sentence. The decision to consider a response correct was in some sense an arbit r a r y one, in that one could argue that the peripheral category members might in fact not be members of the category emotion, but might rather belong to some neighboring category. The rationale for regarding the peripheral exemplars as category members was as follows: F i r s t , each of the terms used was given as a member of the category emotion (by at least 2% of the sample) in the Study One free l i s t i n g s i t u a t i o n . Secondly, when the target terms were rated for goodness of membership (Study 3), the lowest p r o t o t y p i c a l i t y rating received by any of the peripheral terms was 2.49 (respect) on a scale of 1-6, where 1 meant 81 "extremely poor example", and 6 meant "extremely good example". The fact that the lowest rating f e l l between "poor example" and " s l i g h t l y poor example" on the scale suggests that the peripheral category members were indeed considered category members, a l b e i t poor ones. Reactions times were analyzed only for correct responses. Mean reaction times and number correct for each of the emotion and vehicle statements appear in Table 7. Dependent-sample t-tests revealed a s i g n i f i c a n t difference between reaction times to prototypical' and nonprototypical true emotion statements, t_(29)=2.69, p<.0l, one-tailed, such that sentences containing central members were responded to more quickly. The difference between prototypical and nonprototypical false emotion statements was not s i g n i f i c a n t , t(29) = 1..3l, p>.lO. The difference between the number correct for prototypical and nonprototypical true emotion statements was s i g n i f i c a n t , t(29)=2.34, p<.05. Subjects responded "True" more often to True central statements than they did to True peripheral statements. For the vehicle statements, a s i g n i f i c a n t difference was found between prototypical true and nonprototypical true vehicle 'statements, t_( 29) = 1 .86., p<.05, one-tailed. The difference between prototypical and nonprototypical false statements was not s i g n i f i c a n t , t(29)=1.45, p>.10. The number correct d i f f e r e d between prototypical and nonprototypical true vehicle statements t(29)=6.67, p<.000l. The results were as hypothesized. Subjects took longer 82 Table 7 Mean Reaction Time in a Category V e r i f i c a t i o n Task Mean Reaction Mean Time Number Emotion Statements (msec) Correct True Central 1193 8.3 True Peripheral 1361 7.0 False Central 1541 7.6 False Peripheral 1626 7.7 Mean Reaction Mean Time Number Vehicle Statements (msec) Correct True Central 1257 9.3 True Peripheral 1425 6.3 False Central 1306 7.8 False Peripheral " 1380 8.0 Note: Maximum possible number correct i s 10.00. 83 to v e r i f y the truth of a statement l i k e "Awe i s an emotion" than a sentence l i k e "Anger i s an emotion". Central members of the category emotion were i d e n t i f i e d as such more quickly, while more deliberation occurred before peripheral membership was v e r i f i e d . This finding was corroborated in the number correct r e s u l t s . Subjects made more errors in the case of nonprototypical category examplars. The same pattern of results emerged for the vehicle statements, a condition designed to replicate Rosch's findings. Subjects took longer to v e r i f y the truth of a statement l i k e "Wagon is a vehicle" than a sentence l i k e "Truck i s a vehicle". The most errors were made when nonprototypical true statements were presented. Fewest errors were made when prototypical true statements were presented. The fact that subjects made the most errors when nonprototypical true statements were presented suggests that subjects were unsure of, and therefore hesitant to v e r i f y the category membership of the peripheral cases, while that of the central members was not contested. These results suggest that sentences of the form "X i s an emotion" cannot be taken, as the c l a s s i c a l view would have i t , as absolutely true or f a l s e . Rather, their truth seems to be a matter of degree. Study Five Generating the Superordinate Name Rosch (1973) found that more central members ( i . e . good 84 examples of the category) were more l i k e l y to e l i c i t the superordinate category name than were peripheral category members. The primary purpose of t h i s study, therefore, was to discover i f "emotion" would be given as the superordinate name more often for the terms that had been rated as highly t y p i c a l than those which were regarded as less exemplary of the category. Another purpose of t h i s study was to discover what other kinds of superordinates are generated for the 2 0 target emotion terms. Method Subjects. Subjects were 1 2 0 students enrolled in various psychology classes at the University of B r i t i s h Columbia. Pa r t i c i p a t i o n was voluntary. Procedure. The task was described as follows: This study i s part of a larger project on the sorts of things that people have in mind when they hear and use words. On th i s questionnaire, we are interested in the general categories to which things might belong. We w i l l give you a word and you w i l l give us the general category. For example, i f given the word 'truck', you might write in 'vehicle' or 'motor vehicle'. For the word 'polio', you might respond with 'disease' or ' i l l n e s s ' . The l i s t below refers to things you 85 can experience. For each of the items, your task i s to provide the general category to which i t belongs. You may use the same word as often as you wish. Don't worry about whether your answer i s right or wrong. This is not r e a l l y a test of knowledge, but a study of ordinary language. There are actually many possible answers. A l l we want i s your opinion. Four forms of the questionnaire were d i s t r i b u t e d . In three versions, subsets of the 20 target emotions were interspersed with f i l l e r s such as " t i n g l e " , "dizziness", "stubbornness", "moody", "alertness". Subjects generated superordinates for a t o t a l of 20 items. The 20 targets were dis t r i b u t e d as: seven in two forms of the questionnaire, and six in the t h i r d . In the fourth version a l l 20 target emotions were listed--there were no f i l l e r items. The rationale for this last version was that l i s t i n g a l l 20 emotion terms might create a demand c h a r a c t e r i s i t i c such that subjects would be reluctant to give "Emotion" as a response to each item. The purpose of t h i s manipulation, therefore, was to create an i m p l i c i t bias against giving "emotion" and to help discover what superordinates other than "emotion" would be generated. Each form of the questionnaire was administered to 30 subjects. 86 Results and Discussion Each of the 20 target terms generated "emotion" as i t s superordinate. The percentage of subjects who simply gave "emotion" for each term i s shown in Table 8. The word "emotion" embedded in a phrase, or prefaced by an adjective (e.g. "negative emotion") also occurred f a i r l y often. Therefore another set of percentages i s shown for responses that either were "emotion" or included the word "emotion" as part of the response. This was done separately for questionnaires with and without f i l l e r items. These data also appear in Table 8. To assess whether "emotion" was e l i c i t e d as a superordinate label more often for the more prototypical exemplars, the responses for the instances with the 10 highest t y p i c a l i t y ratings (Study Three) were averaged. Responses for the 10 less prototypical instances were also averaged. T-tests between mean per cent emotion responses for these two groups were computed for each of the four conditions l i s t e d . As shown in Table 9, results were s i g n i f i c a n t and in the expected di r e c t i o n in every case. "Emotion" was e l i c i t e d as the superordinate category name s i g n i f i c a n t l y more often for highly prototypical than for less t y p i c a l instances regardless of type of questionnaire or type of response considered. Table 10 depicts the kinds of superordinates given in cases where the word "emotion" was included as only part of the response. Only responses given for two or more terms were 87 Table 8 Generation of the Superordinate Category for 20 Target Emotion Terms "EMOTION" "EMOTION" INCLUDED AS AS THE RESPONSE PART OF THE RESPONSE No f i l l e r s with f i l l e r s No f i l l e r s with f i l l e r s Category % Subjects % Subjects % Subjects % Subjects Love 50. .00 50. .00 64. ,29 60. ,00 Sadness 35. .71 60. .00 53. ,58 66. ,67 Hate 35. .71 50. ,00 46. ,43 50. ,00 Happiness 32. . 1 4 56. .67 46. ,43 60. .00 Joy 32. .14 46. .67 46. .43 50. ,00 Anger 28. .57 46. .67 46. .43 56. .67 Depression 28. .57 10. .00 42. .86 10. .00 Envy 25, .00 30. .00 39. .29 36, .67 Disgust 25. .00 36. .67 35. .71 40. .00 Fear 25, .00 33. .33 39. .29 46, .67 Gu i l t 25, .00 26, .67 39. .29 36, .67 Pride 25, .00 3. .33 35. .71 3, .33 Worry 25, .00 26, .67 39, .29 30, .00 Anxiety 21 , .43 30. .00 35. .71 30, .00 Excitement 21 , .43 30, .00 32. . 14 40, .00 Respect 21 , .43 0 32. . 1 4 3, .33 Awe 17, .86 10, .00 28. .57 10, .00 Embarrassment 17, .86 20. .00 35. .71 23, .33 Boredom 10. .71 3. .33 25. .00 3, .33 Calm 10, .71 6. .67 21 . 43 6, .67 T a b l e 9 Per 'Cent S u b j e c t s G i v i n g "Emotion" as S u p e r o r d i n a t e Category NO FILLERS WITH FILLERS "Emotion" "Emotion" as p a r t as the response of the response . "Emotion" "Emotion" as p a r t as the response of the response P r o t o t y p i c a l Exemplars N o n p r o t o t y p i c a l Exemplars t 31 .43 20.00 3.93* 45.76 32.86 5.28" 4 1 .00 16.67 8 . 27** 47 .67 18.67 7 .94** *p <.01 **p <.001 Table 10 Responses Given for Generation of Superordinate Category Task Category Emotion Negative Emotion Unpleasant Emotion Positive Emotion Spontaneous Emotion Neutral Emotion Unhappy Emotion Happy Emotion Uncontrolled Emotion Controlled Emotion Pleasant Emotion % OF 20 TERMS No f i l l e r s with f i l l e r s 100 100 60 5 60 0 35 5 35 0 30 0 25 0 20 0 20 0 10 0 10 0 90 included in thi s table. When the targets were embedded in other semantically related categories, subjects responded with only the word "emotion" in almost a l l cases. When the targets appeared alone, more s p e c i f i c responses l i k e "pleasant emotion", "unpleasant emotion", for example, occurred with much greater frequency. F i n a l l y , some superordinates did not include the word emotion. "Feeling" ("feeling", "negative f e e l i n g " , or "positive feeling") was given as the superordinate response at least once for a l l 20 emotion categories in both forms of the questionnaire ( f i l l e r s or no f i l l e r s ) . "State" ("state", "state of mind", "state of being") was given as the superordinate response at least once for 17 of the 20 categories in the "no f i l l e r s " version, and for 15 of the 20 categories in the " f i l l e r s " version. No other superordinates were generated for more than one category by more than one subject. The issue of how "emotion" is di f f e r e n t from other possible superordinates i s an important but neglected issue. Perusal of the d e f i n i t i o n s offered in the l i t e r a t u r e on emotion unfold two major interrelated issues. The f i r s t i s : how is emotion defined? The second, related to the f i r s t , i s : how is emotion distinguished from other psychological states? The present data suggest that there i s an overlapping relationship between "emotions", "feelings", and "states". "Feelings" was given as the superordinate at least once for each of the target terms, and "state" for a majority. Perhaps 91 emotion, feeling, and psychological state are a l l terms on one l e v e l of a hierarchy, the l e v e l here c a l l e d superordinate. A l t e r n a t i v e l y , emotion may be a type of feel i n g , i f feeling includes not only emotions, but b e l i e f s , proprioceptive feedback, pains, and i l l n e s s e s . The relationship between these pyschological states invites further exploration. Study Six Family Resemblances A potential source for prototype emergence in semantic categories i s the relationship of par t i c u l a r members of a category to the other members of that category. Rosch and Mervis(l975) demonstrated that the exemplars that share the greatest number of features with other members come to be regarded as protoytpical. This argument i s based on the notion that for some natural language categories, category members do not share the same attributes or same set of att r i b u t e s . Rather, the members are linked through a pattern of shared attributes, and the degree of f i t to such a pattern is referred to as family resemblance. The idea i s that each exemplar shares one or more features with most other category members. To use a human family as a concrete example, several members of a family may have buck teeth, but a l l members need not have th i s feature. Several may have reddish hair, but not a l l are necessarily redheads. While no p a r t i c u l a r family member may have a l l of the family t r a i t s , a l l members probably have several of the t r a i t s , although not necessarily 92 the same ones. The most t y p i c a l member of the family w i l l have buck teeth and red hair. Rosch and Mervis(l975) found that more prototypical category members share more of the family a t t r i b u t e s . Less t y p i c a l members have fewer attributes in common with other category members, and have more attributes in common with neighboring categories than do the prototypical cases. For example, consider members of the category furniture: tables and chairs have many features in common. Rugs and lamps do not share as many attributes with each other, nor do they share many attributes with tables and chairs. The purpose of this study was two-fold. The f i r s t purpose was to see i f a l l members of the category emotion have one or more attributes in common (which would enable us to define the concept by l i s t i n g the c r i t e r i a l a t t r i b u t e s ) . The other purpose was to discover i f the category exemplars sharing more attributes with other instances of the concept were those regarded as prototypical. Method Subjects. Subjects were 145 students enrolled in pyschology classes at the University of B r i t i s h Columbia. Pa r t i c i p a t i o n was voluntary. In the f i r s t phase, 40 respondents generated attributes for the 20 target emotions (Table 2). In the second phase, 105 respondents rated one emotion term for the presence or absence of the attributes that had been generated by subjects in the f i r s t phase. 93 Phase One: Attribute Generation The task was described to subjects in the following way: This i s a simple study to find out the c h a r a c t e r i s t i c s and attributes that are common to people experiencing psychological events. For example, i f you were asked to l i s t the c h a r a c t e r i s t i c s of a person experiencing terror, you might write: -possible danger occurs—may be real l i k e a bear; may be imaginary l i k e a ghost -attention is focused on the threat -heart beats quickly -eyes open wider -eyebrows l i f t -palms and soles sweat -thoughts race through the person's mind -unpleasant sensations are experienced -the person runs as fast as they can -hands tremble - r e l i e f i s experienced after a few minutes It might help to imagine you're explaining the meaning of the word terror to a foreigner or to someone who has never experienced i t . So, include the obvious. T e l l how i t comes about and what 9 4 happens a f t e r . But emphasize a description of how one feels and acts. Try not just to free associate. If "terror" makes you think of elevators, don't write elevators. We're interested in what is common to instances of t e r r o r . Subjects were instructed to take 2 or 3 minutes to l i s t the attributes for each of ten emotions presented in random order. The p a r t i c u l a r emotions rated varied from one subject to the next. Phase Two: Rating Task In the second phase of t h i s study, subjects were required to act as judges, rating emotions with respect to some of the attributes generated in phase one. The following instructions were provided: This is a study about our b e l i e f s concerning important psychological states. This study i s part of a larger project on the topic. Please begin by considering one such state, namely, {X}. Remember several occasions in which you or someone you know has experienced {X}. Pause to consider the various forms i t might take and some of the various events that might be associated with i t . 95 On the remaining pages of thi s booklet is a long l i s t of various events that could be involved in any psychological state. Some occur often, some rarely. Your task in t h i s study w i l l be to rate the extent to which each event goes with or i s part of {X}. The attributes were rated using a scale of 0-4 for each of the 20 target emotions as well as the word "emotion", where 0=never; 1=rarely; 2=sometimes 3=often; and 4=always. Each subject provided ratings for one term on 161 at t r i b u t e s . Results and Discussion Phase One The t o t a l number of responses generated in phase one was 2425. Creating family resemblance scores therefore required some decision as to which of these responses represented the same and which represented d i f f e r e n t a t t r i b u t e s . The attempt was to be conservative, but s t i l l count as the same attribute words or phrases highly similar in meaning. The decision to group responses required a consensus. Two judges, graduate psychology students, performed th i s task. Most groupings were of i d e n t i c a l responses. Some examples of groupings of non-id e n t i c a l responses are "eyes open wider, eyes widen, eyes open, eyes are wide, and widened eyes" were a l l grouped as "eyes open wide". However, "pupil d i l a t i o n " , "eyes averted", "squinted, narrow eyes", and "bright, sparkling eyes" were 96 treated as separate a t t r i b u t e s . This grouping procedure reduced the number of attributes to 642. Next, attributes mentioned only once (315) were eliminated, leaving 327 attributes that were mentioned on two or more occasions, either by di f f e r e n t subjects for one emotion, or by the same subject for d i f f e r e n t emotions. The point here i s that idiosyncratic responses for single emotion categories were omitted from consideration for the family resemblance scores, since the attributes unique to single members do not contribute to the structure of the category per se. Each attribute was then weighted by the number of emotions for which i t had been generated. For example, a weight of 16 was given to "heartrate increases", because i t was l i s t e d as an attibute for 16 of the 20 emotions. This was, in fact, the highest weight obtained. A weight of 1 was given to "doesn't care about appearance" because i t was endorsed (by two or more subjects) for one emotion (depression). The 327 attributes and the weight for each appear in Table 11. Figure 2 depicts how many emotions were credited with a given number of attrib u t e s . As can be seen from Figure 2, the number of attributes decreases as the number of emotions to which i t applies increases. There were no attributes common to a l l 20 emotions. In fact, with one exception, there were no attributes in common to more than half the 20 emotions eyes open w1de(9) eyes a v e r t e d ( 5 ) eyes c o l d and h a r d ( 1 ) sml11ng(7) c l e n c h i n g t e e t h ( 3 ) p u r s e d mouth(2) f i s t s c1enched(5) b r e a t h e 1n r a p i d l y and s h a l l o w ( 4 ) expel a i r s u d d e n l y ( 1 ) sweaty palms(7) v o i c e becomes h i g h - p i t c h e d ( 2 ) k n i t t e d brow(3) head bowed(2) look of c o n t e n t m e n t ( 1 ) s t e r n f a c i a l e x p r e s s 1 o n ( 2 ) doesn't c a r e about appearance(1) c h i n In palm(1) h e a r t r a c e s ( 3 ) h e a r t a c h e ( 2 ) knot 1n stomach(5) b u t t e r f l i e s i n stomach(5) • f e e l i n g l o c a l i z e d i n stomach(2) body ( e s p e c i a l l y hands) i s c o l d ( 3 ) T a b l e 11 A t t r i b u t e s L i s t e d f o r 20 Target Emoti p u p i l d 1 l a t 1 o n ( 2 ) s q u i n t e d , narrow e y e s ( 7 ) g l a r i n g e y e s ( 1 ) laugh1ng(4) f r o w n i n g ( 5 ) mouth drops open(2) t e a r s / c r y 1 n g ( 9 ) b r e a t h l e s s ( 2 ) b r e a t h e heav11y(4) t r e m b l i n g / s h a k i n g ( 1 0 ) r e l a x e d ( 7 ) r a i s e d e y e b r o w s / l i f t e d brow(4) bent over p o s t u r e ( 4 ) p l e a s a n t f a c i a l express1on(2) unhappy, gloomy express 1on(1) h a i r s r a 1 s e ( 2 ) i n c r e a s e d h e a r t r a t e ( 1 6 ) h e a r t r a t e slows(5) h e a r t and p u l s e r a t e r e g u l a r ( 1 ) stomach muscles t1ghten(4) queasy stomach(5) f e e l heat i n body(4) p e a r s burn1ng(2) Categor1es f i x e d eyes(3) bags under eyes(2) b r i g h t , s p a r k l i n g eyes(2) dry mouth and l i p s ( 3 ) scowl(3) 1ips c u r l ( 1) f a c e t u r n s r e d ( 7 ) easy, r e l a x e d breath1ng(2) p e r s p i r a t i o n / s w e a t ( 1 0 ) t i n g l i n g s e n s a t i o n ( 5 ) e r e c t , u p r i g h t p o s t u r e ( 3 ) w r i n k l i n g nose( 1) body s t i f f and r i g i d ( 3 ) look of d i s d a i n on f a c e ( 2 ) s t r o n g f a c i a l e x p r e s s i o n s ( 1 ) f e e l e n e r g e t 1 c ( 4 ) h e a r t pounds(2) heart f l u t t e r s ( 2 ) adrenal i n f l q w ( 7 ) s i c k f e e l i n g 1n stomach(5) stomach c h u r n s ( 3 ) f e e l warm 1ns1de(6) i n c r e a s e d b l o o d pressur.e(2) t e n s e ( 9 ) no energy, 1 e t h a r g 1 c ( 4 ) r i n g i n g 1n the e a r s ( 3 ) l o s s of a p p e t i t e ) 5 ) p l e a s a n t s e n s a t 1 o n s ( 4 ) wlthdrawn(4) shout and scream(4) Impaired v e r b a l 1 z a t 1 o n ( 3 ) ta1kat1ve(6)~, mumbl1ng(2) t a l k s f a s t e r ( 3 ) s1gh1ng(5) s t a r i n g I nto s p a c e ( 4 ) muscle tens1on(4) f e e l i n g of 11ghtheadedness(5) f l o a t i n g f e e l 1 n g ( 4 ) bad t a s t e 1n mouth(1) u n p l e a s a n t sensat1ons(2) Ignore p e o p l e around you(2) ramblIng t a l k ( 2 ) speech1essness(5) d i f f i c u l t to v e r b a l i z e f e e l 1 n g s ( 3 ) s t u t t e r 1 n g ( 2 ) e x c l a m a t o r y speech(4) s 1 ng 1 ng( 3) o n l o o k e r w i t h amazement)1) ca n ' t s l e e p ) 1 ) s l e e p ) 2 ) a c t i n g k i n d t o compensate f o r wrongdo1ng(1) generous)2) hugging p e o p l e ( 3 ) a v o i d o b j e c t of emot1on(2) t r y to h i d e f a c e ( 1 ) c o v e r up by changing s u b j e c t ) 1) • don't f e e l l i k e l a u g h i n g or sm111ng(1) f e e l l i k e e e l e b r a t 1 n g ) 4 ) f 1 d g e t ( 6 ) s h u f f l i n g , t a p p i n g f e e t ( 4 ) hands t w 1 s t i n g ( 2 ) t a p p i n g hands and f 1 n g e r n a i 1 s ( 2 ) jumpy(3) q u i c k to r e a c t ( 4 ) s l o w e r r e f l e x t1me(3) slow movements(5) l i s t e n s to m u s i c ( 2 ) e a t i n g ( 2 ) d r i n k 1 n g ( 3 ) 1n a c o m f o r t a b l e p h y s i c a l p o s 1 t i o n ( 2 ) a v o i d n o t 1 c e ( 2 ) s t a r e or look f1xed1y(3) f e e l t 1 r e d ( 3 ) heightened a r o u s a l ( 4 ) ^ heightened s e n s e s ( 4 ) weak, numb l e g s ( 1 ) q u i e t ( 6 ) t a l k i n g 111oglca11y(5) swearing)3) t a l k l o u d e r ( 4 ) u n t a l k a t 1 v e ( 2 ) t a l k s s o f t l y and q u 1 e t l y ( 2 ) humm1ng)2 ) yawning)1) p l e a s a n t to everyone)1) engage 1n v i o l e n t a c t i o n s ) 1) running(2) escape from o b j e c t of emot1on(2) laugh i t o f f ) 1 ) r e s t l e s s n e s s ( 5 ) doodle on paper(2) hyperact1ve(4) unreact1ve/unrespons1ve(2) Impaired m o b i l i z a t i o n and c o o r d 1nat1 on)5) smoking)2) show o f f ( 1 ) t a l k i n g t h i n g s o u t ( 2 ) make l i g h t c o n v e r s a t 1 o n ( 3 ) agree w i t h o b j e c t of emot1on(1) dandng/jump1ng(4) bouncy(3) r u n n i n g to the bathroom a l l the t1me(1) r a p i d body movements!3) easy body movements!3) remain mot1onless!2) b i t e I1ps(3) n e r v o u s n e s s ! 7 ) uneasy!3) c o n c e n t r a t i o n s h o r t ( 2 ) o b e s s l v e c o n c e r n with s i t u a t i o n ! 1 C c o n c e n t r a t i o n on o b j e c t of emot1on(4) p r e o c c u p a t i o n w i t h e v e n t ( 5 ) thoughts r a c e t h r o u g h m1nd(3) empty thou g h t s ( 2 ) happy or p l e a s a n t thoughts!4) u n p l e a s a n t thoughts!3) e v a l u a t i o n of s 1 t u a t 1 o n ( 4 ) c o n s i d e r s o l u t 1 o n s ( 2 ) t h i n k s l o g i c a l 1y(1) v e r y a t t e n t 1 v e ( 2 ) unaware of s u r r o u n d 1ngs(5) l o s s of i n t e r e s t ! 1 ) f e e l u n s t i m u l a t e d ! 1 ) a c u t e i n t e r e s t ! 1 ) f e e l a g g r e s s i v e toward o b j e c t of embt1on(4) f e e l c o n f u s e d ( 4 ) e x p e r i e n c e contempt and d i s d a i n ! 1 ] f e e l l a z y ( 2 ) motivated!1) e v e r y t h i n g l o o k s d a r k ( 2 ) f e e l i n g of inner emptiness!1) f e e l 1 s o l a t e d ( 2 ) want to d1e(2) f e e l s e l f - p 1 t y ( 3 ) f e e l humble(2) f e e l 1nadequate!5) q u e s t i o n i n g s e l f ( 1 ) f e e l u n c o m f o r t a b l e ( 3 ) f e e l unsure of s e l f ( 4 ) f e e l s u p e r i o r ! 2 ) s e l f i s h ( l ) f e e l i n g of h e l p l e s s n e s s and l o s s of c o n t r o l ( 4 ) pac1ng(5) arms wave about!2) seek p h y s i c a l r e l e a s e O ) i n a c t 1 v e ( 3 ) b i t i n g n a l l s ( 3 ) can't c o n c e n t r a t e ( 8 ) hard to c o n c e n t r a t e on a n y t h i n g e l s e ( 2 ) deep i n thought!2) sense of d i s b e l i e f ( 2 ) unorganized, s c a t t e r e d t h o u g h t s ( 3 ) i r r a t i o n a l thoughts(4) aware!5) t h i n k i n g i s e1sewhere(2) c l osed-minded!2 ) f e e l h o s t i l e ( 2 ) f e e l r e v o l t e d ( 1 ) hopeless f e e l i n g ( 2 ) f e e l 1onely!2) f e e l s t u p i d or f o o l 1 s h ( 1 ) v u l n e r a b l e and e a s i l y h u r t ( 3 ) f e e l i n g of se1f-worth!3) d i s a p p o i n t e d with y o u r s e l f ! 1 ) 1n c o n t r o l ( 1) r e v e n g e f u l ( 1 ) I n d e c i s i v e ) 2 ) u n p l e a s a n t e x p e c t a t 1 o n s ( 3 ) daydreaming)2) a l e r t ( 3 ) p e a c e f u l ) 2 ) f e e l s e c u r e ( 4 ) c h e e r f u l ( 2 ) sense of wel1-be1ng(3) n o t h i n g can p o s s i b l y go wrong(2) e a s y g o i n g a t t i t u d e ) 2 ) good mood) 2) i-j u b l l a n t ( 2 ) p s y c h o l o g i c a l h1gh(2) f e e l s t r o n g ) 3 ) e x p r e s s 1 v e ( 2 ) f e e l i n g of a n g e r ( 8 ) f e a r or d r e a d ) 4 ) . w o r r y ) 3 ) f e e l down or low)2) a p a t h e t i c ( 2 ) r e m o r s e f u l ( 2 ) eager t o p l e a s e ( 2 ) f e e l d i s c o n t e n t ) 1 ) f e e l g u 1 l t y ( 2 ) p r e f e r t o be a l o n e ( 3 ) c o l 1 e c t e d ( 2 ) emotional b u i l d u p ) 2 ) g e t t i n g ready f o r an event) 1) f e e l i n g of u n r e a l 1 t y ( 2 ) senses a r e sharpened(3) s e d a t e ( 3 ) j o y ( 3 ) 1 I g h t h e a r t e d ) 2 ) f e e l I 1 v e l y ( 2 ) n o t h i n g b o t h e r s you(2) e v e r y t h i n g looks good(4) f e e l e x c 1 t e d ( 6 ) e x h l l a r a t 1 o n ( 2 ) f e e l i n g of f u l f i l l m e n t ^ ) admlrat1on)3) f r u s t r a t 1 o n ( 7 ) 1 r r 1 t a b l e ( 6 ) f e e l t1m1d.(2) f e e l d e p r e s s e d ( 4 ) f e e l J e a l o u s ( 2 ) u n p l e a s a n t f e e l 1 n g s ( 3 ) f e e l shame)2) c o o l ( 2 ) ant 1cIpat1on(6 ) f e e l dazed)2) dreamy f e e l 1 n g ( 2 ) f e e l s e l f - c o n f 1 d e n t ( 4 ) g l a d ( 2 ) f e e l happy(6) f e e l c o n t e n t e d and s a t 1 s f 1 e d ( 4 ) few worr1es(2) not upt1ght(2) sense of u n i t y , harmony)1) e1ated(4) bubbly)3) l o y a l t y ( 2 ) p l e a s a n t fee11ng(4) f e e l sad(2) not e a s i l y 1 r r 1 t a t e d ( 2 ) f e a r of outcome)3) f r e t f u l ) 2 ) b i t t e r f e e l 1 n g s ( 2 ) sorrow)2) f e e l s e 1 f - c o n s d o u s ( 2 ) p o l 1 t e / p r o p e r ( 2 ) ) t 3 want what someone e l s e has(1) h a t r e d ( 2 ) embarrassment)2) l i k e to be around o b j e c t of emotion)1) look away from o b j e c t of emotion) 1) seek companlonshfp(2) need to t a l k about s1t u a t 1 o n ( 5 ) want t o f o r g e t s 1 t u a t 1 o n ( 2 ) need to mob1!1ze(2) want t o es c a p e from s 1 t u a t 1 o n ( 4 ) want to remove o b j e c t of emotlon(3) want t o be v e r b a l l y / p h y s i c a l l y v1o1ent(1) perhaps want t o hug someone!1) want to h1de(1) e c s t a s y ( 3 ) c a r i n g f o r person(1) need t o p h y s i c a l l y touch o b j e c t of emot1on(1) a g 1 t a t e d ( 1 ) d e b l 1 1 t a t e d ( 2 ) c o n t i n u e s u n t i l a n o t h e r f e e l i n g o v e r r i d e s 1t(2) f e e l i n g d i s s i p a t e s once I n c i d e n t 1s o v e r ( 3 ) f e e l i n g of r e l a x a t i o n when I n c i d e n t ends(2) l o s i n g someone you l o v e ( 2 ) t r a g i c event o c c u r s to s e l f or o t h e r s ( 1 k1nd(1) d i g n i t y and pr1de(1) s t r e s s e d f a c i a l e x p r e s s 1 o n ( 1 ) f e e l l i k e scream1ng(2) don't f e e l l i k e doing anyth1ng(3) f e e l l i k e cry1ng(2) wish you c o u l d change th1ngs(2) want to share f e e l i n g w i t h o t h e r s ( 1 ) Int1mate(1) f e e l courageous(2) f e e l i n g d i s s i p a t e s q u 1 c k l y ( 4 ) f e e l i n g doesn't go away f a s t ( 3 ) r e l i e f when Incident ends(3) s p e c t u c u l a r sight-seems unbel1evab1e( 1) > f e e l accompl1shment(1) Note: The number In p a r e n t h e s e s Is the weight a s s i g n e d to each a t t r 1 b u t e - - e a c h a t t r i b u t e was weighted by the number emotions f o r which i t was g e n e r a t e d . 102 Figure 2. Number of attributes generated for each number of emotion categories. cn CD a X! •H >-l -P +J ftJ u a) § 1 17 + / / 65 60 55 50 45 40 35 30 25 20 1 5 10 5 0 x x x / / x X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 Number of emotion categories to which attributes apply 103 studied. F i n a l l y , a family resemblance score (termed phase one score) was calculated for each emotion. This score i s simply the sum of the weighted at t r i b u t e scores for that emotion. In t h i s f i r s t phase, the family resemblance scores were highly data bound, meaning that the scores were as close to the att r i b u t e generation data as possible. These scores are given in Table 13. Phase Two A second set of family resemblance scores (termed phase two scores) was more derived than the f i r s t set. F i r s t , the phase one attributes were re-written in a form amenable to rating on the scale provided. Secondly, any attributes that contained emotion words were eliminated to avoid confounding resemblance with s i m i l a r i t y of emotions. F i n a l l y , certain phase one attributes that would have been d i f f i c u l t to rate separately were combined. This procedure reduced the number of attributes to 161, which are l i s t e d in Table 12. Each subject rated the 161 attributes for either one of the 20 target emotions or the word "emotion". For each term, the attributes were rated by five "judges". Phase two family resemblance scores were then calculated. The five judges' ratings for a given term were f i r s t summed for each of the 161 at t r i b u t e s . The maximum possible sum for any attribute was 20; the minimum sum was 0. The obtained maximum sum was 20; the minimum was 0. An "attribute score" was then computed by taking the mean of the 1 04 Table 13 Family Resemblance Scores for 20 Target Emotion Categories Category Phase One Rank Phase Two Rank Excitement 283 1 24632 14 Fear 274 2 28974 11 Worry 261 3 33341 6 Anxiety 248 4 39234 2 Anger 246 5 29459 10 Depression 224 6 29796 9 Joy 220 7 30389 8 Gui l t 217 8 39990 1 Pride 21 1 9 1 4337 20 Love 202 10 28082 13 Happiness 200 1 1 31982 7 Embarrassment 190 1 2 22042 1 5 Hate 188 1 3 33985 5 Disgust 183 1 4 28659 12 Awe 1 64 1 5 36560 3 Sadness 1 55 1 6 35784 4 Respect 1 36 1 7 1 4635 19 Envy 1 32 1 8 1 5494 18 Boredom 1 1 2 19 21351 16 Calm 1 1 1 20 17409 1 7 Note: In the rating task judges were also required to rate category Emotion. The family resemblance score for this category was 17550. T a b l e 12 A t t r i b u t e s Rated f o r 20 Target Emotion C a t e g o r i e s eyes open w i d e r ( 1 4 . 2 ) eyes a r e a v e r t e d (8.7) eyes appear h a r d and c o l d ( 7 . 2 5 ) p e r s o n l a u g h s ( 8 . 2 ) p e r s o n frowns(7.3) mouth drops open(7.85) p e r s o n c r i e s , sheds t e a r s ( 9 . 3 ) b r e a t h i n g 1s easy and slow(8.2) eyes s t a r e ( 1 1 . 0 5 ) p e r s o n s q u i n t s ; narrows eyes(7.5) eyes appear b r i g h t , spark 11ng(8.45) mouth, t h r o a t , or l i p s a re dry(9.35) p e r s o n scowls(6.9) 11ps c u r l ( 6 . 4 ) f a c e t u r n s r e d ; blushes(9.15) p u p i l s d11ate(9.35) bags appear under eyes(8.3) person s m i l e s ( 8 . 8 5 ) person c l e n c h e s teeth(7.55) mouth 1s pursed(8.25) f i s t s a r e clenched(7.2) b r e a t h i n g 1s s h a l l o w and rap1d(10) person p e r s p l r e s ( 9 . 1 ) p e r s o n t r e m b l e s ( 8 . 5 5 ) p o s t u r e 1s e r e c t , s t i f f , o r r1g1d(8.6) brow 1s k n l t t e d ( 7 . 3 5 ) nose w r 1 n k l e s ( 7 ) head bows(7.95) b r e a t h i n g 1s heavy(9) p e r s o n has t i n g l i n g , p r i c k l y , or Itchy sensat1ons(7.25) brow Is I 1 f t e d ( 6 . 7 ) p o s t u r e 1s slumped(7.15) s t r o n g f a c i a l e x p r e s s i o n a p p e a r s ( 9 . 7 ) h a i r s ra1se(8.75) h e a r t r a t e 1 n c r e a s e s ( 1 1 . 2 ) v o i c e changes(10.15) a d r e n a l i n f l o w s ( 9 . 3 ) body f e e l s c o l d ( 6 . 2 ) p e r s o n f e e l s l a c k of energy(8.55) p e r s o n f e e l s 1 1ghthearted(5.9) p e r s o n f e e l s h i g h l y a r o u s e d ( 9 . 6 ) s e n s o r y a b i l i t y 1s he1ghtened(8.85) p e r s o n f e e l s weak or numb In l e g s ( 8 . 2 5 ) p e r s o n f e e l o t h e r s a r e s t a r 1ng(7.6) p e r s o n 1s eager to p l e a s e ( 7 . 9 ) p e r s o n has s t r o n g wants or d e s 1 r e s ( 9 . 6 5 ) p e r s o n f e e l s s e n s a t i o n s 1n stomach(11.7) b l o o d p r e s s u r e r1ses(10.05) p e r s o n f e e l s drowsy(5.9) t h e r e Is r i n g i n g In the ears(6.55) a p p e t i t e Is l o s t ( 9 . 2 ) s e n s a t i o n s a r e p l e a s a n t ( 8 ) person f e e l s energet1c(9.3) h e a r t r a t e slows(6.3)  muscles tense(10.35) person f e e l s d e b l 1 1 t a t e d ( 7 ) person f e e l s as If f1 oat 1ng(7.6) bad t a s t e Is 1n mouth(5.9) s e n s a t i o n s are u n p l e a s a n t ( 9 . 5 ) body f e e l s warm or hot(9.6) person a c t s proper or p o l 1 t e ( 9 . 2 ) person wants to be near a n o t h e r ( 1 0 ) p e r s o n t r i e s t o be l i k e a n o t h e r ( 7 . 4 ) p e r s o n wants t o be away from another(9.15) p e r s o n wants t o run(8.95) p e r s o n wants to w1thdraw(8.75) p e r s o n Is q u i e t , unta1kat1ve(10.15) p e r s o n shouts or screams(7.35) p e r s o n swears or c u r s e s ( 6 . 8 5 ) p e r s o n 1s t a 1 k a t W e ( 8 . 8 ) p e r s o n mumb1es(8.7) p e r s o n s t u t t e r s ( 7 . 1 5 ) p e r s o n s1ghs(8.55) p e r s o n wants to s i n g or hum(6.4) p e r s o n yawns(5.25) p e r s o n c a n ' t s l e e p (7.7) ' p e r s o n a c t s f r i e n d l y and generous(7.35) p e r s o n wants t o escape from someth1ng(8.3) p e r s o n doesn't f e e l l i k e s m i l i n g o r i a u g h l n g ( 8 . 7 5 ) p e r s o n shows u n d i r e c t e d a c t i v i t y ( f i d g e t s , paces, t a p p i n g , s h u f f l i n g , e t c . ) p e r s o n Is h y p e r a c t i v e , q u i c k t o r e a c t ( 9 . 9 ) c o o r d i n a t i o n and m o b i l i z a t i o n a r e slow(8.5) p e r s o n Is 1n a c o m f o r t a b l e pos1t1on(8.15) p e r s o n 1s a g r e e a b l e ( 9 . 5 5 ) walk 1s bouncy(7.7) p e r s o n wants t o be v1o1ent(6.4) p e r s o n seeks r e l e a s e ( 9 . 1 5 ) movements a r e vehement(7.8) p e r s o n wants to show o f f ( 4 . 9 ) p e r s o n b i t e s 11ps or na11s(8.75) thoughts a r e d1sorganlzed(10.45) p e r s o n t r i e s t o e v a l u a t e c u r r e n t s1tuat1on(10.05) thoughts a r e 1 r r a t 1ona1(9.4) thoughts are 1og1ca1(7.25) p e r s o n Is I n a t t e n t i v e o r unaware of c u r r e n t env1roment(9.G5) p e r s o n Is u n a b l e to c o n c e n t r a t e ( 9 . 5 ) p e r s o n 1s preoccup1ed( 1 1 ) thoughts r a c e though m1nd(11.05) thoughts a r e empty(6.5) p e r s o n f e e l s worthy, s u p e r i o r , l"n c o n t r o l , or conf 1 dent (7 . 5) p e r s o n f e e l s unworthy. I n f e r i o r , h e l p l e s s , unconf1dent(8.05) p e r s o n 1s 1 ndec 1 s 1 ve(8.. 9) p e r s o n f e e l s concern f o r another(8.6! t h e r e a r e u n p l e a s a n t e x p e c t a t 1 o n s ( 8 ) p e r s o n f e e l s dazed(9.2) person Ignores o t h e r s ( 8 . 4 ) a b i l i t y to speak 1s Impa1red(7.7) person cannot v e r b a l i z e f e e l 1ngs(8.55) p e r s o n speaks f a s t ( 8 ! 5 ) person s t a r e s Into space(9.3) person .s leeps( 7 . 85 ) person wants to hug another(7.75) person f e e l s r e s t 1 e s s ( 1 1 . 4 5 ) person f e e l s l i k e e e l e b r a t 1 n g ( 7 . 9 ) t r y to h i d e f a c e ( 6 . 6 ) movements and time to r e a c t a r e slow(8.5) person Is opt 1m1st1c(7.2) person t r i e s to a v o i d b e i n g not1ced(8.1) arms wave about(7.65) behav i o r 1s r e p e t l t l v e O . 15) person Is I n a c t i v e or mot Ionl e s s ( 7 .8.) mind 1s c l o s e d ( 8 . 5 5 ) person c o n s i d e r s p l a n s or s o l u t 1 o n s ( 9 . 6 ) person Is very a t t e n t i v e or aware(9.65) person f e e l s unstImulated(7.9) p e r s o n f s deep In thought(10) person has a sense of d l s b e l 1 e f ( 7 . 6 ) thoughts a r e p i e a s a n t ( 8 . 5 5 ) thoughts are unpleasant(9.35) t h e r e Is h i g h ant 1cIpat1on(8.6) person daydreams(8.95) p e r s o n d e s i r e s to e a t , smoke, dr1nk(10.05) p e r s o n f e e l s l o y a l ( 7 . 2 5 ) p e r s o n admires another!8.85) p e r s o n f e e l s confused!9.16) p e r s o n l a c k s mot 1 vat 1on(7.75) p e r s o n wants t o d i e ( 4 . 8 5 ) p e r s o n f e e l s s t u p i d ( 4 . 8 5 ) p e r s o n f e e l s v u l n e r a b l e , e a s i l y h u r t ( 6 . 5 5 ) p e r s o n f e e l s inadequate(8.95) p e r s o n wants to be alon e ( 9 . 5 ) ft p e r s o n wants t o t a l k about c u r r e n t s1tuat1on(9.75) p e r s o n wants to escape(8.7) p e r s o n wants to d i s p o s e o f , or remove s 1 t u a t i o n ( 10.05) p e r s o n wants t o touch a n o t h e r ( 9 . 2 5 ) p e r s o n wants to share w i t h another(9.6) person wants to change s l t u a t l o n t 1 0 . 1 5 ) p e r s o n wants t o h l d e ( 6 . 3 5 ) c u r r e n t s t a t e w i l l cont1nue(7.05) c u r r e n t s t a t e w i l l d i s s i p a t e q u 1 c k l y ( 9 . 2 ) c u r r e n t s i t u a t i o n i s unexpected(9.25) c u r r e n t s i t u a t i o n 1s t r a g i c ( 4 . 8 ) f a c e 1s e x p r e s s 1 o n l e s s ( 5 . 3 ) Note: The number i n p a r e n t h e s e s i s the a t t r i b u t e s c o r e f o r t h a t a t t r i b u t e . An " a t t r i b u t e s c o r e " was computed by t a k i n g the mean of the Judges summed r e s p o n s e s f o r a g i v e n a t t r i b u t e a c r o s s the 20 t a r g e t emotions. person i s pessim1st1c(8.2) person 1s a l e r t ( 9 . 6 ) person f e e l s a l o n e or 1 s o l a t e d ( 9 . 1 ) person f e e l s sense of u n r e a l 1ty(8.35) person seeks company of o t h e r s ( 8 . 4 5 ) person wants to f o r g e t ( 8 . 1 5 ) 108 judges' summed responses for a given attribute across the 20 target emotions. The highest possible was 20; the lowest 0. The obtained attribute scores, which ranged from 5.25 to 14.2, are given in parentheses in Table 12. This attribute score was calculated so that attributes more often associated with emotions would contribute more to family resemblance scores than would attributes less associated. Each term's association with an attribute was then calculated by multiplying the summed judges' ratings by the attribute score. The maximum possible product was 400 (20 X 20); the minimum 0. The obtained maximum product was 227.2; the obtained minimum 0. F i n a l l y , these products were summed across a l l 161 attributes for each term resulting in a family resemblance score for each of the 20 target emotions and the word "emotion". The maximum obtainable score was 64400 (400 X 161); the minimum 0. The obtained range of family resemblance scores was 14337-39990. Phase two family resemblance scores are given in Table 13. Across the 20 emotion terms, phase one family resemblance scores correlated with phase two scores .44 (correlation of ranks i s .34). Each of the Phase Two family resemblance scores was correlated with the score for the word "emotion". These correlations are given in Table 14. Subjects found i t meaningful to generate attributes for the 20 target emotion terms. The attribute "heartrate increases" received the highest weighting. Other 109 Table 14 Correlations between Family Resemblance Scores for "Emotion" and 20 Target Emotion Categories Category Correlation Probability Embarrassment .25 .01 Gu i l t .19 .01 Awe .17 .01 Excitement .17 .01 Hate .16 .02 Anger .14 .04 Envy . 1 3 .05 Happiness . 1 3 .05 Pride .13 .05 Respect .12 .06 Love .10 .11 Worry .10 .10 Joy .08 .15 Fear .07 .20 Anxiety .06 .21 Disgust .03 .36 Sadness .02 .37 Boredom .00 .48 Depression .00 .46 Calm -.04 .30 110 physiological attributes l i k e "perspiration/sweat" and "trembling/shaking" received a weight of 10, which was the second highest weight assigned to any a t t r i b u t e . The reason for t h i s pattern of results i s not readily apparent. In the second phase, based on attribute ratings, the highest ratings were given to attributes l i k e "eyes open wider", "heartrate increases", "breathing i s shallow and rapid", "person feels sensation in stomach", "blood pressure r i s e s " , "muscles tense", and so on. Once again, physiological/physical appearance changes were emphasized. In this phase, five subjects (judges) were required to rate the word "emotion". The a t t r i b u t e for "emotion" that received the highest score was "strong f a c i a l expression appears". The sum of the ratings for t h i s a t t r i b u t e was 18 (the highest possible score was 20). The next highest summed rating (16) was for "person seeks release" and for "thoughts race through mind". A rating of 15 was given to the following a t t r i b u t e s : "person feels highly aroused", "person wants to be near another", and "person shows undirected a c t i v i t y (fidgets, paces, tapping, s h u f f l i n g , e t c . ) " . What emerges from the two sets of family resemblance scores i s that the category emotion does not possess a set of defining a t t r i b u t e s . Rather, members of the category resemble each other in overlapping and c r i s s - c r o s s i n g ways that vary in kind and number, with no one attribute being shared by a l l emotions. Hilgard(1953) commented that, "Emotional states as experienced in ordinary l i f e are complex and l i t t l e is gained 111 by trying to distinguish sharply among the various emotions" (Hilgard, 1953, p.142). It may be the case that nothing i s gained by making sharp d i s t i n c t i o n s , and that in fact, the internal structure of the category defies doing so. GENERAL DISCUSSION OF INTERNAL STRUCTURE It i s through the convergence of operations that the internal structure of a category i s demonstrated. In Study Three, p r o t o t y p i c a l i t y ratings were obtained for some of the emotion terms generated in Study One. The fact that subjects found i t meaningful to rate the terms for goodness of example, and that they agreed with one another, suggests that the category emotion may be i n t e r n a l l y structured. If t h i s was the case, one would expect that the internal sturcture of the category would aff e c t performance on various measures. To recapitulate, the measures employed in thi s study were: 1. Free production of exemplars. 2. Reaction time to v e r i f y category membership. 3. Probability of e l i c i t a t i o n of "emotion" as the category name. 4. Family resemblance scores. Table 15 provides the p r o t o t y p i c a l i t y ratings for the 20 target emotions as well as the scores on the other measures for each of the 20 terms. The correlations between these measures, both for raw scores and ranks appear in Table 16. An examination of Table 16 reveals that the degree of representativeness correlates s i g n i f i c a n t l y with frequency of T a b l e 15 ^ Convergence of O p e r a t i o n s EMOTION as S u p e r o r d i n a t e F a m i l y Resemblances Study 1 Study 1 Hunt & T y p i c a l i t y No C a t e g o r y ( A l l ) Rank ( 1st 4) Rank Hodge Rank Rat ings Rank F11 l e r s Rank F 1 1 l e r s Rank S's Rank Judges Rank Hap p i n e s s 76 .0 1 51 .0 1 29 .75 4 5 .00 5 46 .43 4 .5 60 .00 2 . 5 200 1 1 31982 7 Anger 74 . 5 2 49 .0 • 2 27 .75 5 5 . 15 2 46 • 43 4 .5 56 .67 4 246 5 29459 10' Sadness 68 .0 3 42 .5 3 22 .75 6 5 .04 4 53 .58 2 66 .67 1 155 16 35784 4 Love 62 .0 4 38 . 5 4 52 . 25 2 5 .46 1 64 . 29 1 60, .00 2 .5 202 10 28082 13 Fear 48 .0 5 24 . 5 5 53 . 25 1 4 .78 6 39 . 29 9 .5 46. .67 7 274 2 28974 1 1 Hate 44 .5 6 19 .5 6 44 .50 3 5 .26 3 46 .43 4 .5 50. .00 5 .5 188 13 33985 5 Joy 41 .0 7 « 13 .0 7 15 . 75 7 4 .89 8 46 .63 4 .5 50. .00 .5 220 7 30389 8 Exc i tement 26 . 5 8 6 . 5 8 5 .00 10 4 .58 9 32 . 14 16, .5 40. .00 8 .5 283 1 24632 14 Anx i e t y 25 .0 9 3 . 5 10 7 .00 9 4 . 29 13 35 .71 13, 5 30. .00 12. . 5 248 4 39234 2 D e p r e s s i o n 21 .0 10 4 . 0 9 7, .50 8 4 .73 7 42 .86 7 10. 00 15 .5 224 6 29796 9 D1sgust 13 . 5 1 1 2 . 5 13 1 .00 14 3 .71 15 35. .71 13. ,5 40. 00 8 .5 183 14 28659 12 Gui 11 1 1 .0 12 3 . 0 1 1 . 5 1 . , 25 13 4 .55 • 10 39. , 29 9. ,5 36. 67 10. .5 217 8 39990 1 Embarrassment 10 .00 13 3. .0 1 1 . 5 2 . 5 17 4 . 36 1 1 35 71 13. 5 23. 33 14 190 12 22042 15 Worry 9. .5 14 1 . 5 14 ,75 15 . 5 3 . 84 14 39. 29 9. 5 30. 00 12. 5 261 3 33341 6 Envy 7 , .0 15 5 15. .5 1 . 50 12 4 . 13 12' 39. 29 9. 5 36. 67 10. 5 132 18 15494 18 Pr 1 de 6 O 16 0 18 . 5 75 15. 5 3 . 33 17 35. 71 13. 5 3. 33 19 211 9 14337 20 Ca 1 m 4 . 5 18 0 18 . 5 0 19 2 . 75 18 21 . 43 20 6. 67 17 1 1 1 20 17409 17 Boredom 4 . 0 18 0 18 . 5 0 19 2 . 71 19 25 . 00 19 3. 33 19 1 12 19 21351 16 Respect 3 . 5 19 5 15 . 5 5 15. 5 2 . 49 20 32. 14 16. 5 3. 33 19 136 17 14635 19 Awe 2 . 0 20 0 18 . 5 0 19 3 . 46 16 28 . 57 18 10. 00 15 . 5 164 15 36560 3 T a b l e 16 C o r r e l a t i o n s f o r 20 Target Emotions Among Measures of I n t e r n a l S t r u c t u r e 1 2 3 4 5 6 7 8 1. Frequency of Free L1st1ng-a .97 .80 .80 .86 . 77 .34* .40 2 . Frequency of F r e e L1st1ng-b .98 . 76 . 70 .80 .73 .21* .32* 3 . Frequency of F r e e L1st1ng-c .89 .86 . 70 .69 . 70 .30* .28* 4 . P r o t o t y p i c a l i t y R a t i n g s .92 .93 .87 .85 .83 .56 .56 5 . P r o b a b i l i t y of "Emotion" as S u p e r o r d l n a t e - d .91 .89 .80 .88 .78 .37* .49 6 . P r o b a b i l i t y of "Emotion" as Superord1nate-e .81 .82 . 75 .88 .82 .30* .37 7 . F a m i l y Resemblance (Phase One) . 47 .46 . 39 .42 .28* .29* .44 8 . F a m i l y Resemblance (Phase Two) . 39 .41 . 30* .45 .43 .40 .34* * not s i g n i f i c a n t a t .05 l e v e l ••"Emotion I n c l u d e d as p a r t of the response" d a t a Note: C o r r e l a t i o n s above d i a g o n a l based on raw s c o r e s ; c o r r e l a t i o n s below the di a g o n a l based on ranks. a=Study One ( a l l r e s p o n s e s ) b=Study One ( f i r s t f o u r r e s p o n s e s ) c=Hunt & Hodge (1971) d="Emot1on" I n c l u d e d as p a r t of the response d a t a ( F i l l e r s ) e="Emot1on" i n c l u d e d as p a r t of the response d a t a (No f i l l e r s ) 1 1 4 l i s t i n g in a free production of exemplars task. This relationship held true when a l l Study One responses were considered, as well as when only the f i r s t four were used. The Hunt and Hodge frequency data also correlated s i g n i f i c a n t l y with p r o t o t y p i c a l i t y ratings. As already mentioned, reaction time to v e r i f y the category membership of the prototypical cases was faster than for less t y p i c a l cases. It was also predicted that "emotion" would be given as a response to the question "To what general category does th i s instance belong?" more often for prototypical than nonprototypical cases. As shown in Table 16, ratings of t y p i c a l i t y correlated s i g n i f i c a n t l y with the number of subjects who gave "emotion" as a response for each item. F i n a l l y , two sets of family resemblance scores were computed with the expectation that the prototypical exemplars would have the highest family resemblance scores. Both sets of family resemblance scores correlated s i g n i f i c a n t l y with p r o t o t y p i c a l i t y for the raw score data. Only the f i r s t set of family resemblance scores correlated s i g n i f i c a n t l y with p r o t o t y p i c a l i t y for correlations based on ranks. Family resemblance did not correlate s i g n i f i c a n t l y with many of the other measures. In several studies, measures of internal structure were available for an additional 30 emotion categories. P r o t o t y p i c a l i t y was correlated with the frequency of free l i s t i n g measures. These correlations appear in Table 17. 1 1 5 Table 17 Frequency and Pr o t o t y p i c a l i t y Correlations for 30 Categories of Emot ion 1 2 3 4 1. Frequency of Free L i s t i n g - a .75* .47* .46* 2. Frequency of Free Listing-b .68* .40* .23 3. Frequency of Free Li s t i n g - c .51* .51* .21 4. Pr o t o t y p i c a l i t y Ratings .42 .29 .21 * p <.05 Note: a=Study One ( a l l responses) b=Study One ( f i r s t four responses) c=Hunt & Hodge (1971) 1 1 6 Correlations between frequency and p r o t o t y p i c a l i t y ratings for a l l 50 emotion categories appear in Table 1 8 . If "emotion" was definable in the c l a s s i c a l sense, the category members should not have varied in how much they came to mind. Moreover, i t would have been meaningless for subjects to rate instances for goodness of example, since within a c l a s s i c a l view each instance i s equally representative of the category. S i m i l a r l y , based on the assumption of equivalence of category members, one would not have expected d i f f e r e n t i a l response rates in reaction times to v e r i f y category membership. Each category member should have an equal probability of e l i c i t i n g the category name i f a l l instances were equal, which was not the case in this study. F i n a l l y , within the c l a s s i c a l view, category members would not demonstrate a family resemblance relationship, but would rather possess a common set of c r i t e r i a l a t t r i b u t e s . Based on the converging results obtained in this study, i t appears that the concept "emotion" is more amenable to a prototype, than a c l a s s i c a l , conceptualization. Conclusion The aim of the research reported here was to test the f e a s i b i l i t y of viewing the concept of emotion from the perspective of Rosch's theory of concepts. Because no research had previously been done on thi s topic, these studies were highly exploratory. And yet the results could hardly have been more encouraging. Nearly a l l of the predictions derived from Rosch's theory worked well when 1 1 7 Table 18 Frequency and P r o t o t y p i c a l i t y Correlations for 50 Categories of Emotion 1 2 3 4 •1. Frequency of Free List i n g - a .97 .84 .77 2. Frequency of Free Listing-b .91 .81 .66 3. Frequency of Free Li s t i n g - c .87 .83 .66 4. Pr o t o t y p i c a l i t y Ratings .78 .71 .72 Note: A l l correlations are s i g n i f i c a n t at .01 l e v e l . a=Study One ( a l l responses) b=Study One ( f i r s t four responses) c=Hunt & Hodge (1971) 1 18 extrapolated to and tested in the domain of emotion. The focus so far has been on the issue of a d e f i n i t i o n of emotion. And c e r t a i n l y the success of working within Rosch's framework contrasts with the fr u s t r a t i o n expressed by the writers who attempted to define "emotion" from a c l a s s i c a l perspective of c r i t e r i a l features. It would appear that attempts to c l a s s i c a l l y define the everyday word "emotion" are unlikely to succeed in the future. Atempts to specify the border of the concept emotion (such that, for example, "anger" is and "pride" i s not an emotion), or to specify boundaries between s p e c i f i c emotions, are also unlikely to succeed. Attempts to achieve a d e f i n i t i v e l i s t of "the emotions" w i l l probably not meet with success, either. Yet the real gain from the evidence uncovered here concerns not the e x p l i c i t theory of the experts, but the i m p l i c i t theory of the layperson. This evidence provides a new and interesting picture of how people think about emotions. People can be thought of as possessing , or capable of creating, an i m p l i c i t taxonomy for the categorization of emotional states. The taxonomy i s h i e r a r c h i c a l l y organized but in quite a simple way. At one l e v e l , superordinate, is the category emotion. Below that l e v e l is a middle l e v e l consisting of a large but indeterminate number of categories such as "happiness", "love", "anger", and "fear". Both of these levels are coded with single words in the English lanugage and are thus e a s i l y used. Evidence here indicated that people may be able to create an even lower level by 1 1 9 subdividing the middle level categories, but few such subdivisions are coded in English and subjects do not agree with one another on their subdivisions. In other words, there is no ready-made scheme. Similarly, people can create a le v e l between the middle and superordinate, but again there i s no ready-made scheme and no pre-coded categories.. The major di v i s i o n s were pleasant, neutral, and unpleasant, although other d i v i s i o n s were used as well. Notice that unpleasant-pleasant i s actually just a single feature, and therefore may be c l a s s i c a l l y , rather than pr o t o t y p i c a l l y , defined. Many subordinate l e v e l categories, such as " i r r a t i o n a l anger" were also of thi s sort. Evidence here also indicated that people think of anger, happiness, and love as better examples of an emotion than pride, awe, and respect. This is so even though the l a t t e r three are, nonetheless, s t i l l c l a s s i f i e d as emotions by various c r i t e r i a . These better examples come to mind more readily than do poorer examples when the word "emotion" i s mentioned. People hesitate when v e r i f y i n g that the poorer examples are emotions. These behaviors point to the psychological r e a l i t y of what Rosch c a l l s the internal structure of a concept. If this evidence for the internal structure of emotion i s replicated, then there i s strong reason to believe that subjects w i l l show other sorts of behaviors that have been associated with internal structure. For example, Mervis and Rosch (1981) hypothesize that we d i s t o r t our memory toward the prototype. 120 Thus an emotional incident might be remembered as more pro t o t y p i c a l l y anger, or love, or whatever, than i t actually was. Rips (1975) showed that internal structure influences inductive judgements. Subjects were more w i l l i n g to generalize from prototypical members of a category to nonprototypical members than vice-versa. Told that robins (prototypical birds) have a new disease, subjects were more w i l l i n g to guess that ducks (nonprototypical birds) w i l l catch the disease than they were w i l l i n g to generalize to robins when told that ducks have the disease. One would expect internal structure to influence judgments in the domain of emotion as well. Because Rosch's theory has proved successful in the domain of emotions so far, i t may be worthwhile to mention two additional hypotheses not addressed in the present set of studies. F i r s t , we might hypothesize that actual emotional states consist of a large number of features, no one of which defines "anger", "love", or any other middle l e v e l emotion category. Just as these middle l e v e l categories vary in their representativeness of the superordinate "emotion" ( i . e . as shown here), real world emotional events vary in the extent to which they are representative of "anger", etc. Some occasions involve prototypical anger states, whereas others involve states that only resemble anger to a limited degree. Similar hypotheses can be generated for behaviors, for example (some are more representative of anger than others), 121 or for f a c i a l expressions (some are better expressions of anger than others). A second hypothesis derived from Rosch's theory i s that the middle l e v e l of the emotion hierarchy is a basic l e v e l . According to Rosch and Mervis, "the most cognitively e f f i c i e n t , and therefore the most basic l e v e l i s that at which the information value of attribute clusters i s maximized" (Mervis and Rosch, 1981, p.92). Rosch et a l . (1976a) found that the number of attributes generated between h i e r a r c h i c a l levels varied, with more attributes being generated at the basic l e v e l . Concerning the content of at t r i b u t e s , they discovered that for superordinate categories, attributes of a very general nature were provided. At the basic l e v e l , s i g n i f i c a n t l y more nouns and adjectives were used. Those few attributes which were added at the subordinate l e v e l were almost exclusively adjectives. In addition to free l i s t i n g of attributes for object names, subjects were required to l i s t a t tributes for v i s u a l l y present objects in the Rosch et a l . (1976a) study. The attributes l i s t e d in t h i s s i t u a t i o n did not d i f f e r from those previously generated from memory. Rosch et a l . (1976a) have extended these findings to various other domains. For example, they found that the number of motor movements when interacting with the various categories d i f f e r e d s i g n i f i c a n t l y between categories, such that more motor movements were given for basic l e v e l categories. This number did not s i g n i f i c a n t l y increase at the subordinate l e v e l . The 122 results were replicated when a l i v e model performed the actions. These researchers also overlapped two-dimensional l i n e drawings of items in each category. A large and consistent increase in s i m i l a r i t y of the overall look of objects was obtained for basic l e v e l over superordinate categories. A s i g n i f i c a n t , but s i g n i f i c a n t l y smaller, increase in s i m i l a r i t y was obtained for subordinate over basic l e v e l categories. Similar results were obtained when the pictures were superimposed and the l i n e s averaged. Basic l e v e l averaged shapes were guessed c o r r e c t l y more times than their corresponding superordinate shape. Again, there was no evidence to suggest that subordinate l e v e l shapes were i d e n t i f i e d better than basic l e v e l . How might one attempt to establish basic l e v e l of emotion categories? A dire c t application of Rosch's techniques seems feasible for priming experiments. She found that in a task requiring recognition of a basic l e v e l category name, priming with the basic l e v e l name f a c i l i t a t e d responding to a greater extent than priming with either the superordinate or subordinate category names (Rosch et a l . , 1976a). These findings were replicated when the dependent measure was response latency in picture recognition. In the present research, subjects were required to generate attributes only for the middle l e v e l catgories. In future research, subjects could be asked to generate 1 23 a t t r i b u t e s f o r s u p e r o r d i n a t e , middle, and subordinate c a t e g o r i e s . If the c o n j e c t u r e that c a t e g o r i e s l i k e "happiness", "anger", and " l o v e " are b a s i c l e v e l i s t r u e , we would expect that more a t t r i b u t e s would be generated for these c a t e g o r i e s than f o r the su p e r o r d i n a t e category emotion. Concerning the content of a t t r i b u t e s , we would expect a t t r i b u t e s of a very g e n e r a l nature to be generated f o r "emotion". At the b a s i c l e v e l , we would expect more nouns and a d j e c t i v e s , while a t t r i b u t e s generated f o r the subordinate l e v e l would mostly be a d j e c t i v e s . B a s i c l e v e l i s l e a r n e d f i r s t developmentally and i s coded f i s t i n languages h i s t o r i c a l l y . I f our middle l e v e l c a t e g o r i e s are b a s i c l e v e l , we would expect c h i l d r e n to l e a r n c a t e g o r i e s l i k e "happy", "sad", and "angry" before c a t e g o r i e s l i k e "emotion" ( s u p e r o r d i n a t e ) or "contentment" (a subordinate of "happiness"). B a s i c l e v e l i s more l i k e l y to appear c r o s s - c u l t u r a l l y than other l e v e l s . T h i s f i n d i n g c o u l d a l s o be t e s t e d i n the domain of emotion concepts. In sum, an o p t i m i s t i c p i c t u r e has been p a i n t e d of emotion as a Roschian concept. In order to make t h i s case with g r e a t e r c e r t a i n t y , the s t u d i e s here should be r e p l i c a t e d and the new hypotheses generated should be t e s t e d . 1 24 References A v e r i l l , J.R. A semantic atlas of emotional concepts. JSAS Catalogue of Selected Documents in Psychology, 1975, 5, 330. (Ms. No.421). Calfee, R.C. Human Experimental Psychology. New York: Holt, Rinehart & Winston, 1975, p.222. Cannon, W.B. Neural organization for emotional expression. In M.L. Reymart (Ed.), Feelings and Emotions: The Wittenberg Symposium. Worchester, Mass.: Clark University Press,1928, pp.257-269. Cantor, N. & Mischel, W. Prototypes in person perception. Advances in Experimental Social Psychology, 1979, J_2, 3-52. Cantor, N. & Mischel, W. Tr a i t s as prototypes: Effects on recognition memory. Journal of Personality and Social Psychology, 1977, 35, 38-48. Cantor, N., Smith E.E., French R.D., Mezzich J., Psychiatric diagnosis as prototype categorization. Journal of Abnormal Psychology, 1980, 89, 181-193. Claparede, E. Feelings and emotions. In M.L. Reymart (Ed.), Feelings and Emotions: The Wittenberg Symposium. Worchester, Mass.: Clark University Press, 1928, pp.124-139. Duffy, E. Emotion: An example of the need for reorientation in Psychology. Psychological Review, 1934, 4 J , 184-198. Duffy, E. An explanation of "emotional" phenomena without the use of the concept "emotion". The Journal of General 1 25 Psychology, 1941, 25, 283-293. E l l i s , A. Reason and Emotion in Psychotherapy. New York: Lyle Stewart, 1962. Franks, J.J. & Bransford, J.D. Abstraction of vi s u a l patterns. Journal of Experimental Pychology, 1971, 90, 65-74. Hebb, D.O. Emotional disturbances. In The Organization of Behavior: A Neurpsychological Theory. New York: Wiley, 1949, pp. 235-274. Heider, E.R. "Focal" color areas and the development of color names. Developmental Psychology, 1971, £, 447-455. Heider, E.R. Universals in color naming and memory. Journal of Experimental Psychology, 1972, 93, 10-20. Hilgard, E.R. Emotion and motivation. In Introduction to Psychology. New York: Harcourt, Brace, and Company, 1953. Hunt, K.P. & Hodge M.H. Category-itern frequency and category-name meaningfulness: Taxonomic norms for 84 categories. Psychonomic Monograph Supplements, 1971, Vol 4, (6), (Whole No. 54), pp. 97-121. Izard, C.E. & Beuchler, S. Emotion expressions and personality integration in infancy. In C.E. Izard (Ed.), Emotions in Personality and  Psychopathology. New York: Plenum Press, 1979, pp.447-466. James, W. The V a r i e t i e s of Religious-Experience: A Study in  Human Nature. New York: Longmans, Green & Co., 1929. 126 James, W. What is emotion? Mind, 1884, J_9, 188-205. Lakoff, G. Hedges: A study in the meaning of c r i t e r i a and the logic of fuzzy concepts. Journal of Philosophical Logic, 1973, 2, 458-508. Lazarus, R.S. A cognitively oriented psychologist looks at feedback. American Psychologist, 1975, _3_9_' 553-561. Mervis, C.B., C a t l i n , J., & Rosch, E. Development of the structure of color categories. Developmental Psychology, 1975, JM, 54-60. Mervis, C.B. & Rosch, E. Categorization of natural objects. Annual Review of Pychology, 1981, 32, 89-115. Neisser, U. The concept of i n t e l l i g e n c e . Intelligence, 1979, 3, 217-227. Plutchik, R. Emotion: A Psychoevolutionary synthesis. New York: Harper & Row, 1980. Posner, M.I., Goldsmith, R., & Welton, K.E. Perceived distance and c l a s s i f i c a t i o n of distorted patterns. Journal of Experimental Psychology, 1967, 7_3, 28-38. Reed, S.K. Pattern recognition and categorization. Cognitive Psychology, 1972, 3, 382-407. Rips, L.J. Inductive judgments about natural categories. Journal of Verbal Learning and Verbal Behavior, 1 975, J_4, 665-681 . Rosch, E.H. On the internal structure of perceptual and semantic categories. In T.E. Moore (Ed.), Cognitive Development and the Acquisition of Language. New York: Academic Press, 1973, pp.111-144. 1 27 Rosch, E. L i n g u i s t i c r e l a t i v i t y . In A S i l v e r s t e i n (Ed)., Human Communication: Theoretical Explorations. H i l l s d a l e , N.J.: Erlbaum Press, 1974, pp. 95-121. Rosch, E. Cognitive reference points. Cognitive Psychology, 1975, 7, 532-547. (a) Rosch, E. Cognitive representations of semantic categories. Journal of Experimental Psychology: General, 1975, 104, 192-233. (b) Rosch, E. & Mervis C.B. Family resemblances in the internal structure of categories. Cognitive Psychology, 1976, 7, 573-605. Rosch, E., Mervis, C.B., Gray, W.D., Johnson, D.M., & Boyes-Braem, P. Basic objects in natural categories. Cognitive Psychology, 1976, 8, 382-439. (a) Rosch, E., Mervis, C.B., Gray, W.D., Johnson, D.M., & Boyes-Braem, P. Basic objects in natural categories. (Working Paper No. 40) Language and Behavior Research Laboratory. University of C a l i f o r n i a , Berkeley, Ca. 94720. (b) Rosch, E., Simpson, C , & M i l l e r , R.S., Structural bases of t y p i c a l i t y e f f e c t s . Journal of Experimental Psychology:  Human Perception and Performance, 1976, 2, 491-502. (c) Rosch, E. Human categorization. In N. Warren (Ed.), Studies in Cross-Cultural Psychology (Vol.1). London: Academic Press, 1978. (a) Rosch, E. Princip l e s of categorization. In E. Rosch and B.B. Lloyd (Eds.), Cognition and Categorization. H i l l s d a l e , 128 N.J.: Lawrence Erlbaum Association, 1978. (b) Rosch, E. Prototype c l a s s i f i c a t i o n and l o g i c a l c l a s s i f i c a t i o n : The two systems. Paper presented at a meeting of the Jean Piaget Society, Philadelphia, May, 1981. Schacter, S. Cognitive e f f e c t s on bodily functioning. In D.C. Glass (Ed.), Neurophysiology and Emotion. New York: Rockfeller University Press, 1967. Skinner, B.F. The Behavior of Organisms: An Experimental Analysis. New York: Appleton-Century Crofts, 1938, pp. 406-409. Strongman, K.T. The Psychology of Emotion. London: John Wiley & Sons, 1973. Watson, J.B. Hereditary modes of response: Emotion. In Psychology from the Standpoint of a Behaviorist. Philadelphia, Lippincott, 1924 (2nd ed.), pp. 194-230. Wenger, M.A. Emotional behavior. In Wenger, M.A., Jones, F.N., & Jones, M.H. Physiological Psychology. New York: Holt, Rinehart, and Winston, 1956, pp.339-359. Young, P.T. Nature and bodily mechanisms of emotion. In Motivation and Emotion: A Survey of the Determinants of Human and Animal A c t i v i t y . New York: John Wiley & Sons, 1961, pp. 344-410. 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0095323/manifest

Comment

Related Items