Articulation, the Letter, and the Spirit in the Aesthetics ...
What Would They Think?
A Computational Model of Personal AttitudesArticulation, the Letter, and the Spirit
in the Aesthetics of Narrative
Hugo Liu
MIT Media Laboratory
20 Ames St., Cambridge, MA
hugo@media.mit.edu
Pattie MaesHugo Liu
MIT Media LaboratoryMIT Media Laboratory
20 Ames St., Cambridge, MAMassachusetts Institute of Technology
pattiehugo@media.mit.edu
ABSTRACT
A furthered understanding of the aesthetic aspects of narrative is important to both people and machines who wish to author pleasing narratives. This paper gives an account of the aesthetics of narrative employing the triptych of articulation, the letter, and the spirit as a framework for understanding. The rhetoric of the letter and the spirit, with great intellectual precedent, is used in this work to segregate the mundane and habitual aspects of narrative (the letter) from narrative’s mystified, mythical, and aesthetic aspects (the spirit). Articulation, understood as the interplay between the letter and the spirit, has certain aesthetic modes, and these modes and their relationship to connotation, defamiliarization, and myth are discussed.
Also central to the aesthetic qualification of articulation is the cultural and cognitive backdrop against which an articulation occurs. This paper will argue that in the culture of our contemporary period, media-driven commodification of narratives has led to the saturation of the cultural narrative space with cliché. If a narrative is to be aesthetic in this environment, it must face the additional challenge of resisting hyperarticulation, as hyperarticulation invites unflattering comparison to known narrative forms and techniques. This paper concludes with a presentation of four strategies which support aesthetic articulation in narratives under the current cultural context – intertextuality, unusual representation, aesthetic signature, and personalization.
Categories and Subject Descriptors
J.5 [Arts and Humanities]: Literature, Fine arts.
General Terms
Theory, Design, Human Factors.
Keywords
Articulation, aesthetics, narrative theory, myth, creativity, spirit.Understanding the personalities and dynamics of an online community empowers the community’s potential and existing members. This task has typically required a considerable investment of a user’s time combing through the community’s interaction logs. This paper introduces a novel method for automatically modeling and visualizing the personalities of community members in terms of their individual attitudes and opinions.
“What Would They Think?” is an intelligent user interface which houses a collection of virtual representations of real people reacting to what a user writes or talks about (e.g. a virtual Marvin Minsky may show a highly aroused and disagreeing face when you write “formal logic is the solution to commonsense reasoning in A.I.). These “digital personas” are constructed automatically by analyzing personal texts (weblogs, instant messages, interviews, etc. posted by the person being modeled) using natural language processing techniques and commonsense-based textual-affect sensing.
Evaluations of the automatically generated attitude models are very promising. They support the thesis that the whole application can help a person form a deep understanding of a community that is new to them by constantly showing them the attitudes and disagreements of strong personalities of that community.
Categories and Subject Descriptors
H.5.2 [Information Interfaces and Presentation]: User Interfaces – interaction styles, natural language, theory and methods, graphical user interfaces (GUI); I.2.7 [Artificial Intelligence]: Natural Language Processing – language models, language parsing and understanding, text analysis.
General Terms
Algorithms, Design, Human Factors, Languages, Theory.
Keywords
Affective interfaces, memory, online communities, natural language processing. commonsense reasoning.
INTRODUCTIONINTRODUCTION
Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee.
SRMC’04, October 15, 2004, New York, New York, USA.
Copyright 2004 ACM 1-58113-931-4 /04/0010...$5.00.
Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee.
IUI’04, January 13-16, 2004, Island of Madeira, Portugal.
Copyright 2004 ACM.
Narratives are not only communicative, they also have aesthetic value. Intrinsic in a narrative is motivation for its existence – why it is being told, why it demands to be heard, and how it may be appreciated. Even if a narrative is meant primarily to communicate concepts, such as narratives belonging to the genre of news, a good narrative nevertheless obeys an aesthetic imperative, albeit an invisible one, governing aspects of the narrative such as, inter alia, point-of-view, selective articulation of details, consistency and coherence, dramatic storytelling technique, and the perception of authenticity. Crafting these aesthetic aspects of narrative is not only the trade of the storyteller, but also her art.
Given that the culture of our contemporary period has all but commodified narratives, and given that artificial intelligence and media technology are beginning to enable the mechanization of narrative production, there is some concern that the aesthetics of narrative might continue to take a back seat to its informational aspects, just as turn of the twentieth century technologies for mechanical reproduction prompted many, including Walter Benjamin, to be concerned for the diminutive caveats that those technologies bore for the “aura” of art (1935). Our present concern for the fate of the aesthetic aura of narrative stems from the observation that commodification and mechanization privilege (to indulge in Derridian discourse) the cleaner abstractions of concepts over the liminal, more-difficult-to-quantify contributions of amorphous meaning; yet the origin of narrative aesthetics, as we shall reveal in this paper, lies largely in the amorphous, and is actually weakened by conceptualization. As our contemporary information-obsessed culture plows on with its trend toward conceptualization, will the aesthetics of narrative experience decline, or can they survive? If so, what might be the terms and strategies of their survival?
In pursuing these questions, our inquiry will first focus on the role that modes of articulation play in the aesthetics of narrative, and in art in general. Though we will more fully develop the notion of articulation as a multi-faceted theme, we offer the following as a summary and working definition. Articulation is the application of some interpretative apparatus to distill an amorphous space of meaning into discrete and bounded concepts. Expressed in this way, articulation and our use of the term conceptualization are nearly synonymous, with the only barrier to their interchange being that “articulation” emphasizes a process, while “conceptualization” highlights the eventualities and results of said process.
A corollary notion we introduce is that of hyperarticulation. Because narratives have been packaged into univocal commodities in the information age, the collective memory of the mass culture is arguably more saturated than ever before with caricatures of all the known narrative forms and techniques, with exemplars drawn from the medias of music, film, television, and news. This saturation of our collective memory of narratives, we will argue, constitutes a pollutive backtext which endangers the aesthetic power of storytelling. The greater the degree of articulation in a narrative, the more susceptible the narrative is to being identified with clichés of form, content, and technique. Hyperarticulation, then, is the failure of a narrative to steer clear of the unaesthetic space of narrative clichés.
The position of this paper is that in the context of our contemporary period, only by resisting hyperarticulation can the aesthetics of narrative survive. The subject and implications of this position paper will be of interest to the artificial intelligence narrative understanding and generation community because it attempts to characterize the aesthetic consequences of employing overly explicit narrative representations, and will toward the end of the paper examine some computationally-suggestive strategies for preserving narrative aesthetics, centered around the principle of resisting hyperarticulation. The art and literary criticism communities will be interested in the novel way that cognitive and computational perspective are infused into thinking about narrative aesthetics and articulation.
The rest of this paper begins by first synthesizing a compelling thematics for the notion of articulation and its consequence to aesthetics. Second, we analyze the nature of articulation within the cultural context of the contemporary period, putting forth the position that the space of narratives today is saturated and that this poses specific challenges for the aesthetics of narrative. We conclude by nominating four strategies for preserving narrative aesthetics through the resistance of hyperarticulation.Entering an online community for the first time can be intimidating if a person does not understand the dynamics of the community and the attitudes and opinions espoused by its members. Right now, there seems to only be one option for these first-time entrants – to comb through the interaction logs of the community for clues about people’s personalities, attitudes, and how they would likely react to various situations. Picking up on social and personal cues, and overgeneralizing these cues into personality traits, we begin to paint a picture of a person so lucid that we seem to be able to converse with that person in our heads. Gaining understanding of the community in this manner is time consuming and difficult, especially when the community is complex. For the less dedicated, more casual community entrant, this approach would be undesirable.
[pic]
Figure 1. Virtual personas representing members of the AI community react to typed text. Each virtual persona’s affective reactions are visualized by modulating graphical elements of the icon.
In our research, we are interested in giving people at-a-glance impressions of the attitudes of people in an online community so that they can more quickly and deeply understand the personalities and dynamics of the community.
We have built a system that can automatically generate a model of a person’s attitudes and opinions from an automated analysis of a corpus of personal texts, consisting of, inter alia, weblogs, emails, webpages, instant messages, and interviews. “What Would They Think?” (Fig. 1) displays a handful of these digital personas together, each reacting to inputted text differently. The user can see visually the attitudes and disagreements of strong personalities in a community. Personas are also capable of explaining why they react as they do, by displaying some text quoted from that person when the face is clicked.
To build a digital persona, the attitudes that a person exhibits in his/her personal texts are recorded into an affective memory system. Newly presented text triggers memories from this system and forms the basis for an affective reaction. Mining attitudes from text is achieved through natural language processing and commonsense-based textual affect sensing (Liu et al., 2003). This approach to person modeling is quite novel when compared to previous work on the topic (cf. behavior modeling, e.g. (Sison & Shimura, 1998), and demographic profiling, e.g. questionnaire-derived user profiles).
A related paper on this work (Liu, 2003b) gives a more thorough technical treatment of the system for modeling human affective memory from personal texts. This paper does not dwell on the implementation-level details of the system, but rather, describes the computational model of attitudes in a more practical light, and discusses how these models are incorporated to build the intelligent user interface “What Would They Think?”.
This paper is structured as follows. First, we introduce a computational model of a person’s attitudes, a system for automatically acquiring this model from personal texts, and methods for applying this model to predict a person’s attitudes. Second, we present how a collection of digital personas can portray a community in “What Would They Think?” and an evaluation of our approach. Third, we situate our work in the literature. The paper concludes with further discussion and presents directions for future work.
THEMATICS OF ARTICULATIONCOMPUTING A PERSON’S ATTITUDES
We preface this section by recapitulating our initiatory understanding of articulation as a point of departure. Articulation is the application of some interpretative apparatus to distill an amorphous space of meaning into discrete and bounded concepts.
1 Articulation as
Interplay between Letter and Spirit
The amorphous space of meaning is that rich tapestry of the sensorial, affective, chimerical, and reminiscent existing in every mind, eluding language, and escaping the shackles of definition. The amorphous space, alive with meaning, intimate, and fraught of creative tensions, is a place where the aesthetic, especially of a personal nature, resonates. Because it is at once alive and lively yet ephemeral and elusive, the amorphous is akin to a spirit. If the spirit has an archenemy, his name is letter, that which is definite, rigid, inanimate, and existing only as a fossilized social construction. The letter refers not only to words and language, but also to culture and any social constructs characterized by order and definition. In narratives, the letter is seen in the explicit aspects of the text, such as the words themselves, and in clichéd techniques and forms which are used to structure the narrative. A cliché is an entity which is known to the social consciousness and is thus easily recognizable. While most clichés belong to the letter, a noteable exception is a special cliché called myth, which belongs dually to the letter because it is structured and familiar, but also to the spirit because it is ritualized, sacred, and possesses a wisdom which transcends the mundane. We will defer further discussion of myth until a later point in order to preserve a cleaner characterization of letter and spirit in the present discussion.
The rhetoric of the letter versus the spirit has been used widely throughout history and discourse, though all uses remain remarkably consistent in their characterizations of letter, spirit, and their interplay. As far back as the Bible, the apostle Paul advocated “a new covenant, not of the letter but of the spirit; for the letter kills, but the Spirit gives life” (2 Corinthians 3.6). By “letter,” Paul was likely referring to the Torah, whose rigidity of law and detachment from the self he protested. By contrast, Paul probably meant the Spirit to be something much more capable of personal affectation than the external letter. As the system of law emerged out of religion, this rhetoric was inherited and transcoded into the tension of “the letter versus the spirit of the law.” This rhetorical export illustrates that even in the legal system, where the orderliness of signifiers is highly privileged, there is an admission that laws viewed as narratives, embody more than only the literal aspects of that text; there is also an unwritten, unarticulated spirit which lives intertextually and is somehow more authentic, authoritative, and more aesthetic in nature than the letter of the law.
In the post-structuralist treatises of Lacan and Jameson, the interplay between letter and spirit is even more apparent. The spirit is understood as a mythical source of meaning which underlies the letter and motivates its articulation. The spirit is always forlorn for articulation because that process provides catharsis for the creative tensions which brood within the spirit. But it is easy to forget that it is this same tension which animates the spirit; thus the act of articulation, on the one hand a vehicle of catharsis for the spirit, is also its betrayer and executioner.
The letter is endowed with order and definition, and is inherently a socially constructed and public entity, so articulation into letter allows an idea to be freely communicated; however, there are also disquieting and unaesthetic qualities about the letter. In “The Agency of the Letter in the Unconscious,” (1966), Lacan likens language to bondage, and says that all subjects are “slaves of language.” Jameson extends Lacan’s portrayal by painting the letter as “ideological” and arguing that “ideological closure” undermines a mind’s capability to make new and original meanings (1972).
The relationship between spirit and letter can also be interpreted in terms of Lacan’s three phases of human development from infant to adult: the Real, the Imaginary, and the Symbolic (the trajectory of this series is a progression away from Nature and into Culture). The Real is a place of psychic fullness and completeness, and a state of Nature, and the home of the spirit. By contrast, the Symbolic is a place dominated by the letter and Culture, and is characterized by Lacking rather than fullness. Lacan describes the movement out of the Real into the Symbolic as an irreversible and irretrievable loss of naiveté. Movement from the Real to the Symbolic parallels the process of articulation, in which a wanton and multivocal spirit is shackled and defined into letter. Like the former process, the latter is irreversible, though perhaps tenuously recoverable through forgetting and disintegration.
In summary, the spirit is the amorphous space of meaning that is the substrate to the articulation process. It is personal, alive, and full of tension, and thus the preferred dwelling of the aesthetic. In contrast, the letter is a socially constructed entity with definition; and the system of language, while facilitating communication, is also an oppressive force which ties down the spirit. The spirit craves articulation into the letter for catharsis, but full articulation endangers the spirit, which is d.o.a., dead on articulation.
Why is the spirit such a source of power for a narrative, and why does the aesthetic side with it rather than with the letter?
2 Partial Articulation and
the Power of the Spirit
It is not quite right to say that the spirit exists in a narrative but remains completely unarticulated. If that were the case, it would follow that the spirit might exist completely detached from the perceivable aspects of the text, and the sense is that this is not the case. At the very least, the spirit has a hand in shaping the deliverance of the literal aspects of the text, and more judiciously, the spirit exists partially articulated through connotations and contexts. Partial articulation nudges the reader toward a particular interpretation and appreciation of a narrative without the obtuseness of explicit exposition. The freedom of discovery is preserved for the reader, and thusly a narrative with a strong partial articulation of spirit engages more of a reader’s psychic energies than a narrative which has fully articulated a spirit (by offering pre-interpretation rather than nurturing original interpretation), or one altogether devoid of spirit.
That the spirit entices the reader into discovering it, then, affords the spirit its hypnotic power. The literature of psychoanalysis is particularly multiloquent on the power of the spirit, especially the writings of Jung. One way to interpret the meta-narrative given by Jung in Symbols of Transformation (1912) and Two Essays on Analytical Psychology (1956) is to view the spirit as that which pervades Jung’s realm of symbols. For Jung, symbols are mystified and amorphous, like the ourobouros, and for that reason, they possess “symbolic power” and stay active in the mind, becoming a “transformer of consciousness.” They invite personal interpretation and are free to be associated with personal experiences because they lack clearly delineated a prioris. The equivalent of full articulation into letter would be the attachment of a symbol to a singular static meaning, and Jung explains that in doing so, the psychic powers of the symbol are vanquished.
The spirit can also been seen in the Jungian notion of a psychic libido. In his theory of opposites, Jung conceptualizes mental energy, collectively known as the libido, as the product of the conflict between opposites. He declares that “there is no energy unless there is a tension of opposites” (Jung, Two Essays, 63). The importance of tension to the spirit is paralleled in Blake’s meta-narrative, advocating tension and opposition as sources of creativity.
In addition to Jungian symbols and libido, a third source of power for the spirit is myth. Myth is unique because it possesses both the letter and the spirit, being at once familiar and mystified (however, it naturally resists habitualization). In narratives, myths are frequently employed as patterns to structure storytelling activity.
While a portion of the realm of spirit is determined by a personal unconscious – that unarticulated body of one’s own thoughts, feelings, and urges not usually accessible to consciousness – and both Jung and Freud agree on the existence of a personal unconscious, Jung further postulates that there also exists a collective unconscious inherited and shared by all minds, the basic compositional unit of which is a psychological archetype, a universal pattern imprinted on the collective psyche, which represents a single unit of myth called a mythos. Regardless of the validity of Jung’s ideology about the origin of mythical archetypes, myths possess the spirit because they are widely understood as universal themes, patterns, narratives, and images which are profound because they transcend time and culture.
Like libido, myth is alive, and it is the eternal tension between the everyday and the divine which sustains curiosity in it (a tension not likely to ever be resolved). To some extent of course, myth also possesses the letter, because people are vaguely aware of what it is, and can recall with great familiarity the many instances where it is uttered and recapitulated throughout the arts. However, unlike other manifestations of the spirit, myth resists death-upon-articulation. This is because myth describes that which recurs eternally inside and outside of us, and because it is an omen of the divine, it inherently is impervious to full resolution and definition. Unlike most symbols, myth refuses to be tied down to static meaning; and in this sense, as frequently as mythical patterns are found in narratives, they remain forever partially articulated entities, always retaining the power to intrigue.
However, this is not to say that myth cannot be parodied or caricatured. Certainly over-essentializations and distortions of myth lack the spirit of a genuine myth and thus fall more into the realm of the letter. An important quality of genuine myth is that it remains unconscious rather than awkwardly explicated and forcing conscious consideration, for articulating myth into consciousness corrupts its intimate nature and depletes its psychic power.
Myth is sacred and ritualistic. In Awakening the Heroes Within, Pearson nominates twelve common archetypes which pervade art, literature, and other media: the Innocent, the Orphan, the Warrior, the Caregiver, the Seeker, the Destroyer, the Lover, the Creator, the Ruler, the Magician, the Sage, and the Fool. This cast of characters has been thoroughly ritualized and reappears in narratives across time and place. For example, the Warrior can be seen in King Arthur in the middle ages, and again in Luke Skywalker in contemporary times. The myth of star-crossed lovers has been ritualized and recapitulated from Romeo and Juliet to appearing in nearly every theatrical musical to almost every Bollywood film. We observe that while myth is recognizable, it nonetheless, being of the spirit, remains in the subtext; the Warrior myth is only projected unto Luke Skywalker, and the star-crossed lovers myth is only projected onto the characters of a musical.
While certain characters and story patterns are recurrent elements of myth, their rote invocation in narrative does not guarantee that the resulting narrative has any mythical quality. According to Campbell, “You can keep an old tradition going only by renewing it in terms of current circumstances;” (Campbell, The Power of Myth, 26) this advice can be interpreted to mean that in composing new narratives, old mythical archetypes must be invoked in a way such that they make natural sense with respect with the gestalt context of the new narrative. Above all, narrative authenticity is most important to the engagement of the reader’s energies and psyche; only when a reader has intimately embraced a narrative can we entertain the possibility of awakening the mythical spirit.
In summary, the spirit pervades the connotations, contexts, and subtexts, and manners of speech of narratives, and should remain only partially articulated. The power of the spirit lies in the troika of symbol, libido, and myth. As mystified symbol, the spirit invites and entices interpretation without being tied down to static definition. As libido, the spirit feeds on the creative tension produced by the conflict of oppositions. As myth, the spirit is most resistant of mundane habituation, as the wisdom of universal eternally recurring themes and patterns of ritualistic magniloquence hypnotize the reader in spite of any recognition of pattern.
Our approach to modeling attitudes is based on the analysis of personal texts using natural language parsing and the commonsense-based textual affect sensing work described in (Liu et al., 2003). Personal texts are broken down into units of affective memory, consisting of concepts, situations, and “episodes”, coupled with their emotional value in the text. The whole attitudes model can be seen as an affective memory system that valuates the affect of newly presented concepts, situations, and episodes by the affective memories they trigger.
In this section, we first present a bipartite model of the affective memory system. Second, we describe how such a model is acquired automatically from personal texts. Third, we discuss methods for applying the model to predict a user’s affective reaction to new texts. Fourth, we describe how some advanced features enrich our basic person modeling approach.
3 The Aesthetics of Articulation A Bipartite Affective Memory System
The spirit, rather than the letter, is the primary dwelling of the aesthetic, as evidenced by the observation that personals aesthetics itself is beyond full articulation; thus it follows that the narrative (or any art for that matter), which aims to be aesthetic, must be imbued with the spirit.
A primary reason why the spirit should be more sympathetic toward the aesthetic than the letter is because the spirit entices the reader to grasp it and articulate it or at the very least to acknowledge and laud its presence; by contrast, grasping the habituated and known letter is a mundane exercise. In his treatise on art criticism entitled Art as Experience (1934), Dewey views art as that which engages a subject into active perception, rather than a passive and mundane recognition, paralleling our present narrative on the engagingness of the spirit versus the fixity of the letter.
That the spirit is aesthetic because it activates the mind is also echoed in Shklovsky’s Art as Technique (1917). For Shklovsky, the letter is seen in the habituation of perception, which he describes as a “process of 'algebrization,' the over-automatization of an object.” In light of the numbness of the letter, “art exists that one may recover the sensation of life; it exists to make one feel things, to make the stone stony.” Because full articulation has a literalizing effect, it promotes habitualized perception rather than artful perception. To imbue a narrative or art with spirit, Shklovsky might suggest that we de-articulate the letter so that it has more spirit: “The technique of art is to make objects 'unfamiliar,' to make forms difficult, to increase the difficulty and length of perception because the process of perception is an aesthetic end in itself and must be prolonged.”
The technique of de-articulation, or de-familiarization as it is often called, can be seen widely through art (especially modern art), literature, and media. Its goal is to re-aestheticize the mundanely familiar by de-identifying the signifier, and shifting the signified meaning underneath the signifier. The Surrealist artist Magritte, for example, captions an image of a smoking pipe with “Ceci n’est pas une pipe;” by denying the image the power to signify a real smoking pipe, Magritte calls into question the signifying power of the image. Furthermore, he de-authoritizes our habitual assumption that the word “pipe” even points to a smoking pipe at all; after all, “Pipe” in French carries a second slang entendre of “blowjob,” and Magritte tempts us with this tabooed re-interpretation which also preserves the tautology of the caption. In other examples, the metafiction of Nabakov, Borges, and Cortazar exploit reader expectations for habituated narrative patterns as opportunities for de-familiarization by intentionally violating expectations and reflexively exposing the author of the narrative in the narrative.
While de-articulation denies recognition and thus re-energizes perception, there is a different aesthetic to be found in what we will refer to as re-articulation. Horace, in his Ars Poetica explained that ‘things which are repeated are pleasing.’ This is not to say that all things which are repeated are pleasing, like for example, the unspirited use of cliché. Rather, that repetition should be backed with spirit seems to be implied, especially given that Horace’s domain of discourse was poetics.
As quintessentially spirited and repeated, the deliverance of myth is a prime example of the re-articulation aesthetic. Myth, which is a distillation of the wisdom of human culture, is pleasing when re-articulated because it is evocative of the divine and eternal. Re-articulation is not mundane because the heart of myth transcends the world, and it is not habituated because each enunciation carries with it the weight of grandiosity and a mysterious tension which is irresolvable. The repetition of myth is aesthetic because identification with the sublime comfort of myth is an aesthetic process. That the grand chorus of myth speaks univocally, with absolute clarity and profundity, gives solace to the oft-felt triviality and ephemera of day-to-day human realities.
The profundity-of-myth is itself a meta-narrative, albeit a most powerful and spirited one, but there are other meta-narratives found across art, literature, history, and cultures which are also amenable to re-articulation. Meta-narratives are almost as sublime as myth, though not always as universal. Realism, the Enlightenment project, Buddhism, Modernism, and Postmodernity can all be thought of as meta-narratives. Perhaps the class of meta-narrative is also amenable to the aesthetic of re-articulation because at the heart of each meta-narrative is a well-elaborated aesthetic system which guides perception and action, paralleling the workings of the spirit.
In summary, the spirit of articulation can be characterized as having two aesthetic modes: de-articulation, and re-articulation. De-articulation is an aesthetic that impels the reader to an active perception by subverting the habituated signification process. Re-articulation is an aesthetic in which the repetition of myth and meta-narrative are lauded as being a sublime experience which brings solace and comfort. Rather than mundane familiarity, the aesthetic experience of re-articulation might best be described as Grand Familiarity.
In this section, we tried to cultivate a deeper appreciation for the role of articulation in narrative by exploring its various thematics. First we introduced the opposition of the letter and the spirit and posed articulation as the interplay between spirit and letter. A second thematic explores the spirit as something which should only be partially articulated. This can be achieved by allowing the spirit to exist in connotation, context, subtext, and the manner of speech. Spirit loses its effect if it is too fully articulated because its power stems largely from mystification of the symbol, the libido of creative tension, and sacredness of myth. The aesthetics of articulation are a third thematic. Articulation takes at least two aesthetic modes – de-articulation, which aims to remove known forms and symbols from the realm of familiarity; and re-articulation, whose aesthetic is a Grand Familiarity, experienced through identification with myth and meta-narrative.
Having sketched out a theoretical framework for understanding the aesthetics of narrative in terms of the letter and the spirit, the power of the spirit, and modes of articulation (i.e. partial articulation, full articulation, de-articulation, and re-articulation), we now examine the politics of articulation within the context of today’s media- and information-driven culture.
CULTURAL AND COGNITIVE CONTEXT FOR ARTICULATION
The letter is mundane and unaesthetic, but because it is habitually known to all, the letter is easily recognized; thus it poses the danger of distracting the reader from seeing the spirit. In skillful articulation then, there needs to be a certain economy of the letter, perhaps to prevent drowning out the spirit, for if the spirit is drowned then the aesthetic is sunk.
1 Cultural Narrative Space
The constitution of the letter is not a fixed quantity; it changes as culture changes. The letter is that collection of forms and symbols known to the cultural collective, whose collective unconscious is habituated to their recognition. Because skillful articulation requires the careful negotiation of the space of the letter, the constitution of the letter is of paramount importance. In the realm of narratives, the letter can be regarded as the space of known narratives, forms and techniques.
To understand the politics of articulation in narratives, we must also understand the backdrop of the cultural narrative space, inhabited by the exemplars of narratives, techniques, and forms well known to participants of a cultural system.
In the culture of the contemporary period, this space is saturated like never before in history. With the media and consumerism driving this culture, narratives and other aspects of culture are heavily commodified. The cultural production of narratives has coalesced into mechanized industries like the music, book, magazine, and film industries. Serving as a vehicle for marketing and branding, narratives are found everywhere in advertising, and especially in industries governed by fashion systems. As narrative production comes to be driven by market forces, there is a sense of inevitability that all past narratives will be maximally exploited for every ounce of what aesthetic value they still hold, if it pleases the market.
If a new aesthetic narrative is created, it will in short time be mimicked and commodified. Any spirit born of the new narrative is opportunistically imitated and articulated to the masses and dies as letter. Like a bloodhound, market efficiency relentlessly seeks out new aesthetics, ruthlessly ravaging them until they are spent.
2 Insights from Fashion
The fashion system is the cultural institution responsible for the systematized exploitation of new aesthetics. In Fashion, Culture, and Identity (1994), Davis details the mechanization of aesthetic production in the garment system, but his analysis is generally valid across all the industries governed by fashion.
Garments, Davis explains, follow a sartorial code, which is akin to the letter in our present discourse, and the aesthetics of garments traditionally follows a fashion cycle whose stages are invention, introduction, leadership, increased visibility, and waning. In the initial stage of the fashion cycle, garments are undercoded because the narrative which explicate what a garment represents is only beginning to be articulated. The spirit which motivates new garment narratives are what Davis calls, identity ambivalences, which are the creative tensions which exist in gender, sexuality, and class (they might constitute perhaps what Jung might call a cultural libido). This stage is particularly aesthetic because as we have noted, it represents only a partial articulation of the narrative underlying the new garments. As the garment narrative becomes more fully articulated to the masses, its value wanes and it becomes unaesthetic, having been fossilized into letter.
In Fashion, Culture, and Identity, Davis notes interestingly that the traditional fashion cycle seems to be undergoing transformation. He notes that fashion cycles are so short today that nothing practically ever goes out-of-fashion for very long. There is a new pluralism and polycentrism which now describes the populus. This observation bears an interesting consequence for the constitution of the cultural narrative space. When fashion cycles were lengthy and protracted and fashion was univocal, the space behaved more like a practical memory. Each cycle births some prevailing narrative which is foregrounded and made prominent in memory. As the cycle ends and the next one begins, this narrative recedes slightly into the background and the next prevailing narrative is foregrounded. This memory can be thought of as a serialized stream of narratives, whose recall is degraded over time, and the whole system is capable of forgetting and revision in recall. But degradation and forgetting should be viewed as desirable properties in the sense that as time passes, the letter is decomposing back into the amorphous substrate of the spirit.
When the univocal fashion cycle is replaced by pluralism and polycentrism, the cultural narrative space ceases to behave like a practical memory. There is no forgetting, no degradation, only an ersatz media-driven sustainment of all past narratives, awkwardly fossilized in the present moment. In the age of pluralism and polycentrism, the cultural narrative space is saturated and bloated. All past aesthetic narratives that carry consumer and market cachet today are exploitatively resurrected, commodified, and maintained in the cultural narrative space.
3 Falling into Cliché
The current cultural narrative space is filled with more examples than ever before. While in the past it may have been possible for an individual subject to possess a relatively intuitive understanding of the space, that becomes a more difficult proposition today. In order for an individual subject to accommodate more examples, these examples become further conceptualized and caricaturized. This can be understood as a form of information compression. There is reason for concern because caricaturization amounts to the distillation of spirit into letter, and ideal compression implies the complete eradication of spirit.
That today’s cultural narrative space is bloated with too many examples and that each example is overly conceptualized and caricaturized holds a two-fold consequence for articulation.
First, because the space of narratives is so exhaustively covered, the road to articulation of new narratives is mined with cliché. The narrative exemplars of the cultural narrative space are cliché because they are so compressed, caricaturized, and devoid of spirit. The graveyard of narrative clichés is vast because the market will not allow narratives to die. It is surprisingly easy to fall into cliché because the cognition of reading narratives causes the language of known narratives to be projected onto what is being read, even recognizing cliché even if it was not intended. In any case, the vast size of known exemplars in the cultural narrative space makes it difficult to enunciate new ideas.
Second, in our contemporary culture, articulation of known narratives from the cultural narrative space is rarely aesthetic. This is because the cultural narrative space is fraught with cliché, yet devoid of myth, which is the only form that never loses its aesthetic cachet. Myth and cliché are two edges to the same sword. Both can be familiar and possess of the letter, but whereas myth also possesses of the spirit, cliché has had its spirit eradicated from it in order to improve economy of representation and efficiency of social communication.
4 Decline of the Naïve Narrative
The present culture favors concepts for their economy, and consequently, the letter over the spirit. Among other things, this tendency has led to a saturation of the cultural narrative space, which becomes such a distracting cognitive backdrop of clichés for the subject-reader so as to render many traditional approaches to narrative aesthetically ineffectual.
And indeed, the value of traditional narrative in our culture has declined. In “The Storyteller,” Benjamin attributes the waning importance of storytelling in the twentieth century to a devaluation of experience and the fact that “the communicability of experience is decreasing.” (1936). In our phraseology, Benjamin’s experience is of the spirit because it is rich and personal, and it opposes concepts, which are of the letter. In the “information age,” concepts are favored over experience because they have an economy of representation and a memetic efficiency, which in turn, in the psychology of this culture, lends concepts the cachet of a greater social importance.
Just as traditional notions of the storyteller have declined, myth and Jungian symbolism have also become more inaccessible. In Images and Symbols, Eliade laments the demystification and desacralisation of Western society in the late twentieth century (1961). Eliade implicates the development of modern science and the mechanization of scientific thought as contributors to the erosion of the historic richness and complexity of mythical systems. This analysis is echoed in the works of another scholar of modern myth, Campbell, who wrote the following of modern scientism, “The earth was beginning to be systematically explored, and the old, symbolic, mythological geographies discredited” (Campbell, 1961). However, both Eliade and Campbell hold that myth and the sacred have not completely vanished, but rather they lie neglected and forgotten as repressions of the individual and collective subconscious, resurfacing occasionally and unexpectedly.
Decline of Benjaminian experience and myth is a decline of the spirit, and this, along with the crowding of the cultural narrative space, has made artful narration difficult. Naïve approaches to narrative which simply recapitulate known narratives, techniques and forms, risk being overpowered by a backtext of cliché. Even if a narrative celebrates myth, our present cultural context may not lend itself to an appreciation of its aesthetic value.
In this section we introduced the notion of a cultural narrative space as the embodiment of the agency of the letter for narratives. We reported that the present media- and market-driven culture has hyperarticulated the cultural narrative space with a bloat of clichés of past narratives, techniques, and forms, creating a disruptive backtext which undermines the aesthetic potency of new narratives; furthermore, the power of myth is endangered because myth has evacuated from modern cultures. These problematics render naïve approaches to narrative as hyperarticulate and thus impotent, necessitating more advanced strategies of articulation for aesthetic narratives.
ARTICULATION STRATEGIES
FOR AESTHETIC NARRATIVES
Given that today, the letter is overpowering, and the spirit is repressed, artful narratives seem more and more to resort to psychoanalytic play to liberate the aesthetics of spirit. There is a sense that, in order to maintain a careful balance of letter and spirit, artful narration requires walking on eggshells atop the backtext of the cultural narrative space to resist entropy toward hyperarticulation. In this section, we nominate several articulation strategies that protect and promote the aesthetics of narrative, centered around the principle of resisting hyperarticulation.
1 (Inter)Textuality
Textuality perhaps best represents post-structuralism’s re-conceptualization of the traditional structural narrative as a deconstructed text. In S/Z (1970), Barthes famously synthesizes Derrida’s advocacy to “escape structurality” with Kristeva’s notion of intertextuality to explain how traditional narratives become post-structural texts. Structurality represents the letter, and must be escaped in order to arrive at the more aesthetically interesting space of the intertextual, which represents the spirit.
Traditional narratives, whose structure is dictated by an underlying cultural code, can be re-read as a post-structural text when the reader becomes aware of the “structurality of structure,” and in the process, the reader becomes both cultural critic and semiotician. By foregrounding and habituating away the structurality of cultural codes and forms which underlie cultural narratives, the reader is in a sense inoculated to hyperarticulation, because structurality gets de-authoritized. As a result, a reader is urged to attune to meanings which are only present intertextually; a sort of reading-between-the-lines if you will. When the narrative is freed from its a priori center, narrative coherence is created only through each reader’s own original interpretation of the text, and it is in this original experience that spirit thrives.
The construction of texts which are meant to be read intertextually is itself an artform. Traditional cultural structures are still invoked in a text, but perhaps used quite irreverently and in a manner such as to invite connotation and comparison. Rather than giving the narrative explicitly, the storyteller articulates only the rough ingredients necessary to construct the narrative, deferring its construction to the reader. Of course, the storyteller-psychoanalyst can still influence what narrative a reader is likely to construct by considering how a reader’s psychology and intuition might lead him to interpret what is before him.
Joyce’s Ulysses, Eco’s The Name of the Rose, and Lynch’s television text Twin Peaks are all prime examples of artful post-structural texts. In Twin Peaks for example, Lynch nurtures intertextuality by casting actors from other Lynch texts, and melding different film and television genres (e.g. Westerns, commercials, horror, and sitcom) and aesthetics (e.g. 1950s middle America) into a single soap-operatic detective story.
Despite its aesthetic potentialities, the validity of textuality as an articulation strategy in praxis is called into question when we consider that it puts an incredible onus on the reader to find or create narrative closure. A reader must be quite informed about the cultural space, capable of cultural criticism, and possessed with a high level of attention and dedication to active reading. At its very best, when the readership is qualified, textuality can deliver a slew of transitory aesthetic experiences weaved together in the reader’s mind. However, in the absence of creative reading, textuality may produce narratives judged as fragmented, incoherent, and unstable. Jameson (1983), for example, condemns pastiche – the mimicry of various styles – in post-structural texts as “blank parody;” and the decenteredness of the post-structural text as a schizophrenic experience which becomes increasingly brilliant with iconification yet evermore confusing. Given these caveats, the textuality articulation strategy is only aesthetic if applied appropriately and judiciously.
2 Unusual Representation
The cultural narrative space is saturated with all-too-common narrative forms, styles, mannerisms, and other techniques. We have become collectively habituated to these knowns, so when they appear in a narrative as a vehicle for other story ideas, both the vehicle and that which is carried are recognized as mundanely familiar and unaesthetic, relegated to the bin of known narratives, and unflattering comparison with clichéd works is invited. In order to maintain a reader’s attention and interest, these vehicles of narrative must not be the habitual knowns, but rather, the storyteller should employ unconventional representations. Of course, this is not to say that a representation need only be unusual to do the trick. The representation must also be intuitive, insightful, unique, and coherent.
Artful storytelling employs unusual representations to achieve storytelling goals at various granularities. At the higher granularities are thematic planning and story flow. Here exists an opportunity to vary in unusual ways many of the more established forms and techniques. Playing with and violating expectations of established forms is the technique of many metafictional works, such as those of Cortázar. In his short story, House Taken Over, Cortázar seems to follow a mystery and suspense theme in relating the story of a brother and sister living in a house possessed of unwelcomed occupants. However, the story ends as the unwelcomed occupants take over the house, yet with no articulation of the identity or nature of those occupants. The resolution which is typical at end of suspense and mystery narratives is withheld, thus violating the reader’s expectations of known story form.
Another technique of unusual representation is the use of metaphor in story planning (related to the technique of allegory). In the short story Blowup, Cortázar takes the metaphor of making blow-ups of photographs to create a unique story flow. In the story, a photographer is observing a scene between a man and a woman across a river. While at first he arrives at one analysis of the events transpiring between the man and woman, he progressively reinterprets those events with increasing detail and clarity, just as a photographer enlarges a photo to reveal further detail.
The strategy of unusual representation also applies to the lower granularities of storytelling, such as narrative perspective and discourse style. A single event, when visited through various narrative perspectives, will appear different as each perspective cognitively highlights different sets of features about the event, both quantitatively in which details are told, but also qualitatively in the mannerisms of the telling. Narrative perspective can be understood as a kind of impressionistic observation. Even more granular than narrative perspective is discourse style, or word-choice. Nabakov, for example, prefers long sensorial descriptive vignettes and in particular, synthesthetic cross-sensorial description. The narrators in his fictions describe sounds as having tastes, scents as having colors, they fuse together concepts in interesting combinations, and employ abundant psychologically revealing metonymies.
By violating traditional representations for narratives and developing new representations through intuitive metaphors, a storyteller resists hyperarticulation by avoiding or defamiliarizing known forms and techniques.
3 Aesthetic Signature
We define a narrative’s aesthetic signature as the gestalt of all the small storytelling decisions which sum up to produce a unique narrative voice or style. Aesthetic signature is important to a narrative because it is so unique that it is very difficult to duplicate or make commonplace, thus, it is rather immune to becoming letter, even in the contemporary period’s saturation of the cultural narrative space.
Nabakov is a masterful storyteller who understands the aesthetic value of narrative point-of-view. Each of his fictions and short fictions are narrated with such a strong aesthetic signature in large part because both the narrator and the storyteller (the two are sometimes the same, other times different) employ a unique representation or approach to storytelling and discourse. Nabakov’s narrators wield words as a masterful director wields a camera. He may start focused with a passage on a particular detail, then zooms out to a larger context, then pans across a scene or shuffles through memories or associations. The manner of the narrator and the storyteller and the unique lens they take on the world constitute the aesthetic signature. This signature aestheticizes a narrative by lending it aesthetic closure – a sense of consistency and coherency over the gestalt. The aesthetic signature in its gestalt is articulate and spirited, but because this gestalt is formed out of small, non-textual and non-explicit influences shaping the storytelling, hyperarticulation of the letter can be averted.
The aesthetic signature is a well-authenticated construct because even the storyteller herself cannot claim to have full conscious control over all storytelling choices. Those storytelling decisions invisible to conscious intervention are a product of a storyteller’s intuition, subconscious, and unconscious, all agencies of the spirit rather than of the letter.
4 Personalization
Personalization is the customization of narrative to speak to a particular readership, such as members of a subculture, or even a single person, which is a new affordance granted by computer generated narratives. Personalization resists hyperarticulation on the premise that many of the nuances of a subculture or a person remain unarticulated in the cultural narrative space because these nuances are not common enough to have been articulated among the masses. If we think of the letter as being a popularly dictated agency, the nuances of many subcultures are simply not popular enough to have (yet) been the victim of commodification and caricaturization.
Diaspora is perhaps one subcultural phenomenon whose nuances have not been articulated to death, although admittedly the overall notion of diaspora has been the victim of caricature. Kundera’s novel, Ignorance, takes diasporian experience as its subject, but unlike a novel about the experience of falling in love, there are far fewer clichés to be compared against, so articulating the diasporian experience in an interesting and aesthetic way will face fewer priors for comparison.
Narratives that involve tabooed subjects or values can also be interpreted as a form of personalization because taboo can be viewed as a subcultural safe zone from hyperarticulation since the letter as a socially dictated agency, is generally bounded against articulations of taboo. Much of “contemporary art” relies on narrating taboo as the source of its aesthetic, tackling socially repressed subjects like gender, sexuality, and death.
Finally, artificial intelligence and electronic media now affords the opportunity to customize narrative down to a single person, based on a model of his or her background, experiences, interests, tastes, and personality. Elo’s PLUM story program (1995), for example, reformulates details of news stories to make them more personally relevant for a reader. A news article about a flood devastation in Nigeria reports that 127,000 people were left homeless, and to generate greater contextual relevance to the reader, PLUM augments the story with the fact that “127,000 people is roughly the same as 10 times the people living in Bellefontaine,” where Bellefontaine is the town where the reader lives.
These sorts of personalizations produced by machine narratives are aesthetic for now, but the techniques are still in their infancy. In the future, it is unclear whether or not personalization will remain aesthetic, or possibly be clichéd as more and more machine personalization programs saturate the space and push these techniques into the letter.
CONCLUSION
In this paper, we posed the aesthetics of narrative as problematics of articulation, the letter, and the spirit. We established the letter as the agency of social language, of the explicit, known, mundane, habituated, and thus, unaesthetic. In contrast, the spirit is the agency of the aesthetic; it is an amorphous, anomic space that is alive with meaning, fraught with creative tension, and home to the unarticulated, unarticulatable, mystified, sacred, and mythical. Whereas the letter is socially constructed and maintained, the spirit arises out of the personal and collective unconscious, its chief vehicle to the realm of the conscious being through the agency of intuition.
A central problematic of articulation is that if the spirit is overarticulated into letter, it becomes de-aestheticized, as well demonstrated in the fashion cycle where a new idea becomes increasingly stale as it undergoes massification. Therefore, to maintain the aesthetic integrity of a narrative, the spirit should only be partially articulated, and indeed, there are many subtle spaces that the spirit can be partially articulated into, such as connotations, contexts, and subtexts, and manner of speech. In addition to partial articulation, two other aesthetic modes of articulation, both of which avoid the habituation of the letter, are de-articulation – defamiliarizing a symbol by shifting the signification or meaning beneath it; and, re-articulation – the aesthetic of Grand Familiarity, gained through identification with myths and meta-narratives.
To better understand how artful articulation might be achieved, we also examined the cultural backdrop against which the quality of art can be evaluated. We termed the sum of all commonplace narratives, techniques, and forms which are present in the collective consciousness the cultural narrative space, and reported that in our contemporary period, the media-driven commodification of narratives, together with trends toward pluralism and polycentrism, have led to the saturation and bloating of this narrative space with cliché. This fact, together with the decline of myth in modern cultures, have condemned the traditional, “naïve” structuralist narrative to the realm of unaesthetic because such narratives hyperarticulate what already belongs to cliché.
Given our current cultural context, the production of aesthetic narratives then becomes a question of resisting hyperarticulation. In the penultimate section of this paper, we nominated four articulation strategies based on the resistance or avoidance of the letter of the cultural narrative space – (inter)textuality, unusual representation, aesthetic signature, and personalization. Central to all of these strategies is that they avoid the letter by relying on intuition, the gateway to the spirit, to generate or interpret narrative.
Of course, in spite of this paper’s advocacy for the spirit, the letter should not be discounted wholesale, as the tension between the letter and the spirit is itself a source of the aesthetic, just as Nietzsche’s Apollonian and Dionysian forces interplay to form tragedy, or Bergson’s intuitive and conceptual thinking combine to form a dynamic understanding of reality. But we shall end here, resisting the temptation to articulate any further.
A person’s affective reaction to a concept, topic, or situation can be thought of as either instinctive, due to attitudes and opinions conditioned over time, or reasoned, due to the effect of a particularly vivid recalled memory. Borrowing from cognitive models of human memory function, attitudes that are conditioned over time can be best seen as a reflexive memory, while attitudes resulting from the recall of a past event can be represented as a long-term episodic memory (LTEM). Memory psychologist Endel Tulving equates LTEM with “remembering” and reflexive memory with “knowing” and describes their functions as complementary (Tulving, 1983). We combine the strengths of these two types of memories to form a bipartite, episode-reflex model of the affective memory system.
1 Affective long-term episodic memory
Long-term episodic memory (LTEM) is a relatively stable memory capturing significant experiences and events. The basic unit of memory captures a coherent series of sequential events, and is known as an episode. Episodes are content-addressable, meaning, that they can be retrieved through a variety of cues encoded in the episode, such as a person, location, or action. LTEM can be powerful because even events that happen only once can become salient memories and serve to recurrently influence a person’s future thinking. In modeling attitudes, we must account for the influence of these particularly powerful one-time events.
In our affective memory system, we compute an affective LTEM as an episode frame, coupled with an affect valence score that best characterizes that episode. In Fig. 2, we show an episode frame for the following example episode: “John and I were at the park. John was eating an ice cream. I asked him for a taste but he refused. I thought he was selfish for doing that.”
Figure 2. An episode frame in affective LTEM.
As illustrated in Fig. 2, An episode frame decomposes the text of an identified episode into simple verb-subject-argument propositions like (eat John “ice cream”). Together, these constitute the subevents of the episode. The moral of an episode is important because the episode-affect can be most directly attributed to it. Extraction of the moral, or root cause, is done through heuristics which are discussed elsewhere (Liu, 2003b). Tulving’s encoding specificity hypothesis (1983) suggests that contexts such as date, location, and topic are useful to record because an episode is more likely to be triggered when current conditions match the encoding conditions. The affect valence score is a numeric triple representing (pleasure, arousal, dominance). This will be covered in more detail later in the paper.
2 Affective reflexive memory
While long-term episodic memory deals in salient, one-time events and must generally be consciously recalled, reflexive memory is full of automatic, instant, almost instinctive associations. Whereas LTEM is content-addressable and requires pattern-matching the current situation with that of the episode, reflexive memory is like a simple lookup-table that directly associates a cue with a reaction, thereby abstracting away the content. In humans, reflexive memories are generally formed through repeated exposures rather than one-time events, though subsequent exposures may simply be recalls of a particularly strong primary exposure (Locke, 1689). In addition to frequency of exposures, the strength of an experience is also considered. Complementing the event-specific affective LTEM with an event-independent affective reflexive memory makes sense because there may not always be an appropriate distinct episode which shapes our appraisal of a situation; often, we react reflexively – our present attitudes deriving from an amalgamation of our past experiences now collapsed into something instinctive.
Because humans undergo forgetting, belief revision, and theory change, update policies for human reflexive memory may actually be quite complex. In our computational, we adopt a more simplistic representation and update policy that is not cognitively motivated, but instead, exploits the ability of a computer system to compute an affect valence at runtime.
The affective reflexive memory is represented by a lookup-table. The lookup-keys are simple concepts which can be semantically recognized as a person, action, object, activity, or named event. These keys act as the simple linguistic cues that can trigger the recall of some affect. Associated with each key is a list of exposures, where each exposure represents a distinct instance of that concept appearing in the personal texts. An exposure, E, is represented by the triple: (date, affect valence score V, saliency S). At runtime, the affect valence score associated with a given conceptual cue can be computed using the formula given in Eq. (1).
[pic] (1)
where n = the number of exposures of the concept
This formula returns the valence of a conceptual cue averaged over a particular time period. The term, [pic], rewards frequency of exposures, while the term, [pic], rewards the saliency of an exposure. In this simple model of an affective reflexive memory, we do not consider phenomena such as belief revision, reflexes conditioned over contexts, or forgetting.
To give an example of how affective reflexive memories are acquired from personal texts, consider Fig. 3, which shows two excerpts of text from a weblog and a snapshot sketch of a portion of the resulting reflexive memory.
Figure 3. How reflexive memories get recorded from excerpts.
In the above example, two text excerpts are processed with textual affect sensing and concepts, both simple (e.g. telemarketer, dinner, phone), and compound (e.g. telemarketer::call, interrupt::dinner, phone::ring) are extracted. The saliency of each exposure is determined by heuristics such as the degree to which a particular concept in topicalized in a paragraph. The resulting reflexive memory can be queried using Eq. (1). Note that while a query on 3 Oct 01 for “telemarketer” returns an affect valence score of (-.15, .25, .1), a query on 5 Oct 01 for the same concept returns a score of (-.24, .29, .11). Recalling that the valence scores correspond to (pleasure, arousal, dominance), we can interpret the second annoying intrusion of a telemarketer’s call as having conditioned a further displeasure and a further arousal to the word “telemarketer”.
Of course, concepts like “phone” and “dinner” also unintentionally inherit some negative affect, though with dinner, that negative affect is not as substantial because the saliency of the exposure is lower than with “telemarketer.” (“dinner” is not so much the topic of that episode as “telemarketer”). Also, if successive exposures of “phone” are affectively ambiguous (sometimes used positively, other times negatively), Eq. (1) tends to cancel out inconsistent affect valence scores, resulting in a more neutral valence.
In summary, we have motivated and characterized the two components of the affective memory system: an episodic component emphasizing the affect of one-time salient memories, and a reflexive component, emphasizing instinctive reactions to conceptual cues that are conditioned over time. In the following subsection, we propose how this bipartite affective memory system can be acquired automatically from personal texts.
2 Model Acquisition from Personal Texts
The bipartite model of the affective memory system presented above can be acquired automatically from an analysis of a corpus of personal texts. Fig. 4 illustrates the model acquisition architecture. [pic]
Figure 4. An architecture for acquiring the affective memory system from personal texts.
Though there are some challenging tasks in the natural language extraction of episodes and concepts, such as the heuristic extraction of episode frames, these details are discussed elsewhere (Liu, 2003b). In this subsection, we focus on three aspects of model acquisition, namely, establishing the suitability criteria for personal texts, choosing an affective representation of attitudes, and assessing the affective valence of episodes and concepts.
1 What Personal Texts are Suitable?
In deciding the suitability of personal texts, it’s important to keep in mind that we want a text that is both a rich source of opinion, and also amenable to natural language processing by the computer. First, texts should be first-person, opinion narratives. It is still rather difficult to extract a person’s attitudes given a non-autobiographical text because the natural language processing system would have to robustly decide which opinions belong to which persons (we save this for future work). It is also important that the text be of a personal nature, relating personal experiences or opinions. Attitudes and opinions are not easily accessible in third-person texts or objective writing, especially for a rather naïve computer reading program. Second, texts should explore a sufficient breadth of topics to be interesting. An insufficiently broad model gives a poor and disproportional sampling of a person and would hardly justify the embodiment of such a model into a digital persona. It should be noted however, that there is plausible reason to intentionally partition a person’s text corpus into two or more digital personas. Perhaps it would be interesting to contrast an old Marvin Minsky versus a young one, or a Marvin who is passionate about music versus a Marvin who is passionate about A.I. Third, texts should cover everyday events, situations, and topics, because that is the optimal discourse domain of recognition of the mechanism with which we will judge the affect of text. Fourth, texts should ideally be organized into episodes, occurring over a substantial period of time relative to the length of a person’s life. This is a softer requirement because it is still possible to build a reflexive memory without episode partitioning. Weblogs are an ideal input source because of their episodic organization, although instant messages, newsgroups, and interview transcripts are also good input sources because they are so often rich in opinion.
Representing Affect using the PAD Model
Affect valence pervading the proposed models can take one of two potential representations. They take an atomistic view that emotions existing as a part of some finite repertoire, as exemplified by Manfred Clyne’s “sentics” schema (1977). Or, they can take the form of a dimensional model, represented prominently by Albert Mehrabian’s Pleasure-Arousal-Dominance (PAD) model (1995). In this model, the three nearly independent dimensions are Pleasure-Displeasure (i.e., feeling happy or unhappy), Arousal-Nonarousal (i.e., arousing one’s attention), and Dominance-Submissiveness (i.e., the amount of confidence/lack-of-confidence felt). Each dimension can assume values from –100% to +100%, and a PAD valence score is a 3-tuple of these values (e.g. [-.51, .59, .25] might represent anger).
We chose a dimensional model, namely, Mehrabian’s PAD model, over the discrete canonical emotion model because PAD represents a sub-symbolic, continuous account of affect, where different symbolic affects can be unified along one of the three dimensions. This model has robustness implications for the affective classification of text. For example, in the affective reflexive memory, a conceptual cue may be variously associated with anger, fear, and surprise, which can be unified along the Arousal dimension of the PAD model, thus enabling the affect association to be coherent and focused.
3 Affective Appraisal of Personal Text
Judging the affect of a personal text has three chief considerations. First, the mechanism for judging the affect should be robust and comprehensive enough to correctly appraise the affect of a breadth of concepts. Second, to aid in the determination of saliency, the mechanism must be able to appraise the affect of very little text, such as on the sentence-level. Third, the mechanism should recognize specific emotions rather than convolving affect onto any single dimension.
Several common approaches fail to meet the criteria. The naïve keyword spotting approach looks for surface language features like keywords. However, this approach is not acceptably robust on its own because affect is often conveyed without mood keywords. Statistical affect classification using statistical learning models such as latent semantic analysis (Deerwester et al., 1990) generally require large inputs for acceptable accuracy because it is a semantically weak method. Hand-crafted models and rules are not broad enough to analyze the desired breadth of phenomena.
To analyze personal text with the desired robustness, granularity, and specificity, we employ a model of textual affect sensing using real-world knowledge, proposed by Liu et al. (2003). In this model, defeasible knowledge of everyday people, things, places, events, and situations is leveraged to sense the affect of a text by evaluating the affective implications of each event or situation. For example, to evaluate the affect of “I got fired today,” this model evaluates the consequences of this situation and characterizes it using negative emotions such as fear, sadness, and anger. This model, coupled with a naïve keyword spotting approach, provides rather comprehensive and robust affective classification. Since the model uses knowledge rather than word statistics, it is semantically strong enough to evaluate text on the sentence level, classifying each sentence into a six-tuple of valences (ranging from a value of 0.0 to 1.0) for each of the six basic Ekman emotions of happy, sad, angry, surprised, fearful, and disgusted (an atomistic view of emotions) (Ekman, 1993). These emotions are then mapped to the PAD model.
One point of potential paradox should be addressed. The real-world knowledge-based model of affect sensing is based on defeasible commonsense knowledge from the Open Mind Commonsense corpus (Singh et al., 2002), which is in turn, gathered from a web community of some 11,000 teachers. Therefore, the affective assessment of text made by such a model represents the judgment of a typical person. However, sometimes a personal judgment of affect is contradicted by the typical judgment. Thus, it would seem paradoxical to attempt to learn that a situation has a personally negative affect when the typical person judges the situation as positive. To overcome this difficulty, we implement, in parallel, a mood keyword-spotting affect sensing mechanism to confirm and contradict the assessment of the primary model. In addition, we make the assumption that although a personal affect judgment may deviate from that of a typical person on small particulars, it will not deviate on average, when examining a large text. The implication of this is that on a slightly larger granularity than a sentence, the affective appraisal is more likely to be accurate. In fact, accuracy should increase proportional to the size of the textual context being considered. The evaluation of Liu et al.’s affective navigation system (2003b) yields some indirect support for the idea that accuracy increases with the size of the textual context. In that user study, users found affective categorizations of textual units on the order of chapters to be more accurate and useful to information navigation than affective categorizations of small textual units such as paragraphs.
To assess the affect of a sentence, we factor in the affective assessment of not only the sentence itself, but also of the paragraph, section, and whole journal entry or episode. Because so much context is factored into the affect judgment, only a modest amount of affective information can learned for any given sentence. Thus we rely on the confirming effects of being able to encounter an attitude multiple times. In exchange for only being able to learn a modest amount from a sentence, we also minimize the impact of erroneous judgments.
In summary, digital personas can be automatically acquired from personal texts. These texts should feature the explicit expression of the opinions of the person to be modeled, and should be of a certain form required by the natural language processing. Natural language processed texts are analyzed for its affective content at varying textual granularities (e.g. sentence-, paragraph-, and section- level) so as to minimize the possibility of error. This is necessary because our textual affect sensing tool evaluates a typical person’s affective reaction to a text, and not any particular person’s. Affect valence is represented using the PAD dimensional model of affect, whose continuity allows affect valences to be more easily summed together. The resulting affect valence is recorded with a concept in the reflexive memory, and an episode in the episodic memory.
3 Predicting Attitudes using the Model
Having acquired the model, the digital persona attempts to predict the attitudes of the person being modeled by offering some affective reaction when it is fed some new text. This reaction is based on how the new text triggers the reflex concepts and the recall of episodes in the affective memory system. When a reflex memory or episode is triggered, the affective valence score associated with that memory gets attached to the affective context of the new text. The gestalt reaction to the new text is a weighted summation of the affect valence scores of the triggered memories.
The triggering process is somewhat complex. The triggering of episodes requires the detection of an episode in the new text, and heuristically pattern matching this new episode frame to the library of episode frames. The range of concepts that can trigger a reflex memory is increased by the addition of conceptual analogy using OMCSNet, a semantic network of commonsense knowledge. The details of the triggering process is omitted here, but is discussed elsewhere (Liu, 2003b).
This process of valuating some new text by triggering memories out of the context in which they were encoded, and inheriting their affect valences, is error prone. We rely on the observation that if many memories are triggered, their contextual intersection is more likely to be accurate. Ultimately, the performance of the digital persona in reproducing the attitudes of the person being model is determined by the breadth and quality of the corpus of personal texts gathered on the person. The digital persona cannot predict attitudes that are not explicitly exhibited in the personal texts.
4 Enriching the Basic Model
The basic model of a person’s attitudes focuses on applying a person’s self-described memories to valuate new textual episodes. While this basic model is sufficient to produce reactions to text for which there exists some relevant personal memories, the generated digital personas are often quite “sparse” in what they can react to. We have proposed and evaluated some advancements to the basic model. In particular, we have looked at how a person’s attitude model can be enriched by the attitude models of people with whom the modeled person fashions himself/herself after – perhaps a good friend or mentor. More technically, we mean an imprimer.
Marvin Minsky describes an imprimer as someone to which one becomes attached. (Minsky, forthcoming) He introduces the concept in the context of attachment-learning of goals, and suggests that imprimers help to shape a child’s values. Imprimers can be a parent, mentor, cartoon character, a cult, or a person-type. The two most important criteria for an imprimer are that 1) the imprimer embodies some image, filled with goals, ideas, or intentions, and that 2) one feels attachment to the imprimer.
We extend this idea in the affect realm and make the further claim that internal imprimers can do more than to critique our goals; our attachment to them leads us to the willful emulation of a portion of their values and attitudes. Keeping a collection of these internal imprimers, they help to support our identity. From the supposition that we conform to many of the attitudes of our internal imprimers, we hypothesize that affective memory models of these imprimers, if known, can complement the person’s own affective memory model in helping to predict a person’s attitudes. This hypothesis is supported by much of the work in psychoanalysis. Sigmund Freud (1991) wrote of a process he called introjection, in which children unconsciously emulate aspects of their parents, such as the assumption of their parent’s personalities and values. Other psychologists have referred to introjection by terms like identification, internalization, and incorporation.
We propose the following model of internal imprimers to support attitude prediction. First, it is necessary to identify people, groups, and images that may possibly be a person’s imprimer. We can do so but analyzing the affective memory. From a list of all conceptual cues from both the episodic and reflexive memories, we use semantic recognizers to identify all people, groups (e.g. “my company”) and images (e.g. “dog”=> “dog-person”) that on average, elicit high Arousal and high Submissiveness, show high frequency of exposure in the reflexive memory, and collocate in past episodes with self-conscious emotion keywords like “proud”, “embarrassed”, “ashamed”.
[pic]
Figure 5. Affective models of internal imprimers, organized into personas, complements one’s own affective model
Once imprimers are identified, we also wish to identify the context under which an imprimer’s attitudes show influence. Shown in Fig. 5, we propose organizing the internal imprimer space into personas representing different contextual realms. There is good reason to believe that humans organize imprimers by persona because we are different people for different reasons. One might like Warren Buffett’s ideas about business but probably not about cooking. Personas can also prevent internal conflicts but allowing a person to maintain separate systems of attitudes in different contexts. To identify an imprimer’s context, we must first agree on an ontology of personas, which can be person-general (as the personas in Fig. 5 are) or person-specific. Once imprimers are associated with personae, we gather as much “personal” text from each imprimer as desired and acquire only the reflexive memory model, thus relaxing the constraint that texts have episodic organization. In this augmented attitude prediction strategy (depicted in Fig. 3), when conceptual cues are unfamiliar to the self, we identify internal imprimers whose persona matches the genre of the new episode, and give them an opportunity to react to the cue. These affective reactions are multiplied by a coefficient representing the ability of this self to be influenced, and the valence score is added on to the episode. Rather than maintaining all attitudes in the self, internal imprimers enable judgments about certain things to be mentally outsourced to the persona-appropriate imprimers.
We have implemented and evaluated the automated identification and modeling acquisition of imprimer personas in cases where the imprimers are people. Our implemented system is not yet able to use abstract non-person imprimers, e.g. “dog-person”.
[pic]
Figure 6. The imprimer-augmented attitude prediction strategy. Edges represent memory triggers.
In summary, we have presented a reflex-episode model of affective memory as a memory-based representation of a person’s attitudes. The model can be acquired automatically from personal text using natural language processing and textual affect analysis. The model can be applied over new textual episodes to produce affective reactions that aim to emulate the actual reactions of the person being modeled. (Fig. 6). We have also discussed how the basic attitudes model can be enriched with added information about the attitudes of the mentors of the person being modeled.
In the following section, we abstract away the details of the attitudes model presented in this section to examine how digital personas can be portrayed graphically and how a collection of digital personas can portray the personalities of a community.
WHAT WOULD THEY THINK?
While modeling a person’s attitudes is fun in the abstract, it lacks the motivation and the verifiability of a real application of the theory and technology. What Would They Think? (Fig. 1) is a graphical realization of the modeling theory discussed in the previous section. What Would They Think? has been implemented and is currently being evaluated through user studies, though the underlying attitude models have already been evaluated in a separate study. In this section, we discuss the design of our interface, present some scenarios for its use, and report how this work has been evaluated.
1 Interface Design
Digital personas acquired from an automatic analysis of personal text, are represented visually with pictures of faces, which occupy a matrix. Given some new text typed or spoken into the “fodder” box, each persona expresses an affective reaction through modulations in the graphical elements of the face icon. Each digital persona is also capable of some introspection. When clicked, a face can explain what motivated its reaction by displaying a salient quote from its personal text.
Why a static face? Visualizing a digital persona’s attitudes and reactions with the face of the person being represented is better than with something textual or abstract. There are several reasons why a face is a superior representation. People are already wired with a cognitive faculty for quickly recognizing and remembering faces, and a face acts as a unique cognitive container for a person’s individual identity and personality. In the user task of understanding a person’s personality, it is easier to attribute personality traits and attitudes to a face than to text or an abstract graphic. For example, people-watching is a past-time in which we imagine the personality and identity behind a stranger’s face (Whyte, 1988). A community of faces is more socially evocative than either a community of textual labels or abstract representations, for those representations are not designed as convenient containers of identity and personality.
Having decided on a face representation, should the face be abstract or real, static or animated? While verisimilitude is the goal for many facial interfaces, we must be careful to not portray more detail in the face than our attitude model is capable of elucidating, for the face is fraught with social cues, and unjustified cues could do more harm than good. By conveying attitudes through modulations in the graphical elements of a static face image, rather than through modulations of expression and gaze in an animated face, we are emphasizing the representational aspect of the face, over the real. Scott McCloud has explored extensively the representational-vs.-real tradeoff of face drawing in comics (1993).
Modulating the Face. In the expression of an affective reaction, it is nice to be able to preserve the detail of the continuous, dimensional output of the digital persona. The information should also be conveyed as intuitively as possible. Thus an intuitive mapping may be best achieved through the use of visual metaphors to represent affective states of the person (Lakoff & Johnson, 1980). We often describe a happy person as being “colorful”, while “face turns colorless” usually represents negative emotions like fear and melancholy. A person whose attention or passion is aroused has a face that “lights up”. And someone who isn’t sure or confident about a topic feels “fuzzy” toward it. Taking these metaphors into consideration, a rather straightforward scheme is used to map the three affect dimensions of pleasure, arousal, and dominance onto the three graphical dimensions of color saturation, brightness, and focus, respectively. A pleasurable reaction is manifested by a face with high color saturation, while a displeasurable reaction maps to an unsaturated, colorless face. This mapping creates an implicit constraint that the face icon be in color. An aroused reaction results in a brightly lit icon, while a non-aroused reaction results in a dimly lit icon. A dominant (confident) reaction maps to a sharp, crisp image, while a submissive (unconfident) reaction maps to a blurry, unfocused image. While better mapping schemes may exist, our experience with users who have worked with this interface tells us that the current scheme conveys the affect reaction quite intuitively. This makes the assumption that the original face icons are all of good quality – in color, bright enough, and in focus.
Populating a Community. An n x n matrix can hold a small collection of digital personas. The matrix can either be configured automatically or manually. Each matrix cell can be manually configured to house a digital persona by specifying a persona .MIND file and a face icon. A user can build and later augment a digital persona by specifying a weblog url, homepage url, or some personal text pasted into the window. The matrix can also be configured automatically to represent a community. Plug-in scripts have been created to automatically populate the matrix with certain types of communities, including a niche community of weblogs known as a “blog ring,” a circle of friends in the online networking community called “,” a group of potential mates on an online dating website called “,” and a usenet community.
Currently, only a blog ring community can generate fully specified digital personas. The Friendster and communities’ personal text corpora are rather small profile texts. As a result, only a fairly shallow reflexive memory can be built. The episodic memory is not meaningful for these texts. The personal texts for usenet communities are rather inconsistent in quality. For example, a usenet community based on question and answers will not be as good a source of explicit opinions as a community based on discussion of issues. Also, usenet communities pose the problem of not providing a face icon for each user. In this case, the text of each person’s name labels each matrix cell, accompanied by a default face icon in the background, which is necessary to convey the affective reaction.
Introspection. A digital persona is capable of some limited introspection. To inquire what motivated a persona to express a certain reaction to some text, the face icon can be clicked. An explanation will be offered taking the form of a quote or a series of quotes from the personal text. These quotes are generated by backpointers to the text associated with each affective memory. For episodic memory, a particularly salient episode can justify a reaction, while there may need to be many quotes to justify a triggered reflex memory. With the capability for some introspection and explanation, a user can verify whether or not an affective reaction is indeed justified. This lends the interface some fail-softness, as a user will not be completely mislead when a person’s attitude is erroneously represented by the system.
2 Use Cases
How can a person use the What Would They Think? interface to understand the personalities and attitudes of people in a community? The system supports several use cases.
In the basic use case, the user, a new entrant to a community, is presented with an automatically generated matrix of some people in the community. The user can employ a hypothesis-testing approach to understanding personalities. The user types some very opinionated statements into the “fodder” box as a litmus test in understanding the attitudes of the different people toward that statement. Faces that lighting up in color versus black and white provide an illustrative contrast of the strong disagreements in the community. A user can inquire as to the source of strong opinions by clicking on a face and viewing a motivating quote. A user can reorganize the matrix so as to cluster personalities perceived to be similar. Assuming that the personal texts for each persona in the community is of comparable length, depth, and quality, the user may notice over a series of interactions that certain personas are negative more often than not, or certain other personas are aroused more intensely more often than other personas. These may lead a user to conclude that certain personalities are more cynical, and others more easily excitable.
Another use case is gaging the interests and expertise of people in a community. Because people generally talk more about things that interest them and have more to say on topics they are more familiar with, a digital persona modeled on such texts will necessarily exhibit more reaction to texts that are interesting to the person being or falls in their area of expertise. In this use case, a user can, for example, copy-and-paste a news article into the fodder box and assess which personas are interested or have expertise toward a particular topic.
A third use case involves community-assisted reading. The matrix fodder box can be linked to a cursor position in a text file browser. As a user reads through a webpage, story, or news article, he/she can get a sense of how the community might read and react to the text currently being read.
3 Evaluation
The quality of the attitude prediction in What Would They Think? has been formally evaluated through user studies. We are also currently conducting user studies to evaluate the effectiveness of the matrix interface in assisting a person to learn about and understand a community. These results will be available by press time.
The quality of attitude prediction was evaluated experimentally, working with four subjects. Subjects were between the ages of 18 and 28, and have kept diary-style weblogs for at least 2 years, with an average entry interval of three-to-four days. Subjects submitted their weblog urls, for the generation of affective memory models. An imprimer identification routine was run, and the examiner hand-picked the top one imprimer for each of the three persona domains implemented: social, business, and domestic. A personal text corpus was built, and imprimer reflexive memory models were generated. The subjects were engaged in an interview-style experiment with the examiner.
In the interview, subject and their corresponding PERSONA models were asked to evaluate 12 short texts representative of three genres: social, business, and domestic (corresponding to the ontology of personas in the tested implementation). The same set of texts was presented to each participant and the examiner chose texts that were generally evocative. They were asked to summarize their reaction by rating three factors on Likert-5 scales.
Feel negative about it (1)…. Feel positive about it (5)
• Feel indifferent about it (1) … Feel intensely about it (5)
Don’t feel control over it (1)… Feel control over it (5)
These factors are mapped onto the PAD valence format, assuming the following correspondence: 1(-1.0, 2( -0.5, 3(0.0, 4( +0.5, and 5( +1.0. Subjects’ responses were not normalized. To assess the quality of attitude prediction, we record the spread between the human-assessed and computer-assessed valences,
[pic] (2)
We computed the mean spread and standard deviation across all episodes along each PAD dimension. On the –1.0 to +1.0 valence scale, the maximum spread is 2.0. Table 1 summarizes the results.
Table 1. Performance of attitude prediction, measured as the spread between human and computer judged values.
| |Pleasure |Arousal |Dominance |
| |mean |std. |mean |std. |mean |std. |
| |spread |dev. |spread |dev. |spread |dev. |
|SUBJECT 1 |0.39 |0.38 |0.27 |0.24 |0.44 |0.35 |
|SUBJECT 2 |0.42 |0.47 |0.21 |0.23 |0.48 |0.31 |
|SUBJECT 3 |0.22 |0.21 |0.16 |0.14 |0.38 |0.38 |
|SUBJECT 4 |0.38 |0.33 |0.22 |0.20 |0.41 |0.32 |
Assuming that human reactions obeyed a uniform distribution over the Likert-5 scale, we give two baselines, which were simulated over 100,000 trials. In BASELINE 1, [pic] is fixed at 0.0 (neutral reaction to all text). In BASELINE 2, [pic] is given a random value over the interval [-1.0,1.0] with a uniform distribution (arbitrary reaction to all text). It should be pointed out however, that in the context of an interactive sociable computer, BASELINE 1 is not a fair comparison, because it would never produce any behavior.
On average, our approach performed noticeably better than both baselines, excelling particularly in predicting arousal, and having the most difficulty predicting dominance. The standard deviations were very high, reflecting the observation that predictions were often either very close to the actual valence, or very far. This can be attributed to one of several causes. First, multiple episodes described in the same journal entries may have caused the wrong associations to be learned. Second, the reflexive memory model does not account for conflicting word senses. Third, personal texts inputted for the imprimers often generated models skewed to positive or negative because text did not always have an episodic organization. While results along the pleasure and dominance dimensions are weaker, the arousal dimension recorded a mean spread of 0.22, suggesting the possibility that it alone may have immediate applicability.
Table 2. Performance of attitude prediction that can be attributed to imprimers and episodic memory
In the experiment, we also analyzed how often the episodic memory, reflexive memory, and imprimers were triggered. Episodes were on average, 4 sentences long. For each episode, reflexive memory was triggered an average of 21.5 times, episodic memory 0.8 times, and imprimer reflexive memory 4.2 times. To measure the effect of imprimers and episodic memories, we re-ran the experiment turning off imprimers only, episodic memory only, and both. Table 2 summarizes the results.
These results suggest that the positive effect of episodic memory was negligible on the results. This certainly has to do with its low rate of triggering, and the fact that episodic memories were weighted only slightly more than reflexive memories. The low trigger rate of episodic memory can also be attributed to the strict criteria that three conceptual cues in an episode frame must trigger in order for the whole episode to trigger. These results also suggest that imprimers played a measurable role in improving performance, which is a very promising result.
Overall, the evaluation demonstrates that the proposed attitude prediction approach is promising, but needs further refinement. The randomized BASELINE 2 is a good comparison when considering possible entertainment applications, whose interaction is more fail-soft. The approach does quite well against the active BASELINE 2, and is within the performance range of these applications. Taking into account possible erroneous reactions, we were careful to pose What Would They Think? as a fail-soft interface. The reacting faces are evocative, and encourage the user to click on a face for further explanation. Used in this manner, the application is fail-soft because users can decide on the basis of the explanations whether the reaction is justified or mistaken. We expect that ongoing studies of the usefulness of the What Would They Think? intelligent will show that its use is fail-soft: the generated reactions are evocative and encourage the user to further verify and investigate a purported attitude. We do not suggest that the approach is yet ready for fail-hard applications, such as deployment as a sociable software agent, because fallout (bad predictions) can be very costly in the realm of affective communication (Nass et al., 1994).
RELATED WORK
The community of personalities metaphor has been previously explored with Guides (Oren et al., 1990), a multi-character interface that assisted users in browsing a hypermedia database. Each guide embodied a specific character (e.g. preacher, miner, settler) with a unique “life story.” Presented with the current document that a user is browsing, each guide suggested a recommended follow-up document, motivated by the guide’s own point-of-view. Each guide’s recommendations is based on a manually constructed bag of “interests” keywords.
Our affective memory -based approach to modeling a person’s attitudes appears to be unique in the literature. Existing approaches to person modeling are of two kinds: behavior modeling, and demographic profiling. The former approach models the actions that users take within the context of an application domain. For example, intelligent tutoring systems track a person’s test performance (Sison & Shimura, 1998), while online bookstores track user purchasing and browsing habits and combine this with collaborative filtering to group similar users (Shardanand & Maes, 1995). The latter approach uses gathered demographic information about a user, such as a “user profile”, to draw generalized conclusions about user preferences and behavior.
Neither of the existing approaches are appropriate to the modeling of “digital personas.” In behavior modeling, knowledge of user action sequences are generally only meaningful in the context of a particular application and does not significantly contribute to a picture of a person’s attitudes and opinions. Demographic profiling tends to overgeneralize people by the categories they fit into, is not motivated by personal experience, and often requires additional user action such as filling out a user profile.
Memory-based modeling approaches have also been tried in related work on assistive agents. Brad Rhode’s Remembrance Agent (Rhodes & Starner, 1996) uses an associative memory to proactively suggest relevant information. Sunil Vemuri’s project, “What Was I Thinking?” (2004) is a memory prosthesis that records audio from a wearable device, and intelligently segments the audio into episodes, allowing the “audio memory” to be more easily browsed.
CONCLUSION
Learning about the personalities and dynamics of online communities has been up to now a difficult problem with no good technological solutions. In this paper, we propose What Would They Think? an interactive visual representation of the personalities in a community. A matrix of digital personas reacts visually to what a user types or says to the interface, based on predictions of attitudes actually held by the persons being modeled. Each digital persona’s model of attitudes is generated automatically from an analysis of some personal text (e.g. weblog), using natural language processing and textual affect sensing to populate an associative affective memory system. The whole application enables a person to understand the personalities in a community through interaction rather than by reading narratives. Patterns of reactions observed over a history of interactions can illustrate qualities of a person’s personality (e.g. negativity, excitability), interests and expertise, and also qualities about the social dynamics in a community, such as the consenses and disagreements held by a group of individuals.
The automated, memory-based personality modeling approach introduced in this paper represents a new direction in person modeling. Whereas behavior modeling only yields information about a person within some narrow application context, and whereas demographic profiling paints an overly generalized picture of a person and often requires a profile to be filled out, our modeling of a person’s attitudes from a “memory” of personal experiences paints a richer, better-motivated picture about a person that has a wider range of potential applications than application-specific user models. User studies concerning the quality of the attitude prediction technology are promising and suggest that the currently implemented approach is strong enough to be used in fail-soft applications. In What Would They Think? the interface is designed to be fail-soft. The reactions given by the digital personas are meant to be evocative. The user is encouraged to further verify and investigate a purported attitude by clicking on a persona and viewing a textual explanation of the reaction.
In future work, we intend to further develop the modeling of attitudes by investigating how particularly strong beliefs such as “I love dogs” can help to create a model of a person’s identity. We also intend to investigate other applications for our person modeling approach, such as virtual mentors and guides, marketing, and document recommendation.
ACKNOWLEDGMENTS
The authors would like to thank Deb Roy, Barbara Barry, Push Singh, Andrea Lockerd, Marvin Minsky, Henry Lieberman, and Ted Selker for their comments on this work.
REFERENCESWORKS CITED
1]
2] Roland Barthes: 1970, S/Z. French & European Pubns.
3] Walter Benjamin: 1935, “The Work of Art in the Age of Mechanical Reproduction.” Zeitschrift für SozialforschungC.
4] Walter Benjamin: 1985, “The Storyteller.” Illuminations. Schocken Books. (Original work published 1936)
5] Joseph Campbell: 1961, The Impact of Science on Myth – Myths to Live By. New York: Bantam Books
6] Fred Davis: 1994, Fashion, Culture, and Identity, University of Chicago Press, Chicago.
7] John Dewey: 1967, Art as Experience, JA Boydston (Ed.), The Collected Works of John Dewey. Carbondale: Southern Illinois University Press. (Original work published 1934)
8] Mircea Eliade: 1961, Images and Symbols. New York: Sheed and Hard
9]
10] Sarah Elo: 1995, PLUM: Contextualizing News for Communities through Augmentation. MIT Media Lab S.M. Thesis.
11] Frederic Jameson: 1972, The Prison-House of Language: A Critical Account of Structuralism and Russian Formalism. Princeton, NJ: Princeton UP.
12] Frederic Jameson: 1998, “Postmodernism and Consumer Society,” The Cultural Turn. Verso. (Original work published 1983)
13] Carl Gustav Jung: 1977, Symbols of Transformation. Princeton NJ: Princeton UP (Original work published 1912)
14] Carl Gustav Jung: 1956, Two Essays on Analytical Psychology. New York: Meridian.
15] Jacques Lacan: 1977, “The agency of the letter in the unconscious or reason since Freud,” A. Sheridan (trans.), Ecrits. New York: W.W. Norton. (Original work published 1966).
16] Friedrich Nietzsche: 1872, Birth of Tragedy. Penguin.
17] Carol S. Pearson: 1991, Awakening the Heroes Within: Twelve Archetypes to Help Us Find Ourselves and Transform the World. New York, NY: HarperSanFrancisco.
18] Victor Shklovsky: 1988, Art as Technique. Lodge (Ed.), Modern Criticism and Theory: A Reader. London: Longmans, pp. 16-30. (Original work published 1917).
Deerwester, S. et al. (1990). Indexing by latent semantic anlysis. Journal of the American Society of Information science:416(6), pp 391-407.
19] Ekman, P. (1993). Facial expression of emotion. American Psychologist, 48, 384-392.
20] Freud, S. (1991). The essentials of psycho-analysis: the definitive collection of Sigmund Freud's writing selected, with an introduction and commentaries, by Anna Freud. London: Penguin.
21] Lakoff, G. & Johnson, M. (1980). Metaphors We Live By, University of Chicago Press.
22] Liu, H. (2003b). A Computational Model of Human Affective Memory and Its Application to Mindreading. Submitted to FLAIRS 2004. Draft available at: /~hugo/publications/drafts/Affective-Mindreading-Liu.doc
23] Liu, H., Lieberman, H., Selker, T. (2003). A Model of Textual Affect Sensing using Real-World Knowledge. Proceedings of IUI 2003, pp. 125-132.
24] Liu, H., Selker, T., Lieberman, H. (2003b). Visualizing the Affective Structure of a Text Document. Proceedings of CHI 2003, pp. 740-741.
25] Locke, J. (1689). Essay Concerning Human Understanding Hypertext by ITL at Columbia University, 1995. Print version ed. P.H. Nidditch. Oxford, 1975.
26] McCloud, S. (1993). Understanding Comics, Kitchen Sink Press, Northhampton, Maine,
27] Mehrabian, A. (1995). for a comprehensive system of measures of emotional states: The PAD Model. (Available from Albert Mehrabian, 1130 Alta Mesa Road, Monterey, CA, USA 93940).
28] Minsky, M., (forthcoming). The Emotion Machine, Pantheon, New York. Several chapters are available at: .
29] Nass, C.I., Stener, J.S., and Tanber, E. (1994) Computers are social actors. In Proceedings of CHI ’94, (Boston, MA), pp. 72-78, April 1994.
30] Oren, T., Salomon, G., Kreitman, K. and Don, A. (1990). Guides: characterizing the interface. In Laurel, B. (Eds.) The art of human-computer interface design. Addison-Wesley.
31] Rhodes, B. and Starner, T. (1996). The Remembrance Agent: A continuously running automated information retrieval system. Proceedings of PAAM '96, pp. 487-495.
32] Shardanand, U. and Maes, P. (1995). Social information filtering: Algorithms for automating "word of mouth", Proceedings of CHI'95, 210-217.
33] Singh, P., (2002). The public acquisition of commonsense knowledge. Proceedings of AAAI Spring Symposium. Palo Alto, CA, AAAI.
34] Sison, R. and Shimura, M. (1998). Student modeling and machine learning. International Journal of Artificial Intelligence in Education, 9:128-158.
Tulving, E. (1983). Elements of episodic memory. Oxford: New York.
V
Whyte, W. (1988). City. Doubleday, New York.
35]
-----------------------
::: EPISODE FRAME :::
SUBEVENTS:
(eat John “ice cream”),
(ask I John “for taste”),
(refuse John)
MORAL: (selfish John)
CONTEXTS: (date), (park), ()
EPISODE-IMPORTANCE: 0.8
EPISODE-AFFECT: (-0.8,0.7,0)
(refuse John)
Text Excerpts
…2 Oct 01… Telemarketers called harassed me again today, interrupting my dinner. I’m really upset…
…4 Oct 01… The phone rang, and of course, it was a telemarketer. Damn it!
::: REFLEXIVE MEMORY :::
telemarketer = {
[2oct01, (-.3, .5, .2), .5],
[4oct01, (-.8, .8, .3), .4] } ;
dinner = {
[2oct01, (-.3, .5, .2), .2]}
“interrupt dinner” = {…} ;
….
Personality
traits
Explicit
attitudes
Implicit
attitudes
................
................
In order to avoid copyright disputes, this page is only a partial summary.
To fulfill the demand for quickly locating and searching documents.
It is intelligent file search solution for home and business.
Related download
- the rise of youth counter culture after world war ii and
- articulation the letter and the spirit in the aesthetics
- doctoral dissertations by author
- american grand narrative in postmodern locality
- memories ii the cultural meaning of forrest gump
- popular culture and narrative—introduction
- montclair state university
- reading the us cultural past
Related searches
- the government and the economy
- health and wellness programs in the workplace
- health and wellness ideas in the workplace
- find the adjective in the sentence calculator
- is the us in the european union
- all the prayers in the bible pdf
- the wicked in the bible
- the torah and the talmud
- narrowing of the spine in the neck
- explain the new and the old testament
- all the words in the english language
- all the words in the world list