{ "paper_id": "W02-0224", "header": { "generated_with": "S2ORC 1.0.0", "date_generated": "2023-01-19T05:16:01.370110Z" }, "title": "Grounding styles of aged dyads: an exploratory study", "authors": [ { "first": "Atsue", "middle": [], "last": "Takeoka", "suffix": "", "affiliation": {}, "email": "takeoka@jaist.ac.jp" }, { "first": "Atsushi", "middle": [], "last": "Shimojima", "suffix": "", "affiliation": {}, "email": "ashimoji@jaist.ac.jp" } ], "year": "", "venue": null, "identifiers": {}, "abstract": "This paper reports an exploratory study of the grounding styles of older dyads, namely, the characteristic ways in which they mutually agree to have shared a piece of information in dialogue. On the basis of Traum's classification of grounding acts, we conducted an exploratory comparison of dialogue data on older and younger dyads, and found that a fairly clear contrast holds mainly in the types of acknowledgement utterances used by the two groups. We will discuss the implications of this contrast, concerning how some of the negative stereotypes about conversations with older people may arise from this difference in grounding styles.", "pdf_parse": { "paper_id": "W02-0224", "_pdf_hash": "", "abstract": [ { "text": "This paper reports an exploratory study of the grounding styles of older dyads, namely, the characteristic ways in which they mutually agree to have shared a piece of information in dialogue. On the basis of Traum's classification of grounding acts, we conducted an exploratory comparison of dialogue data on older and younger dyads, and found that a fairly clear contrast holds mainly in the types of acknowledgement utterances used by the two groups. We will discuss the implications of this contrast, concerning how some of the negative stereotypes about conversations with older people may arise from this difference in grounding styles.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Abstract", "sec_num": null } ], "body_text": [ { "text": "1 Introduction Clark (1996) developed the notion of grounding to capture intricate processes with which conversants establish a set of information as a shared common ground for on-going (and possibly subsequent) dialogue. In a grounding process, a conversant presents a piece of information verbally or non-verbally, and the partner issues a particular public signal of the receipt, understanding, or acceptance of that information. Once a piece of information is thus incorporated into a common ground, it is available as something than can be presupposed in subsequent communication, in which still another change to the common ground is worked out. On the basis of Clark's work, Traum (1994) shows a way of substantiating the process of grounding in a finite-state transition model, which specifies what contributions to a grounding process can be made by individual utterances in dialogue and how individual contributions combine themselves to make a sequence that completes the grounding. In Traum's terms, such contributions are grounding acts, and the units of utterance that can perform grounding acts are utterance units, and the sequences of utterance units that complete are discourse units.", "cite_spans": [ { "start": 15, "end": 27, "text": "Clark (1996)", "ref_id": "BIBREF1" }, { "start": 668, "end": 694, "text": "Clark's work, Traum (1994)", "ref_id": null } ], "ref_spans": [], "eq_spans": [], "section": "", "sec_num": null }, { "text": "Now, both Clark's and Traum's models of grounding are generic, in the sense that they both admit a significant range of forms or styles in which actual processes of grounding may take place. That is, their models tell us what forms of grounding sequences are possible or well-formed, but do not tell what particular forms are most frequent in actual dialogues or, more specifically, what forms are most frequent in what sorts of dialogue situations. Thus, on the basis of Clark's and Traum's work, natural questions arise as to grounding styles, namely, the possibility of characteristic forms of grounding sequences adopted by particular types of conversants in particular types of occasions. This paper investigates this question in its special application to casual, non-task-oriented dialogues conducted by aged people. Do people develop a characteristic style of grounding as they age? If so, what exactly distinguish their grounding styles in casual, non-task-oriented settings?", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "", "sec_num": null }, { "text": "To address these questions, we collected a spoken dialogue data comprising both older dyads and younger dyads, and analyzed the grounding processes in our data on the basis of Traum's finite-state model.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "", "sec_num": null }, { "text": "Why are we particularly interested in possible grounding styles of aged people? It is rather common that younger people find difficult or get frustrated in conversing with aged people. Thus, a common stereotype says that conversations with aged people are slow-paced, repetitive, and therefore boring. It is perhaps because of these oft-felt problems that younger people often adopt \"patronizing speech styles\" or \"secondary baby talk\" toward aged people, characterized by restricted topics, childlike expressions, directive speech, and exaggerated nonverbal signals (Ryan & Hummert, 1995) or by slower speech, varied paraphrases, decreased grammatical complexity, lower propositional density, and other modifications (Kemper & Vandeputte, 1995) . Now, it seems customary that these communicative problems between older people and younger people are attributed to the side of aged people only, to their difficulties in auditory reception, alleged slowness of understanding, and alleged stubbornness in accepting new ideas. However, they may not be problems with one side only, nor problems with cognitive capacities or personal characters of aged people, but may largely come from different grounding styles adopted by the two groups of people. That is, the ways they agree to have a piece of information shared may have different recurrent patterns, and this difference may be a real cause of the difficulty and the negative stereotype. If so, digging out this difference and identifying the characteristic grounding styles of aged people are a necessity for scientific treatments of communicative problems commonly attributed to aged people, as well as informed design of dialogue systems that interact with aged people.", "cite_spans": [ { "start": 567, "end": 589, "text": "(Ryan & Hummert, 1995)", "ref_id": null }, { "start": 718, "end": 745, "text": "(Kemper & Vandeputte, 1995)", "ref_id": "BIBREF2" } ], "ref_spans": [], "eq_spans": [], "section": "", "sec_num": null }, { "text": "Nevertheless, most studies related to communications of aged people have focused on the influence of aging on basic linguistic competence such as word retrievals (Tatsumi 1997, Ministry of Health and Welfare 1999) , and studies of their behaviors in actual dialogues have been quite scarce. The studies by Ryan & Hummert (1995) and Kemper & Vandeputte (1995) did touch on the conversational styles adopted by aged people, but their focus was more on the conversational styles adopted by younger people toward aged people. The study by Bortfeld et al. (2001) was exceptional in that it measured the number of fillers, restarts, and repairs to evaluate disfluency of aged people's speech in conversational settings. Yet, the focus of this study was on the characteristics of speech by individual conversants, rather than the ways in which they interact. Thus, conversational styles of aged people, let alone their grounding styles, have been seldom studied systematically.", "cite_spans": [ { "start": 162, "end": 199, "text": "(Tatsumi 1997, Ministry of Health and", "ref_id": null }, { "start": 200, "end": 213, "text": "Welfare 1999)", "ref_id": null }, { "start": 306, "end": 327, "text": "Ryan & Hummert (1995)", "ref_id": null }, { "start": 332, "end": 358, "text": "Kemper & Vandeputte (1995)", "ref_id": "BIBREF2" }, { "start": 535, "end": 557, "text": "Bortfeld et al. (2001)", "ref_id": "BIBREF0" } ], "ref_spans": [], "eq_spans": [], "section": "", "sec_num": null }, { "text": "Under this situation, we have no specific hypothesis as to whether aged people have any characteristic grounding styles and if so, what they are. Our purpose in this study is therefore exploratory, dedicated to a search for plausible hypotheses about grounding styles of aged dyads, rather than a verification of their existence and specifics. Although our study is thus hypothetical in nature, we did identify two clearly distinguishable grounding styles in our data, correlated with the age ranges of conversants. In the following, we will report what they are, how they are found, and what implications their difference has as to some negative stereotypes about conversations with aged people.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "", "sec_num": null }, { "text": "The details of the corpus, its collection, and the assignment of tags are given below.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Method", "sec_num": "2" }, { "text": "Our data consist of 5 dialogues by older dyads (age range: 80-88, sd: 2.1) and 5 dialogues by younger dyads (age range: 23-39, sd: 5.2). Most of the older participants were therefore so-called \"older-olds,\" meaning that they were between 75 and 85 years of age. Familiarities of the older dyads were strong: they had known each other since they were children or after they moved into the current areas of residence, although they started talking each other frequently after they joined an common local association several years ago. The younger dyads were also familiar with each other: they had worked in the same department of an institute for at least 3 months and up to 5 years and had talked with each other almost every working day. Dialogues of older dyads were videotaped in a community center that the subjects regularly visit for activities of their association. Videotaping of younger adults took place in a lounge space of their work place. The dialogues were casual conversations in Japanese, involving no specific objectives or tasks. The dialogue topics between older dyads included community events, common friends, their girlhood, and garden work, while the topics between younger adults included their job, their children, their cars, and common friends. The length of a dialogue in either group was from 30 to 50 minutes. Older adults participated in our data collection on the voluntary basis, while we paid a small amount of money for participation of younger adults.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Dialogue Data", "sec_num": "2.1" }, { "text": "We transcribed a 5-minute portion after 3 minutes from the beginning of each dialogue. All words and word fragments audible to the ear were transcribed, including overlapping speech, nonlexical fillers (such as \"uh\"), and other vocalizations (such as laughter and whistle). After each dialogue was transcribed, we checked the videotape to add data of all noddings and salient gestures (facial expressions and hand movements) to the transcription. The transcriber solicited the help of some local residents to understand some of the dialectical expressions contained used in dialogues by older dyads.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Transcription", "sec_num": null }, { "text": "On the basis of Traum's finite-state transition model of grounding, we divided all speech in our dialogue data into utterance units (UUs), namely, \"continuous speech by the same speaker, punctuated by prosodic boundaries (including pauses of significant length and boundary tones)\" (Traum, 1994) . One of the present authors then classified each utterance unit into one of seven categories of grounding acts, according to its contribution to a grounding process. Table 1 shows the seven categories of grounding acts and their definitions. An acknowledgement claiming or demonstrating understanding of a previous utterance. It may be either a repetition or paraphrase of all or part of the utterance, an explicit signal of comprehension such as \"ok\" or \"uh huh\", or an implicit signaling of understanding.", "cite_spans": [ { "start": 282, "end": 295, "text": "(Traum, 1994)", "ref_id": "BIBREF5" } ], "ref_spans": [ { "start": 463, "end": 470, "text": "Table 1", "ref_id": "TABREF0" } ], "eq_spans": [], "section": "Coding", "sec_num": null }, { "text": "Changes the content of the current DU.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Repair (repair)", "sec_num": null }, { "text": "A request for a repair by the other party.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "ReqRepair (reqRepair)", "sec_num": null }, { "text": "Attempt to get the other agent to acknowledge the previous utterance. Cancel (cancel) Closes off the current DU as ungrounded.", "cite_spans": [ { "start": 77, "end": 85, "text": "(cancel)", "ref_id": null } ], "ref_spans": [], "eq_spans": [], "section": "ReqAck (reqAck)", "sec_num": null }, { "text": "To check the reliability of our coding scheme, we asked two independent subjects to code a portion of our data and compared their results with our own. As an instruction to these independent coders, we produced a coding manual that lists Traum's definitions of the seven categories of grounding acts (see Table 1 ), sample utterances of each category, and some general notes on treatments of marginal cases such as fillers, simultaneous utterances, and utterances with both Init and Ack functions. The subjects did some exercise coding before actual coding, and they were encouraged to ask questions about the coding standard during the exercise period. Both transcribed text and audio recording were available to the subjects during the actual coding. The subjects coded two partial dialogues one from older dyads and the other from younger dyads. Each dialogue was about 2.5 minute in length.", "cite_spans": [], "ref_spans": [ { "start": 305, "end": 312, "text": "Table 1", "ref_id": "TABREF0" } ], "eq_spans": [], "section": "Reliability", "sec_num": null }, { "text": "The interrater reliability among the original coder and these two subjects was excellent; there was an 83% agreement, with a Cohen's Kappa of .78, for dialogues by older dyads (UUs = 138, coders = 3, categories = 9) and a 78% agreement, with a Cohen's Kappa .77, for dialogues by younger dyads (UUs = 128, coders = 3, categories = 9).", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Reliability", "sec_num": null }, { "text": "Thus, our data comprise both dialogues by older dyads and ones by younger dyads, but these two groups of dialogues were not tightly controlled in their settings and therefore are not appropriate for exact comparison. For the limited purpose of exploratory data analysis, however, it is convenient to tentatively divide our data into these two groups and compare them from various points of view. This section reports four main results of such exploratory comparisons, in terms of (1) the length of disourse units, (2) the types of frequently used acknowledgments, and (3) the frequencies of collaborative completions and echoic responses.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Results", "sec_num": "3" }, { "text": "The two groups of dialogues had no significant differences in the total numbers of discourse units (older adults, 446; younger adults, 482) and the total numbers of utterance units (older adults, 1406; younger adults, 1309). This roughly means that older dyads grounded information as frequently as younger dyads, and the numbers of utterances needed for each grounding are about the same in older dyads and in younger dyads. Also, the frequencies of occurrences of the seven categories of grounding acts were about the same in older dyads and younger dyads (Figure 1 ). ", "cite_spans": [], "ref_spans": [ { "start": 558, "end": 567, "text": "(Figure 1", "ref_id": null } ], "eq_spans": [], "section": "Utterance units and discourse units", "sec_num": "3.1" }, { "text": "Generally, an \"acknowledgement act\" appears at the end of a grounding unit, indicating that conversants have mutually understood what the presenter said. In contrast, an \"initiate act\" appears at the beginning of a grounding unit, initiating the presentation of a new information unit. Sometimes, a speaker performs both initiation and acknowledgement at the same time, by producing an initiation utterance whose content presupposes the understanding of the preceding presentation. An example is found in dialogue excerpt 1from an older dyad, where one of B's utterances, \"a-hon-na-ikatta (that's good),\" acknowledges the previous utterance of speaker A while initiating B's presentation of her own information. We found that acknowledgements of this type accounts for a 27% of all acknowledgement utterances made by younger adults, while they account for only a 10% of those by older adults (see the top section of each column in Figure 2) . The difference was significant (t(5)=2.87, .01=w 5,5 (0.01)=39, two-tailed), the ratio of general-purpose acknowledgements (W 0 >=w 5,5 (0.025)=38, two-tailed) and the counts of solicited post-grounding acknowledgements (E(R)=27.5,", "cite_spans": [], "ref_spans": [ { "start": 310, "end": 317, "text": "Table 2", "ref_id": null } ], "eq_spans": [], "section": "Two Grounding Styles", "sec_num": "4.4" }, { "text": "V(R)=36, Z 0 =3.75, Z(a)=3.09, p<0", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Two Grounding Styles", "sec_num": "4.4" }, { "text": ".002, two-tailed). Thus, our hypothesis associating style A to older dyads and style B to younger dyads receives some initial supports from this analysis.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Two Grounding Styles", "sec_num": "4.4" }, { "text": "On the other hand, Table 2 also shows certain exceptions to this association, namely, some older dyads with style B and some younger dyads with style A. This is only natural as we can easily imagine older people who talk like younger people as well as younger people who talk like older people. Styles are styles, and they are not natural traits exclusively possessed by particular species of creatures. Table 2 . Rank order of ten dialogues according to the ratios of dedicated acknowledgements, the ratios of general-purpose acknowledgements, and the counts of post-grounding acknowledgements (solicited acknowledgements and spontaneous acknowledgements) ", "cite_spans": [], "ref_spans": [ { "start": 19, "end": 26, "text": "Table 2", "ref_id": null }, { "start": 404, "end": 411, "text": "Table 2", "ref_id": null } ], "eq_spans": [], "section": "Two Grounding Styles", "sec_num": "4.4" }, { "text": "As a pilot study of the grounding styles of older dyads, we analyzed dialogue data featuring both older and younger dyads, and identified a certain contrast in grounding style that separates the two groups fairly clearly. In the style associated with older dyads, people use more dedicated acknowledgements than dual-functional ones, use more general-purpose acknowledgements than special-purpose ones, and use relatively many post-grounding acknowledgements. In the style associated with younger dyads, the opposite tendencies on the use of acknowledgements hold. On the basis of this contrast, we also discussed the implications of the adoption of the first grounding style, and conjectured that some of the negative stereotypes about conversations with older people may be based on this grounding style they adopt. Certainly, as the result of an exploratory study, all these observations are purely hypothetical. We only hope that this study has provided enough materials to set a stage for a further, more controlled comparison of grounding styles of older and younger dyads.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Conclusion", "sec_num": "5" } ], "back_matter": [], "bib_entries": { "BIBREF0": { "ref_id": "b0", "title": "Disfluency Rates in Conversation: Effects of Age, Relationship, Topic, Role, and Gender", "authors": [ { "first": "H", "middle": [], "last": "Bortfeld", "suffix": "" }, { "first": "S", "middle": [ "D" ], "last": "Leon", "suffix": "" }, { "first": "J", "middle": [ "E" ], "last": "Bloom", "suffix": "" }, { "first": "M", "middle": [ "F" ], "last": "Schober", "suffix": "" }, { "first": "S", "middle": [ "E" ], "last": "Brennan", "suffix": "" } ], "year": 2001, "venue": "Language and Speech", "volume": "44", "issue": "2", "pages": "123--147", "other_ids": {}, "num": null, "urls": [], "raw_text": "Bortfeld, H., Leon, S.D., Bloom, J.E., Schober, M.F., & Brennan, S.E. 2001. Disfluency Rates in Conversation: Effects of Age, Relationship, Topic, Role, and Gender. Language and Speech, 44(2): 123-147", "links": null }, "BIBREF1": { "ref_id": "b1", "title": "Using language", "authors": [ { "first": "H", "middle": [ "H" ], "last": "Clark", "suffix": "" } ], "year": 1996, "venue": "", "volume": "", "issue": "", "pages": "", "other_ids": {}, "num": null, "urls": [], "raw_text": "Clark, H.H. 1996. Using language. Cambridge: Cambridge University Press.", "links": null }, "BIBREF2": { "ref_id": "b2", "title": "Communication predicaments of aging: patronizing behavior toward elder adults", "authors": [ { "first": "J", "middle": [ "L ;" ], "last": "Fleiss", "suffix": "" }, { "first": "S", "middle": [], "last": "Kemper", "suffix": "" }, { "first": "D", "middle": [], "last": "Vandeputte", "suffix": "" } ], "year": 1981, "venue": "Statistical Methods of Rates and Proportions", "volume": "14", "issue": "", "pages": "144--166", "other_ids": {}, "num": null, "urls": [], "raw_text": "Fleiss, J.L. 1981. Statistical Methods of Rates and Proportions. John Wiley & Sons, Inc. Kemper, S. & Vandeputte, D. 1995. Speech adjustments to aging during a referential communication task. Journal of Language & Social Psychology, 14(1/2): 40-59 Ministry of Health and Welfare. 1999. White paper. Tokyo: Gyousei. Ryan, E.B. & Hummert, M.L. 1995. Communication predicaments of aging: patronizing behavior toward elder adults. Journal of Language & Social Psychology, 14'1/2): 144-166.", "links": null }, "BIBREF3": { "ref_id": "b3", "title": "Informational and dialogue-coordinating functions of prosodic features of Japanese echoic responses", "authors": [ { "first": "A", "middle": [], "last": "Shimojima", "suffix": "" }, { "first": "Y", "middle": [], "last": "Katagiri", "suffix": "" }, { "first": "H", "middle": [], "last": "Koiso", "suffix": "" }, { "first": "M", "middle": [], "last": "Swerts", "suffix": "" } ], "year": 1999, "venue": "Speech Communication", "volume": "36", "issue": "", "pages": "113--132", "other_ids": {}, "num": null, "urls": [], "raw_text": "Shimojima, A., Katagiri, Y., Koiso, H, & Swerts, M. 1999. Informational and dialogue-coordinating functions of prosodic features of Japanese echoic responses. Speech communication. Speech Communication, 36(2002):113-132", "links": null }, "BIBREF4": { "ref_id": "b4", "title": "Word finding difficulty as aging", "authors": [ { "first": "I", "middle": [], "last": "Tatsumi", "suffix": "" } ], "year": 1997, "venue": "Gengo", "volume": "26", "issue": "13", "pages": "26--33", "other_ids": {}, "num": null, "urls": [], "raw_text": "Tatsumi, I. 1997. Word finding difficulty as aging, Gengo, 26(13): 26-33", "links": null }, "BIBREF5": { "ref_id": "b5", "title": "A Computational Theory of Grounding in Natural Language Conversation", "authors": [ { "first": "D", "middle": [], "last": "Tannen", "suffix": "" }, { "first": "D", "middle": [ "R" ], "last": "Traum", "suffix": "" } ], "year": 1989, "venue": "Taking Voices: Repetition, Dialogue, and Imagery in Conversational Discourse", "volume": "", "issue": "", "pages": "", "other_ids": {}, "num": null, "urls": [], "raw_text": "Tannen, D. 1989. Taking Voices: Repetition, Dialogue, and Imagery in Conversational Discourse. Cambridge: Cambridge University Press. Traum, D.R. 1994. A Computational Theory of Grounding in Natural Language Conversation. Technical Report 545. University of Rochester.", "links": null } }, "ref_entries": { "FIGREF0": { "type_str": "figure", "uris": null, "num": null, "text": "Ratios of the three types of acknowledgements in older dyads and younger dyads" }, "TABREF0": { "type_str": "table", "num": null, "content": "
InitiateAn initial utterance component of a
(init)Discourse unit.
ContinueA continuation of a previous act
(cont)performed by the same speaker.
Acknowl-
edgement
(ack)
", "html": null, "text": "Seven categories of grounding acts" } } } }