Herbert H. Clark’s research while affiliated with Stanford University and other places

What is this page?


This page lists works of an author who doesn't have a ResearchGate profile or hasn't added the works to their profile yet. It is automatically generated from public (personal) data to further our legitimate goal of comprehensive and accurate scientific recordkeeping. If you are this author and want this page removed, please let us know.

Publications (87)


Talking as if.
  • Conference Paper

March 2008

·

12 Reads

·

4 Citations

Herbert H. Clark

If ordinary people are to work with humanoid robots, they will need to communicate with them. But how will they do that? For many in the field, the goal is to design robots that people can talk to just as they talk to actual people. But a close look at actual communication suggests that this goal isn't realist. It may even be untenable in principle. An alternative goal is to design robots that people can talk to just as they talk to dynamic depictions of other people-what I will call characters. As it happens, ordinary people have a great deal of experience in interpreting the speech, gestures, and other actions of characters, and even in interacting with them. But talking to characters is different from talking to actual people. So once we view robots as characters, we will need a model of communication based on different principles. That, in turn, may change our ideas of what people can and cannot do with robots.




Mixing virtual and actual

November 2006

·

8 Reads

People often communicate with a mixture of virtual and actual elements. On the telephone, my sister and I and what we say are actual, even though our voices are virtual. In the London Underground, the warning expressed in the recording "Stand clear of the doors" is actual, even though the person making it is virtual. In the theater, Shakespeare, the actors, and I are actual, even though Romeo and Juliet and what they say are virtual. Mixtures like these cannot be accounted for in standard models of communication-for a variety of reasons. In this talk I introduce the notion of displaced actions (as on the telephone, in the London Underground, and in the theater) and characterize how they are used and interpreted in communication with a range of modern-day technologies.


Encyclopedia of Cognitive Science

January 2006

·

32 Reads

·

4 Citations

Conversations emerge as people use dialogue to coordinate on joint activities they engage in. People proceed turn by turn as they reach local agreements on the course of each section and subsection, including the opening and closing of the conversation itself.


Coordinating with each other in a material world

October 2005

·

105 Reads

·

172 Citations

Discourse Studies

In everyday joint activities, people coordinate with each other by means not only of linguistic signals, but also of material signals – signals in which they indicate things by deploying material objects, locations, or actions around them. Material signals fall into two main classes: directing-to and placing-for. In directing-to, people request addressees to direct their attention to objects, events, or themselves. In placing-for, people place objects, actions, or themselves in special sites for addressees to interpret. Both classes have many subtypes. Features of material signals were examined in pairs of people who were videotaped as they assembled TV stands, built Lego models, planned furnishings for a house, played piano duets, or bought coffee at Starbucks. In these activities, the pointing and placements were often sustained, creating three phases of signals – initiation, maintenance, and termination – each with its own interpretation.


Speaking While Monitoring Addressees for Understanding

January 2004

·

369 Reads

·

638 Citations

Journal of Memory and Language

Speakers monitor their own speech and, when they discover problems, make repairs. In the proposal examined here, speakers also monitor addressees for understanding and, when necessary, alter their utterances in progress. Addressees cooperate by displaying and signaling their understanding in progress. Pairs of participants were videotaped as a director instructed a builder in assembling 10 Lego models. In one group, directors could see the builders’ workspace; in a second, they could not; in a third, they gave instructions by audiotape. Two partners were much slower when directors could not see the builders’ workspace, and they made many more errors when the instructions were audiotaped. When their workspace was visible, builders communicated with directors by exhibiting, poising, pointing at, placing, and orienting blocks, and by eye gaze, head nods, and head shakes, all timed with precision. Directors often responded by altering their utterances midcourse, also timed with precision.


FIGURE 1 Mean percentage distribution of different project markers relative to topic entry point in the Switchboard corpus. 
TABLE 1 Production Rates of Okay, Uh-Huh, and Yeah in the Switchboard Corpus (Call Beginnings, Endings and Overall). Number of Occurrences, Mean Rate per 1,000 Words, and t Value of Difference From Overall Corpus 
FIGURE 2 Mean percentage distribution of different project markers relative to topic exit point in the Switchboard corpus. 
TABLE 2 Predecessors of Five Types of Project Markers (in percentages) in the Directory Enquiry Corpus 
Figure 5 of 5
Navigating Joint Projects in Telephone Conversations
  • Article
  • Full-text available

January 2004

·

315 Reads

·

48 Citations

Discourse Processes

Conversation coordinates joint activities and the joint projects that compose them. Participants coordinate (1) vertical transitions on entering and exiting joint pro-jects; and (2) horizontal transitions in continuing within them. Transitions are co-ordinated using project markers such as uh-huh, yeah, right, and okay. In the au-thors' proposal, participants use uh-huh, yeah, and right to continue within joint projects, and okay and all right to enter and exit them. This was examined in 2 telephone conversation corpora. Telephone conversations divide into an entry, body, and exit phase, each of which is a joint project. Okay and all right were used to transit from the entry to body and from body to exit, whereas uh-huh, yeah, and right were used within the body. JOINT PROJECTS IN CONVERSATION In conversation, the participants do not just speak—they do things together. These joint actions are normally the reason for their encounter, and their talk is shaped by the need to coordinate them. To understand what people are doing in conversation, one must understand the joint activities they are engaged in. Outside of conversation, individual and joint activities have long been analyzed into hierarchies of projects and subprojects (Cranach, Kalbermatten, Indermühle, & Gugler, 1982; Grosz & Sidner, 1986; Leont'ev, 1981; Miller, Galanter, & Pribram, 1960; Newell & Simon, 1972). Take Betty and Camilla baking a cake to-gether. That joint project will include subprojects, such as preparing the batter, pouring it into a cake pan, and baking it in the oven. Preparing the batter might it-DISCOURSE PROCESSES, 37(1), 1–23 Copyright © 2004, Lawrence Erlbaum Associates, Inc.

Download

Changing Ideas about Reference

January 2004

·

115 Reads

·

143 Citations

How do people refer to things? At first, the answer seems simple: they produce the right expression in the right situation. According to John Searle (1969), for example, to refer to a dog, speakers must produce a referring expression (such as the dog she had with her) with the intention that it pick out or identify the dog for their addressees. But is the answer really this simple? Accounts of how people refer have changed again and again since about 1960, often dramatically. But how have they changed, and why? In this chapter, we offer a selective, largely personal history of these changes as they have played out in the experimental study of reference. Our goal is not a complete history — an impossible ambition — but a better understanding of what reference really is.


Navigating joint projects with dialogue

March 2003

·

256 Reads

·

148 Citations

Cognitive Science A Multidisciplinary Journal

Dialogue has its origins in joint activities, which it serves to coordinate. Joint activities, in turn, usually emerge in hierarchically nested projects and subprojects. We propose that participants use dialogue to coordinate two kinds of transitions in these joint projects: vertical transitions, or entering and exiting joint projects; and horizontal transitions, or continuing within joint projects. The participants help signal these transitions with project markers, words such as uh-huh, m-hm, yeah, okay, or all right. These words have been studied mainly as signals of listener feedback (back-channel signals) or turn-taking devices (acknowledgment tokens). We present evidence from several types of well-defined tasks that they are also part of a system of contrasts specialized for navigating joint projects. Uh-huh, m-hm and yeah are used for horizontal transitions, and okay and all right for vertical transitions.


Citations (77)


... People make sense of the present interaction through the histories they have experienced over time. Thus, successful communication depends on participants building, recognizing, and drawing on shared histories of participation in activity, what Herbert Clark calls common ground (see Clark & Wilkes-Gibbs, 1992). ...

Reference:

Enriching Communicative Environments: Leveraging Advances in Neuroplasticity for Improving Outcomes in Neurogenic Communication Disorders
Referring as a Collaborative Process
  • Citing Chapter
  • June 1990

... Based on these findings, our present studies aim to extend existing knowledge on algorithm aversion and algorithm ambivalence with two contributions. First, in our Study 1 we explore whether the key finding of the algorithm aversion literature (harsher penalties for algorithmic vs. human errors) can be found in a different field: collaborating with an agent-supposedly an AI or a human-in a referential communication task (Brennan & Clark, 1996;Krauss & Weinheimer, 1964;Yule, 2013). Knowing whether increased penalties for algorithmic errors can also found beyond classical delegation tasks would broaden the scope of fields of application. ...

Conceptual Pacts and Lexical Choice in Conversation

Journal of Experimental Psychology Learning Memory and Cognition

... In some cases, this involves the speaker putting extra individual effort into message planning in order to facilitate their partner's comprehension (Clark & Brennan, 1991;Clark & Schaefer, 1989;Clark & Wilkes-Gibbs, 1986;Schober, 1995). In other words, the speakers try to increase the efficacy of the conversation and to reduce the collaborative effort by producing messages that are designed and adapted to their addressee, even if doing so involves increased individual effort, a mechanism called audience design (Clark & Murphy, 1982;Clark & Wilkes-Gibbs, 1986;Fussell & Krauss, 1989;Gann & Barr, 2014;Harris et al., 1980;Nückles et al., 2006;Turner & Knutsen, 2021). Consider these two messages during a conversation between friends planning a meeting: ...

La visée vers l'auditoire dans la signification et la référence
  • Citing Article
  • January 1982

Bulletin de psychologie

... Moreover, several studies suggested that comparative judgements require high-level semantic and linguistic cognitive factors 8-11 . These high-level cognitive factors have been considered the semantic component of comparative judgment 8,[12][13][14][15][16] . The influence of the semantic component on comparative judgement has been first supported by Banks et al. 8 . ...

The locus of the semantic congruity effect in comparative judgments

Journal of Experimental Psychology Human Perception & Performance

... L'examen des interactions produites dans le cadre de l'expérience des Tangrams pour l'étude de la référence amène les auteurs à dégager neuf stratégies linguistiques dont huit éloignées des types d'expressions traditionnellement associées à l'acte de référence 48 (Clark & Bangerter, 2004 : Au cours de la tâche, le directeur aide l'apparieur à placer son jeu dans le même ordre que la séquence cible. Les échanges sont enregistrés afin que les actes référentiels puissent être analysés (Newlands, 1998 : 30 Ces stratégies configurées dans et par le discours sont exploitées par les participants dans la dynamique interactionnelle de l'échange pour i. construire collaborativement une référence précise ii. ...

Changing Ideas about Reference
  • Citing Chapter
  • January 2004

... Languages permit speakers some freedom in how they order information in an utterance. For example, ordering freedom appears through syntactic alternations such as the dative and genitive alternations, shown below, in which roughly equivalent meanings are expressed through different constructions that order the key elements differently: While both forms are possible in such cases, speakers show stable preferences for one order over the other as a function of the words involved, their meanings, and their status within an ongoing discourse (57)(58)(59)(60)(61). Across different syntactic alternations and languages, the words that are chosen to go earlier have a common set of properties: They are typically short, frequent, definite, and discourse-given, meaning that they refer to an entity which has already been mentioned in the discourse context (62)(63)(64). ...

Universals, relativity, and language processing
  • Citing Article
  • January 1978

... Speakers commonly consider their listeners when talking and adjust the characteristics of their speech accordingly (Bell, 1984;Clark & Murphy, 1982). The concept of audience design applies to many different types of listeners, such as those who are hard-of-hearing, those who speak a different language, AI devices, and infants and children (Cohn et al., 2022;Lam & Kitamura, 2012;Uther et al., 2007). ...

Audience Design in Meaning and Reference
  • Citing Chapter
  • December 1982

... e xed effects in our model were our independent variables (i.e., phonotactic probability, lexicality, and presentation order) which were allowed to interact. All analyses had random intercepts of Subject and Item Clark, 1973). e nal and reduced model contained a maximal random effect structure (Barr et al., 2013). ...

The Language-as-Fixed-Effect Fallacy: A Critique of Language Statistics in Psychological Research
  • Citing Article
  • August 1973

Journal of Verbal Learning and Verbal Behavior