Restriction of referential domains 1 Running Head Referential domains in conversation This uncorrected manuscript has been accepted for publication in Cognitive Science Real time investigation of referential domains in unscripted conversation a targeted language game approach Sarah Brown Schmidt Michael K Tanenhaus Department of Brain and Cognitive Sciences University of Rochester Please direct correspondence to Sarah Brown Schmidt Beckman Institute 405 N Mathews Ave University of Illinois at Urbana Champaign Urbana IL 61801 tel 217 244 4787 fax 217 333 2922 e mail brownsch uiuc edu Restriction of referential domains 2 Abstract Two experiments examined the restriction of referential domains during unscripted conversation by analyzing the modification and on line interpretation of referring expressions Experiment 1 demonstrated that from the earliest moments of processing addressees interpreted referring expressions with respect to referential domains constrained by the conversation Analysis of eye movements during the conversation showed elimination of standard competition effects seen with scripted language Results from Experiment 2 pinpointed two pragmatic factors responsible for restriction of the referential domains used by speakers to design referential expressions and demonstrated that the same factors predict whether addressees consider local competitors to be potential referents during on line interpretation of the same expressions These experiments demonstrate for the first time that on line interpretation of referring expressions in conversation is facilitated by referential domains constrained by pragmatic factors which predict when addressees are likely to encounter temporary ambiguity in language processing Keywords Eye tracking conversation alignment on line language processing referential communication cohort point of disambiguation Restriction of referential domains 3 1 Introduction Most psycholinguistic research on spoken language comprehension can be divided into one of two traditions each with its roots in seminal work from the 1960s Clark 1992 Trueswell Tanenhaus 2005 and each with its own characteristic theoretical concerns and dominant methodologies The language as product tradition has its roots in George Miller s synthesis of the then emerging information processing paradigm and Chomsky s theory of transformational grammar e g Miller 1962 Miller Chomsky 1963 The product tradition emphasizes the individual cognitive processes by which listeners recover linguistic representations the products of language comprehension Psycholinguistic research within the product tradition typically examines moment by moment processes in real time language processing using carefully controlled stimuli scripted materials and fine grained on line measures that are closely time locked to the input The motivation for on line measures comes from two observations The first is that speech unfolds over time as a series of rapidly changing acoustic events The second is that listeners continuously integrate the input making provisional commitments at multiple levels of representations Marslen Wilson 1973 1975 Tanenhaus Spivey Knowlton Eberhard Sedivy 1995 As a consequence evaluating models of how linguistic representations are accessed constructed and integrated requires data that can only be obtained from response measures that are closely time locked to the input and sensitive to how the listener s representations change as the input unfolds in time One case in point is temporary ambiguity One of the consequences of the combination of sequential input and continuous processing is that listeners are continuously faced with resolving temporary ambiguity at multiple levels of representation For example the initial sounds of clown are briefly consistent with both cloud and clown Response measures that are closely time locked to the input have revealed that as a listener hears clown both cloud and clown are briefly activated with activation to cloud decreasing as soon as coarticulatory information in the vowel becomes more consistent with clown Dahan Magnuson Tanenhaus Hogan 2001 Zwitserlood 1989 Similarly in a context that includes both a large silver fork and a large silver spoon a listener hearing the large silver spoon will actively consider both as potential referents until the disambiguating sounds at the onset of spoon Eberhard Spivey Restriction of referential domains 4 Knowlton Sedivy Tanenhaus 1995 Hanna Tanenhaus Trueswell 2003 Response measures that are not closely time locked to the input are of limited value for examining processes like these which are central to language processing The language as action tradition has its roots in work by the Oxford philosophers of language use e g Austin 1962 Grice 1957 and Searle 1969 and work on conversational analysis e g Schegloff and Sachs 1973 The action tradition focuses on how people use language to perform acts in conversation the most basic form of language use Psycholinguistic research within the action tradition typically examines unscripted interactive conversation involving two or more participants engaged in a cooperative task typically with real world referents and well defined behavioral goals One reason is that many aspects of utterances in a conversation can only be understood with respect to the context of the language use which includes the time place and participant s conversational goals as well as the collaborative processes that are intrinsic to conversation Moreover many characteristic features of conversation emerge only when interlocutors have joint goals and when they participate in a dialogue both as a speaker and an addressee Detailed analyses of participants linguistic behavior and actions in cooperative tasks have provided important insights into how interlocutors track information to achieve successful communication They demonstrate that many aspects of communication establishing successful reference for instance are not simply individual cognitive processes they are achieved as the result of coordinated actions among two or more individuals across multiple linguistic and nonlinguistic exchanges Bangerter 2004 Clark Wilkes Gibbs 1986 Clark Krych 2004 Gergle Kraut Fussell 2004 Schober Brennan 2003 The coordination that emerges through participation in a conversation appears to change how language is understood in a conversation One way to quantify the contribution of this coordination while controlling
View Full Document
Unlocking...