Academic journal article Psychonomic Bulletin & Review

The Role of Executive Function in Perspective Taking during Online Language Comprehension

Academic journal article Psychonomic Bulletin & Review

The Role of Executive Function in Perspective Taking during Online Language Comprehension

Article excerpt

During conversation, interlocutors build on the set of shared beliefs known as common ground. Although there is general agreement that interlocutors maintain representations of common ground, there is no consensus regarding whether common-ground representations constrain initial language interpretation processes. Here, I propose that executive functioning-specifically, failures in inhibition control-can account for some occasional insensitivities to common-ground information. The present article presents the results of an experiment that demonstrates that individual differences in inhibition control determine the degree to which addressees successfully inhibit perspective-inappropriate interpretations of temporary referential ambiguities in their partner's speech. Whether mentioned information was grounded or not also played a role, suggesting that addressees may show sensitivity to common ground only when it is established collaboratively. The results suggest that, in conversation, perspective information routinely guides online language processing and that occasional insensitivities to perspective can be attributed partly to difficulties in inhibiting perspective-inappropriate interpretations.

Interlocutors bring separate perspectives to a conversation. For any pair of individuals, some aspects of their perspectives are shared, known as common ground (Stalnaker, 1978), whereas other aspects are private, known as privileged ground. As a conversation progresses, common ground grows when interlocutors introduce new information into the discourse. Common ground is thought to include shared community and cultural experiences, the physically copresent environment, and shared linguistic exchanges (Clark & Marshall, 1981).

Although there is general agreement that interlocutors maintain representations of common ground, the results of studies examining whether perspective guides online interpretation processes have been equivocal, with some results pointing to early use of perspective in online processing (Brown-Schmidt, Gunlogson, & Tanenhaus, 2008; Hanna & Tanenhaus, 2004; Hanna, Tanenhaus, & Trueswell, 2003; Heller, Grodner, & Tanenhaus, 2008; Nadig & Sedivy, 2002) and other results pointing to unreliable or late use of perspective (Keysar, Barr, Balin, & Brauner, 2000; Keysar, Lin, & Barr, 2003). Thus, the challenge for models of the role of perspective in language processing is to account for why perspective sometimes constrains online processing and sometimes does not.

According to perspective-adjustment models, listeners generate perspective-inappropriate interpretations during online interpretation either due to strategic egocentric processing strategies (Keysar et al., 2000) or because perspective-inappropriate interpretations are autonomously activated (Barr, 2008). On the strategic account, online perspective taking occurs only in special circumstances (i.e., when global ambiguities require perspective use). On the autonomous account, a combination of anticipatory perspective taking (e.g., anticipating reference to a shared object) and a failure to use perspective information as words are interpreted online yields results that misleadingly appear to show online use of perspective (Barr, 2008).

Alternatively, constraint-based models maintain that perspective is one of multiple probabilistic, partial constraints that guide online interpretation; perspective- inappropriate interpretations are likely to occur when evidence supports the perspective-inappropriate interpretation, or when perspective information is unreliable or uncertain (Heller et al., 2008).

Each of these accounts provides partial explanations for the competing results in the literature. However, little to no work experimentally manipulates or predicts when online perspective-taking failures are and are not likely to occur. In the present article, I explore the possibility that subtle individual differences in inhibitory control can account, in part, for adults' occasional failure to use perspective information to inhibit perspective-inappropriate interpretations. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.