Short CommunicationIndividual differences in switching and inhibition predict perspective-taking across the lifespan
Introduction
During interactive discourse, we often rely on estimates about what is shared with an interlocutor (common ground) and what is not (privileged ground). Such estimates typically require perspective-taking to consider another’s knowledge and how it may differ from one’s own. The process by which people consider others’ perspectives is essential to communication, yet questions remain regarding its underlying cognitive mechanisms, and about possible variation in individual perspective-taking abilities.
A central question in language research is the degree to which linguistic behaviors reflect language-specific or domain-general mechanisms. For perspective-taking, executive functions (EF) are theorized to play a role in inhibiting privileged information when considering common ground. Some studies show that differences in inhibitory control and working memory predict communicative perspective-taking performance (Brown-Schmidt, 2009, Lin et al., 2010, Wardlow, 2013), whereas others have failed to replicate these patterns (Brown-Schmidt and Fraundorf, 2015, Ryskin et al., 2015, Ryskin et al., 2014).
This disparity may reflect the participant populations: the aforementioned studies focused exclusively on college-aged students. Compared to children and elderly adults, whose cognitive control exhibit substantial variability, young adults as a group likely operate at peak cognitive capacity, potentially concealing any influence of individual differences (Brown-Schmidt and Fraundorf, 2015, Cepeda et al., 2001, Comalli et al., 1962, Zelazo et al., 2004). This performance advantage in early adulthood extends to interactive dialogue: younger adults use more succinct, contextually-relevant, partner-specific language, whereas older adults are often less effective in making adjustments for particular partners (Bortfeld et al., 2001, Healey and Grossman, 2016, Horton and Spieler, 2007, Lysander and Horton, 2012).
In this context, it is reasonable to ask whether age-related communicative patterns are mediated by underlying differences in EF. In children, inhibitory control is negatively correlated with communicative egocentrism (Nilsen & Graham, 2009). At the other end of the lifespan, Wardlow, Ivanova, and Gollan (2014) observed that perspective-taking correlates more strongly with EF in Alzheimer’s patients than in healthy age-matched controls. However, those EF measures were simplified for the patients, leading to ceiling-level performance in controls and possibly obscuring a relationship between perspective-taking and cognitive mechanisms in older adults. The current study addresses this by testing healthy adults of all ages.
As noted above, EF capacities targeted in prior perspective-taking work have been primarily limited to inhibition and working memory. Equally important, however, may be the ability to efficiently switch attention between perspectives, mediated by mechanisms of attentional shifting (Miyake et al., 2000) involving a combination of both inhibition and release from inhibition/refocusing of attention. People restrict attention to perspective-relevant information less efficiently when switching from a previous perspective, as shown in comparisons of trials that require a perspective shift from a previous context with trials that do not (Bradford et al., 2015, Ryskin et al., 2016, Ryskin et al., 2014). This suggests a role for domain-general switching capacities in perspective-taking, alongside inhibition.
Here, we explore the simultaneous contributions of inhibition and switching to performance in a conversational perspective-taking task. Interestingly, these EF capacities are associated with two semi-independent (yet possibly concurrently engaged) modes of cognitive control. The first is a ‘proactive’ (Braver, 2012) or ‘goal-shielding’ (Goschke & Dreisbach, 2008) mode, which prioritizes the maintenance of internal goals, preventing interference from irrelevant information at the price of ignoring potentially significant contextual cues. The second is a ‘reactive’ or ‘background monitoring’ mode, which enhances the sensitivity to contextual cues at the expense of goal-maintenance. In conversation, speakers must balance the salience of their own perspectives against the need to attend to the interlocutor’s. These pressures may require both the inhibition of salient-but-irrelevant information along with the readiness to refocus attention on appropriate contextual information. An individual’s ‘proactive’ goal maintenance could be taken as the ability to consistently inhibit privileged context. In contrast, a ‘reactive’ mode allows for enhanced sensitivity to contextual cues, requiring modulation of inhibition when a speaker switches perspectives.
To measure these capacities, we used the Test of Everyday Attention (TEA) (Robertson, Ward, Ridgeway, & Nimmo-Smith, 1994), a well-established clinical test with one subtest examining inhibition alone and another examining switching (jointly tapping into inhibition and release from inhibition) in a closely-related task. Recent work on bilingualism and language learning has used the TEA (Bak et al., 2016, Bak et al., 2014, Vega-Mendoza et al., 2015). However, it has not been used in linguistic perspective-taking research. Thus, we hope to diversify approaches to analyzing EF capacities in communicative contexts.
Our perspective-taking study adapts a referential communication task from prior research (e.g., Wardlow, 2013, Wardlow et al., 2014) whereby a speaker identifies target objects presented in 4-object displays for a listener. On experimental trials, a size-contrasting competitor is also present. For common ground (CG) trials, both the target and competitor are mutually visible, while for privileged ground (PG) trials, the target is visible but the competitor is occluded from the listener’s view. Successful perspective-taking is indexed by the relative frequency with which speakers include appropriate modification on CG trials but refrain from doing so on PG trials.
Section snippets
Participants
Participants (N = 121) were recruited from Scottish educational institutions, including the University of Edinburgh Psychology Volunteer Panel, the University of Edinburgh Centre for Open Learning, and Sabhal Mòr Ostaig. Written informed consent was obtained. Prior to analysis, we removed data from 21 participants: 18 non-native speakers of English, 1 aphasiac, 1 with abnormally low TEA scores, and 1 due to technical malfunction. We report data from 100 native English-speaking participants aged
Results
As expected, performance was at ceiling (M = 99%) on the TEA Elevator Task, so this will not be considered further.2 An effect of participant age was found for the TEA switching subtest (linear regression: β = −8.415, p < 0.05) but not for the inhibition subtest (β = −1.424, p = 0.44). Following Brown-Schmidt and Fraundorf
Discussion
Based on the performance of a large sample of individuals varying widely in age, we provide support for the claim that individual differences—both in age and domain-general cognitive capacities—contribute to variability in communicative perspective-taking. While we cannot rule out other contributing factors, like how comfortable older participants were in responding via iPad, our results reveal striking age-related differences in the influence of both inhibition and switching: for young adults,
Acknowledgements
We are very grateful to all those who participated in this study. We would also like to thank the University of Edinburgh PPLS Research Support Grants and Principal's Go Abroad Fund for generously funding this research, all institutions that helped with recruitment including the University of Edinburgh and Sabhal Mòr Ostaig, as well as Casey Noble for helping to create the communication task.
References (31)
- et al.
Random effects structure for confirmatory hypothesis testing: Keep it maximal
Journal of Memory and Language
(2013) - et al.
From self to social cognition: Theory of Mind mechanisms and their relation to executive functioning
Cognition
(2015) The variable nature of cognitive control: A dual mechanisms framework
Trends in Cognitive Sciences
(2012)- et al.
Interpretation of informational questions modulated by joint knowledge and intonational contours
Journal of Memory and Language
(2015) - et al.
Watching the eyes when talking about size: An investigation of message formulation and utterance planning
Journal of Memory and Language
(2006) - et al.
Reflexively mindblind: Using theory of mind to interpret behavior requires effortful attention
Journal of Experimental Social Psychology
(2010) - et al.
The unity and diversity of executive functions and their contributions to complex “frontal lobe” tasks: A latent variable analysis
Cognitive Psychology
(2000) - et al.
The relations between children’s communicative perspective-taking and executive functioning
Cognitive Psychology
(2009) - et al.
Visuospatial perspective-taking in conversation and the role of bilingual experience
Journal of Memory and Language
(2014) - et al.
Listeners use speaker identity to access representations of spatial perspective during online language comprehension
Cognition
(2016)