Elsevier

Cognition

Volume 157, December 2016, Pages 352-357
Cognition

Brief article
When we cannot speak: Eye contact disrupts resources available to cognitive control processes during verb generation

https://doi.org/10.1016/j.cognition.2016.10.002Get rights and content

Abstract

Although eye contact and verbal processing appear independent, people frequently avert their eyes from interlocutors during conversation. This suggests that there is interference between these processes. We hypothesized that such interference occurs because both processes share cognitive resources of a domain-general system and explored the influence of eye contact on simultaneous verb generation processes (i.e., retrieval and selection). In the present experiment, viewing a movie of faces with eyes directed toward the viewer delayed verbal generation more than a movie of faces with averted eyes; however, this effect was only present when both retrieval and selection demands were high. The results support the hypothesis that eye contact shares domain-general cognitive resource with verb generation. This further indicates that a full understanding of functional and dysfunctional communication must consider the interaction and interference of verbal and non-verbal channels.

Introduction

Face-to-face communication is a complex and important aspect of human behavior. The simultaneous functioning of verbal and non-verbal processes, such as verb generation for nouns (Snyder & Munakata, 2008) and eye contact (Senju & Johnson, 2009), is essential for face-to-face communication. The difficulty of this simultaneous functioning is reflected in the frequent observation that people avert their eyes from interlocutors during conversation (Doherty-Sneddon, Bonner, & Bruce, 2001), despite the importance of face-to-face perception in regulating interactions (Kleinke, 1986, Uono and Hietanen, 2015). Indeed, the phenomenon that face-to-face contact disrupts verbal processing has been shown experimentally (e.g., answering general knowledge/mathematics questions; Glenberg, Schroeder, & Robertson, 1998).

Eye contact during face-to-face perception appears to have unique effect on cognitive control processes. For instance, Markson and Paterson (2009) found that maintaining eye contact with an experimenter disrupted participants’ visuospatial imagination to a greater degree than did viewing a static or dynamic visual stimulus, and the effect got stronger as the task became more difficult (Buchanan et al., 2014). Additionally, a study using the visual word-color Stroop task with simultaneous presentation of gaze stimuli showed that being watched selectively disrupts inhibitory control of the predominant response (Conty, Gimmig, Belletier, George, & Huguet, 2010). Although these intriguing results suggest that eye contact affects cognitive control processes, it remains unclear what aspects of cognitive control processes are affected by eye contact because these previous tasks included some visual factors that potentially compete with eye contact. For example, eye contact may compete with visuospatial imagination for visual domain-specific cognitive processes (Markson & Paterson, 2009), while the simultaneously presented eye stimulus during the visual color-word Stroop task could not be attended to at the same time as the Stroop stimulus (Conty et al., 2010), meaning that the results reflect the effect of being watched rather than the effect of eye contact.

In the present study, we attempted to clarify how eye contact interferes with cognitive control processes during verb generation which does not recruit visual processes but still comprises multiple, dissociable processes (Snyder and Munakata, 2008, Snyder et al., 2010). Verb generation requires the retrieval and selection of words from multiple possible alternatives. As cognitive demands of retrieval and selection processes vary depending on word properties (Fig. 1A), using the auditory version of the verb generation task (VGT), which empirically controls cognitive demands (Kajimura, Inohara, Utsumi, & Nomura, submitted for publication), would help to demonstrate which cognitive control processes are affected by eye contact, following the assumptions below.

For the cognitive control processes of verb generation, we drew on the assumptions of working memory research regarding the conceptualization of domain-specific and domain-general processes (Fig. 1B; Baddeley, 2007, Logie, 2011): namely, (1) highly demanding cognitive processes require more cognitive resources for domain-specific processes than do less demanding processes, and (2) when cognitive demands are beyond the capacities of domain-specific cognitive resources or coordinated operation of multiple domains is needed, the domain-general cognitive system becomes influential. According to these assumptions, if eye contact recruits the domain-general cognitive system rather than the domain-specific system, it would disrupt verb generation only when the cognitive demands of verb generation are beyond the capacities of retrieval- and selection-specific resources or coordinated operation of retrieval and selection is needed (Fig. 1B represents the latter possibility). On the other hand, if eye contact recruits either domain-specific process, it would show a main effect; in other words, interference would occur in all demand conditions because the domain-specific conditions should be involved even in the condition with the lowest demand.

Section snippets

Power analysis

To determine a sufficient sample size, we conducted a power analysis using G∗power software (version 3.1; Faul, Erdfelder, Lang, & Buchner, 2007) before starting the experiment. We utilized conservative criteria for conducting a three-way repeated measures ANOVA. Specifically, we aimed to achieve 80% statistical power for a medium effect size (η2  0.06) on a three-way between-subjects ANOVA, because within design enable to eliminate the variance derived to individual difference from variance of

Ratings of face stimuli

Before the main analyses, differences in the ratings of face stimuli between the directed- and averted-gaze faces were analyzed using two-sample t-tests. The results are shown in Table 1. The t-tests showed that every aspect except arousal was significantly higher in directed-gaze faces compared to averted-gaze faces. Besides that, we quit to include them as covariates because the variables had interactions with independent variables, which violate an ANCOVA assumption (Stevens, 1996).

Trials eliminated due to gaze not being maintained

Trials in

Discussion

The present study explored how eye contact interferes with the cognitive control processes for verbal processing that are the foundation of face-to-face communication, using controlled visual stimuli and a verb generation task that does not share visual-specific processes with eye contact and comprises difficulty-graded and separable domain-specific processes. The current results (1) support recent findings on eye contact effects, in that eye contact automatically recruits and/or interferes

Conflict of interest

The authors declared that they had no conflicts of interest with respect to their authorship or the publication of this article.

Funding

This work was supported by grants from the Japan Society for the Promotion of Science (23500329, 24240041).

Acknowledgments

We thank professor Sakiko Yoshikawa at KRC for letting us use the FED. We thank Ryunosuke Oka, Kaichi Yanaoka, Daiki Hiraoka, and Satoru Nishiyama for help with data collection. We thank Masataka Nakayama for helpful comments.

References (24)

  • A.G. Greenwald

    Within-subjects designs: To use or not to use?

    Psychological Bulletin

    (1976)
  • Kajimura, S., Inohara, K., Utsumi, A., & Nomura, M. (submitted for publication). Dissociable effects of retrieval and...
  • Cited by (0)

    View full text