Eyes that bind us: Gaze leading induces an implicit sense of agency

Abstract

Humans feel a sense of agency over the effects their motor system causes. This is the case for manual actions such as pushing buttons, kicking footballs, and all acts that affect the physical environment. We ask whether initiating joint attention - causing another person to follow our eye movement - can elicit an implicit sense of agency over this congruent gaze response. Eye movements themselves cannot directly affect the physical environment, but joint attention is an example of how eye movements can indirectly cause social outcomes. Here we show that leading the gaze of an on-screen face induces an underestimation of the temporal gap between action and consequence (Experiments 1 and 2). This underestimation effect, named 'temporal binding,' is thought to be a measure of an implicit sense of agency. Experiment 3 asked whether merely making an eye movement in a non-agentic, non-social context might also affect temporal estimation, and no reliable effects were detected, implying that inconsequential oculomotor acts do not reliably affect temporal estimations under these conditions. Together, these findings suggest that an implicit sense of agency is generated when initiating joint attention interactions. This is important for understanding how humans can efficiently detect and understand the social consequences of their actions.

In natural interaction with embodied robots we prefer it when they follow our gaze: a gaze-contingent mobile eyetracking study

“…The interaction effect between identity and contingency for the return-to-face saccades indicates that attentional engagement with the robot was facilitated after initiating joint attention, but only if the robot typically followed the participant's gaze. This partly replicates previous findings that people are not merely sensitive to establishing ad-hoc joint attention, but that this sensitivity depends on implicit expectations set by previous interaction (8,12,24,25). Whereas a similar -but screen-based-study found a main-effect in which return-to-face saccades were faster towards the joint-attention robot avatar than towards the disjoint-attention one (8), the current study suggests that an agent's joint-attention disposition drives further interaction.…”

A naturalistic paradigm simulating gaze-based social interactions for the investigation of social agency

“…More precisely, gaze congruency (specifically gaze following and mutual gaze) and gaze contingency (reaction latency) seem to contribute toward these experiences. These findings are in line with previous noninteractive research using still pictures of faces, demonstrating a perceived sense of agency induced by gaze cues (Stephenson et al, 2018). In contrast to the implicit measure of agency employed in the study by Stephenson et al (the intentional binding effect), the present experiment focuses on an explicit sense of social agency by asking participants to rate the intensity of the perceived relatedness of their own gaze towards the gaze of the interaction partner.…”

“…However, the characteristics and the underlying mechanisms of social agency are still ambiguous. Previous research provides important evidence that both an implicit and explicit sense of agency is present within a social context (Pfister, Obhi, Rieger, & Wenke, 2014;Stephenson, Edwards, Howard, & Bayliss, 2018;Weiss, Herwig, & Schütz-Bosbach, 2011) and that gaze behavior provides a useful aspect to study agency both in general (Gregori Grgič, Crespi, & de'Sperati, 2016) as well as in social situations (Pfeiffer et al, 2012(Pfeiffer et al, , 2014Pfeiffer, Timmermans, Bente, Vogeley, & Schilbach, 2011;Stephenson et al, 2018). In a recent publication, Stephenson et al applied the widely used intentional binding paradigm including social cues.…”

Building blocks of joint attention: Early sensitivity to having one’s own gaze followed

“…Since IJA may be a more sensitive index of socio-cognitive development than RJA ( Tomasello, 1995 ), as well as a more robust symptom of neurodevelopmental disorders such as ASD ( Mundy, 2018 ; Mundy et al, 2016 ), research aimed at elucidating the early building blocks of IJA is of particular importance. Accordingly, our study was designed to investigate the emergence of a critical, but largely neglected, component of IJA: detecting when one’s own gaze has been followed ( Stephenson et al, 2018 ). Essential for determining whether or not one has been successful in directing someone else’s gaze, an early sensitivity to having one’s gaze followed is likely foundational for the intentional directing of another’s attention later on in infancy ( Mundy, 2018 ).…”