Gaze direction conveys a wealth of personal information, allowing one to quickly follow other people’s current focus of attention and infer their intentions as well as mental states (Colombatto et al., 2020; Frischen et al., 2007; Kuhn & Kingstone, 2009). Utilizing information on gaze direction from social partners to deploy our attention resources accordingly is considered an essential ability in social interaction (Dalmaso et al., 2016; Rogers et al., 2014).
It has been shown that observers tend to automatically follow the gaze of others, and this has been investigated by a modified Posner paradigm in previous studies, i.e. the gaze-cueing paradigm (Friesen & Kingstone, 1998; Frischen & Tipper, 2004; Kuhn & Kingstone, 2009). During this type of task, a face with a direct gaze is presented in the centre of a screen, which then looks to the left or right side of the screen. After a short variable stimulus onset asynchrony (SOA), a peripheral target randomly emerges on the left or right side of the screen. Participants are required to detect the target as quickly and accurately as possible. Although the gaze direction of the central face is not predictive or counterpredictive of the target location, participants’ reaction times (RTs) to targets at the gazed-at location are shorter than targets at the non-gazed-at location (Dalmaso et al., 2020; Friesen & Kingstone, 1998; Frischen & Tipper, 2004), a social phenomenon termed the gaze-cueing effect (GCE).
More recent studies have found that the GCE is modulated by top-down factors, such as political affiliation (Cazzato et al., 2015; Liuzza et al., 2011), social exclusion (Capellini et al., 2019), competitiveness (Ciardo et al., 2015), and trustworthiness (Süßenbach & Schönbrodt, 2014). For example, Capellini (2019) used a Cyberball game to make participants feel excluded before performing a gaze-cueing task. They found that participants who felt excluded in the prior task exhibited a reduced GCE compared with those who were included. Moreover, faces associated with competitive relationships trigger a stronger GCE relative to faces associated with cooperative relationships (Ciardo et al., 2015). Dalmaso (2016) suggested that faces that had engaged in joint gaze with participants in a previous task elicited a standard GCE relative to those engaged in disjoint gaze, showing that knowledge of previous interactions could also influence later gaze-following behaviours in these people. The relationship between the observer and cue face can modulate the GCE, suggesting that social interaction with others influences how their gaze shifts our attention.
Guilt, as a social emotion, arises in and regulates social interactions (Pfeiffer et al., 2013). In an interpersonal context, we feel guilty when our fault inflicts loss upon others (Yu et al., 2014). As a moral emotion, guilt is related to understanding a victim’s thoughts, feelings, and attitudes towards transgressors (Yu et al., 2014; Zhu, Wu, et al., 2019). The feeling of guilt makes individuals consider the concerns of others and increases reparative intentions, which helps enhance social relationships (de Hooge et al., 2011). Moreover, participants who felt guilty avoided making eye contact that would lead them to become involved in an interaction (Van Dillen et al., 2017; Yu et al., 2017). Yu et al. (2017) found that participants fixated less on a partner’s eyes in a high-guilt condition in which participants caused the partner’s pain than in a control condition in which the partner caused pain. In this experiment, the arousal of guilt measured by skin conductance response was influenced by whether or not the victim’s eyes were seen. It has been demonstrated that interpersonal guilt is processed as a social threat (Furukawa et al., 2019). After experiencing guilt, making eye contact with the victim induces anxiety. Thus, participants who feel guilty are more likely to be influenced by the victim’s eyes (Yu et al., 2017). Moreover, previous research has found that the superior temporal sulcus region, which is known to play a prominent role in gaze cueing (Burra et al., 2017), is also involved in the processing of guilt (Takahashi et al., 2004). Additional evidence from neuroimaging studies suggests that the processing of guilt engages the same brain regions involved in social cognition and theory of mind, i.e. the prefrontal and temporal regions (Li et al., 2020; Takahashi et al., 2004; Zhu, Feng, et al., 2019). Thus, guilt might have important effects on the perception of gaze direction and gaze-oriented attention.
The present study investigated whether the GCE is modulated by a guilt-directed face in a prior interaction. In the first task, we employed a dot-estimation task to induce guilt and manipulate the relationship between the participants and the cueing faces, with one face for the guilt condition (participant-caused loss) and the other for the control condition (both win). This method of inducing guilt has been employed in previous studies (Leng et al., 2017; Sánchez-García et al., 2019; Yu et al., 2014). The faces were then used as the central face cue in a subsequent gaze-cueing task. Given that previous research has found that guilt could cause individuals to avoid eye contact with a victim, we anticipated that participants might follow the gaze shift less in the guilt condition compared with the control condition. In other words, participants might exhibit a reduced GCE when the cues were victims’ faces.