Foot cues can elicit covert orienting of attention

A broad and increasing body of literature has shown that people tend to orient their attention towards the same location indicated by a variety of spatial signals provided by their conspecifics, such as eye-gaze direction (Dalmaso et al., 2020c; Frischen et al., 2007; McKay et al., 2021), pointing gestures (Ariga & Watanabe, 2009; Langton & Bruce, 2000), and head and body turns (Azarian et al., 2017; Langton & Bruce, 1999). The ability to pay attention to the same location as another individual, which is often referred to as ‘social attention’ (Kingstone, 2009), is essential in daily-life social interactions, as it allows people to establish fluent relationships and interactions with both others and the physical environment in which they act (Capozzi & Ristic, 2018; Emery, 2000).

The impact of social stimuli on visual orienting of attention has generally been investigated by adopting spatial cueing tasks derived from the paradigm proposed by Posner (1980). In these tasks, participants are initially presented with a task-irrelevant social cue, communicating a spatial direction (e.g., a pointing finger oriented rightwards), and appearing at a central position. After a variable stimulus onset asynchrony (SOA), they are asked to provide a motor response (e.g., a key press) to a target appearing in the periphery. Behavioural advantages (i.e., shorter reaction time and greater accuracy) are generally observed when the target appears at a spatial location indicated by the cue (i.e., a congruent trial) than when it appears at a different spatial location (i.e., an incongruent trial; see, e.g., McKay et al., 2021).

It is interesting to note that some studies have also reported attentional orienting in response to stimuli related to the feet, suggesting that our social attention system would be sensitive even to spatial signals coming from the lower part of the body of others. Most studies focused on the biological motion associated with feet and presented participants with highly impoverished stimuli consisting of moving light dots that mimic the movement of an individual (Bardi et al., 2015; Troje & Westhoff, 2006; Wang et al., 2014). For instance, in Wang et al. (2014), participants were engaged in a spatial cueing task in which two moving light dots, obtained by tracking the walking of a real individual with motion markers attached to his ankles, were used as cues, and the task consisted of localising (with a manual response) a peripheral target appearing 600 ms after cue onset. Evidence of a reliable orienting of attention emerged.

Other studies used more concrete and ecological foot-related stimuli, reporting mixed results. On the one hand, a recent study (Dalmaso, 2023) presented participants with footprint stimuli, which represent an indirect sign of the passage of another person within the environment. In this case, different types of footprint stimuli (i.e., human barefoot, shoes, and animal footprints) were used as central spatial cues, while participants were asked to discriminate (with a manual response) a peripheral target that appeared after a variable SOA (i.e., 200, 600, or 1000 ms). Evidence of a reliable orienting of attention emerged regardless of the type of cue, suggesting that humans are sensitive to footprints of different natures.

On the other hand, and of particular interest for the present work, another recent study (Chen et al., 2020) employed a spatial cueing task in which the picture of a human hand with the index finger pointing left or right, and the picture of a naked human foot oriented leftward or rightward, were used as spatial cues and presented in two distinct blocks of trials. After either 100 or 1000 ms, a peripheral target appeared that was to be localised with either a manual or foot response. A reliable orienting of attention emerged for the pointing finger but not for the foot stimulus, and this emerged regardless of the responding effector. Chen et al. (2020) concluded that the lack of an attentional orienting effect for the foot stimulus could reflect the fact that, in our everyday social interactions, we scarcely use our feet to convey a message of spatial nature, which is something that certainly cannot be said for finger-pointing stimuli. Furthermore, Chen et al. (2020) also suggested that the foot stimulus used in their study belonged to a stationary person (i.e., an individual standing upright) and was therefore not associated with a clear and strong pointing position. Indeed, the direction of the feet of person standing upright does not necessarily indicate that such an individual is paying attention in the same direction pointed her/his feet are pointed, or that, more generally, she/he is interested in such a spatial direction (i.e., during a face-to-face interaction, our feet can be directed towards a variety of different spatial locations without necessarily having a specific spatial meaning). On the other hand, the foot stimuli associated with an action-oriented posture (e.g., walking/running) would be a more direct and less ambiguous index of spatial pointing. Therefore, a reliable orienting response to foot stimuli could be expected for action-oriented foot stimuli. This possibility was tested here.

The current study, inspired by the work of Chen et al. (2020), represents a further attempt to reveal covert orienting of attention in response to the picture of a real foot. Participants were presented with the picture of a naked human foot oriented leftward or rightward. The foot cue belonged to a person in a neutral posture (i.e., standing upright), as in Chen et al. (2020), or in an action-oriented posture (i.e., walking/running). The hypothesis was straightforward: If it is true that a clear and strong pointing position associated with feet is a key element to elicit an attention-orienting response, a reliable orienting of attention was therefore expected for the foot stimulus associated with an action-oriented posture, but not for the foot stimulus associated with a neutral posture (in line with Chen et al., 2020).

The purpose of Experiment 1 was to extend the results reported by Chen et al. (2020). For this reason, the task was similar to that employed by Chen et al. (2020): Participants were presented with task-irrelevant central foot stimuli associated with a neutral or an action-oriented posture, while a target to be localised appeared in the periphery. To adhere as much as possible to the design adopted by Chen et al. (2020), the two types of stimuli (neutral versus action-oriented foot cues) were presented in two distinct blocks.

The rationale underlying Experiment 2 was based on research on social attention showing that a modulatory role in visual orienting of a social variable can be observed when different types of social stimuli are presented in an intermixed rather than blocked fashion (see, e.g., Dalmaso et al., 2020a; Pavan et al., 2011; Zhang et al., 2023). The general idea underlying this approach is that a given stimulus can be less or more relevant to participants if it is presented with a concurrent comparative stimulus (Carvalho & Goldstone, 2017). As for the social attention literature, for example, Zhang et al. (2023) reported a greater gaze-cueing effect—the covert orienting of attention elicited by averted-gaze faces—for faces belonging to White individuals (vs. Asian individuals), but only when these two types of faces were presented intermixed. Indeed, when the two types of faces were presented in two distinct blocks of trials, the magnitude of gaze-mediated orienting was not modulated by face type. The intermixed presentation would favour the activation of categorization processes based on a continuous comparison of different types of stimuli, while this comparison is, by definition, lacking when a blocked presentation is adopted. A similar rationale could also be applied to the present context. More precisely, the block design adopted in Experiment 1 suggested that our social attention system treated the two types of foot cues similarly. However, and in line with previous studies (Dalmaso et al., 2020a; Pavan et al., 2011; Zhang et al., 2023), an intermixed presentation of stimuli could increase the saliency of the action-oriented foot cue over the neutral foot cue, thus favouring the emergence of an orienting response for the former cue. Therefore, this possibility was examined in Experiment 2.

This work explored whether the picture of a human foot, oriented leftwards or rightwards, can elicit attentional orienting towards the same spatial location in an observer. In two experiments, participants were required to locate a peripheral target while neutral posture and action-oriented foot cues appeared in the centre of the screen in a blocked (Experiment 1) or intermixed (Experiment 2) fashion. The main result showed that the predicted congruency × cue interaction was significant only in Experiment 2, with subsequent comparisons indicating that reliable attention-orienting emerged for the action-oriented foot cue but not for the neutral foot cue. Overall, the results emerging from the present study provide an extension to what was reported by Chen et al. (2020) and, more generally, provide additional supporting evidence for the notion that foot-related stimuli can impact the social attention system (Dalmaso, 2023; Wang et al., 2014).

A consistent result between the two experiments was the presence of a significant congruency × SOA interaction, although none of the subsequent comparisons revealed a statistically significant attention-orienting response for foot stimuli. However, it should be noted with caution that the means seem to suggest an increase in the magnitude of the attention-orienting response effect with SOA. Even if the link between social orienting and SOA is still a matter of debate in the social attention literature, the available evidence indicates that social orienting would diminish at relatively long SOAs. This is what was observed in recent meta-analytic work on the gaze-cueing effect (McKay et al., 2021), according to which the magnitude of such an effect would be higher for relatively short SOAs (e.g., smaller than 200 ms until about 600–800 ms), then diminished at relatively longer SOAs (i.e., after about 800 ms). Nevertheless, it is important to keep in mind that eye-gaze stimuli are likely the most relevant spatial signals for humans, and evolution seems to have developed specific neurocognitive systems dedicated to eye-gaze processing and gaze-mediated orienting (Emery, 2000). Therefore, the high relevance of eye-gaze stimuli could explain the possible behavioural differences that could emerge between eye-gaze and other social signals that convey spatial direction. Additional studies are needed to address the temporal dynamics underlying the attention-orienting response to different types of social stimuli.

The results of Chen et al. (2020) and those reported here suggest that the orienting response elicited by foot stimuli is less robust and automatic compared to the orienting response elicited by cues that are more commonly used to guide others’ attention, such as eye-gaze or pointing gestures. In this regard, a recent theoretical framework called eyeTUNE (Dalmaso et al., 2020c) has been proposed to describe the possible social modulations of the gaze-cueing effect. According to this framework, an orienting response to eye-gaze stimuli would represent a default condition due to the great relevance of such stimuli for human beings. In fact, following the direction of another’s gaze could help identify a possible relevant stimulus in the environment, and therefore gaze-mediated orienting should appear by default. This would explain why reliable gaze-cueing effects have been documented for a variety of different eye-gaze stimuli, including simple (and socially impoverished) cues such as schematic eyes (Friesen & Kingstone, 1998) and at different levels of SOAs, including extremely very brief durations such as 14 ms (Hietanen & Leppänen, 2003). In sum, the picture emerging from Chen et al. (2020) and the present study suggests that at least two factors play a crucial role in the emergence of an attentional response to foot cues, which are (1) the presence of a cue with an evident action-oriented nature and (2) the adoption of a context (i.e., an intermixed presentation of stimuli) designed to incentivize a continuous comparison between neutral and action-oriented cues and favouring an increment in the saliency of the latter. Furthermore, the results provided by the analyses carried out on a sub-sample of trials of Experiment 2 revealed the presence of a peculiar trial-by-trial modulation, according to which the spatial nature of the n-1 trial influenced the performance on the n trial (see also, e.g., Ciardo et al., 2019). This modulation could be related to apparent motion, which likely appeared to participants as particularly evident when an action-oriented cue was preceded by a neutral cue pointing towards the opposite direction. However, given the explorative nature of these analyses, and the fact that the role of trial order is still a largely unexplored topic in social attention, future studies are needed to get a precise overview of such modulations in the present context.

One may wonder whether the experimental setting employed in Experiments 1 and 2 is suitable for detecting gaze-cueing of attention. Therefore, a follow-up experiment was conducted, which was similar to Experiments 1 and 2 but with averted-gaze faces instead of foot cues.⁵ A reliable gaze-cueing effect was found, thus confirming previous studies (e.g., Friesen & Kingstone, 1998; for other studies investigating gaze-mediated orienting through online tasks, see also, e.g., Dalmaso et al., 2021a, 2021b; Gregory, 2022; Villani et al., 2022). Furthermore, exploratory analyses compared the magnitude of the gaze-cueing effect with the magnitude of the orienting response elicited by the action-oriented foot cues in Experiment 3, but no differences emerged. This aligns with previous evidence showing that, at least at the behavioural level, gaze cues can elicit a similar orienting response as compared to other social cues, such as, for instance, pointing gestures (e.g., Cazzato et al., 2012; Dalmaso et al., 2015).

Future studies should further test the boundary conditions linking social attention and foot stimuli, for example, by employing different tasks and measures (e.g., eye movements; Dalmaso et al., 2020b; Kuhn & Kingstone, 2009) that are known to tap into attentional mechanisms from alternative perspectives (see also Hommel et al., 2019). In addition, even if in Chen et al. (2020) participants were immersed in an experimental context similar to that employed here, it is important to recall that in that work responses were provided with the hands or the feet on different blocks of trials. Since there is evidence showing that pairing different types of responses could activate different attentional mechanisms with pairs of responses within a trial (see, e.g., Taylor & Klein, 2000; Hilchey et al., 2012), another avenue for future work could be to manipulate response context (e.g., hand vs. foot responses) and the contexts concerning stimulus type and presentation (e.g., intermixed vs. blocked presentation of neutral vs. action-related foot cues).

The present work represents an additional piece of evidence showing that the modulatory role of a given variable becomes evident when an intermixed, rather than blocked, presentation of different stimuli is adopted, which is in line with some previous studies on the gaze-cueing effect (e.g., Dalmaso et al., 2020a; Pavan et al., 2011; Zhang et al., 2023). Additional, albeit indirect, evidence for this notion can also be found in studies examining spatial orientation in response to body pictures depicting individuals with neutral or action-oriented postures (Azarian et al., 2016, 2017). Indeed, in Azarian et al. (2016), evidence for an orienting response emerged for bodies (oriented leftwards or rightwards) with a posture communicating a threat (fear or anger), but not for bodies with a neutral posture. However, in a subsequent follow-up study, Azarian et al. (2017) found that bodies with a neutral posture can also elicit a reliable orienting response. The authors posited that this might be due to the fact that while in Azarian et al. (2016) all stimuli were presented intermixed (thus leading to the emergence of a comparative context which would have favored the processing of the threat stimuli over the neutral ones), in Azarian et al. (2017) only neutral stimuli were employed, thus avoiding any comparison with other types of stimuli. In sum, the picture emerging from social attention literature clearly indicates that the mechanisms governing the attentional response to different social stimuli appears to be influenced by the interplay between the specific nature of the stimuli and the context in which such stimuli are presented.