When 1 + 1 = 1: The unification of independent actors revealed through joint Simon effects in crossed and uncrossed effector conditions

Abstract

The “Simon effect” describes a pattern of reaction times (RTs) where responses to symbolic information are shorter when the information is presented on the same side of space as the desired response than when it is on the opposite side of space. For example, if right hand responses are required for green targets and left hand responses for red targets, RTs with the right hand are shorter when the green target appears on the right side than on the left side. It has been reported that Simon effects also appear when two individuals perform independent components of a Simon effect task. It has been suggested that such joint Simon effects occur because participants represent the action of their partner. It is unclear, however, if the joint Simon effect emerges because: (1) each partner represents the other’s action; (2) each partner is using the other person or their response as an environmental reference; or (3) an intra-hemispheric processing advantage due to the lateralized cerebral organization of perceptual and motor systems. The present study distinguished between these possibilities by asking pairs of participants to perform in conditions in which they crossed their arms into the other person’s space. Consistent with within-person Simon effects, joint Simon effects were observed in uncrossed- and crossed limb conditions. These results support a response co-representation explanation of joint Simon effects. It is suggested that the processes underlying the evoked representations have developed to allow two independent agents to form temporary synergies to facilitate efficient task completion.

Introduction

Recent research has revealed that many of the same behavioral phenomena that are observed when an individual performs a task on their own are also observed when two people perform separate, but related, tasks in the same environment. For example, Welsh et al. (2005) and Welsh et al. (2007b) have observed that participants are slower at responding to a target that appears in the same place as a previous target when someone else had recently engaged that target – a between-person inhibition of return effect. In another series of studies, Sebanz, Knoblich, and Prinz (2003, 2005a) discovered what was called a joint Simon effect. As in a conventional Simon effect task, participants in the Sebanz et al. studies responded by pressing either a left or right button when a red or green target was presented, respectively. Targets were presented as rings on the picture of an index finger on a hand that pointed to the left, to the right, or centrally. Thus, the relevant feature (target color) was presented with an irrelevant feature (direction of the pointing finger). Consistent with previous Simon effect experiments, when participants performed the task on their own and were required to complete both the left and right response (two-choice task), reaction times (RTs) were shortest when the relevant and irrelevant features overlapped (e.g., a red target on a leftward pointing finger) and were longest when the features did not overlap (e.g., a red target on a rightward pointing finger). The novel finding of the Sebanz et al. (2003) work was that this spatial compatibility effect was also observed when individual participants were asked to respond to only one of the targets (a simple go-nogo task), but were sitting next to a partner who was asked to only respond to the other target – a joint Simon effect.

It has been suggested that between-person information processing effects, such as the joint Simon effect, are elicited because the observation or the knowledge of another person’s response activates a representation of that response in the observer and that this observation-evoked response subsequently activates the same set of processes in the observer as an internally-generated response representation would activate – attention- or response-based inhibition in the case of the between-person IOR effect (Welsh et al., 2005) or response conflict in the case of the joint Simon effect (Sebanz et al., 2003). In support of the notion that response co-representation is responsible for these effects, the Simon effect that is detected when two independent responding agents are sitting next to each other is no longer observed when only one of the individuals performs the same go-nogo task on their own (i.e., without a partner responding to the other relevant stimulus feature – Sebanz et al., 2003, Tsai et al., 2008, Tsai et al., 2006, Welsh et al., 2007a). The suggested reason for the absence of a Simon effect in the individual go-nogo task is that there is only one preactived response code in the performer because that individual is performing the task without another acting intentional agent. In the absence of another responding agent, there is no preactivation of an alternate response code and, subsequently, no need for time-consuming response conflict resolution and inhibitory processing (Sebanz et al., 2003, Welsh et al., 2007a; see Tsai et al., 2006, Tsai et al., 2008 for ERP evidence in favor of this interpretation). Further, the critical role that the observation of a response from an intentional agent plays in the activation of the observation-evoked response code is supported by the findings that joint Simon effects are not present when individuals witness the movement of a non-intentional object (Tsai & Brass, 2007).

Although there is growing evidence in favor of a response co-representation interpretation of the joint Simon effect, it could be that the joint Simon effect emerges because one partner is using the presence of another intentional and responding agent as a spatial reference point and that it is the spatial relation between people that is enhancing the saliency of the spatial location of the target in the joint conditions. Specifically, in each previous experiment (Sebanz et al., 2003, Tsai and Brass, 2007, Welsh et al., 2007a), the responding individuals always completed their responses on a button in front of their body (even if vision of the actual response was prevented by a box covering the responding hand; see Sebanz et al., 2003). As such, the individual on the right could just have coded themselves as “right” and the other person as “left” and, as a result, increased the saliency of the irrelevant location component of the stimulus information. The increased saliency of ipsilateral target locations could have ultimately lead to the “appearance” of a joint Simon effect.¹

The purpose of the present study was to test this alternative explanation by asking participants to complete a joint Simon effect task in uncrossed and crossed limb conditions (Figs. 1c–f). One of the strongest pieces of evidence in favor of a response conflict explanation of the conventional within-person (two-choice) Simon effect (Kronblum, Hasbroucq, & Osman, 1990) over an anatomical explanation is the finding that the Simon effect is tied to the side of space of the response and not the anatomical origin of the effector that is responsible for the response. To elucidate, in the conventional uncrossed choice RT condition when the right hand is in right space and the left limb is in left space (Fig. 1a), RTs are shorter when the assigned target (e.g., a green square) is presented in right space than when it is in left space. It is argued that the Simon effect emerges because both the relevant (color) and irrelevant (location) stimulus features activate relatively independent response codes. On compatible trials in which the relevant and irrelevant features active the same response codes, the response selection process is relatively efficient leading to short RTs. In contrast, on incompatible trials in which the relevant and irrelevant features activate different response codes, the two opposing response codes lead to conflict and the response selection process is relatively inefficient leading to longer RTs. Under these uncrossed conditions, however, it is also possible that the shorter RTs for ipsilateral targets could occur because of the relative efficiencies of intra- versus interhemispheric processing (e.g., Poffenberger, 1912). Specifically, the left hemisphere receives visual information from the right side of space and controls the right hand and the right hemisphere receives visual information from the left side of space and controls the left hand. Hence, “spatial compatibility” effects might arise because the processing of the visual information and the response is maintained within a single hemisphere on compatible trials, whereas the processing of the visual information and the response must involve both hemispheres and cross the corpus callosum on incompatible trials. Because intra-hemispheric processing is likely more efficient than inter-hemisphere processing (due to the involvement of the corpus callosum at minimum), the Simon effect might emerge because of this relative processing efficiency advantage.

To address this issue, studies have been conducted in which participants cross their limbs such that the right hand is responding on the left side and the left limb is on the right side (see Fig. 1b). If the Simon effect was due to intra- versus interhemispheric processing efficiencies alone, then RTs for the right hand would still be shorter for targets presented in right space than for targets in left space, even though the right hand was responding in left space. Consistent with the response conflict account, however, RTs are shorter when the assigned target is presented on the same side as the responding effector than when it is presented on the same side as the origin of the effector (Brebner et al., 1972, Simon et al., 1970, Wallace, 1971). Thus, it is concluded that it is the irrelevant response codes evoked by the presentation of the lateralized targets, and not the lateralization of perceptual and motor processes, that are responsible for the information processing inefficiencies that lead to the Simon effect.

The uncrossed/crossed limb manipulation provides a simple method for disentangling response co-representation from the mapping of personal location for the same reasons that it addressed similar issues in the two-choice tasks. Thus, participants in the present study completed the two-choice and joint Simon effect task in uncrossed and crossed arm conditions. In the joint uncrossed conditions, participants responded with a button that was directly in front of them (Figs. 1c and e). In the joint crossed conditions, participants responded with a button that was directly in front of their partner (Figs. 1d and f). The uncrossed conditions were similar to those used in previous studies of the joint Simon effect (Sebanz et al., 2003, Welsh et al., 2007a) and, as such, it was anticipated that RTs for compatible targets (e.g., green square presented in right space for the person sitting and responding on the right) would be shorter than those for incompatible targets (e.g., green square presented in left space for the person sitting and responding on the right).

The key conditions in the present study were the crossed limb conditions. If the joint Simon effect is based on the spatial coding of personal location, then RTs for targets presented on the same side of space as the responding individual (e.g., green square presented in right space for the person sitting on the right but responding on the left) would continue to be shorter than those presented in front of their partner (e.g., green square presented in left space for the person sitting on the right but responding on the left). The joint Simon effect would remain with the location of the individual because individuals would still be coding themselves as “right” or “left” and maintaining the salience of stimuli presented on their side of space. In contrast, if the joint Simon effect is caused by response co-representation, then RTs for targets presented on the same side of space as the effector (e.g., green square presented in left space for the person sitting on the right but responding on the left) would be shorter than those presented in front of themselves (e.g., green square presented in right space for the person sitting on the right but responding on the left). The joint Simon effect should move with the effector because the relevant and irrelevant features should evoke similar or competing response codes as they do in the conventional two-choice tasks.

In addition to testing the role of the relative location of the person and response in the joint Simon effect, the potential influence that the actual responding limb plays in the joint Simon effect was also examined. Although not explicitly stated in the methods of Sebanz et al. (2003), the participants in Welsh et al. (2007a) used the hand on the same side of space as they were sitting to respond in the joint Simon effect task – the person on the left responded with the left hand and the person on the right responded with the right hand (see also,Tsai et al., 2006, Tsai et al., 2008, Tsai and Brass, 2007). Thus, it is also possible that the joint Simon effect was driven by hand-space compatibility or some mediation of the intra- versus interhemispheric processing advantage. For example, for the person sitting on the right, the right (outside) hand might have had shorter RTs for targets presented in right space because of a special perception–action (compatibility) relationship or privileged access to the information processed within the same hemispheric system. To determine if the hand used plays a role in the joint Simon effect, participants completed the joint tasks twice – once with their “inside” hands (Figs. 1c and d) and once with their “outside” hands (Figs. 1e and f). If the hand used contributes to the joint Simon effect, then joint Simon effects should diminish or even reverse in the “inside” hand conditions because the responses with these hands may be more compatible with, or have a more direct access to, the information presented in the opposite side of space. However, if the joint Simon effect results from response co-representation and response conflict, then the hand used should not influence the direction of the Simon effect because each individual will be coding their response as “right” or “left” regardless of the actual hand that was used.