Two Trackers Are Better than One: Information about the Co-actor's Actions and Performance Scores Contribute to the Collective Benefit in a Joint Visuospatial Task

Strategic Task Decomposition in Joint Action

The core of human cooperation is people's ability to perform joint actions. Frequently, this requires effectively decomposing a joint task into individual subtasks, for example, when jointly shopping at the market to buy food. Surprisingly, little is known about how collaborators balance the costs of establishing a joint strategy for such decompositions and its expected benefits for a joint goal. We created a new online task that required pairs of randomly matched participants to jointly collect colored items. We then systematically varied the cognitive costs and benefits of applying a color-splitting strategy. The results showed that pairs adopted a color-splitting strategy more often when necessary to lower cognitive costs. However, once the strategy was jointly adopted, it continued to be used even when the cost-benefits changed. Our results provide first insights on how people decompose joint tasks into individual components and how decomposition strategies may evolve into conventions.

Cooperación en el rastreo visual de objetos: resultados preliminares

La cooperación tiene sus costos, pero también sus beneficios. En una tarea de rastreo visual de objetos con tres niveles de dificultad (4, 5 y 6 objetivos a rastrear respectivamente) se comparó el rendimiento de cada individuo en dos condiciones: rendimiento individual vs. rendimiento del individuo como parte de una díada. El propósito del estudio fue explorar si la diferencia en rendimiento entre estas dos condiciones se puede atribuir a la cooperación. Diez estudiantes universitarios respondieron un cuestionario de personalidad de auto informe (Big Five) y realizaron la tarea de rastreo visual de manera individual y como parte de una díada. Las díadas se grabaron en video para identificar las estrategias de comunicación, resolución de problemas y acuerdos en la selección de objetivos. Las díadas mostraron puntajes más altos y tiempos de reacción más lentos en comparación con el rendimiento individual. Los tiempos de reacción más lentos se asociaron positivamente con los intercambios verbales y las estrategias de división del trabajo en la díada. La amabilidad y la extraversión, medidas por el cuestionario de autoinforme, se asociaron positivamente con el acuerdo en la selección de objetivos. La interacción entre los miembros de una díada y sus rasgos de personalidad podrían ser relevantes para comprender la cooperación exitosa y sus costos asociados.

Effects of individual and dyadic decision-making and normative reference on delay discounting decisions

The tendency to devaluate delayed rewards, a phenomenon referred to as ‘discounting behaviour’, has been studied by wide-ranging research examining individuals choosing between sooner but smaller or later but larger rewards. Despite the fact that many real-life choices are embedded in a social context, the question of whether or not social collaboration can have an impact on such choices has not been addressed empirically. With this research, we aimed to fill this gap experimentally by implementing a novel choice selection procedure in order to study the interactive dynamics between two participants. This selection procedure allowed us to dissect the sequence of decision-making into its elements, starting from the very first individual preference to the solution of possible conflicting preferences in the dyad. In Experiment 1, we studied group decision-making on classical intertemporal choices to reveal the possible benefit of social collaboration on discounting and identified that the knowledge of the social situation in collective decision-making causes a reduction in discounting. In a pre-registered Experiment 2, we compared classical intertemporal choices with choices in a gamified version of a discounting paradigm in which the participants had a real-time experience trial by trial and for which a normative reference was present. We found that collective decision-making had a substantial impact on intertemporal decision-making, but was shaped by different types of choices. Classical intertemporal choices were rather susceptible to the contextual factors of decision-making, whereas in the gamified version that included a normative reference the decisions were reliably influenced by social collaboration and resulted in a lower discounting. The results in this paradigm replicate our original findings from former research.

Better together? Social distance affects joint probability discounting

Deciding together is common in our everyday life. However, the process of this joint decision-making plays out across different levels, for example language, intonation, or non-verbal behaviour. Here we focused on non-verbal interaction dynamics between two participants in probability discounting. We applied a gamified decision-making task in which participants performed a series of choices between a small but safe and a large but risky reward. In two experiments, we found that joint decision-making resulted in lower discounting and higher efficiency. In order to understand the underlying mechanisms in greater detail, we studied through which process this variation occurred and whether this process would be modulated by the social distance between both participants. Our findings suggested that socially close participants managed to reduce their discounting by interactive processes while socially distant participants were influenced by the social context itself. However, a higher level of efficiency was achieved through interactive processes for both groups. In summary, this study served as a fine-grained investigation of collaborative interaction processes and its significant impact on the outcome of choices with probabilistic consequences.

Coordination effort in joint action is reflected in pupil size

Humans often perform visual tasks together, and when doing so, they tend to devise division of labor strategies to share the load. Implementing such strategies, however, is effortful as co-actors need to coordinate their actions. We tested if pupil size - a physiological correlate of mental effort - can detect such a coordination effort in a multiple object tracking task (MOT). Participants performed the MOT task jointly with a computer partner and either devised a division of labor strategy (main experiment) or the labor division was already pre-determined (control experiment). We observed that pupil sizes increase relative to performing the MOT task alone in the main experiment while this is not the case in the control experiment. These findings suggest that pupil size can detect a rise in coordination effort, extending the view that pupil size indexes mental effort across a wide range of cognitively demanding tasks.

Humans share task load with a computer partner if (they believe that) it acts human-like

In the near future humans will increasingly be required to cooperate and share task load with artificial agents in joint tasks as they will be able to greatly assist humans in various types of tasks and contexts. In the present study, we investigated humans' willingness to share task load with a computer partner in a joint visuospatial task. The partner was described as either behaving in a human-like or machine-like way and followed a pre-defined behaviour that was either human-like or non-human-like. We found that participants successfully shared task load when the partner behaved in a human-like way. Critically, the successful collaboration was sustained throughout the experiment only when the partner was also described as behaving in a human-like way beforehand. These findings suggest that not only the behaviour of a computer partner but also the prior description of the partner is a critical factor influencing humans' willingness to share task load.

Dyadic and triadic search: Benefits, costs, and predictors of group performance

In daily life, humans often perform visual tasks, such as solving puzzles or searching for a friend in a crowd. Performing these visual searches jointly with a partner can be beneficial: The two task partners can devise effective division of labor strategies and thereby outperform individuals who search alone. To date, it is unknown whether these group benefits scale up to triads or whether the cost of coordinating with others offsets any potential benefit for group sizes above two. To address this question, we compare participants' performance in a visual search task that they perform either alone, in dyads, or in triads. When the search task is performed jointly, co-actors receive information about each other's gaze location. After controlling for speed-accuracy trade-offs, we found that triads searched faster than dyads, suggesting that group benefits do scale up to triads. Moreover, we found that the triads' divided the search space in accordance with the co-actors' individual search performances but searched less efficiently than dyads. We also present a linear model to predict group benefits, which accounts for 70% of the variance. The model includes our experimental factors and a set of non-redundant predictors, quantifying the similarities in the individual performances, the collaboration between co-actors, and the estimated benefits that co-actors would attain without collaborating. Overall, the present study demonstrates that group benefits scale up to larger group sizes, but the additional gains are attenuated by the increased costs associated with devising effective division of labor strategies.

Labor division in joint tasks: Humans maximize use of their individual attentional capacities

In daily life, humans frequently perform visuospatial tasks together (e.g., visual search) and distribute the labor in such tasks. Previous research has shown that humans prefer a left and right labor division in a joint multiple object tracking (MOT) task. Yet, findings from studies investigating individuals' tracking ability suggest attentional capacities may be more maximally used with a top and bottom labor division. We investigated whether co-actors' labor division preference is influenced by how they are seated (neighboring vs. opposite of each other) or how the MOT task is displayed (portrait vs. landscape). We find that pairs attain a higher performance using a top and bottom labor division and preferred this labor division compared to a left and right division. This preference was unaffected by the seating arrangement. For the landscape display, however, we find that participants no longer attain a higher performance for the top and bottom labor division and accordingly participants' preference for this labor division was greatly reduced as well. Overall, we propose that co-actors are sensitive to changes within their environment, which allows them to choose a labor division that maximizes use of their individual attentional capacities.

Neurophysiological correlates of collective perceptual decision-making

Humans frequently perform tasks collaboratively in daily life. Collaborating with others may or may not result in higher task performance than if one were to complete the task alone (i.e., a collective benefit). A recent study on collective benefits in perceptual decision-making showed that dyad members with similar individual performances attain collective benefit. However, little is known about the physiological basis of these results. Here, we replicate this earlier work and also investigate the neurophysiological correlates of decision-making using EEG. In a two-interval forced-choice task, co-actors individually indicated presence of a target stimulus with a higher contrast and then indicated their confidence on a rating scale. Viewing the individual ratings, dyads made a joint decision. Replicating earlier work, we found a positive correlation between the similarity of individual performances and collective benefit. We analyzed event-related potentials (ERPs) in three phases (i.e., stimulus onset, response and feedback) using explorative cluster mass permutation tests. At stimulus onset, ERPs were significantly linearly related to our manipulation of contrast differences, validating our manipulation of task difficulty. For individual and joint responses, we found a significant centro-parietal error-related positivity for correct versus incorrect responses, which suggests that accuracy is already evaluated at the response level. At feedback presentation, we found a significant late positive fronto-central potential elicited by incorrect joint responses. In sum, these results demonstrate that response- and feedback-related components elicited by an error-monitoring system differentially integrate conflicting information exchanged during the joint decision-making process.

The Social Situation Affects How We Process Feedback About Our Actions

Humans achieve their goals in joint action tasks either by cooperation or competition. In the present study, we investigated the neural processes underpinning error and monetary rewards processing in such cooperative and competitive situations. We used electroencephalography (EEG) and analyzed event-related potentials (ERPs) triggered by feedback in both social situations. 26 dyads performed a joint four-alternative forced choice (4AFC) visual task either cooperatively or competitively. At the end of each trial, participants received performance feedback about their individual and joint errors and accompanying monetary rewards. Furthermore, the outcome, i.e., resulting positive, negative, or neutral rewards, was dependent on the pay-off matrix, defining the social situation either as cooperative or competitive. We used linear mixed effects models to analyze the feedback-related-negativity (FRN) and used the Threshold-free cluster enhancement (TFCE) method to explore activations of all electrodes and times. We found main effects of the outcome and social situation, but no interaction at mid-line frontal electrodes. The FRN was more negative for losses than wins in both social situations. However, the FRN amplitudes differed between social situations. Moreover, we compared monetary with neutral outcomes in both social situations. Our exploratory TFCE analysis revealed that processing of feedback differs between cooperative and competitive situations at right temporo-parietal electrodes where the cooperative situation elicited more positive amplitudes. Further, the differences induced by the social situations were stronger in participants with higher scores on a perspective taking test. In sum, our results replicate previous studies about the FRN and extend them by comparing neurophysiological responses to positive and negative outcomes in a task that simultaneously engages two participants in competitive and cooperative situations.

Social modulation of object-directed but not image-directed actions

There has recently been an increased research focus on the influence of social factors on human cognition, attention, and action. While this represents an important step towards an ecologically valid description of real-world behaviour, this work has primarily examined dyads interacting with virtual stimuli i.e. on-screen images of objects. Though differences between actions to images and real items are known, their relative sensitivity to social factors is largely unknown. We argue that because images and real items elicit different neural representations, patterns of attention, and hand actions, a direct comparison between the magnitude of social effects while interacting with images and real objects is demanded. We examined patterns of reaching as individuals performed a shape-matching game. Images and real objects were used as stimuli, and social context was manipulated via the proximity of an observer. We found that social context interacted with stimulus type to modulate behaviour. Specifically, there was a delay in reaching for distant objects when a participant was facing another individual but this social effect only occurred when the stimuli were real objects. Our data suggest that even when images and real objects are arranged to share the affordance of reachability, they differ in their sensitivity to social influences. Therefore, the measurement of social effects using on-screen stimuli may poorly predict the social effects of actions directed towards real objects. Accordingly, generalizations between these two domains should be treated with caution.

Is Your Color My Color? Dividing the Labor of the Stroop Task Between Co-actors

Performing a task with other actors involves two opposing forces, division of labor between co-acting individuals and integration of divided parts of the task into a shared mental representation (co-representation). Previous studies have focused primarily on the integration of task representations and limited attention has paid to the division of labor. The present study devised a test of the integration and the division in a joint task setting. A joint version of the Stroop task was developed, in which pairs of actors were assigned different sets of target colors. If the actors integrate their co-actor’s task, the colors assigned to their co-actor should be represented as if they were the actor’s own target colors; the Stroop effect should be as large when distractor color words denote their co-actor’s target colors as when these words denote the actor’s own target colors. If the actors divide the labor of the Stroop task, the colors assigned to their partner should be represented as non-target colors; the Stroop effect should be smaller when the distractor color words denote the co-actor’s target colors than when these words denote the actor’s own target colors. The results of response time did not provide clear support for either position, while those of response accuracy supported the division of labor. Possible cognitive mechanisms that support the division of labor and the integration of task representation are discussed.

The Influence of Co-action on a Simple Attention Task: A Shift Back to the Status Quo

There is a growing consensus among researchers that a complete description of human attention and action should include information about how these processes are informed by social context. When we actively engage in co-action with others, there are characteristic changes in action kinematics, reaction time, search behavior, as well as other processes (see Sebanz et al., 2003; Becchio et al., 2010; Wahn et al., 2017). It is now important to identify precisely what is shared between co-actors in these joint action situations. One group recently found that participants seem to withdraw their attention away from a partner and toward themselves when co-engaged in a line bisection judgment task (Szpak et al., 2016). This effect runs counter to the typical finding that attention is drawn toward social items in the environment (Birmingham et al., 2008, 2009; Foulsham et al., 2011). As such, the result suggests that joint action can uniquely lead to the withdrawal of covert attention in a manner detectable by a line bisection task performed on a computer screen. This task could therefore act as a simple and elegant measure of interpersonal effects on attention within particular pairs of participants. For this reason, the present work attempted to replicate and extend the finding that attention, as measured by a line-bisection task, is withdrawn away from nearby co-actors. Overall our study found no evidence of social modulation of covert attention. This suggests that the line bisection task may not be sensitive enough to reliably measure interpersonal attention effects – at least when one looks at overall group performance. However, our data also hint at the possibility that the effect of nearby others on the distribution of attention may be modulated by individual differences.

Let's Move It Together: A Review of Group Benefits in Joint Object Control

In daily life, humans frequently engage in object-directed joint actions, be it carrying a table together or jointly pulling a rope. When two or more individuals control an object together, they may distribute control by performing complementary actions, e.g., when two people hold a table at opposite ends. Alternatively, several individuals may execute control in a redundant manner by performing the same actions, e.g., when jointly pulling a rope in the same direction. Previous research has investigated whether dyads can outperform individuals in tasks where control is either distributed or redundant. The aim of the present review is to integrate findings for these two types of joint control to determine common principles and explain differing results. In sum, we find that when control is distributed, individuals tend to outperform dyads or attain similar performance levels. For redundant control, conversely, dyads have been shown to outperform individuals. We suggest that these differences can be explained by the possibility to freely divide control: Having the option to exercise control redundantly allows co-actors to coordinate individual contributions in line with individual capabilities, enabling them to maximize the benefit of the available skills in the group. In contrast, this freedom to adopt and adapt customized coordination strategies is not available when the distribution of control is determined from the outset.

Group benefits in joint perceptual tasks-a review

In daily life, humans often perform perceptual tasks together to reach a shared goal. In these situations, individuals may collaborate (e.g., by distributing task demands) to perform the task better than when the task is performed alone (i.e., attain a group benefit). In this review, we identify the factors influencing if, and to what extent, a group benefit is attained and provide a framework of measures to assess group benefits in perceptual tasks. In particular, we integrate findings from two frequently investigated joint perceptual tasks: visuospatial tasks and decision-making tasks. For both task types, we find that an exchange of information between coactors is critical to improve joint performance. Yet, the type of exchanged information and how coactors collaborate differs between tasks. In visuospatial tasks, coactors exchange information about the performed actions to distribute task demands. In perceptual decision-making tasks, coactors exchange their confidence on their individual perceptual judgments to negotiate a joint decision. We argue that these differences can be explained by the task structure: coactors distribute task demands if a joint task allows for a spatial division and stimuli can be accurately processed by one individual. Otherwise, they perform the task individually and then integrate their individual judgments.

Performance similarities predict collective benefits in dyadic and triadic joint visual search

When humans perform tasks together, they may reach a higher performance in comparison to the best member of a group (i.e., a collective benefit). Earlier research showed that interindividual performance similarities predict collective benefits for several joint tasks. Yet, researchers did not test whether this is the case for joint visuospatial tasks. Also, researchers did not investigate whether dyads and triads reach a collective benefit when they are forbidden to exchange any information while performing a visuospatial task. In this study, participants performed a joint visual search task either alone, in dyads, or in triads, and were not allowed to exchange any information while doing the task. We found that dyads reached a collective benefit. Triads did outperform their best individual member and dyads—yet, they did not outperform the best dyad pairing within the triad. In addition, similarities in performance significantly predicted the collective benefit for dyads and triads. Furthermore, we find that the dyads’ and triads’ search performances closely match a simulated performance based on the individual search performances, which assumed that members of a group act independently. Overall, the present study supports the view that performance similarities predict collective benefits in joint tasks. Moreover, it provides a basis for future studies to investigate the benefits of exchanging information between co-actors in joint visual search tasks.

Can Limitations of Visuospatial Attention Be Circumvented? A Review

In daily life, humans are bombarded with visual input. Yet, their attentional capacities for processing this input are severely limited. Several studies have investigated factors that influence these attentional limitations and have identified methods to circumvent them. Here, we provide a review of these findings. We first review studies that have demonstrated limitations of visuospatial attention and investigated physiological correlates of these limitations. We then review studies in multisensory research that have explored whether limitations in visuospatial attention can be circumvented by distributing information processing across several sensory modalities. Finally, we discuss research from the field of joint action that has investigated how limitations of visuospatial attention can be circumvented by distributing task demands across people and providing them with multisensory input. We conclude that limitations of visuospatial attention can be circumvented by distributing attentional processing across sensory modalities when tasks involve spatial as well as object-based attentional processing. However, if only spatial attentional processing is required, limitations of visuospatial attention cannot be circumvented by distributing attentional processing. These findings from multisensory research are applicable to visuospatial tasks that are performed jointly by two individuals. That is, in a joint visuospatial task requiring object-based as well as spatial attentional processing, joint performance is facilitated when task demands are distributed across sensory modalities. Future research could further investigate how applying findings from multisensory research to joint action research may facilitate joint performance. Generally, findings are applicable to real-world scenarios such as aviation or car-driving to circumvent limitations of visuospatial attention.