Spontaneous Theory of Mind is reduced for nonhuman-like agents as compared to human-like agents

Evidence against implicit belief processing in a blindfold task

Understanding what other people think is crucial to our everyday interactions. We seem to be affected by the perspective of others even in situations where it is irrelevant to us. This intrusion from others' perspectives has been referred to as altercentric bias and has been suggested to reflect implicit belief processing. There is an ongoing debate about how robust such altercentric effects are and whether they indeed reflect true mentalizing or result from simpler, domain-general processes. As a critical test for true mentalizing, the blindfold manipulation has been proposed. That is, participants are familiarized with a blindfold that is either transparent or opaque. When they then observe a person wearing this blindfold, they can only infer what this person can or cannot see based on their knowledge of the blindfold's transparency. Here, we used this blindfold manipulation to test whether participants' reaction times in detecting an object depended on the agent's belief about the object's location, itself manipulated with a blindfold. As a second task, we asked participants to detect where the agent was going to look for the object. Across two experiments with a large participant pool (N = 234) and different settings (online/lab), we found evidence against altercentric biases in participants' response times in detecting the object. We did, however, replicate a well-documented reality congruency effect. When asked to detect the agent's action, in turn, participants were biased by their own knowledge of where the object should be, in line with egocentric biases previously found in false belief reasoning. These findings suggests that altercentric biases do not reflect belief processing but lower-level processes, or alternatively, that implicit belief processing does not occur when the belief needs to be inferred from one's own experience.

Human-like interactions prompt people to take a robot's perspective

The increasing need for human-robot interaction requires not only robots to understand how humans think, but also humans to understand robots. Interestingly, little attention has been given to how humans interpret robots' behaviors. In this study, we adopted a social mental rotation task and investigated whether socially engaging behaviors could influence how people take a robot's perspectives. In a real lab, two android robots with neutral appearance sat opposite each other by a table with conflicting perspectives. Before the participant started the experiment, one of the robots behaved more interactively than the other by showing more socially engaging behaviors. Then the participant was required to identify rotated normal or mirrored digits presented inbetween the two robots. Results revealed a significant interactive effect between the digits type (normal; mirrored) and robot type (interactive; noninteractive). When digits were oriented to the interactive robot, we found a larger RT difference between normal and mirrored digits. In general, these findings suggested that robots' interactive behaviors could influence how people spontaneously consider the robot's perspective. Future studies may further consider how interactive behaviors can shape human-robot relationships and facilitate human-robot interaction.

Left and right temporal-parietal junctions (TPJs) as "match/mismatch" hedonic machines: A unifying account of TPJ function

Experimental and theoretical studies have tried to gain insights into the involvement of the Temporal Parietal Junction (TPJ) in a broad range of cognitive functions like memory, attention, language, self-agency and theory of mind. Recent investigations have demonstrated the partition of the TPJ in discrete subsectors. Nonetheless, whether these subsectors play different roles or implement an overarching function remains debated. Here, based on a review of available evidence, we propose that the left TPJ codes both matches and mismatches between expected and actual sensory, motor, or cognitive events while the right TPJ codes mismatches. These operations help keeping track of statistical contingencies in personal, environmental, and conceptual space. We show that this hypothesis can account for the participation of the TPJ in disparate cognitive functions, including "humour", and explain: a) the higher incidence of spatial neglect in right brain damage; b) the different emotional reactions that follow left and right brain damage; c) the hemispheric lateralisation of optimistic bias mechanisms; d) the lateralisation of mechanisms that regulate routine and novelty behaviours. We propose that match and mismatch operations are aimed at approximating "free energy", in terms of the free energy principle of decision-making. By approximating "free energy", the match/mismatch TPJ system supports both information seeking to update one's own beliefs and the pleasure of being right in one's own' current choices. This renewed view of the TPJ has relevant clinical implications because the misfunctioning of TPJ-related "match" and "mismatch" circuits in unilateral brain damage can produce low-dimensional deficits of active-inference and predictive coding that can be associated with different neuropsychological disorders.

Closeness impeded self-perspective inhibition whereas facilitated explicit perspective calculation

This study examined whether and how closeness affected the calculation and selection processes underlying perspective taking. Using the visual dot perspective taking task, we introduced a close friend and a stranger from the participants' university as the perspective-taking targets. Friend and stranger trials were mixed in a block in Experiment 1 but separated in different blocks in Experiment 2. Results revealed a significant effect of closeness on egocentric but not altercentric interference. The analyses on other-consistent and other-inconsistent trials suggested that closeness impeded responding from the avatar's perspective when self and other perspectives differed but facilitated responding from the avatar's perspective when self and other perspectives were consistent. However, the analyses on self-consistent and self-inconsistent trials revealed that the processing cost induced by implicit perspective calculation and other-inhibition was comparable between friends and strangers. These suggested that closeness selectively impeded self-perspective inhibition whereas facilitated explicit perspective calculation.

Knowledge before Belief

Research on the capacity to understand others' minds has tended to focus on representations of beliefs, which are widely taken to be among the most central and basic theory of mind representations. Representations of knowledge, by contrast, have received comparatively little attention and have often been understood as depending on prior representations of belief. After all, how could one represent someone as knowing something if one doesn't even represent them as believing it? Drawing on a wide range of methods across cognitive science, we ask whether belief or knowledge is the more basic kind of representation. The evidence indicates that non-human primates attribute knowledge but not belief, that knowledge representations arise earlier in human development than belief representations, that the capacity to represent knowledge may remain intact in patient populations even when belief representation is disrupted, that knowledge (but not belief) attributions are likely automatic, and that explicit knowledge attributions are made more quickly than equivalent belief attributions. Critically, the theory of mind representations uncovered by these various methods exhibit a set of signature features clearly indicative of knowledge: they are not modality-specific, they are factive, they are not just true belief, and they allow for representations of egocentric ignorance. We argue that these signature features elucidate the primary function of knowledge representation: facilitating learning from others about the external world. This suggests a new way of understanding theory of mind-one that is focused on understanding others' minds in relation to the actual world, rather than independent from it.

Visibly constraining an agent modulates observers' automatic false-belief tracking

Our motor system can generate representations which carry information about the goals of another agent’s actions. However, it is not known whether motor representations play a deeper role in social understanding, and, in particular, whether they enable tracking others’ beliefs. Here we show that, for adult observers, reliably manifesting an ability to track another’s false belief critically depends on representing the agent’s potential actions motorically. One signature of motor representations is that they can be disrupted by constraints on an observed agent’s action capacities. We therefore used a ‘mummification’ technique to manipulate whether the agent in a visual ball-detection task was free to act or whether he was visibly constrained from acting. Adults’ reaction times reliably reflected the agent’s beliefs only when the agent was free to act on the ball and not when the agent was visibly constrained from acting. Furthermore, it was the agent’s constrained action capabilities, rather than any perceptual novelty, that determined whether adult observers’ reaction times reliably reflected the agent’s beliefs. These findings signal that our motor system may underpin more of social cognition than previously imagined, and, in particular, that motor representations may underpin automatic false-belief tracking.