Reduced egocentric bias when perspective-taking compared with working from rules.

Previous research has suggested that adults are sometimes egocentric, erroneously attributing their current beliefs, perspectives, and opinions to others. Interestingly, this egocentricity is sometimes stronger when perspective-taking than when working from functionally identical but non-perspectival rules. Much of our knowledge of egocentric bias comes from Level 1 perspective-taking (e.g., judging whether something is seen) and judgements made about narrated characters or avatars rather than truly social stimuli such as another person in the same room. We tested whether adults would be egocentric on a Level 2 perspective-taking task (judging how something appears), in which they were instructed to indicate on a continuous colour scale the colour of an object as seen through a filter. In our first experiment, we manipulated the participants' knowledge of the object's true colour. We also asked participants to judge either what the filtered colour looked like to themselves or to another person present in the room. We found participants' judgements did not vary across conditions. In a second experiment, we instead manipulated how much participants knew about the object's colour when it was filtered. We found that participants were biased towards the true colour of the object when making judgements about targets they could not see relative to targets they could, but that this bias disappeared when the instruction was to imagine what the object looked like to another person. We interpret these findings as indicative of reduced egocentricity when considering other people's experiences of events relative to considering functionally identical but abstract rules.

Where was I? Taking alternative visual perspectives can make us (briefly) misplace our own.

How do we imagine what the world looks like from another visual perspective? The two most common proposals-embodiment and array rotation-imply that we must briefly imagine either movement of the self (embodiment) or movement of the scene (array rotation). What is not clear is what this process might mean for our real, egocentric perspective of the world. We present a novel task in which participants had to locate a target from an alternative perspective but make a manual response consistent with their own. We found that when errors occurred they were usually manual responses that would have been correct from the computed alternative perspective. This was the case both when participants were instructed to find the target from another perspective and when they were asked to imagine the scene itself rotated. We interpret this as direct evidence that perspective-taking leads to the brief adoption of a computed perspective-a new imagined relationship between ourselves and the scene-to the detriment of our own, egocentric point of view.

Mirror self-recognition and its relationship to social cognition in chimpanzees.

Chimpanzees and humans are capable of recognizing their own reflection in mirrors. Little is understood about the selective pressures that led to this evolved trait and about the mechanisms that underlie it. Here, we investigated the hypothesis that mirror self-recognition in chimpanzees is the byproduct of a developed form of self-awareness that was naturally selected for its adaptive use in social cognitive behaviors. We present here the first direct attempt to assess the social cognition hypothesis by analyzing the association between mirror self-recognition in chimpanzees, as measured by a mirror-mark test, and their performance on a variety of social cognition tests. Consistent with the social cognition hypothesis, chimpanzees who showed evidence of mirror self-recognition in the mark test tended to perform significantly better on the social cognition tasks than those who failed the mark test. Additionally, the data as a whole fit the social cognition hypothesis better than the main competing hypothesis of mirror self-recognition in great apes, the secondary representation hypothesis. Our findings strongly suggest that the evolutionary origins of great apes' and humans' capacity to understand ourselves, as revealed by our capacity to recognize ourselves in mirrors, are intimately linked to our ability to understand others.

Is Language Required to Represent Others' Mental States? Evidence From Beliefs and Other Representations.

An important part of our Theory of Mind-the ability to reason about other people's unobservable mental states-is the ability to attribute false beliefs to others. We investigated whether processing these false beliefs, as well as similar but nonmental representations, is reliant on language. Participants watched videos in which a protagonist hides a gift and either takes a photo of it or writes a text about its location before a second person inadvertently moves the present to a different location, thereby rendering the belief and either the photo or text false. At the same time, participants performed either a concurrent verbal interference task (rehearsing strings of digits) or a visual interference task (remembering a visual pattern). Results showed that performance on false belief trials did not decline under verbal interference relative to visual interference. We interpret these findings as further support for the view that language does not form an essential part of the process of reasoning online ("in the moment") about false beliefs.

The unreliability of egocentric bias across self-other and memory-belief distinctions in the Sandbox Task.

Humans are often considered egocentric creatures, particularly (and ironically) when we are supposed to take another person's perspective over our own (i.e. when we use our theory of mind). We investigated the underlying causes of this phenomenon. We gave young adult participants a false belief task (Sandbox Task) in which objects were first hidden at one location by a protagonist and then moved to a second location within the same space but in the protagonist's absence. Participants were asked to indicate either where the protagonist remembered the item to be (reasoning about another's memory), believed it to be (reasoning about another's false belief), or where the protagonist would look for it (action prediction of another based on false belief). The distance away from Location A (the original one) towards Location B (the new location) was our measure of egocentric bias. We found no evidence that egocentric bias varied according to reasoning type, and no evidence that participants actually were more biased when reasoning about another person than when simply recalling the first location from memory. We conclude that the Sandbox Task paradigm may not be sensitive enough to draw out consistent effects related to mental state reasoning in young adults.

Egocentric bias across mental and non-mental representations in the Sandbox Task.

In the Sandbox Task, participants indicate where a protagonist who has a false belief about the location of an object will look for that object in a trough filled with a substrate that conceals the hidden object's location. Previous findings that participants tend to indicate a location closer to where they themselves know the object to be located have been interpreted as evidence of egocentric bias when attributing mental states to others. We tested the assumption that such biases occur as a result of reasoning about mental states specifically. We found that participants showed more egocentric bias when reasoning from a protagonist's false belief than from their own memory, but found equivalent levels of bias when they were asked to indicate where a false film would depict the object as when they were asked about a protagonist's false belief. Our findings suggest that that egocentric biases found in adult false belief tasks are more likely due to a general difficulty with reasoning about false representations than a specialised difficulty with reasoning about false mental states.

Chimpanzees gesture to humans in mirrors: using reflection to dissociate seeing from line of gaze.

There is much experimental evidence suggesting that chimpanzees understand that others see. However, previous research has never experimentally ruled out the alternative explanation that chimpanzees are just responding to the geometric cue of 'direct line of gaze', the observable correlate of seeing in others. Here, we sought to resolve this ambiguity by dissociating seeing from direct line of gaze using a mirror. We investigated the frequency of chimpanzees' visual gestures towards a human experimenter who could see them (as a result of looking into a mirror) but who lacked a direct line of gaze to them (as a result of having his/her head turned away). Chimpanzees produced significantly more visual gestures when the experimenter could see them than when he/she could not, even when the experimenter did not have a direct line of gaze to them. Results suggest that chimpanzees, through a possible process of experience projection based on their own prior experience with mirrors, infer that an experimenter looking at the mirror can see them. We discuss our results in relation to the theory of mind hypothesis that chimpanzees understand seeing in others, and we evaluate possible alternative low-level explanations.

Error rate on the director's task is influenced by the need to take another's perspective but not the type of perspective.

Adults are prone to responding erroneously to another's instructions based on what they themselves see and not what the other person sees. Previous studies have indicated that in instruction-following tasks participants make more errors when required to infer another's perspective than when following a rule. These inference-induced errors may occur because the inference process itself is error-prone or because they are a side effect of the inference process. Crucially, if the inference process is error-prone, then higher error rates should be found when the perspective to be inferred is more complex. Here, we found that participants were no more error-prone when they had to judge how an item appeared (Level 2 perspective-taking) than when they had to judge whether an item could or could not be seen (Level 1 perspective-taking). However, participants were more error-prone in the perspective-taking variants of the task than in a version that only required them to follow a rule. These results suggest that having to represent another's perspective induces errors when following their instructions but that error rates are not directly linked to errors in inferring another's perspective.

I know you see it wrong! Children use others' false perceptions to predict their behaviors.

Research on children's ability to attribute false mental states to others has focused exclusively on false beliefs. We developed a novel paradigm that focuses instead on another type of false mental state: false perceptions. From approximately 4years of age, children begin to recognize that their perception of an illusory object can be at odds with its true properties. Our question was whether they also recognize that another individual viewing the object will similarly experience a false perception. We tested 33 preschool children with a task in which distorting lenses caused a small object to appear large and a large object to appear small. To succeed, children needed to recognize that a naive agent would falsely perceive the relative size of the objects and to correctly anticipate the agent's actions on that basis. Children performed significantly better than chance in our false perception test, and there was a developmental progression in performance from 4 to 5years of age similar to that seen in standard false belief tests. Our findings demonstrate that preschool children are capable of understanding that other individuals will be perceptually misled by illusory objects and that these false perceptions will influence their actions in predictable ways.

Smoke and mirrors: Testing the scope of chimpanzees' appearance-reality understanding.

The ability to make appearance-reality (AR) discriminations is an important higher-order cognitive adaptation in humans but is still poorly understood in our closest primate relatives. Previous research showed that chimpanzees are capable of AR discrimination when choosing between food items that appear, due to the effects of distorting lenses, to be smaller or larger than they actually are (Krachun, Call, & Tomasello, 2009). In the current study, we investigated the scope and flexibility of chimpanzees' AR discrimination abilities by presenting them with a wider range of illusory stimuli. In addition to using lenses to change the apparent size of food items (Experiment 1), we used a mirror to change the apparent number of items (Experiment 2), and tinted filters to change their apparent color (Experiment 3). In all three experiments, some chimpanzees were able to maximize their food rewards by making a choice based on the real properties of the stimuli in contrast to their manifest apparent properties. These results replicate the earlier findings for size illusions and extend them to additional situations involving illusory number and color. Control tests, together with findings from previous studies, ruled out lower-level explanations for the chimpanzees' performance. The findings thus support the hypothesis that chimpanzees are capable of making AR discriminations with a range of illusory stimuli.

How to solve the distinguishability problem: Triangulation without explicit training.

Heyes's (1998) triangulation approach to distinguishing a "theory" of mind (ToM) from a "theory" of behavior (ToB) in chimpanzees fails. The ToB theorist can appeal to the explicit training sessions and analogical reasoning to explain/predict the chimpanzees' behaviors. An alternative triangulation experiment is sketched, demonstrating how the removal of such training sessions paves the way toward solving the distinguishability problem.