Sally-Anne Task

The Sally-Anne task is a widely recognized experimental paradigm designed to probe a child's understanding that others can hold beliefs that differ from their own and from reality. Developed by Simon Baron-Cohen, Alan Leslie, and Uta Frith in 1985, it has become a cornerstone in the study of Theory of Mind (ToM), a fundamental human cognitive ability to infer and understand the mental states of others. The task's importance in evolutionary psychology stems from the hypothesis that ToM is a species-typical adaptation, crucial for navigating complex social environments, cooperation, and deception, and that its development may offer insights into its evolutionary history.

The Task Design and Procedure

The Sally-Anne task involves two dolls, Sally and Anne, and two objects, a basket and a box, along with a marble. The procedure typically unfolds as follows:

1. Introduction: The child is introduced to Sally and Anne. The experimenter ensures the child can identify each doll.
2. Scenario: Sally places her marble in her basket. The experimenter confirms the child understands this action.
3. False Belief Induction: Sally leaves the scene. While Sally is absent, Anne moves the marble from Sally's basket to her own box. The experimenter confirms the child understands this transfer.
4. Test Question: Sally returns. The child is then asked, "Where will Sally look for her marble?"

To pass the task, the child must answer, "In the basket." This answer indicates an understanding that Sally, despite the marble's actual location, still believes it is in the basket, as she did not witness Anne move it. A child who fails the task typically answers, "In the box," indicating that they are unable to disassociate Sally's belief from their own knowledge of reality. Control questions are also administered to ensure the child remembers who put the marble where and where it is now, ruling out memory or comprehension issues as reasons for failure.

Developmental Trajectory and Clinical Applications

Typically developing children reliably pass the Sally-Anne task around four to five years of age. Before this age, most children fail, suggesting a significant developmental shift in their ToM abilities. This age-related pattern has been replicated across various cultures, lending support to the idea of a universal developmental timetable for this cognitive capacity.

The task gained particular prominence in the study of autism spectrum disorder (ASD). Baron-Cohen, Leslie, and Frith (1985) famously used the Sally-Anne task to investigate the "mind-blindness" hypothesis, which posits that individuals with ASD have specific deficits in ToM. Their seminal study found that a significantly higher proportion of children with autism failed the Sally-Anne task compared to typically developing children and children with Down syndrome, who served as a control group for general developmental delay. This finding provided strong evidence for a specific ToM impairment in autism, contributing substantially to the understanding of the cognitive profile of ASD.

Evolutionary Significance and Critiques

From an evolutionary perspective, the ability to attribute mental states to others is considered a critical adaptation for complex social living. It allows individuals to predict and explain the behavior of conspecifics, engage in strategic social interactions, form alliances, deceive, and cooperate. The universality of its developmental trajectory and its apparent impairment in conditions like autism suggest a dedicated cognitive mechanism, potentially with an evolutionary history rooted in the demands of primate sociality.

However, the Sally-Anne task, and false-belief tasks in general, have faced several critiques. Some researchers argue that passing the task might not reflect a full-fledged ToM but rather a more basic understanding of representational states or behavioral rules. Dennett (1978) distinguished between first-order (understanding another's belief) and second-order (understanding another's belief about another's belief) ToM, and the Sally-Anne task primarily tests first-order ToM. More complex tasks, such as the "ice cream truck task" (Perner and Wimmer, 1985), are used to assess second-order false beliefs, which typically develop later.

Furthermore, some critics, such as Bloom and German (2000), suggest that failure on the Sally-Anne task might be due to executive function demands (e.g., inhibition of a prepotent response to point to the actual location of the marble) rather than a complete absence of ToM. Studies using 'implicit' or 'spontaneous' false-belief tasks, which measure looking times or anticipatory gaze rather than verbal responses, have shown that infants as young as 15 months might demonstrate an implicit understanding of false beliefs (Onishi and Baillargeon, 2005). This suggests that the capacity for ToM may emerge much earlier than indicated by explicit false-belief tasks like Sally-Anne, and that the task itself may underestimate infants' and young children's true ToM abilities due to linguistic or executive demands.

Despite these critiques, the Sally-Anne task remains a foundational tool in cognitive development and evolutionary psychology. It continues to inform research on the origins, development, and neural underpinnings of Theory of Mind, providing a standardized measure for comparing ToM abilities across different populations and developmental stages. Its simplicity and clarity have made it an enduring classic in the study of social cognition.