Olson, 1993). Theoretically, it seems that a 4-year- old child who has coordinated belief and desire rea- soning abstractly enough to understand false belief tasks would have no trouble comprehending desire- based emotions (Yuill, Perner, Peerbhoy, & van den Ende, 1996) and further should comprehend belief- based emotions. While there is evidence that sup- ports this possibility (Pasquale, 2001; Wellman & Bartsch, 1988; Wellman & Banerjee, 1991), another school of thought claims there is a developmental lag between a child's understanding of false-belief and his understanding of belief-based emotions (Hadwin & Perner, 1991; MacLaren & Olson, 1993; Ruffman & Keenan, 1996). It is a debate of the emo- tional consequences of beliefs and desires versus the cognitive theories of emotions in children, respec- tively (MacLaren & Olson, 1993).
In a study investigating the former, Wellman and Bartsch (1988) demonstrated that 4-year-olds have the ability to detect another individual's sur- prise when the emotion is the result of an incon- gruity between belief and reality. Children were told a story about a character whose likes and thoughts were explicitly detailed. They were then shown a picture of an outcome and were asked whether the character felt surprised or not. The children under- stood that the character would feel surprised when the outcome contradicted her belief.
Hadwin and Perner (1991), however, obtained conflicting results. In a study with similar procedures, Hadwin and Perner (1991) found that 4- year-olds who could pass false belief tasks explicitly understood the desire-based nature of happiness but had difficulty with the belief-based nature of sur- prise. Only 5-year-olds and 6-year-olds understood that a character would feel surprised if his belief was inconsistent with reality. Crucial differences may account for the discrepancy in results between Hadwin and Perner's (1991) procedures and those of Wellman and Bartsch (1988). First, Wellman and Bartsch (1988) read subjects warm-up stories and provided feedback for the subjects' responses, which could have taught the subjects how to respond. Furthermore, in Wellman and Bartsch's (1988) pro- tocol, the subjects discovered the story outcomes at the same time as the story character, which may have influenced the subjects' own reactions to the situa- tion.
In follow-up experiments, Hadwin and Perner (1991) found that 4-year-olds' success with Perspectives in Psychology
the surprise tasks in Wellman and Bartsch's (1988) study was most likely an artifact of teaching. They showed that telling children what a protagonist feels increases the likelihood of accurate performance during subsequent stories. Furthermore, children seemed to have just as much trouble identifying a character's belief-dependent happiness as surprise. Children generally seemed to interpret surprise as something pleasant rather than something unexpect- ed, indicating a delay between a child's understand- ing of false belief and that belief's emotional conse- quences.
In another attempt to outline the emotional consequences of beliefs and desires of young chil- dren, Wellman and Banerjee (1991) utilized back- wards reasoning tasks, which the experimenters argued were more sensitive than the forward reason- ing tasks applied by Wellman and Bartsch (1988) and Hadwin and Perner (1991). Responses to the open-ended questions were limited, so analysis relied on the data obtained from leading, potentially suggestible follow-up questions, a plausible short-
coming of the analysis 1991).
(Ruffman and Keenan,
Wellman and Banerjee (1991) concluded that most 3-year-olds understand the belief-based nature of surprise, based on their belief-based responses to the follow-up questions. They claimed as well that their procedures circumvented young children's potential to misinterpret specific emotion terms, such as the word surprise, and thus served to simpli- fy the dilemma further for the subjects. These find- ings do not directly contradict those of Hadwin and Perner (1991), but rather provide insight into the developmental progression. Three-year-olds seem to have a developing sense of belief-based reasoning that is implemented in certain, but not all, knowl- edge- and emotion-related problems.
MacLaren and Olson (1993) sought to fur- ther explain the discrepancy in findings and the pos- sibility of a developmental lag. They found that failed expectations are harder developmentally to interpret than unexpected treats. Their results were consistent with Hadwin and Perner's (1991). Using backwards reasoning tasks similar to those of Wellman and Banerjee (1991), MacLaren and Olson (1993) concluded that 3-year-olds have a concept of surprise that is developmentally different from that of slightly older children. They contended that it is not until five or six years of age that children rely on