University of Louisville
Reasoning About Mental Illness: a Study of the Influence of Causal Knowledge on Categorization
Institution
University of Louisville
Faculty Advisor/ Mentor
Patrick Shafto
Abstract
Kim & Ahn’s (2002) study sought to explore reasoning about mental illness by asking participants to draw arrows indicating causal relationships between symptoms of common mental disorders, and then assign numeric strengths to those relationships, and rate the importance of each symptom in diagnosis. These processes were intended to determine causal and conceptual centrality. The authors found a positive correlation between the causal and conceptual centrality ratings, suggesting that features are “conceptually central to the extent that they cause other features.” We propose that Kim & Ahn’s methodology is insufficient to explicitly link participants’ responses to their beliefs about causal relationships among symptoms of each mental disorder. An additional measure will be incorporated into the study’s replication of Kim & Ahn’s experiment, in which participants will be asked to judge the likelihood of each symptom in the presence or absence of the other two. This study aims to determine whether or not causal theories are the source of participant’s responses concerning relationships between symptoms of common mental disorders, or whether a prototype account is more consistent with people’s judgments. A third possibility is that judgments are consistent with both, in which case an amalgamated model including aspects of both approaches will be proposed. If either of the initial cases are found to be true, our current methods of diagnosis, including the diagnostic methods currently condoned by the current edition of the Diagnostic and Statistical Manual, may have to be seriously reviewed and possibly augmented in order to better accommodate the way people actually reason.
Reasoning About Mental Illness: a Study of the Influence of Causal Knowledge on Categorization
Kim & Ahn’s (2002) study sought to explore reasoning about mental illness by asking participants to draw arrows indicating causal relationships between symptoms of common mental disorders, and then assign numeric strengths to those relationships, and rate the importance of each symptom in diagnosis. These processes were intended to determine causal and conceptual centrality. The authors found a positive correlation between the causal and conceptual centrality ratings, suggesting that features are “conceptually central to the extent that they cause other features.” We propose that Kim & Ahn’s methodology is insufficient to explicitly link participants’ responses to their beliefs about causal relationships among symptoms of each mental disorder. An additional measure will be incorporated into the study’s replication of Kim & Ahn’s experiment, in which participants will be asked to judge the likelihood of each symptom in the presence or absence of the other two. This study aims to determine whether or not causal theories are the source of participant’s responses concerning relationships between symptoms of common mental disorders, or whether a prototype account is more consistent with people’s judgments. A third possibility is that judgments are consistent with both, in which case an amalgamated model including aspects of both approaches will be proposed. If either of the initial cases are found to be true, our current methods of diagnosis, including the diagnostic methods currently condoned by the current edition of the Diagnostic and Statistical Manual, may have to be seriously reviewed and possibly augmented in order to better accommodate the way people actually reason.