Date of Award


Document Type


Degree Name

Doctor of Psychology (PsyD)



First Advisor

Susan Zimny, Ph.D.

Second Advisor

Donald Robertson, Ph.D.

Third Advisor

Lynda Federoff, Ph.D.


This study explored the phenomenon of stereotype threat in a population of undergraduate psychology students at a small, rural, state university in Pennsylvania. The primary goal was to attempt to establish a clear and concise mechanism of operation to explain the origin of the phenomenon, using Walter Kintsch’s (1986) Construction-Integration model of text processing as a base. Participants were asked to read a series of brief texts. Following each text, participants were asked to determine whether or not a number of sentences had appeared in the preceding texts. The analysis of the pattern of reading times and recognition of sentences was used to test the hypothesis that participants would experience disruption of performance, primarily through the activation of threatening evaluative information at the situation model level of representation. Long-Term Working Memory structures associated with stereotypic information were thought to be a key factor in decreased processing speed and memory disruption under conditions of stereotype threat. Black participants primed to approach a task as if it were a measure of their intelligence were expected to suffer performance deficits through the activation of task-unrelated memory constructs associated with stereotypes. Meanwhile, those individuals operating under stereotype threat who were exposed to a stereotype-relevant text were expected to show heightened performance, due to the relevance of the stereotypic information. The results from the analyses of the reading time and recognition data suggested a consistent pattern of activation of stereotype threat among the White participants, with little reaction to this manipulation among the Black students. Black students were noted to consistently focus their attention on the surface level of representation, while the White students focused primarily on the situation model level. This pattern had not been previously investigated in stereotype threat research and several potential explanations were discussed. Of significance, these patterns would not have become evident were it not for the application of a well-defined model of text processing, such as the Construction-Integration model.