Leadership categorization theory proposes that individuals hold multiple implicit leadership theories (ILTs), each tailored to a specific context and organized hierarchically from superordinate to subordinate levels. A central assumption of this model is that ILT attributes maintain consistent meanings across levels, forming nested cognitive structures. This study questions that assumption by examining how the commonly cited attribute of intelligence is defined at hierarchical levels in a political leadership context. Drawing on serial cognitive and connectionist models of schema activation, we use cognitive mapping and verbal protocols with 30 participants to analyze intelligence at superordinate, basic, and subordinate levels. Our findings suggest that ILT attributes often shift in meaning across these levels, undermining the assumption of stable, coherent hierarchies. Quantitatively, adjacent levels shared about half of their attributes, with the rest unique to each level, underscoring both the commonality and the variability of leadership prototypes. Rather than nested subsets, ILTs reflect context-sensitive configurations that vary with level of abstraction. These results refine leadership categorization theory by challenging its structural assumptions and support a more dynamic model of leadership categorization. We conclude with implications for leadership development, performance evaluation, and succession planning, highlighting the importance of recognizing shifting expectations when assessing leaders.
Pitsi et al. (Mon,) studied this question.