The neuropsychological construct of fluency refers to productivity under timed and limited search conditions. Verbal fluency tasks, such as the Controlled Oral Word Association Test (COWAT), require production of words under specific constraints (i.e. words that begin with "F"), whereas nonverbal fluency tests, such as the Design Fluency Test (DFT), require generation of unique and unnameable designs. Asymmetric deficits have been obtained, with verbal impairments associated with left frontal pathology and nonverbal impairments with right frontal involvement. These fluency constructs, however, have not been systematically examined in conjunction with one another, or with other frontally-mediated tasks. Therefore, this experiment evaluated verbal and nonverbal fluency, and their relationship to tasks purported to be sensitive to frontal lobe functioning. A double dissociation was predicted; the COWAT would be associated with left but not right frontal tasks, but the reverse pattern would be evident for the DFT. Left frontal tasks included the Trail-Making Tests and the Stroop Color-Word Test, whereas right frontal tasks consisted of the Ruff Figural Fluency Test (RFFT) and newly created measures of facial accuracy and intensity. Bilateral motor tasks (i.e. dynamometer and finger tapping) were also evaluated.

In this sample of male college students (N = 60), multivariate analyses (MANOVAS) indicated that the left-hemisphere variables successfully discriminated subjects classified as verbally fluent or verbally nonfluent, but did not do so with nonverbal fluency. The opposite pattern was obtained with right hemisphere variables, which was due to the RFFT rather than the facial measures. Measures of facial accuracy and intensity, as well as motor tasks, were minimally related to either of the fluency measures. Additionally, univariate analyses (ANOVAS) revealed that verbally fluent subjects performed better than verbally nonfluent subjects on Trails B, but not A, and on selected components of the Stroop.

These findings are discussed in regard to their capacity to demonstrate asymmetries in frontal lobe functioning. Relationships between the presumed frontal measures and fluency constructs is also used to examine the unique ways in which fluency measures may challenge the frontal lobes in higher-order cognitive processing. Recommendations for the development of improved verbal and nonverbal fluency measures are provided.