Reference Work Entry

Encyclopedia of Clinical Neuropsychology

pp 703-706

Controlled Oral Word Association Test

  • Janet PattersonAffiliated withDepartment of Communicartive Sciences and Disorders, California State University East Bay


Category fluency; CFL test; COWA; COWAT; F-A-S test; Letter fluency; Phonemic fluency; Verbal fluency


The Controlled Oral Word Association Test (COWAT) is a measure of verbal fluency and is a subtest of the Multilingual Aphasia Examination (MAE; Benton, Hamsher, & Sivan, 1994). The COWAT uses the three letter set of C, F, and L to assess phonemic fluency. Individuals are given 1 min to name as many words as possible beginning with one of the letters. The procedure is then repeated for the remaining two letters (see Strauss, Sherman, & Spreen, 2006 and Benton, Hamsher, Rey, & Sivan, 1994 for specific administration instructions). Several tests of phonemic fluency exist, some of which are part of larger test batteries (e.g., the MAE or the Neurosensory Center Comprehensive Examination for Aphasia; Spreen & Benton, 1977) and others that can be administered independently (e.g., the F-A-S Test).

Verbal fluency is a cognitive function that facilitates information retrieval from memory. Successful retrieval requires executive control over cognitive process such as selective attention, mental set shifting, internal response generation, and self-monitoring. Tests of verbal fluency evaluate an individual’s ability to retrieve specific information within restricted search parameters (Lezak, Howieson, Loring, Hannay, & Fischer, 2004). The two most common parameters are semantic fluency and phonemic fluency.

Historical Background

Borkowski, Benton, and Spreen (1967) were early proponents of systematically examining word fluency in persons with brain damage. They recognized that so-called word fluency tasks had been used in neuropsychological investigation of patients with brain damage and undertook two studies of the task. The first established the relations between word fluency and various English letters; the second examined word fluency data for persons with brain damage and control patients. They presented normative data for the first study and comparative data for the second, clearly supporting their hypotheses of the utility of word fluency assessment.

Since publication of these data, word fluency tasks, and the COWAT in particular, have been investigated in detail. Spreen and Risser (2003) suggest that this assessment tool in its various forms has been one of the most frequently and thoroughly investigated neuropsychological assessment measures for unimpaired and neurologically impaired persons. A search of electronic databases confirms this suggestion.

Psychometric Data

Psychometric data for the COWAT and other phonemic fluency tests, as well as other verbal fluency tasks (e.g., semantic fluency) are readily available. Norms have been published for children and adults of varying ages, levels of education, ethnic diversity, and geographical diversity (Loonstra, Tarlow, & Sellers, 2001; Strauss et al., 2006). Some differences have been noted between test forms, most notably, between the COWAT and F-A-S Test (Barry, Bates, & Labouvie, 2008); the CFL form appeared more difficult, while results for the F-A-S form appeared more variable. In addition, COWAT scores have been correlated with neuropsychological measures such as reading tests and IQ tests.

Clinical Uses

Scoring for the COWAT and other verbal fluency tests is straightforward. The examiner writes each word as it is produced by the individual. The transcript is reviewed and inadmissible words (i.e., repetitions, proper names, or slang) are eliminated. The test score is the total number of different words produced for all three letters (see Strauss et al., 2006 and Benton et al., 1994 for specific administration instructions).

Supplementary scoring measures for the COWAT and other phonemic fluency tests provide additional information in clinical diagnosis and treatment. Supplementary scorning measures are error analysis, and cluster and switching analysis (see Table 1). In error analysis, the examiner notes any observable pattern of production of errors that suggests a loosening of executive control over cognitive processes that would result in errors. For example, a pattern of multiple repetitions of previous responses suggests perseveration and inefficient self-monitoring. Error patterns provide qualitative performance data and may appear as common patterns such as repetition of a word, or idiosyncratic patterns. Clustering and switching analyses evaluate the depth of the searchable knowledge base (clusters) and the cognitive flexibility within and across categories (switching) (Troyer, Moscovitcvh, & Winocur, 1977). An example of an efficient search strategy would be identifying a cluster or subcategory within the category (e.g., words that begin with “cr” in the COWAT task of naming words that begin with C) and producing as many items in that category as possible and then switching clusters (e.g., to words beginning with “cl”). Clusters are scored by counting the number of clusters and calculating the mean cluster size; switches are scored by counting the number of transitions between clusters. Rules for scoring cluster size and number of switches appear in Troyer et al. (1997) and normative data for healthy adults appear in Troyer (2000).
Controlled Oral Word Association Test. Table 1

An example of F-A-S Test results and cluster and switch scoring for a person with aphasia



Clusters and Switches

FAS-Test: “F”

fast, fun, fickle, fuschsia, finger of fate,

four, fifty-four, forty, fornicate

Words = 9

Clusters = 8 (mean size = 1.1)





finger of fate



forty, fornicate

Switches = 7

FAS-Test: “A”

apple, aardvark, alpaca,

ammonia, arsenic

Words = 5

Clusters = 5 (mean size = 1)






Switches = 4

FAS-Test: “S”

substantial, sum, subtraction,

stuck, structure, symbol, sympathy, stroke, sixty

Words = 9

Clusters = 5 (mean size = 1.8)

substantial, sum, subtraction

stuck, structure

symbol, sympathy



Switches = 4


Words = 23

Clusters = 18

Mean size = 1.27

Switches = 15

Scores from the COWAT are useful in evaluation of persons with neurogenic communication disorders, such as aphasia following stroke, traumatic brain injury, and dementia. Studies have included COWAT in the diagnostic batteries given to several patient populations and also in treatment studies as measures of behavioral change. The utility of the COWAT in identifying the nature and severity of performance deficits in clinical populations has been supported; however, conflicting findings have been reported. Typically, the total number of acceptable responses is the reported test result; however, increasingly cluster and switching scores are reported as well. The COWAT is valuable in detecting cognitive dysfunction, but it requires further study before definitive conclusions are possible with regard to performance patterns that can be linked with specific neurogenic behavioral deficits.

Cross References


Aphasia Tests

Benton, Arthur

Boston Diagnostic Aphasia Examination



Cognitive-Communication Disorder

Cognitive Functioning

Cognitive Processing


Cued Recall

Error Recognition and Correction

Free Recall

Mayos Older American Normative Studies (MOANS)

National Adult Reading Test

Phonemic Cue

Semantic Cue

Semantic Fluency

Verbal Mediation

Western Aphasia Battery

Copyright information

© Springer Science+Business Media, LLC 2011
Show all