Test Anxiety and Working Memory Impairment

Anxiety related to test performance appears to affect a significant number of individuals.  The prevalence of test anxiety in school children has been estimated as up to 15% to 20%.  Some studies have found the prevalence higher in girls than in boys--this would be consistent with the gender prevalence patterns found in other childhood anxiety disorders.

Children with test anxiety show educational assessment levels approximately 1/2 standard deviation below their nonanxious peers.  This translates into approximately a 12 percentile difference in these two groups.

In the U.S. there is increased emphasis on improvement in standardized testing performance with scrutiny of individual teacher and school performance effects.  Any intervention that would potentially increase performance in a subgroup of students (i.e. those with text anxiety) could be an important intervention in raising testing scores.

However, the effect of interventions on test anxiety on test performance is unclear.  Well-designed large-scale test anxiety interventions are limited.  Better understanding of the relationship between test anxiety and neuropsychological performance is needed.

Gass and Curiel recently published a study of test anxiety and neuropsychological performance in a group of adult veterans.  Two hundred and eighty six veterans referred for neuropsychological testing completed a comprehensive neuropsychological inventory known as the Halstead-Russell Neuropsychological Evaluation System-Revised (HRNERS).

Subjects were grouped by level of test anxiety related to response of a inventory known as the Test Anxiety Profile (TAP). Groups were then compared on a variety of neuropsychological domains with the following key results:

• High TAP scores more commonly found in those with lower levels of educational attainment (years of education)
• High TAP scores correlated with decreased performance on working memory performance
• High TAP scores did not correlate with impairment in processing speed, verbal comprehension, perceptual organization or overall neuropsychological test index performance

The authors note these findings suggest relatively intact neuropsychological performance in adults with self-reported test anxiety.  Neuropsychological testing appears valid even in those who report problems with test anxiety.

The findings also suggest that test anxiety may be a consequence rather than a cause of working memory impairment.  The direction and meaning of the correlation remains unclear.  If test anxiety is a consequence of reduced working memory, there may be limited change with interventions that target reducing test anxiety.

One problem with interpretation of the Cass and Curiel study is the age of the study population.  The mean age of subjects was 59.3 years of age.  Participants were long past the time of their formal educational period.  How well their rating of test anxiety correlated with actual test anxiety during their school years is unknown.

More detailed research of the correlates of test anxiety and neuropsychological performance in children and adolescents is needed.  These types of studies might suggest more effective interventions for improving cognitive performance (and standardized testing) in our schools.

Photo of Juno Beach, Florida sunrise from author's collection.

Gass CS, & Curiel RE (2011). Test anxiety in relation to measures of cognitive and intellectual functioning. Archives of clinical neuropsychology : the official journal of the National Academy of Neuropsychologists, 26 (5), 396-404 PMID: 21636602