The Mathematics Continuous Performance Test (MATH-CPT) is a new measure designed differently from other CPT-type tests. Thus far, the developer has found no other CPT-type tests that use mathematical problems as stimuli. There are four major differences between this new test and other CPT-type tests: 1) because the stimuli shown require more complex cognitive processing of calculations, responding requires more time. The intention of this change was to resolve the floor effects found in many CPT tests that use simple stimuli (Halperin, Sharma, Greenblatt, & Schwartz, 1991; Greenberg et al., 1999; Pei-Chun et al., 2005). On other tests, one unintentional mistake may lead to an erroneous ADHD diagnosis. 2) No time limit is set for answering each question in the MATH-CPT to allow better assessment of the pace of each individual tested. 3) Analyses of the results are divided into three parts of 150 problems each. Each of these parts is then divided into three more parts of 50 problems each (for a total of nine parts per test). This procedure allows better analysis of possible difficulties in sustained attention. 4) Sustained attention is reported as one number for the following four measures: total time, standard deviation of reaction time, total impulsive responses (a combination of the impulsive-guessing answers and the fast-wrong responses) and number of correct responses. A formula was developed to report worsening or improvement in the performance of the individual tested on the nine different parts of the test.
In other CPT-type tests, the stimuli appear on the screen for approximately 200 to 500 milliseconds. After that, an inter-stimulus interval is given, usually between 1 to 4 seconds, followed by the response of the individual tested. This method forces the examinee to respond within the designated inter-stimulus interval. In the MATH-CPT, the stimulus changes to the next stimulus only after the examinee responds to the stimulus appearing on the screen. The change to this self-paced procedure in the MATH-CPT is intended to encourage the examinee’s natural tendency to answer either quickly or slowly, as opposed to forcing the examinee to respond quickly, as in other CPT-type tests. In 2001, Milich, Ballentine, and Lynam introduced the term Slow Cognitive Tempo, which describes a behavioral pattern appearing in some children with ADHD and characterized by passivity and daydreaming or by hypo-activity, confusion, moving slowly and sluggish responses. A self-paced procedure, as used in the MATH-CPT, aims at assessing Slow Cognitive Tempo more accurately.
The MATH-CPT was programmed on a PC computer using the Visual Basic programming language. There are 450 simple arithmetical problems of addition, subtraction, multiplication, and division whose answer is not larger than 9. In the MATH-CPT, the stimulus of a mathematical problem is shown on the computer screen with an answer that is either correct or incorrect (i.e., 1 + 3 = 4 or 7 – 5 = 3). The program uses white numbers positioned on an approximately 1.0 centimeter square matrix against a black background.
The test includes the following main measures: final overall attention level formula to assess the participant’s overall attention level; reaction time (time taken to complete the test); standard deviation of reaction time (a measure of reaction time variability); correct answers—considered a measure of attention; impulsive responses (measured by a fast guessing response of less than 0.5 seconds and by incorrect responses that are faster than each participant’s average mean reaction time to all the problems). The results are then presented in three performance blocks from the beginning to the end of the test, with 150 arithmetical problems in each block of problems. This method allows assessment of sustained attention by subjective observation of the responses to check for a decline, an improvement or no change in individual measures as the examinee progresses through the performance blocks. The results are also presented in nine performance blocks from the beginning to the end of the test, with 50 mathematical problems in each block. This method allows for a more refined assessment of sustained attention through subjective observation of the responses to check for decline, improvement or no change in individual measures as the examinee progresses through the nine performance blocks test. This procedure is common in many other CPT-type tests (Halperin, et al., 1991; Conners, 2000).
To improve assessment of sustained attention, we constructed a unique formula to report this variable. The following measures assessed sustained attention by means of a single number: sustained-time (a measure of sustained attention of reaction time over nine parts of the test); sustained-SD (sustained attention of reaction time SD over nine parts of the test); sustained-correct responses (sustained attention of correct responses over nine parts of the test); and sustained impulsivity (a measure of sustained attention of impulsivity over nine parts of the test). These four measures of sustained attention assess a possible decline in performance on these measures in their respective areas and are based on an algorithm designed specifically to assess these domains. The algorithm for assessing sustained attention is based on a calculation of the nine blocks. From the first to the ninth block in ascending order, each block contributes more to the measure of sustained attention than its predecessor does. The result is a single number assessing sustained attention on the four measures mentioned above.
Eight research papers were published so far using the MATH-CPT as the research instrument:
Lufi, D., & Fichman, N. (2012). The Development and Use of a Computerized Test, MATH-CPT, to Assess Attention. Perceptual and Motor Skills,114, 59-74.
Lufi, D. (2011). Validation the Mathematics Continuous Performance Test (MATH-CPT), Circles of the Mind (in Hebrew) 6, 61-71.
Lufi, D., Tzchishinky, O., & Hadar, S. (2011). Delaying school starting times by one hour: effects on attention level in adolescents. Journal of Clinical Sleep Medicine, 7, 138-143.
Lufi, D. (2013). The Relation between Sleep Patterns and Attention Levels. Neurocase, 20, 591-598.
Lufi, D., Bassin-Savion, S., & Rubel, L. (2015). The Effect of Methylphenidate on Sustained Attention. Neurocase, 6, 802-808.
Lufi, D., Segev, S., Blum, A. Rosen, T., & Haimov, I. (2015). The Effect of Age on Attention Level: A Comparison of Two Age Groups. In print at the International Journal of Aging and Human Development. 81,176-188.
Lufi, D. & Pan, S. (2015). Comparing the Mathematics Continuous Performance Test with other Measures of Tests of Attention. European Journal of Psychological Assessment. 32, 291-297.
Lufi, D., Yun-Kwok Wing & Ngan-Yin, C. (2017). Cross-Cultural Assessment of Attention Level of Chinese and Israeli Adolescents. Journal of Behavioral Health, 6, 52-57.
Conners, C. K. (2000) Conners’ Continuous Performance Test II, technical guide and software manual. North Tonawanda, NY: Multi-Health Systems Inc.
Greenberg, L. M., Kindschi, C. L., & Corman, C. M. (1999). T.O.V.A. Test of Variables of Attention, clinical guide. Katelal: Universal Attention Disorders, Inc.