An Interpretative Analysis of Five Commonly Used Processing Speed Measures

Abstract

Processing speed subtests are components of widely used intellectual assessment instruments. Many researchers interpret these measures as assessing a unitary construct, but there is a question concerning the constructs assessed by these measures and, ultimately, their interpretative utility. Coding and Symbol Search from the Wechsler Intelligence Scale for Children-Third Edition (Wechsler, 1991), Visual Matching and Cross Out from the Woodcock Johnson Tests of Cognitive Ability-Revised (Woodcock & Johnson, 1989), and Speed of Information Processing from the Differential Ability Scales (Elliott, 1990) were administered to 102 volunteer participants. Using regression analyses, performance on each of these tests was predicted by motor speed and/or number facility factors. Individual differences in motor speed were found to be related to each of the five processing speed measures, whereas number facility was related to three of the measures.

Latent Factors Underlying Individual Differences in Attention Measures

“…In order to specify empirically the extent to which processing speed contributes to the Perceptual attention factor in model D1, the influence of processing speed needs to be partialled out by implementing a processing speed factor. Processing speed measures should be selected carefully since performance in commonly used speed measures may be related to many different skills, e.g., visual memory, attention, working memory, and visual scanning (see Feldman, Kelly, & Diehl, 2004). Inasmuch as processing speed indicators tap attention-related skills, the contribution of Processing speed to the Perceptual attention factor would be inflated.…”

Differential effects of reasoning and speed training in children

“…We examined the effects of cognitive training on two different PS measures (Feldmann, Kelly & Diehl, 2004): Cross Out from Woodcock‐Johnson‐Revised and Coding B from Wechsler Intelligence Scale for Children IV. Cross Out is a timed test in which one must rapidly identify and put a line through each instance of a specific symbol in a row of similar symbols.…”

Construct validity of the symbol-digit modalities test

“…This differs from Gr tasks which require novel responses and place a greater demand on more complex cognitive functions. 3 The history of the SDMT as covered by Benedict et al 1 shares roots with the familiar coding test from the Wechsler Intelligence Scales-still a gold standard for intelligence tests. An analysis of five common Gs tests including coding found that all were robustly intercorrelated as expected and showed poorer relationship to higher order timed tasks such as speeded arithmetic.…”

“…An analysis of five common Gs tests including coding found that all were robustly intercorrelated as expected and showed poorer relationship to higher order timed tasks such as speeded arithmetic. 3 More recent factor analyses have found that coding is an excellent measure of Gs with no evidence of cross-loading on other broad abilities. 4,5 We wholeheartedly agree that SDMT involves more than processing speed, and one should note that the psychology factor structure literature is based on the study of healthy persons.…”

“…Indeed, even in 'stable' patients, a subtle and more gradual decline of cognitive performance as well as the lack of an expected practice effect cannot be excluded. In a previous work, 3 we applied reliable change methodology to a group of 54 healthy controls (34 women, mean age 38.3 ± 9.1 years, mean education 12.9 ± 3.3 years) receiving a neuropsychological evaluation using the Brief Repeatable Battery and Stroop Test 3 times, with an 18-month interval. As for the SDMT, the observed standard deviation of the difference between the first two assessments was 8.26, and practice effect 1.43 points.…”