Guide to ReportsNEUROPSYCHOLOGICAL ASSESSMENT LABPAUL JONES, ED.D.
UNIVERSITY OF NEVADA, LAS VEGAS
____________________________________________________________________
ASSESSMENT OF PLANNING FUNCTION:
Assessment Lab Report 3-1
June/1999
Luria (1973, 1980) proposed a model of neuropsychological function involving three "blocks" or functional units, necessary for any type of mental activity. Block 1, identified as attention or arousal, is primarily involved with enabling a person to focus attention, take in information, and maintain sufficient alertness to attempt problem solving. Block 2 involves the overall capability of an individual to input and process information. This unit is further subdivided into a sequential/successive processing function associated with integrating information in serial order, and a simultaneous function associated with more holistic, often spatial, integration. Block 3, typical identified as the planning function is associated most directly with the prefrontal areas of the brain. It involves the capabilities to generate and implement hypotheses and use decision-making strategies to evaluate them.
Luria's proposal is arguably the most influential contemporary theoretical model for neuropsychological function. Recent test instruments built around this model include the Kaufman Short Neuropsychological Assessment Procedure (Kaufman & Kaufman, 1994) and Naglieri and Das’s (1997) Cognitive Assessment System (CAS). In the CAS, the four functional units are identified with the acronym Planning-Attention-Simultaneous processing-Sequential Processing.
Prior studies in this lab in the spring of 1998 and in the fall of 1998 with cognitive scales from the ANAM/TWB battery suggest that the attention function is tapped with the Simple 2-Choice Reaction Time scale and the Procedural Reaction Time scale. The simultaneous function is tapped with the Spatial Processing and the Matching To Sample scales. The ANAM/TWB mathematical processing scale appears to tap the successive function. Data suggest that two special scales, identified as CogAttention and CogMemory, designed for use by persons with visual disability are associated with the attention and successive processing scales, respectively.
Inspection of the CAS scales used for assessment of the planning function suggested that the key element was in how quickly the individual "catches on", recognizes a pattern in the task. Differentiation is based on speed of response; essentially 100% accuracy would be expected on the tasks if there were no time limit.
This study was designed to test the hypothesis that adaptation of one of the instruments in the ANAM/TWB battery could provide a comparable task for measuring the planning function. The scale selected for adaptation is identified in the ANAM/TWB battery as Grammatical Reasoning. A summary of the procedures used for adaptation of this scale is in the instrumentation section below.
Method
Participants
The participant sample for this study was comprised of upper division teacher education students enrolled in a required educational psychology course in the spring of 1999. A total of 82 subjects provided complete data for this study.
Instrumentation
From the ANAM/TWB battery, participants in this study completed the Simple 2-Choice Reaction and Procedural Reaction Time cognitive scales and a revised adjective checklist for state personality characteristics from the Mood Scale 2 included with the ANAM/TWB Tester's Workbench.
Each participant also completed two of the Planning subtests from the Cognitive Assessment System (CAS). Although norms for the CAS are available only through age 17, perusal of the CAS norms tables suggested that performance on the planning scale appeared to be stabilizing near the upper age range. It thus appeared reasonable to assume that the stabilization could be to an extent to provide sufficient discrimination at age levels beyond the norms, and the data obtained from this participant sample supported this assumption. In the Basic CAS battery, the two subtests used in this study are combined to form a Planning score with mean = 100 and s.d. = 15. Using the highest available age norms (17-8 to 17-11), the upper division students in this study obtained a mean of 109.3 with a standard deviation of 14.81. This outcome is consistent with expectations for the ability level of the sample.
The standard ANAM/TWB Grammatical Reasoning scale was designed to tap the higher mental processes associated with logical reasoning. For the present study, however, the scale was adapted with objective to create a task with outcome hypothesized to be comparable to the results in CAS Planning scales.
In the ANAM/TWB Grammatical Reasoning scale, the user is first given two symbols separated by a relationship identifier (e.g. * AFTER #). This is then followed by another statement (e.g. & AFTER *), and then by the test stimulus (e.g. # * &). In standard application, the user responds with "match" if both of the statements are true or if both of the statements are false and responds "nonmatch" if only one of the statements is true.
The adaptation for this study involved first making a change in the internal structure of the scale such that one of the statements would always be true. The user would then be asked to respond by indicating whether both statements are true (e.g. the test stimulus is correctly described by both statements) or whether one of the statements is false. With this modification, the user who "catches on" to the pattern would recognize that if the first statement is false, it is not necessary to attend to the second statement. The proposed "score" on the test would be based on the difference in efficiency (accuracy/speed) of response between instances in which it was necessary to attend to both statements and instances in which it was not.
Procedures
Each participant scheduled a one-half hour appointment in the lab which began with administration by the research assistant of the two CAS scales using standard procedures from the test manual. During the same appointment, the participants then completed the following battery of instruments from the ANAM/TWB battery.
1. Simple 2-Choice Reaction
2. Adapted Grammatical Reasoning
3. Adjective Checklist from Mood Scale 2
4. Procedural Reaction (5 trials)
CAS scales were hand scored by the research assistant. ANAM/TWB scales were both administered and scored by the computer. The standard ANAM/TWB cognitive processing scales have a low ceiling on the accuracy score with participants typically obtaining scores of 90% or above. For these scales, particularly with nonclinical samples, the cognitive efficiency score has been suggested (Levinson & Reeves, 1997) as the more useful score. Cognitive efficiency, originally labeled as "throughput", is a function of both accuracy and speed (correct responses/minute). The efficiency score is the primary unit of analysis for the cognitive processing scales in this study.
The Simple 2-Choice Response and Procedural Reaction test trials were immediately proceeded by a short practice trial. For the grammatical reasoning scale, both orientation and practice trials were included.
The intent of the orientation trials for the grammatical reasoning scale was to provide familiarity with the general format before the actual test trial. For the orientation trials, the participant was presented with a simpler form of the actual task. The participant was given (on the screen) just one statement using two symbols separated by a relationship identifier (e.g. * AFTER #) and the test stimulus (e.g. * # %). The task in the orientation trials was simply to identify whether the statement correctly described the order in the test stimulus. Participants responded to five orientation practice items before the 10 items in the orientation trial.
The grammatical reasoning orientation trials were then immediately followed by a practice trial with the actual test which used two statements before the test stimulus, constructed with the three conditions described above. There were five practice items for the actual test and 30 items in the test.
The actual test trial with the adapted grammatical reasoning scale was comprised of 30 items, constructed such that the first statement was incorrect (Condition A) for 10 of the 30 items, the second statement was incorrect (Condition B) on 10 of the 30 items, and both statements were correct (Condition C) on 10 of the 30 items. Items for the three conditions were randomly dispersed through the trial.
Results
Table 1 below provides the intercorrelation matrix and descriptive data for the cognitive scales. It was hypothesized that participants could have a higher level of performance on Condition A of the grammatical reasoning scale because a correct response could be given after reading only the first statement in each test item. Conditions B and C required reading both statements in order to provide a correct response. These data were consistent with this hypothesis with mean scores for Conditions A, B, and C of 14.0, 11.6, and 10.8, respectively. As anticipated, there were statistically significant correlations between CAS scores and the various conditions of the grammatical reasoning scale.
| 2-choice m=132.4 sd=18.55 | react-1 m=104.9 sd=13.13 | numMatch m=21.0 sd=5.27 | planCode m=117.0 sd=28.08 | planTot m=109.3 sd=14.81 |
gramA m=14.0 sd=6.03 | gramB m=11.6 sd=4.53 | gramC m=10.8 sd=3.91 | gramDiff m=2.4 sd=3.61 | |
|---|---|---|---|---|---|---|---|---|---|
| 2-cho | 1.0 | .60 | .15 | .29* | .28* | .22* | .33* | .31* | -.05 |
| react-1 | x | 1.0 | .21 | .31* | .32* | .26* | .38* | .33* | -.04 |
| nuMtch | x | x | 1.0 | .49* | .87* | .38* | .32* | .31* | .24* |
| plCod | x | x | x | 1.0 | .84* | .44* | .42* | .43* | .20 |
| plTot | x | x | x | x | 1.0 | .47* | .43* | .43* | .25* |
| gramA | x | x | x | x | x | 1.0 | .80* | .77* | .66* |
| gramB | x | x | x | x | x | x | 1.0 | .82* | .09 |
| gramC | x | x | x | x | x | x | x | 1.0 | .26* |
____________________________________________________________________* significant at .05 level
2-choice: Simple Two-Choice Reaction Time Efficiency Score
react-1: Procedural Reaction Time Efficiency Score
numMatch: CAS Matching Numbers Raw Score
planCode: CAS Planned Codes Raw Score
planTot: CAS Planning Scaled Score
gramA: Grammatical Reasoning-Condition A Efficiency Score
gramB: Grammatical Reasoning-Condition B Efficiency Score
gramDiff: Difference Between gramA and gramB Efficiency scores
Participants with Total CAS Planning scores above the national norm of 100 (54 of the 82 participants in this study) had higher scores on condition A of the grammatical reasoning scale (t (80) = 3.71, p < .001), on condition B (t (80) = 3.20, p < .01), on condition C (t (80) = 3.25, p < .01), and on the score calculated as the absolute difference between Condition A and Condition B (t (80) = 1.99, p < .05). This finding was also not unanticipated given the likelihood of a general mental processing efficiency capability which would be expected to influence all scales on which there is evident cognitive demand. Higher performance by persons above the national norm on the CAS Planning scale was also evident on the Procedural Reaction Time scale (t (80) = 2.59, p < .05) but not on the Simple 2-Choice Reaction scale (t (80) = 1.32, p > .05), the scale with minimal cognitive demand.
With one exception, there was no evident relationship between performance on the cognitive scales and the self-report of personality mood state on scales for activity, attention, anxiety, and depression. The exception was a statistically significant coefficient of .33 between the scale for activity (adjectives: energetic, restless, easily excited, impulsive, loud, curious) and the first trial of the Procedural Reaction Test. Only the first trial of the Procedural Reaction Test is used as a variable in this study, but five trials were actually completed by participants, the additional trials included for normative purposes. Correlation coefficients of the activity scale with the other Procedural Reaction Time trials were only .19, .17, .14, and -.05, respectively. This suggests that the initial correlation may well have been simply an artifact and not indicative of significant relationship.
The primary research question in this study was the hypothesis that the adapted Grammatical Reasoning scale would provide an ANAM/TWB alternative with outcome consistent with assessment of the planning function as measured by the CAS. Data which appear to support this hypothesis are evident in Tables 2-5 below. Some caution is needed in interpreting the results portrayed in these tables because of the relatively small n per factor. However, given this caution, the exploratory factor analysis data appear consistent with the hypothesized relationships.
Previous studies in this lab have suggested that the Simple 2-Choice Reaction and the Procedural Reaction Time tests will load on a single factor identified as attention. CAS studies (Naglieri & Das, 1997) provide evidence that the Matching Numbers and Planned Codes scales load on a planning factor. Data in Table 2 indicate that Condition A of the adapted grammatical reasoning scale, as hypothesized, loads with the CAS scales. In Condition A, the test stimulus did not follow the rule in the first statement, so a correct response did not require the participant to attend to the second statement and thus more rapid response was possible. In effect, as in the CAS scales, the participant had opportunity to "catch on" to the pattern.
Tables 3 and 4 provide comparable analysis with grammatical reasoning Conditions B and C, respectively. In Condition B, the test stimulus followed the rule in the first statement but not in the second statement. In Condition C, the test stimulus followed the rule in both statements. Thus, in both conditions it was necessary for the participant to attend to both statements in order to give a correct response. Data in Tables 3 and 4 indicate that, unlike Condition A, the efficiency scores under Conditions B and C did not load with the CAS scales.
In Table 5 the Grammatical Reasoning score was the absolute difference between efficiency under Condition A and efficiency under Condition B. In both instances, the test stimulus did not follow the rule in one of the two statements, but in Condition B it was necessary to attend to both statements in order to give a correct response. The A minus B efficiency score, as hypothesized, loaded with the CAS scales.
| Factor 1 | Factor 2 | ||||
|---|---|---|---|---|---|
| Simple 2-Choice Reaction | .09 | .90* | |||
| Procedural Reaction | .17 | .87* | |||
| CAS Number Matching | .82* | .04 | |||
| CAS Planned Codes | .77* | .26 | |||
| Gramm. Reasoning A | .73* | .18 | |||
| percent of variance | 37% | 33% |
| Factor 1 | Factor 2 | ||||
|---|---|---|---|---|---|
| Simple 2-Choice Reaction | .88* | .08 | |||
| Procedural Reaction | .86* | .16 | |||
| CAS Number Matching | .01 | .86* | |||
| CAS Planned Codes | .24 | .80* | |||
| Gramm. Reasoning B | .45 | .55 | |||
| percent of variance | 36% | 34% |
| Factor 1 | Factor 2 | ||||
|---|---|---|---|---|---|
| Simple 2-Choice Reaction | .09 | .89* | |||
| Procedural Reaction | .16 | .86* | |||
| CAS Number Matching | .84* | .01 | |||
| CAS Planned Codes | .80* | .24 | |||
| Gramm. Reasoning C | .59 | .38 | |||
| percent of variance | 35% | 35% |
| Factor 1 | Factor 2 | ||||
|---|---|---|---|---|---|
| Simple 2-Choice Reaction | .87* | .06 | |||
| Procedural Reaction | .86* | .11 | |||
| CAS Number Matching | .19 | .78* | |||
| CAS Planned Codes | .39 | .71* | |||
| Gramm. Reasoning A-B | -.28 | .71* | |||
| percent of variance | 35% | 32% |
Summary
These data appear to clearly support additional exploration of the use of the adapted grammatical reasoning scale as a computer-based alternative for assessment of the planning function. Further study is needed to insure that these results were not simply an artifact of this sample group and to explore development of optimal norms for performance on this scale.
A particular value in using the "difference" score may be in providing for the possibility of web based assessment. Timing of responses is an essential element in the efficiency scores, and there is appropriate concern that the necessary precision to allow normative comparisons in timed instruments may not be available when tests are administered via the Internet/WWW. However, when the base is a contrast in time of response under two conditions of administration (e.g. Condition A vs. Condition B), such administration may be feasible. Future lab studies will examine this possibility.
References
Kaufman, A.S., & Kaufman, N.L. (1994). Kaufman short neuropsychological assessment procedure. Circle Pines, MN: American Guidance Service, Inc.
Levinson, D.M., & Reeves, D.L. (1997). Monitoring recovery from traumatic brain injury using Automated Neuropsychological Assessment Metrics. Archives of Clinical Neuropsychology, 12, 155-166.
Luria, A.R. (1973). the working brain: An introduction to neuropsychology. New York: Basic Books.
Luria, A.R. (1980). Higher cortical functions in man.(2nd ed.). New York: Basic Books.
Naglieri, J.A., &; Das, J.P. (1997). Cognitive Assessment System. Itasca, IL: Riverside Publishing.
________________________________________________________________________________________________________________________________________