Grammatical Reasoning Test

Grammatical Reasoning Test

Article Information
Category:

Description

The grammatical reasoning test is a test of the ability to carry out mental operations involving chains of logical reasoning.

Discussion/References

References to the use of the method in ATC or similar situations.

  • Baddeley A D, A 3 MIN REASONING TEST BASED ON GRAMMATICAL TRANSFORMATION, 1988, Psychonomic Science, Vol 10(10), pp 341 342.
  • Carter Robert C, Kennedy Robert S and Bitter Alvah C, GRAMMATICAL REASONING: A STABLE PERFORMANCE YARDSTICK, 1981, Human Factors, 1981 Vol 23(5), pp 587 591.
  • EUROCONTROL Experimental Centre Report 183 (David 1986) found that there was a strong cross-cultural bias in this test. German language speakers found the negative passive construction particularly difficult to follow. Most controllers, whatever their native language, preferred to use the English version of the test.
Categories
Human Factors Methods
Generics
Type of method Mental Capacity Test
Target of method Strain
This test is applied before and after an exercise, with the intention of measuring a possible decrease in mental efficiency during the exercise.
Time Scale of method Hours
This is a before-and-after test, which is applied at the start and end of an exercise. It can therefore provide no greater precision than over the entire exercise.
Portability of method Yes
Paper and pencil (or laptop).
Observer Effect No
No observer effects to be expected.
Context of studies
Laboratory studies Avoid
 
Simulation studies Avoid
 
Field studies Avoid
 
Potential problems with the method
Failure risk High
This test aims at identifying a decrement in logical reasoning following heavy mental work. It has produced significant results in other contexts (nitrogen narcosis, and as a secondary task while driving). Unfortunately, the inevitable time lag between the end of the Real Time simulation Run appears to allow time for the controller to recover from any work-related effects.
Bias risk High
A specific cross-cultural bias appears to occur with this test. Controllers whose native language is German appear to find the passive negative construction particularly difficult. (Some controllers could not answer the question: ‘B is not preceded by A -AB’). (All controllers preferred to take this test in English).
Ethical problems None
This test is generally available for research work.
Costs of the method
Staff Cost Low
The test forms must be distributed and collected at the start and end of each run. Either a technician or, better, the project leader or assistant project leader can do this.
Set-up Cost Low
Tests can be generated by a cheap public domain generator, which also prepares scoring sheets.
Running Cost Low
Forms/Computer storage.
Analysis Cost Low
The cost of analysis is mainly the time and speed of transcription of the controllers’ responses.
Analysis data
Analysis Speed Moderate
Although each test can be scored using the prepared score-sheet, the results must usually be transcribed manually.
Data Automation Yes
A computer-based version of the test, with individually presented and scored questions could be developed if warranted.
Analysis Automation Yes
Most data analysis is more or less automatised in the current circumstances. No specific human interpretation problems expected.
Status Rejected
This test appears to work in cases of physiological stress, and where it can be used in addition to an on-going task. It does not appear to work as a before-and-after test.

SKYbrary Partners:

Safety knowledge contributed by: