Predicting potential intelligence

Abstract

Veterans A~r n~n j~~r u~~~Hospital Denver, Colorado Many writers have stressed the usefulness of distinguishing between a measured intelligence test score and some higher score, which the S is presumed capable of obtaining. The hypothetical higher level of intellectual functioning has been described as intellectual altitude@), intellectual potentiality (*), available intelligence@), intelligence level as distinguished from the functioning intellect (19 21, potential intelligen~e(~, 9), intellectual capacitycs), usable intelligence (4), and other terms acknowledging this distinction. These terms may be grouped together as referring to a global concept of "potential intelligence" as distinguished from measured intelligence. PROBLEMMany techniques have been proposed to estimate a potential intelligence score from the profile of subtest scores on the Wechsler-Bellevue (W-B) test. Recently, a method has been proposed as a useful measure of potential intelligence, yielding a "potential I& score", and described elsewhere('). Briefly, after the standard administration of the W-B, the S is asked to give a number of additional answers to those questions in which he did not obtain full credit. A potential I& score is then calculated based on the highest credit for each item regardless of whether the highest credit answer was given on the standard administration or on the additional testing. This method elicits answers that are said to be "potential," but which have not been elicited during the standard administration. This potential I& method, however, has been found to be wearing and laborious. Therefore, the purpose of this study was to determine the degree of relationship between the potential I& score and each of four scores, derived from commonly used and less laborious techniques of estimating intellectual potential. METHOD Our subjects were 60 neuropsychiatric pat#ients admitted consecutively to the Clinical Psychology Service of Fitzsimons Army Hospital, exclusive of military prisoners, children, and emergency cases. The mean W-B I& of this sample was 100.56 with a S.D. of 15.43. Fifty-one patients were male and nine female. Fiftythree were inpatients and seven outpatients. The ages ranged from 17 to 47 with a mean age of 24.18 years and a S.D. of 5.90 years.Each S was first given the standard W-B administration on five verbal subtests: Information, Comprehension, Arithmetic, Similarities, and Vocabulary. Immediately following the standard administration of the W-B, the Ss were readministered the test according to the potential I& method. This procedure yielded a standard W-B I& and a potential I& score for each S. Four techniques, felt to be commonly used in clinical settings, were used to predict the potential I& scores from the subtests of the standard W-B administration:1. The variance was computed for the distribution of the five subtest scaled scores from each S's standard W-B protocol and used as an index of intersubtest variability. Variability is common1 used as an indication of deviation in intellectual pe...

How universal can an intelligence test be?

“…This is a well-known problem in animal cognition (and also happens with other places where intelligence is sought, such as plants, bacteria, SETI, etc.). Moreover, this under-estimation also happens in human psychometrics, where the term 'potential intelligence' is applied to 'test potential' (Mahrer, 1958;Thorp & Mahrer, 1959;Little & Bailey, 1972), which is not to be confused with the term 'potential intelligence' applied to the ability of becoming intelligent . From a computational point of view, we can say that evaluating intelligence (in a universal context) is, at most, semi-computable (i.e.…”

On the universality of cognitive tests

“…This is a well-known problem in animal cognition (and also happens with other places where intelligence is sought, such as plants, bacteria, SETI, etc.). Moreover, this under-estimation also happens in human psychometrics, where the term "potential intelligence" is applied to "test potential" [67,93,63], (not to be confused with the term "potential intelligence" applied to the ability of becoming intelligent [42,43]).…”

On Potential Cognitive Abilities in the Machine Kingdom

“…Note that with this interpretation, the 'potential' would be the right (or corrected) value of an ability, while measurements would be approximations, which are typically -but not always-below that value. This has led to approaches to convert this lower bound into a less biased estimate, trying to predict 'potential' intelligence [51] or calculating how far a test score can be from the actual measure [37]. In a nutshell, the difference between actual and 'potential' would really be applied to the measurement, but not to the individual.…”

“…6], bad interfaces, etc. It is not then the problem referred to by (other uses of) the term potential in psychology [39,51,37], as discussed in section 2. Instead, there are two specific problems for measuring potential cognitive abilities.…”