If it is agreed that intelligence research has (thus far, and in general) been deformed and devalued by its origins in population-level and statistically-orientated research methods (especially warped by aiming at group--level social scientific uses in education research, demographics and epidemiology)
-then there is a clear need for individual-level research methods;
without which we can never understand the qualitative nature of 'general intelligence' but will remain stuck in the present (and for the past century plus) situation of merely an abstract and statistical and correlational non-definition of general intelligence, which utterly fails to describe or explain 'what intelligence actually is'.
This problem seems very obvious to me, coming into intelligence research (as I do) from a background in medicine, psychiatry and evolutionary biology.
It is clear that mainstream intelligence research is and always has been (to put it crudely, but accurately) non-biological.
One remedy would be to develop new strands of intelligence research which focus on individuals; in other words case study methodologies.
Specifically, intensive 'clinical' study of individuals who are selected for their unusual attributes: in other words an adaptation of the approach pioneered in Cognitive Neuropsychology (in which individuals with unusual forms of localized brain damage were studied intensively).
This was later extended to psychiatry as Cognitive Neuropsychiatry - for example, the 'looking for exceptions to a theoretically-grounded prediction' (or 'black swan') approach I used in Walston F, Blennerhassett RC, Charlton BG. (2000). ‘Theory of mind’, persecutory delusions and the somatic marker mechanism. Cognitive Neuropsychiatry 5: 161-174.
(Authors who have written in this tradition include Oliver Sacks, Antonio Damasio and (in a useful textbook) Tim Shallice. - see my references in the previous blog post.)
Focus of case studies
While traditional, population-based intelligence research has focused on as-large-as-possible, norm-based, representative samples - and on studying population differences, average effects and correlations...
in contrast, case study based research focuses on the unusual.
For example, while Terman's 1920s longitudinal, follow-up cohort research on high IQ children, or Deary's recent analysis of a more-representative cohort from the same era, focus upon normal, averages, correlations etc. - case based research would focus upon the exceptions to these correlations.
Possible suitable and potentially-fruitful subjects for a Cognitive Neurointelligence approach would be:
Those people who had high IQ as children but who did NOT have successful careers as adults. In other words, Grady M Towers 'Outsiders' -
The misfits, the people who sometimes congregates in high (and ultra-high) IQ societies; whose personalities stand in the path of successful careers (e.g. Grady Towers was a night watchman).
And/ or very high IQ people who suffer illness, localized brain damage, who are psychopathic, neurotic, psychotic - who have poor concentration, a small working-memory...
Or people who have very fast simple reaction times, probably indicating fast processing and highly efficient brains - but who have low IQ test scores.
In such studies population norms ('averages') are used in a qualitative fashion which does not depend upon great precision and is not sensitive to the exact representativeness of population sampling - measures simply to indicate whether the person has about-average performance - or significantly above or significantly below average performance.
So, test measures are used in a yes/no fashion - and a profiles can be built up for each individual in terms of three categories:
1. About average
2. Certainly above average
3. Certainly below average
Why? because is it (for instance) much easier to understand the nature of intelligence when intelligence is pretty much the only advantage a person has - when they have high intelligence but - say - poor concentration - what can they do and what can they not do?
When somebody has high intelligence but a psychopathic personality - how do they function socially, educationally, vocationally, under formal testing - what can they do at normal levels, what better than average, what worse than average?
What kind of a person shows very fast reaction times but performs badly on IQ tests - are there any subtests (with modifications) tests on which they perform very well - if so, what characteristics do these subtests have?
The great thing about this kind of research is that it can be done (in fact, needs to be done) by many people of many types (including, of course, self-study, detailed autobiography, and self-experimentation) - it can be amateur science - and thereby escape from the wholesale careerism, bureaucratization and in-general corruption of mainstream modern 'science'.