A proposal for the use of Bayesian probabilities in neuropsychological assessment.

Abstract

A measurement model based on Bayesian probabilities is proposed as an alternative to approaches based on classical measurement theory. The suggested model includes use of interactive microcomputer programs to integrate clinical im~ pressions into the metric assessment procedure. The Bayesian model provides a new focus in neuropsychological assessment for both cognitive and affective functioning.

The Value of Bayesian Methods for Accurate and Efficient Neuropsychological Assessment

“…The implications of Bayes' theorem for clinical neuropsychological diagnosis have been discussed as early as in 1955 (Crawford, Garthwaite, & Betkowska, 2009;Elwood, 1993;Frederick & Bowden, 2009;Gavett, 2015;Gordon, 1977;Gouvier, 2001;Ivnik et al, 2001;Jones, 1989;Labarge et al, 2003;Meehl & Rosen, 1955;O'Bryant & Lucas, 2006;Satz et al, 1970;Slick, 2006). Despite these early insights, a review in 2003 observed that it was uncommon to consider the impact of base rates in diagnostic accuracy studies (Woods et al, 2003).…”

“…Although Bayes theorem is not new in the field of neuropsychology (e.g., Elwood, 1993;Gavett, 2015;Gelman & Geurts, 2017;Jones, 1989), Bayesian hypothesis testing has received less attention.…”

“…Although Bayes theorem is not new in the field of neuropsychology (e.g., Elwood, 1993;Gavett, 2015;Gelman & Geurts, 2017;Jones, 1989), Bayesian hypothesis testing has received less attention. Bayesian inference has been critiqued to be too difficult, which has often impeded widespread adoption of Bayesian methods.…”