Misc,

Assessment of discrimination skills in individuals with autism: Validity of the assessment of basic learning abilities (abla)

.
(2003)UM: AAINQ74662; MT: Print; FO: Print; PO: Human.

Abstract

The Assessment of Basic Learning Abilities (ABLA) is a reliable and valid measure of discrimination skills in individuals with Developmental Disabilities (DD), but there are several gaps in the few studies that have attempted to extend its use to individuals with Autism and accompanying mental retardation. The study was designed (a) to determine whether ABLA performance is indicative of real world learning and (b) to examine whether the ABLA has clinical value beyond current assessment practices. Each experimental task had six discrimination levels designed to be analogous to the ABLA. One set of tasks used materials that were virtually identical to the ABLA (differing by colour), while the other used common everyday materials. ABLA Performance was compared to performance on these tasks delivered as static probes, and as dynamic teaching tasks. The ABLA accurately predicted learning (95% agreement) on virtually identical tasks, but was less accurate at predicting everyday discrimination skill learning (78%) than static baseline probes (89%), and took significantly longer to administer. Performance on the ABLA did not predict skill acquisition on everyday tasks, in this sample with autism, to the degree that it has for individuals with DD. Further, the hierarchy of task difficulty of everyday skills was consistent with the disordered development that is characteristic of Autism. The lack of generalization of skills across materials was related to differences in task difficulty and stimulus overselectivity. A hierarchical multiple regression procedure was used to test a model of everyday discrimination skill learning with: chronological age, mental age, adaptive skill level, communication ability, severity of autism, and ABLA level as predictors. Chronological age, mental age, and adaptive behaviour accounted for 50% of the variance in the model. The ABLA contributed a small portion (6%) of unique variance over and above the standardized psychological assessment protocol. Overall, the ABLA does not appear to contribute meaningful information not already provided by a thorough developmental and diagnostic assessment. This, combined with the lack of generalization of performance to everyday discrimination tasks, suggests that the ABLA (in its present form) is not an adequate measure of discrimination skills for individuals with Autism. (PsycINFO Database Record (c) 2006 APA, all rights reserved)

Tags

Users

  • @quinn

Comments and Reviews