Wednesday, March 07, 2007

IMFAR 2007 abstracts

The 2007 International Meeting for Autism Research (Seattle this year) schedule for oral and poster presentations has been posted here. I'm involved in three studies at IMFAR this year. Here are the abstracts for two of them:


Pervasive Developmental Disorders Specialized Clinic, Rivière des Prairies Hospital, University of Montreal

HOW MANY HOURS IS FORTY HOURS? RANGE OF TREATMENT INTENSITY IN LOVAAS (1987)
M. Dawson, L. Mottron

Background: 40hrs/wk of one-to-one treatment--the reported intensity of ABA-based intervention received by the experimental group in Lovaas (1987) and McEachin, Smith and Lovaas (1993)--is a benchmark in autism research.

Objective: To compare accounts of hrs/wk received by the experimental group in Lovaas (1987) and McEachin, Smith and Lovaas (1993).

Method: We compared detailed 2004 sworn legal testimony from a behavior analyst (Leaf) directly involved in treatment of 10 of the 19 experimental group children reported in Lovaas (1987) and McEachin, Smith and Lovaas (1993), to descriptions in published articles authored by Lovaas, Smith and/or McEachin.

Results:
34 published descriptions of hrs/wk in the experimental group authored by Lovaas, Smith and/or McEachin, report 40hrs/wk with no range; 4 descriptions (including 1 in both Lovaas, 1987, and McEachin, Smith & Lovaas, 1993) specify 40hrs/wk was a minimum (e.g. "40 hours or more per week", "more than 40 hours"); 7 specify an average of 40hrs/wk, including the only account by Leaf; 3 specify "normal functioning" children received 40hrs/wk; 1 is ambiguous; 2 specify "approximately" 40hrs/wk; and the rest are unqualified. Leaf testified (from Wynberg et al. v. Ontario; Leaf, pp. 16687-16692 [Compendium, Vol. II, Tab 23, pp. 692-697]) under oath that 40 hrs/wk was a group average; range in the experimental group was 18 to "in the 50 range" and that 2 of the 9 "normal functioning" children received 18 hrs/wk which diminished over time.

Conclusion: Leaf's testimony questions the influential premise that high intensity of treatment in Lovaas (1987) was essential in achieving "normal functioning". In the current absence of ABA controlled trials correlating treatment intensity with outcome measures, and considering the landmark status of Lovaas (1987) and its follow-up, we suggest that accurate data re treatment intensity and its relation with outcome measures be provided by the authors.

Sponsor: CIHR



Université de Montréal

INTELLIGENCE IN AUTISM: WHAT ARE THE GOOD PREDICTORS?
L. MOTTRON, I. Soulières, M. Dawson, M. Gernsbacher

Background: Recent findings of discrepancies between intelligence in autism as measured by Wechsler and Raven's Progressive Matrices (Dawson et al, in press) lead to the reconsideration of established relations between variables characterizing the autistic phenotype and intelligence level reached.

Objectives: To establish how early developmental milestones and cross-sectional adaptive level predicts the intelligence level measured by various instruments, including the Raven's Progressive Matrices.

Methods: All measures were extracted from the socio-demographic data of Rivière-des-Prairies Hospital's database, which contains diagnostic and cognitive information on approximately 200 ADI and ADOS-G positive autistics. Correlations were computed between age at first words/phrases-word and other ADI items at age 4-5 and subtests, subscales and global Wechsler intelligence level (WISC-III and WAIS-III), Raven intelligence level, and Vineland adaptation level in school-age children and adults.

Results: Age at first words/phrases was not significantly correlated with intelligence level achieved later in childhood or adulthood, whether measured with Wechsler Scales or Raven's Progressive Matrices. Furthermore, ADI scores (social, communication and repetitive behaviours) was not significantly correlated with intelligence in adulthood.

Conclusion: The age at which first words or phrases are spoken by in autistic children does not predict intelligence level achieved later in childhood or adulthood. Also, ADI scores, often taken to index the so-called "severity" of autistic symptoms, does not predict cognitive outcome in adulthood.

Sponsors: CIHR