Part of the problem with the available evidence suggesting a link between autism and pesticide exposure is that it is, to a large extent, observational and correlational. As per the paper from Janie Shelton and colleagues , results are based on variables such as looking at what types and amounts of pesticides are used in a specific area and correlating proximity to that area as a function of a diagnosis of autism or not. Similar to the work looking at other environmental factors linked (or not) to autism such as air pollution (see here), it's all about where and when you were, followed by calculations on your probable exposure being tied into risk.
The paper from Alessia De Felice and colleagues  (open-access) approaches the research area of pesticide exposure and autism (a model of autism) from a slightly different angle. Their findings were based on the use of a mouse model of autism and what happened to mouse offspring when mother mice were exposed to a particular pesticide "at doses devoid of maternal or systemic toxicity."
The mouse model in question was a favourite in autism research circles, the BTBR T+tf/J strain (see here). The pesticide used was "the organophosphate insecticide chlorpyrifos (CPF)." Organophosphate (OP) by the way, refers to the chemical make-up of the compound and its particular effects on the enzyme acetylcholinesterase working in a similar fashion to certain warfare agents.
The De Felice paper is open-access but as ever, a few details are provided:
- Recognising that pesticide exposure already has quite a long history in terms of adverse health effects (pesticide applicators of the world take note), the authors set about looking to "evaluate in the offspring of both sexes the effect of the gestational CPF exposure on spontaneous locomotion, ultrasonic vocalization and neurodevelopment during the first two weeks of postnatal life, to evidence early changes in behavioral profile." Further they decided to "evaluate the long-term effects of CPF on selected markers of the peculiar behavioral repertoire of this mouse strain." You'll note that the word autism does not appear in either of those aims.
- Male and female BTBR mice were allowed to 'get it on' in breeding cages. Females mice were regularly inspected for evidence of carrying a baby mouse and 14 days into the pregnancy were either given peanut oil (vehicle) or CPF dissolved in peanut oil by oral gavage for 4 days. This was given as a single dose.
- Baby mice were born ("Twenty-four litters (13 Vehicle-treated and 11 CPF-treated"). Various behavioural assessments were conducted including an analysis of ultrasonic vocalisations and spontaneous movements. Various other behavioural measures were also made when the baby mice grew into adult mice.
- Results: "prenatal CPF exposure significantly modifies spontaneous motor activity in BTBR pups, delaying their motor development and further enhancing the abnormally high vocalization rates." The authors also reported reduced weight gain in CPF exposed offspring during the early days.
- Looking into adulthood, CPF exposure seemed to show more pronounced effects for males over females with an "altered pattern of investigation of a sexual partner." This finding has however been downplayed by the authors to a certain extent. They conclude with the need for further research in this area "to evaluate the role of environmental chemicals in the etiology of neurodevelopment disorders."
Taking a few steps back, one has to remember that this was a study of mouse exposure to pesticides not humans. Indeed, other similar work from some of the authors has also recently seen the light of day . As per other discussions, mice might be good 'models' of something like autism but they are never ever going to reflect autism (and its important comorbidities) in their entirety. Given also that the BTBR mouse "exhibit a 100% absence of the corpus callosum and a severely reduced hippocampal commissure" according to suppliers, one has to be careful about any sweeping generalisations to all autism. The timing and route of CPF exposure described by De Felice et al might also be seen as potentially important variables insofar as their focus some time into pregnancy and sole reliance on the oral route of exposure.
That all being said, these are potentially important results not least because of their focus on how both genes and environment might be implicated in [models of] cases of autism. As I've mentioned, this was a study of offspring BTBR mice thus implying that there may be some familial predisposition to mimicking the effects of autism anyway. The idea that in predisposed individuals there may be additional effects from an environmental factor is an important one and perhaps complements the whole 'air pollution effects modified by genotype' suggestion (see here) with autism in mind. Certainly, I think there is quite a bit more research to see and do in this area.
Music: Buddy Holly - Rave on!
 Shelton JF. et al. Neurodevelopmental disorders and prenatal residential proximity to agricultural pesticides: The CHARGE study. Environ Health Perspect. 2014: June 23.
 De Felice A. et al. Prenatal Exposure to a Common Organophosphate Insecticide Delays Motor Development in a Mouse Model of Idiopathic Autism. PLoS One. 2015 Mar 24;10(3):e0121663.
 Venerosi A. et al. Effects of maternal chlorpyrifos diet on social investigation and brain neuroendocrine markers in the offspring – a mouse study. Environmental Health 2015, 14:32.
De Felice, A., Scattoni, M., Ricceri, L., & Calamandrei, G. (2015). Prenatal Exposure to a Common Organophosphate Insecticide Delays Motor Development in a Mouse Model of Idiopathic Autism PLOS ONE, 10 (3) DOI: 10.1371/journal.pone.0121663